# GNU gettext' utilities

Next: , Previous: (dir), Up: (dir)

## GNU gettext utilities

This manual documents the GNU gettext tools and the GNU libintl library, version 0.17.

--- The Detailed Node Listing ---

Introduction

The User's View

Setting the POSIX Locale

Preparing Program Sources

Making the PO Template File

Creating a New PO File

Updating Existing PO Files

Editing PO Files

Emacs's PO File Editor

Using Translation Compendia

Manipulating PO Files

Highlighting parts of PO files

Producing Binary MO Files

The Programmer's View

About catgets

About gettext

Temporary Notes for the Programmers Chapter

The Translator's View

Organization

National Teams

The Maintainer's View

Files You Must Create or Alter

Autoconf macros for use in configure.ac

Integrating with CVS

Other Programming Languages

The Translator's View

Individual Programming Languages

sh - Shell Script

Perl

Internationalizable Data

Concluding Remarks

Language Codes

Next: , Previous: Top, Up: Top

## 1 Introduction

This chapter explains the goals sought in the creation of GNU gettext and the free Translation Project. Then, it explains a few broad concepts around Native Language Support, and positions message translation with regard to other aspects of national and cultural variance, as they apply to programs. It also surveys those files used to convey the translations. It explains how the various tools interact in the initial generation of these files, and later, how the maintenance cycle should usually operate.

In this manual, we use he when speaking of the programmer or maintainer, she when speaking of the translator, and they when speaking of the installers or end users of the translated program. This is only a convenience for clarifying the documentation. It is absolutely not meant to imply that some roles are more appropriate to males or females. Besides, as you might guess, GNU gettext is meant to be useful for people using computers, whatever their sex, race, religion or nationality!

Please send suggestions and corrections to:

     Internet address:
bug-gnu-gettext@gnu.org


Please include the manual's edition number and update date in your messages.

Next: , Previous: Introduction, Up: Introduction

### 1.1 The Purpose of GNU gettext

Usually, programs are written and documented in English, and use English at execution time to interact with users. This is true not only of GNU software, but also of a great deal of proprietary and free software. Using a common language is quite handy for communication between developers, maintainers and users from all countries. On the other hand, most people are less comfortable with English than with their own native language, and would prefer to use their mother tongue for day to day's work, as far as possible. Many would simply love to see their computer screen showing a lot less of English, and far more of their own language.

However, to many people, this dream might appear so far fetched that they may believe it is not even worth spending time thinking about it. They have no confidence at all that the dream might ever become true. Yet some have not lost hope, and have organized themselves. The Translation Project is a formalization of this hope into a workable structure, which has a good chance to get all of us nearer the achievement of a truly multi-lingual set of programs.

GNU gettext is an important step for the Translation Project, as it is an asset on which we may build many other steps. This package offers to programmers, translators and even users, a well integrated set of tools and documentation. Specifically, the GNU gettext utilities are a set of tools that provides a framework within which other free packages may produce multi-lingual messages. These tools include

• A set of conventions about how programs should be written to support message catalogs.
• A directory and file naming organization for the message catalogs themselves.
• A runtime library supporting the retrieval of translated messages.
• A few stand-alone programs to massage in various ways the sets of translatable strings, or already translated strings.
• A library supporting the parsing and creation of files containing translated messages.
• A special mode for Emacs1 which helps preparing these sets and bringing them up to date.

GNU gettext is designed to minimize the impact of internationalization on program sources, keeping this impact as small and hardly noticeable as possible. Internationalization has better chances of succeeding if it is very light weighted, or at least, appear to be so, when looking at program sources.

The Translation Project also uses the GNU gettext distribution as a vehicle for documenting its structure and methods. This goes beyond the strict technicalities of documenting the GNU gettext proper. By so doing, translators will find in a single place, as far as possible, all they need to know for properly doing their translating work. Also, this supplemental documentation might also help programmers, and even curious users, in understanding how GNU gettext is related to the remainder of the Translation Project, and consequently, have a glimpse at the big picture.

Next: , Previous: Why, Up: Introduction

### 1.2 I18n, L10n, and Such

Two long words appear all the time when we discuss support of native language in programs, and these words have a precise meaning, worth being explained here, once and for all in this document. The words are internationalization and localization. Many people, tired of writing these long words over and over again, took the habit of writing i18n and l10n instead, quoting the first and last letter of each word, and replacing the run of intermediate letters by a number merely telling how many such letters there are. But in this manual, in the sake of clarity, we will patiently write the names in full, each time...

By internationalization, one refers to the operation by which a program, or a set of programs turned into a package, is made aware of and able to support multiple languages. This is a generalization process, by which the programs are untied from calling only English strings or other English specific habits, and connected to generic ways of doing the same, instead. Program developers may use various techniques to internationalize their programs. Some of these have been standardized. GNU gettext offers one of these standards. See Programmers.

By localization, one means the operation by which, in a set of programs already internationalized, one gives the program all needed information so that it can adapt itself to handle its input and output in a fashion which is correct for some native language and cultural habits. This is a particularisation process, by which generic methods already implemented in an internationalized program are used in specific ways. The programming environment puts several functions to the programmers disposal which allow this runtime configuration. The formal description of specific set of cultural habits for some country, together with all associated translations targeted to the same native language, is called the locale for this language or country. Users achieve localization of programs by setting proper values to special environment variables, prior to executing those programs, identifying which locale should be used.

In fact, locale message support is only one component of the cultural data that makes up a particular locale. There are a whole host of routines and functions provided to aid programmers in developing internationalized software and which allow them to access the data stored in a particular locale. When someone presently refers to a particular locale, they are obviously referring to the data stored within that particular locale. Similarly, if a programmer is referring to “accessing the locale routines”, they are referring to the complete suite of routines that access all of the locale's information.

One uses the expression Native Language Support, or merely NLS, for speaking of the overall activity or feature encompassing both internationalization and localization, allowing for multi-lingual interactions in a program. In a nutshell, one could say that internationalization is the operation by which further localizations are made possible.

Also, very roughly said, when it comes to multi-lingual messages, internationalization is usually taken care of by programmers, and localization is usually taken care of by translators.

Next: , Previous: Concepts, Up: Introduction

### 1.3 Aspects in Native Language Support

For a totally multi-lingual distribution, there are many things to translate beyond output messages.

• As of today, GNU gettext offers a complete toolset for translating messages output by C programs. Perl scripts and shell scripts will also need to be translated. Even if there are today some hooks by which this can be done, these hooks are not integrated as well as they should be.
• Some programs, like autoconf or bison, are able to produce other programs (or scripts). Even if the generating programs themselves are internationalized, the generated programs they produce may need internationalization on their own, and this indirect internationalization could be automated right from the generating program. In fact, quite usually, generating and generated programs could be internationalized independently, as the effort needed is fairly orthogonal.
• A few programs include textual tables which might need translation themselves, independently of the strings contained in the program itself. For example, RFC 1345 gives an English description for each character which the recode program is able to reconstruct at execution. Since these descriptions are extracted from the RFC by mechanical means, translating them properly would require a prior translation of the RFC itself.
• Almost all programs accept options, which are often worded out so to be descriptive for the English readers; one might want to consider offering translated versions for program options as well.
• Many programs read, interpret, compile, or are somewhat driven by input files which are texts containing keywords, identifiers, or replies which are inherently translatable. For example, one may want gcc to allow diacriticized characters in identifiers or use translated keywords; ‘rm -i’ might accept something else than ‘y’ or ‘n’ for replies, etc. Even if the program will eventually make most of its output in the foreign languages, one has to decide whether the input syntax, option values, etc., are to be localized or not.
• The manual accompanying a package, as well as all documentation files in the distribution, could surely be translated, too. Translating a manual, with the intent of later keeping up with updates, is a major undertaking in itself, generally.

As we already stressed, translation is only one aspect of locales. Other internationalization aspects are system services and are handled in GNU libc. There are many attributes that are needed to define a country's cultural conventions. These attributes include beside the country's native language, the formatting of the date and time, the representation of numbers, the symbols for currency, etc. These local rules are termed the country's locale. The locale represents the knowledge needed to support the country's native attributes.

There are a few major areas which may vary between countries and hence, define what a locale must describe. The following list helps putting multi-lingual messages into the proper context of other tasks related to locales. See the GNU libc manual for details.

Characters and Codesets
The codeset most commonly used through out the USA and most English speaking parts of the world is the ASCII codeset. However, there are many characters needed by various locales that are not found within this codeset. The 8-bit ISO 8859-1 code set has most of the special characters needed to handle the major European languages. However, in many cases, choosing ISO 8859-1 is nevertheless not adequate: it doesn't even handle the major European currency. Hence each locale will need to specify which codeset they need to use and will need to have the appropriate character handling routines to cope with the codeset.
Currency
The symbols used vary from country to country as does the position used by the symbol. Software needs to be able to transparently display currency figures in the native mode for each locale.
Dates
The format of date varies between locales. For example, Christmas day in 1994 is written as 12/25/94 in the USA and as 25/12/94 in Australia. Other countries might use ISO 8601 dates, etc.

Time of the day may be noted as hh:mm, hh.mm, or otherwise. Some locales require time to be specified in 24-hour mode rather than as AM or PM. Further, the nature and yearly extent of the Daylight Saving correction vary widely between countries.

Numbers
Numbers can be represented differently in different locales. For example, the following numbers are all written correctly for their respective locales:
          12,345.67       English
12.345,67       German
12345,67       French
1,2345.67       Asia


Some programs could go further and use different unit systems, like English units or Metric units, or even take into account variants about how numbers are spelled in full.

Messages
The most obvious area is the language support within a locale. This is where GNU gettext provides the means for developers and users to easily change the language that the software uses to communicate to the user.

These areas of cultural conventions are called locale categories. It is an unfortunate term; locale aspects or locale feature categories would be a better term, because each “locale category” describes an area or task that requires localization. The concrete data that describes the cultural conventions for such an area and for a particular culture is also called a locale category. In this sense, a locale is composed of several locale categories: the locale category describing the codeset, the locale category describing the formatting of numbers, the locale category containing the translated messages, and so on.

Components of locale outside of message handling are standardized in the ISO C standard and the POSIX:2001 standard (also known as the SUSV3 specification). GNU libc fully implements this, and most other modern systems provide a more or less reasonable support for at least some of the missing components.

Next: , Previous: Aspects, Up: Introduction

### 1.4 Files Conveying Translations

The letters PO in .po files means Portable Object, to distinguish it from .mo files, where MO stands for Machine Object. This paradigm, as well as the PO file format, is inspired by the NLS standard developed by Uniforum, and first implemented by Sun in their Solaris system.

PO files are meant to be read and edited by humans, and associate each original, translatable string of a given package with its translation in a particular target language. A single PO file is dedicated to a single target language. If a package supports many languages, there is one such PO file per language supported, and each package has its own set of PO files. These PO files are best created by the xgettext program, and later updated or refreshed through the msgmerge program. Program xgettext extracts all marked messages from a set of C files and initializes a PO file with empty translations. Program msgmerge takes care of adjusting PO files between releases of the corresponding sources, commenting obsolete entries, initializing new ones, and updating all source line references. Files ending with .pot are kind of base translation files found in distributions, in PO file format.

MO files are meant to be read by programs, and are binary in nature. A few systems already offer tools for creating and handling MO files as part of the Native Language Support coming with the system, but the format of these MO files is often different from system to system, and non-portable. The tools already provided with these systems don't support all the features of GNU gettext. Therefore GNU gettext uses its own format for MO files. Files ending with .gmo are really MO files, when it is known that these files use the GNU format.

Previous: Files, Up: Introduction

### 1.5 Overview of GNU gettext

The following diagram summarizes the relation between the files handled by GNU gettext and the tools acting on these files. It is followed by somewhat detailed explanations, which you should read while keeping an eye on the diagram. Having a clear understanding of these interrelations will surely help programmers, translators and maintainers.

     Original C Sources ───> Preparation ───> Marked C Sources ───╮
│
╭─────────<─── GNU gettext Library             │
╭─── make <───┤                                              │
│             ╰─────────<────────────────────┬───────────────╯
│                                            │
│   ╭─────<─── PACKAGE.pot <─── xgettext <───╯   ╭───<─── PO Compendium
│   │                                            │              ↑
│   │                                            ╰───╮          │
│   ╰───╮                                            ├───> PO editor ───╮
│       ├────> msgmerge ──────> LANG.po ────>────────╯                  │
│   ╭───╯                                                               │
│   │                                                                   │
│   ╰─────────────<───────────────╮                                     │
│                                 ├─── New LANG.po <────────────────────╯
│   ╭─── LANG.gmo <─── msgfmt <───╯
│   │
│   ╰───> install ───> /.../LANG/PACKAGE.mo ───╮
│                                              ├───> "Hello world!"
╰───────> install ───> /.../bin/PROGRAM ───────╯


As a programmer, the first step to bringing GNU gettext into your package is identifying, right in the C sources, those strings which are meant to be translatable, and those which are untranslatable. This tedious job can be done a little more comfortably using emacs PO mode, but you can use any means familiar to you for modifying your C sources. Beside this some other simple, standard changes are needed to properly initialize the translation library. See Sources, for more information about all this.

For newly written software the strings of course can and should be marked while writing it. The gettext approach makes this very easy. Simply put the following lines at the beginning of each file or in a central header file:

     #define _(String) (String)
#define N_(String) String
#define textdomain(Domain)
#define bindtextdomain(Package, Directory)


Doing this allows you to prepare the sources for internationalization. Later when you feel ready for the step to use the gettext library simply replace these definitions by the following:

     #include <libintl.h>
#define _(String) gettext (String)
#define gettext_noop(String) String
#define N_(String) gettext_noop (String)


and link against libintl.a or libintl.so. Note that on GNU systems, you don't need to link with libintl because the gettext library functions are already contained in GNU libc. That is all you have to change.

Once the C sources have been modified, the xgettext program is used to find and extract all translatable strings, and create a PO template file out of all these. This package.pot file contains all original program strings. It has sets of pointers to exactly where in C sources each string is used. All translations are set to empty. The letter t in .pot marks this as a Template PO file, not yet oriented towards any particular language. See xgettext Invocation, for more details about how one calls the xgettext program. If you are really lazy, you might be interested at working a lot more right away, and preparing the whole distribution setup (see Maintainers). By doing so, you spare yourself typing the xgettext command, as make should now generate the proper things automatically for you!

The first time through, there is no lang.po yet, so the msgmerge step may be skipped and replaced by a mere copy of package.pot to lang.po, where lang represents the target language. See Creating for details.

Then comes the initial translation of messages. Translation in itself is a whole matter, still exclusively meant for humans, and whose complexity far overwhelms the level of this manual. Nevertheless, a few hints are given in some other chapter of this manual (see Translators). You will also find there indications about how to contact translating teams, or becoming part of them, for sharing your translating concerns with others who target the same native language.

While adding the translated messages into the lang.po PO file, if you are not using one of the dedicated PO file editors (see Editing), you are on your own for ensuring that your efforts fully respect the PO file format, and quoting conventions (see PO Files). This is surely not an impossible task, as this is the way many people have handled PO files around 1995. On the other hand, by using a PO file editor, most details of PO file format are taken care of for you, but you have to acquire some familiarity with PO file editor itself.

If some common translations have already been saved into a compendium PO file, translators may use PO mode for initializing untranslated entries from the compendium, and also save selected translations into the compendium, updating it (see Compendium). Compendium files are meant to be exchanged between members of a given translation team.

Programs, or packages of programs, are dynamic in nature: users write bug reports and suggestion for improvements, maintainers react by modifying programs in various ways. The fact that a package has already been internationalized should not make maintainers shy of adding new strings, or modifying strings already translated. They just do their job the best they can. For the Translation Project to work smoothly, it is important that maintainers do not carry translation concerns on their already loaded shoulders, and that translators be kept as free as possible of programming concerns.

The only concern maintainers should have is carefully marking new strings as translatable, when they should be, and do not otherwise worry about them being translated, as this will come in proper time. Consequently, when programs and their strings are adjusted in various ways by maintainers, and for matters usually unrelated to translation, xgettext would construct package.pot files which are evolving over time, so the translations carried by lang.po are slowly fading out of date.

It is important for translators (and even maintainers) to understand that package translation is a continuous process in the lifetime of a package, and not something which is done once and for all at the start. After an initial burst of translation activity for a given package, interventions are needed once in a while, because here and there, translated entries become obsolete, and new untranslated entries appear, needing translation.

The msgmerge program has the purpose of refreshing an already existing lang.po file, by comparing it with a newer package.pot template file, extracted by xgettext out of recent C sources. The refreshing operation adjusts all references to C source locations for strings, since these strings move as programs are modified. Also, msgmerge comments out as obsolete, in lang.po, those already translated entries which are no longer used in the program sources (see Obsolete Entries). It finally discovers new strings and inserts them in the resulting PO file as untranslated entries (see Untranslated Entries). See msgmerge Invocation, for more information about what msgmerge really does.

Whatever route or means taken, the goal is to obtain an updated lang.po file offering translations for all strings.

The temporal mobility, or fluidity of PO files, is an integral part of the translation game, and should be well understood, and accepted. People resisting it will have a hard time participating in the Translation Project, or will give a hard time to other participants! In particular, maintainers should relax and include all available official PO files in their distributions, even if these have not recently been updated, without exerting pressure on the translator teams to get the job done. The pressure should rather come from the community of users speaking a particular language, and maintainers should consider themselves fairly relieved of any concern about the adequacy of translation files. On the other hand, translators should reasonably try updating the PO files they are responsible for, while the package is undergoing pretest, prior to an official distribution.

Once the PO file is complete and dependable, the msgfmt program is used for turning the PO file into a machine-oriented format, which may yield efficient retrieval of translations by the programs of the package, whenever needed at runtime (see MO Files). See msgfmt Invocation, for more information about all modes of execution for the msgfmt program.

Finally, the modified and marked C sources are compiled and linked with the GNU gettext library, usually through the operation of make, given a suitable Makefile exists for the project, and the resulting executable is installed somewhere users will find it. The MO files themselves should also be properly installed. Given the appropriate environment variables are set (see Setting the POSIX Locale), the program should localize itself automatically, whenever it executes.

The remainder of this manual has the purpose of explaining in depth the various steps outlined above.

Next: , Previous: Introduction, Up: Top

## 2 The User's View

Nowadays, when users log into a computer, they usually find that all their programs show messages in their native language – at least for users of languages with an active free software community, like French or German; to a lesser extent for languages with a smaller participation in free software and the GNU project, like Hindi and Filipino.

How does this work? How can the user influence the language that is used by the programs? This chapter will answer it.

Next: , Previous: Users, Up: Users

### 2.1 Operating System Installation

The default language is often already specified during operating system installation. When the operating system is installed, the installer typically asks for the language used for the installation process and, separately, for the language to use in the installed system. Some OS installers only ask for the language once.

This determines the system-wide default language for all users. But the installers often give the possibility to install extra localizations for additional languages. For example, the localizations of KDE (the K Desktop Environment) and OpenOffice.org are often bundled separately, as one installable package per language.

At this point it is good to consider the intended use of the machine: If it is a machine designated for personal use, additional localizations are probably not necessary. If, however, the machine is in use in an organization or company that has international relationships, one can consider the needs of guest users. If you have a guest from abroad, for a week, what could be his preferred locales? It may be worth installing these additional localizations ahead of time, since they cost only a bit of disk space at this point.

The system-wide default language is the locale configuration that is used when a new user account is created. But the user can have his own locale configuration that is different from the one of the other users of the same machine. He can specify it, typically after the first login, as described in the next section.

Next: , Previous: System Installation, Up: Users

### 2.2 Setting the Locale Used by GUI Programs

The immediately available programs in a user's desktop come from a group of programs called a “desktop environment”; it usually includes the window manager, a web browser, a text editor, and more. The most common free desktop environments are KDE, GNOME, and Xfce.

The locale used by GUI programs of the desktop environment can be specified in a configuration screen called “control center”, “language settings” or “country settings”.

Individual GUI programs that are not part of the desktop environment can have their locale specified either in a settings panel, or through environment variables.

For some programs, it is possible to specify the locale through environment variables, possibly even to a different locale than the desktop's locale. This means, instead of starting a program through a menu or from the file system, you can start it from the command-line, after having set some environment variables. The environment variables can be those specified in the next section (Setting the POSIX Locale); for some versions of KDE, however, the locale is specified through a variable KDE_LANG, rather than LANG or LC_ALL.

Next: , Previous: Setting the GUI Locale, Up: Users

### 2.3 Setting the Locale through Environment Variables

As a user, if your language has been installed for this package, in the simplest case, you only have to set the LANG environment variable to the appropriate ‘ll_CC’ combination. For example, let's suppose that you speak German and live in Germany. At the shell prompt, merely execute ‘setenv LANG de_DE (in csh), ‘export LANG; LANG=de_DE (in sh) or ‘export LANG=de_DE (in bash). This can be done from your .login or .profile file, once and for all.

Next: , Previous: Setting the POSIX Locale, Up: Setting the POSIX Locale

#### 2.3.1 Locale Names

A locale name usually has the form ‘ll_CC’. Here ‘ll’ is an ISO 639 two-letter language code, and ‘CC’ is an ISO 3166 two-letter country code. For example, for German in Germany, ll is de, and CC is DE. You find a list of the language codes in appendix Language Codes and a list of the country codes in appendix Country Codes.

You might think that the country code specification is redundant. But in fact, some languages have dialects in different countries. For example, ‘de_AT’ is used for Austria, and ‘pt_BR’ for Brazil. The country code serves to distinguish the dialects.

Many locale names have an extended syntax ‘ll_CC.encoding’ that also specifies the character encoding. These are in use because between 2000 and 2005, most users have switched to locales in UTF-8 encoding. For example, the German locale on glibc systems is nowadays ‘de_DE.UTF-8’. The older name ‘de_DE’ still refers to the German locale as of 2000 that stores characters in ISO-8859-1 encoding – a text encoding that cannot even accomodate the Euro currency sign.

Some locale names use ‘ll_CC.@variant’ instead of ‘ll_CC’. The ‘@variant’ can denote any kind of characteristics that is not already implied by the language ll and the country CC. It can denote a particular monetary unit. For example, on glibc systems, ‘de_DE@euro’ denotes the locale that uses the Euro currency, in contrast to the older locale ‘de_DE’ which implies the use of the currency before 2002. It can also denote a dialect of the language, or the script used to write text (for example, ‘sr_RS@latin’ uses the Latin script, whereas ‘sr_RS’ uses the Cyrillic script to write Serbian), or the orthography rules, or similar.

On other systems, some variations of this scheme are used, such as ‘ll’. You can get the list of locales supported by your system for your language by running the command ‘locale -a | grep '^ll'’.

There is also a special locale, called ‘C’. When it is used, it disables all localization: in this locale, all programs standardized by POSIX use English messages and an unspecified character encoding (often US-ASCII, but sometimes also ISO-8859-1 or UTF-8, depending on the operating system).

Next: , Previous: Locale Names, Up: Setting the POSIX Locale

#### 2.3.2 Locale Environment Variables

A locale is composed of several locale categories, see Aspects. When a program looks up locale dependent values, it does this according to the following environment variables, in priority order:

1. LANGUAGE
2. LC_ALL
3. LC_xxx, according to selected locale category: LC_CTYPE, LC_NUMERIC, LC_TIME, LC_COLLATE, LC_MONETARY, LC_MESSAGES, ...
4. LANG

Variables whose value is set but is empty are ignored in this lookup.

LANG is the normal environment variable for specifying a locale. As a user, you normally set this variable (unless some of the other variables have already been set by the system, in /etc/profile or similar initialization files).

LC_CTYPE, LC_NUMERIC, LC_TIME, LC_COLLATE, LC_MONETARY, LC_MESSAGES, and so on, are the environment variables meant to override LANG and affecting a single locale category only. For example, assume you are a Swedish user in Spain, and you want your programs to handle numbers and dates according to Spanish conventions, and only the messages should be in Swedish. Then you could create a locale named ‘sv_ES’ or ‘sv_ES.UTF-8’ by use of the localedef program. But it is simpler, and achieves the same effect, to set the LANG variable to es_ES.UTF-8 and the LC_MESSAGES variable to sv_SE.UTF-8; these two locales come already preinstalled with the operating system.

LC_ALL is an environment variable that overrides all of these. It is typically used in scripts that run particular programs. For example, configure scripts generated by GNU autoconf use LC_ALL to make sure that the configuration tests don't operate in locale dependent ways.

Some systems, unfortunately, set LC_ALL in /etc/profile or in similar initialization files. As a user, you therefore have to unset this variable if you want to set LANG and optionally some of the other LC_xxx variables.

The LANGUAGE variable is described in the next subsection.

Previous: Locale Environment Variables, Up: Setting the POSIX Locale

#### 2.3.3 Specifying a Priority List of Languages

Not all programs have translations for all languages. By default, an English message is shown in place of a nonexistent translation. If you understand other languages, you can set up a priority list of languages. This is done through a different environment variable, called LANGUAGE. GNU gettext gives preference to LANGUAGE over LC_ALL and LANG for the purpose of message handling, but you still need to have LANG (or LC_ALL) set to the primary language; this is required by other parts of the system libraries. For example, some Swedish users who would rather read translations in German than English for when Swedish is not available, set LANGUAGE to ‘sv:de’ while leaving LANG to ‘sv_SE’.

Special advice for Norwegian users: The language code for Norwegian bokma*l changed from ‘no’ to ‘nb’ recently (in 2003). During the transition period, while some message catalogs for this language are installed under ‘nb’ and some older ones under ‘no’, it is recommended for Norwegian users to set LANGUAGE to ‘nb:no’ so that both newer and older translations are used.

In the LANGUAGE environment variable, but not in the other environment variables, ‘ll_CC’ combinations can be abbreviated as ‘ll’ to denote the language's main dialect. For example, ‘de’ is equivalent to ‘de_DE’ (German as spoken in Germany), and ‘pt’ to ‘pt_PT’ (Portuguese as spoken in Portugal) in this context.

Note: The variable LANGUAGE is ignored if the locale is set to ‘C’. In other words, you have to first enable localization, by setting LANG (or LC_ALL) to a value other than ‘C’, before you can use a language priority list through the LANGUAGE variable.

Previous: Setting the POSIX Locale, Up: Users

### 2.4 Installing Translations for Particular Programs

Languages are not equally well supported in all packages using GNU gettext, and more translations are added over time. Usually, you use the translations that are shipped with the operating system or with particular packages that you install afterwards. But you can also install newer localizations directly. For doing this, you will need an understanding where each localization file is stored on the file system.

For programs that participate in the Translation Project, you can start looking for translations here: http://translationproject.org/team/index.html. A snapshot of this information is also found in the ABOUT-NLS file that is shipped with GNU gettext.

For programs that are part of the KDE project, the starting point is: http://i18n.kde.org/.

For programs that are part of the GNOME project, the starting point is: http://www.gnome.org/i18n/.

For other programs, you may check whether the program's source code package contains some ll.po files; often they are kept together in a directory called po/. Each ll.po file contains the message translations for the language whose abbreviation of ll.

Next: , Previous: Users, Up: Top

## 3 The Format of PO Files

The GNU gettext toolset helps programmers and translators at producing, updating and using translation files, mainly those PO files which are textual, editable files. This chapter explains the format of PO files.

A PO file is made up of many entries, each entry holding the relation between an original untranslated string and its corresponding translation. All entries in a given PO file usually pertain to a single project, and all translations are expressed in a single target language. One PO file entry has the following schematic structure:

     white-space
#: reference...
#, flag...
#| msgid previous-untranslated-string
msgid untranslated-string
msgstr translated-string


The general structure of a PO file should be well understood by the translator. When using PO mode, very little has to be known about the format details, as PO mode takes care of them for her.

A simple entry can look like this:

     #: lib/error.c:116
msgid "Unknown system error"
msgstr "Error desconegut del sistema"


Entries begin with some optional white space. Usually, when generated through GNU gettext tools, there is exactly one blank line between entries. Then comments follow, on lines all starting with the character #. There are two kinds of comments: those which have some white space immediately following the # - the translator comments -, which comments are created and maintained exclusively by the translator, and those which have some non-white character just after the # - the automatic comments -, which comments are created and maintained automatically by GNU gettext tools. Comment lines starting with #. contain comments given by the programmer, directed at the translator; these comments are called extracted comments because the xgettext program extracts them from the program's source code. Comment lines starting with #: contain references to the program's source code. Comment lines starting with #, contain flags; more about these below. Comment lines starting with #| contain the previous untranslated string for which the translator gave a translation.

All comments, of either kind, are optional.

After white space and comments, entries show two strings, namely first the untranslated string as it appears in the original program sources, and then, the translation of this string. The original string is introduced by the keyword msgid, and the translation, by msgstr. The two strings, untranslated and translated, are quoted in various ways in the PO file, using " delimiters and \ escapes, but the translator does not really have to pay attention to the precise quoting format, as PO mode fully takes care of quoting for her.

The msgid strings, as well as automatic comments, are produced and managed by other GNU gettext tools, and PO mode does not provide means for the translator to alter these. The most she can do is merely deleting them, and only by deleting the whole entry. On the other hand, the msgstr string, as well as translator comments, are really meant for the translator, and PO mode gives her the full control she needs.

The comment lines beginning with #, are special because they are not completely ignored by the programs as comments generally are. The comma separated list of flags is used by the msgfmt program to give the user some better diagnostic messages. Currently there are two forms of flags defined:

fuzzy
This flag can be generated by the msgmerge program or it can be inserted by the translator herself. It shows that the msgstr string might not be a correct translation (anymore). Only the translator can judge if the translation requires further modification, or is acceptable as is. Once satisfied with the translation, she then removes this fuzzy attribute. The msgmerge program inserts this when it combined the msgid and msgstr entries after fuzzy search only. See Fuzzy Entries.
c-format
no-c-format
These flags should not be added by a human. Instead only the xgettext program adds them. In an automated PO file processing system as proposed here the user changes would be thrown away again as soon as the xgettext program generates a new template file.

The c-format flag tells that the untranslated string and the translation are supposed to be C format strings. The no-c-format flag tells that they are not C format strings, even though the untranslated string happens to look like a C format string (with ‘%’ directives).

In case the c-format flag is given for a string the msgfmt does some more tests to check to validity of the translation. See msgfmt Invocation, c-format Flag and c-format.

objc-format
no-objc-format
Likewise for Objective C, see objc-format.
sh-format
no-sh-format
Likewise for Shell, see sh-format.
python-format
no-python-format
Likewise for Python, see python-format.
lisp-format
no-lisp-format
Likewise for Lisp, see lisp-format.
elisp-format
no-elisp-format
Likewise for Emacs Lisp, see elisp-format.
librep-format
no-librep-format
Likewise for librep, see librep-format.
scheme-format
no-scheme-format
Likewise for Scheme, see scheme-format.
smalltalk-format
no-smalltalk-format
Likewise for Smalltalk, see smalltalk-format.
java-format
no-java-format
Likewise for Java, see java-format.
csharp-format
no-csharp-format
Likewise for C#, see csharp-format.
awk-format
no-awk-format
Likewise for awk, see awk-format.
object-pascal-format
no-object-pascal-format
Likewise for Object Pascal, see object-pascal-format.
ycp-format
no-ycp-format
Likewise for YCP, see ycp-format.
tcl-format
no-tcl-format
Likewise for Tcl, see tcl-format.
perl-format
no-perl-format
Likewise for Perl, see perl-format.
perl-brace-format
no-perl-brace-format
Likewise for Perl brace, see perl-format.
php-format
no-php-format
Likewise for PHP, see php-format.
gcc-internal-format
no-gcc-internal-format
Likewise for the GCC sources, see gcc-internal-format.
qt-format
no-qt-format
Likewise for Qt, see qt-format.
kde-format
no-kde-format
Likewise for KDE, see kde-format.
boost-format
no-boost-format
Likewise for Boost, see boost-format.

It is also possible to have entries with a context specifier. They look like this:

     white-space
#: reference...
#, flag...
#| msgctxt previous-context
#| msgid previous-untranslated-string
msgctxt context
msgid untranslated-string
msgstr translated-string


The context serves to disambiguate messages with the same untranslated-string. It is possible to have several entries with the same untranslated-string in a PO file, provided that they each have a different context. Note that an empty context string and an absent msgctxt line do not mean the same thing.

A different kind of entries is used for translations which involve plural forms.

     white-space
#: reference...
#, flag...
#| msgid previous-untranslated-string-singular
#| msgid_plural previous-untranslated-string-plural
msgid untranslated-string-singular
msgid_plural untranslated-string-plural
msgstr[0] translated-string-case-0
...
msgstr[N] translated-string-case-n


Such an entry can look like this:

     #: src/msgcmp.c:338 src/po-lex.c:699
#, c-format
msgid "found %d fatal error"
msgid_plural "found %d fatal errors"
msgstr[0] "s'ha trobat %d error fatal"
msgstr[1] "s'han trobat %d errors fatals"


Here also, a msgctxt context can be specified before msgid, like above.

The previous-untranslated-string is optionally inserted by the msgmerge program, at the same time when it marks a message fuzzy. It helps the translator to see which changes were done by the developers on the untranslated-string.

It happens that some lines, usually whitespace or comments, follow the very last entry of a PO file. Such lines are not part of any entry, and will be dropped when the PO file is processed by the tools, or may disturb some PO file editors.

The remainder of this section may be safely skipped by those using a PO file editor, yet it may be interesting for everybody to have a better idea of the precise format of a PO file. On the other hand, those wishing to modify PO files by hand should carefully continue reading on.

Each of untranslated-string and translated-string respects the C syntax for a character string, including the surrounding quotes and embedded backslashed escape sequences. When the time comes to write multi-line strings, one should not use escaped newlines. Instead, a closing quote should follow the last character on the line to be continued, and an opening quote should resume the string at the beginning of the following PO file line. For example:

     msgid ""
"Here is an example of how one might continue a very long string\n"
"for the common case the string represents multi-line output.\n"


In this example, the empty string is used on the first line, to allow better alignment of the H from the word ‘Here’ over the f from the word ‘for’. In this example, the msgid keyword is followed by three strings, which are meant to be concatenated. Concatenating the empty string does not change the resulting overall string, but it is a way for us to comply with the necessity of msgid to be followed by a string on the same line, while keeping the multi-line presentation left-justified, as we find this to be a cleaner disposition. The empty string could have been omitted, but only if the string starting with ‘Here’ was promoted on the first line, right after msgid.2 It was not really necessary either to switch between the two last quoted strings immediately after the newline ‘\n’, the switch could have occurred after any other character, we just did it this way because it is neater.

One should carefully distinguish between end of lines marked as ‘\ninside quotes, which are part of the represented string, and end of lines in the PO file itself, outside string quotes, which have no incidence on the represented string.

Outside strings, white lines and comments may be used freely. Comments start at the beginning of a line with ‘#’ and extend until the end of the PO file line. Comments written by translators should have the initial ‘#’ immediately followed by some white space. If the ‘#’ is not immediately followed by white space, this comment is most likely generated and managed by specialized GNU tools, and might disappear or be replaced unexpectedly when the PO file is given to msgmerge.

Next: , Previous: PO Files, Up: Top

## 4 Preparing Program Sources

For the programmer, changes to the C source code fall into three categories. First, you have to make the localization functions known to all modules needing message translation. Second, you should properly trigger the operation of GNU gettext when the program initializes, usually from the main function. Last, you should identify, adjust and mark all constant strings in your program needing translation.

Next: , Previous: Sources, Up: Sources

### 4.1 Importing the gettext declaration

Presuming that your set of programs, or package, has been adjusted so all needed GNU gettext files are available, and your Makefile files are adjusted (see Maintainers), each C module having translated C strings should contain the line:

     #include <libintl.h>


Similarly, each C module containing printf()/fprintf()/... calls with a format string that could be a translated C string (even if the C string comes from a different C module) should contain the line:

     #include <libintl.h>


Next: , Previous: Importing, Up: Sources

### 4.2 Triggering gettext Operations

The initialization of locale data should be done with more or less the same code in every program, as demonstrated below:

     int
main (int argc, char *argv[])
{
...
setlocale (LC_ALL, "");
bindtextdomain (PACKAGE, LOCALEDIR);
textdomain (PACKAGE);
...
}


PACKAGE and LOCALEDIR should be provided either by config.h or by the Makefile. For now consult the gettext or hello sources for more information.

The use of LC_ALL might not be appropriate for you. LC_ALL includes all locale categories and especially LC_CTYPE. This latter category is responsible for determining character classes with the isalnum etc. functions from ctype.h which could especially for programs, which process some kind of input language, be wrong. For example this would mean that a source code using the ç (c-cedilla character) is runnable in France but not in the U.S.

Some systems also have problems with parsing numbers using the scanf functions if an other but the LC_ALL locale category is used. The standards say that additional formats but the one known in the "C" locale might be recognized. But some systems seem to reject numbers in the "C" locale format. In some situation, it might also be a problem with the notation itself which makes it impossible to recognize whether the number is in the "C" locale or the local format. This can happen if thousands separator characters are used. Some locales define this character according to the national conventions to '.' which is the same character used in the "C" locale to denote the decimal point.

So it is sometimes necessary to replace the LC_ALL line in the code above by a sequence of setlocale lines

     {
...
setlocale (LC_CTYPE, "");
setlocale (LC_MESSAGES, "");
...
}


On all POSIX conformant systems the locale categories LC_CTYPE, LC_MESSAGES, LC_COLLATE, LC_MONETARY, LC_NUMERIC, and LC_TIME are available. On some systems which are only ISO C compliant, LC_MESSAGES is missing, but a substitute for it is defined in GNU gettext's <libintl.h> and in GNU gnulib's <locale.h>.

Note that changing the LC_CTYPE also affects the functions declared in the <ctype.h> standard header and some functions declared in the <string.h> and <stdlib.h> standard headers. If this is not desirable in your application (for example in a compiler's parser), you can use a set of substitute functions which hardwire the C locale, such as found in the modules ‘c-ctype’, ‘c-strcase’, ‘c-strcasestr’, ‘c-strtod’, ‘c-strtold’ in the GNU gnulib source distribution.

It is also possible to switch the locale forth and back between the environment dependent locale and the C locale, but this approach is normally avoided because a setlocale call is expensive, because it is tedious to determine the places where a locale switch is needed in a large program's source, and because switching a locale is not multithread-safe.

Next: , Previous: Triggering, Up: Sources

### 4.3 Preparing Translatable Strings

Before strings can be marked for translations, they sometimes need to be adjusted. Usually preparing a string for translation is done right before marking it, during the marking phase which is described in the next sections. What you have to keep in mind while doing that is the following.

• Decent English style.
• Entire sentences.
• Split at paragraphs.
• Use format strings instead of string concatenation.
• Avoid unusual markup and unusual control characters.

Let's look at some examples of these guidelines.

Translatable strings should be in good English style. If slang language with abbreviations and shortcuts is used, often translators will not understand the message and will produce very inappropriate translations.

     "%s: is parameter\n"


This is nearly untranslatable: Is the displayed item a parameter or the parameter?

     "No match"


The ambiguity in this message makes it unintelligible: Is the program attempting to set something on fire? Does it mean "The given object does not match the template"? Does it mean "The template does not fit for any of the objects"?

In both cases, adding more words to the message will help both the translator and the English speaking user.

Translatable strings should be entire sentences. It is often not possible to translate single verbs or adjectives in a substitutable way.

     printf ("File %s is %s protected", filename, rw ? "write" : "read");


Most translators will not look at the source and will thus only see the string "File %s is %s protected", which is unintelligible. Change this to

     printf (rw ? "File %s is write protected" : "File %s is read protected",
filename);


This way the translator will not only understand the message, she will also be able to find the appropriate grammatical construction. A French translator for example translates "write protected" like "protected against writing".

Entire sentences are also important because in many languages, the declination of some word in a sentence depends on the gender or the number (singular/plural) of another part of the sentence. There are usually more interdependencies between words than in English. The consequence is that asking a translator to translate two half-sentences and then combining these two half-sentences through dumb string concatenation will not work, for many languages, even though it would work for English. That's why translators need to handle entire sentences.

Often sentences don't fit into a single line. If a sentence is output using two subsequent printf statements, like this

     printf ("Locale charset \"%s\" is different from\n", lcharset);
printf ("input file charset \"%s\".\n", fcharset);


the translator would have to translate two half sentences, but nothing in the POT file would tell her that the two half sentences belong together. It is necessary to merge the two printf statements so that the translator can handle the entire sentence at once and decide at which place to insert a line break in the translation (if at all):

     printf ("Locale charset \"%s\" is different from\n\
input file charset \"%s\".\n", lcharset, fcharset);


     puts ("Apollo 13 scenario: Stack overflow handling failed.");
puts ("On the next stack overflow we will crash!!!");


Should these two statements merged into a single one? I would recommend to merge them if the two sentences are related to each other, because then it makes it easier for the translator to understand and translate both. On the other hand, if one of the two messages is a stereotypic one, occurring in other places as well, you will do a favour to the translator by not merging the two. (Identical messages occurring in several places are combined by xgettext, so the translator has to handle them once only.)

Translatable strings should be limited to one paragraph; don't let a single message be longer than ten lines. The reason is that when the translatable string changes, the translator is faced with the task of updating the entire translated string. Maybe only a single word will have changed in the English string, but the translator doesn't see that (with the current translation tools), therefore she has to proofread the entire message.

Many GNU programs have a ‘--help’ output that extends over several screen pages. It is a courtesy towards the translators to split such a message into several ones of five to ten lines each. While doing that, you can also attempt to split the documented options into groups, such as the input options, the output options, and the informative output options. This will help every user to find the option he is looking for.

Hardcoded string concatenation is sometimes used to construct English strings:

     strcpy (s, "Replace ");
strcat (s, object1);
strcat (s, " with ");
strcat (s, object2);
strcat (s, "?");


In order to present to the translator only entire sentences, and also because in some languages the translator might want to swap the order of object1 and object2, it is necessary to change this to use a format string:

     sprintf (s, "Replace %s with %s?", object1, object2);


A similar case is compile time concatenation of strings. The ISO C 99 include file <inttypes.h> contains a macro PRId64 that can be used as a formatting directive for outputting an ‘int64_t’ integer through printf. It expands to a constant string, usually "d" or "ld" or "lld" or something like this, depending on the platform. Assume you have code like

     printf ("The amount is %0" PRId64 "\n", number);


The gettext tools and library have special support for these <inttypes.h> macros. You can therefore simply write

     printf (gettext ("The amount is %0" PRId64 "\n"), number);


The PO file will contain the string "The amount is %0<PRId64>\n". The translators will provide a translation containing "%0<PRId64>" as well, and at runtime the gettext function's result will contain the appropriate constant string, "d" or "ld" or "lld".

This works only for the predefined <inttypes.h> macros. If you have defined your own similar macros, let's say ‘MYPRId64’, that are not known to xgettext, the solution for this problem is to change the code like this:

     char buf1[100];
sprintf (buf1, "%0" MYPRId64, number);
printf (gettext ("The amount is %s\n"), buf1);


This means, you put the platform dependent code in one statement, and the internationalization code in a different statement. Note that a buffer length of 100 is safe, because all available hardware integer types are limited to 128 bits, and to print a 128 bit integer one needs at most 54 characters, regardless whether in decimal, octal or hexadecimal.

All this applies to other programming languages as well. For example, in Java and C#, string concatenation is very frequently used, because it is a compiler built-in operator. Like in C, in Java, you would change

     System.out.println("Replace "+object1+" with "+object2+"?");


into a statement involving a format string:

     System.out.println(
MessageFormat.format("Replace {0} with {1}?",
new Object[] { object1, object2 }));


Similarly, in C#, you would change

     Console.WriteLine("Replace "+object1+" with "+object2+"?");


into a statement involving a format string:

     Console.WriteLine(
String.Format("Replace {0} with {1}?", object1, object2));


Unusual markup or control characters should not be used in translatable strings. Translators will likely not understand the particular meaning of the markup or control characters.

For example, if you have a convention that ‘|’ delimits the left-hand and right-hand part of some GUI elements, translators will often not understand it without specific comments. It might be better to have the translator translate the left-hand and right-hand part separately.

Another example is the ‘argp’ convention to use a single ‘\v’ (vertical tab) control character to delimit two sections inside a string. This is flawed. Some translators may convert it to a simple newline, some to blank lines. With some PO file editors it may not be easy to even enter a vertical tab control character. So, you cannot be sure that the translation will contain a ‘\v’ character, at the corresponding position. The solution is, again, to let the translator translate two separate strings and combine at run-time the two translated strings with the ‘\v’ required by the convention.

HTML markup, however, is common enough that it's probably ok to use in translatable strings. But please bear in mind that the GNU gettext tools don't verify that the translations are well-formed HTML.

Next: , Previous: Preparing Strings, Up: Sources

### 4.4 How Marks Appear in Sources

All strings requiring translation should be marked in the C sources. Marking is done in such a way that each translatable string appears to be the sole argument of some function or preprocessor macro. There are only a few such possible functions or macros meant for translation, and their names are said to be marking keywords. The marking is attached to strings themselves, rather than to what we do with them. This approach has more uses. A blatant example is an error message produced by formatting. The format string needs translation, as well as some strings inserted through some ‘%s’ specification in the format, while the result from sprintf may have so many different instances that it is impractical to list them all in some ‘error_string_out()’ routine, say.

This marking operation has two goals. The first goal of marking is for triggering the retrieval of the translation, at run time. The keyword is possibly resolved into a routine able to dynamically return the proper translation, as far as possible or wanted, for the argument string. Most localizable strings are found in executable positions, that is, attached to variables or given as parameters to functions. But this is not universal usage, and some translatable strings appear in structured initializations. See Special cases.

The second goal of the marking operation is to help xgettext at properly extracting all translatable strings when it scans a set of program sources and produces PO file templates.

The canonical keyword for marking translatable strings is ‘gettext’, it gave its name to the whole GNU gettext package. For packages making only light use of the ‘gettext’ keyword, macro or function, it is easily used as is. However, for packages using the gettext interface more heavily, it is usually more convenient to give the main keyword a shorter, less obtrusive name. Indeed, the keyword might appear on a lot of strings all over the package, and programmers usually do not want nor need their program sources to remind them forcefully, all the time, that they are internationalized. Further, a long keyword has the disadvantage of using more horizontal space, forcing more indentation work on sources for those trying to keep them within 79 or 80 columns.

Many packages use ‘_’ (a simple underline) as a keyword, and write ‘_("Translatable string")’ instead of ‘gettext ("Translatable string")’. Further, the coding rule, from GNU standards, wanting that there is a space between the keyword and the opening parenthesis is relaxed, in practice, for this particular usage. So, the textual overhead per translatable string is reduced to only three characters: the underline and the two parentheses. However, even if GNU gettext uses this convention internally, it does not offer it officially. The real, genuine keyword is truly ‘gettext’ indeed. It is fairly easy for those wanting to use ‘_’ instead of ‘gettext’ to declare:

     #include <libintl.h>
#define _(String) gettext (String)


instead of merely using ‘#include <libintl.h>’.

The marking keywords ‘gettext’ and ‘_’ take the translatable string as sole argument. It is also possible to define marking functions that take it at another argument position. It is even possible to make the marked argument position depend on the total number of arguments of the function call; this is useful in C++. All this is achieved using xgettext's ‘--keyword’ option.

Note also that long strings can be split across lines, into multiple adjacent string tokens. Automatic string concatenation is performed at compile time according to ISO C and ISO C++; xgettext also supports this syntax.

Later on, the maintenance is relatively easy. If, as a programmer, you add or modify a string, you will have to ask yourself if the new or altered string requires translation, and include it within ‘_()’ if you think it should be translated. For example, ‘"%s"’ is an example of string not requiring translation. But ‘"%s: %d"does require translation, because in French, unlike in English, it's customary to put a space before a colon.

Next: , Previous: Mark Keywords, Up: Sources

### 4.5 Marking Translatable Strings

In PO mode, one set of features is meant more for the programmer than for the translator, and allows him to interactively mark which strings, in a set of program sources, are translatable, and which are not. Even if it is a fairly easy job for a programmer to find and mark such strings by other means, using any editor of his choice, PO mode makes this work more comfortable. Further, this gives translators who feel a little like programmers, or programmers who feel a little like translators, a tool letting them work at marking translatable strings in the program sources, while simultaneously producing a set of translation in some language, for the package being internationalized.

The set of program sources, targeted by the PO mode commands describe here, should have an Emacs tags table constructed for your project, prior to using these PO file commands. This is easy to do. In any shell window, change the directory to the root of your project, then execute a command resembling:

     etags src/*.[hc] lib/*.[hc]


presuming here you want to process all .h and .c files from the src/ and lib/ directories. This command will explore all said files and create a TAGS file in your root directory, somewhat summarizing the contents using a special file format Emacs can understand.

For packages following the GNU coding standards, there is a make goal tags or TAGS which constructs the tag files in all directories and for all files containing source code.

Once your TAGS file is ready, the following commands assist the programmer at marking translatable strings in his set of sources. But these commands are necessarily driven from within a PO file window, and it is likely that you do not even have such a PO file yet. This is not a problem at all, as you may safely open a new, empty PO file, mainly for using these commands. This empty PO file will slowly fill in while you mark strings as translatable in your program sources.

,
Search through program sources for a string which looks like a candidate for translation (po-tags-search).
M-,
Mark the last string found with ‘_()’ (po-mark-translatable).
M-.
Mark the last string found with a keyword taken from a set of possible keywords. This command with a prefix allows some management of these keywords (po-select-mark-and-mark).

The , (po-tags-search) command searches for the next occurrence of a string which looks like a possible candidate for translation, and displays the program source in another Emacs window, positioned in such a way that the string is near the top of this other window. If the string is too big to fit whole in this window, it is positioned so only its end is shown. In any case, the cursor is left in the PO file window. If the shown string would be better presented differently in different native languages, you may mark it using M-, or M-.. Otherwise, you might rather ignore it and skip to the next string by merely repeating the , command.

A string is a good candidate for translation if it contains a sequence of three or more letters. A string containing at most two letters in a row will be considered as a candidate if it has more letters than non-letters. The command disregards strings containing no letters, or isolated letters only. It also disregards strings within comments, or strings already marked with some keyword PO mode knows (see below).

If you have never told Emacs about some TAGS file to use, the command will request that you specify one from the minibuffer, the first time you use the command. You may later change your TAGS file by using the regular Emacs command M-x visit-tags-table, which will ask you to name the precise TAGS file you want to use. See Tag Tables.

Each time you use the , command, the search resumes from where it was left by the previous search, and goes through all program sources, obeying the TAGS file, until all sources have been processed. However, by giving a prefix argument to the command (C-u ,), you may request that the search be restarted all over again from the first program source; but in this case, strings that you recently marked as translatable will be automatically skipped.

Using this , command does not prevent using of other regular Emacs tags commands. For example, regular tags-search or tags-query-replace commands may be used without disrupting the independent , search sequence. However, as implemented, the initial , command (or the , command is used with a prefix) might also reinitialize the regular Emacs tags searching to the first tags file, this reinitialization might be considered spurious.

The M-, (po-mark-translatable) command will mark the recently found string with the ‘_’ keyword. The M-. (po-select-mark-and-mark) command will request that you type one keyword from the minibuffer and use that keyword for marking the string. Both commands will automatically create a new PO file untranslated entry for the string being marked, and make it the current entry (making it easy for you to immediately proceed to its translation, if you feel like doing it right away). It is possible that the modifications made to the program source by M-, or M-. render some source line longer than 80 columns, forcing you to break and re-indent this line differently. You may use the O command from PO mode, or any other window changing command from Emacs, to break out into the program source window, and do any needed adjustments. You will have to use some regular Emacs command to return the cursor to the PO file window, if you want command , for the next string, say.

The M-. command has a few built-in speedups, so you do not have to explicitly type all keywords all the time. The first such speedup is that you are presented with a preferred keyword, which you may accept by merely typing <RET> at the prompt. The second speedup is that you may type any non-ambiguous prefix of the keyword you really mean, and the command will complete it automatically for you. This also means that PO mode has to know all your possible keywords, and that it will not accept mistyped keywords.

If you reply ? to the keyword request, the command gives a list of all known keywords, from which you may choose. When the command is prefixed by an argument (C-u M-.), it inhibits updating any program source or PO file buffer, and does some simple keyword management instead. In this case, the command asks for a keyword, written in full, which becomes a new allowed keyword for later M-. commands. Moreover, this new keyword automatically becomes the preferred keyword for later commands. By typing an already known keyword in response to C-u M-., one merely changes the preferred keyword and does nothing more.

All keywords known for M-. are recognized by the , command when scanning for strings, and strings already marked by any of those known keywords are automatically skipped. If many PO files are opened simultaneously, each one has its own independent set of known keywords. There is no provision in PO mode, currently, for deleting a known keyword, you have to quit the file (maybe using q) and reopen it afresh. When a PO file is newly brought up in an Emacs window, only ‘gettext’ and ‘_’ are known as keywords, and ‘gettext’ is preferred for the M-. command. In fact, this is not useful to prefer ‘_’, as this one is already built in the M-, command.

Next: , Previous: Marking, Up: Sources

### 4.6 Special Comments preceding Keywords

In C programs strings are often used within calls of functions from the printf family. The special thing about these format strings is that they can contain format specifiers introduced with %. Assume we have the code

     printf (gettext ("String %s' has %d characters\n"), s, strlen (s));


A possible German translation for the above string might be:

     "%d Zeichen lang ist die Zeichenkette %s'"


A C programmer, even if he cannot speak German, will recognize that there is something wrong here. The order of the two format specifiers is changed but of course the arguments in the printf don't have. This will most probably lead to problems because now the length of the string is regarded as the address.

To prevent errors at runtime caused by translations the msgfmt tool can check statically whether the arguments in the original and the translation string match in type and number. If this is not the case and the ‘-c’ option has been passed to msgfmt, msgfmt will give an error and refuse to produce a MO file. Thus consequent use of ‘msgfmt -c’ will catch the error, so that it cannot cause cause problems at runtime.

If the word order in the above German translation would be correct one would have to write

     "%2$d Zeichen lang ist die Zeichenkette %1$s'"


The routines in msgfmt know about this special notation.

Because not all strings in a program must be format strings it is not useful for msgfmt to test all the strings in the .po file. This might cause problems because the string might contain what looks like a format specifier, but the string is not used in printf.

Therefore the xgettext adds a special tag to those messages it thinks might be a format string. There is no absolute rule for this, only a heuristic. In the .po file the entry is marked using the c-format flag in the #, comment line (see PO Files).

The careful reader now might say that this again can cause problems. The heuristic might guess it wrong. This is true and therefore xgettext knows about a special kind of comment which lets the programmer take over the decision. If in the same line as or the immediately preceding line to the gettext keyword the xgettext program finds a comment containing the words xgettext:c-format, it will mark the string in any case with the c-format flag. This kind of comment should be used when xgettext does not recognize the string as a format string but it really is one and it should be tested. Please note that when the comment is in the same line as the gettext keyword, it must be before the string to be translated.

This situation happens quite often. The printf function is often called with strings which do not contain a format specifier. Of course one would normally use fputs but it does happen. In this case xgettext does not recognize this as a format string but what happens if the translation introduces a valid format specifier? The printf function will try to access one of the parameters but none exists because the original code does not pass any parameters.

xgettext of course could make a wrong decision the other way round, i.e. a string marked as a format string actually is not a format string. In this case the msgfmt might give too many warnings and would prevent translating the .po file. The method to prevent this wrong decision is similar to the one used above, only the comment to use must contain the string xgettext:no-c-format.

If a string is marked with c-format and this is not correct the user can find out who is responsible for the decision. See xgettext Invocation to see how the --debug option can be used for solving this problem.

Next: , Previous: c-format Flag, Up: Sources

### 4.7 Special Cases of Translatable Strings

The attentive reader might now point out that it is not always possible to mark translatable string with gettext or something like this. Consider the following case:

     {
static const char *messages[] = {
"some very meaningful message",
"and another one"
};
const char *string;
...
string
= index > 1 ? "a default message" : messages[index];

fputs (string);
...
}


While it is no problem to mark the string "a default message" it is not possible to mark the string initializers for messages. What is to be done? We have to fulfill two tasks. First we have to mark the strings so that the xgettext program (see xgettext Invocation) can find them, and second we have to translate the string at runtime before printing them.

The first task can be fulfilled by creating a new keyword, which names a no-op. For the second we have to mark all access points to a string from the array. So one solution can look like this:

     #define gettext_noop(String) String

{
static const char *messages[] = {
gettext_noop ("some very meaningful message"),
gettext_noop ("and another one")
};
const char *string;
...
string
= index > 1 ? gettext ("a default message") : gettext (messages[index]);

fputs (string);
...
}


Please convince yourself that the string which is written by fputs is translated in any case. How to get xgettext know the additional keyword gettext_noop is explained in xgettext Invocation.

The above is of course not the only solution. You could also come along with the following one:

     #define gettext_noop(String) String

{
static const char *messages[] = {
gettext_noop ("some very meaningful message",
gettext_noop ("and another one")
};
const char *string;
...
string
= index > 1 ? gettext_noop ("a default message") : messages[index];

fputs (gettext (string));
...
}


But this has a drawback. The programmer has to take care that he uses gettext_noop for the string "a default message". A use of gettext could have in rare cases unpredictable results.

One advantage is that you need not make control flow analysis to make sure the output is really translated in any case. But this analysis is generally not very difficult. If it should be in any situation you can use this second method in this situation.

Next: , Previous: Special cases, Up: Sources

### 4.8 Letting Users Report Translation Bugs

Code sometimes has bugs, but translations sometimes have bugs too. The users need to be able to report them. Reporting translation bugs to the programmer or maintainer of a package is not very useful, since the maintainer must never change a translation, except on behalf of the translator. Hence the translation bugs must be reported to the translators.

Here is a way to organize this so that the maintainer does not need to forward translation bug reports, nor even keep a list of the addresses of the translators or their translation teams.

Every program has a place where is shows the bug report address. For GNU programs, it is the code which handles the “–help” option, typically in a function called “usage”. In this place, instruct the translator to add her own bug reporting address. For example, if that code has a statement

     printf (_("Report bugs to <%s>.\n"), PACKAGE_BUGREPORT);


you can add some translator instructions like this:

     /* TRANSLATORS: The placeholder indicates the bug-reporting address
"Report translation bugs to <...>\n" with the address for translation
printf (_("Report bugs to <%s>.\n"), PACKAGE_BUGREPORT);


These will be extracted by ‘xgettext’, leading to a .pot file that contains this:

     #. TRANSLATORS: The placeholder indicates the bug-reporting address
#. "Report translation bugs to <...>\n" with the address for translation
#: src/hello.c:178
#, c-format
msgid "Report bugs to <%s>.\n"
msgstr ""


### 4.9 Marking Proper Names for Translation

Should names of persons, cities, locations etc. be marked for translation or not? People who only know languages that can be written with Latin letters (English, Spanish, French, German, etc.) are tempted to say “no”, because names usually do not change when transported between these languages. However, in general when translating from one script to another, names are translated too, usually phonetically or by transliteration. For example, Russian or Greek names are converted to the Latin alphabet when being translated to English, and English or French names are converted to the Katakana script when being translated to Japanese. This is necessary because the speakers of the target language in general cannot read the script the name is originally written in.

As a programmer, you should therefore make sure that names are marked for translation, with a special comment telling the translators that it is a proper name and how to pronounce it. Like this:

     printf (_("Written by %s.\n"),
/* TRANSLATORS: This is a proper name.  See the gettext
manual, section Names.  Note this is actually a non-ASCII
name: The first name is (with Unicode escapes)
"Fran\u00e7ois" or (with HTML entities) "Fran&ccedil;ois".
Pronunciation is like "fraa-swa pee-nar".  */
_("Francois Pinard"));


As a translator, you should use some care when translating names, because it is frustrating if people see their names mutilated or distorted. If your language uses the Latin script, all you need to do is to reproduce the name as perfectly as you can within the usual character set of your language. In this particular case, this means to provide a translation containing the c-cedilla character. If your language uses a different script and the people speaking it don't usually read Latin words, it means transliteration; but you should still give, in parentheses, the original writing of the name – for the sake of the people that do read the Latin script. Here is an example, using Greek as the target script:

     #. This is a proper name.  See the gettext
#. manual, section Names.  Note this is actually a non-ASCII
#. name: The first name is (with Unicode escapes)
#. "Fran\u00e7ois" or (with HTML entities) "Fran&ccedil;ois".
#. Pronunciation is like "fraa-swa pee-nar".
msgid "Francois Pinard"
msgstr "\phi\rho\alpha\sigma\omicron\alpha \pi\iota\nu\alpha\rho"
" (Francois Pinard)"


Because translation of names is such a sensitive domain, it is a good idea to test your translation before submitting it.

The translation project http://sourceforge.net/projects/translation has set up a POT file and translation domain consisting of program author names, with better facilities for the translator than those presented here. Namely, there the original name is written directly in Unicode (rather than with Unicode escapes or HTML entities), and the pronunciation is denoted using the International Phonetic Alphabet (see http://www.wikipedia.org/wiki/International_Phonetic_Alphabet).

However, we don't recommend this approach for all POT files in all packages, because this would force translators to use PO files in UTF-8 encoding, which is - in the current state of software (as of 2003) - a major hassle for translators using GNU Emacs or XEmacs with po-mode.

Previous: Names, Up: Sources

### 4.10 Preparing Library Sources

When you are preparing a library, not a program, for the use of gettext, only a few details are different. Here we assume that the library has a translation domain and a POT file of its own. (If it uses the translation domain and POT file of the main program, then the previous sections apply without changes.)

1. The library code doesn't call setlocale (LC_ALL, ""). It's the responsibility of the main program to set the locale. The library's documentation should mention this fact, so that developers of programs using the library are aware of it.
2. The library code doesn't call textdomain (PACKAGE), because it would interfere with the text domain set by the main program.
3. The initialization code for a program was
            setlocale (LC_ALL, "");
bindtextdomain (PACKAGE, LOCALEDIR);
textdomain (PACKAGE);


For a library it is reduced to

            bindtextdomain (PACKAGE, LOCALEDIR);


If your library's API doesn't already have an initialization function, you need to create one, containing at least the bindtextdomain invocation. However, you usually don't need to export and document this initialization function: It is sufficient that all entry points of the library call the initialization function if it hasn't been called before. The typical idiom used to achieve this is a static boolean variable that indicates whether the initialization function has been called. Like this:

          static bool libfoo_initialized;

static void
libfoo_initialize (void)
{
bindtextdomain (PACKAGE, LOCALEDIR);
libfoo_initialized = true;
}

/* This function is part of the exported API.  */
struct foo *
create_foo (...)
{
/* Must ensure the initialization is performed.  */
if (!libfoo_initialized)
libfoo_initialize ();
...
}

/* This function is part of the exported API.  The argument must be
non-NULL and have been created through create_foo().  */
int
foo_refcount (struct foo *argument)
{
/* No need to invoke the initialization function here, because
create_foo() must already have been called before.  */
...
}

4. The usual declaration of the ‘_’ macro in each source file was
          #include <libintl.h>
#define _(String) gettext (String)


for a program. For a library, which has its own translation domain, it reads like this:

          #include <libintl.h>
#define _(String) dgettext (PACKAGE, String)


In other words, dgettext is used instead of gettext. Similarly, the dngettext function should be used in place of the ngettext function.

Next: , Previous: Sources, Up: Top

## 5 Making the PO Template File

After preparing the sources, the programmer creates a PO template file. This section explains how to use xgettext for this purpose.

xgettext creates a file named domainname.po. You should then rename it to domainname.pot. (Why doesn't xgettext create it under the name domainname.pot right away? The answer is: for historical reasons. When xgettext was specified, the distinction between a PO file and PO file template was fuzzy, and the suffix ‘.pot’ wasn't in use at that time.)

Previous: Template, Up: Template

### 5.1 Invoking the xgettext Program

     xgettext [option] [inputfile] ...


The xgettext program extracts translatable strings from given input files.

#### 5.1.1 Input file location

inputfile ...
Input files.
-f file
--files-from=file
Read the names of the input files from file instead of getting them from the command line.
-D directory
--directory=directory
Add directory to the list of directories. Source files are searched relative to this list of directories. The resulting .po file will be written relative to the current directory, though.

If inputfile is ‘-’, standard input is read.

#### 5.1.2 Output file location

-d name
--default-domain=name
Use name.po for output (instead of messages.po).
-o file
--output=file
Write output to specified file (instead of name.po or messages.po).
-p dir
--output-dir=dir
Output files will be placed in directory dir.

If the output file is ‘-’ or ‘/dev/stdout’, the output is written to standard output.

#### 5.1.3 Choice of input file language

-L name
--language=name
Specifies the language of the input files. The supported languages are C, C++, ObjectiveC, PO, Python, Lisp, EmacsLisp, librep, Scheme, Smalltalk, Java, JavaProperties, C#, awk, YCP, Tcl, Perl, PHP, GCC-source, NXStringTable, RST, Glade.
-C
--c++
This is a shorthand for --language=C++.

By default the language is guessed depending on the input file name extension.

#### 5.1.4 Input file interpretation

--from-code=name
Specifies the encoding of the input files. This option is needed only if some untranslated message strings or their corresponding comments contain non-ASCII characters. Note that Tcl and Glade input files are always assumed to be in UTF-8, regardless of this option.

By default the input files are assumed to be in ASCII.

#### 5.1.5 Operation mode

-j
--join-existing
Join messages with existing file.
-x file
--exclude-file=file
Entries from file are not extracted. file should be a PO or POT file.
-c [tag]
Place comment block with tag (or those preceding keyword lines) in output file.

#### 5.1.6 Language specific options

-a
--extract-all
Extract all strings.

This option has an effect with most languages, namely C, C++, ObjectiveC, Shell, Python, Lisp, EmacsLisp, librep, Java, C#, awk, Tcl, Perl, PHP, GCC-source, Glade.

-k keywordspec
--keyword[=keywordspec]
Additional keyword to be looked for (without keywordspec means not to use default keywords).

If keywordspec is a C identifier id, xgettext looks for strings in the first argument of each call to the function or macro id. If keywordspec is of the form ‘id:argnum’, xgettext looks for strings in the argnumth argument of the call. If keywordspec is of the form ‘id:argnum1,argnum2’, xgettext looks for strings in the argnum1st argument and in the argnum2nd argument of the call, and treats them as singular/plural variants for a message with plural handling. Also, if keywordspec is of the form ‘id:contextargnumc,argnum’ or ‘id:argnum,contextargnumc’, xgettext treats strings in the contextargnumth argument as a context specifier. And, as a special-purpose support for GNOME, if keywordspec is of the form ‘id:argnumg’, xgettext recognizes the argnumth argument as a string with context, using the GNOME glib syntax ‘"msgctxt|msgid"’.
Furthermore, if keywordspec is of the form ‘id:...,totalnumargst’, xgettext recognizes this argument specification only if the number of actual arguments is equal to totalnumargs. This is useful for disambiguating overloaded function calls in C++.
Finally, if keywordspec is of the form ‘id:argnum...,"xcomment"’, xgettext, when extracting a message from the specified argument strings, adds an extracted comment xcomment to the message. Note that when used through a normal shell command line, the double-quotes around the xcomment need to be escaped.

This option has an effect with most languages, namely C, C++, ObjectiveC, Shell, Python, Lisp, EmacsLisp, librep, Java, C#, awk, Tcl, Perl, PHP, GCC-source, Glade.

The default keyword specifications, which are always looked for if not explicitly disabled, are language dependent. They are:

• For C, C++, and GCC-source: gettext, dgettext:2, dcgettext:2, ngettext:1,2, dngettext:2,3, dcngettext:2,3, gettext_noop, and pgettext:1c,2, dpgettext:2c,3, dcpgettext:2c,3, npgettext:1c,2,3, dnpgettext:2c,3,4, dcnpgettext:2c,3,4.
• For Objective C: Like for C, and also NSLocalizedString, _, NSLocalizedStaticString, __.
• For Shell scripts: gettext, ngettext:1,2, eval_gettext, eval_ngettext:1,2.
• For Python: gettext, ugettext, dgettext:2, ngettext:1,2, ungettext:1,2, dngettext:2,3, _.
• For Lisp: gettext, ngettext:1,2, gettext-noop.
• For EmacsLisp: _.
• For librep: _.
• For Scheme: gettext, ngettext:1,2, gettext-noop.
• For Java: GettextResource.gettext:2, GettextResource.ngettext:2,3, GettextResource.pgettext:2c,3, GettextResource.npgettext:2c,3,4, gettext, ngettext:1,2, pgettext:1c,2, npgettext:1c,2,3, getString.
• For C#: GetString, GetPluralString:1,2, GetParticularString:1c,2, GetParticularPluralString:1c,2,3.
• For awk: dcgettext, dcngettext:1,2.
• For Tcl: ::msgcat::mc.
• For Perl: gettext, %gettext, $gettext, dgettext:2, dcgettext:2, ngettext:1,2, dngettext:2,3, dcngettext:2,3, gettext_noop. • For PHP: _, gettext, dgettext:2, dcgettext:2, ngettext:1,2, dngettext:2,3, dcngettext:2,3. • For Glade 1: label, title, text, format, copyright, comments, preview_text, tooltip. To disable the default keyword specifications, the option ‘-k’ or ‘--keyword’ or ‘--keyword=’, without a keywordspec, can be used. --flag=word:arg:flag Specifies additional flags for strings occurring as part of the argth argument of the function word. The possible flags are the possible format string indicators, such as ‘c-format’, and their negations, such as ‘no-c-format’, possibly prefixed with ‘pass-’. The meaning of --flag=function:arg:lang-format is that in language lang, the specified function expects as argth argument a format string. (For those of you familiar with GCC function attributes, --flag=function:arg:c-format is roughly equivalent to the declaration ‘__attribute__ ((__format__ (__printf__, arg, ...)))’ attached to function in a C source file.) For example, if you use the ‘error’ function from GNU libc, you can specify its behaviour through --flag=error:3:c-format. The effect of this specification is that xgettext will mark as format strings all gettext invocations that occur as argth argument of function. This is useful when such strings contain no format string directives: together with the checks done by ‘msgfmt -c’ it will ensure that translators cannot accidentally use format string directives that would lead to a crash at runtime. The meaning of --flag=function:arg:pass-lang-format is that in language lang, if the function call occurs in a position that must yield a format string, then its argth argument must yield a format string of the same type as well. (If you know GCC function attributes, the --flag=function:arg:pass-c-format option is roughly equivalent to the declaration ‘__attribute__ ((__format_arg__ (arg)))’ attached to function in a C source file.) For example, if you use the ‘_’ shortcut for the gettext function, you should use --flag=_:1:pass-c-format. The effect of this specification is that xgettext will propagate a format string requirement for a _("string") call to its first argument, the literal "string", and thus mark it as a format string. This is useful when such strings contain no format string directives: together with the checks done by ‘msgfmt -c’ it will ensure that translators cannot accidentally use format string directives that would lead to a crash at runtime. This option has an effect with most languages, namely C, C++, ObjectiveC, Shell, Python, Lisp, EmacsLisp, librep, Scheme, Java, C#, awk, YCP, Tcl, Perl, PHP, GCC-source. -T --trigraphs Understand ANSI C trigraphs for input. This option has an effect only with the languages C, C++, ObjectiveC. --qt Recognize Qt format strings. This option has an effect only with the language C++. --kde Recognize KDE 4 format strings. This option has an effect only with the language C++. --boost Recognize Boost format strings. This option has an effect only with the language C++. --debug Use the flags c-format and possible-c-format to show who was responsible for marking a message as a format string. The latter form is used if the xgettext program decided, the format form is used if the programmer prescribed it. By default only the c-format form is used. The translator should not have to care about these details. This implementation of xgettext is able to process a few awkward cases, like strings in preprocessor macros, ANSI concatenation of adjacent strings, and escaped end of lines for continued strings. #### 5.1.7 Output details --force-po Always write an output file even if no message is defined. -i --indent Write the .po file using indented style. --no-location Do not write ‘#: filename:line’ lines. Note that using this option makes it harder for technically skilled translators to understand each message's context. -n --add-location Generate ‘#: filename:line’ lines (default). --strict Write out a strict Uniforum conforming PO file. Note that this Uniforum format should be avoided because it doesn't support the GNU extensions. --properties-output Write out a Java ResourceBundle in Java .properties syntax. Note that this file format doesn't support plural forms and silently drops obsolete messages. --stringtable-output Write out a NeXTstep/GNUstep localized resource file in .strings syntax. Note that this file format doesn't support plural forms. -w number --width=number Set the output page width. Long strings in the output files will be split across multiple lines in order to ensure that each line's width (= number of screen columns) is less or equal to the given number. --no-wrap Do not break long message lines. Message lines whose width exceeds the output page width will not be split into several lines. Only file reference lines which are wider than the output page width will be split. -s --sort-output Generate sorted output. Note that using this option makes it much harder for the translator to understand each message's context. -F --sort-by-file Sort output by file location. --omit-header Don't write header with ‘msgid ""’ entry. This is useful for testing purposes because it eliminates a source of variance for generated .gmo files. With --omit-header, two invocations of xgettext on the same files with the same options at different times are guaranteed to produce the same results. Note that using this option will lead to an error if the resulting file would not entirely be in ASCII. --copyright-holder=string Set the copyright holder in the output. string should be the copyright holder of the surrounding package. (Note that the msgstr strings, extracted from the package's sources, belong to the copyright holder of the package.) Translators are expected to transfer or disclaim the copyright for their translations, so that package maintainers can distribute them without legal risk. If string is empty, the output files are marked as being in the public domain; in this case, the translators are expected to disclaim their copyright, again so that package maintainers can distribute them without legal risk. The default value for string is the Free Software Foundation, Inc., simply because xgettext was first used in the GNU project. --foreign-user Omit FSF copyright in output. This option is equivalent to ‘--copyright-holder=''’. It can be useful for packages outside the GNU project that want their translations to be in the public domain. --package-name=package Set the package name in the header of the output. --package-version=version Set the package version in the header of the output. This option has an effect only if the ‘--package-name’ option is also used. --msgid-bugs-address=email@address Set the reporting address for msgid bugs. This is the email address or URL to which the translators shall report bugs in the untranslated strings: • Strings which are not entire sentences, see the maintainer guidelines in Preparing Strings. • Strings which use unclear terms or require additional context to be understood. • Strings which make invalid assumptions about notation of date, time or money. • Pluralisation problems. • Incorrect English spelling. • Incorrect formatting. It can be your email address, or a mailing list address where translators can write to without being subscribed, or the URL of a web page through which the translators can contact you. The default value is empty, which means that translators will be clueless! Don't forget to specify this option. -m [string] --msgstr-prefix[=string] Use string (or "" if not specified) as prefix for msgstr entries. -M [string] --msgstr-suffix[=string] Use string (or "" if not specified) as suffix for msgstr entries. #### 5.1.8 Informative output -h --help Display this help and exit. -V --version Output version information and exit. Next: , Previous: Template, Up: Top ## 6 Creating a New PO File When starting a new translation, the translator creates a file called LANG.po, as a copy of the package.pot template file with modifications in the initial comments (at the beginning of the file) and in the header entry (the first entry, near the beginning of the file). The easiest way to do so is by use of the ‘msginit’ program. For example: $ cd PACKAGE-VERSION
$cd po$ msginit


The alternative way is to do the copy and modifications by hand. To do so, the translator copies package.pot to LANG.po. Then she modifies the initial comments and the header entry of this file.

Next: , Previous: Creating, Up: Creating

### 6.1 Invoking the msginit Program

     msginit [option]


The msginit program creates a new PO file, initializing the meta information with values from the user's environment.

#### 6.1.1 Input file location

-i inputfile
--input=inputfile
Input POT file.

If no inputfile is given, the current directory is searched for the POT file. If it is ‘-’, standard input is read.

#### 6.1.2 Output file location

-o file
--output-file=file
Write output to specified PO file.

If no output file is given, it depends on the ‘--locale’ option or the user's locale setting. If it is ‘-’, the results are written to standard output.

#### 6.1.3 Input file syntax

-P
--properties-input
Assume the input file is a Java ResourceBundle in Java .properties syntax, not in PO file syntax.
--stringtable-input
Assume the input file is a NeXTstep/GNUstep localized resource file in .strings syntax, not in PO file syntax.

#### 6.1.4 Output details

-l ll_CC
--locale=ll_CC
Set target locale. ll should be a language code, and CC should be a country code. The command ‘locale -a’ can be used to output a list of all installed locales. The default is the user's locale setting.
--no-translator
Declares that the PO file will not have a human translator and is instead automatically generated.
-p
--properties-output
Write out a Java ResourceBundle in Java .properties syntax. Note that this file format doesn't support plural forms and silently drops obsolete messages.
--stringtable-output
Write out a NeXTstep/GNUstep localized resource file in .strings syntax. Note that this file format doesn't support plural forms.
-w number
--width=number
Set the output page width. Long strings in the output files will be split across multiple lines in order to ensure that each line's width (= number of screen columns) is less or equal to the given number.
--no-wrap
Do not break long message lines. Message lines whose width exceeds the output page width will not be split into several lines. Only file reference lines which are wider than the output page width will be split.

#### 6.1.5 Informative output

-h
--help
Display this help and exit.
-V
--version
Output version information and exit.

Previous: msginit Invocation, Up: Creating

### 6.2 Filling in the Header Entry

The initial comments "SOME DESCRIPTIVE TITLE", "YEAR" and "FIRST AUTHOR <EMAIL@ADDRESS>, YEAR" ought to be replaced by sensible information. This can be done in any text editor; if Emacs is used and it switched to PO mode automatically (because it has recognized the file's suffix), you can disable it by typing M-x fundamental-mode.

Modifying the header entry can already be done using PO mode: in Emacs, type M-x po-mode RET and then RET again to start editing the entry. You should fill in the following fields.

Project-Id-Version
This is the name and version of the package. Fill it in if it has not already been filled in by xgettext.
Report-Msgid-Bugs-To
This has already been filled in by xgettext. It contains an email address or URL where you can report bugs in the untranslated strings:
• Strings which are not entire sentences, see the maintainer guidelines in Preparing Strings.
• Strings which use unclear terms or require additional context to be understood.
• Strings which make invalid assumptions about notation of date, time or money.
• Pluralisation problems.
• Incorrect English spelling.
• Incorrect formatting.

POT-Creation-Date
This has already been filled in by xgettext.
PO-Revision-Date
You don't need to fill this in. It will be filled by the PO file editor when you save the file.
Last-Translator
Language-Team
Fill in the English name of the language, and the email address or homepage URL of the language team you are part of.

Before starting a translation, it is a good idea to get in touch with your translation team, not only to make sure you don't do duplicated work, but also to coordinate difficult linguistic issues.

In the Free Translation Project, each translation team has its own mailing list. The up-to-date list of teams can be found at the Free Translation Project's homepage, http://translationproject.org/, in the "Teams" area.

Content-Type
Replace ‘CHARSET’ with the character encoding used for your language, in your locale, or UTF-8. This field is needed for correct operation of the msgmerge and msgfmt programs, as well as for users whose locale's character encoding differs from yours (see Charset conversion).

You get the character encoding of your locale by running the shell command ‘locale charmap’. If the result is ‘C’ or ‘ANSI_X3.4-1968’, which is equivalent to ‘ASCII’ (= ‘US-ASCII’), it means that your locale is not correctly configured. In this case, ask your translation team which charset to use. ‘ASCII’ is not usable for any language except Latin.

Because the PO files must be portable to operating systems with less advanced internationalization facilities, the character encodings that can be used are limited to those supported by both GNU libc and GNU libiconv. These are: ASCII, ISO-8859-1, ISO-8859-2, ISO-8859-3, ISO-8859-4, ISO-8859-5, ISO-8859-6, ISO-8859-7, ISO-8859-8, ISO-8859-9, ISO-8859-13, ISO-8859-14, ISO-8859-15, KOI8-R, KOI8-U, KOI8-T, CP850, CP866, CP874, CP932, CP949, CP950, CP1250, CP1251, CP1252, CP1253, CP1254, CP1255, CP1256, CP1257, GB2312, EUC-JP, EUC-KR, EUC-TW, BIG5, BIG5-HKSCS, GBK, GB18030, SHIFT_JIS, JOHAB, TIS-620, VISCII, GEORGIAN-PS, UTF-8.

In the GNU system, the following encodings are frequently used for the corresponding languages.

• ISO-8859-1 for Afrikaans, Albanian, Basque, Breton, Catalan, Cornish, Danish, Dutch, English, Estonian, Faroese, Finnish, French, Galician, German, Greenlandic, Icelandic, Indonesian, Irish, Italian, Malay, Manx, Norwegian, Occitan, Portuguese, Spanish, Swedish, Tagalog, Uzbek, Walloon,
• ISO-8859-2 for Bosnian, Croatian, Czech, Hungarian, Polish, Romanian, Serbian, Slovak, Slovenian,
• ISO-8859-3 for Maltese,
• ISO-8859-5 for Macedonian, Serbian,
• ISO-8859-6 for Arabic,
• ISO-8859-7 for Greek,
• ISO-8859-8 for Hebrew,
• ISO-8859-9 for Turkish,
• ISO-8859-13 for Latvian, Lithuanian, Maori,
• ISO-8859-14 for Welsh,
• ISO-8859-15 for Basque, Catalan, Dutch, English, Finnish, French, Galician, German, Irish, Italian, Portuguese, Spanish, Swedish, Walloon,
• KOI8-R for Russian,
• KOI8-U for Ukrainian,
• KOI8-T for Tajik,
• CP1251 for Bulgarian, Byelorussian,
• GB2312, GBK, GB18030 for simplified writing of Chinese,
• BIG5, BIG5-HKSCS for traditional writing of Chinese,
• EUC-JP for Japanese,
• EUC-KR for Korean,
• TIS-620 for Thai,
• GEORGIAN-PS for Georgian,
• UTF-8 for any language, including those listed above.

When single quote characters or double quote characters are used in translations for your language, and your locale's encoding is one of the ISO-8859-* charsets, it is best if you create your PO files in UTF-8 encoding, instead of your locale's encoding. This is because in UTF-8 the real quote characters can be represented (single quote characters: U+2018, U+2019, double quote characters: U+201C, U+201D), whereas none of ISO-8859-* charsets has them all. Users in UTF-8 locales will see the real quote characters, whereas users in ISO-8859-* locales will see the vertical apostrophe and the vertical double quote instead (because that's what the character set conversion will transliterate them to).

To enter such quote characters under X11, you can change your keyboard mapping using the xmodmap program. The X11 names of the quote characters are "leftsinglequotemark", "rightsinglequotemark", "leftdoublequotemark", "rightdoublequotemark", "singlelowquotemark", "doublelowquotemark".

Note that only recent versions of GNU Emacs support the UTF-8 encoding: Emacs 20 with Mule-UCS, and Emacs 21. As of January 2001, XEmacs doesn't support the UTF-8 encoding.

The character encoding name can be written in either upper or lower case. Usually upper case is preferred.

Content-Transfer-Encoding
Set this to 8bit.
Plural-Forms
This field is optional. It is only needed if the PO file has plural forms. You can find them by searching for the ‘msgid_plural’ keyword. The format of the plural forms field is described in Plural forms.

Next: , Previous: Creating, Up: Top

## 7 Updating Existing PO Files

Previous: Updating, Up: Updating

### 7.1 Invoking the msgmerge Program

     msgmerge [option] def.po ref.pot


The msgmerge program merges two Uniforum style .po files together. The def.po file is an existing PO file with translations which will be taken over to the newly created file as long as they still match; comments will be preserved, but extracted comments and file positions will be discarded. The ref.pot file is the last created PO file with up-to-date source references but old translations, or a PO Template file (generally created by xgettext); any translations or comments in the file will be discarded, however dot comments and file positions will be preserved. Where an exact match cannot be found, fuzzy matching is used to produce better results.

#### 7.1.1 Input file location

def.po
Translations referring to old sources.
ref.pot
References to the new sources.
-D directory
--directory=directory
Add directory to the list of directories. Source files are searched relative to this list of directories. The resulting .po file will be written relative to the current directory, though.
-C file
--compendium=file
Specify an additional library of message translations. See Compendium. This option may be specified more than once.

#### 7.1.2 Operation mode

-U
--update
Update def.po. Do nothing if def.po is already up to date.

#### 7.1.3 Output file location

-o file
--output-file=file
Write output to specified file.

The results are written to standard output if no output file is specified or if it is ‘-’.

#### 7.1.4 Output file location in update mode

The result is written back to def.po.

--backup=control
Make a backup of def.po
--suffix=suffix
Override the usual backup suffix.

The version control method may be selected via the --backup option or through the VERSION_CONTROL environment variable. Here are the values:

none
off
Never make backups (even if --backup is given).
numbered
t
Make numbered backups.
existing
nil
Make numbered backups if numbered backups for this file already exist, otherwise make simple backups.
simple
never
Always make simple backups.

The backup suffix is ‘~’, unless set with --suffix or the SIMPLE_BACKUP_SUFFIX environment variable.

#### 7.1.5 Operation modifiers

-m
--multi-domain
Apply ref.pot to each of the domains in def.po.
-N
--no-fuzzy-matching
Do not use fuzzy matching when an exact match is not found. This may speed up the operation considerably.
--previous
Keep the previous msgids of translated messages, marked with ‘#|’, when adding the fuzzy marker to such messages.

#### 7.1.6 Input file syntax

-P
--properties-input
Assume the input files are Java ResourceBundles in Java .properties syntax, not in PO file syntax.
--stringtable-input
Assume the input files are NeXTstep/GNUstep localized resource files in .strings syntax, not in PO file syntax.

#### 7.1.7 Output details

--force-po
Always write an output file even if it contains no message.
-i
--indent
Write the .po file using indented style.
--no-location
Do not write ‘#: filename:line’ lines.
Generate ‘#: filename:line’ lines (default).
--strict
Write out a strict Uniforum conforming PO file. Note that this Uniforum format should be avoided because it doesn't support the GNU extensions.
-p
--properties-output
Write out a Java ResourceBundle in Java .properties syntax. Note that this file format doesn't support plural forms and silently drops obsolete messages.
--stringtable-output
Write out a NeXTstep/GNUstep localized resource file in .strings syntax. Note that this file format doesn't support plural forms.
-w number
--width=number
Set the output page width. Long strings in the output files will be split across multiple lines in order to ensure that each line's width (= number of screen columns) is less or equal to the given number.
--no-wrap
Do not break long message lines. Message lines whose width exceeds the output page width will not be split into several lines. Only file reference lines which are wider than the output page width will be split.
-s
--sort-output
Generate sorted output. Note that using this option makes it much harder for the translator to understand each message's context.
-F
--sort-by-file
Sort output by file location.

#### 7.1.8 Informative output

-h
--help
Display this help and exit.
-V
--version
Output version information and exit.
-v
--verbose
Increase verbosity level.
-q
--quiet
--silent
Suppress progress indicators.

Next: , Previous: Updating, Up: Top

## 8 Editing PO Files

Next: , Previous: Editing, Up: Editing

### 8.1 KDE's PO File Editor

Next: , Previous: KBabel, Up: Editing

### 8.2 GNOME's PO File Editor

Next: , Previous: Gtranslator, Up: Editing

### 8.3 Emacs's PO File Editor

For those of you being the lucky users of Emacs, PO mode has been specifically created for providing a cozy environment for editing or modifying PO files. While editing a PO file, PO mode allows for the easy browsing of auxiliary and compendium PO files, as well as for following references into the set of C program sources from which PO files have been derived. It has a few special features, among which are the interactive marking of program strings as translatable, and the validation of PO files with easy repositioning to PO file lines showing errors.

For the beginning, besides main PO mode commands (see Main PO Commands), you should know how to move between entries (see Entry Positioning), and how to handle untranslated entries (see Untranslated Entries).

Next: , Previous: PO Mode, Up: PO Mode

#### 8.3.1 Completing GNU gettext Installation

Once you have received, unpacked, configured and compiled the GNU gettext distribution, the ‘make install’ command puts in place the programs xgettext, msgfmt, gettext, and msgmerge, as well as their available message catalogs. To top off a comfortable installation, you might also want to make the PO mode available to your Emacs users.

During the installation of the PO mode, you might want to modify your file .emacs, once and for all, so it contains a few lines looking like:

     (setq auto-mode-alist
(cons '("\\.po\\'\\|\\.po\\." . po-mode) auto-mode-alist))
(autoload 'po-mode "po-mode" "Major mode for translators to edit PO files" t)


Later, whenever you edit some .po file, or any file having the string ‘.po.’ within its name, Emacs loads po-mode.elc (or po-mode.el) as needed, and automatically activates PO mode commands for the associated buffer. The string PO appears in the mode line for any buffer for which PO mode is active. Many PO files may be active at once in a single Emacs session.

If you are using Emacs version 20 or newer, and have already installed the appropriate international fonts on your system, you may also tell Emacs how to determine automatically the coding system of every PO file. This will often (but not always) cause the necessary fonts to be loaded and used for displaying the translations on your Emacs screen. For this to happen, add the lines:

     (modify-coding-system-alist 'file "\\.po\\'\\|\\.po\\."
'po-find-file-coding-system)


to your .emacs file. If, with this, you still see boxes instead of international characters, try a different font set (via Shift Mouse button 1).

Next: , Previous: Installation, Up: PO Mode

#### 8.3.2 Main PO mode Commands

After setting up Emacs with something similar to the lines in Installation, PO mode is activated for a window when Emacs finds a PO file in that window. This puts the window read-only and establishes a po-mode-map, which is a genuine Emacs mode, in a way that is not derived from text mode in any way. Functions found on po-mode-hook, if any, will be executed.

When PO mode is active in a window, the letters ‘PO’ appear in the mode line for that window. The mode line also displays how many entries of each kind are held in the PO file. For example, the string ‘132t+3f+10u+2o’ would tell the translator that the PO mode contains 132 translated entries (see Translated Entries, 3 fuzzy entries (see Fuzzy Entries), 10 untranslated entries (see Untranslated Entries) and 2 obsolete entries (see Obsolete Entries). Zero-coefficients items are not shown. So, in this example, if the fuzzy entries were unfuzzied, the untranslated entries were translated and the obsolete entries were deleted, the mode line would merely display ‘145t’ for the counters.

The main PO commands are those which do not fit into the other categories of subsequent sections. These allow for quitting PO mode or for managing windows in special ways.

_
Undo last modification to the PO file (po-undo).
Q
Quit processing and save the PO file (po-quit).
q
Quit processing, possibly after confirmation (po-confirm-and-quit).
0
Temporary leave the PO file window (po-other-window).
?
h
Show help about PO mode (po-help).
=
Give some PO file statistics (po-statistics).
V
Batch validate the format of the whole PO file (po-validate).

The command _ (po-undo) interfaces to the Emacs undo facility. See Undoing Changes. Each time U is typed, modifications which the translator did to the PO file are undone a little more. For the purpose of undoing, each PO mode command is atomic. This is especially true for the <RET> command: the whole edition made by using a single use of this command is undone at once, even if the edition itself implied several actions. However, while in the editing window, one can undo the edition work quite parsimoniously.

The commands Q (po-quit) and q (po-confirm-and-quit) are used when the translator is done with the PO file. The former is a bit less verbose than the latter. If the file has been modified, it is saved to disk first. In both cases, and prior to all this, the commands check if any untranslated messages remain in the PO file and, if so, the translator is asked if she really wants to leave off working with this PO file. This is the preferred way of getting rid of an Emacs PO file buffer. Merely killing it through the usual command C-x k (kill-buffer) is not the tidiest way to proceed.

The command 0 (po-other-window) is another, softer way, to leave PO mode, temporarily. It just moves the cursor to some other Emacs window, and pops one if necessary. For example, if the translator just got PO mode to show some source context in some other, she might discover some apparent bug in the program source that needs correction. This command allows the translator to change sex, become a programmer, and have the cursor right into the window containing the program she (or rather he) wants to modify. By later getting the cursor back in the PO file window, or by asking Emacs to edit this file once again, PO mode is then recovered.

The command h (po-help) displays a summary of all available PO mode commands. The translator should then type any character to resume normal PO mode operations. The command ? has the same effect as h.

The command = (po-statistics) computes the total number of entries in the PO file, the ordinal of the current entry (counted from 1), the number of untranslated entries, the number of obsolete entries, and displays all these numbers.

The command V (po-validate) launches msgfmt in checking and verbose mode over the current PO file. This command first offers to save the current PO file on disk. The msgfmt tool, from GNU gettext, has the purpose of creating a MO file out of a PO file, and PO mode uses the features of this program for checking the overall format of a PO file, as well as all individual entries.

The program msgfmt runs asynchronously with Emacs, so the translator regains control immediately while her PO file is being studied. Error output is collected in the Emacs ‘*compilation*’ buffer, displayed in another window. The regular Emacs command C-x (next-error), as well as other usual compile commands, allow the translator to reposition quickly to the offending parts of the PO file. Once the cursor is on the line in error, the translator may decide on any PO mode action which would help correcting the error.

Next: , Previous: Main PO Commands, Up: PO Mode

#### 8.3.3 Entry Positioning

The cursor in a PO file window is almost always part of an entry. The only exceptions are the special case when the cursor is after the last entry in the file, or when the PO file is empty. The entry where the cursor is found to be is said to be the current entry. Many PO mode commands operate on the current entry, so moving the cursor does more than allowing the translator to browse the PO file, this also selects on which entry commands operate.

Some PO mode commands alter the position of the cursor in a specialized way. A few of those special purpose positioning are described here, the others are described in following sections (for a complete list try C-h m):

.
Redisplay the current entry (po-current-entry).
n
Select the entry after the current one (po-next-entry).
p
Select the entry before the current one (po-previous-entry).
<
Select the first entry in the PO file (po-first-entry).
>
Select the last entry in the PO file (po-last-entry).
m
Record the location of the current entry for later use (po-push-location).
r
Return to a previously saved entry location (po-pop-location).
x
Exchange the current entry location with the previously saved one (po-exchange-location).

Any Emacs command able to reposition the cursor may be used to select the current entry in PO mode, including commands which move by characters, lines, paragraphs, screens or pages, and search commands. However, there is a kind of standard way to display the current entry in PO mode, which usual Emacs commands moving the cursor do not especially try to enforce. The command . (po-current-entry) has the sole purpose of redisplaying the current entry properly, after the current entry has been changed by means external to PO mode, or the Emacs screen otherwise altered.

It is yet to be decided if PO mode helps the translator, or otherwise irritates her, by forcing a rigid window disposition while she is doing her work. We originally had quite precise ideas about how windows should behave, but on the other hand, anyone used to Emacs is often happy to keep full control. Maybe a fixed window disposition might be offered as a PO mode option that the translator might activate or deactivate at will, so it could be offered on an experimental basis. If nobody feels a real need for using it, or a compulsion for writing it, we should drop this whole idea. The incentive for doing it should come from translators rather than programmers, as opinions from an experienced translator are surely more worth to me than opinions from programmers thinking about how others should do translation.

The commands n (po-next-entry) and p (po-previous-entry) move the cursor the entry following, or preceding, the current one. If n is given while the cursor is on the last entry of the PO file, or if p is given while the cursor is on the first entry, no move is done.

The commands < (po-first-entry) and > (po-last-entry) move the cursor to the first entry, or last entry, of the PO file. When the cursor is located past the last entry in a PO file, most PO mode commands will return an error saying ‘After last entry’. Moreover, the commands < and > have the special property of being able to work even when the cursor is not into some PO file entry, and one may use them for nicely correcting this situation. But even these commands will fail on a truly empty PO file. There are development plans for the PO mode for it to interactively fill an empty PO file from sources. See Marking.

The translator may decide, before working at the translation of a particular entry, that she needs to browse the remainder of the PO file, maybe for finding the terminology or phraseology used in related entries. She can of course use the standard Emacs idioms for saving the current cursor location in some register, and use that register for getting back, or else, use the location ring.

PO mode offers another approach, by which cursor locations may be saved onto a special stack. The command m (po-push-location) merely adds the location of current entry to the stack, pushing the already saved locations under the new one. The command r (po-pop-location) consumes the top stack element and repositions the cursor to the entry associated with that top element. This position is then lost, for the next r will move the cursor to the previously saved location, and so on until no locations remain on the stack.

If the translator wants the position to be kept on the location stack, maybe for taking a look at the entry associated with the top element, then go elsewhere with the intent of getting back later, she ought to use m immediately after r.

The command x (po-exchange-location) simultaneously repositions the cursor to the entry associated with the top element of the stack of saved locations, and replaces that top element with the location of the current entry before the move. Consequently, repeating the x command toggles alternatively between two entries. For achieving this, the translator will position the cursor on the first entry, use m, then position to the second entry, and merely use x for making the switch.

Next: , Previous: Entry Positioning, Up: PO Mode

#### 8.3.4 Normalizing Strings in Entries

There are many different ways for encoding a particular string into a PO file entry, because there are so many different ways to split and quote multi-line strings, and even, to represent special characters by backslashed escaped sequences. Some features of PO mode rely on the ability for PO mode to scan an already existing PO file for a particular string encoded into the msgid field of some entry. Even if PO mode has internally all the built-in machinery for implementing this recognition easily, doing it fast is technically difficult. To facilitate a solution to this efficiency problem, we decided on a canonical representation for strings.

A conventional representation of strings in a PO file is currently under discussion, and PO mode experiments with a canonical representation. Having both xgettext and PO mode converging towards a uniform way of representing equivalent strings would be useful, as the internal normalization needed by PO mode could be automatically satisfied when using xgettext from GNU gettext. An explicit PO mode normalization should then be only necessary for PO files imported from elsewhere, or for when the convention itself evolves.

So, for achieving normalization of at least the strings of a given PO file needing a canonical representation, the following PO mode command is available:

M-x po-normalize
Tidy the whole PO file by making entries more uniform.

The special command M-x po-normalize, which has no associated keys, revises all entries, ensuring that strings of both original and translated entries use uniform internal quoting in the PO file. It also removes any crumb after the last entry. This command may be useful for PO files freshly imported from elsewhere, or if we ever improve on the canonical quoting format we use. This canonical format is not only meant for getting cleaner PO files, but also for greatly speeding up msgid string lookup for some other PO mode commands.

M-x po-normalize presently makes three passes over the entries. The first implements heuristics for converting PO files for GNU gettext 0.6 and earlier, in which msgid and msgstr fields were using K&R style C string syntax for multi-line strings. These heuristics may fail for comments not related to obsolete entries and ending with a backslash; they also depend on subsequent passes for finalizing the proper commenting of continued lines for obsolete entries. This first pass might disappear once all oldish PO files would have been adjusted. The second and third pass normalize all msgid and msgstr strings respectively. They also clean out those trailing backslashes used by XView's msgfmt for continued lines.

Having such an explicit normalizing command allows for importing PO files from other sources, but also eases the evolution of the current convention, evolution driven mostly by aesthetic concerns, as of now. It is easy to make suggested adjustments at a later time, as the normalizing command and eventually, other GNU gettext tools should greatly automate conformance. A description of the canonical string format is given below, for the particular benefit of those not having Emacs handy, and who would nevertheless want to handcraft their PO files in nice ways.

Right now, in PO mode, strings are single line or multi-line. A string goes multi-line if and only if it has embedded newlines, that is, if it matches ‘[^\n]\n+[^\n]’. So, we would have:

     msgstr "\n\nHello, world!\n\n\n"


but, replacing the space by a newline, this becomes:

     msgstr ""
"\n"
"\n"
"Hello,\n"
"world!\n"
"\n"
"\n"


We are deliberately using a caricatural example, here, to make the point clearer. Usually, multi-lines are not that bad looking. It is probable that we will implement the following suggestion. We might lump together all initial newlines into the empty string, and also all newlines introducing empty lines (that is, for n > 1, the n-1'th last newlines would go together on a separate string), so making the previous example appear:

     msgstr "\n\n"
"Hello,\n"
"world!\n"
"\n\n"


There are a few yet undecided little points about string normalization, to be documented in this manual, once these questions settle.

Next: , Previous: Normalizing, Up: PO Mode

#### 8.3.5 Translated Entries

Each PO file entry for which the msgstr field has been filled with a translation, and which is not marked as fuzzy (see Fuzzy Entries), is said to be a translated entry. Only translated entries will later be compiled by GNU msgfmt and become usable in programs. Other entry types will be excluded; translation will not occur for them.

Some commands are more specifically related to translated entry processing.

t
Find the next translated entry (po-next-translated-entry).
T
Find the previous translated entry (po-previous-translated-entry).

The commands t (po-next-translated-entry) and T (po-previous-translated-entry) move forwards or backwards, chasing for an translated entry. If none is found, the search is extended and wraps around in the PO file buffer.

Translated entries usually result from the translator having edited in a translation for them, Modifying Translations. However, if the variable po-auto-fuzzy-on-edit is not nil, the entry having received a new translation first becomes a fuzzy entry, which ought to be later unfuzzied before becoming an official, genuine translated entry. See Fuzzy Entries.

Next: , Previous: Translated Entries, Up: PO Mode

#### 8.3.6 Fuzzy Entries

Each PO file entry may have a set of attributes, which are qualities given a name and explicitly associated with the translation, using a special system comment. One of these attributes has the name fuzzy, and entries having this attribute are said to have a fuzzy translation. They are called fuzzy entries, for short.

Fuzzy entries, even if they account for translated entries for most other purposes, usually call for revision by the translator. Those may be produced by applying the program msgmerge to update an older translated PO files according to a new PO template file, when this tool hypothesises that some new msgid has been modified only slightly out of an older one, and chooses to pair what it thinks to be the old translation for the new modified entry. The slight alteration in the original string (the msgid string) should often be reflected in the translated string, and this requires the intervention of the translator. For this reason, msgmerge might mark some entries as being fuzzy.

Also, the translator may decide herself to mark an entry as fuzzy for her own convenience, when she wants to remember that the entry has to be later revisited. So, some commands are more specifically related to fuzzy entry processing.

z
Find the next fuzzy entry (po-next-fuzzy-entry).
Z
Find the previous fuzzy entry (po-previous-fuzzy-entry).
<TAB>
Remove the fuzzy attribute of the current entry (po-unfuzzy).

The commands z (po-next-fuzzy-entry) and Z (po-previous-fuzzy-entry) move forwards or backwards, chasing for a fuzzy entry. If none is found, the search is extended and wraps around in the PO file buffer.

The command <TAB> (po-unfuzzy) removes the fuzzy attribute associated with an entry, usually leaving it translated. Further, if the variable po-auto-select-on-unfuzzy has not the nil value, the <TAB> command will automatically chase for another interesting entry to work on. The initial value of po-auto-select-on-unfuzzy is nil.

The initial value of po-auto-fuzzy-on-edit is nil. However, if the variable po-auto-fuzzy-on-edit is set to t, any entry edited through the <RET> command is marked fuzzy, as a way to ensure some kind of double check, later. In this case, the usual paradigm is that an entry becomes fuzzy (if not already) whenever the translator modifies it. If she is satisfied with the translation, she then uses <TAB> to pick another entry to work on, clearing the fuzzy attribute on the same blow. If she is not satisfied yet, she merely uses <SPC> to chase another entry, leaving the entry fuzzy.

The translator may also use the <DEL> command (po-fade-out-entry) over any translated entry to mark it as being fuzzy, when she wants to easily leave a trace she wants to later return working at this entry.

Also, when time comes to quit working on a PO file buffer with the q command, the translator is asked for confirmation, if fuzzy string still exists.

Next: , Previous: Fuzzy Entries, Up: PO Mode

#### 8.3.7 Untranslated Entries

When xgettext originally creates a PO file, unless told otherwise, it initializes the msgid field with the untranslated string, and leaves the msgstr string to be empty. Such entries, having an empty translation, are said to be untranslated entries. Later, when the programmer slightly modifies some string right in the program, this change is later reflected in the PO file by the appearance of a new untranslated entry for the modified string.

The usual commands moving from entry to entry consider untranslated entries on the same level as active entries. Untranslated entries are easily recognizable by the fact they end with ‘msgstr "".

The work of the translator might be (quite naively) seen as the process of seeking for an untranslated entry, editing a translation for it, and repeating these actions until no untranslated entries remain. Some commands are more specifically related to untranslated entry processing.

u
Find the next untranslated entry (po-next-untranslated-entry).
U
Find the previous untranslated entry (po-previous-untransted-entry).
k
Turn the current entry into an untranslated one (po-kill-msgstr).

The commands u (po-next-untranslated-entry) and U (po-previous-untransted-entry) move forwards or backwards, chasing for an untranslated entry. If none is found, the search is extended and wraps around in the PO file buffer.

An entry can be turned back into an untranslated entry by merely emptying its translation, using the command k (po-kill-msgstr). See Modifying Translations.

Also, when time comes to quit working on a PO file buffer with the q command, the translator is asked for confirmation, if some untranslated string still exists.

Next: , Previous: Untranslated Entries, Up: PO Mode

#### 8.3.8 Obsolete Entries

By obsolete PO file entries, we mean those entries which are commented out, usually by msgmerge when it found that the translation is not needed anymore by the package being localized.

The usual commands moving from entry to entry consider obsolete entries on the same level as active entries. Obsolete entries are easily recognizable by the fact that all their lines start with #, even those lines containing msgid or msgstr.

Commands exist for emptying the translation or reinitializing it to the original untranslated string. Commands interfacing with the kill ring may force some previously saved text into the translation. The user may interactively edit the translation. All these commands may apply to obsolete entries, carefully leaving the entry obsolete after the fact.

Moreover, some commands are more specifically related to obsolete entry processing.

o
Find the next obsolete entry (po-next-obsolete-entry).
O
Find the previous obsolete entry (po-previous-obsolete-entry).
<DEL>
Make an active entry obsolete, or zap out an obsolete entry (po-fade-out-entry).

The commands o (po-next-obsolete-entry) and O (po-previous-obsolete-entry) move forwards or backwards, chasing for an obsolete entry. If none is found, the search is extended and wraps around in the PO file buffer.

PO mode does not provide ways for un-commenting an obsolete entry and making it active, because this would reintroduce an original untranslated string which does not correspond to any marked string in the program sources. This goes with the philosophy of never introducing useless msgid values.

However, it is possible to comment out an active entry, so making it obsolete. GNU gettext utilities will later react to the disappearance of a translation by using the untranslated string. The command <DEL> (po-fade-out-entry) pushes the current entry a little further towards annihilation. If the entry is active (it is a translated entry), then it is first made fuzzy. If it is already fuzzy, then the entry is merely commented out, with confirmation. If the entry is already obsolete, then it is completely deleted from the PO file. It is easy to recycle the translation so deleted into some other PO file entry, usually one which is untranslated. See Modifying Translations.

Here is a quite interesting problem to solve for later development of PO mode, for those nights you are not sleepy. The idea would be that PO mode might become bright enough, one of these days, to make good guesses at retrieving the most probable candidate, among all obsolete entries, for initializing the translation of a newly appeared string. I think it might be a quite hard problem to do this algorithmically, as we have to develop good and efficient measures of string similarity. Right now, PO mode completely lets the decision to the translator, when the time comes to find the adequate obsolete translation, it merely tries to provide handy tools for helping her to do so.

Next: , Previous: Obsolete Entries, Up: PO Mode

#### 8.3.9 Modifying Translations

PO mode prevents direct modification of the PO file, by the usual means Emacs gives for altering a buffer's contents. By doing so, it pretends helping the translator to avoid little clerical errors about the overall file format, or the proper quoting of strings, as those errors would be easily made. Other kinds of errors are still possible, but some may be caught and diagnosed by the batch validation process, which the translator may always trigger by the V command. For all other errors, the translator has to rely on her own judgment, and also on the linguistic reports submitted to her by the users of the translated package, having the same mother tongue.

When the time comes to create a translation, correct an error diagnosed mechanically or reported by a user, the translators have to resort to using the following commands for modifying the translations.

<RET>
Interactively edit the translation (po-edit-msgstr).
<LFD>
C-j
Reinitialize the translation with the original, untranslated string (po-msgid-to-msgstr).
k
Save the translation on the kill ring, and delete it (po-kill-msgstr).
w
Save the translation on the kill ring, without deleting it (po-kill-ring-save-msgstr).
y
Replace the translation, taking the new from the kill ring (po-yank-msgstr).

The command <RET> (po-edit-msgstr) opens a new Emacs window meant to edit in a new translation, or to modify an already existing translation. The new window contains a copy of the translation taken from the current PO file entry, all ready for edition, expunged of all quoting marks, fully modifiable and with the complete extent of Emacs modifying commands. When the translator is done with her modifications, she may use C-c C-c to close the subedit window with the automatically requoted results, or C-c C-k to abort her modifications. See Subedit, for more information.

The command <LFD> (po-msgid-to-msgstr) initializes, or reinitializes the translation with the original string. This command is normally used when the translator wants to redo a fresh translation of the original string, disregarding any previous work.

It is possible to arrange so, whenever editing an untranslated entry, the <LFD> command be automatically executed. If you set po-auto-edit-with-msgid to t, the translation gets initialised with the original string, in case none exists already. The default value for po-auto-edit-with-msgid is nil.

In fact, whether it is best to start a translation with an empty string, or rather with a copy of the original string, is a matter of taste or habit. Sometimes, the source language and the target language are so different that is simply best to start writing on an empty page. At other times, the source and target languages are so close that it would be a waste to retype a number of words already being written in the original string. A translator may also like having the original string right under her eyes, as she will progressively overwrite the original text with the translation, even if this requires some extra editing work to get rid of the original.

The command k (po-kill-msgstr) merely empties the translation string, so turning the entry into an untranslated one. But while doing so, its previous contents is put apart in a special place, known as the kill ring. The command w (po-kill-ring-save-msgstr) has also the effect of taking a copy of the translation onto the kill ring, but it otherwise leaves the entry alone, and does not remove the translation from the entry. Both commands use exactly the Emacs kill ring, which is shared between buffers, and which is well known already to Emacs lovers.

The translator may use k or w many times in the course of her work, as the kill ring may hold several saved translations. From the kill ring, strings may later be reinserted in various Emacs buffers. In particular, the kill ring may be used for moving translation strings between different entries of a single PO file buffer, or if the translator is handling many such buffers at once, even between PO files.

To facilitate exchanges with buffers which are not in PO mode, the translation string put on the kill ring by the k command is fully unquoted before being saved: external quotes are removed, multi-line strings are concatenated, and backslash escaped sequences are turned into their corresponding characters. In the special case of obsolete entries, the translation is also uncommented prior to saving.

The command y (po-yank-msgstr) completely replaces the translation of the current entry by a string taken from the kill ring. Following Emacs terminology, we then say that the replacement string is yanked into the PO file buffer. See Yanking. The first time y is used, the translation receives the value of the most recent addition to the kill ring. If y is typed once again, immediately, without intervening keystrokes, the translation just inserted is taken away and replaced by the second most recent addition to the kill ring. By repeating y many times in a row, the translator may travel along the kill ring for saved strings, until she finds the string she really wanted.

When a string is yanked into a PO file entry, it is fully and automatically requoted for complying with the format PO files should have. Further, if the entry is obsolete, PO mode then appropriately push the inserted string inside comments. Once again, translators should not burden themselves with quoting considerations besides, of course, the necessity of the translated string itself respective to the program using it.

Note that k or w are not the only commands pushing strings on the kill ring, as almost any PO mode command replacing translation strings (or the translator comments) automatically saves the old string on the kill ring. The main exceptions to this general rule are the yanking commands themselves.

To better illustrate the operation of killing and yanking, let's use an actual example, taken from a common situation. When the programmer slightly modifies some string right in the program, his change is later reflected in the PO file by the appearance of a new untranslated entry for the modified string, and the fact that the entry translating the original or unmodified string becomes obsolete. In many cases, the translator might spare herself some work by retrieving the unmodified translation from the obsolete entry, then initializing the untranslated entry msgstr field with this retrieved translation. Once this done, the obsolete entry is not wanted anymore, and may be safely deleted.

When the translator finds an untranslated entry and suspects that a slight variant of the translation exists, she immediately uses m to mark the current entry location, then starts chasing obsolete entries with o, hoping to find some translation corresponding to the unmodified string. Once found, she uses the <DEL> command for deleting the obsolete entry, knowing that <DEL> also kills the translation, that is, pushes the translation on the kill ring. Then, r returns to the initial untranslated entry, and y then yanks the saved translation right into the msgstr field. The translator is then free to use <RET> for fine tuning the translation contents, and maybe to later use u, then m again, for going on with the next untranslated string.

When some sequence of keys has to be typed over and over again, the translator may find it useful to become better acquainted with the Emacs capability of learning these sequences and playing them back under request. See Keyboard Macros.

Next: , Previous: Modifying Translations, Up: PO Mode

Any translation work done seriously will raise many linguistic difficulties, for which decisions have to be made, and the choices further documented. These documents may be saved within the PO file in form of translator comments, which the translator is free to create, delete, or modify at will. These comments may be useful to herself when she returns to this PO file after a while.

Comments not having whitespace after the initial ‘#’, for example, those beginning with ‘#.’ or ‘#:’, are not translator comments, they are exclusively created by other gettext tools. So, the commands below will never alter such system added comments, they are not meant for the translator to modify. See PO Files.

The following commands are somewhat similar to those modifying translations, so the general indications given for those apply here. See Modifying Translations.

#
Interactively edit the translator comments (po-edit-comment).
K
Save the translator comments on the kill ring, and delete it (po-kill-comment).
W
Save the translator comments on the kill ring, without deleting it (po-kill-ring-save-comment).
Y
Replace the translator comments, taking the new from the kill ring (po-yank-comment).

These commands parallel PO mode commands for modifying the translation strings, and behave much the same way as they do, except that they handle this part of PO file comments meant for translator usage, rather than the translation strings. So, if the descriptions given below are slightly succinct, it is because the full details have already been given. See Modifying Translations.

The command # (po-edit-comment) opens a new Emacs window containing a copy of the translator comments on the current PO file entry. If there are no such comments, PO mode understands that the translator wants to add a comment to the entry, and she is presented with an empty screen. Comment marks (#) and the space following them are automatically removed before edition, and reinstated after. For translator comments pertaining to obsolete entries, the uncommenting and recommenting operations are done twice. Once in the editing window, the keys C-c C-c allow the translator to tell she is finished with editing the comment. See Subedit, for further details.

Functions found on po-subedit-mode-hook, if any, are executed after the string has been inserted in the edit buffer.

The command K (po-kill-comment) gets rid of all translator comments, while saving those comments on the kill ring. The command W (po-kill-ring-save-comment) takes a copy of the translator comments on the kill ring, but leaves them undisturbed in the current entry. The command Y (po-yank-comment) completely replaces the translator comments by a string taken at the front of the kill ring. When this command is immediately repeated, the comments just inserted are withdrawn, and replaced by other strings taken along the kill ring.

On the kill ring, all strings have the same nature. There is no distinction between translation strings and translator comments strings. So, for example, let's presume the translator has just finished editing a translation, and wants to create a new translator comment to document why the previous translation was not good, just to remember what was the problem. Foreseeing that she will do that in her documentation, the translator may want to quote the previous translation in her translator comments. To do so, she may initialize the translator comments with the previous translation, still at the head of the kill ring. Because editing already pushed the previous translation on the kill ring, she merely has to type M-w prior to #, and the previous translation will be right there, all ready for being introduced by some explanatory text.

On the other hand, presume there are some translator comments already and that the translator wants to add to those comments, instead of wholly replacing them. Then, she should edit the comment right away with #. Once inside the editing window, she can use the regular Emacs commands C-y (yank) and M-y (yank-pop) to get the previous translation where she likes.

Next: , Previous: Modifying Comments, Up: PO Mode

#### 8.3.11 Details of Sub Edition

The PO subedit minor mode has a few peculiarities worth being described in fuller detail. It installs a few commands over the usual editing set of Emacs, which are described below.

C-c C-c
Complete edition (po-subedit-exit).
C-c C-k
Abort edition (po-subedit-abort).
C-c C-a
Consult auxiliary PO files (po-subedit-cycle-auxiliary).

The window's contents represents a translation for a given message, or a translator comment. The translator may modify this window to her heart's content. Once this is done, the command C-c C-c (po-subedit-exit) may be used to return the edited translation into the PO file, replacing the original translation, even if it moved out of sight or if buffers were switched.

If the translator becomes unsatisfied with her translation or comment, to the extent she prefers keeping what was existent prior to the <RET> or # command, she may use the command C-c C-k (po-subedit-abort) to merely get rid of edition, while preserving the original translation or comment. Another way would be for her to exit normally with C-c C-c, then type U once for undoing the whole effect of last edition.

The command C-c C-a (po-subedit-cycle-auxiliary) allows for glancing through translations already achieved in other languages, directly while editing the current translation. This may be quite convenient when the translator is fluent at many languages, but of course, only makes sense when such completed auxiliary PO files are already available to her (see Auxiliary).

Functions found on po-subedit-mode-hook, if any, are executed after the string has been inserted in the edit buffer.

While editing her translation, the translator should pay attention to not inserting unwanted <RET> (newline) characters at the end of the translated string if those are not meant to be there, or to removing such characters when they are required. Since these characters are not visible in the editing buffer, they are easily introduced by mistake. To help her, <RET> automatically puts the character < at the end of the string being edited, but this < is not really part of the string. On exiting the editing window with C-c C-c, PO mode automatically removes such < and all whitespace added after it. If the translator adds characters after the terminating <, it looses its delimiting property and integrally becomes part of the string. If she removes the delimiting <, then the edited string is taken as is, with all trailing newlines, even if invisible. Also, if the translated string ought to end itself with a genuine <, then the delimiting < may not be removed; so the string should appear, in the editing window, as ending with two < in a row.

When a translation (or a comment) is being edited, the translator may move the cursor back into the PO file buffer and freely move to other entries, browsing at will. If, with an edition pending, the translator wanders in the PO file buffer, she may decide to start modifying another entry. Each entry being edited has its own subedit buffer. It is possible to simultaneously edit the translation and the comment of a single entry, or to edit entries in different PO files, all at once. Typing <RET> on a field already being edited merely resumes that particular edit. Yet, the translator should better be comfortable at handling many Emacs windows!

Pending subedits may be completed or aborted in any order, regardless of how or when they were started. When many subedits are pending and the translator asks for quitting the PO file (with the q command), subedits are automatically resumed one at a time, so she may decide for each of them.

Next: , Previous: Subedit, Up: PO Mode

#### 8.3.12 C Sources Context

PO mode is particularly powerful when used with PO files created through GNU gettext utilities, as those utilities insert special comments in the PO files they generate. Some of these special comments relate the PO file entry to exactly where the untranslated string appears in the program sources.

When the translator gets to an untranslated entry, she is fairly often faced with an original string which is not as informative as it normally should be, being succinct, cryptic, or otherwise ambiguous. Before choosing how to translate the string, she needs to understand better what the string really means and how tight the translation has to be. Most of the time, when problems arise, the only way left to make her judgment is looking at the true program sources from where this string originated, searching for surrounding comments the programmer might have put in there, and looking around for helping clues of any kind.

Surely, when looking at program sources, the translator will receive more help if she is a fluent programmer. However, even if she is not versed in programming and feels a little lost in C code, the translator should not be shy at taking a look, once in a while. It is most probable that she will still be able to find some of the hints she needs. She will learn quickly to not feel uncomfortable in program code, paying more attention to programmer's comments, variable and function names (if he dared choosing them well), and overall organization, than to the program code itself.

The following commands are meant to help the translator at getting program source context for a PO file entry.

s
Resume the display of a program source context, or cycle through them (po-cycle-source-reference).
M-s
Display of a program source context selected by menu (po-select-source-reference).
S
Add a directory to the search path for source files (po-consider-source-path).
M-S
Delete a directory from the search path for source files (po-ignore-source-path).

The commands s (po-cycle-source-reference) and M-s (po-select-source-reference) both open another window displaying some source program file, and already positioned in such a way that it shows an actual use of the string to be translated. By doing so, the command gives source program context for the string. But if the entry has no source context references, or if all references are unresolved along the search path for program sources, then the command diagnoses this as an error.

Even if s (or M-s) opens a new window, the cursor stays in the PO file window. If the translator really wants to get into the program source window, she ought to do it explicitly, maybe by using command O.

When s is typed for the first time, or for a PO file entry which is different of the last one used for getting source context, then the command reacts by giving the first context available for this entry, if any. If some context has already been recently displayed for the current PO file entry, and the translator wandered off to do other things, typing s again will merely resume, in another window, the context last displayed. In particular, if the translator moved the cursor away from the context in the source file, the command will bring the cursor back to the context. By using s many times in a row, with no other commands intervening, PO mode will cycle to the next available contexts for this particular entry, getting back to the first context once the last has been shown.

The command M-s behaves differently. Instead of cycling through references, it lets the translator choose a particular reference among many, and displays that reference. It is best used with completion, if the translator types <TAB> immediately after M-s, in response to the question, she will be offered a menu of all possible references, as a reminder of which are the acceptable answers. This command is useful only where there are really many contexts available for a single string to translate.

Program source files are usually found relative to where the PO file stands. As a special provision, when this fails, the file is also looked for, but relative to the directory immediately above it. Those two cases take proper care of most PO files. However, it might happen that a PO file has been moved, or is edited in a different place than its normal location. When this happens, the translator should tell PO mode in which directory normally sits the genuine PO file. Many such directories may be specified, and all together, they constitute what is called the search path for program sources. The command S (po-consider-source-path) is used to interactively enter a new directory at the front of the search path, and the command M-S (po-ignore-source-path) is used to select, with completion, one of the directories she does not want anymore on the search path.

Previous: C Sources Context, Up: PO Mode

#### 8.3.13 Consulting Auxiliary PO Files

PO mode is able to help the knowledgeable translator, being fluent in many languages, at taking advantage of translations already achieved in other languages she just happens to know. It provides these other language translations as additional context for her own work. Moreover, it has features to ease the production of translations for many languages at once, for translators preferring to work in this way.

An auxiliary PO file is an existing PO file meant for the same package the translator is working on, but targeted to a different mother tongue language. Commands exist for declaring and handling auxiliary PO files, and also for showing contexts for the entry under work.

Here are the auxiliary file commands available in PO mode.

a
Seek auxiliary files for another translation for the same entry (po-cycle-auxiliary).
C-c C-a
Switch to a particular auxiliary file (po-select-auxiliary).
A
Declare this PO file as an auxiliary file (po-consider-as-auxiliary).
M-A
Remove this PO file from the list of auxiliary files (po-ignore-as-auxiliary).

Command A (po-consider-as-auxiliary) adds the current PO file to the list of auxiliary files, while command M-A (po-ignore-as-auxiliary just removes it.

The command a (po-cycle-auxiliary) seeks all auxiliary PO files, round-robin, searching for a translated entry in some other language having an msgid field identical as the one for the current entry. The found PO file, if any, takes the place of the current PO file in the display (its window gets on top). Before doing so, the current PO file is also made into an auxiliary file, if not already. So, a in this newly displayed PO file will seek another PO file, and so on, so repeating a will eventually yield back the original PO file.

The command C-c C-a (po-select-auxiliary) asks the translator for her choice of a particular auxiliary file, with completion, and then switches to that selected PO file. The command also checks if the selected file has an msgid field identical as the one for the current entry, and if yes, this entry becomes current. Otherwise, the cursor of the selected file is left undisturbed.

For all this to work fully, auxiliary PO files will have to be normalized, in that way that msgid fields should be written exactly the same way. It is possible to write msgid fields in various ways for representing the same string, different writing would break the proper behaviour of the auxiliary file commands of PO mode. This is not expected to be much a problem in practice, as most existing PO files have their msgid entries written by the same GNU gettext tools.

However, PO files initially created by PO mode itself, while marking strings in source files, are normalised differently. So are PO files resulting of the ‘M-x normalize’ command. Until these discrepancies between PO mode and other GNU gettext tools get fully resolved, the translator should stay aware of normalisation issues.

Previous: PO Mode, Up: Editing

### 8.4 Using Translation Compendia

A compendium is a special PO file containing a set of translations recurring in many different packages. The translator can use gettext tools to build a new compendium, to add entries to her compendium, and to initialize untranslated entries, or to update already translated entries, from translations kept in the compendium.

Next: , Previous: Compendium, Up: Compendium

#### 8.4.1 Creating Compendia

Basically every PO file consisting of translated entries only can be declared as a valid compendium. Often the translator wants to have special compendia; let's consider two cases: concatenating PO files and extracting a message subset from a PO file.

##### 8.4.1.1 Concatenate PO Files

To concatenate several valid PO files into one compendium file you can use ‘msgcomm’ or ‘msgcat’ (the latter preferred):

     msgcat -o compendium.po file1.po file2.po


By default, msgcat will accumulate divergent translations for the same string. Those occurrences will be marked as fuzzy and highly visible decorated; calling msgcat on file1.po:

     #: src/hello.c:200
#, c-format
msgid "Report bugs to <%s>.\n"
msgstr "Comunicar bugs' a <%s>.\n"


and file2.po:

     #: src/bye.c:100
#, c-format
msgid "Report bugs to <%s>.\n"
msgstr "Comunicar \"bugs\" a <%s>.\n"


will result in:

     #: src/hello.c:200 src/bye.c:100
#, fuzzy, c-format
msgid "Report bugs to <%s>.\n"
msgstr ""
"#-#-#-#-#  file1.po  #-#-#-#-#\n"
"Comunicar bugs' a <%s>.\n"
"#-#-#-#-#  file2.po  #-#-#-#-#\n"
"Comunicar \"bugs\" a <%s>.\n"


The translator will have to resolve this “conflict” manually; she has to decide whether the first or the second version is appropriate (or provide a new translation), to delete the “marker lines”, and finally to remove the fuzzy mark.

If the translator knows in advance the first found translation of a message is always the best translation she can make use to the ‘--use-first’ switch:

     msgcat --use-first -o compendium.po file1.po file2.po


A good compendium file must not contain fuzzy or untranslated entries. If input files are “dirty” you must preprocess the input files or postprocess the result using ‘msgattrib --translated --no-fuzzy’.

##### 8.4.1.2 Extract a Message Subset from a PO File

Nobody wants to translate the same messages again and again; thus you may wish to have a compendium file containing getopt.c messages.

To extract a message subset (e.g., all getopt.c messages) from an existing PO file into one compendium file you can use ‘msggrep’:

     msggrep --location src/getopt.c -o compendium.po file.po


Previous: Creating Compendia, Up: Compendium

#### 8.4.2 Using Compendia

You can use a compendium file to initialize a translation from scratch or to update an already existing translation.

##### 8.4.2.1 Initialize a New Translation File

Since a PO file with translations does not exist the translator can merely use /dev/null to fake the “old” translation file.

     msgmerge --compendium compendium.po -o file.po /dev/null file.pot

##### 8.4.2.2 Update an Existing Translation File

Concatenate the compendium file(s) and the existing PO, merge the result with the POT file and remove the obsolete entries (optional, here done using ‘sed’):

     msgcat --use-first -o update.po compendium1.po compendium2.po file.po
msgmerge update.po file.pot | msgattrib --no-obsolete > file.po


Next: , Previous: Editing, Up: Top

## 9 Manipulating PO Files

Sometimes it is necessary to manipulate PO files in a way that is better performed automatically than by hand. GNU gettext includes a complete set of tools for this purpose.

When merging two packages into a single package, the resulting POT file will be the concatenation of the two packages' POT files. Thus the maintainer must concatenate the two existing package translations into a single translation catalog, for each language. This is best performed using ‘msgcat’. It is then the translators' duty to deal with any possible conflicts that arose during the merge.

When a translator takes over the translation job from another translator, but she uses a different character encoding in her locale, she will convert the catalog to her character encoding. This is best done through the ‘msgconv’ program.

When a maintainer takes a source file with tagged messages from another package, he should also take the existing translations for this source file (and not let the translators do the same job twice). One way to do this is through ‘msggrep’, another is to create a POT file for that source file and use ‘msgmerge’.

When a translator wants to adjust some translation catalog for a special dialect or orthography — for example, German as written in Switzerland versus German as written in Germany — she needs to apply some text processing to every message in the catalog. The tool for doing this is ‘msgfilter’.

Another use of msgfilter is to produce approximately the POT file for which a given PO file was made. This can be done through a filter command like ‘msgfilter sed -e d | sed -e '/^# /d'’. Note that the original POT file may have had different comments and different plural message counts, that's why it's better to use the original POT file if available.

When a translator wants to check her translations, for example according to orthography rules or using a non-interactive spell checker, she can do so using the ‘msgexec’ program.

When third party tools create PO or POT files, sometimes duplicates cannot be avoided. But the GNU gettext tools give an error when they encounter duplicate msgids in the same file and in the same domain. To merge duplicates, the ‘msguniq’ program can be used.

msgcomm’ is a more general tool for keeping or throwing away duplicates, occurring in different files.

msgcmp’ can be used to check whether a translation catalog is completely translated.

msgattrib’ can be used to select and extract only the fuzzy or untranslated messages of a translation catalog.

msgen’ is useful as a first step for preparing English translation catalogs. It copies each message's msgid to its msgstr.

Finally, for those applications where all these various programs are not sufficient, a library ‘libgettextpo’ is provided that can be used to write other specialized programs that process PO files.

Next: , Previous: Manipulating, Up: Manipulating

### 9.1 Invoking the msgcat Program

     msgcat [option] [inputfile]...


The msgcat program concatenates and merges the specified PO files. It finds messages which are common to two or more of the specified PO files. By using the --more-than option, greater commonality may be requested before messages are printed. Conversely, the --less-than option may be used to specify less commonality before messages are printed (i.e. ‘--less-than=2’ will only print the unique messages). Translations, comments and extract comments will be cumulated, except that if --use-first is specified, they will be taken from the first PO file to define them. File positions from all PO files will be cumulated.

#### 9.1.1 Input file location

inputfile ...
Input files.
-f file
--files-from=file
Read the names of the input files from file instead of getting them from the command line.
-D directory
--directory=directory
Add directory to the list of directories. Source files are searched relative to this list of directories. The resulting .po file will be written relative to the current directory, though.

If inputfile is ‘-’, standard input is read.

#### 9.1.2 Output file location

-o file
--output-file=file
Write output to specified file.

The results are written to standard output if no output file is specified or if it is ‘-’.

#### 9.1.3 Message selection

-< number
--less-than=number
Print messages with less than number definitions, defaults to infinite if not set.
-> number
--more-than=number
Print messages with more than number definitions, defaults to 0 if not set.
-u
--unique
Shorthand for ‘--less-than=2’. Requests that only unique messages be printed.

#### 9.1.4 Input file syntax

-P
--properties-input
Assume the input files are Java ResourceBundles in Java .properties syntax, not in PO file syntax.
--stringtable-input
Assume the input files are NeXTstep/GNUstep localized resource files in .strings syntax, not in PO file syntax.

#### 9.1.5 Output details

-t
--to-code=name
Specify encoding for output.
--use-first
Use first available translation for each message. Don't merge several translations into one.
--color
--color=when
Specify whether or when to use colors and other text attributes. See The –color option for details.
--style=style_file
Specify the CSS style rule file to use for --color. See The –style option for details.
--force-po
Always write an output file even if it contains no message.
-i
--indent
Write the .po file using indented style.
--no-location
Do not write ‘#: filename:line’ lines.
-n
Generate ‘#: filename:line’ lines (default).
--strict
Write out a strict Uniforum conforming PO file. Note that this Uniforum format should be avoided because it doesn't support the GNU extensions.
-p
--properties-output
Write out a Java ResourceBundle in Java .properties syntax. Note that this file format doesn't support plural forms and silently drops obsolete messages.
--stringtable-output
Write out a NeXTstep/GNUstep localized resource file in .strings syntax. Note that this file format doesn't support plural forms.
-w number
--width=number
Set the output page width. Long strings in the output files will be split across multiple lines in order to ensure that each line's width (= number of screen columns) is less or equal to the given number.
--no-wrap
Do not break long message lines. Message lines whose width exceeds the output page width will not be split into several lines. Only file reference lines which are wider than the output page width will be split.
-s
--sort-output
Generate sorted output. Note that using this option makes it much harder for the translator to understand each message's context.
-F
--sort-by-file
Sort output by file location.

#### 9.1.6 Informative output

-h
--help
Display this help and exit.
-V
--version
Output version information and exit.

Next: , Previous: msgcat Invocation, Up: Manipulating

### 9.2 Invoking the msgconv Program

     msgconv [option] [inputfile]


The msgconv program converts a translation catalog to a different character encoding.

#### 9.2.1 Input file location

inputfile
Input PO file.
-D directory
--directory=directory
Add directory to the list of directories. Source files are searched relative to this list of directories. The resulting .po file will be written relative to the current directory, though.

If no inputfile is given or if it is ‘-’, standard input is read.

#### 9.2.2 Output file location

-o file
--output-file=file
Write output to specified file.

The results are written to standard output if no output file is specified or if it is ‘-’.

#### 9.2.3 Conversion target

-t
--to-code=name
Specify encoding for output.

The default encoding is the current locale's encoding.

#### 9.2.4 Input file syntax

-P
--properties-input
Assume the input file is a Java ResourceBundle in Java .properties syntax, not in PO file syntax.
--stringtable-input
Assume the input file is a NeXTstep/GNUstep localized resource file in .strings syntax, not in PO file syntax.

#### 9.2.5 Output details

--force-po
Always write an output file even if it contains no message.
-i
--indent
Write the .po file using indented style.
--no-location
Do not write ‘#: filename:line’ lines.
Generate ‘#: filename:line’ lines (default).
--strict
Write out a strict Uniforum conforming PO file. Note that this Uniforum format should be avoided because it doesn't support the GNU extensions.
-p
--properties-output
Write out a Java ResourceBundle in Java .properties syntax. Note that this file format doesn't support plural forms and silently drops obsolete messages.
--stringtable-output
Write out a NeXTstep/GNUstep localized resource file in .strings syntax. Note that this file format doesn't support plural forms.
-w number
--width=number
Set the output page width. Long strings in the output files will be split across multiple lines in order to ensure that each line's width (= number of screen columns) is less or equal to the given number.
--no-wrap
Do not break long message lines. Message lines whose width exceeds the output page width will not be split into several lines. Only file reference lines which are wider than the output page width will be split.
-s
--sort-output
Generate sorted output. Note that using this option makes it much harder for the translator to understand each message's context.
-F
--sort-by-file
Sort output by file location.

#### 9.2.6 Informative output

-h
--help
Display this help and exit.
-V
--version
Output version information and exit.

Next: , Previous: msgconv Invocation, Up: Manipulating

### 9.3 Invoking the msggrep Program

     msggrep [option] [inputfile]


The msggrep program extracts all messages of a translation catalog that match a given pattern or belong to some given source files.

#### 9.3.1 Input file location

inputfile
Input PO file.
-D directory
--directory=directory
Add directory to the list of directories. Source files are searched relative to this list of directories. The resulting .po file will be written relative to the current directory, though.

If no inputfile is given or if it is ‘-’, standard input is read.

#### 9.3.2 Output file location

-o file
--output-file=file
Write output to specified file.

The results are written to standard output if no output file is specified or if it is ‘-’.

#### 9.3.3 Message selection

       [-N sourcefile]... [-M domainname]...
[-J msgctxt-pattern] [-K msgid-pattern] [-T msgstr-pattern]
[-C comment-pattern]


A message is selected if

• it comes from one of the specified source files,
• or if it comes from one of the specified domains,
• or if ‘-J’ is given and its context (msgctxt) matches msgctxt-pattern,
• or if ‘-K’ is given and its key (msgid or msgid_plural) matches msgid-pattern,
• or if ‘-T’ is given and its translation (msgstr) matches msgstr-pattern,
• or if ‘-C’ is given and the translator's comment matches comment-pattern.

When more than one selection criterion is specified, the set of selected messages is the union of the selected messages of each criterion.

msgctxt-pattern or msgid-pattern or msgstr-pattern syntax:

       [-E | -F] [-e pattern | -f file]...


patterns are basic regular expressions by default, or extended regular expressions if -E is given, or fixed strings if -F is given.

-N sourcefile
--location=sourcefile
Select messages extracted from sourcefile. sourcefile can be either a literal file name or a wildcard pattern.
-M domainname
--domain=domainname
Select messages belonging to domain domainname.
-J
--msgctxt
Start of patterns for the msgctxt.
-K
--msgid
Start of patterns for the msgid.
-T
--msgstr
Start of patterns for the msgstr.
-C
--comment
Start of patterns for the translator's comment.
-X
--extracted-comment
Start of patterns for the extracted comments.
-E
--extended-regexp
Specify that pattern is an extended regular expression.
-F
--fixed-strings
Specify that pattern is a set of newline-separated strings.
-e pattern
--regexp=pattern
Use pattern as a regular expression.
-f file
--file=file
Obtain pattern from file.
-i
--ignore-case
Ignore case distinctions.
-v
--invert-match
Output only the messages that do not match any selection criterion, instead of the messages that match a selection criterion.

#### 9.3.4 Input file syntax

-P
--properties-input
Assume the input file is a Java ResourceBundle in Java .properties syntax, not in PO file syntax.
--stringtable-input
Assume the input file is a NeXTstep/GNUstep localized resource file in .strings syntax, not in PO file syntax.

#### 9.3.5 Output details

--force-po
Always write an output file even if it contains no message.
--indent
Write the .po file using indented style.
--no-location
Do not write ‘#: filename:line’ lines.
Generate ‘#: filename:line’ lines (default).
--strict
Write out a strict Uniforum conforming PO file. Note that this Uniforum format should be avoided because it doesn't support the GNU extensions.
-p
--properties-output
Write out a Java ResourceBundle in Java .properties syntax. Note that this file format doesn't support plural forms and silently drops obsolete messages.
--stringtable-output
Write out a NeXTstep/GNUstep localized resource file in .strings syntax. Note that this file format doesn't support plural forms.
-w number
--width=number
Set the output page width. Long strings in the output files will be split across multiple lines in order to ensure that each line's width (= number of screen columns) is less or equal to the given number.
--no-wrap
Do not break long message lines. Message lines whose width exceeds the output page width will not be split into several lines. Only file reference lines which are wider than the output page width will be split.
--sort-output
Generate sorted output. Note that using this option makes it much harder for the translator to understand each message's context.
--sort-by-file
Sort output by file location.

#### 9.3.6 Informative output

-h
--help
Display this help and exit.
-V
--version
Output version information and exit.

#### 9.3.7 Examples

To extract the messages that come from the source files gnulib-lib/error.c and gnulib-lib/getopt.c:

     msggrep -N gnulib-lib/error.c -N gnulib-lib/getopt.c input.po


To extract the messages that contain the string “Please specify” in the original string:

     msggrep --msgid -F -e 'Please specify' input.po


To extract the messages that have a context specifier of either “Menu>File” or “Menu>Edit” or a submenu of them:

     msggrep --msgctxt -E -e '^Menu>(File|Edit)' input.po


To extract the messages whose translation contains one of the strings in the file wordlist.txt:

     msggrep --msgstr -F -f wordlist.txt input.po


Next: , Previous: msggrep Invocation, Up: Manipulating

### 9.4 Invoking the msgfilter Program

     msgfilter [option] filter [filter-option]


The msgfilter program applies a filter to all translations of a translation catalog.

#### 9.4.1 Input file location

-i inputfile
--input=inputfile
Input PO file.
-D directory
--directory=directory
Add directory to the list of directories. Source files are searched relative to this list of directories. The resulting .po file will be written relative to the current directory, though.

If no inputfile is given or if it is ‘-’, standard input is read.

#### 9.4.2 Output file location

-o file
--output-file=file
Write output to specified file.

The results are written to standard output if no output file is specified or if it is ‘-’.

#### 9.4.3 The filter

The filter can be any program that reads a translation from standard input and writes a modified translation to standard output. A frequently used filter is ‘sed’. A few particular built-in filters are also recognized.

Note: If the filter is not a built-in filter, you have to care about encodings: It is your responsibility to ensure that the filter can cope with input encoded in the translation catalog's encoding. If the filter wants input in a particular encoding, you can in a first step convert the translation catalog to that encoding using the ‘msgconv’ program, before invoking ‘msgfilter’. If the filter wants input in the locale's encoding, but you want to avoid the locale's encoding, then you can first convert the translation catalog to UTF-8 using the ‘msgconv’ program and then make ‘msgfilter’ work in an UTF-8 locale, by using the LC_ALL environment variable.

Note: Most translations in a translation catalog don't end with a newline character. For this reason, it is important that the filter recognizes its last input line even if it ends without a newline, and that it doesn't add an undesired trailing newline at the end. The ‘sed’ program on some platforms is known to ignore the last line of input if it is not terminated with a newline. You can use GNU sed instead; it does not have this limitation.

#### 9.4.4 Useful filter-options when the filter is ‘sed’

-e script
--expression=script
Add script to the commands to be executed.
-f scriptfile
--file=scriptfile
Add the contents of scriptfile to the commands to be executed.
-n
--quiet
--silent
Suppress automatic printing of pattern space.

#### 9.4.5 Built-in filters

The filter ‘recode-sr-latin’ is recognized as a built-in filter. The command ‘recode-sr-latin’ converts Serbian text, written in the Cyrillic script, to the Latin script. The command ‘msgfilter recode-sr-latin’ applies this conversion to the translations of a PO file. Thus, it can be used to convert an sr.po file to an sr@latin.po file.

The use of built-in filters is not sensitive to the current locale's encoding. Moreover, when used with a built-in filter, ‘msgfilter’ can automatically convert the message catalog to the UTF-8 encoding when needed.

#### 9.4.6 Input file syntax

-P
--properties-input
Assume the input file is a Java ResourceBundle in Java .properties syntax, not in PO file syntax.
--stringtable-input
Assume the input file is a NeXTstep/GNUstep localized resource file in .strings syntax, not in PO file syntax.

#### 9.4.7 Output details

--force-po
Always write an output file even if it contains no message.
--indent
Write the .po file using indented style.
Keep the header entry, i.e. the message with ‘msgid ""’, unmodified, instead of filtering it. By default, the header entry is subject to filtering like any other message.
--no-location
Do not write ‘#: filename:line’ lines.
Generate ‘#: filename:line’ lines (default).
--strict
Write out a strict Uniforum conforming PO file. Note that this Uniforum format should be avoided because it doesn't support the GNU extensions.
-p
--properties-output
Write out a Java ResourceBundle in Java .properties syntax. Note that this file format doesn't support plural forms and silently drops obsolete messages.
--stringtable-output
Write out a NeXTstep/GNUstep localized resource file in .strings syntax. Note that this file format doesn't support plural forms.
-w number
--width=number
Set the output page width. Long strings in the output files will be split across multiple lines in order to ensure that each line's width (= number of screen columns) is less or equal to the given number.
--no-wrap
Do not break long message lines. Message lines whose width exceeds the output page width will not be split into several lines. Only file reference lines which are wider than the output page width will be split.
-s
--sort-output
Generate sorted output. Note that using this option makes it much harder for the translator to understand each message's context.
-F
--sort-by-file
Sort output by file location.

#### 9.4.8 Informative output

-h
--help
Display this help and exit.
-V
--version
Output version information and exit.

#### 9.4.9 Examples

To convert German translations to Swiss orthography (in an UTF-8 locale):

     msgconv -t UTF-8 de.po | msgfilter sed -e 's/ß/ss/g'


To convert Serbian translations in Cyrillic script to Latin script:

     msgfilter recode-sr-latin sr.po


Next: , Previous: msgfilter Invocation, Up: Manipulating

### 9.5 Invoking the msguniq Program

     msguniq [option] [inputfile]


The msguniq program unifies duplicate translations in a translation catalog. It finds duplicate translations of the same message ID. Such duplicates are invalid input for other programs like msgfmt, msgmerge or msgcat. By default, duplicates are merged together. When using the ‘--repeated’ option, only duplicates are output, and all other messages are discarded. Comments and extracted comments will be cumulated, except that if ‘--use-first’ is specified, they will be taken from the first translation. File positions will be cumulated. When using the ‘--unique’ option, duplicates are discarded.

#### 9.5.1 Input file location

inputfile
Input PO file.
-D directory
--directory=directory
Add directory to the list of directories. Source files are searched relative to this list of directories. The resulting .po file will be written relative to the current directory, though.

If no inputfile is given or if it is ‘-’, standard input is read.

#### 9.5.2 Output file location

-o file
--output-file=file
Write output to specified file.

The results are written to standard output if no output file is specified or if it is ‘-’.

#### 9.5.3 Message selection

-d
--repeated
Print only duplicates.
-u
--unique
Print only unique messages, discard duplicates.

#### 9.5.4 Input file syntax

-P
--properties-input
Assume the input file is a Java ResourceBundle in Java .properties syntax, not in PO file syntax.
--stringtable-input
Assume the input file is a NeXTstep/GNUstep localized resource file in .strings syntax, not in PO file syntax.

#### 9.5.5 Output details

-t
--to-code=name
Specify encoding for output.
--use-first
Use first available translation for each message. Don't merge several translations into one.
--force-po
Always write an output file even if it contains no message.
-i
--indent
Write the .po file using indented style.
--no-location
Do not write ‘#: filename:line’ lines.
-n
Generate ‘#: filename:line’ lines (default).
--strict
Write out a strict Uniforum conforming PO file. Note that this Uniforum format should be avoided because it doesn't support the GNU extensions.
-p
--properties-output
Write out a Java ResourceBundle in Java .properties syntax. Note that this file format doesn't support plural forms and silently drops obsolete messages.
--stringtable-output
Write out a NeXTstep/GNUstep localized resource file in .strings syntax. Note that this file format doesn't support plural forms.
-w number
--width=number
Set the output page width. Long strings in the output files will be split across multiple lines in order to ensure that each line's width (= number of screen columns) is less or equal to the given number.
--no-wrap
Do not break long message lines. Message lines whose width exceeds the output page width will not be split into several lines. Only file reference lines which are wider than the output page width will be split.
-s
--sort-output
Generate sorted output. Note that using this option makes it much harder for the translator to understand each message's context.
-F
--sort-by-file
Sort output by file location.

#### 9.5.6 Informative output

-h
--help
Display this help and exit.
-V
--version
Output version information and exit.

Next: , Previous: msguniq Invocation, Up: Manipulating

### 9.6 Invoking the msgcomm Program

     msgcomm [option] [inputfile]...


The msgcomm program finds messages which are common to two or more of the specified PO files. By using the --more-than option, greater commonality may be requested before messages are printed. Conversely, the --less-than option may be used to specify less commonality before messages are printed (i.e. ‘--less-than=2’ will only print the unique messages). Translations, comments and extract comments will be preserved, but only from the first PO file to define them. File positions from all PO files will be cumulated.

#### 9.6.1 Input file location

inputfile ...
Input files.
-f file
--files-from=file
Read the names of the input files from file instead of getting them from the command line.
-D directory
--directory=directory
Add directory to the list of directories. Source files are searched relative to this list of directories. The resulting .po file will be written relative to the current directory, though.

If inputfile is ‘-’, standard input is read.

#### 9.6.2 Output file location

-o file
--output-file=file
Write output to specified file.

The results are written to standard output if no output file is specified or if it is ‘-’.

#### 9.6.3 Message selection

-< number
--less-than=number
Print messages with less than number definitions, defaults to infinite if not set.
-> number
--more-than=number
Print messages with more than number definitions, defaults to 1 if not set.
-u
--unique
Shorthand for ‘--less-than=2’. Requests that only unique messages be printed.

#### 9.6.4 Input file syntax

-P
--properties-input
Assume the input files are Java ResourceBundles in Java .properties syntax, not in PO file syntax.
--stringtable-input
Assume the input files are NeXTstep/GNUstep localized resource files in .strings syntax, not in PO file syntax.

#### 9.6.5 Output details

--force-po
Always write an output file even if it contains no message.
-i
--indent
Write the .po file using indented style.
--no-location
Do not write ‘#: filename:line’ lines.
-n
Generate ‘#: filename:line’ lines (default).
--strict
Write out a strict Uniforum conforming PO file. Note that this Uniforum format should be avoided because it doesn't support the GNU extensions.
-p
--properties-output
Write out a Java ResourceBundle in Java .properties syntax. Note that this file format doesn't support plural forms and silently drops obsolete messages.
--stringtable-output
Write out a NeXTstep/GNUstep localized resource file in .strings syntax. Note that this file format doesn't support plural forms.
-w number
--width=number
Set the output page width. Long strings in the output files will be split across multiple lines in order to ensure that each line's width (= number of screen columns) is less or equal to the given number.
--no-wrap
Do not break long message lines. Message lines whose width exceeds the output page width will not be split into several lines. Only file reference lines which are wider than the output page width will be split.
-s
--sort-output
Generate sorted output. Note that using this option makes it much harder for the translator to understand each message's context.
-F
--sort-by-file
Sort output by file location.
Don't write header with ‘msgid ""’ entry.

#### 9.6.6 Informative output

-h
--help
Display this help and exit.
-V
--version
Output version information and exit.

Next: , Previous: msgcomm Invocation, Up: Manipulating

### 9.7 Invoking the msgcmp Program

     msgcmp [option] def.po ref.pot


The msgcmp program compares two Uniforum style .po files to check that both contain the same set of msgid strings. The def.po file is an existing PO file with the translations. The ref.pot file is the last created PO file, or a PO Template file (generally created by xgettext). This is useful for checking that you have translated each and every message in your program. Where an exact match cannot be found, fuzzy matching is used to produce better diagnostics.

#### 9.7.1 Input file location

def.po
Translations.
ref.pot
References to the sources.
-D directory
--directory=directory
Add directory to the list of directories. Source files are searched relative to this list of directories.

#### 9.7.2 Operation modifiers

-m
--multi-domain
Apply ref.pot to each of the domains in def.po.
--use-fuzzy
Consider fuzzy messages in the def.po file like translated messages. Note that using this option is usually wrong, because fuzzy messages are exactly those which have not been validated by a human translator.
--use-untranslated
Consider untranslated messages in the def.po file like translated messages. Note that using this option is usually wrong.

#### 9.7.3 Input file syntax

-P
--properties-input
Assume the input files are Java ResourceBundles in Java .properties syntax, not in PO file syntax.
--stringtable-input
Assume the input files are NeXTstep/GNUstep localized resource files in .strings syntax, not in PO file syntax.

#### 9.7.4 Informative output

-h
--help
Display this help and exit.
-V
--version
Output version information and exit.

Next: , Previous: msgcmp Invocation, Up: Manipulating

### 9.8 Invoking the msgattrib Program

     msgattrib [option] [inputfile]


The msgattrib program filters the messages of a translation catalog according to their attributes, and manipulates the attributes.

#### 9.8.1 Input file location

inputfile
Input PO file.
-D directory
--directory=directory
Add directory to the list of directories. Source files are searched relative to this list of directories. The resulting .po file will be written relative to the current directory, though.

If no inputfile is given or if it is ‘-’, standard input is read.

#### 9.8.2 Output file location

-o file
--output-file=file
Write output to specified file.

The results are written to standard output if no output file is specified or if it is ‘-’.

#### 9.8.3 Message selection

--translated
Keep translated messages, remove untranslated messages.
--untranslated
Keep untranslated messages, remove translated messages.
--no-fuzzy
Remove ‘fuzzy’ marked messages.
--only-fuzzy
Keep ‘fuzzy’ marked messages, remove all other messages.
--no-obsolete
Remove obsolete #~ messages.
--only-obsolete
Keep obsolete #~ messages, remove all other messages.

#### 9.8.4 Attribute manipulation

Attributes are modified after the message selection/removal has been performed. If the ‘--only-file’ or ‘--ignore-file’ option is specified, the attribute modification is applied only to those messages that are listed in the only-file and not listed in the ignore-file.

--set-fuzzy
Set all messages ‘fuzzy’.
--clear-fuzzy
Set all messages non-‘fuzzy’.
--set-obsolete
Set all messages obsolete.
--clear-obsolete
Set all messages non-obsolete.
--clear-previous
Remove the “previous msgid” (‘#|’) comments from all messages.
--only-file=file
Limit the attribute changes to entries that are listed in file. file should be a PO or POT file.
--ignore-file=file
Limit the attribute changes to entries that are not listed in file. file should be a PO or POT file.
--fuzzy
Synonym for ‘--only-fuzzy --clear-fuzzy’: It keeps only the fuzzy messages and removes their ‘fuzzy’ mark.
--obsolete
Synonym for ‘--only-obsolete --clear-obsolete’: It keeps only the obsolete messages and makes them non-obsolete.

#### 9.8.5 Input file syntax

-P
--properties-input
Assume the input file is a Java ResourceBundle in Java .properties syntax, not in PO file syntax.
--stringtable-input
Assume the input file is a NeXTstep/GNUstep localized resource file in .strings syntax, not in PO file syntax.

#### 9.8.6 Output details

--force-po
Always write an output file even if it contains no message.
-i
--indent
Write the .po file using indented style.
--no-location
Do not write ‘#: filename:line’ lines.
-n
Generate ‘#: filename:line’ lines (default).
--strict
Write out a strict Uniforum conforming PO file. Note that this Uniforum format should be avoided because it doesn't support the GNU extensions.
-p
--properties-output
Write out a Java ResourceBundle in Java .properties syntax. Note that this file format doesn't support plural forms and silently drops obsolete messages.
--stringtable-output
Write out a NeXTstep/GNUstep localized resource file in .strings syntax. Note that this file format doesn't support plural forms.
-w number
--width=number
Set the output page width. Long strings in the output files will be split across multiple lines in order to ensure that each line's width (= number of screen columns) is less or equal to the given number.
--no-wrap
Do not break long message lines. Message lines whose width exceeds the output page width will not be split into several lines. Only file reference lines which are wider than the output page width will be split.
-s
--sort-output
Generate sorted output. Note that using this option makes it much harder for the translator to understand each message's context.
-F
--sort-by-file
Sort output by file location.

#### 9.8.7 Informative output

-h
--help
Display this help and exit.
-V
--version
Output version information and exit.

Next: , Previous: msgattrib Invocation, Up: Manipulating

### 9.9 Invoking the msgen Program

     msgen [option] inputfile


The msgen program creates an English translation catalog. The input file is the last created English PO file, or a PO Template file (generally created by xgettext). Untranslated entries are assigned a translation that is identical to the msgid.

Note: ‘msginit --no-translator --locale=en’ performs a very similar task. The main difference is that msginit cares specially about the header entry, whereas msgen doesn't.

#### 9.9.1 Input file location

inputfile
Input PO or POT file.
-D directory
--directory=directory
Add directory to the list of directories. Source files are searched relative to this list of directories. The resulting .po file will be written relative to the current directory, though.

If inputfile is ‘-’, standard input is read.

#### 9.9.2 Output file location

-o file
--output-file=file
Write output to specified file.

The results are written to standard output if no output file is specified or if it is ‘-’.

#### 9.9.3 Input file syntax

-P
--properties-input
Assume the input file is a Java ResourceBundle in Java .properties syntax, not in PO file syntax.
--stringtable-input
Assume the input file is a NeXTstep/GNUstep localized resource file in .strings syntax, not in PO file syntax.

#### 9.9.4 Output details

--force-po
Always write an output file even if it contains no message.
-i
--indent
Write the .po file using indented style.
--no-location
Do not write ‘#: filename:line’ lines.
Generate ‘#: filename:line’ lines (default).
--strict
Write out a strict Uniforum conforming PO file. Note that this Uniforum format should be avoided because it doesn't support the GNU extensions.
-p
--properties-output
Write out a Java ResourceBundle in Java .properties syntax. Note that this file format doesn't support plural forms and silently drops obsolete messages.
--stringtable-output
Write out a NeXTstep/GNUstep localized resource file in .strings syntax. Note that this file format doesn't support plural forms.
-w number
--width=number
Set the output page width. Long strings in the output files will be split across multiple lines in order to ensure that each line's width (= number of screen columns) is less or equal to the given number.
--no-wrap
Do not break long message lines. Message lines whose width exceeds the output page width will not be split into several lines. Only file reference lines which are wider than the output page width will be split.
-s
--sort-output
Generate sorted output. Note that using this option makes it much harder for the translator to understand each message's context.
-F
--sort-by-file
Sort output by file location.

#### 9.9.5 Informative output

-h
--help
Display this help and exit.
-V
--version
Output version information and exit.

Next: , Previous: msgen Invocation, Up: Manipulating

### 9.10 Invoking the msgexec Program

     msgexec [option] command [command-option]


The msgexec program applies a command to all translations of a translation catalog. The command can be any program that reads a translation from standard input. It is invoked once for each translation. Its output becomes msgexec's output. msgexec's return code is the maximum return code across all invocations.

A special builtin command called ‘0’ outputs the translation, followed by a null byte. The output of ‘msgexec 0’ is suitable as input for ‘xargs -0’.

During each command invocation, the environment variable MSGEXEC_MSGID is bound to the message's msgid, and the environment variable MSGEXEC_LOCATION is bound to the location in the PO file of the message. If the message has a context, the environment variable MSGEXEC_MSGCTXT is bound to the message's msgctxt, otherwise it is unbound.

Note: It is your responsibility to ensure that the command can cope with input encoded in the translation catalog's encoding. If the command wants input in a particular encoding, you can in a first step convert the translation catalog to that encoding using the ‘msgconv’ program, before invoking ‘msgexec’. If the command wants input in the locale's encoding, but you want to avoid the locale's encoding, then you can first convert the translation catalog to UTF-8 using the ‘msgconv’ program and then make ‘msgexec’ work in an UTF-8 locale, by using the LC_ALL environment variable.

#### 9.10.1 Input file location

-i inputfile
--input=inputfile
Input PO file.
-D directory
--directory=directory
Add directory to the list of directories. Source files are searched relative to this list of directories. The resulting .po file will be written relative to the current directory, though.

If no inputfile is given or if it is ‘-’, standard input is read.

#### 9.10.2 Input file syntax

-P
--properties-input
Assume the input file is a Java ResourceBundle in Java .properties syntax, not in PO file syntax.
--stringtable-input
Assume the input file is a NeXTstep/GNUstep localized resource file in .strings syntax, not in PO file syntax.

#### 9.10.3 Informative output

-h
--help
Display this help and exit.
-V
--version
Output version information and exit.

Next: , Previous: msgexec Invocation, Up: Manipulating

### 9.11 Highlighting parts of PO files

Translators are usually only interested in seeing the untranslated and fuzzy messages of a PO file. Also, when a message is set fuzzy because the msgid changed, they want to see the differences between the previous msgid and the current one (especially if the msgid is long and only few words in it have changed). Finally, it's always welcome to highlight the different sections of a message in a PO file (comments, msgid, msgstr, etc.).

Such highlighting is possible through the msgcat options ‘--color’ and ‘--style’.

Next: , Up: Colorizing

#### 9.11.1 The --color option

The ‘--color=when’ option specifies under which conditions colorized output should be generated. The when part can be one of the following:

always
yes
The output will be colorized.
never
no
The output will not be colorized.
auto
tty
The output will be colorized if the output device is a tty, i.e. when the output goes directly to a text screen or terminal emulator window.
html
The output will be colorized and be in HTML format.

--color’ is equivalent to ‘--color=yes’. The default is ‘--color=auto’.

Thus, a command like ‘msgcat vi.po’ will produce colorized output when called by itself in a command window. Whereas in a pipe, such as ‘msgcat vi.po | less -R’, it will not produce colorized output. To get colorized output in this situation nevertheless, use the command ‘msgcat --color vi.po | less -R’.

The ‘--color=html’ option will produce output that can be viewed in a browser. This can be useful, for example, for Indic languages, because the renderic of Indic scripts in browser is usually better than in terminal emulators.

Note that the output produced with the --color option is not a valid PO file in itself. It contains additional terminal-specific escape sequences or HTML tags. A PO file reader will give a syntax error when confronted with such content. Except for the ‘--color=html’ case, you therefore normally don't need to save output produced with the --color option in a file.

Next: , Previous: The --color option, Up: Colorizing

#### 9.11.2 The environment variable TERM

The environment variable TERM contains a identifier for the text window's capabilities. You can get a detailed list of these cababilities by using the ‘infocmp’ command, using ‘man 5 terminfo’ as a reference.

When producing text with embedded color directives, msgcat looks at the TERM variable. Text windows today typically support at least 8 colors. Often, however, the text window supports 16 or more colors, even though the TERM variable is set to a identifier denoting only 8 supported colors. It can be worth setting the TERM variable to a different value in these cases:

xterm
xterm is in most cases built with support for 16 colors. It can also be built with support for 88 or 256 colors (but not both). You can try to set TERM to either xterm-16color, xterm-88color, or xterm-256color.
rxvt
rxvt is often built with support for 16 colors. You can try to set TERM to rxvt-16color.
konsole
konsole too is often built with support for 16 colors. You can try to set TERM to konsole-16color or xterm-16color.

After setting TERM, you can verify it by invoking ‘msgcat --color=test’ and seeing whether the output looks like a reasonable color map.

Next: , Previous: The TERM variable, Up: Colorizing

#### 9.11.3 The --style option

The ‘--style=style_file’ option specifies the style file to use when colorizing. It has an effect only when the --color option is effective.

If the --style option is not specified, the environment variable PO_STYLE is considered. It is meant to point to the user's preferred style for PO files.

The default style file is $prefix/share/gettext/styles/po-default.css, where $prefix is the installation location.

A few style files are predefined:

po-vim.css
This style imitates the look used by vim 7.
po-emacs-x.css
This style imitates the look used by GNU Emacs 21 and 22 in an X11 window.
po-emacs-xterm.css
po-emacs-xterm16.css
po-emacs-xterm256.css
This style imitates the look used by GNU Emacs 22 in a terminal of type ‘xterm’ (8 colors) or ‘xterm-16color’ (16 colors) or ‘xterm-256color’ (256 colors), respectively.

You can use these styles without specifying a directory. They are actually located in $prefix/share/gettext/styles/, where $prefix is the installation location.

You can also design your own styles. This is described in the next section.

Next: , Previous: The --style option, Up: Colorizing

#### 9.11.4 Style rules for PO files

The same style file can be used for styling of a PO file, for terminal output and for HTML output. It is written in CSS (Cascading Style Sheet) syntax. See http://www.w3.org/TR/css2/cover.html for a formal definition of CSS. Many HTML authoring tutorials also contain explanations of CSS.

In the case of HTML output, the style file is embedded in the HTML output. In the case of text output, the style file is interpreted by the msgcat program. This means, in particular, that when @import is used with relative file names, the file names are

• relative to the resulting HTML file, in the case of HTML output,
• relative to the style sheet containing the @import, in the case of text output. (Actually, @imports are not yet supported in this case, due to a limitation in libcroco.)

CSS rules are built up from selectors and declarations. The declarations specify graphical properties; the selectors specify specify when they apply.

In PO files, the following simple selectors (based on "CSS classes", see the CSS2 spec, section 5.8.3) are supported.

• Selectors that apply to entire messages:
.header
This matches the header entry of a PO file.
.translated
This matches a translated message.
.untranslated
This matches an untranslated message (i.e. a message with empty translation).
.fuzzy
This matches a fuzzy message (i.e. a message which has a translation that needs review by the translator).
.obsolete
This matches an obsolete message (i.e. a message that was translated but is not needed by the current POT file any more).
• Selectors that apply to parts of a message in PO syntax. Recall the general structure of a message in PO syntax:
          white-space
#: reference...
#, flag...
#| msgid previous-untranslated-string
msgid untranslated-string
msgstr translated-string

.comment
.translator-comment
.extracted-comment
This matches the extracted comments, i.e. the comments placed by the programmer at the attention of the translator.
.reference-comment
This matches the source file reference comments (entire lines).
.reference
This matches the individual source file references inside the source file reference comment lines.
.flag-comment
This matches the flag comment lines (entire lines).
.flag
This matches the individual flags inside flag comment lines.
.fuzzy-flag
This matches the fuzzy' flag inside flag comment lines.
.previous-comment
This matches the comments containing the previous untranslated string (entire lines).
.previous
This matches the previous untranslated string including the string delimiters, the associated keywords (msgid etc.) and the spaces between them.
.msgid
This matches the untranslated string including the string delimiters, the associated keywords (msgid etc.) and the spaces between them.
.msgstr
This matches the translated string including the string delimiters, the associated keywords (msgstr etc.) and the spaces between them.
.keyword
This matches the keywords (msgid, msgstr, etc.).
.string
This matches strings, including the string delimiters (double quotes).
• Selectors that apply to parts of strings:
.text
This matches the entire contents of a string (excluding the string delimiters, i.e. the double quotes).
.escape-sequence
This matches an escape sequence (starting with a backslash).
.format-directive
This matches a format string directive (starting with a ‘%’ sign in the case of most programming languages, with a ‘{’ in the case of java-format and csharp-format, with a ‘~’ in the case of lisp-format and scheme-format, or with ‘$’ in the case of sh-format). .invalid-format-directive This matches an invalid format string directive. .added In an untranslated string, this matches a part of the string that was not present in the previous untranslated string. (Not yet implemented in this release.) .changed In an untranslated string or in a previous untranslated string, this matches a part of the string that is changed or replaced. (Not yet implemented in this release.) .removed In a previous untranslated string, this matches a part of the string that is not present in the current untranslated string. (Not yet implemented in this release.) These selectors can be combined to hierarchical selectors. For example,  .msgstr .invalid-format-directive { color: red; }  will highlight the invalid format directives in the translated strings. In text mode, pseudo-classes (CSS2 spec, section 5.11) and pseudo-elements (CSS2 spec, section 5.12) are not supported. The declarations in HTML mode are not limited; any graphical attribute supported by the browsers can be used. The declarations in text mode are limited to the following properties. Other properties will be silently ignored. color (CSS2 spec, section 14.1) background-color (CSS2 spec, section 14.2.1) These properties is supported. Colors will be adjusted to match the terminal's capabilities. Note that many terminals support only 8 colors. font-weight (CSS2 spec, section 15.2.3) This property is supported, but most terminals can only render two different weights: normal and bold. Values >= 600 are rendered as bold. font-style (CSS2 spec, section 15.2.3) This property is supported. The values italic and oblique are rendered the same way. text-decoration (CSS2 spec, section 16.3.1) This property is supported, limited to the values none and underline. Previous: Style rules, Up: Colorizing #### 9.11.5 Customizing less for viewing PO files The ‘less’ program is a popular text file browser for use in a text screen or terminal emulator. It also supports text with embedded escape sequences for colors and text decorations. You can use less to view a PO file like this (assuming an UTF-8 environment):  msgcat --to-code=UTF-8 --color xyz.po | less -R  You can simplify this to this simple command:  less xyz.po  after these three preparations: 1. Add the options ‘-R’ and ‘-f’ to the LESS environment variable. In sh shells: $ LESS="$LESS -R -f"$ export LESS

2. If your system does not already have the lessopen.sh and lessclose.sh scripts, create them and set the LESSOPEN and LESSCLOSE environment variables, as indicated in the manual page (‘man less’).
3. Add to lessopen.sh a piece of script that recognizes PO files through their file extension and invokes msgcat on them, producing a temporary file. Like this:
          case "$1" in *.po) tmpfile=mktemp "${TMPDIR-/tmp}/less.XXXXXX"
msgcat --to-code=UTF-8 --color "$1" > "$tmpfile"
echo "tmpfile" exit 0 ;; esac  Previous: Colorizing, Up: Manipulating ### 9.12 Writing your own programs that process PO files For the tasks for which a combination of ‘msgattrib’, ‘msgcat’ etc. is not sufficient, a set of C functions is provided in a library, to make it possible to process PO files in your own programs. When you use this library, you don't need to write routines to parse the PO file; instead, you retrieve a pointer in memory to each of messages contained in the PO file. Functions for writing PO files are not provided at this time. The functions are declared in the header file ‘<gettext-po.h>’, and are defined in a library called ‘libgettextpo’. — Data Type: po_file_t This is a pointer type that refers to the contents of a PO file, after it has been read into memory. — Data Type: po_message_iterator_t This is a pointer type that refers to an iterator that produces a sequence of messages. — Data Type: po_message_t This is a pointer type that refers to a message of a PO file, including its translation. — Function: po_file_t po_file_read (const char *filename) The po_file_read function reads a PO file into memory. The file name is given as argument. The return value is a handle to the PO file's contents, valid until po_file_free is called on it. In case of error, the return value is NULL, and errno is set. — Function: void po_file_free (po_file_t file) The po_file_free function frees a PO file's contents from memory, including all messages that are only implicitly accessible through iterators. — Function: const char * const * po_file_domains (po_file_t file) The po_file_domains function returns the domains for which the given PO file has messages. The return value is a NULL terminated array which is valid as long as the file handle is valid. For PO files which contain no ‘domain’ directive, the return value contains only one domain, namely the default domain "messages". — Function: po_message_iterator_t po_message_iterator (po_file_t file, const char *domain) The po_message_iterator returns an iterator that will produce the messages of file that belong to the given domain. If domain is NULL, the default domain is used instead. To list the messages, use the function po_next_message repeatedly. — Function: void po_message_iterator_free (po_message_iterator_t iterator) The po_message_iterator_free function frees an iterator previously allocated through the po_message_iterator function. — Function: po_message_t po_next_message (po_message_iterator_t iterator) The po_next_message function returns the next message from iterator and advances the iterator. It returns NULL when the iterator has reached the end of its message list. The following functions returns details of a po_message_t. Recall that the results are valid as long as the file handle is valid. — Function: const char * po_message_msgid (po_message_t message) The po_message_msgid function returns the msgid (untranslated English string) of a message. This is guaranteed to be non-NULL. — Function: const char * po_message_msgid_plural (po_message_t message) The po_message_msgid_plural function returns the msgid_plural (untranslated English plural string) of a message with plurals, or NULL for a message without plural. — Function: const char * po_message_msgstr (po_message_t message) The po_message_msgstr function returns the msgstr (translation) of a message. For an untranslated message, the return value is an empty string. — Function: const char * po_message_msgstr_plural (po_message_t message, int index) The po_message_msgstr_plural function returns the msgstr[index] of a message with plurals, or NULL when the index is out of range or for a message without plural. Here is an example code how these functions can be used.  const char *filename = ...; po_file_t file = po_file_read (filename); if (file == NULL) error (EXIT_FAILURE, errno, "couldn't open the PO file %s", filename); { const char * const *domains = po_file_domains (file); const char * const *domainp; for (domainp = domains; *domainp; domainp++) { const char *domain = *domainp; po_message_iterator_t iterator = po_message_iterator (file, domain); for (;;) { po_message_t *message = po_next_message (iterator); if (message == NULL) break; { const char *msgid = po_message_msgid (message); const char *msgstr = po_message_msgstr (message); ... } } po_message_iterator_free (iterator); } } po_file_free (file);  Next: , Previous: Manipulating, Up: Top ## 10 Producing Binary MO Files Next: , Previous: Binaries, Up: Binaries ### 10.1 Invoking the msgfmt Program  msgfmt [option] filename.po ...  The msgfmt programs generates a binary message catalog from a textual translation description. #### 10.1.1 Input file location filename.po ... -D directory --directory=directory Add directory to the list of directories. Source files are searched relative to this list of directories. The resulting .po file will be written relative to the current directory, though. If an input file is ‘-’, standard input is read. #### 10.1.2 Operation mode -j --java Java mode: generate a Java ResourceBundle class. --java2 Like –java, and assume Java2 (JDK 1.2 or higher). --csharp C# mode: generate a .NET .dll file containing a subclass of GettextResourceSet. --csharp-resources C# resources mode: generate a .NET .resources file. --tcl Tcl mode: generate a tcl/msgcat .msg file. --qt Qt mode: generate a Qt .qm file. #### 10.1.3 Output file location -o file --output-file=file Write output to specified file. --strict Direct the program to work strictly following the Uniforum/Sun implementation. Currently this only affects the naming of the output file. If this option is not given the name of the output file is the same as the domain name. If the strict Uniforum mode is enabled the suffix .mo is added to the file name if it is not already present. We find this behaviour of Sun's implementation rather silly and so by default this mode is not selected. If the output file is ‘-’, output is written to standard output. #### 10.1.4 Output file location in Java mode -r resource --resource=resource Specify the resource name. -l locale --locale=locale Specify the locale name, either a language specification of the form ll or a combined language and country specification of the form ll_CC. -d directory Specify the base directory of classes directory hierarchy. The class name is determined by appending the locale name to the resource name, separated with an underscore. The ‘-d’ option is mandatory. The class is written under the specified directory. #### 10.1.5 Output file location in C# mode -r resource --resource=resource Specify the resource name. -l locale --locale=locale Specify the locale name, either a language specification of the form ll or a combined language and country specification of the form ll_CC. -d directory Specify the base directory for locale dependent .dll files. The ‘-l’ and ‘-d’ options are mandatory. The .dll file is written in a subdirectory of the specified directory whose name depends on the locale. #### 10.1.6 Output file location in Tcl mode -l locale --locale=locale Specify the locale name, either a language specification of the form ll or a combined language and country specification of the form ll_CC. -d directory Specify the base directory of .msg message catalogs. The ‘-l’ and ‘-d’ options are mandatory. The .msg file is written in the specified directory. #### 10.1.7 Input file syntax -P --properties-input Assume the input files are Java ResourceBundles in Java .properties syntax, not in PO file syntax. --stringtable-input Assume the input files are NeXTstep/GNUstep localized resource files in .strings syntax, not in PO file syntax. #### 10.1.8 Input file interpretation -c --check Perform all the checks implied by --check-format, --check-header, --check-domain. --check-format Check language dependent format strings. If the string represents a format string used in a printf-like function both strings should have the same number of ‘%’ format specifiers, with matching types. If the flag c-format or possible-c-format appears in the special comment <#,> for this entry a check is performed. For example, the check will diagnose using ‘%.*s’ against ‘%s’, or ‘%d’ against ‘%s’, or ‘%d’ against ‘%x’. It can even handle positional parameters. Normally the xgettext program automatically decides whether a string is a format string or not. This algorithm is not perfect, though. It might regard a string as a format string though it is not used in a printf-like function and so msgfmt might report errors where there are none. To solve this problem the programmer can dictate the decision to the xgettext program (see c-format). The translator should not consider removing the flag from the <#,> line. This "fix" would be reversed again as soon as msgmerge is called the next time. --check-header Verify presence and contents of the header entry. See Header Entry, for a description of the various fields in the header entry. --check-domain Check for conflicts between domain directives and the --output-file option -C --check-compatibility Check that GNU msgfmt behaves like X/Open msgfmt. This will give an error when attempting to use the GNU extensions. --check-accelerators[=char] Check presence of keyboard accelerators for menu items. This is based on the convention used in some GUIs that a keyboard accelerator in a menu item string is designated by an immediately preceding ‘&’ character. Sometimes a keyboard accelerator is also called "keyboard mnemonic". This check verifies that if the untranslated string has exactly one ‘&’ character, the translated string has exactly one ‘&’ as well. If this option is given with a char argument, this char should be a non-alphanumeric character and is used as keyboard accelerator mark instead of ‘&’. -f --use-fuzzy Use fuzzy entries in output. Note that using this option is usually wrong, because fuzzy messages are exactly those which have not been validated by a human translator. #### 10.1.9 Output details -a number --alignment=number Align strings to number bytes (default: 1). --no-hash Don't include a hash table in the binary file. Lookup will be more expensive at run time (binary search instead of hash table lookup). #### 10.1.10 Informative output -h --help Display this help and exit. -V --version Output version information and exit. --statistics Print statistics about translations. -v --verbose Increase verbosity level. Next: , Previous: msgfmt Invocation, Up: Binaries ### 10.2 Invoking the msgunfmt Program  msgunfmt [option] [file]...  The msgunfmt program converts a binary message catalog to a Uniforum style .po file. #### 10.2.1 Operation mode -j --java Java mode: input is a Java ResourceBundle class. --csharp C# mode: input is a .NET .dll file containing a subclass of GettextResourceSet. --csharp-resources C# resources mode: input is a .NET .resources file. --tcl Tcl mode: input is a tcl/msgcat .msg file. #### 10.2.2 Input file location file ... Input .mo files. If no input file is given or if it is ‘-’, standard input is read. #### 10.2.3 Input file location in Java mode -r resource --resource=resource Specify the resource name. -l locale --locale=locale Specify the locale name, either a language specification of the form ll or a combined language and country specification of the form ll_CC. The class name is determined by appending the locale name to the resource name, separated with an underscore. The class is located using the CLASSPATH. #### 10.2.4 Input file location in C# mode -r resource --resource=resource Specify the resource name. -l locale --locale=locale Specify the locale name, either a language specification of the form ll or a combined language and country specification of the form ll_CC. -d directory Specify the base directory for locale dependent .dll files. The ‘-l’ and ‘-d’ options are mandatory. The .msg file is located in a subdirectory of the specified directory whose name depends on the locale. #### 10.2.5 Input file location in Tcl mode -l locale --locale=locale Specify the locale name, either a language specification of the form ll or a combined language and country specification of the form ll_CC. -d directory Specify the base directory of .msg message catalogs. The ‘-l’ and ‘-d’ options are mandatory. The .msg file is located in the specified directory. #### 10.2.6 Output file location -o file --output-file=file Write output to specified file. The results are written to standard output if no output file is specified or if it is ‘-’. #### 10.2.7 Output details --force-po Always write an output file even if it contains no message. -i --indent Write the .po file using indented style. --strict Write out a strict Uniforum conforming PO file. Note that this Uniforum format should be avoided because it doesn't support the GNU extensions. -p --properties-output Write out a Java ResourceBundle in Java .properties syntax. Note that this file format doesn't support plural forms and silently drops obsolete messages. --stringtable-output Write out a NeXTstep/GNUstep localized resource file in .strings syntax. Note that this file format doesn't support plural forms. -w number --width=number Set the output page width. Long strings in the output files will be split across multiple lines in order to ensure that each line's width (= number of screen columns) is less or equal to the given number. --no-wrap Do not break long message lines. Message lines whose width exceeds the output page width will not be split into several lines. Only file reference lines which are wider than the output page width will be split. -s --sort-output Generate sorted output. Note that using this option makes it much harder for the translator to understand each message's context. #### 10.2.8 Informative output -h --help Display this help and exit. -V --version Output version information and exit. -v --verbose Increase verbosity level. Previous: msgunfmt Invocation, Up: Binaries ### 10.3 The Format of GNU MO Files The format of the generated MO files is best described by a picture, which appears below. The first two words serve the identification of the file. The magic number will always signal GNU MO files. The number is stored in the byte order of the generating machine, so the magic number really is two numbers: 0x950412de and 0xde120495. The second word describes the current revision of the file format. For now the revision is 0. This might change in future versions, and ensures that the readers of MO files can distinguish new formats from old ones, so that both can be handled correctly. The version is kept separate from the magic number, instead of using different magic numbers for different formats, mainly because /etc/magic is not updated often. It might be better to have magic separated from internal format version identification. Follow a number of pointers to later tables in the file, allowing for the extension of the prefix part of MO files without having to recompile programs reading them. This might become useful for later inserting a few flag bits, indication about the charset used, new tables, or other things. Then, at offset O and offset T in the picture, two tables of string descriptors can be found. In both tables, each string descriptor uses two 32 bits integers, one for the string length, another for the offset of the string in the MO file, counting in bytes from the start of the file. The first table contains descriptors for the original strings, and is sorted so the original strings are in increasing lexicographical order. The second table contains descriptors for the translated strings, and is parallel to the first table: to find the corresponding translation one has to access the array slot in the second array with the same index. Having the original strings sorted enables the use of simple binary search, for when the MO file does not contain an hashing table, or for when it is not practical to use the hashing table provided in the MO file. This also has another advantage, as the empty string in a PO file GNU gettext is usually translated into some system information attached to that particular MO file, and the empty string necessarily becomes the first in both the original and translated tables, making the system information very easy to find. The size S of the hash table can be zero. In this case, the hash table itself is not contained in the MO file. Some people might prefer this because a precomputed hashing table takes disk space, and does not win that much speed. The hash table contains indices to the sorted array of strings in the MO file. Conflict resolution is done by double hashing. The precise hashing algorithm used is fairly dependent on GNU gettext code, and is not documented here. As for the strings themselves, they follow the hash file, and each is terminated with a <NUL>, and this <NUL> is not counted in the length which appears in the string descriptor. The msgfmt program has an option selecting the alignment for MO file strings. With this option, each string is separately aligned so it starts at an offset which is a multiple of the alignment value. On some RISC machines, a correct alignment will speed things up. Contexts are stored by storing the concatenation of the context, a <EOT> byte, and the original string, instead of the original string. Plural forms are stored by letting the plural of the original string follow the singular of the original string, separated through a <NUL> byte. The length which appears in the string descriptor includes both. However, only the singular of the original string takes part in the hash table lookup. The plural variants of the translation are all stored consecutively, separated through a <NUL> byte. Here also, the length in the string descriptor includes all of them. Nothing prevents a MO file from having embedded <NUL>s in strings. However, the program interface currently used already presumes that strings are <NUL> terminated, so embedded <NUL>s are somewhat useless. But the MO file format is general enough so other interfaces would be later possible, if for example, we ever want to implement wide characters right in MO files, where <NUL> bytes may accidentally appear. (No, we don't want to have wide characters in MO files. They would make the file unnecessarily large, and the ‘wchar_t’ type being platform dependent, MO files would be platform dependent as well.) This particular issue has been strongly debated in the GNU gettext development forum, and it is expectable that MO file format will evolve or change over time. It is even possible that many formats may later be supported concurrently. But surely, we have to start somewhere, and the MO file format described here is a good start. Nothing is cast in concrete, and the format may later evolve fairly easily, so we should feel comfortable with the current approach.  byte +------------------------------------------+ 0 | magic number = 0x950412de | | | 4 | file format revision = 0 | | | 8 | number of strings | == N | | 12 | offset of table with original strings | == O | | 16 | offset of table with translation strings | == T | | 20 | size of hashing table | == S | | 24 | offset of hashing table | == H | | . . . (possibly more entries later) . . . | | O | length & offset 0th string ----------------. O + 8 | length & offset 1st string ------------------. ... ... | | O + ((N-1)*8)| length & offset (N-1)th string | | | | | | | T | length & offset 0th translation ---------------. T + 8 | length & offset 1st translation -----------------. ... ... | | | | T + ((N-1)*8)| length & offset (N-1)th translation | | | | | | | | | | | H | start hash table | | | | | ... ... | | | | H + S * 4 | end hash table | | | | | | | | | | | | NUL terminated 0th string <----------------' | | | | | | | | | NUL terminated 1st string <------------------' | | | | | | ... ... | | | | | | | NUL terminated 0th translation <---------------' | | | | | NUL terminated 1st translation <-----------------' | | ... ... | | +------------------------------------------+  Next: , Previous: Binaries, Up: Top ## 11 The Programmer's View One aim of the current message catalog implementation provided by GNU gettext was to use the system's message catalog handling, if the installer wishes to do so. So we perhaps should first take a look at the solutions we know about. The people in the POSIX committee did not manage to agree on one of the semi-official standards which we'll describe below. In fact they couldn't agree on anything, so they decided only to include an example of an interface. The major Unix vendors are split in the usage of the two most important specifications: X/Open's catgets vs. Uniforum's gettext interface. We'll describe them both and later explain our solution of this dilemma. Next: , Previous: Programmers, Up: Programmers ### 11.1 About catgets The catgets implementation is defined in the X/Open Portability Guide, Volume 3, XSI Supplementary Definitions, Chapter 5. But the process of creating this standard seemed to be too slow for some of the Unix vendors so they created their implementations on preliminary versions of the standard. Of course this leads again to problems while writing platform independent programs: even the usage of catgets does not guarantee a unique interface. Another, personal comment on this that only a bunch of committee members could have made this interface. They never really tried to program using this interface. It is a fast, memory-saving implementation, an user can happily live with it. But programmers hate it (at least I and some others do...) But we must not forget one point: after all the trouble with transferring the rights on Unix(tm) they at last came to X/Open, the very same who published this specification. This leads me to making the prediction that this interface will be in future Unix standards (e.g. Spec1170) and therefore part of all Unix implementation (implementations, which are allowed to wear this name). Next: , Previous: catgets, Up: catgets #### 11.1.1 The Interface The interface to the catgets implementation consists of three functions which correspond to those used in file access: catopen to open the catalog for using, catgets for accessing the message tables, and catclose for closing after work is done. Prototypes for the functions and the needed definitions are in the <nl_types.h> header file. catopen is used like in this:  nl_catd catd = catopen ("catalog_name", 0);  The function takes as the argument the name of the catalog. This usual refers to the name of the program or the package. The second parameter is not further specified in the standard. I don't even know whether it is implemented consistently among various systems. So the common advice is to use 0 as the value. The return value is a handle to the message catalog, equivalent to handles to file returned by open. This handle is of course used in the catgets function which can be used like this:  char *translation = catgets (catd, set_no, msg_id, "original string");  The first parameter is this catalog descriptor. The second parameter specifies the set of messages in this catalog, in which the message described by msg_id is obtained. catgets therefore uses a three-stage addressing:  catalog name set number message ID translation  The fourth argument is not used to address the translation. It is given as a default value in case when one of the addressing stages fail. One important thing to remember is that although the return type of catgets is char * the resulting string must not be changed. It should better be const char *, but the standard is published in 1988, one year before ANSI C. The last of these functions is used and behaves as expected:  catclose (catd);  After this no catgets call using the descriptor is legal anymore. Previous: Interface to catgets, Up: catgets #### 11.1.2 Problems with the catgets Interface?! Now that this description seemed to be really easy — where are the problems we speak of? In fact the interface could be used in a reasonable way, but constructing the message catalogs is a pain. The reason for this lies in the third argument of catgets: the unique message ID. This has to be a numeric value for all messages in a single set. Perhaps you could imagine the problems keeping such a list while changing the source code. Add a new message here, remove one there. Of course there have been developed a lot of tools helping to organize this chaos but one as the other fails in one aspect or the other. We don't want to say that the other approach has no problems but they are far more easy to manage. Next: , Previous: catgets, Up: Programmers ### 11.2 About gettext The definition of the gettext interface comes from a Uniforum proposal. It was submitted there by Sun, who had implemented the gettext function in SunOS 4, around 1990. Nowadays, the gettext interface is specified by the OpenI18N standard. The main point about this solution is that it does not follow the method of normal file handling (open-use-close) and that it does not burden the programmer with so many tasks, especially the unique key handling. Of course here also a unique key is needed, but this key is the message itself (how long or short it is). See Comparison for a more detailed comparison of the two methods. The following section contains a rather detailed description of the interface. We make it that detailed because this is the interface we chose for the GNU gettext Library. Programmers interested in using this library will be interested in this description. Next: , Previous: gettext, Up: gettext #### 11.2.1 The Interface The minimal functionality an interface must have is a) to select a domain the strings are coming from (a single domain for all programs is not reasonable because its construction and maintenance is difficult, perhaps impossible) and b) to access a string in a selected domain. This is principally the description of the gettext interface. It has a global domain which unqualified usages reference. Of course this domain is selectable by the user.  char *textdomain (const char *domain_name);  This provides the possibility to change or query the current status of the current global domain of the LC_MESSAGE category. The argument is a null-terminated string, whose characters must be legal in the use in filenames. If the domain_name argument is NULL, the function returns the current value. If no value has been set before, the name of the default domain is returned: messages. Please note that although the return value of textdomain is of type char * no changing is allowed. It is also important to know that no checks of the availability are made. If the name is not available you will see this by the fact that no translations are provided. To use a domain set by textdomain the function  char *gettext (const char *msgid);  is to be used. This is the simplest reasonable form one can imagine. The translation of the string msgid is returned if it is available in the current domain. If it is not available, the argument itself is returned. If the argument is NULL the result is undefined. One thing which should come into mind is that no explicit dependency to the used domain is given. The current value of the domain is used. If this changes between two executions of the same gettext call in the program, both calls reference a different message catalog. For the easiest case, which is normally used in internationalized packages, once at the beginning of execution a call to textdomain is issued, setting the domain to a unique name, normally the package name. In the following code all strings which have to be translated are filtered through the gettext function. That's all, the package speaks your language. Next: , Previous: Interface to gettext, Up: gettext #### 11.2.2 Solving Ambiguities While this single name domain works well for most applications there might be the need to get translations from more than one domain. Of course one could switch between different domains with calls to textdomain, but this is really not convenient nor is it fast. A possible situation could be one case subject to discussion during this writing: all error messages of functions in the set of common used functions should go into a separate domain error. By this mean we would only need to translate them once. Another case are messages from a library, as these have to be independent of the current domain set by the application. For this reasons there are two more functions to retrieve strings:  char *dgettext (const char *domain_name, const char *msgid); char *dcgettext (const char *domain_name, const char *msgid, int category);  Both take an additional argument at the first place, which corresponds to the argument of textdomain. The third argument of dcgettext allows to use another locale category but LC_MESSAGES. But I really don't know where this can be useful. If the domain_name is NULL or category has an value beside the known ones, the result is undefined. It should also be noted that this function is not part of the second known implementation of this function family, the one found in Solaris. A second ambiguity can arise by the fact, that perhaps more than one domain has the same name. This can be solved by specifying where the needed message catalog files can be found.  char *bindtextdomain (const char *domain_name, const char *dir_name);  Calling this function binds the given domain to a file in the specified directory (how this file is determined follows below). Especially a file in the systems default place is not favored against the specified file anymore (as it would be by solely using textdomain). A NULL pointer for the dir_name parameter returns the binding associated with domain_name. If domain_name itself is NULL nothing happens and a NULL pointer is returned. Here again as for all the other functions is true that none of the return value must be changed! It is important to remember that relative path names for the dir_name parameter can be trouble. Since the path is always computed relative to the current directory different results will be achieved when the program executes a chdir command. Relative paths should always be avoided to avoid dependencies and unreliabilities. Next: , Previous: Ambiguities, Up: gettext #### 11.2.3 Locating Message Catalog Files Because many different languages for many different packages have to be stored we need some way to add these information to file message catalog files. The way usually used in Unix environments is have this encoding in the file name. This is also done here. The directory name given in bindtextdomains second argument (or the default directory), followed by the name of the locale, the locale category, and the domain name are concatenated:  dir_name/locale/LC_category/domain_name.mo  The default value for dir_name is system specific. For the GNU library, and for packages adhering to its conventions, it's:  /usr/local/share/locale  locale is the name of the locale category which is designated by LC_category. For gettext and dgettext this LC_category is always LC_MESSAGES.3 The name of the locale category is determined through setlocale (LC_category, NULL). 4 When using the function dcgettext, you can specify the locale category through the third argument. Next: , Previous: Locating Catalogs, Up: gettext #### 11.2.4 How to specify the output character set gettext uses gettext not only looks up a translation in a message catalog. It also converts the translation on the fly to the desired output character set. This is useful if the user is working in a different character set than the translator who created the message catalog, because it avoids distributing variants of message catalogs which differ only in the character set. The output character set is, by default, the value of nl_langinfo (CODESET), which depends on the LC_CTYPE part of the current locale. But programs which store strings in a locale independent way (e.g. UTF-8) can request that gettext and related functions return the translations in that encoding, by use of the bind_textdomain_codeset function. Note that the msgid argument to gettext is not subject to character set conversion. Also, when gettext does not find a translation for msgid, it returns msgid unchanged – independently of the current output character set. It is therefore recommended that all msgids be US-ASCII strings. — Function: char * bind_textdomain_codeset (const char *domainname, const char *codeset) The bind_textdomain_codeset function can be used to specify the output character set for message catalogs for domain domainname. The codeset argument must be a valid codeset name which can be used for the iconv_open function, or a null pointer. If the codeset parameter is the null pointer, bind_textdomain_codeset returns the currently selected codeset for the domain with the name domainname. It returns NULL if no codeset has yet been selected. The bind_textdomain_codeset function can be used several times. If used multiple times with the same domainname argument, the later call overrides the settings made by the earlier one. The bind_textdomain_codeset function returns a pointer to a string containing the name of the selected codeset. The string is allocated internally in the function and must not be changed by the user. If the system went out of core during the execution of bind_textdomain_codeset, the return value is NULL and the global variable errno is set accordingly. Next: , Previous: Charset conversion, Up: gettext #### 11.2.5 Using contexts for solving ambiguities One place where the gettext functions, if used normally, have big problems is within programs with graphical user interfaces (GUIs). The problem is that many of the strings which have to be translated are very short. They have to appear in pull-down menus which restricts the length. But strings which are not containing entire sentences or at least large fragments of a sentence may appear in more than one situation in the program but might have different translations. This is especially true for the one-word strings which are frequently used in GUI programs. As a consequence many people say that the gettext approach is wrong and instead catgets should be used which indeed does not have this problem. But there is a very simple and powerful method to handle this kind of problems with the gettext functions. Contexts can be added to strings to be translated. A context dependent translation lookup is when a translation for a given string is searched, that is limited to a given context. The translation for the same string in a different context can be different. The different translations of the same string in different contexts can be stored in the in the same MO file, and can be edited by the translator in the same PO file. The gettext.h include file contains the lookup macros for strings with contexts. They are implemented as thin macros and inline functions over the functions from <libintl.h>.  const char *pgettext (const char *msgctxt, const char *msgid);  In a call of this macro, msgctxt and msgid must be string literals. The macro returns the translation of msgid, restricted to the context given by msgctxt. The msgctxt string is visible in the PO file to the translator. You should try to make it somehow canonical and never changing. Because every time you change an msgctxt, the translator will have to review the translation of msgid. Finding a canonical msgctxt string that doesn't change over time can be hard. But you shouldn't use the file name or class name containing the pgettext call – because it is a common development task to rename a file or a class, and it shouldn't cause translator work. Also you shouldn't use a comment in the form of a complete English sentence as msgctxt – because orthography or grammar changes are often applied to such sentences, and again, it shouldn't force the translator to do a review. The ‘p’ in ‘pgettext’ stands for “particular”: pgettext fetches a particular translation of the msgid.  const char *dpgettext (const char *domain_name, const char *msgctxt, const char *msgid); const char *dcpgettext (const char *domain_name, const char *msgctxt, const char *msgid, int category);  These are generalizations of pgettext. They behave similarly to dgettext and dcgettext, respectively. The domain_name argument defines the translation domain. The category argument allows to use another locale category than LC_MESSAGES. As as example consider the following fictional situation. A GUI program has a menu bar with the following entries:  +------------+------------+--------------------------------------+ | File | Printer | | +------------+------------+--------------------------------------+ | Open | | Select | | New | | Open | +----------+ | Connect | +----------+  To have the strings File, Printer, Open, New, Select, and Connect translated there has to be at some point in the code a call to a function of the gettext family. But in two places the string passed into the function would be Open. The translations might not be the same and therefore we are in the dilemma described above. What distinguishes the two places is the menu path from the menu root to the particular menu entries:  Menu|File Menu|Printer Menu|File|Open Menu|File|New Menu|Printer|Select Menu|Printer|Open Menu|Printer|Connect  The context is thus the menu path without its last part. So, the calls look like this:  pgettext ("Menu|", "File") pgettext ("Menu|", "Printer") pgettext ("Menu|File|", "Open") pgettext ("Menu|File|", "New") pgettext ("Menu|Printer|", "Select") pgettext ("Menu|Printer|", "Open") pgettext ("Menu|Printer|", "Connect")  Whether or not to use the ‘|’ character at the end of the context is a matter of style. For more complex cases, where the msgctxt or msgid are not string literals, more general macros are available:  const char *pgettext_expr (const char *msgctxt, const char *msgid); const char *dpgettext_expr (const char *domain_name, const char *msgctxt, const char *msgid); const char *dcpgettext_expr (const char *domain_name, const char *msgctxt, const char *msgid, int category);  Here msgctxt and msgid can be arbitrary string-valued expressions. These macros are more general. But in the case that both argument expressions are string literals, the macros without the ‘_expr’ suffix are more efficient. Next: , Previous: Contexts, Up: gettext #### 11.2.6 Additional functions for plural forms The functions of the gettext family described so far (and all the catgets functions as well) have one problem in the real world which have been neglected completely in all existing approaches. What is meant here is the handling of plural forms. Looking through Unix source code before the time anybody thought about internationalization (and, sadly, even afterwards) one can often find code similar to the following:  printf ("%d file%s deleted", n, n == 1 ? "" : "s");  After the first complaints from people internationalizing the code people either completely avoided formulations like this or used strings like "file(s)". Both look unnatural and should be avoided. First tries to solve the problem correctly looked like this:  if (n == 1) printf ("%d file deleted", n); else printf ("%d files deleted", n);  But this does not solve the problem. It helps languages where the plural form of a noun is not simply constructed by adding an ‘s’ but that is all. Once again people fell into the trap of believing the rules their language is using are universal. But the handling of plural forms differs widely between the language families. For example, Rafal Maszkowski <rzm@mat.uni.torun.pl> reports: In Polish we use e.g. plik (file) this way:  1 plik 2,3,4 pliki 5-21 pliko'w 22-24 pliki 25-31 pliko'w  and so on (o' means 8859-2 oacute which should be rather okreska, similar to aogonek). There are two things which can differ between languages (and even inside language families); • The form how plural forms are built differs. This is a problem with languages which have many irregularities. German, for instance, is a drastic case. Though English and German are part of the same language family (Germanic), the almost regular forming of plural noun forms (appending an ‘s’) is hardly found in German. • The number of plural forms differ. This is somewhat surprising for those who only have experiences with Romanic and Germanic languages since here the number is the same (there are two). But other language families have only one form or many forms. More information on this in an extra section. The consequence of this is that application writers should not try to solve the problem in their code. This would be localization since it is only usable for certain, hardcoded language environments. Instead the extended gettext interface should be used. These extra functions are taking instead of the one key string two strings and a numerical argument. The idea behind this is that using the numerical argument and the first string as a key, the implementation can select using rules specified by the translator the right plural form. The two string arguments then will be used to provide a return value in case no message catalog is found (similar to the normal gettext behavior). In this case the rules for Germanic language is used and it is assumed that the first string argument is the singular form, the second the plural form. This has the consequence that programs without language catalogs can display the correct strings only if the program itself is written using a Germanic language. This is a limitation but since the GNU C library (as well as the GNU gettext package) are written as part of the GNU package and the coding standards for the GNU project require program being written in English, this solution nevertheless fulfills its purpose. — Function: char * ngettext (const char *msgid1, const char *msgid2, unsigned long int n) The ngettext function is similar to the gettext function as it finds the message catalogs in the same way. But it takes two extra arguments. The msgid1 parameter must contain the singular form of the string to be converted. It is also used as the key for the search in the catalog. The msgid2 parameter is the plural form. The parameter n is used to determine the plural form. If no message catalog is found msgid1 is returned if n == 1, otherwise msgid2. An example for the use of this function is:  printf (ngettext ("%d file removed", "%d files removed", n), n);  Please note that the numeric value n has to be passed to the printf function as well. It is not sufficient to pass it only to ngettext. In the English singular case, the number – always 1 – can be replaced with "one":  printf (ngettext ("One file removed", "%d files removed", n), n);  This works because the ‘printf’ function discards excess arguments that are not consumed by the format string. It is also possible to use this function when the strings don't contain a cardinal number:  puts (ngettext ("Delete the selected file?", "Delete the selected files?", n));  In this case the number n is only used to choose the plural form. — Function: char * dngettext (const char *domain, const char *msgid1, const char *msgid2, unsigned long int n) The dngettext is similar to the dgettext function in the way the message catalog is selected. The difference is that it takes two extra parameter to provide the correct plural form. These two parameters are handled in the same way ngettext handles them. — Function: char * dcngettext (const char *domain, const char *msgid1, const char *msgid2, unsigned long int n, int category) The dcngettext is similar to the dcgettext function in the way the message catalog is selected. The difference is that it takes two extra parameter to provide the correct plural form. These two parameters are handled in the same way ngettext handles them. Now, how do these functions solve the problem of the plural forms? Without the input of linguists (which was not available) it was not possible to determine whether there are only a few different forms in which plural forms are formed or whether the number can increase with every new supported language. Therefore the solution implemented is to allow the translator to specify the rules of how to select the plural form. Since the formula varies with every language this is the only viable solution except for hardcoding the information in the code (which still would require the possibility of extensions to not prevent the use of new languages). The information about the plural form selection has to be stored in the header entry of the PO file (the one with the empty msgid string). The plural form information looks like this:  Plural-Forms: nplurals=2; plural=n == 1 ? 0 : 1;  The nplurals value must be a decimal number which specifies how many different plural forms exist for this language. The string following plural is an expression which is using the C language syntax. Exceptions are that no negative numbers are allowed, numbers must be decimal, and the only variable allowed is n. Spaces are allowed in the expression, but backslash-newlines are not; in the examples below the backslash-newlines are present for formatting purposes only. This expression will be evaluated whenever one of the functions ngettext, dngettext, or dcngettext is called. The numeric value passed to these functions is then substituted for all uses of the variable n in the expression. The resulting value then must be greater or equal to zero and smaller than the value given as the value of nplurals. The following rules are known at this point. The language with families are listed. But this does not necessarily mean the information can be generalized for the whole family (as can be easily seen in the table below).5 Only one form: Some languages only require one single form. There is no distinction between the singular and plural form. An appropriate header entry would look like this:  Plural-Forms: nplurals=1; plural=0;  Languages with this property include: Asian family Japanese, Korean, Vietnamese Turkic/Altaic family Turkish Two forms, singular used for one only This is the form used in most existing programs since it is what English is using. A header entry would look like this:  Plural-Forms: nplurals=2; plural=n != 1;  (Note: this uses the feature of C expressions that boolean expressions have to value zero or one.) Languages with this property include: Germanic family Danish, Dutch, English, Faroese, German, Norwegian, Swedish Finno-Ugric family Estonian, Finnish Latin/Greek family Greek Semitic family Hebrew Romanic family Italian, Portuguese, Spanish Artificial Esperanto Another language using the same header entry is: Finno-Ugric family Hungarian Hungarian does not appear to have a plural if you look at sentences involving cardinal numbers. For example, “1 apple” is “1 alma”, and “123 apples” is “123 alma”. But when the number is not explicit, the distinction between singular and plural exists: “the apple” is “az alma”, and “the apples” is “az almák”. Since ngettext has to support both types of sentences, it is classified here, under “two forms”. Two forms, singular used for zero and one Exceptional case in the language family. The header entry would be:  Plural-Forms: nplurals=2; plural=n>1;  Languages with this property include: Romanic family French, Brazilian Portuguese Three forms, special case for zero The header entry would be:  Plural-Forms: nplurals=3; plural=n%10==1 && n%100!=11 ? 0 : n != 0 ? 1 : 2;  Languages with this property include: Baltic family Latvian Three forms, special cases for one and two The header entry would be:  Plural-Forms: nplurals=3; plural=n==1 ? 0 : n==2 ? 1 : 2;  Languages with this property include: Celtic Gaeilge (Irish) Three forms, special case for numbers ending in 00 or [2-9][0-9] The header entry would be:  Plural-Forms: nplurals=3; \ plural=n==1 ? 0 : (n==0 || (n%100 > 0 && n%100 < 20)) ? 1 : 2;  Languages with this property include: Romanic family Romanian Three forms, special case for numbers ending in 1[2-9] The header entry would look like this:  Plural-Forms: nplurals=3; \ plural=n%10==1 && n%100!=11 ? 0 : \ n%10>=2 && (n%100<10 || n%100>=20) ? 1 : 2;  Languages with this property include: Baltic family Lithuanian Three forms, special cases for numbers ending in 1 and 2, 3, 4, except those ending in 1[1-4] The header entry would look like this:  Plural-Forms: nplurals=3; \ plural=n%10==1 && n%100!=11 ? 0 : \ n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2;  Languages with this property include: Slavic family Croatian, Serbian, Russian, Ukrainian Three forms, special cases for 1 and 2, 3, 4 The header entry would look like this:  Plural-Forms: nplurals=3; \ plural=(n==1) ? 0 : (n>=2 && n<=4) ? 1 : 2;  Languages with this property include: Slavic family Slovak, Czech Three forms, special case for one and some numbers ending in 2, 3, or 4 The header entry would look like this:  Plural-Forms: nplurals=3; \ plural=n==1 ? 0 : \ n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2;  Languages with this property include: Slavic family Polish Four forms, special case for one and all numbers ending in 02, 03, or 04 The header entry would look like this:  Plural-Forms: nplurals=4; \ plural=n%100==1 ? 0 : n%100==2 ? 1 : n%100==3 || n%100==4 ? 2 : 3;  Languages with this property include: Slavic family Slovenian You might now ask, ngettext handles only numbers n of type ‘unsigned long’. What about larger integer types? What about negative numbers? What about floating-point numbers? About larger integer types, such as ‘uintmax_t’ or ‘unsigned long long’: they can be handled by reducing the value to a range that fits in an ‘unsigned long’. Simply casting the value to ‘unsigned long’ would not do the right thing, since it would treat ULONG_MAX + 1 like zero, ULONG_MAX + 2 like singular, and the like. Here you can exploit the fact that all mentioned plural form formulas eventually become periodic, with a period that is a divisor of 100 (or 1000 or 1000000). So, when you reduce a large value to another one in the range [1000000, 1999999] that ends in the same 6 decimal digits, you can assume that it will lead to the same plural form selection. This code does this:  #include <inttypes.h> uintmax_t nbytes = ...; printf (ngettext ("The file has %"PRIuMAX" byte.", "The file has %"PRIuMAX" bytes.", (nbytes > ULONG_MAX ? (nbytes % 1000000) + 1000000 : nbytes)), nbytes);  Negative and floating-point values usually represent physical entities for which singular and plural don't clearly apply. In such cases, there is no need to use ngettext; a simple gettext call with a form suitable for all values will do. For example:  printf (gettext ("Time elapsed: %.3f seconds"), num_milliseconds * 0.001);  Even if num_milliseconds happens to be a multiple of 1000, the output  Time elapsed: 1.000 seconds  is acceptable in English, and similarly for other languages. Previous: Plural forms, Up: gettext #### 11.2.7 Optimization of the *gettext functions At this point of the discussion we should talk about an advantage of the GNU gettext implementation. Some readers might have pointed out that an internationalized program might have a poor performance if some string has to be translated in an inner loop. While this is unavoidable when the string varies from one run of the loop to the other it is simply a waste of time when the string is always the same. Take the following example:  { while (...) { puts (gettext ("Hello world")); } }  When the locale selection does not change between two runs the resulting string is always the same. One way to use this is:  { str = gettext ("Hello world"); while (...) { puts (str); } }  But this solution is not usable in all situation (e.g. when the locale selection changes) nor does it lead to legible code. For this reason, GNU gettext caches previous translation results. When the same translation is requested twice, with no new message catalogs being loaded in between, gettext will, the second time, find the result through a single cache lookup. Next: , Previous: gettext, Up: Programmers ### 11.3 Comparing the Two Interfaces The following discussion is perhaps a little bit colored. As said above we implemented GNU gettext following the Uniforum proposal and this surely has its reasons. But it should show how we came to this decision. First we take a look at the developing process. When we write an application using NLS provided by gettext we proceed as always. Only when we come to a string which might be seen by the users and thus has to be translated we use gettext("...") instead of "...". At the beginning of each source file (or in a central header file) we define  #define gettext(String) (String)  Even this definition can be avoided when the system supports the gettext function in its C library. When we compile this code the result is the same as if no NLS code is used. When you take a look at the GNU gettext code you will see that we use _("...") instead of gettext("..."). This reduces the number of additional characters per translatable string to 3 (in words: three). When now a production version of the program is needed we simply replace the definition  #define _(String) (String)  by  #include <libintl.h> #define _(String) gettext (String)  Additionally we run the program xgettext on all source code file which contain translatable strings and that's it: we have a running program which does not depend on translations to be available, but which can use any that becomes available. The same procedure can be done for the gettext_noop invocations (see Special cases). One usually defines gettext_noop as a no-op macro. So you should consider the following code for your project:  #define gettext_noop(String) String #define N_(String) gettext_noop (String)  N_ is a short form similar to _. The Makefile in the po/ directory of GNU gettext knows by default both of the mentioned short forms so you are invited to follow this proposal for your own ease. Now to catgets. The main problem is the work for the programmer. Every time he comes to a translatable string he has to define a number (or a symbolic constant) which has also be defined in the message catalog file. He also has to take care for duplicate entries, duplicate message IDs etc. If he wants to have the same quality in the message catalog as the GNU gettext program provides he also has to put the descriptive comments for the strings and the location in all source code files in the message catalog. This is nearly a Mission: Impossible. But there are also some points people might call advantages speaking for catgets. If you have a single word in a string and this string is used in different contexts it is likely that in one or the other language the word has different translations. Example:  printf ("%s: %d", gettext ("number"), number_of_errors) printf ("you should see %d %s", number_count, number_count == 1 ? gettext ("number") : gettext ("numbers"))  Here we have to translate two times the string "number". Even if you do not speak a language beside English it might be possible to recognize that the two words have a different meaning. In German the first appearance has to be translated to "Anzahl" and the second to "Zahl". Now you can say that this example is really esoteric. And you are right! This is exactly how we felt about this problem and decide that it does not weight that much. The solution for the above problem could be very easy:  printf ("%s %d", gettext ("number:"), number_of_errors) printf (number_count == 1 ? gettext ("you should see %d number") : gettext ("you should see %d numbers"), number_count)  We believe that we can solve all conflicts with this method. If it is difficult one can also consider changing one of the conflicting string a little bit. But it is not impossible to overcome. catgets allows same original entry to have different translations, but gettext has another, scalable approach for solving ambiguities of this kind: See Ambiguities. Next: , Previous: Comparison, Up: Programmers ### 11.4 Using libintl.a in own programs Starting with version 0.9.4 the library libintl.h should be self-contained. I.e., you can use it in your own programs without providing additional functions. The Makefile will put the header and the library in directories selected using the (prefix).

Next: , Previous: Using libintl.a, Up: Programmers

### 11.5 Being a gettext grok

NOTE: This documentation section is outdated and needs to be revised.

To fully exploit the functionality of the GNU gettext library it is surely helpful to read the source code. But for those who don't want to spend that much time in reading the (sometimes complicated) code here is a list comments:

• Changing the language at runtime For interactive programs it might be useful to offer a selection of the used language at runtime. To understand how to do this one need to know how the used language is determined while executing the gettext function. The method which is presented here only works correctly with the GNU implementation of the gettext functions.

In the function dcgettext at every call the current setting of the highest priority environment variable is determined and used. Highest priority means here the following list with decreasing priority:

1. LANGUAGE
2. LC_ALL
3. LC_xxx, according to selected locale category
4. LANG

Afterwards the path is constructed using the found value and the translation file is loaded if available.

What happens now when the value for, say, LANGUAGE changes? According to the process explained above the new value of this variable is found as soon as the dcgettext function is called. But this also means the (perhaps) different message catalog file is loaded. In other words: the used language is changed.

But there is one little hook. The code for gcc-2.7.0 and up provides some optimization. This optimization normally prevents the calling of the dcgettext function as long as no new catalog is loaded. But if dcgettext is not called the program also cannot find the LANGUAGE variable be changed (see Optimized gettext). A solution for this is very easy. Include the following code in the language switching function.

            /* Change language.  */
setenv ("LANGUAGE", "fr", 1);

/* Make change known.  */
{
extern int  _nl_msg_cat_cntr;
++_nl_msg_cat_cntr;
}


The variable _nl_msg_cat_cntr is defined in loadmsgcat.c. You don't need to know what this is for. But it can be used to detect whether a gettext implementation is GNU gettext and not non-GNU system's native gettext implementation.

Previous: gettext grok, Up: Programmers

### 11.6 Temporary Notes for the Programmers Chapter

NOTE: This documentation section is outdated and needs to be revised.

Next: , Previous: Temp Programmers, Up: Temp Programmers

#### 11.6.1 Temporary - Two Possible Implementations

There are two competing methods for language independent messages: the X/Open catgets method, and the Uniforum gettext method. The catgets method indexes messages by integers; the gettext method indexes them by their English translations. The catgets method has been around longer and is supported by more vendors. The gettext method is supported by Sun, and it has been heard that the COSE multi-vendor initiative is supporting it. Neither method is a POSIX standard; the POSIX.1 committee had a lot of disagreement in this area.

Neither one is in the POSIX standard. There was much disagreement in the POSIX.1 committee about using the gettext routines vs. catgets (XPG). In the end the committee couldn't agree on anything, so no messaging system was included as part of the standard. I believe the informative annex of the standard includes the XPG3 messaging interfaces, “...as an example of a messaging system that has been implemented...

They were very careful not to say anywhere that you should use one set of interfaces over the other. For more on this topic please see the Programming for Internationalization FAQ.

Next: , Previous: Temp Implementations, Up: Temp Programmers

#### 11.6.2 Temporary - About catgets

There have been a few discussions of late on the use of catgets as a base. I think it important to present both sides of the argument and hence am opting to play devil's advocate for a little bit.

I'll not deny the fact that catgets could have been designed a lot better. It currently has quite a number of limitations and these have already been pointed out.

However there is a great deal to be said for consistency and standardization. A common recurring problem when writing Unix software is the myriad portability problems across Unix platforms. It seems as if every Unix vendor had a look at the operating system and found parts they could improve upon. Undoubtedly, these modifications are probably innovative and solve real problems. However, software developers have a hard time keeping up with all these changes across so many platforms.

And this has prompted the Unix vendors to begin to standardize their systems. Hence the impetus for Spec1170. Every major Unix vendor has committed to supporting this standard and every Unix software developer waits with glee the day they can write software to this standard and simply recompile (without having to use autoconf) across different platforms.

As I understand it, Spec1170 is roughly based upon version 4 of the X/Open Portability Guidelines (XPG4). Because catgets and friends are defined in XPG4, I'm led to believe that catgets is a part of Spec1170 and hence will become a standardized component of all Unix systems.

Next: , Previous: Temp catgets, Up: Temp Programmers

#### 11.6.3 Temporary - Why a single implementation

Now it seems kind of wasteful to me to have two different systems installed for accessing message catalogs. If we do want to remedy catgets deficiencies why don't we try to expand catgets (in a compatible manner) rather than implement an entirely new system. Otherwise, we'll end up with two message catalog access systems installed with an operating system - one set of routines for packages using GNU gettext for their internationalization, and another set of routines (catgets) for all other software. Bloated?

Supposing another catalog access system is implemented. Which do we recommend? At least for Linux, we need to attract as many software developers as possible. Hence we need to make it as easy for them to port their software as possible. Which means supporting catgets. We will be implementing the libintl code within our libc, but does this mean we also have to incorporate another message catalog access scheme within our libc as well? And what about people who are going to be using the libintl + non-catgets routines. When they port their software to other platforms, they're now going to have to include the front-end (libintl) code plus the back-end code (the non-catgets access routines) with their software instead of just including the libintl code with their software.

Message catalog support is however only the tip of the iceberg. What about the data for the other locale categories? They also have a number of deficiencies. Are we going to abandon them as well and develop another duplicate set of routines (should libintl expand beyond message catalog support)?

Like many parts of Unix that can be improved upon, we're stuck with balancing compatibility with the past with useful improvements and innovations for the future.

Previous: Temp WSI, Up: Temp Programmers

#### 11.6.4 Temporary - Notes

X/Open agreed very late on the standard form so that many implementations differ from the final form. Both of my system (old Linux catgets and Ultrix-4) have a strange variation.

OK. After incorporating the last changes I have to spend some time on making the GNU/Linux libc gettext functions. So in future Solaris is not the only system having gettext.

Next: , Previous: Programmers, Up: Top

## 12 The Translator's View

Next: , Previous: Translators, Up: Translators

### 12.1 Introduction 0

NOTE: This documentation section is outdated and needs to be revised.

Free software is going international! The Translation Project is a way to get maintainers, translators and users all together, so free software will gradually become able to speak many native languages.

The GNU gettext tool set contains everything maintainers need for internationalizing their packages for messages. It also contains quite useful tools for helping translators at localizing messages to their native language, once a package has already been internationalized.

To achieve the Translation Project, we need many interested people who like their own language and write it well, and who are also able to synergize with other translators speaking the same language. If you'd like to volunteer to work at translating messages, please send mail to your translating team.

Each team has its own mailing list, courtesy of Linux International. You may reach your translating team at the address ll@li.org, replacing ll by the two-letter ISO 639 code for your language. Language codes are not the same as country codes given in ISO 3166. The following translating teams exist:

Chinese zh, Czech cs, Danish da, Dutch nl, Esperanto eo, Finnish fi, French fr, Irish ga, German de, Greek el, Italian it, Japanese ja, Indonesian in, Norwegian no, Polish pl, Portuguese pt, Russian ru, Spanish es, Swedish sv and Turkish tr.

For example, you may reach the Chinese translating team by writing to zh@li.org. When you become a member of the translating team for your own language, you may subscribe to its list. For example, Swedish people can send a message to sv-request@li.org, having this message body:

     subscribe


Keep in mind that team members should be interested in working at translations, or at solving translational difficulties, rather than merely lurking around. If your team does not exist yet and you want to start one, please write to coordinator@translationproject.org; you will then reach the coordinator for all translator teams.

A handful of GNU packages have already been adapted and provided with message translations for several languages. Translation teams have begun to organize, using these packages as a starting point. But there are many more packages and many languages for which we have no volunteer translators. If you would like to volunteer to work at translating messages, please send mail to coordinator@translationproject.org indicating what language(s) you can work on.

Next: , Previous: Trans Intro 0, Up: Translators

### 12.2 Introduction 1

NOTE: This documentation section is outdated and needs to be revised.

This is now official, GNU is going international! Here is the announcement submitted for the January 1995 GNU Bulletin:

A handful of GNU packages have already been adapted and provided with message translations for several languages. Translation teams have begun to organize, using these packages as a starting point. But there are many more packages and many languages for which we have no volunteer translators. If you'd like to volunteer to work at translating messages, please send mail to ‘coordinator@translationproject.org’ indicating what language(s) you can work on.

This document should answer many questions for those who are curious about the process or would like to contribute. Please at least skim over it, hoping to cut down a little of the high volume of e-mail generated by this collective effort towards internationalization of free software.

Most free programming which is widely shared is done in English, and currently, English is used as the main communicating language between national communities collaborating to free software. This very document is written in English. This will not change in the foreseeable future.

However, there is a strong appetite from national communities for having more software able to write using national language and habits, and there is an on-going effort to modify free software in such a way that it becomes able to do so. The experiments driven so far raised an enthusiastic response from pretesters, so we believe that internationalization of free software is dedicated to succeed.

For suggestion clarifications, additions or corrections to this document, please e-mail to coordinator@translationproject.org.

Next: , Previous: Trans Intro 1, Up: Translators

### 12.3 Discussions

NOTE: This documentation section is outdated and needs to be revised.

Facing this internationalization effort, a few users expressed their concerns. Some of these doubts are presented and discussed, here.

• Smaller groups

Some languages are not spoken by a very large number of people, so people speaking them sometimes consider that there may not be all that much demand such versions of free software packages. Moreover, many people being into computers, in some countries, generally seem to prefer English versions of their software.

On the other end, people might enjoy their own language a lot, and be very motivated at providing to themselves the pleasure of having their beloved free software speaking their mother tongue. They do themselves a personal favor, and do not pay that much attention to the number of people benefiting of their work.

• Misinterpretation

Other users are shy to push forward their own language, seeing in this some kind of misplaced propaganda. Someone thought there must be some users of the language over the networks pestering other people with it.

But any spoken language is worth localization, because there are people behind the language for whom the language is important and dear to their hearts.

• Odd translations

The biggest problem is to find the right translations so that everybody can understand the messages. Translations are usually a little odd. Some people get used to English, to the extent they may find translations into their own language “rather pushy, obnoxious and sometimes even hilarious.” As a French speaking man, I have the experience of those instruction manuals for goods, so poorly translated in French in Korea or Taiwan...

The fact is that we sometimes have to create a kind of national computer culture, and this is not easy without the collaboration of many people liking their mother tongue. This is why translations are better achieved by people knowing and loving their own language, and ready to work together at improving the results they obtain.

• Dependencies over the GPL or LGPL

Some people wonder if using GNU gettext necessarily brings their package under the protective wing of the GNU General Public License or the GNU Library General Public License, when they do not want to make their program free, or want other kinds of freedom. The simplest answer is “normally not”.

The gettext-runtime part of GNU gettext, i.e. the contents of libintl, is covered by the GNU Library General Public License. The gettext-tools part of GNU gettext, i.e. the rest of the GNU gettext package, is covered by the GNU General Public License.

The mere marking of localizable strings in a package, or conditional inclusion of a few lines for initialization, is not really including GPL'ed or LGPL'ed code. However, since the localization routines in libintl are under the LGPL, the LGPL needs to be considered. It gives the right to distribute the complete unmodified source of libintl even with non-free programs. It also gives the right to use libintl as a shared library, even for non-free programs. But it gives the right to use libintl as a static library or to incorporate libintl into another library only to free software.

Next: , Previous: Discussions, Up: Translators

### 12.4 Organization

NOTE: This documentation section is outdated and needs to be revised.

On a larger scale, the true solution would be to organize some kind of fairly precise set up in which volunteers could participate. I gave some thought to this idea lately, and realize there will be some touchy points. I thought of writing to Richard Stallman to launch such a project, but feel it might be good to shake out the ideas between ourselves first. Most probably that Linux International has some experience in the field already, or would like to orchestrate the volunteer work, maybe. Food for thought, in any case!

I guess we have to setup something early, somehow, that will help many possible contributors of the same language to interlock and avoid work duplication, and further be put in contact for solving together problems particular to their tongue (in most languages, there are many difficulties peculiar to translating technical English). My Swedish contributor acknowledged these difficulties, and I'm well aware of them for French.

This is surely not a technical issue, but we should manage so the effort of locale contributors be maximally useful, despite the national team layer interface between contributors and maintainers.

The Translation Project needs some setup for coordinating language coordinators. Localizing evolving programs will surely become a permanent and continuous activity in the free software community, once well started. The setup should be minimally completed and tested before GNU gettext becomes an official reality. The e-mail address coordinator@translationproject.org has been set up for receiving offers from volunteers and general e-mail on these topics. This address reaches the Translation Project coordinator.

Next: , Previous: Organization, Up: Organization

#### 12.4.1 Central Coordination

I also think GNU will need sooner than it thinks, that someone set up a way to organize and coordinate these groups. Some kind of group of groups. My opinion is that it would be good that GNU delegates this task to a small group of collaborating volunteers, shortly. Perhaps in gnu.announce a list of this national committee's can be published.

My role as coordinator would simply be to refer to Ulrich any German speaking volunteer interested to localization of free software packages, and maybe helping national groups to initially organize, while maintaining national registries for until national groups are ready to take over. In fact, the coordinator should ease volunteers to get in contact with one another for creating national teams, which should then select one coordinator per language, or country (regionalized language). If well done, the coordination should be useful without being an overwhelming task, the time to put delegations in place.

Next: , Previous: Central Coordination, Up: Organization

#### 12.4.2 National Teams

I suggest we look for volunteer coordinators/editors for individual languages. These people will scan contributions of translation files for various programs, for their own languages, and will ensure high and uniform standards of diction.

From my current experience with other people in these days, those who provide localizations are very enthusiastic about the process, and are more interested in the localization process than in the program they localize, and want to do many programs, not just one. This seems to confirm that having a coordinator/editor for each language is a good idea.

We need to choose someone who is good at writing clear and concise prose in the language in question. That is hard—we can't check it ourselves. So we need to ask a few people to judge each others' writing and select the one who is best.

I announce my prerelease to a few dozen people, and you would not believe all the discussions it generated already. I shudder to think what will happen when this will be launched, for true, officially, world wide. Who am I to arbitrate between two Czekolsovak users contradicting each other, for example?

I assume that your German is not much better than my French so that I would not be able to judge about these formulations. What I would suggest is that for each language there is a group for people who maintain the PO files and judge about changes. I suspect there will be cultural differences between how such groups of people will behave. Some will have relaxed ways, reach consensus easily, and have anyone of the group relate to the maintainers, while others will fight to death, organize heavy administrations up to national standards, and use strict channels.

The German team is putting out a good example. Right now, they are maybe half a dozen people revising translations of each other and discussing the linguistic issues. I do not even have all the names. Ulrich Drepper is taking care of coordinating the German team. He subscribed to all my pretest lists, so I do not even have to warn him specifically of incoming releases.

I'm sure, that is a good idea to get teams for each language working on translations. That will make the translations better and more consistent.

Next: , Previous: National Teams, Up: National Teams
##### 12.4.2.1 Sub-Cultures

Taking French for example, there are a few sub-cultures around computers which developed diverging vocabularies. Picking volunteers here and there without addressing this problem in an organized way, soon in the project, might produce a distasteful mix of internationalized programs, and possibly trigger endless quarrels among those who really care.

Keeping some kind of unity in the way French localization of internationalized programs is achieved is a difficult (and delicate) job. Knowing the latin character of French people (:-), if we take this the wrong way, we could end up nowhere, or spoil a lot of energies. Maybe we should begin to address this problem seriously before GNU gettext become officially published. And I suspect that this means soon!

Previous: Sub-Cultures, Up: National Teams
##### 12.4.2.2 Organizational Ideas

I expect the next big changes after the official release. Please note that I use the German translation of the short GPL message. We need to set a few good examples before the localization goes out for true in the free software community. Here are a few points to discuss:

• Each group should have one FTP server (at least one master).
• The files on the server should reflect the latest version (of course!) and it should also contain a RCS directory with the corresponding archives (I don't have this now).
• There should also be a ChangeLog file (this is more useful than the RCS archive but can be generated automatically from the later by Emacs).
• A core group should judge about questionable changes (for now this group consists solely by me but I ask some others occasionally; this also seems to work).

Previous: National Teams, Up: Organization

#### 12.4.3 Mailing Lists

If we get any inquiries about GNU gettext, send them on to:

     coordinator@translationproject.org


The *-pretest lists are quite useful to me, maybe the idea could be generalized to many GNU, and non-GNU packages. But each maintainer his/her way!

François, we have a mechanism in place here at gnu.ai.mit.edu to track teams, support mailing lists for them and log members. We have a slight preference that you use it. If this is OK with you, I can get you clued in.

Things are changing! A few years ago, when Daniel Fekete and I asked for a mailing list for GNU localization, nested at the FSF, we were politely invited to organize it anywhere else, and so did we. For communicating with my pretesters, I later made a handful of mailing lists located at iro.umontreal.ca and administrated by majordomo. These lists have been very dependable so far...

I suspect that the German team will organize itself a mailing list located in Germany, and so forth for other countries. But before they organize for true, it could surely be useful to offer mailing lists located at the FSF to each national team. So yes, please explain me how I should proceed to create and handle them.

We should create temporary mailing lists, one per country, to help people organize. Temporary, because once regrouped and structured, it would be fair the volunteers from country bring back their list in there and manage it as they want. My feeling is that, in the long run, each team should run its own list, from within their country. There also should be some central list to which all teams could subscribe as they see fit, as long as each team is represented in it.

Next: , Previous: Organization, Up: Translators

### 12.5 Information Flow

NOTE: This documentation section is outdated and needs to be revised.

There will surely be some discussion about this messages after the packages are finally released. If people now send you some proposals for better messages, how do you proceed? Jim, please note that right now, as I put forward nearly a dozen of localizable programs, I receive both the translations and the coordination concerns about them.

If I put one of my things to pretest, Ulrich receives the announcement and passes it on to the German team, who make last minute revisions. Then he submits the translation files to me as the maintainer. For free packages I do not maintain, I would not even hear about it. This scheme could be made to work for the whole Translation Project, I think. For security reasons, maybe Ulrich (national coordinators, in fact) should update central registry kept at the Translation Project (Jim, me, or Len's recruits) once in a while.

In December/January, I was aggressively ready to internationalize all of GNU, giving myself the duty of one small GNU package per week or so, taking many weeks or months for bigger packages. But it does not work this way. I first did all the things I'm responsible for. I've nothing against some missionary work on other maintainers, but I'm also loosing a lot of energy over it—same debates over again.

And when the first localized packages are released we'll get a lot of responses about ugly translations :-). Surely, and we need to have beforehand a fairly good idea about how to handle the information flow between the national teams and the package maintainers.

Please start saving somewhere a quick history of each PO file. I know for sure that the file format will change, allowing for comments. It would be nice that each file has a kind of log, and references for those who want to submit comments or gripes, or otherwise contribute. I sent a proposal for a fast and flexible format, but it is not receiving acceptance yet by the GNU deciders. I'll tell you when I have more information about this.

Previous: Information Flow, Up: Translators

### 12.6 Prioritizing messages: How to determine which messages to translate first

A translator sometimes has only a limited amount of time per week to spend on a package, and some packages have quite large message catalogs (over 1000 messages). Therefore she wishes to translate the messages first that are the most visible to the user, or that occur most frequently. This section describes how to determine these "most urgent" messages. It also applies to determine the "next most urgent" messages after the message catalog has already been partially translated.

In a first step, she uses the programs like a user would do. While she does this, the GNU gettext library logs into a file the not yet translated messages for which a translation was requested from the program.

In a second step, she uses the PO mode to translate precisely this set of messages.

Here a more details. The GNU libintl library (but not the corresponding functions in GNU libc) supports an environment variable GETTEXT_LOG_UNTRANSLATED. The GNU libintl library will log into this file the messages for which gettext() and related functions couldn't find the translation. If the file doesn't exist, it will be created as needed. On systems with GNU libc a shared library ‘preloadable_libintl.so’ is provided that can be used with the ELF ‘LD_PRELOAD’ mechanism.

So, in the first step, the translator uses these commands on systems with GNU libc:

     $LD_PRELOAD=/usr/local/lib/preloadable_libintl.so$ export LD_PRELOAD
$GETTEXT_LOG_UNTRANSLATED=$HOME/gettextlogused
$export GETTEXT_LOG_UNTRANSLATED  and these commands on other systems: $ GETTEXT_LOG_UNTRANSLATED=$HOME/gettextlogused$ export GETTEXT_LOG_UNTRANSLATED


Then she uses and peruses the programs. (It is a good and recommended practice to use the programs for which you provide translations: it gives you the needed context.) When done, she removes the environment variables:

     $unset LD_PRELOAD$ unset GETTEXT_LOG_UNTRANSLATED


The second step starts with removing duplicates:

     $msguniq$HOME/gettextlogused > missing.po


The result is a PO file, but needs some preprocessing before a PO file editor can be used with it. First, it is a multi-domain PO file, containing messages from many translation domains. Second, it lacks all translator comments and source references. Here is how to get a list of the affected translation domains:

     $sed -n -e 's,^domain "$$.*$$"$,\1,p' < missing.po | sort | uniq


Then the translator can handle the domains one by one. For simplicity, let's use environment variables to denote the language, domain and source package.

     $lang=nl # your language$ domain=coreutils    # the name of the domain to be handled
$package=/usr/src/gnu/coreutils-4.5.4 # the package where it comes from  She takes the latest copy of$lang.po from the Translation Project, or from the package (in most cases, $package/po/$lang.po), or creates a fresh one if she's the first translator (see Creating). She then uses the following commands to mark the not urgent messages as "obsolete". (This doesn't mean that these messages - translated and untranslated ones - will go away. It simply means that the PO file editor will ignore them in the following editing session.)

     $msggrep --domain=$domain missing.po | grep -v '^domain' \
> $domain-missing.po$ msgattrib --set-obsolete --ignore-file $domain-missing.po$domain.$lang.po \ >$domain.$lang-urgent.po  The she translates$domain.$lang-urgent.po by use of a PO file editor (see Editing). (FIXME: I don't know whether KBabel and gtranslator also preserve obsolete messages, as they should.) Finally she restores the not urgent messages (with their earlier translations, for those which were already translated) through this command: $ msgmerge --no-fuzzy-matching $domain.$lang-urgent.po $package/po/$domain.pot \
> $domain.$lang.po


Then she can submit $domain.$lang.po and proceed to the next domain.

Next: , Previous: Translators, Up: Top

## 13 The Maintainer's View

The maintainer of a package has many responsibilities. One of them is ensuring that the package will install easily on many platforms, and that the magic we described earlier (see Users) will work for installers and end users.

Of course, there are many possible ways by which GNU gettext might be integrated in a distribution, and this chapter does not cover them in all generality. Instead, it details one possible approach which is especially adequate for many free software distributions following GNU standards, or even better, Gnits standards, because GNU gettext is purposely for helping the internationalization of the whole GNU project, and as many other good free packages as possible. So, the maintainer's view presented here presumes that the package already has a configure.ac file and uses GNU Autoconf.

Nevertheless, GNU gettext may surely be useful for free packages not following GNU standards and conventions, but the maintainers of such packages might have to show imagination and initiative in organizing their distributions so gettext work for them in all situations. There are surely many, out there.

Even if gettext methods are now stabilizing, slight adjustments might be needed between successive gettext versions, so you should ideally revise this chapter in subsequent releases, looking for changes.

Next: , Previous: Maintainers, Up: Maintainers

### 13.1 Flat or Non-Flat Directory Structures

Some free software packages are distributed as tar files which unpack in a single directory, these are said to be flat distributions. Other free software packages have a one level hierarchy of subdirectories, using for example a subdirectory named doc/ for the Texinfo manual and man pages, another called lib/ for holding functions meant to replace or complement C libraries, and a subdirectory src/ for holding the proper sources for the package. These other distributions are said to be non-flat.

We cannot say much about flat distributions. A flat directory structure has the disadvantage of increasing the difficulty of updating to a new version of GNU gettext. Also, if you have many PO files, this could somewhat pollute your single directory. Also, GNU gettext's libintl sources consist of C sources, shell scripts, sed scripts and complicated Makefile rules, which don't fit well into an existing flat structure. For these reasons, we recommend to use non-flat approach in this case as well.

Maybe because GNU gettext itself has a non-flat structure, we have more experience with this approach, and this is what will be described in the remaining of this chapter. Some maintainers might use this as an opportunity to unflatten their package structure.

Next: , Previous: Flat and Non-Flat, Up: Maintainers

### 13.2 Prerequisite Works

There are some works which are required for using GNU gettext in one of your package. These works have some kind of generality that escape the point by point descriptions used in the remainder of this chapter. So, we describe them here.

• Before attempting to use gettextize you should install some other packages first. Ensure that recent versions of GNU m4, GNU Autoconf and GNU gettext are already installed at your site, and if not, proceed to do this first. If you get to install these things, beware that GNU m4 must be fully installed before GNU Autoconf is even configured.

To further ease the task of a package maintainer the automake package was designed and implemented. GNU gettext now uses this tool and the Makefiles in the intl/ and po/ therefore know about all the goals necessary for using automake and libintl in one project.

Those four packages are only needed by you, as a maintainer; the installers of your own package and end users do not really need any of GNU m4, GNU Autoconf, GNU gettext, or GNU automake for successfully installing and running your package, with messages properly translated. But this is not completely true if you provide internationalized shell scripts within your own package: GNU gettext shall then be installed at the user site if the end users want to see the translation of shell script messages.

• Your package should use Autoconf and have a configure.ac or configure.in file. If it does not, you have to learn how. The Autoconf documentation is quite well written, it is a good idea that you print it and get familiar with it.
• Your C sources should have already been modified according to instructions given earlier in this manual. See Sources.
• Your po/ directory should receive all PO files submitted to you by the translator teams, each having ll.po as a name. This is not usually easy to get translation work done before your package gets internationalized and available! Since the cycle has to start somewhere, the easiest for the maintainer is to start with absolutely no PO files, and wait until various translator teams get interested in your package, and submit PO files.

It is worth adding here a few words about how the maintainer should ideally behave with PO files submissions. As a maintainer, your role is to authenticate the origin of the submission as being the representative of the appropriate translating teams of the Translation Project (forward the submission to coordinator@translationproject.org in case of doubt), to ensure that the PO file format is not severely broken and does not prevent successful installation, and for the rest, to merely put these PO files in po/ for distribution.

As a maintainer, you do not have to take on your shoulders the responsibility of checking if the translations are adequate or complete, and should avoid diving into linguistic matters. Translation teams drive themselves and are fully responsible of their linguistic choices for the Translation Project. Keep in mind that translator teams are not driven by maintainers. You can help by carefully redirecting all communications and reports from users about linguistic matters to the appropriate translation team, or explain users how to reach or join their team. The simplest might be to send them the ABOUT-NLS file.

Maintainers should never ever apply PO file bug reports themselves, short-cutting translation teams. If some translator has difficulty to get some of her points through her team, it should not be an option for her to directly negotiate translations with maintainers. Teams ought to settle their problems themselves, if any. If you, as a maintainer, ever think there is a real problem with a team, please never try to solve a team's problem on your own.

Next: , Previous: Prerequisites, Up: Maintainers

### 13.3 Invoking the gettextize Program

The gettextize program is an interactive tool that helps the maintainer of a package internationalized through GNU gettext. It is used for two purposes:

• As a wizard, when a package is modified to use GNU gettext for the first time.
• As a migration tool, for upgrading the GNU gettext support in a package from a previous to a newer version of GNU gettext.

This program performs the following tasks:

• It copies into the package some files that are consistently and identically needed in every package internationalized through GNU gettext.
• It performs as many of the tasks mentioned in the next section Adjusting Files as can be performed automatically.
• It removes obsolete files and idioms used for previous GNU gettext versions to the form recommended for the current GNU gettext version.
• It prints a summary of the tasks that ought to be done manually and could not be done automatically by gettextize.

It can be invoked as follows:

     gettextize [ option... ] [ directory ]


and accepts the following options:

-f
--force
Force replacement of files which already exist.
--intl
Install the libintl sources in a subdirectory named intl/. This libintl will be used to provide internationalization on systems that don't have GNU libintl installed. If this option is omitted, the call to AM_GNU_GETTEXT in configure.ac should read: ‘AM_GNU_GETTEXT([external])’, and internationalization will not be enabled on systems lacking GNU gettext.
--po-dir=dir
Specify a directory containing PO files. Such a directory contains the translations into various languages of a particular POT file. This option can be specified multiple times, once for each translation domain. If it is not specified, the directory named po/ is updated.
--no-changelog
Don't update or create ChangeLog files. By default, gettextize logs all changes (file additions, modifications and removals) in a file called ‘ChangeLog’ in each affected directory.
Make symbolic links instead of copying the needed files. This can be useful to save a few kilobytes of disk space, but it requires extra effort to create self-contained tarballs, it may disturb some mechanism the maintainer applies to the sources, and it is likely to introduce bugs when a newer version of gettext is installed on the system.
-n
--dry-run
Print modifications but don't perform them. All actions that gettextize would normally execute are inhibited and instead only listed on standard output.
--help
Display this help and exit.
--version
Output version information and exit.

If directory is given, this is the top level directory of a package to prepare for using GNU gettext. If not given, it is assumed that the current directory is the top level directory of such a package.

The program gettextize provides the following files. However, no existing file will be replaced unless the option --force (-f) is specified.

1. The ABOUT-NLS file is copied in the main directory of your package, the one being at the top level. This file gives the main indications about how to install and use the Native Language Support features of your program. You might elect to use a more recent copy of this ABOUT-NLS file than the one provided through gettextize, if you have one handy. You may also fetch a more recent copy of file ABOUT-NLS from Translation Project sites, and from most GNU archive sites.
2. A po/ directory is created for eventually holding all translation files, but initially only containing the file po/Makefile.in.in from the GNU gettext distribution (beware the double ‘.in’ in the file name) and a few auxiliary files. If the po/ directory already exists, it will be preserved along with the files it contains, and only Makefile.in.in and the auxiliary files will be overwritten.

If ‘--po-dir’ has been specified, this holds for every directory specified through ‘--po-dir’, instead of po/.

3. Only if ‘--intl’ has been specified: A intl/ directory is created and filled with most of the files originally in the intl/ directory of the GNU gettext distribution. Also, if option --force (-f) is given, the intl/ directory is emptied first.
4. The file config.rpath is copied into the directory containing configuration support files. It is needed by the AM_GNU_GETTEXT autoconf macro.
5. Only if the project is using GNU automake: A set of autoconf macro files is copied into the package's autoconf macro repository, usually in a directory called m4/.

If your site support symbolic links, gettextize will not actually copy the files into your package, but establish symbolic links instead. This avoids duplicating the disk space needed in all packages. Merely using the ‘-h’ option while creating the tar archive of your distribution will resolve each link by an actual copy in the distribution archive. So, to insist, you really should use ‘-h’ option with tar within your dist goal of your main Makefile.in.

Furthermore, gettextize will update all Makefile.am files in each affected directory, as well as the top level configure.ac or configure.in file.

It is interesting to understand that most new files for supporting GNU gettext facilities in one package go in intl/, po/ and m4/ subdirectories. One distinction between intl/ and the two other directories is that intl/ is meant to be completely identical in all packages using GNU gettext, while the other directories will mostly contain package dependent files.

The gettextize program makes backup files for all files it replaces or changes, and also write ChangeLog entries about these changes. This way, the careful maintainer can check after running gettextize whether its changes are acceptable to him, and possibly adjust them. An exception to this rule is the intl/ directory, which is added or replaced or removed as a whole.

It is important to understand that gettextize can not do the entire job of adapting a package for using GNU gettext. The amount of remaining work depends on whether the package uses GNU automake or not. But in any case, the maintainer should still read the section Adjusting Files after invoking gettextize.

In particular, if after using ‘gettexize’, you get an error ‘AC_COMPILE_IFELSE was called before AC_GNU_SOURCE’ or ‘AC_RUN_IFELSE was called before AC_GNU_SOURCE’, you can fix it by modifying configure.ac, as described in configure.ac.

It is also important to understand that gettextize is not part of the GNU build system, in the sense that it should not be invoked automatically, and not be invoked by someone who doesn't assume the responsibilities of a package maintainer. For the latter purpose, a separate tool is provided, see autopoint Invocation.

Next: , Previous: gettextize Invocation, Up: Maintainers

### 13.4 Files You Must Create or Alter

Besides files which are automatically added through gettextize, there are many files needing revision for properly interacting with GNU gettext. If you are closely following GNU standards for Makefile engineering and auto-configuration, the adaptations should be easier to achieve. Here is a point by point description of the changes needed in each.

So, here comes a list of files, each one followed by a description of all alterations it needs. Many examples are taken out from the GNU gettext 0.17 distribution itself, or from the GNU hello distribution (http://www.franken.de/users/gnu/ke/hello or http://www.gnu.franken.de/ke/hello/) You may indeed refer to the source code of the GNU gettext and GNU hello packages, as they are intended to be good examples for using GNU gettext functionality.

#### 13.4.1 POTFILES.in in po/

The po/ directory should receive a file named POTFILES.in. This file tells which files, among all program sources, have marked strings needing translation. Here is an example of such a file:

     # List of source files containing translatable strings.
# Copyright (C) 1995 Free Software Foundation, Inc.

# Common library files
lib/error.c
lib/getopt.c
lib/xmalloc.c

# Package source files
src/gettext.c
src/msgfmt.c
src/xgettext.c


Hash-marked comments and white lines are ignored. All other lines list those source files containing strings marked for translation (see Mark Keywords), in a notation relative to the top level of your whole distribution, rather than the location of the POTFILES.in file itself.

When a C file is automatically generated by a tool, like flex or bison, that doesn't introduce translatable strings by itself, it is recommended to list in po/POTFILES.in the real source file (ending in .l in the case of flex, or in .y in the case of bison), not the generated C file.

Next: , Previous: po/POTFILES.in, Up: Adjusting Files

#### 13.4.2 LINGUAS in po/

The po/ directory should also receive a file named LINGUAS. This file contains the list of available translations. It is a whitespace separated list. Hash-marked comments and white lines are ignored. Here is an example file:

     # Set of available languages.
de fr


This example means that German and French PO files are available, so that these languages are currently supported by your package. If you want to further restrict, at installation time, the set of installed languages, this should not be done by modifying the LINGUAS file, but rather by using the LINGUAS environment variable (see Installers).

It is recommended that you add the "languages" ‘en@quot’ and ‘en@boldquot’ to the LINGUAS file. en@quot is a variant of English message catalogs (en) which uses real quotation marks instead of the ugly looking asymmetric ASCII substitutes ‘’ and ‘'’. en@boldquot is a variant of en@quot that additionally outputs quoted pieces of text in a bold font, when used in a terminal emulator which supports the VT100 escape sequences (such as xterm or the Linux console, but not Emacs in M-x shell mode).

These extra message catalogs ‘en@quot’ and ‘en@boldquot’ are constructed automatically, not by translators; to support them, you need the files Rules-quot, quot.sed, boldquot.sed, en@quot.header, en@boldquot.header, insert-header.sin in the po/ directory. You can copy them from GNU gettext's po/ directory; they are also installed by running gettextize.

Next: , Previous: po/LINGUAS, Up: Adjusting Files

#### 13.4.3 Makevars in po/

The po/ directory also has a file named Makevars. It contains variables that are specific to your project. po/Makevars gets inserted into the po/Makefile when the latter is created. The variables thus take effect when the POT file is created or updated, and when the message catalogs get installed.

The first three variables can be left unmodified if your package has a single message domain and, accordingly, a single po/ directory. Only packages which have multiple po/ directories at different locations need to adjust the three first variables defined in Makevars.

As an alternative to the XGETTEXT_OPTIONS variables, it is also possible to specify xgettext options through the AM_XGETTEXT_OPTION autoconf macro. See AM_XGETTEXT_OPTION.

Next: , Previous: po/Makevars, Up: Adjusting Files

#### 13.4.4 Extending Makefile in po/

All files called Rules-* in the po/ directory get appended to the po/Makefile when it is created. They present an opportunity to add rules for special PO files to the Makefile, without needing to mess with po/Makefile.in.in.

GNU gettext comes with a Rules-quot file, containing rules for building catalogs en@quot.po and en@boldquot.po. The effect of en@quot.po is that people who set their LANGUAGE environment variable to ‘en@quot’ will get messages with proper looking symmetric Unicode quotation marks instead of abusing the ASCII grave accent and the ASCII apostrophe for indicating quotations. To enable this catalog, simply add en@quot to the po/LINGUAS file. The effect of en@boldquot.po is that people who set LANGUAGE to ‘en@boldquot’ will get not only proper quotation marks, but also the quoted text will be shown in a bold font on terminals and consoles. This catalog is useful only for command-line programs, not GUI programs. To enable it, similarly add en@boldquot to the po/LINGUAS file.

Similarly, you can create rules for building message catalogs for the sr@latin locale – Serbian written with the Latin alphabet – from those for the sr locale – Serbian written with Cyrillic letters. See msgfilter Invocation.

Next: , Previous: po/Rules-*, Up: Adjusting Files

#### 13.4.5 configure.ac at top level

configure.ac or configure.in - this is the source from which autoconf generates the configure script.

1. Declare the package and version. This is done by a set of lines like these:
          PACKAGE=gettext
VERSION=0.17
AC_DEFINE_UNQUOTED(PACKAGE, "$PACKAGE") AC_DEFINE_UNQUOTED(VERSION, "$VERSION")
AC_SUBST(PACKAGE)
AC_SUBST(VERSION)


or, if you are using GNU automake, by a line like this:

          AM_INIT_AUTOMAKE(gettext, 0.17)


Of course, you replace ‘gettext’ with the name of your package, and ‘0.17’ by its version numbers, exactly as they should appear in the packaged tar file name of your distribution (gettext-0.17.tar.gz, here).

2. Check for internationalization support.

Here is the main m4 macro for triggering internationalization support. Just add this line to configure.ac:

          AM_GNU_GETTEXT


This call is purposely simple, even if it generates a lot of configure time checking and actions.

If you have suppressed the intl/ subdirectory by calling gettextize without ‘--intl’ option, this call should read

          AM_GNU_GETTEXT([external])

3. Have output files created.

The AC_OUTPUT directive, at the end of your configure.ac file, needs to be modified in two ways:

          AC_OUTPUT([existing configuration files intl/Makefile po/Makefile.in],


The modification to the first argument to AC_OUTPUT asks for substitution in the intl/ and po/ directories. Note the ‘.in’ suffix used for po/ only. This is because the distributed file is really po/Makefile.in.in.

If you have suppressed the intl/ subdirectory by calling gettextize without ‘--intl’ option, then you don't need to add intl/Makefile to the AC_OUTPUT line.

If, after doing the recommended modifications, a command like ‘aclocal -I m4’ or ‘autoconf’ or ‘autoreconf’ fails with a trace similar to this:

     configure.ac:44: warning: AC_COMPILE_IFELSE was called before AC_GNU_SOURCE
../../lib/autoconf/specific.m4:335: AC_GNU_SOURCE is expanded from...
m4/lock.m4:224: gl_LOCK is expanded from...
m4/gettext.m4:571: gt_INTL_SUBDIR_CORE is expanded from...
m4/gettext.m4:472: AM_INTL_SUBDIR is expanded from...
m4/gettext.m4:347: AM_GNU_GETTEXT is expanded from...
configure.ac:44: the top level
configure.ac:44: warning: AC_RUN_IFELSE was called before AC_GNU_SOURCE


you need to add an explicit invocation of ‘AC_GNU_SOURCE’ in the configure.ac file - after ‘AC_PROG_CC’ but before ‘AM_GNU_GETTEXT’, most likely very close to the ‘AC_PROG_CC’ invocation. This is necessary because of ordering restrictions imposed by GNU autoconf.

Next: , Previous: configure.ac, Up: Adjusting Files

#### 13.4.6 config.guess, config.sub at top level

If you haven't suppressed the intl/ subdirectory, you need to add the GNU config.guess and config.sub files to your distribution. They are needed because the intl/ directory has platform dependent support for determining the locale's character encoding and therefore needs to identify the platform.

You can obtain the newest version of config.guess and config.sub from the CVS of the ‘config’ project at http://savannah.gnu.org/. The commands to fetch them are

     $wget 'http://savannah.gnu.org/cgi-bin/viewcvs/*checkout*/config/config/config.guess'$ wget 'http://savannah.gnu.org/cgi-bin/viewcvs/*checkout*/config/config/config.sub'


Less recent versions are also contained in the GNU automake and GNU libtool packages.

Normally, config.guess and config.sub are put at the top level of a distribution. But it is also possible to put them in a subdirectory, altogether with other configuration support files like install-sh, ltconfig, ltmain.sh or missing. All you need to do, other than moving the files, is to add the following line to your configure.ac.

     AC_CONFIG_AUX_DIR([subdir])


Next: , Previous: config.guess, Up: Adjusting Files

#### 13.4.7 mkinstalldirs at top level

With earlier versions of GNU gettext, you needed to add the GNU mkinstalldirs script to your distribution. This is not needed any more. You can remove it if you not also using an automake version older than automake 1.9.

Next: , Previous: mkinstalldirs, Up: Adjusting Files

#### 13.4.8 aclocal.m4 at top level

If you do not have an aclocal.m4 file in your distribution, the simplest is to concatenate the files codeset.m4, gettext.m4, glibc2.m4, glibc21.m4, iconv.m4, intdiv0.m4, intl.m4, intldir.m4, intlmacosx.m4, intmax.m4, inttypes_h.m4, inttypes-pri.m4, lcmessage.m4, lib-ld.m4, lib-link.m4, lib-prefix.m4, lock.m4, longlong.m4, nls.m4, po.m4, printf-posix.m4, progtest.m4, size_max.m4, stdint_h.m4, uintmax_t.m4, visibility.m4, wchar_t.m4, wint_t.m4, xsize.m4 from GNU gettext's m4/ directory into a single file. If you have suppressed the intl/ directory, only gettext.m4, iconv.m4, lib-ld.m4, lib-link.m4, lib-prefix.m4, nls.m4, po.m4, progtest.m4 need to be concatenated.

If you are not using GNU automake 1.8 or newer, you will need to add a file mkdirp.m4 from a newer automake distribution to the list of files above.

If you already have an aclocal.m4 file, then you will have to merge the said macro files into your aclocal.m4. Note that if you are upgrading from a previous release of GNU gettext, you should most probably replace the macros (AM_GNU_GETTEXT, etc.), as they usually change a little from one release of GNU gettext to the next. Their contents may vary as we get more experience with strange systems out there.

If you are using GNU automake 1.5 or newer, it is enough to put these macro files into a subdirectory named m4/ and add the line

     ACLOCAL_AMFLAGS = -I m4


These macros check for the internationalization support functions and related informations. Hopefully, once stabilized, these macros might be integrated in the standard Autoconf set, because this piece of m4 code will be the same for all projects using GNU gettext.

Next: , Previous: aclocal, Up: Adjusting Files

#### 13.4.9 acconfig.h at top level

Earlier GNU gettext releases required to put definitions for ENABLE_NLS, HAVE_GETTEXT and HAVE_LC_MESSAGES, HAVE_STPCPY, PACKAGE and VERSION into an acconfig.h file. This is not needed any more; you can remove them from your acconfig.h file unless your package uses them independently from the intl/ directory.

Next: , Previous: acconfig, Up: Adjusting Files

#### 13.4.10 config.h.in at top level

The include file template that holds the C macros to be defined by configure is usually called config.h.in and may be maintained either manually or automatically.

If gettextize has created an intl/ directory, this file must be called config.h.in and must be at the top level. If, however, you have suppressed the intl/ directory by calling gettextize without ‘--intl’ option, then you can choose the name of this file and its location freely.

If it is maintained automatically, by use of the ‘autoheader’ program, you need to do nothing about it. This is the case in particular if you are using GNU automake.

If it is maintained manually, and if gettextize has created an intl/ directory, you should switch to using ‘autoheader’. The list of C macros to be added for the sake of the intl/ directory is just too long to be maintained manually; it also changes between different versions of GNU gettext.

If it is maintained manually, and if on the other hand you have suppressed the intl/ directory by calling gettextize without ‘--intl’ option, then you can get away by adding the following lines to config.h.in:

     /* Define to 1 if translation of program messages to the user's
native language is requested. */
#undef ENABLE_NLS


Next: , Previous: config.h.in, Up: Adjusting Files

#### 13.4.11 Makefile.in at top level

Here are a few modifications you need to make to your main, top-level Makefile.in file.

1. Add the following lines near the beginning of your Makefile.in, so the ‘dist:’ goal will work properly (as explained further down):
          PACKAGE = @PACKAGE@
VERSION = @VERSION@

2. Add file ABOUT-NLS to the DISTFILES definition, so the file gets distributed.
3. Wherever you process subdirectories in your Makefile.in, be sure you also process the subdirectories ‘intl’ and ‘po’. Special rules in the Makefiles take care for the case where no internationalization is wanted.

If you are using Makefiles, either generated by automake, or hand-written so they carefully follow the GNU coding standards, the effected goals for which the new subdirectories must be handled include ‘installdirs’, ‘install’, ‘uninstall’, ‘clean’, ‘distclean’.

Here is an example of a canonical order of processing. In this example, we also define SUBDIRS in Makefile.in for it to be further used in the ‘dist:’ goal.

          SUBDIRS = doc intl lib src po


Note that you must arrange for ‘make’ to descend into the intl directory before descending into other directories containing code which make use of the libintl.h header file. For this reason, here we mention intl before lib and src.

4. A delicate point is the ‘dist:’ goal, as both intl/Makefile and po/Makefile will later assume that the proper directory has been set up from the main Makefile. Here is an example at what the ‘dist:’ goal might look like:
          distdir = $(PACKAGE)-$(VERSION)
dist: Makefile
rm -fr $(distdir) mkdir$(distdir)
chmod 777 $(distdir) for file in$(DISTFILES); do \
ln $$file (distdir) 2>/dev/null || cp -p$$file $(distdir); \ done for subdir in$(SUBDIRS); do \
mkdir $(distdir)/$$subdir || exit 1; \ chmod 777 (distdir)/$$subdir; \ (cd $$subdir && (MAKE) @) || exit 1; \ done tar chozf (distdir).tar.gz (distdir) rm -fr (distdir)  Note that if you are using GNU automake, Makefile.in is automatically generated from Makefile.am, and all needed changes to Makefile.am are already made by running ‘gettextize’. Next: , Previous: Makefile, Up: Adjusting Files #### 13.4.12 Makefile.in in src/ Some of the modifications made in the main Makefile.in will also be needed in the Makefile.in from your package sources, which we assume here to be in the src/ subdirectory. Here are all the modifications needed in src/Makefile.in: 1. In view of the ‘dist:’ goal, you should have these lines near the beginning of src/Makefile.in:  PACKAGE = @PACKAGE@ VERSION = @VERSION@  2. If not done already, you should guarantee that top_srcdir gets defined. This will serve for cpp include files. Just add the line:  top_srcdir = @top_srcdir@  3. You might also want to define subdir as ‘src’, later allowing for almost uniform ‘dist:’ goals in all your Makefile.in. At list, the ‘dist:’ goal below assume that you used:  subdir = src  4. The main function of your program will normally call bindtextdomain (see see Triggering), like this:  bindtextdomain (PACKAGE, LOCALEDIR); textdomain (PACKAGE);  To make LOCALEDIR known to the program, add the following lines to Makefile.in:  datadir = @datadir@ localedir = (datadir)/locale DEFS = -DLOCALEDIR=\"(localedir)\" @DEFS@  Note that @datadir@ defaults to ‘(prefix)/share’, thus (localedir) defaults to ‘(prefix)/share/locale’. 5. You should ensure that the final linking will use @LIBINTL@ or @LTLIBINTL@ as a library. @LIBINTL@ is for use without libtool, @LTLIBINTL@ is for use with libtool. An easy way to achieve this is to manage that it gets into LIBS, like this:  LIBS = @LIBINTL@ @LIBS@  In most packages internationalized with GNU gettext, one will find a directory lib/ in which a library containing some helper functions will be build. (You need at least the few functions which the GNU gettext Library itself needs.) However some of the functions in the lib/ also give messages to the user which of course should be translated, too. Taking care of this, the support library (say libsupport.a) should be placed before @LIBINTL@ and @LIBS@ in the above example. So one has to write this:  LIBS = ../lib/libsupport.a @LIBINTL@ @LIBS@  6. You should also ensure that directory intl/ will be searched for C preprocessor include files in all circumstances. So, you have to manage so both ‘-I../intl’ and ‘-I(top_srcdir)/intl’ will be given to the C compiler. 7. Your ‘dist:’ goal has to conform with others. Here is a reasonable definition for it:  distdir = ../(PACKAGE)-(VERSION)/(subdir) dist: Makefile (DISTFILES) for file in (DISTFILES); do \ ln$$file$(distdir) 2>/dev/null || cp -p $$file (distdir) || exit 1; \ done  Note that if you are using GNU automake, Makefile.in is automatically generated from Makefile.am, and the first three changes and the last change are not necessary. The remaining needed Makefile.am modifications are the following: 1. To make LOCALEDIR known to the program, add the following to Makefile.am:  <module>_CPPFLAGS = -DLOCALEDIR=\"(localedir)\"  for each specific module or compilation unit, or  AM_CPPFLAGS = -DLOCALEDIR=\"(localedir)\"  for all modules and compilation units together. Furthermore, add this line to define ‘localedir’:  localedir = (datadir)/locale  2. To ensure that the final linking will use @LIBINTL@ or @LTLIBINTL@ as a library, add the following to Makefile.am:  <program>_LDADD = @LIBINTL@  for each specific program, or  LDADD = @LIBINTL@  for all programs together. Remember that when you use libtool to link a program, you need to use @LTLIBINTL@ instead of @LIBINTL@ for that program. 3. If you have an intl/ directory, whose contents is created by gettextize, then to ensure that it will be searched for C preprocessor include files in all circumstances, add something like this to Makefile.am:  AM_CPPFLAGS = -I../intl -I(top_srcdir)/intl  Previous: src/Makefile, Up: Adjusting Files #### 13.4.13 gettext.h in lib/ Internationalization of packages, as provided by GNU gettext, is optional. It can be turned off in two situations: • When the installer has specified ‘./configure --disable-nls’. This can be useful when small binaries are more important than features, for example when building utilities for boot diskettes. It can also be useful in order to get some specific C compiler warnings about code quality with some older versions of GCC (older than 3.0). • When the package does not include the intl/ subdirectory, and the libintl.h header (with its associated libintl library, if any) is not already installed on the system, it is preferable that the package builds without internationalization support, rather than to give a compilation error. A C preprocessor macro can be used to detect these two cases. Usually, when libintl.h was found and not explicitly disabled, the ENABLE_NLS macro will be defined to 1 in the autoconf generated configuration file (usually called config.h). In the two negative situations, however, this macro will not be defined, thus it will evaluate to 0 in C preprocessor expressions. gettext.h is a convenience header file for conditional use of <libintl.h>, depending on the ENABLE_NLS macro. If ENABLE_NLS is set, it includes <libintl.h>; otherwise it defines no-op substitutes for the libintl.h functions. We recommend the use of "gettext.h" over direct use of <libintl.h>, so that portability to older systems is guaranteed and installers can turn off internationalization if they want to. In the C code, you will then write  #include "gettext.h"  instead of  #include <libintl.h>  The location of gettext.h is usually in a directory containing auxiliary include files. In many GNU packages, there is a directory lib/ containing helper functions; gettext.h fits there. In other packages, it can go into the src directory. Do not install the gettext.h file in public locations. Every package that needs it should contain a copy of it on its own. Next: , Previous: Adjusting Files, Up: Maintainers ### 13.5 Autoconf macros for use in configure.ac GNU gettext installs macros for use in a package's configure.ac or configure.in. See Introduction. The primary macro is, of course, AM_GNU_GETTEXT. Next: , Previous: autoconf macros, Up: autoconf macros #### 13.5.1 AM_GNU_GETTEXT in gettext.m4 The AM_GNU_GETTEXT macro tests for the presence of the GNU gettext function family in either the C library or a separate libintl library (shared or static libraries are both supported) or in the package's intl/ directory. It also invokes AM_PO_SUBDIRS, thus preparing the po/ directories of the package for building. AM_GNU_GETTEXT accepts up to three optional arguments. The general syntax is  AM_GNU_GETTEXT([intlsymbol], [needsymbol], [intldir])  intlsymbol can be ‘external’ or ‘no-libtool’. The default (if it is not specified or empty) is ‘no-libtool’. intlsymbol should be ‘external’ for packages with no intl/ directory. For packages with an intl/ directory, you can either use an intlsymbol equal to ‘no-libtool’, or you can use ‘external’ and override by using the macro AM_GNU_GETTEXT_INTL_SUBDIR elsewhere. The two ways to specify the existence of an intl/ directory are equivalent. At build time, a static library (top_builddir)/intl/libintl.a will then be created. If needsymbol is specified and is ‘need-ngettext’, then GNU gettext implementations (in libc or libintl) without the ngettext() function will be ignored. If needsymbol is specified and is ‘need-formatstring-macros’, then GNU gettext implementations that don't support the ISO C 99 <inttypes.h> formatstring macros will be ignored. Only one needsymbol can be specified. These requirements can also be specified by using the macro AM_GNU_GETTEXT_NEED elsewhere. To specify more than one requirement, just specify the strongest one among them, or invoke the AM_GNU_GETTEXT_NEED macro several times. The hierarchy among the various alternatives is as follows: ‘need-formatstring-macros’ implies ‘need-ngettext’. intldir is used to find the intl libraries. If empty, the value ‘(top_builddir)/intl/’ is used. The AM_GNU_GETTEXT macro determines whether GNU gettext is available and should be used. If so, it sets the USE_NLS variable to ‘yes’; it defines ENABLE_NLS to 1 in the autoconf generated configuration file (usually called config.h); it sets the variables LIBINTL and LTLIBINTL to the linker options for use in a Makefile (LIBINTL for use without libtool, LTLIBINTL for use with libtool); it adds an ‘-I’ option to CPPFLAGS if necessary. In the negative case, it sets USE_NLS to ‘no’; it sets LIBINTL and LTLIBINTL to empty and doesn't change CPPFLAGS. The complexities that AM_GNU_GETTEXT deals with are the following: • Some operating systems have gettext in the C library, for example glibc. Some have it in a separate library libintl. GNU libintl might have been installed as part of the GNU gettext package. • GNU libintl, if installed, is not necessarily already in the search path (CPPFLAGS for the include file search path, LDFLAGS for the library search path). • Except for glibc, the operating system's native gettext cannot exploit the GNU mo files, doesn't have the necessary locale dependency features, and cannot convert messages from the catalog's text encoding to the user's locale encoding. • GNU libintl, if installed, is not necessarily already in the run time library search path. To avoid the need for setting an environment variable like LD_LIBRARY_PATH, the macro adds the appropriate run time search path options to the LIBINTL and LTLIBINTL variables. This works on most systems, but not on some operating systems with limited shared library support, like SCO. • GNU libintl relies on POSIX/XSI iconv. The macro checks for linker options needed to use iconv and appends them to the LIBINTL and LTLIBINTL variables. Next: , Previous: AM_GNU_GETTEXT, Up: autoconf macros #### 13.5.2 AM_GNU_GETTEXT_VERSION in gettext.m4 The AM_GNU_GETTEXT_VERSION macro declares the version number of the GNU gettext infrastructure that is used by the package. The use of this macro is optional; only the autopoint program makes use of it (see CVS Issues). Next: , Previous: AM_GNU_GETTEXT_VERSION, Up: autoconf macros #### 13.5.3 AM_GNU_GETTEXT_NEED in gettext.m4 The AM_GNU_GETTEXT_NEED macro declares a constraint regarding the GNU gettext implementation. The syntax is  AM_GNU_GETTEXT_NEED([needsymbol])  If needsymbol is ‘need-ngettext’, then GNU gettext implementations (in libc or libintl) without the ngettext() function will be ignored. If needsymbol is ‘need-formatstring-macros’, then GNU gettext implementations that don't support the ISO C 99 <inttypes.h> formatstring macros will be ignored. The optional second argument of AM_GNU_GETTEXT is also taken into account. The AM_GNU_GETTEXT_NEED invocations can occur before or after the AM_GNU_GETTEXT invocation; the order doesn't matter. Next: , Previous: AM_GNU_GETTEXT_NEED, Up: autoconf macros #### 13.5.4 AM_GNU_GETTEXT_INTL_SUBDIR in intldir.m4 The AM_GNU_GETTEXT_INTL_SUBDIR macro specifies that the AM_GNU_GETTEXT macro, although invoked with the first argument ‘external’, should also prepare for building the intl/ subdirectory. The AM_GNU_GETTEXT_INTL_SUBDIR invocation can occur before or after the AM_GNU_GETTEXT invocation; the order doesn't matter. The use of this macro requires GNU automake 1.10 or newer and GNU autoconf 2.61 or newer. Next: , Previous: AM_GNU_GETTEXT_INTL_SUBDIR, Up: autoconf macros #### 13.5.5 AM_PO_SUBDIRS in po.m4 The AM_PO_SUBDIRS macro prepares the po/ directories of the package for building. This macro should be used in internationalized programs written in other programming languages than C, C++, Objective C, for example sh, Python, Lisp. See Programming Languages for a list of programming languages that support localization through PO files. The AM_PO_SUBDIRS macro determines whether internationalization should be used. If so, it sets the USE_NLS variable to ‘yes’, otherwise to ‘no’. It also determines the right values for Makefile variables in each po/ directory. Next: , Previous: AM_PO_SUBDIRS, Up: autoconf macros #### 13.5.6 AM_XGETTEXT_OPTION in po.m4 The AM_XGETTEXT_OPTION macro registers a command-line option to be used in the invocations of xgettext in the po/ directories of the package. For example, if you have a source file that defines a function ‘error_at_line’ whose fifth argument is a format string, you can use  AM_XGETTEXT_OPTION([--flag=error_at_line:5:c-format])  to instruct xgettext to mark all translatable strings in ‘gettext’ invocations that occur as fifth argument to this function as ‘c-format’. See xgettext Invocation for the list of options that xgettext accepts. The use of this macro is an alternative to the use of the ‘XGETTEXT_OPTIONS’ variable in po/Makevars. Previous: AM_XGETTEXT_OPTION, Up: autoconf macros #### 13.5.7 AM_ICONV in iconv.m4 The AM_ICONV macro tests for the presence of the POSIX/XSI iconv function family in either the C library or a separate libiconv library. If found, it sets the am_cv_func_iconv variable to ‘yes’; it defines HAVE_ICONV to 1 in the autoconf generated configuration file (usually called config.h); it defines ICONV_CONST to ‘const’ or to empty, depending on whether the second argument of iconv() is of type ‘const char **’ or ‘char **’; it sets the variables LIBICONV and LTLIBICONV to the linker options for use in a Makefile (LIBICONV for use without libtool, LTLIBICONV for use with libtool); it adds an ‘-I’ option to CPPFLAGS if necessary. If not found, it sets LIBICONV and LTLIBICONV to empty and doesn't change CPPFLAGS. The complexities that AM_ICONV deals with are the following: • Some operating systems have iconv in the C library, for example glibc. Some have it in a separate library libiconv, for example OSF/1 or FreeBSD. Regardless of the operating system, GNU libiconv might have been installed. In that case, it should be used instead of the operating system's native iconv. • GNU libiconv, if installed, is not necessarily already in the search path (CPPFLAGS for the include file search path, LDFLAGS for the library search path). • GNU libiconv is binary incompatible with some operating system's native iconv, for example on FreeBSD. Use of an iconv.h and libiconv.so that don't fit together would produce program crashes. • GNU libiconv, if installed, is not necessarily already in the run time library search path. To avoid the need for setting an environment variable like LD_LIBRARY_PATH, the macro adds the appropriate run time search path options to the LIBICONV variable. This works on most systems, but not on some operating systems with limited shared library support, like SCO. iconv.m4 is distributed with the GNU gettext package because gettext.m4 relies on it. Next: , Previous: autoconf macros, Up: Maintainers ### 13.6 Integrating with CVS Many projects use CVS for distributed development, version control and source backup. This section gives some advice how to manage the uses of cvs, gettextize, autopoint and autoconf. Next: , Previous: CVS Issues, Up: CVS Issues #### 13.6.1 Avoiding version mismatch in distributed development In a project development with multiple developers, using CVS, there should be a single developer who occasionally - when there is desire to upgrade to a new gettext version - runs gettextize and performs the changes listed in Adjusting Files, and then commits his changes to the CVS. It is highly recommended that all developers on a project use the same version of GNU gettext in the package. In other words, if a developer runs gettextize, he should go the whole way, make the necessary remaining changes and commit his changes to the CVS. Otherwise the following damages will likely occur: • Apparent version mismatch between developers. Since some gettext specific portions in configure.ac, configure.in and Makefile.am, Makefile.in files depend on the gettext version, the use of infrastructure files belonging to different gettext versions can easily lead to build errors. • Hidden version mismatch. Such version mismatch can also lead to malfunctioning of the package, that may be undiscovered by the developers. The worst case of hidden version mismatch is that internationalization of the package doesn't work at all. • Release risks. All developers implicitly perform constant testing on a package. This is important in the days and weeks before a release. If the guy who makes the release tar files uses a different version of GNU gettext than the other developers, the distribution will be less well tested than if all had been using the same gettext version. For example, it is possible that a platform specific bug goes undiscovered due to this constellation. Next: , Previous: Distributed CVS, Up: CVS Issues #### 13.6.2 Files to put under CVS version control There are basically three ways to deal with generated files in the context of a CVS repository, such as configure generated from configure.ac, parser.c generated from parser.y, or po/Makefile.in.in autoinstalled by gettextize or autopoint. 1. All generated files are always committed into the repository. 2. All generated files are committed into the repository occasionally, for example each time a release is made. 3. Generated files are never committed into the repository. Each of these three approaches has different advantages and drawbacks. 1. The advantage is that anyone can check out the CVS at any moment and gets a working build. The drawbacks are: 1a. It requires some frequent "cvs commit" actions by the maintainers. 1b. The repository grows in size quite fast. 2. The advantage is that anyone can check out the CVS, and the usual "./configure; make" will work. The drawbacks are: 2a. The one who checks out the repository needs tools like GNU automake, GNU autoconf, GNU m4 installed in his PATH; sometimes he even needs particular versions of them. 2b. When a release is made and a commit is made on the generated files, the other developers get conflicts on the generated files after doing "cvs update". Although these conflicts are easy to resolve, they are annoying. 3. The advantage is less work for the maintainers. The drawback is that anyone who checks out the CVS not only needs tools like GNU automake, GNU autoconf, GNU m4 installed in his PATH, but also that he needs to perform a package specific pre-build step before being able to "./configure; make". For the first and second approach, all files modified or brought in by the occasional gettextize invocation and update should be committed into the CVS. For the third approach, the maintainer can omit from the CVS repository all the files that gettextize mentions as "copy". Instead, he adds to the configure.ac or configure.in a line of the form  AM_GNU_GETTEXT_VERSION(0.17)  and adds to the package's pre-build script an invocation of ‘autopoint’. For everyone who checks out the CVS, this autopoint invocation will copy into the right place the gettext infrastructure files that have been omitted from the CVS. The version number used as argument to AM_GNU_GETTEXT_VERSION is the version of the gettext infrastructure that the package wants to use. It is also the minimum version number of the ‘autopoint’ program. So, if you write AM_GNU_GETTEXT_VERSION(0.11.5) then the developers can have any version >= 0.11.5 installed; the package will work with the 0.11.5 infrastructure in all developers' builds. When the maintainer then runs gettextize from, say, version 0.12.1 on the package, the occurrence of AM_GNU_GETTEXT_VERSION(0.11.5) will be changed into AM_GNU_GETTEXT_VERSION(0.12.1), and all other developers that use the CVS will henceforth need to have GNU gettext 0.12.1 or newer installed. Previous: Files under CVS, Up: CVS Issues #### 13.6.3 Invoking the autopoint Program  autopoint [option]...  The autopoint program copies standard gettext infrastructure files into a source package. It extracts from a macro call of the form AM_GNU_GETTEXT_VERSION(version), found in the package's configure.in or configure.ac file, the gettext version used by the package, and copies the infrastructure files belonging to this version into the package. ##### 13.6.3.1 Options -f --force Force overwriting of files that already exist. -n --dry-run Print modifications but don't perform them. All file copying actions that autopoint would normally execute are inhibited and instead only listed on standard output. ##### 13.6.3.2 Informative output --help Display this help and exit. --version Output version information and exit. autopoint supports the GNU gettext versions from 0.10.35 to the current one, 0.17. In order to apply autopoint to a package using a gettext version newer than 0.17, you need to install this same version of GNU gettext at least. In packages using GNU automake, an invocation of autopoint should be followed by invocations of aclocal and then autoconf and autoheader. The reason is that autopoint installs some autoconf macro files, which are used by aclocal to create aclocal.m4, and the latter is used by autoconf to create the package's configure script and by autoheader to create the package's config.h.in include file template. The name ‘autopoint’ is an abbreviation of ‘auto-po-intl-m4’; the tool copies or updates mostly files in the po, intl, m4 directories. Previous: CVS Issues, Up: Maintainers ### 13.7 Creating a Distribution Tarball In projects that use GNU automake, the usual commands for creating a distribution tarball, ‘make dist’ or ‘make distcheck’, automatically update the PO files as needed. If GNU automake is not used, the maintainer needs to perform this update before making a release:   ./configure  (cd po; make update-po)  make distclean  Next: , Previous: Maintainers, Up: Top ## 14 The Installer's and Distributor's View By default, packages fully using GNU gettext, internally, are installed in such a way that they to allow translation of messages. At configuration time, those packages should automatically detect whether the underlying host system already provides the GNU gettext functions. If not, the GNU gettext library should be automatically prepared and used. Installers may use special options at configuration time for changing this behavior. The command ‘./configure --with-included-gettext’ bypasses system gettext to use the included GNU gettext instead, while ‘./configure --disable-nls’ produces programs totally unable to translate messages. Internationalized packages have usually many ll.po files. Unless translations are disabled, all those available are installed together with the package. However, the environment variable LINGUAS may be set, prior to configuration, to limit the installed set. LINGUAS should then contain a space separated list of two-letter codes, stating which languages are allowed. Next: , Previous: Installers, Up: Top ## 15 Other Programming Languages While the presentation of gettext focuses mostly on C and implicitly applies to C++ as well, its scope is far broader than that: Many programming languages, scripting languages and other textual data like GUI resources or package descriptions can make use of the gettext approach. Next: , Previous: Programming Languages, Up: Programming Languages ### 15.1 The Language Implementor's View All programming and scripting languages that have the notion of strings are eligible to supporting gettext. Supporting gettext means the following: 1. You should add to the language a syntax for translatable strings. In principle, a function call of gettext would do, but a shorthand syntax helps keeping the legibility of internationalized programs. For example, in C we use the syntax _("string"), and in GNU awk we use the shorthand _"string". 2. You should arrange that evaluation of such a translatable string at runtime calls the gettext function, or performs equivalent processing. 3. Similarly, you should make the functions ngettext, dcgettext, dcngettext available from within the language. These functions are less often used, but are nevertheless necessary for particular purposes: ngettext for correct plural handling, and dcgettext and dcngettext for obeying other locale-related environment variables than LC_MESSAGES, such as LC_TIME or LC_MONETARY. For these latter functions, you need to make the LC_* constants, available in the C header <locale.h>, referenceable from within the language, usually either as enumeration values or as strings. 4. You should allow the programmer to designate a message domain, either by making the textdomain function available from within the language, or by introducing a magic variable called TEXTDOMAIN. Similarly, you should allow the programmer to designate where to search for message catalogs, by providing access to the bindtextdomain function. 5. You should either perform a setlocale (LC_ALL, "") call during the startup of your language runtime, or allow the programmer to do so. Remember that gettext will act as a no-op if the LC_MESSAGES and LC_CTYPE locale categories are not both set. 6. A programmer should have a way to extract translatable strings from a program into a PO file. The GNU xgettext program is being extended to support very different programming languages. Please contact the GNU gettext maintainers to help them doing this. If the string extractor is best integrated into your language's parser, GNU xgettext can function as a front end to your string extractor. 7. The language's library should have a string formatting facility where the arguments of a format string are denoted by a positional number or a name. This is needed because for some languages and some messages with more than one substitutable argument, the translation will need to output the substituted arguments in different order. See c-format Flag. 8. If the language has more than one implementation, and not all of the implementations use gettext, but the programs should be portable across implementations, you should provide a no-i18n emulation, that makes the other implementations accept programs written for yours, without actually translating the strings. 9. To help the programmer in the task of marking translatable strings, which is sometimes performed using the Emacs PO mode (see Marking), you are welcome to contact the GNU gettext maintainers, so they can add support for your language to po-mode.el. On the implementation side, three approaches are possible, with different effects on portability and copyright: • You may integrate the GNU gettext's intl/ directory in your package, as described in Maintainers. This allows you to have internationalization on all kinds of platforms. Note that when you then distribute your package, it legally falls under the GNU General Public License, and the GNU project will be glad about your contribution to the Free Software pool. • You may link against GNU gettext functions if they are found in the C library. For example, an autoconf test for gettext() and ngettext() will detect this situation. For the moment, this test will succeed on GNU systems and not on other platforms. No severe copyright restrictions apply. • You may emulate or reimplement the GNU gettext functionality. This has the advantage of full portability and no copyright restrictions, but also the drawback that you have to reimplement the GNU gettext features (such as the LANGUAGE environment variable, the locale aliases database, the automatic charset conversion, and plural handling). Next: , Previous: Language Implementors, Up: Programming Languages ### 15.2 The Programmer's View For the programmer, the general procedure is the same as for the C language. The Emacs PO mode marking supports other languages, and the GNU xgettext string extractor recognizes other languages based on the file extension or a command-line option. In some languages, setlocale is not needed because it is already performed by the underlying language runtime. Next: , Previous: Programmers for other Languages, Up: Programming Languages ### 15.3 The Translator's View The translator works exactly as in the C language case. The only difference is that when translating format strings, she has to be aware of the language's particular syntax for positional arguments in format strings. Next: , Previous: Translators for other Languages, Up: Translators for other Languages #### 15.3.1 C Format Strings C format strings are described in POSIX (IEEE P1003.1 2001), section XSH 3 fprintf(), http://www.opengroup.org/onlinepubs/007904975/functions/fprintf.html. See also the fprintf() manual page, http://www.linuxvalley.it/encyclopedia/ldp/manpage/man3/printf.3.php, http://informatik.fh-wuerzburg.de/student/i510/man/printf.html. Although format strings with positions that reorder arguments, such as  "Only %2d bytes free on '%1s'."  which is semantically equivalent to  "'%s' has only %d bytes free."  are a POSIX/XSI feature and not specified by ISO C 99, translators can rely on this reordering ability: On the few platforms where printf(), fprintf() etc. don't support this feature natively, libintl.a or libintl.so provides replacement functions, and GNU <libintl.h> activates these replacement functions automatically. As a special feature for Farsi (Persian) and maybe Arabic, translators can insert an ‘I’ flag into numeric format directives. For example, the translation of "%d" can be "%Id". The effect of this flag, on systems with GNU libc, is that in the output, the ASCII digits are replaced with the ‘outdigits’ defined in the LC_CTYPE locale category. On other systems, the gettext function removes this flag, so that it has no effect. Note that the programmer should not put this flag into the untranslated string. (Putting the ‘I’ format directive flag into an msgid string would lead to undefined behaviour on platforms without glibc when NLS is disabled.) Next: , Previous: c-format, Up: Translators for other Languages #### 15.3.2 Objective C Format Strings Objective C format strings are like C format strings. They support an additional format directive: "@", which when executed consumes an argument of type Object *. Next: , Previous: objc-format, Up: Translators for other Languages #### 15.3.3 Shell Format Strings Shell format strings, as supported by GNU gettext and the ‘envsubst’ program, are strings with references to shell variables in the form variable or {variable}. References of the form {variable-default}, {variable:-default}, {variable=default}, {variable:=default}, {variable+replacement}, {variable:+replacement}, {variable?ignored}, {variable:?ignored}, that would be valid inside shell scripts, are not supported. The variable names must consist solely of alphanumeric or underscore ASCII characters, not start with a digit and be nonempty; otherwise such a variable reference is ignored. Next: , Previous: sh-format, Up: Translators for other Languages #### 15.3.4 Python Format Strings Python format strings are described in Python Library reference / 2. Built-in Types, Exceptions and Functions / 2.2. Built-in Types / 2.2.6. Sequence Types / 2.2.6.2. String Formatting Operations. http://www.python.org/doc/2.2.1/lib/typesseq-strings.html. Next: , Previous: python-format, Up: Translators for other Languages #### 15.3.5 Lisp Format Strings Lisp format strings are described in the Common Lisp HyperSpec, chapter 22.3 Formatted Output, http://www.lisp.org/HyperSpec/Body/sec_22-3.html. Next: , Previous: lisp-format, Up: Translators for other Languages #### 15.3.6 Emacs Lisp Format Strings Emacs Lisp format strings are documented in the Emacs Lisp reference, section Formatting Strings, http://www.gnu.org/manual/elisp-manual-21-2.8/html_chapter/elisp_4.html#SEC75. Note that as of version 21, XEmacs supports numbered argument specifications in format strings while FSF Emacs doesn't. Next: , Previous: elisp-format, Up: Translators for other Languages #### 15.3.7 librep Format Strings librep format strings are documented in the librep manual, section Formatted Output, http://librep.sourceforge.net/librep-manual.html#Formatted%20Output, http://www.gwinnup.org/research/docs/librep.html#SEC122. Next: , Previous: librep-format, Up: Translators for other Languages #### 15.3.8 Scheme Format Strings Scheme format strings are documented in the SLIB manual, section Format Specification. Next: , Previous: scheme-format, Up: Translators for other Languages #### 15.3.9 Smalltalk Format Strings Smalltalk format strings are described in the GNU Smalltalk documentation, class CharArray, methods ‘bindWith:’ and ‘bindWithArguments:’. http://www.gnu.org/software/smalltalk/gst-manual/gst_68.html#SEC238. In summary, a directive starts with ‘%’ and is followed by ‘%’ or a nonzero digit (‘1’ to ‘9’). Next: , Previous: smalltalk-format, Up: Translators for other Languages #### 15.3.10 Java Format Strings Java format strings are described in the JDK documentation for class java.text.MessageFormat, http://java.sun.com/j2se/1.4/docs/api/java/text/MessageFormat.html. See also the ICU documentation http://oss.software.ibm.com/icu/apiref/classMessageFormat.html. Next: , Previous: java-format, Up: Translators for other Languages #### 15.3.11 C# Format Strings C# format strings are described in the .NET documentation for class System.String and in http://msdn.microsoft.com/library/default.asp?url=/library/en-us/cpguide/html/cpConFormattingOverview.asp. Next: , Previous: csharp-format, Up: Translators for other Languages #### 15.3.12 awk Format Strings awk format strings are described in the gawk documentation, section Printf, http://www.gnu.org/manual/gawk/html_node/Printf.html#Printf. Next: , Previous: awk-format, Up: Translators for other Languages #### 15.3.13 Object Pascal Format Strings Where is this documented? Next: , Previous: object-pascal-format, Up: Translators for other Languages #### 15.3.14 YCP Format Strings YCP sformat strings are described in the libycp documentation file:/usr/share/doc/packages/libycp/YCP-builtins.html. In summary, a directive starts with ‘%’ and is followed by ‘%’ or a nonzero digit (‘1’ to ‘9’). Next: , Previous: ycp-format, Up: Translators for other Languages #### 15.3.15 Tcl Format Strings Tcl format strings are described in the format.n manual page, http://www.scriptics.com/man/tcl8.3/TclCmd/format.htm. Next: , Previous: tcl-format, Up: Translators for other Languages #### 15.3.16 Perl Format Strings There are two kinds format strings in Perl: those acceptable to the Perl built-in function printf, labelled as ‘perl-format’, and those acceptable to the libintl-perl function __x, labelled as ‘perl-brace-format’. Perl printf format strings are described in the sprintf section of ‘man perlfunc’. Perl brace format strings are described in the Locale::TextDomain(3pm) manual page of the CPAN package libintl-perl. In brief, Perl format uses placeholders put between braces (‘{’ and ‘}’). The placeholder must have the syntax of simple identifiers. Next: , Previous: perl-format, Up: Translators for other Languages #### 15.3.17 PHP Format Strings PHP format strings are described in the documentation of the PHP function sprintf, in phpdoc/manual/function.sprintf.html or http://www.php.net/manual/en/function.sprintf.php. Next: , Previous: php-format, Up: Translators for other Languages #### 15.3.18 GCC internal Format Strings These format strings are used inside the GCC sources. In such a format string, a directive starts with ‘%’, is optionally followed by a size specifier ‘l’, an optional flag ‘+’, another optional flag ‘#’, and is finished by a specifier: ‘%’ denotes a literal percent sign, ‘c’ denotes a character, ‘s’ denotes a string, ‘i’ and ‘d’ denote an integer, ‘o’, ‘u’, ‘x’ denote an unsigned integer, ‘.*s’ denotes a string preceded by a width specification, ‘H’ denotes a ‘location_t *’ pointer, ‘D’ denotes a general declaration, ‘F’ denotes a function declaration, ‘T’ denotes a type, ‘A’ denotes a function argument, ‘C’ denotes a tree code, ‘E’ denotes an expression, ‘L’ denotes a programming language, ‘O’ denotes a binary operator, ‘P’ denotes a function parameter, ‘Q’ denotes an assignment operator, ‘V’ denotes a const/volatile qualifier. Next: , Previous: gcc-internal-format, Up: Translators for other Languages #### 15.3.19 Qt Format Strings Qt format strings are described in the documentation of the QString class file:/usr/lib/qt-4.3.0/doc/html/qstring.html. In summary, a directive consists of a ‘%’ followed by a digit. The same directive cannot occur more than once in a format string. Next: , Previous: qt-format, Up: Translators for other Languages #### 15.3.20 KDE Format Strings KDE 4 format strings are defined as follows: A directive consists of a ‘%’ followed by a non-zero decimal number. If a ‘%n’ occurs in a format strings, all of ‘%1’, ..., ‘%(n-1)’ must occur as well, except possibly one of them. Previous: kde-format, Up: Translators for other Languages #### 15.3.21 Boost Format Strings Boost format strings are described in the documentation of the boost::format class, at http://www.boost.org/libs/format/doc/format.html. In summary, a directive has either the same syntax as in a C format string, such as ‘%1+5d’, or may be surrounded by vertical bars, such as ‘%|1+5d|’ or ‘%|1+5|’, or consists of just an argument number between percent signs, such as ‘%1%’. Next: , Previous: Translators for other Languages, Up: Programming Languages ### 15.4 The Maintainer's View For the maintainer, the general procedure differs from the C language case in two ways. • For those languages that don't use GNU gettext, the intl/ directory is not needed and can be omitted. This means that the maintainer calls the gettextize program without the ‘--intl’ option, and that he invokes the AM_GNU_GETTEXT autoconf macro via ‘AM_GNU_GETTEXT([external])’. • If only a single programming language is used, the XGETTEXT_OPTIONS variable in po/Makevars (see po/Makevars) should be adjusted to match the xgettext options for that particular programming language. If the package uses more than one programming language with gettext support, it becomes necessary to change the POT file construction rule in po/Makefile.in.in. It is recommended to make one xgettext invocation per programming language, each with the options appropriate for that language, and to combine the resulting files using msgcat. Next: , Previous: Maintainers for other Languages, Up: Programming Languages ### 15.5 Individual Programming Languages Next: , Previous: List of Programming Languages, Up: List of Programming Languages #### 15.5.1 C, C++, Objective C RPMs gcc, gpp, gobjc, glibc, gettext File extension For C: c, h. For C++: C, c++, cc, cxx, cpp, hpp. For Objective C: m. String syntax "abc" gettext shorthand _("abc") gettext/ngettext functions gettext, dgettext, dcgettext, ngettext, dngettext, dcngettext textdomain textdomain function bindtextdomain bindtextdomain function setlocale Programmer must call setlocale (LC_ALL, "") Prerequisite #include <libintl.h> #include <locale.h> #define _(string) gettext (string) Use or emulate GNU gettext Use Extractor xgettext -k_ Formatting with positions fprintf "%2d %1d" In C++: autosprintf "%2d %1d" (see Introduction) Portability autoconf (gettext.m4) and #if ENABLE_NLS po-mode marking yes The following examples are available in the examples directory: hello-c, hello-c-gnome, hello-c++, hello-c++-qt, hello-c++-kde, hello-c++-gnome, hello-c++-wxwidgets, hello-objc, hello-objc-gnustep, hello-objc-gnome. Next: , Previous: C, Up: List of Programming Languages #### 15.5.2 sh - Shell Script RPMs bash, gettext File extension sh String syntax "abc", 'abc', abc gettext shorthand "gettext \"abc\"" gettext/ngettext functions gettext, ngettext programs eval_gettext, eval_ngettext shell functions textdomain environment variable TEXTDOMAIN bindtextdomain environment variable TEXTDOMAINDIR setlocale automatic Prerequisite . gettext.sh Use or emulate GNU gettext use Extractor xgettext Formatting with positions Portability fully portable po-mode marking An example is available in the examples directory: hello-sh. Next: , Previous: sh, Up: sh ##### 15.5.2.1 Preparing Shell Scripts for Internationalization Preparing a shell script for internationalization is conceptually similar to the steps described in Sources. The concrete steps for shell scripts are as follows. 1. Insert the line  . gettext.sh  near the top of the script. gettext.sh is a shell function library that provides the functions eval_gettext (see eval_gettext Invocation) and eval_ngettext (see eval_ngettext Invocation). You have to ensure that gettext.sh can be found in the PATH. 2. Set and export the TEXTDOMAIN and TEXTDOMAINDIR environment variables. Usually TEXTDOMAIN is the package or program name, and TEXTDOMAINDIR is the absolute pathname corresponding to prefix/share/locale, where prefix is the installation location.  TEXTDOMAIN=@PACKAGE@ export TEXTDOMAIN TEXTDOMAINDIR=@LOCALEDIR@ export TEXTDOMAINDIR  3. Prepare the strings for translation, as described in Preparing Strings. 4. Simplify translatable strings so that they don't contain command substitution ("..." or "(...)"), variable access with defaulting (like {variable-default}), access to positional arguments (like 0, 1, ...) or highly volatile shell variables (like ?). This can always be done through simple local code restructuring. For example,  echo "Usage: 0 [OPTION] FILE..."  becomes  program_name=0 echo "Usage: program_name [OPTION] FILE..."  Similarly,  echo "Remaining files: ls | wc -l"  becomes  filecount="ls | wc -l" echo "Remaining files: filecount"  5. For each translatable string, change the output command ‘echo’ or ‘echo’ to ‘gettext’ (if the string contains no references to shell variables) or to ‘eval_gettext’ (if it refers to shell variables), followed by a no-argument ‘echo’ command (to account for the terminating newline). Similarly, for cases with plural handling, replace a conditional ‘echo’ command with an invocation of ‘ngettext’ or ‘eval_ngettext’, followed by a no-argument ‘echo’ command. When doing this, you also need to add an extra backslash before the dollar sign in references to shell variables, so that the ‘eval_gettext’ function receives the translatable string before the variable values are substituted into it. For example,  echo "Remaining files: filecount"  becomes  eval_gettext "Remaining files: \filecount"; echo  If the output command is not ‘echo’, you can make it use ‘echo’ nevertheless, through the use of backquotes. However, note that inside backquotes, backslashes must be doubled to be effective (because the backquoting eats one level of backslashes). For example, assuming that ‘error’ is a shell function that signals an error,  error "file not found: filename"  is first transformed into  error "echo \"file not found: \filename\""  which then becomes  error "eval_gettext \"file not found: \\\filename\""  Next: , Previous: Preparing Shell Scripts, Up: sh ##### 15.5.2.2 Contents of gettext.sh gettext.sh, contained in the run-time package of GNU gettext, provides the following: • echo The variable echo is set to a command that outputs its first argument and a newline, without interpreting backslashes in the argument string. • eval_gettext See eval_gettext Invocation. • eval_ngettext See eval_ngettext Invocation. Next: , Previous: gettext.sh, Up: sh ##### 15.5.2.3 Invoking the gettext program  gettext [option] [[textdomain] msgid] gettext [option] -s [msgid]...  The gettext program displays the native language translation of a textual message. Arguments -d textdomain --domain=textdomain Retrieve translated messages from textdomain. Usually a textdomain corresponds to a package, a program, or a module of a program. -e Enable expansion of some escape sequences. This option is for compatibility with the ‘echo’ program or shell built-in. The escape sequences ‘\a’, ‘\b’, ‘\c’, ‘\f’, ‘\n’, ‘\r’, ‘\t’, ‘\v’, ‘\\’, and ‘\’ followed by one to three octal digits, are interpreted like the System V ‘echo’ program did. -E This option is only for compatibility with the ‘echo’ program or shell built-in. It has no effect. -h --help Display this help and exit. -n Suppress trailing newline. By default, gettext adds a newline to the output. -V --version Output version information and exit. [textdomain] msgid Retrieve translated message corresponding to msgid from textdomain. If the textdomain parameter is not given, the domain is determined from the environment variable TEXTDOMAIN. If the message catalog is not found in the regular directory, another location can be specified with the environment variable TEXTDOMAINDIR. When used with the -s option the program behaves like the ‘echo’ command. But it does not simply copy its arguments to stdout. Instead those messages found in the selected catalog are translated. Next: , Previous: gettext Invocation, Up: sh ##### 15.5.2.4 Invoking the ngettext program  ngettext [option] [textdomain] msgid msgid-plural count  The ngettext program displays the native language translation of a textual message whose grammatical form depends on a number. Arguments -d textdomain --domain=textdomain Retrieve translated messages from textdomain. Usually a textdomain corresponds to a package, a program, or a module of a program. -e Enable expansion of some escape sequences. This option is for compatibility with the ‘gettext’ program. The escape sequences ‘\a’, ‘\b’, ‘\c’, ‘\f’, ‘\n’, ‘\r’, ‘\t’, ‘\v’, ‘\\’, and ‘\’ followed by one to three octal digits, are interpreted like the System V ‘echo’ program did. -E This option is only for compatibility with the ‘gettext’ program. It has no effect. -h --help Display this help and exit. -V --version Output version information and exit. textdomain Retrieve translated message from textdomain. msgid msgid-plural Translate msgid (English singular) / msgid-plural (English plural). count Choose singular/plural form based on this value. If the textdomain parameter is not given, the domain is determined from the environment variable TEXTDOMAIN. If the message catalog is not found in the regular directory, another location can be specified with the environment variable TEXTDOMAINDIR. Next: , Previous: ngettext Invocation, Up: sh ##### 15.5.2.5 Invoking the envsubst program  envsubst [option] [shell-format]  The envsubst program substitutes the values of environment variables. Operation mode -v --variables Output the variables occurring in shell-format. Informative output -h --help Display this help and exit. -V --version Output version information and exit. In normal operation mode, standard input is copied to standard output, with references to environment variables of the form VARIABLE or {VARIABLE} being replaced with the corresponding values. If a shell-format is given, only those environment variables that are referenced in shell-format are substituted; otherwise all environment variables references occurring in standard input are substituted. These substitutions are a subset of the substitutions that a shell performs on unquoted and double-quoted strings. Other kinds of substitutions done by a shell, such as {variable-default} or (command-list) or command-list, are not performed by the envsubst program, due to security reasons. When --variables is used, standard input is ignored, and the output consists of the environment variables that are referenced in shell-format, one per line. Next: , Previous: envsubst Invocation, Up: sh ##### 15.5.2.6 Invoking the eval_gettext function  eval_gettext msgid  This function outputs the native language translation of a textual message, performing dollar-substitution on the result. Note that only shell variables mentioned in msgid will be dollar-substituted in the result. Previous: eval_gettext Invocation, Up: sh ##### 15.5.2.7 Invoking the eval_ngettext function  eval_ngettext msgid msgid-plural count  This function outputs the native language translation of a textual message whose grammatical form depends on a number, performing dollar-substitution on the result. Note that only shell variables mentioned in msgid or msgid-plural will be dollar-substituted in the result. Next: , Previous: sh, Up: List of Programming Languages #### 15.5.3 bash - Bourne-Again Shell Script GNU bash 2.0 or newer has a special shorthand for translating a string and substituting variable values in it: "msgid". But the use of this construct is discouraged, due to the security holes it opens and due to its portability problems. The security holes of "..." come from the fact that after looking up the translation of the string, bash processes it like it processes any double-quoted string: dollar and backquote processing, like ‘eval’ does. 1. In a locale whose encoding is one of BIG5, BIG5-HKSCS, GBK, GB18030, SHIFT_JIS, JOHAB, some double-byte characters have a second byte whose value is 0x60. For example, the byte sequence \xe0\x60 is a single character in these locales. Many versions of bash (all versions up to bash-2.05, and newer versions on platforms without mbsrtowcs() function) don't know about character boundaries and see a backquote character where there is only a particular Chinese character. Thus it can start executing part of the translation as a command list. This situation can occur even without the translator being aware of it: if the translator provides translations in the UTF-8 encoding, it is the gettext() function which will, during its conversion from the translator's encoding to the user's locale's encoding, produce the dangerous \x60 bytes. 2. A translator could - voluntarily or inadvertently - use backquotes "..." or dollar-parentheses "(...)" in her translations. The enclosed strings would be executed as command lists by the shell. The portability problem is that bash must be built with internationalization support; this is normally not the case on systems that don't have the gettext() function in libc. Next: , Previous: bash, Up: List of Programming Languages #### 15.5.4 Python RPMs python File extension py String syntax 'abc', u'abc', r'abc', ur'abc', "abc", u"abc", r"abc", ur"abc", '''abc''', u'''abc''', r'''abc''', ur'''abc''', """abc""", u"""abc""", r"""abc""", ur"""abc""" gettext shorthand _('abc') etc. gettext/ngettext functions gettext.gettext, gettext.dgettext, gettext.ngettext, gettext.dngettext, also ugettext, ungettext textdomain gettext.textdomain function, or gettext.install(domain) function bindtextdomain gettext.bindtextdomain function, or gettext.install(domain,localedir) function setlocale not used by the gettext emulation Prerequisite import gettext Use or emulate GNU gettext emulate Extractor xgettext Formatting with positions '...%(ident)d...' % { 'ident': value } Portability fully portable po-mode marking An example is available in the examples directory: hello-python. Next: , Previous: Python, Up: List of Programming Languages #### 15.5.5 GNU clisp - Common Lisp RPMs clisp 2.28 or newer File extension lisp String syntax "abc" gettext shorthand (_ "abc"), (ENGLISH "abc") gettext/ngettext functions i18n:gettext, i18n:ngettext textdomain i18n:textdomain bindtextdomain i18n:textdomaindir setlocale automatic Prerequisite Use or emulate GNU gettext use Extractor xgettext -k_ -kENGLISH Formatting with positions format "~1@*~D ~0@*~D" Portability On platforms without gettext, no translation. po-mode marking An example is available in the examples directory: hello-clisp. Next: , Previous: Common Lisp, Up: List of Programming Languages #### 15.5.6 GNU clisp C sources RPMs clisp File extension d String syntax "abc" gettext shorthand ENGLISH ? "abc" : "" GETTEXT("abc") GETTEXTL("abc") gettext/ngettext functions clgettext, clgettextl textdomain bindtextdomain setlocale automatic Prerequisite #include "lispbibl.c" Use or emulate GNU gettext use Extractor clisp-xgettext Formatting with positions fprintf "%2d %1d" Portability On platforms without gettext, no translation. po-mode marking Next: , Previous: clisp C, Up: List of Programming Languages #### 15.5.7 Emacs Lisp RPMs emacs, xemacs File extension el String syntax "abc" gettext shorthand (_"abc") gettext/ngettext functions gettext, dgettext (xemacs only) textdomain domain special form (xemacs only) bindtextdomain bind-text-domain function (xemacs only) setlocale automatic Prerequisite Use or emulate GNU gettext use Extractor xgettext Formatting with positions format "%2d %1d" Portability Only XEmacs. Without I18N3 defined at build time, no translation. po-mode marking Next: , Previous: Emacs Lisp, Up: List of Programming Languages #### 15.5.8 librep RPMs librep 0.15.3 or newer File extension jl String syntax "abc" gettext shorthand (_"abc") gettext/ngettext functions gettext textdomain textdomain function bindtextdomain bindtextdomain function setlocale Prerequisite (require 'rep.i18n.gettext) Use or emulate GNU gettext use Extractor xgettext Formatting with positions format "%2d %1d" Portability On platforms without gettext, no translation. po-mode marking An example is available in the examples directory: hello-librep. Next: , Previous: librep, Up: List of Programming Languages #### 15.5.9 GNU guile - Scheme RPMs guile File extension scm String syntax "abc" gettext shorthand (_ "abc") gettext/ngettext functions gettext, ngettext textdomain textdomain bindtextdomain bindtextdomain setlocale (catch #t (lambda () (setlocale LC_ALL "")) (lambda args #f)) Prerequisite (use-modules (ice-9 format)) Use or emulate GNU gettext use Extractor xgettext -k_ Formatting with positions Portability On platforms without gettext, no translation. po-mode marking An example is available in the examples directory: hello-guile. Next: , Previous: Scheme, Up: List of Programming Languages #### 15.5.10 GNU Smalltalk RPMs smalltalk File extension st String syntax 'abc' gettext shorthand NLS ? 'abc' gettext/ngettext functions LcMessagesDomain>>#at:, LcMessagesDomain>>#at:plural:with: textdomain LcMessages>>#domain:localeDirectory: (returns a LcMessagesDomain object). Example: I18N Locale default messages domain: 'gettext' localeDirectory: /usr/local/share/locale' bindtextdomain LcMessages>>#domain:localeDirectory:, see above. setlocale Automatic if you use I18N Locale default. Prerequisite PackageLoader fileInPackage: 'I18N'! Use or emulate GNU gettext emulate Extractor xgettext Formatting with positions '%1 %2' bindWith: 'Hello' with: 'world' Portability fully portable po-mode marking An example is available in the examples directory: hello-smalltalk. Next: , Previous: Smalltalk, Up: List of Programming Languages #### 15.5.11 Java RPMs java, java2 File extension java String syntax "abc" gettext shorthand _("abc") gettext/ngettext functions GettextResource.gettext, GettextResource.ngettext, GettextResource.pgettext, GettextResource.npgettext textdomain —, use ResourceBundle.getResource instead bindtextdomain —, use CLASSPATH instead setlocale automatic Prerequisite Use or emulate GNU gettext —, uses a Java specific message catalog format Extractor xgettext -k_ Formatting with positions MessageFormat.format "{1,number} {0,number}" Portability fully portable po-mode marking Before marking strings as internationalizable, uses of the string concatenation operator need to be converted to MessageFormat applications. For example, "file "+filename+" not found" becomes MessageFormat.format("file {0} not found", new Object[] { filename }). Only after this is done, can the strings be marked and extracted. GNU gettext uses the native Java internationalization mechanism, namely ResourceBundles. There are two formats of ResourceBundles: .properties files and .class files. The .properties format is a text file which the translators can directly edit, like PO files, but which doesn't support plural forms. Whereas the .class format is compiled from .java source code and can support plural forms (provided it is accessed through an appropriate API, see below). To convert a PO file to a .properties file, the msgcat program can be used with the option --properties-output. To convert a .properties file back to a PO file, the msgcat program can be used with the option --properties-input. All the tools that manipulate PO files can work with .properties files as well, if given the --properties-input and/or --properties-output option. To convert a PO file to a ResourceBundle class, the msgfmt program can be used with the option --java or --java2. To convert a ResourceBundle back to a PO file, the msgunfmt program can be used with the option --java. Two different programmatic APIs can be used to access ResourceBundles. Note that both APIs work with all kinds of ResourceBundles, whether GNU gettext generated classes, or other .class or .properties files. 1. The java.util.ResourceBundle API. In particular, its getString function returns a string translation. Note that a missing translation yields a MissingResourceException. This has the advantage of being the standard API. And it does not require any additional libraries, only the msgcat generated .properties files or the msgfmt generated .class files. But it cannot do plural handling, even if the resource was generated by msgfmt from a PO file with plural handling. 2. The gnu.gettext.GettextResource API. Reference documentation in Javadoc 1.1 style format is in the javadoc2 directory. Its gettext function returns a string translation. Note that when a translation is missing, the msgid argument is returned unchanged. This has the advantage of having the ngettext function for plural handling and the pgettext and npgettext for strings constraint to a particular context. To use this API, one needs the libintl.jar file which is part of the GNU gettext package and distributed under the LGPL. Four examples, using the second API, are available in the examples directory: hello-java, hello-java-awt, hello-java-swing, hello-java-qtjambi. Now, to make use of the API and define a shorthand for ‘getString’, there are three idioms that you can choose from: • (This one assumes Java 1.5 or newer.) In a unique class of your project, say ‘Util’, define a static variable holding the ResourceBundle instance and the shorthand:  private static ResourceBundle myResources = ResourceBundle.getBundle("domain-name"); public static String _(String s) { return myResources.getString(s); }  All classes containing internationalized strings then contain  import static Util._;  and the shorthand is used like this:  System.out.println(_("Operation completed."));  • In a unique class of your project, say ‘Util’, define a static variable holding the ResourceBundle instance:  public static ResourceBundle myResources = ResourceBundle.getBundle("domain-name");  All classes containing internationalized strings then contain  private static ResourceBundle res = Util.myResources; private static String _(String s) { return res.getString(s); }  and the shorthand is used like this:  System.out.println(_("Operation completed."));  • You add a class with a very short name, say ‘S’, containing just the definition of the resource bundle and of the shorthand:  public class S { public static ResourceBundle myResources = ResourceBundle.getBundle("domain-name"); public static String _(String s) { return myResources.getString(s); } }  and the shorthand is used like this:  System.out.println(S._("Operation completed."));  Which of the three idioms you choose, will depend on whether your project requires portability to Java versions prior to Java 1.5 and, if so, whether copying two lines of codes into every class is more acceptable in your project than a class with a single-letter name. Next: , Previous: Java, Up: List of Programming Languages #### 15.5.12 C# RPMs pnet, pnetlib 0.6.2 or newer, or mono 0.29 or newer File extension cs String syntax "abc", @"abc" gettext shorthand _("abc") gettext/ngettext functions GettextResourceManager.GetString, GettextResourceManager.GetPluralString GettextResourceManager.GetParticularString GettextResourceManager.GetParticularPluralString textdomain new GettextResourceManager(domain) bindtextdomain —, compiled message catalogs are located in subdirectories of the directory containing the executable setlocale automatic Prerequisite Use or emulate GNU gettext —, uses a C# specific message catalog format Extractor xgettext -k_ Formatting with positions String.Format "{1} {0}" Portability fully portable po-mode marking Before marking strings as internationalizable, uses of the string concatenation operator need to be converted to String.Format invocations. For example, "file "+filename+" not found" becomes String.Format("file {0} not found", filename). Only after this is done, can the strings be marked and extracted. GNU gettext uses the native C#/.NET internationalization mechanism, namely the classes ResourceManager and ResourceSet. Applications use the ResourceManager methods to retrieve the native language translation of strings. An instance of ResourceSet is the in-memory representation of a message catalog file. The ResourceManager loads and accesses ResourceSet instances as needed to look up the translations. There are two formats of ResourceSets that can be directly loaded by the C# runtime: .resources files and .dll files. • The .resources format is a binary file usually generated through the resgen or monoresgen utility, but which doesn't support plural forms. .resources files can also be embedded in .NET .exe files. This only affects whether a file system access is performed to load the message catalog; it doesn't affect the contents of the message catalog. • On the other hand, the .dll format is a binary file that is compiled from .cs source code and can support plural forms (provided it is accessed through the GNU gettext API, see below). Note that these .NET .dll and .exe files are not tied to a particular platform; their file format and GNU gettext for C# can be used on any platform. To convert a PO file to a .resources file, the msgfmt program can be used with the option ‘--csharp-resources’. To convert a .resources file back to a PO file, the msgunfmt program can be used with the option ‘--csharp-resources’. You can also, in some cases, use the resgen program (from the pnet package) or the monoresgen program (from the mono/mcs package). These programs can also convert a .resources file back to a PO file. But beware: as of this writing (January 2004), the monoresgen converter is quite buggy and the resgen converter ignores the encoding of the PO files. To convert a PO file to a .dll file, the msgfmt program can be used with the option --csharp. The result will be a .dll file containing a subclass of GettextResourceSet, which itself is a subclass of ResourceSet. To convert a .dll file containing a GettextResourceSet subclass back to a PO file, the msgunfmt program can be used with the option --csharp. The advantages of the .dll format over the .resources format are: 1. Freedom to localize: Users can add their own translations to an application after it has been built and distributed. Whereas when the programmer uses a ResourceManager constructor provided by the system, the set of .resources files for an application must be specified when the application is built and cannot be extended afterwards. 2. Plural handling: A message catalog in .dll format supports the plural handling function GetPluralString. Whereas .resources files can only contain data and only support lookups that depend on a single string. 3. Context handling: A message catalog in .dll format supports the query-with-context functions GetParticularString and GetParticularPluralString. Whereas .resources files can only contain data and only support lookups that depend on a single string. 4. The GettextResourceManager that loads the message catalogs in .dll format also provides for inheritance on a per-message basis. For example, in Austrian (de_AT) locale, translations from the German (de) message catalog will be used for messages not found in the Austrian message catalog. This has the consequence that the Austrian translators need only translate those few messages for which the translation into Austrian differs from the German one. Whereas when working with .resources files, each message catalog must provide the translations of all messages by itself. 5. The GettextResourceManager that loads the message catalogs in .dll format also provides for a fallback: The English msgid is returned when no translation can be found. Whereas when working with .resources files, a language-neutral .resources file must explicitly be provided as a fallback. On the side of the programmatic APIs, the programmer can use either the standard ResourceManager API and the GNU GettextResourceManager API. The latter is an extension of the former, because GettextResourceManager is a subclass of ResourceManager. 1. The System.Resources.ResourceManager API. This API works with resources in .resources format. The creation of the ResourceManager is done through  new ResourceManager(domainname, Assembly.GetExecutingAssembly())  The GetString function returns a string's translation. Note that this function returns null when a translation is missing (i.e. not even found in the fallback resource file). 2. The GNU.Gettext.GettextResourceManager API. This API works with resources in .dll format. Reference documentation is in the csharpdoc directory. The creation of the ResourceManager is done through  new GettextResourceManager(domainname)  The GetString function returns a string's translation. Note that when a translation is missing, the msgid argument is returned unchanged. The GetPluralString function returns a string translation with plural handling, like the ngettext function in C. The GetParticularString function returns a string's translation, specific to a particular context, like the pgettext function in C. Note that when a translation is missing, the msgid argument is returned unchanged. The GetParticularPluralString function returns a string translation, specific to a particular context, with plural handling, like the npgettext function in C. To use this API, one needs the GNU.Gettext.dll file which is part of the GNU gettext package and distributed under the LGPL. You can also mix both approaches: use the GNU.Gettext.GettextResourceManager constructor, but otherwise use only the ResourceManager type and only the GetString method. This is appropriate when you want to profit from the tools for PO files, but don't want to change an existing source code that uses ResourceManager and don't (yet) need the GetPluralString method. Two examples, using the second API, are available in the examples directory: hello-csharp, hello-csharp-forms. Now, to make use of the API and define a shorthand for ‘GetString’, there are two idioms that you can choose from: • In a unique class of your project, say ‘Util’, define a static variable holding the ResourceManager instance:  public static GettextResourceManager MyResourceManager = new GettextResourceManager("domain-name");  All classes containing internationalized strings then contain  private static GettextResourceManager Res = Util.MyResourceManager; private static String _(String s) { return Res.GetString(s); }  and the shorthand is used like this:  Console.WriteLine(_("Operation completed."));  • You add a class with a very short name, say ‘S’, containing just the definition of the resource manager and of the shorthand:  public class S { public static GettextResourceManager MyResourceManager = new GettextResourceManager("domain-name"); public static String _(String s) { return MyResourceManager.GetString(s); } }  and the shorthand is used like this:  Console.WriteLine(S._("Operation completed."));  Which of the two idioms you choose, will depend on whether copying two lines of codes into every class is more acceptable in your project than a class with a single-letter name. Next: , Previous: C#, Up: List of Programming Languages #### 15.5.13 GNU awk RPMs gawk 3.1 or newer File extension awk String syntax "abc" gettext shorthand _"abc" gettext/ngettext functions dcgettext, missing dcngettext in gawk-3.1.0 textdomain TEXTDOMAIN variable bindtextdomain bindtextdomain function setlocale automatic, but missing setlocale (LC_MESSAGES, "") in gawk-3.1.0 Prerequisite Use or emulate GNU gettext use Extractor xgettext Formatting with positions printf "%2d %1d" (GNU awk only) Portability On platforms without gettext, no translation. On non-GNU awks, you must define dcgettext, dcngettext and bindtextdomain yourself. po-mode marking An example is available in the examples directory: hello-gawk. Next: , Previous: gawk, Up: List of Programming Languages #### 15.5.14 Pascal - Free Pascal Compiler RPMs fpk File extension pp, pas String syntax 'abc' gettext shorthand automatic gettext/ngettext functions —, use ResourceString data type instead textdomain —, use TranslateResourceStrings function instead bindtextdomain —, use TranslateResourceStrings function instead setlocale automatic, but uses only LANG, not LC_MESSAGES or LC_ALL Prerequisite {mode delphi} or {mode objfpc} uses gettext; Use or emulate GNU gettext emulate partially Extractor ppc386 followed by xgettext or rstconv Formatting with positions uses sysutils; format "%1:d %0:d" Portability ? po-mode marking The Pascal compiler has special support for the ResourceString data type. It generates a .rst file. This is then converted to a .pot file by use of xgettext or rstconv. At runtime, a .mo file corresponding to translations of this .pot file can be loaded using the TranslateResourceStrings function in the gettext unit. An example is available in the examples directory: hello-pascal. Next: , Previous: Pascal, Up: List of Programming Languages #### 15.5.15 wxWidgets library RPMs wxGTK, gettext File extension cpp String syntax "abc" gettext shorthand _("abc") gettext/ngettext functions wxLocale::GetString, wxGetTranslation textdomain wxLocale::AddCatalog bindtextdomain wxLocale::AddCatalogLookupPathPrefix setlocale wxLocale::Init, wxSetLocale Prerequisite #include <wx/intl.h> Use or emulate GNU gettext emulate, see include/wx/intl.h and src/common/intl.cpp Extractor xgettext Formatting with positions wxString::Format supports positions if and only if the system has wprintf(), vswprintf() functions and they support positions according to POSIX. Portability fully portable po-mode marking yes Next: , Previous: wxWidgets, Up: List of Programming Languages #### 15.5.16 YCP - YaST2 scripting language RPMs libycp, libycp-devel, yast2-core, yast2-core-devel File extension ycp String syntax "abc" gettext shorthand _("abc") gettext/ngettext functions _() with 1 or 3 arguments textdomain textdomain statement bindtextdomain setlocale Prerequisite Use or emulate GNU gettext use Extractor xgettext Formatting with positions sformat "%2 %1" Portability fully portable po-mode marking An example is available in the examples directory: hello-ycp. Next: , Previous: YCP, Up: List of Programming Languages #### 15.5.17 Tcl - Tk's scripting language RPMs tcl File extension tcl String syntax "abc" gettext shorthand [_ "abc"] gettext/ngettext functions ::msgcat::mc textdomain bindtextdomain —, use ::msgcat::mcload instead setlocale automatic, uses LANG, but ignores LC_MESSAGES and LC_ALL Prerequisite package require msgcat proc _ {s} {return [::msgcat::mc s]} Use or emulate GNU gettext —, uses a Tcl specific message catalog format Extractor xgettext -k_ Formatting with positions format "%2\d %1\d" Portability fully portable po-mode marking Two examples are available in the examples directory: hello-tcl, hello-tcl-tk. Before marking strings as internationalizable, substitutions of variables into the string need to be converted to format applications. For example, "file filename not found" becomes [format "file %s not found" filename]. Only after this is done, can the strings be marked and extracted. After marking, this example becomes [format [_ "file %s not found"] filename] or [msgcat::mc "file %s not found" filename]. Note that the msgcat::mc function implicitly calls format when more than one argument is given. Next: , Previous: Tcl, Up: List of Programming Languages #### 15.5.18 Perl RPMs perl File extension pl, PL, pm, cgi String syntax • "abc" • 'abc' • qq (abc) • q (abc) • qr /abc/ • qx (/bin/date) • /pattern match/ • ?pattern match? • s/substitution/operators/ • tied_hash{"message"} • tied_hash_reference->{"message"} • etc., issue the command ‘man perlsyn’ for details gettext shorthand __ (double underscore) gettext/ngettext functions gettext, dgettext, dcgettext, ngettext, dngettext, dcngettext textdomain textdomain function bindtextdomain bindtextdomain function bind_textdomain_codeset bind_textdomain_codeset function setlocale Use setlocale (LC_ALL, ""); Prerequisite use POSIX; use Locale::TextDomain; (included in the package libintl-perl which is available on the Comprehensive Perl Archive Network CPAN, http://www.cpan.org/). Use or emulate GNU gettext platform dependent: gettext_pp emulates, gettext_xs uses GNU gettext Extractor xgettext -k__ -k\__ -k%__ -k__x -k__n:1,2 -k__nx:1,2 -k__xn:1,2 -kN__ -k Formatting with positions Both kinds of format strings support formatting with positions. printf "%2\d %1\d", ... (requires Perl 5.8.0 or newer) __expand("[new] replaces [old]", old => oldvalue, new => newvalue) Portability The libintl-perl package is platform independent but is not part of the Perl core. The programmer is responsible for providing a dummy implementation of the required functions if the package is not installed on the target system. po-mode marking Documentation Included in libintl-perl, available on CPAN (http://www.cpan.org/). An example is available in the examples directory: hello-perl. The xgettext parser backend for Perl differs significantly from the parser backends for other programming languages, just as Perl itself differs significantly from other programming languages. The Perl parser backend offers many more string marking facilities than the other backends but it also has some Perl specific limitations, the worst probably being its imperfectness. Next: , Up: Perl ##### 15.5.18.1 General Problems Parsing Perl Code It is often heard that only Perl can parse Perl. This is not true. Perl cannot be parsed at all, it can only be executed. Perl has various built-in ambiguities that can only be resolved at runtime. The following example may illustrate one common problem:  print gettext "Hello World!";  Although this example looks like a bullet-proof case of a function invocation, it is not:  open gettext, ">testfile" or die; print gettext "Hello world!"  In this context, the string gettext looks more like a file handle. But not necessarily:  use Locale::Messages qw (:libintl_h); open gettext ">testfile" or die; print gettext "Hello world!";  Now, the file is probably syntactically incorrect, provided that the module Locale::Messages found first in the Perl include path exports a function gettext. But what if the module Locale::Messages really looks like this?  use vars qw (*gettext); 1;  In this case, the string gettext will be interpreted as a file handle again, and the above example will create a file testfile and write the string “Hello world!” into it. Even advanced control flow analysis will not really help:  if (0.5 < rand) { eval "use Sane"; } else { eval "use InSane"; } print gettext "Hello world!";  If the module Sane exports a function gettext that does what we expect, and the module InSane opens a file for writing and associates the handle gettext with this output stream, we are clueless again about what will happen at runtime. It is completely unpredictable. The truth is that Perl has so many ways to fill its symbol table at runtime that it is impossible to interpret a particular piece of code without executing it. Of course, xgettext will not execute your Perl sources while scanning for translatable strings, but rather use heuristics in order to guess what you meant. Another problem is the ambiguity of the slash and the question mark. Their interpretation depends on the context:  # A pattern match. print "OK\n" if /foobar/; # A division. print 1 / 2; # Another pattern match. print "OK\n" if ?foobar?; # Conditional. print x ? "foo" : "bar";  The slash may either act as the division operator or introduce a pattern match, whereas the question mark may act as the ternary conditional operator or as a pattern match, too. Other programming languages like awk present similar problems, but the consequences of a misinterpretation are particularly nasty with Perl sources. In awk for instance, a statement can never exceed one line and the parser can recover from a parsing error at the next newline and interpret the rest of the input stream correctly. Perl is different, as a pattern match is terminated by the next appearance of the delimiter (the slash or the question mark) in the input stream, regardless of the semantic context. If a slash is really a division sign but mis-interpreted as a pattern match, the rest of the input file is most probably parsed incorrectly. If you find that xgettext fails to extract strings from portions of your sources, you should therefore look out for slashes and/or question marks preceding these sections. You may have come across a bug in xgettext's Perl parser (and of course you should report that bug). In the meantime you should consider to reformulate your code in a manner less challenging to xgettext. Next: , Previous: General Problems, Up: Perl ##### 15.5.18.2 Which keywords will xgettext look for? Unless you instruct xgettext otherwise by invoking it with one of the options --keyword or -k, it will recognize the following keywords in your Perl sources: • gettext • dgettext • dcgettext • ngettext:1,2 The first (singular) and the second (plural) argument will be extracted. • dngettext:1,2 The first (singular) and the second (plural) argument will be extracted. • dcngettext:1,2 The first (singular) and the second (plural) argument will be extracted. • gettext_noop • %gettext The keys of lookups into the hash %gettext will be extracted. • gettext The keys of lookups into the hash reference gettext will be extracted. Next: , Previous: Default Keywords, Up: Perl ##### 15.5.18.3 How to Extract Hash Keys Translating messages at runtime is normally performed by looking up the original string in the translation database and returning the translated version. The “natural” Perl implementation is a hash lookup, and, of course, xgettext supports such practice.  print __"Hello world!"; print __{"Hello world!"}; print __->{"Hello world!"}; print$$__{"Hello world!"};


The above four lines all do the same thing. The Perl module Locale::TextDomain exports by default a hash %__ that is tied to the function __(). It also exports a reference $__ to %__. If an argument to the xgettext option --keyword, resp. -k starts with a percent sign, the rest of the keyword is interpreted as the name of a hash. If it starts with a dollar sign, the rest of the keyword is interpreted as a reference to a hash. Note that you can omit the quotation marks (single or double) around the hash key (almost) whenever Perl itself allows it:  print$gettext{Error};


The exact rule is: You can omit the surrounding quotes, when the hash key is a valid C (!) identifier, i.e. when it starts with an underscore or an ASCII letter and is followed by an arbitrary number of underscores, ASCII letters or digits. Other Unicode characters are not allowed, regardless of the use utf8 pragma.

Next: , Previous: Special Keywords, Up: Perl
##### 15.5.18.4 What are Strings And Quote-like Expressions?

Perl offers a plethora of different string constructs. Those that can be used either as arguments to functions or inside braces for hash lookups are generally supported by xgettext.

• double-quoted strings
          print gettext "Hello World!";

• single-quoted strings
          print gettext 'Hello World!';

• the operator qq
          print gettext qq |Hello World!|;
print gettext qq <E-mail: <guido\@imperia.net>>;


The operator qq is fully supported. You can use arbitrary delimiters, including the four bracketing delimiters (round, angle, square, curly) that nest.

• the operator q
          print gettext q |Hello World!|;
print gettext q <E-mail: <guido@imperia.net>>;


The operator q is fully supported. You can use arbitrary delimiters, including the four bracketing delimiters (round, angle, square, curly) that nest.

• the operator qx
          print gettext qx ;LANGUAGE=C /bin/date;
print gettext qx [/usr/bin/ls | grep '^[A-Z]*'];


The operator qx is fully supported. You can use arbitrary delimiters, including the four bracketing delimiters (round, angle, square, curly) that nest.

The example is actually a useless use of gettext. It will invoke the gettext function on the output of the command specified with the qx operator. The feature was included in order to make the interface consistent (the parser will extract all strings and quote-like expressions).

• here documents
          print gettext <<'EOF';
program not found in $PATH EOF print ngettext <<EOF, <<"EOF"; one file deleted EOF several files deleted EOF  Here-documents are recognized. If the delimiter is enclosed in single quotes, the string is not interpolated. If it is enclosed in double quotes or has no quotes at all, the string is interpolated. Delimiters that start with a digit are not supported! Next: , Previous: Quote-like Expressions, Up: Perl ##### 15.5.18.5 Invalid Uses Of String Interpolation Perl is capable of interpolating variables into strings. This offers some nice features in localized programs but can also lead to problems. A common error is a construct like the following:  print gettext "This is the program$0!\n";


Perl will interpolate at runtime the value of the variable $0 into the argument of the gettext() function. Hence, this argument is not a string constant but a variable argument ($0 is a global variable that holds the name of the Perl script being executed). The interpolation is performed by Perl before the string argument is passed to gettext() and will therefore depend on the name of the script which can only be determined at runtime. Consequently, it is almost impossible that a translation can be looked up at runtime (except if, by accident, the interpolated string is found in the message catalog).

The xgettext program will therefore terminate parsing with a fatal error if it encounters a variable inside of an extracted string. In general, this will happen for all kinds of string interpolations that cannot be safely performed at compile time. If you absolutely know what you are doing, you can always circumvent this behavior:

     my $know_what_i_am_doing = "This is program$0!\n";
print gettext $know_what_i_am_doing;  Since the parser only recognizes strings and quote-like expressions, but not variables or other terms, the above construct will be accepted. You will have to find another way, however, to let your original string make it into your message catalog. If invoked with the option --extract-all, resp. -a, variable interpolation will be accepted. Rationale: You will generally use this option in order to prepare your sources for internationalization. Please see the manual page ‘man perlop’ for details of strings and quote-like expressions that are subject to interpolation and those that are not. Safe interpolations (that will not lead to a fatal error) are: • the escape sequences \t (tab, HT, TAB), \n (newline, NL), \r (return, CR), \f (form feed, FF), \b (backspace, BS), \a (alarm, bell, BEL), and \e (escape, ESC). • octal chars, like \033 Note that octal escapes in the range of 400-777 are translated into a UTF-8 representation, regardless of the presence of the use utf8 pragma. • hex chars, like \x1b • wide hex chars, like \x{263a} Note that this escape is translated into a UTF-8 representation, regardless of the presence of the use utf8 pragma. • control chars, like \c[ (CTRL-[) • named Unicode chars, like \N{LATIN CAPITAL LETTER C WITH CEDILLA} Note that this escape is translated into a UTF-8 representation, regardless of the presence of the use utf8 pragma. The following escapes are considered partially safe: • \l lowercase next char • \u uppercase next char • \L lowercase till \E • \U uppercase till \E • \E end case modification • \Q quote non-word characters till \E These escapes are only considered safe if the string consists of ASCII characters only. Translation of characters outside the range defined by ASCII is locale-dependent and can actually only be performed at runtime; xgettext doesn't do these locale-dependent translations at extraction time. Except for the modifier \Q, these translations, albeit valid, are generally useless and only obfuscate your sources. If a translation can be safely performed at compile time you can just as well write what you mean. Next: , Previous: Interpolation I, Up: Perl ##### 15.5.18.6 Valid Uses Of String Interpolation Perl is often used to generate sources for other programming languages or arbitrary file formats. Web applications that output HTML code make a prominent example for such usage. You will often come across situations where you want to intersperse code written in the target (programming) language with translatable messages, like in the following HTML example:  print gettext <<EOF; <h1>My Homepage</h1> <script language="JavaScript"><!-- for (i = 0; i < 100; ++i) { alert ("Thank you so much for visiting my homepage!"); } //--></script> EOF  The parser will extract the entire here document, and it will appear entirely in the resulting PO file, including the JavaScript snippet embedded in the HTML code. If you exaggerate with constructs like the above, you will run the risk that the translators of your package will look out for a less challenging project. You should consider an alternative expression here:  print <<EOF; <h1>$gettext{"My Homepage"}</h1>
<script language="JavaScript"><!--
for (i = 0; i < 100; ++i) {
alert ("$gettext{'Thank you so much for visiting my homepage!'}"); } //--></script> EOF  Only the translatable portions of the code will be extracted here, and the resulting PO file will begrudgingly improve in terms of readability. You can interpolate hash lookups in all strings or quote-like expressions that are subject to interpolation (see the manual page ‘man perlop’ for details). Double interpolation is invalid, however:  # TRANSLATORS: Replace "the earth" with the name of your planet. print gettext qq{Welcome to$gettext->{"the earth"}};


The qq-quoted string is recognized as an argument to xgettext in the first place, and checked for invalid variable interpolation. The dollar sign of hash-dereferencing will therefore terminate the parser with an “invalid interpolation” error.

It is valid to interpolate hash lookups in regular expressions:

     if ($var =~ /$gettext{"the earth"}/) {
print gettext "Match!\n";
}
s/$gettext{"U. S. A."}/$gettext{"U. S. A."} $gettext{"(dial +0)"}/g;  Next: , Previous: Interpolation II, Up: Perl ##### 15.5.18.7 When To Use Parentheses In Perl, parentheses around function arguments are mostly optional. xgettext will always assume that all recognized keywords (except for hashes and hash references) are names of properly prototyped functions, and will (hopefully) only require parentheses where Perl itself requires them. All constructs in the following example are therefore ok to use:  print gettext ("Hello World!\n"); print gettext "Hello World!\n"; print dgettext ($package => "Hello World!\n");
print dgettext $package, "Hello World!\n"; # The "fat comma" => turns the left-hand side argument into a # single-quoted string! print dgettext smellovision => "Hello World!\n"; # The following assignment only works with prototyped functions. # Otherwise, the functions will act as "greedy" list operators and # eat up all following arguments. my$anonymous_hash = {
planet => gettext "earth",
cakes => ngettext "one cake", "several cakes", $n, still =>$works,
};
# The same without fat comma:
my $other_hash = { 'planet', gettext "earth", 'cakes', ngettext "one cake", "several cakes",$n,
'still', $works, }; # Parentheses are only significant for the first argument. print dngettext 'package', ("one cake", "several cakes",$n), $discarded;  Next: , Previous: Parentheses, Up: Perl ##### 15.5.18.8 How To Grok with Long Lines The necessity of long messages can often lead to a cumbersome or unreadable coding style. Perl has several options that may prevent you from writing unreadable code, and xgettext does its best to do likewise. This is where the dot operator (the string concatenation operator) may come in handy:  print gettext ("This is a very long" . " message that is still" . " readable, because" . " it is split into" . " multiple lines.\n");  Perl is smart enough to concatenate these constant string fragments into one long string at compile time, and so is xgettext. You will only find one long message in the resulting POT file. Note that the future Perl 6 will probably use the underscore (‘_’) as the string concatenation operator, and the dot (‘.’) for dereferencing. This new syntax is not yet supported by xgettext. If embedded newline characters are not an issue, or even desired, you may also insert newline characters inside quoted strings wherever you feel like it:  print gettext ("<em>In HTML output embedded newlines are generally no problem, since adjacent whitespace is always rendered into a single space character.</em>");  You may also consider to use here documents:  print gettext <<EOF; <em>In HTML output embedded newlines are generally no problem, since adjacent whitespace is always rendered into a single space character.</em> EOF  Please do not forget that the line breaks are real, i.e. they translate into newline characters that will consequently show up in the resulting POT file. Previous: Long Lines, Up: Perl ##### 15.5.18.9 Bugs, Pitfalls, And Things That Do Not Work The foregoing sections should have proven that xgettext is quite smart in extracting translatable strings from Perl sources. Yet, some more or less exotic constructs that could be expected to work, actually do not work. One of the more relevant limitations can be found in the implementation of variable interpolation inside quoted strings. Only simple hash lookups can be used there:  print <<EOF;$gettext{"The dot operator"
. " does not work"
. "here!"}
Likewise, you cannot @{[ gettext ("interpolate function calls") ]}
inside quoted strings or quote-like expressions.
EOF


This is valid Perl code and will actually trigger invocations of the gettext function at runtime. Yet, the Perl parser in xgettext will fail to recognize the strings. A less obvious example can be found in the interpolation of regular expressions:

     s/<!--START_OF_WEEK-->/gettext ("Sunday")/e;


The modifier e will cause the substitution to be interpreted as an evaluable statement. Consequently, at runtime the function gettext() is called, but again, the parser fails to extract the string “Sunday”. Use a temporary variable as a simple workaround if you really happen to need this feature:

     my $sunday = gettext "Sunday"; s/<!--START_OF_WEEK-->/$sunday/;


Hash slices would also be handy but are not recognized:

     my @weekdays = @gettext{'Sunday', 'Monday', 'Tuesday', 'Wednesday',
'Thursday', 'Friday', 'Saturday'};
# Or even:
@weekdays = @gettext{qw (Sunday Monday Tuesday Wednesday Thursday
Friday Saturday) };


This is perfectly valid usage of the tied hash %gettext but the strings are not recognized and therefore will not be extracted.

Another caveat of the current version is its rudimentary support for non-ASCII characters in identifiers. You may encounter serious problems if you use identifiers with characters outside the range of 'A'-'Z', 'a'-'z', '0'-'9' and the underscore '_'.

Maybe some of these missing features will be implemented in future versions, but since you can always make do without them at minimal effort, these todos have very low priority.

A nasty problem are brace format strings that already contain braces as part of the normal text, for example the usage strings typically encountered in programs:

     die "usage: $0 {OPTIONS} FILENAME...\n";  If you want to internationalize this code with Perl brace format strings, you will run into a problem:  die __x ("usage: {program} {OPTIONS} FILENAME...\n", program =>$0);


Whereas ‘{program}’ is a placeholder, ‘{OPTIONS}’ is not and should probably be translated. Yet, there is no way to teach the Perl parser in xgettext to recognize the first one, and leave the other one alone.

There are two possible work-arounds for this problem. If you are sure that your program will run under Perl 5.8.0 or newer (these Perl versions handle positional parameters in printf()) or if you are sure that the translator will not have to reorder the arguments in her translation – for example if you have only one brace placeholder in your string, or if it describes a syntax, like in this one –, you can mark the string as no-perl-brace-format and use printf():

     # xgettext: no-perl-brace-format
die sprintf ("usage: %s {OPTIONS} FILENAME...\n", $0);  If you want to use the more portable Perl brace format, you will have to do put placeholders in place of the literal braces:  die __x ("usage: {program} {[}OPTIONS{]} FILENAME...\n", program =>$0, '[' => '{', ']' => '}');


Perl brace format strings know no escaping mechanism. No matter how this escaping mechanism looked like, it would either give the programmer a hard time, make translating Perl brace format strings heavy-going, or result in a performance penalty at runtime, when the format directives get executed. Most of the time you will happily get along with printf() for this special case.

Next: , Previous: Perl, Up: List of Programming Languages

#### 15.5.19 PHP Hypertext Preprocessor

RPMs
mod_php4, mod_php4-core, phpdoc
File extension
php, php3, php4
String syntax
"abc", 'abc'
gettext shorthand
_("abc")
gettext/ngettext functions
gettext, dgettext, dcgettext; starting with PHP 4.2.0 also ngettext, dngettext, dcngettext
textdomain
textdomain function
bindtextdomain
bindtextdomain function
setlocale
Programmer must call setlocale (LC_ALL, "")
Prerequisite

Use or emulate GNU gettext
use
Extractor
xgettext
Formatting with positions
printf "%2\$d %1\$d"
Portability
On platforms without gettext, the functions are not available.
po-mode marking

An example is available in the examples directory: hello-php.

Next: , Previous: PHP, Up: List of Programming Languages

#### 15.5.20 Pike

RPMs
roxen
File extension
pike
String syntax
"abc"
gettext shorthand

gettext/ngettext functions
gettext, dgettext, dcgettext
textdomain
textdomain function
bindtextdomain
bindtextdomain function
setlocale
setlocale function
Prerequisite
import Locale.Gettext;
Use or emulate GNU gettext
use
Extractor

Formatting with positions

Portability
On platforms without gettext, the functions are not available.
po-mode marking

Previous: Pike, Up: List of Programming Languages

#### 15.5.21 GNU Compiler Collection sources

RPMs
gcc
File extension
c, h.
String syntax
"abc"
gettext shorthand
_("abc")
gettext/ngettext functions
gettext, dgettext, dcgettext, ngettext, dngettext, dcngettext
textdomain
textdomain function
bindtextdomain
bindtextdomain function
setlocale
Programmer must call setlocale (LC_ALL, "")
Prerequisite
#include "intl.h"
Use or emulate GNU gettext
Use
Extractor
xgettext -k_
Formatting with positions

Portability
Uses autoconf macros
po-mode marking
yes

Previous: List of Programming Languages, Up: Programming Languages

### 15.6 Internationalizable Data

Here is a list of other data formats which can be internationalized using GNU gettext.

Next: , Previous: List of Data Formats, Up: List of Data Formats

#### 15.6.1 POT - Portable Object Template

RPMs
gettext
File extension
pot, po
Extractor
xgettext

Next: , Previous: POT, Up: List of Data Formats

#### 15.6.2 Resource String Table

RPMs
fpk
File extension
rst
Extractor
xgettext, rstconv

Previous: RST, Up: List of Data Formats

#### 15.6.3 Glade - GNOME user interface description

RPMs
File extension
glade, glade2
Extractor
xgettext, libglade-xgettext, xml-i18n-extract, intltool-extract

Next: , Previous: Programming Languages, Up: Top

## 16 Concluding Remarks

We would like to conclude this GNU gettext manual by presenting an history of the Translation Project so far. We finally give a few pointers for those who want to do further research or readings about Native Language Support matters.

Next: , Previous: Conclusion, Up: Conclusion

### 16.1 History of GNU gettext

Internationalization concerns and algorithms have been informally and casually discussed for years in GNU, sometimes around GNU libc, maybe around the incoming Hurd, or otherwise (nobody clearly remembers). And even then, when the work started for real, this was somewhat independently of these previous discussions.

This all began in July 1994, when Patrick D'Cruze had the idea and initiative of internationalizing version 3.9.2 of GNU fileutils. He then asked Jim Meyering, the maintainer, how to get those changes folded into an official release. That first draft was full of #ifdefs and somewhat disconcerting, and Jim wanted to find nicer ways. Patrick and Jim shared some tries and experimentations in this area. Then, feeling that this might eventually have a deeper impact on GNU, Jim wanted to know what standards were, and contacted Richard Stallman, who very quickly and verbally described an overall design for what was meant to become glocale, at that time.

Jim implemented glocale and got a lot of exhausting feedback from Patrick and Richard, of course, but also from Mitchum DSouza (who wrote a catgets-like package), Roland McGrath, maybe David MacKenzie, François Pinard, and Paul Eggert, all pushing and pulling in various directions, not always compatible, to the extent that after a couple of test releases, glocale was torn apart. In particular, Paul Eggert – always keeping an eye on developments in Solaris – advocated the use of the gettext API over glocale's catgets-based API.

While Jim took some distance and time and became dad for a second time, Roland wanted to get GNU libc internationalized, and got Ulrich Drepper involved in that project. Instead of starting from glocale, Ulrich rewrote something from scratch, but more conforming to the set of guidelines who emerged out of the glocale effort. Then, Ulrich got people from the previous forum to involve themselves into this new project, and the switch from glocale to what was first named msgutils, renamed nlsutils, and later gettext, became officially accepted by Richard in May 1995 or so.

Let's summarize by saying that Ulrich Drepper wrote GNU gettext in April 1995. The first official release of the package, including PO mode, occurred in July 1995, and was numbered 0.7. Other people contributed to the effort by providing a discussion forum around Ulrich, writing little pieces of code, or testing. These are quoted in the THANKS file which comes with the GNU gettext distribution.

While this was being done, François adapted half a dozen of GNU packages to glocale first, then later to gettext, putting them in pretest, so providing along the way an effective user environment for fine tuning the evolving tools. He also took the responsibility of organizing and coordinating the Translation Project. After nearly a year of informal exchanges between people from many countries, translator teams started to exist in May 1995, through the creation and support by Patrick D'Cruze of twenty unmoderated mailing lists for that many native languages, and two moderated lists: one for reaching all teams at once, the other for reaching all willing maintainers of internationalized free software packages.

François also wrote PO mode in June 1995 with the collaboration of Greg McGary, as a kind of contribution to Ulrich's package. He also gave a hand with the GNU gettext Texinfo manual.

In 1997, Ulrich Drepper released the GNU libc 2.0, which included the gettext, textdomain and bindtextdomain functions.

In 2000, Ulrich Drepper added plural form handling (the ngettext function) to GNU libc. Later, in 2001, he released GNU libc 2.2.x, which is the first free C library with full internationalization support.

Ulrich being quite busy in his role of General Maintainer of GNU libc, he handed over the GNU gettext maintenance to Bruno Haible in 2000. Bruno added the plural form handling to the tools as well, added support for UTF-8 and CJK locales, and wrote a few new tools for manipulating PO files.

Previous: History, Up: Conclusion

NOTE: This documentation section is outdated and needs to be revised.

Eugene H. Dorr (dorre@well.com) maintains an interesting bibliography on internationalization matters, called Internationalization Reference List, which is available as:

     ftp://ftp.ora.com/pub/examples/nutshell/ujip/doc/i18n-books.txt


Michael Gschwind (mike@vlsivie.tuwien.ac.at) maintains a Frequently Asked Questions (FAQ) list, entitled Programming for Internationalisation. This FAQ discusses writing programs which can handle different language conventions, character sets, etc.; and is applicable to all character set encodings, with particular emphasis on ISO 8859-1. It is regularly published in Usenet groups comp.unix.questions, comp.std.internat, comp.software.international, comp.lang.c, comp.windows.x, comp.std.c, comp.answers and news.answers. The home location of this document is:

     ftp://ftp.vlsivie.tuwien.ac.at/pub/8bit/ISO-programming


Patrick D'Cruze (pdcruze@li.org) wrote a tutorial about NLS matters, and Jochen Hein (Hein@student.tu-clausthal.de) took over the responsibility of maintaining it. It may be found as:

     ftp://sunsite.unc.edu/pub/Linux/utils/nls/catalogs/Incoming/...
...locale-tutorial-0.8.txt.gz


This site is mirrored in:

     ftp://ftp.ibp.fr/pub/linux/sunsite/


A French version of the same tutorial should be findable at:

     ftp://ftp.ibp.fr/pub/linux/french/docs/
`

together with French translations of many Linux-related documents.

Next: , Previous: Conclusion, Up: Top

## Appendix A Language Codes

The ISO 639 standard defines two-letter codes for many languages, and three-letter codes for more rarely used languages. All abbreviations for languages used in the Translation Project should come from this standard.

Next: , Previous: Language Codes, Up: Language Codes

### A.1 Usual Language Codes

For the commonly used languages, the ISO 639-1 standard defines two-letter codes.

aa
Afar.
ab
Abkhazian.
ae
Avestan.
af
Afrikaans.
ak
Akan.
am
Amharic.
an
Aragonese.
ar
Arabic.
as
Assamese.
av
Avaric.
ay
Aymara.
az
Azerbaijani.
ba
Bashkir.
be
Byelorussian; Belarusian.
bg
Bulgarian.
bh
Bihari.
bi
Bislama.
bm
Bambara.
bn
Bengali; Bangla.
bo
Tibetan.
br
Breton.
bs
Bosnian.
ca
Catalan.
ce
Chechen.
ch
Chamorro.
co
Corsican.
cr
Cree.
cs
Czech.
cu
Church Slavic.
cv
Chuvash.
cy
Welsh.
da
Danish.
de
German.
dv
Divehi; Maldivian.
dz
Dzongkha; Bhutani.
ee
Éwé.
el
Greek.
en
English.
eo
Esperanto.
es
Spanish.
et
Estonian.
eu
Basque.
fa
Persian.
ff
Fulah.
fi
Finnish.
fj
Fijian; Fiji.
fo
Faroese.
fr
French.
fy
Western Frisian.
ga
Irish.
gd
Scots; Gaelic.
gl
Galician.
gn
Guarani.
gu
Gujarati.
gv
Manx.
ha
Hausa.
he
Hebrew (formerly iw).
hi
Hindi.
ho
Hiri Motu.
hr
Croatian.
ht
Haitian; Haitian Creole.
hu
Hungarian.
hy
Armenian.
hz
Herero.
ia
Interlingua.
id
Indonesian (formerly in).
ie
Interlingue.
ig
Igbo.
ii
Sichuan Yi.
ik
Inupiak; Inupiaq.
io
Ido.
is
Icelandic.
it
Italian.
iu
Inuktitut.
ja
Japanese.
jv
Javanese.
ka
Georgian.
kg
Kongo.
ki
Kikuyu; Gikuyu.
kj
Kuanyama; Kwanyama.
kk
Kazakh.
kl
Kalaallisut; Greenlandic.
km
Khmer; Cambodian.
kn
ko
Korean.
kr
Kanuri.
ks
Kashmiri.
ku
Kurdish.
kv
Komi.
kw
Cornish.
ky
Kirghiz.
la
Latin.
lb
Letzeburgesch; Luxembourgish.
lg
Ganda.
li
Limburgish; Limburger; Limburgan.
ln
Lingala.
lo
Lao; Laotian.
lt
Lithuanian.
lu
Luba-Katanga.
lv
Latvian; Lettish.
mg
Malagasy.
mh
Marshallese.
mi
Maori.
mk
Macedonian.
ml
Malayalam.
mn
Mongolian.
mo
Moldavian.
mr
Marathi.
ms
Malay.
mt
Maltese.
my
Burmese.
na
Nauru.
nb
Norwegian Bokmål.
nd
Ndebele, North.
ne
Nepali.
ng
Ndonga.
nl
Dutch.
nn
Norwegian Nynorsk.
no
Norwegian.
nr
Ndebele, South.
nv
Navajo; Navaho.
ny
Chichewa; Nyanja.
oc
Occitan; Provençal.
oj
Ojibwa.
om
(Afan) Oromo.
or
Oriya.
os
Ossetian; Ossetic.
pa
Panjabi; Punjabi.
pi
Pali.
pl
Polish.
ps
Pashto, Pushto.
pt
Portuguese.
qu
Quechua.
rm
Rhaeto-Romance.
rn
Rundi; Kirundi.
ro
Romanian.
ru
Russian.
rw
Kinyarwanda.
sa
Sanskrit.
sc
Sardinian.
sd
Sindhi.
se
Northern Sami.
sg
Sango; Sangro.
si
Sinhala; Sinhalese.
sk
Slovak.
sl
Slovenian.
sm
Samoan.
sn
Shona.
so
Somali.
sq
Albanian.
sr
Serbian.
ss
Swati; Siswati.
st
Sesotho; Sotho, Southern.
su
Sundanese.
sv
Swedish.
sw
Swahili.
ta
Tamil.
te
Telugu.
tg
Tajik.
th
Thai.
ti
Tigrinya.
tk
Turkmen.
tl
Tagalog.
tn
Tswana; Setswana.
to
Tonga.
tr
Turkish.
ts
Tsonga.
tt
Tatar.
tw
Twi.
ty
Tahitian.
ug
Uighur.
uk
Ukrainian.
ur
Urdu.
uz
Uzbek.
ve
Venda.
vi
Vietnamese.
vo
Volapük; Volapuk.
wa
Walloon.
wo
Wolof.
xh
Xhosa.
yi
Yiddish (formerly ji).
yo
Yoruba.
za
Zhuang.
zh
Chinese.
zu
Zulu.

Previous: Usual Language Codes, Up: Language Codes

### A.2 Rare Language Codes

For rarely used languages, the ISO 639-2 standard defines three-letter codes. Here is the current list, reduced to only living languages with at least one million of speakers.

ace
Achinese.
awa
Banda.
bal
Baluchi.
ban
Balinese.
bem
Bemba.
bho
Bhojpuri.
bik
Bikol.
bin
Bini.
btk
Batak (Indonesia).
bug
Buginese.
ceb
Cebuano.
din
Dinka.
doi
Dogri.
fil
Filipino; Pilipino.
fon
Fon.
gon
Gondi.
gsw
Alemani; Swiss German.
hil
Hiligaynon.
hmn
Hmong.
ilo
Iloko.
kab
Kabyle.
kam
Kamba.
kbd
Kabardian.
kmb
Kimbundu.
kok
Konkani.
kru
Kurukh.
lua
Luba-Lulua.
luo
Luo (Kenya and Tanzania).
mag
Magahi.
mai
Maithili.
mak
Makasar.
man
Mandingo.
men
Mende.
min
Minangkabau.
mni
Manipuri.
mos
Mossi.
mwr
Marwari.
nap
Neapolitan.
nso
Pedi; Sepedi; Northern Sotho.
nym
Nyamwezi.
nyn
Nyankole.
pag
Pangasinan.
pam
Pampanga.
raj
Rajasthani.
sas
Sasak.
sat
Santali.
scn
Sicilian.
shn
Shan.
sid
Sidamo.
srr
Serer.
suk
Sukuma.
sus
Susu.
tem
Timne.
tiv
Tiv.
tum
Tumbuka.
umb
Umbundu.
wal
Walamo.