glibc/manual/message.texi

@node Message Translation, Searching and Sorting, Locales, Top
@c %MENU% How to make the program speak the user's language
@chapter Message Translation

The program's interface with the user should be designed to ease the user's
task.  One way to ease the user's task is to use messages in whatever
language the user prefers.

Printing messages in different languages can be implemented in different
ways.  One could add all the different languages in the source code and
choose among the variants every time a message has to be printed.  This is
certainly not a good solution since extending the set of languages is
cumbersome (the code must be changed) and the code itself can become
really big with dozens of message sets.

A better solution is to keep the message sets for each language
in separate files which are loaded at runtime depending on the language
selection of the user.

@Theglibc{} provides two different sets of functions to support
message translation.  The problem is that neither of the interfaces is
officially defined by the POSIX standard.  The @code{catgets} family of
functions is defined in the X/Open standard but this is derived from
industry decisions and therefore not necessarily based on reasonable
decisions.

As mentioned above, the message catalog handling provides easy
extendability by using external data files which contain the message
translations.  I.e., these files contain for each of the messages used
in the program a translation for the appropriate language.  So the tasks
of the message handling functions are

@itemize @bullet
@item
locate the external data file with the appropriate translations
@item
load the data and make it possible to address the messages
@item
map a given key to the translated message
@end itemize

The two approaches mainly differ in the implementation of this last
step.  Decisions made in the last step influence the rest of the design.

@menu
* Message catalogs a la X/Open::  The @code{catgets} family of functions.
* The Uniforum approach::         The @code{gettext} family of functions.
@end menu


@node Message catalogs a la X/Open
@section X/Open Message Catalog Handling

The @code{catgets} functions are based on the simple scheme:

@quotation
Associate every message to translate in the source code with a unique
identifier.  To retrieve a message from a catalog file solely the
identifier is used.
@end quotation

This means for the author of the program that s/he will have to make
sure the meaning of the identifier in the program code and in the
message catalogs is always the same.

Before a message can be translated the catalog file must be located.
The user of the program must be able to guide the responsible function
to find whatever catalog the user wants.  This is separated from what
the programmer had in mind.

All the types, constants and functions for the @code{catgets} functions
are defined/declared in the @file{nl_types.h} header file.

@menu
* The catgets Functions::      The @code{catgets} function family.
* The message catalog files::  Format of the message catalog files.
* The gencat program::         How to generate message catalogs files which
                                can be used by the functions.
* Common Usage::               How to use the @code{catgets} interface.
@end menu


@node The catgets Functions
@subsection The @code{catgets} function family

@comment nl_types.h
@comment X/Open
@deftypefun nl_catd catopen (const char *@var{cat_name}, int @var{flag})
@safety{@prelim{}@mtsafe{@mtsenv{}}@asunsafe{@ascuheap{}}@acunsafe{@acsmem{}}}
@c catopen @mtsenv @ascuheap @acsmem
@c  strchr ok
@c  setlocale(,NULL) ok
@c  getenv @mtsenv
@c  strlen ok
@c  alloca ok
@c  stpcpy ok
@c  malloc @ascuheap @acsmem
@c  __open_catalog @ascuheap @acsmem
@c   strchr ok
@c   open_not_cancel_2 @acsfd
@c   strlen ok
@c   ENOUGH ok
@c    alloca ok
@c    memcpy ok
@c   fxstat64 ok
@c   __set_errno ok
@c   mmap @acsmem
@c   malloc dup @ascuheap @acsmem
@c   read_not_cancel ok
@c   free dup @ascuheap @acsmem
@c   munmap ok
@c   close_not_cancel_no_status ok
@c  free @ascuheap @acsmem
The @code{catopen} function tries to locate the message data file named
@var{cat_name} and loads it when found.  The return value is of an
opaque type and can be used in calls to the other functions to refer to
this loaded catalog.

The return value is @code{(nl_catd) -1} in case the function failed and
no catalog was loaded.  The global variable @var{errno} contains a code
for the error causing the failure.  But even if the function call
succeeded this does not mean that all messages can be translated.

Locating the catalog file must happen in a way which lets the user of
the program influence the decision.  It is up to the user to decide
about the language to use and sometimes it is useful to use alternate
catalog files.  All this can be specified by the user by setting some
environment variables.

The first problem is to find out where all the message catalogs are
stored.  Every program could have its own place to keep all the
different files but usually the catalog files are grouped by languages
and the catalogs for all programs are kept in the same place.

@cindex NLSPATH environment variable
To tell the @code{catopen} function where the catalog for the program
can be found the user can set the environment variable @code{NLSPATH} to
a value which describes her/his choice.  Since this value must be usable
for different languages and locales it cannot be a simple string.
Instead it is a format string (similar to @code{printf}'s).  An example
is

@smallexample
/usr/share/locale/%L/%N:/usr/share/locale/%L/LC_MESSAGES/%N
@end smallexample

First one can see that more than one directory can be specified (with
the usual syntax of separating them by colons).  The next things to
observe are the format string, @code{%L} and @code{%N} in this case.
The @code{catopen} function knows about several of them and the
replacement for all of them is of course different.

@table @code
@item %N
This format element is substituted with the name of the catalog file.
This is the value of the @var{cat_name} argument given to
@code{catgets}.

@item %L
This format element is substituted with the name of the currently
selected locale for translating messages.  How this is determined is
explained below.

@item %l
(This is the lowercase ell.) This format element is substituted with the
language element of the locale name.  The string describing the selected
locale is expected to have the form
@code{@var{lang}[_@var{terr}[.@var{codeset}]]} and this format uses the
first part @var{lang}.

@item %t
This format element is substituted by the territory part @var{terr} of
the name of the currently selected locale.  See the explanation of the
format above.

@item %c
This format element is substituted by the codeset part @var{codeset} of
the name of the currently selected locale.  See the explanation of the
format above.

@item %%
Since @code{%} is used as a meta character there must be a way to
express the @code{%} character in the result itself.  Using @code{%%}
does this just like it works for @code{printf}.
@end table


Using @code{NLSPATH} allows arbitrary directories to be searched for
message catalogs while still allowing different languages to be used.
If the @code{NLSPATH} environment variable is not set, the default value
is

@smallexample
@var{prefix}/share/locale/%L/%N:@var{prefix}/share/locale/%L/LC_MESSAGES/%N
@end smallexample

@noindent
where @var{prefix} is given to @code{configure} while installing @theglibc{}
(this value is in many cases @code{/usr} or the empty string).

The remaining problem is to decide which must be used.  The value
decides about the substitution of the format elements mentioned above.
First of all the user can specify a path in the message catalog name
(i.e., the name contains a slash character).  In this situation the
@code{NLSPATH} environment variable is not used.  The catalog must exist
as specified in the program, perhaps relative to the current working
directory.  This situation in not desirable and catalogs names never
should be written this way.  Beside this, this behavior is not portable
to all other platforms providing the @code{catgets} interface.

@cindex LC_ALL environment variable
@cindex LC_MESSAGES environment variable
@cindex LANG environment variable
Otherwise the values of environment variables from the standard
environment are examined (@pxref{Standard Environment}).  Which
variables are examined is decided by the @var{flag} parameter of
@code{catopen}.  If the value is @code{NL_CAT_LOCALE} (which is defined
in @file{nl_types.h}) then the @code{catopen} function uses the name of
the locale currently selected for the @code{LC_MESSAGES} category.

If @var{flag} is zero the @code{LANG} environment variable is examined.
This is a left-over from the early days when the concept of locales
had not even reached the level of POSIX locales.

The environment variable and the locale name should have a value of the
form @code{@var{lang}[_@var{terr}[.@var{codeset}]]} as explained above.
If no environment variable is set the @code{"C"} locale is used which
prevents any translation.

The return value of the function is in any case a valid string.  Either
it is a translation from a message catalog or it is the same as the
@var{string} parameter.  So a piece of code to decide whether a
translation actually happened must look like this:

@smallexample
@{
  char *trans = catgets (desc, set, msg, input_string);
  if (trans == input_string)
    @{
      /* Something went wrong.  */
    @}
@}
@end smallexample

@noindent
When an error occurs the global variable @var{errno} is set to

@table @var
@item EBADF
The catalog does not exist.
@item ENOMSG
The set/message tuple does not name an existing element in the
message catalog.
@end table

While it sometimes can be useful to test for errors programs normally
will avoid any test.  If the translation is not available it is no big
problem if the original, untranslated message is printed.  Either the
user understands this as well or s/he will look for the reason why the
messages are not translated.
@end deftypefun

Please note that the currently selected locale does not depend on a call
to the @code{setlocale} function.  It is not necessary that the locale
data files for this locale exist and calling @code{setlocale} succeeds.
The @code{catopen} function directly reads the values of the environment
variables.


@deftypefun {char *} catgets (nl_catd @var{catalog_desc}, int @var{set}, int @var{message}, const char *@var{string})
@safety{@prelim{}@mtsafe{}@assafe{}@acsafe{}}
The function @code{catgets} has to be used to access the message catalog
previously opened using the @code{catopen} function.  The
@var{catalog_desc} parameter must be a value previously returned by
@code{catopen}.

The next two parameters, @var{set} and @var{message}, reflect the
internal organization of the message catalog files.  This will be
explained in detail below.  For now it is interesting to know that a
catalog can consist of several sets and the messages in each thread are
individually numbered using numbers.  Neither the set number nor the
message number must be consecutive.  They can be arbitrarily chosen.
But each message (unless equal to another one) must have its own unique
pair of set and message numbers.

Since it is not guaranteed that the message catalog for the language
selected by the user exists the last parameter @var{string} helps to
handle this case gracefully.  If no matching string can be found
@var{string} is returned.  This means for the programmer that

@itemize @bullet
@item
the @var{string} parameters should contain reasonable text (this also
helps to understand the program seems otherwise there would be no hint
on the string which is expected to be returned.
@item
all @var{string} arguments should be written in the same language.
@end itemize
@end deftypefun

It is somewhat uncomfortable to write a program using the @code{catgets}
functions if no supporting functionality is available.  Since each
set/message number tuple must be unique the programmer must keep lists
of the messages at the same time the code is written.  And the work
between several people working on the same project must be coordinated.
We will see how some of these problems can be relaxed a bit (@pxref{Common
Usage}).

@deftypefun int catclose (nl_catd @var{catalog_desc})
@safety{@prelim{}@mtsafe{}@asunsafe{@ascuheap{}}@acunsafe{@acucorrupt{} @acsmem{}}}
@c catclose @ascuheap @acucorrupt @acsmem
@c  __set_errno ok
@c  munmap ok
@c  free @ascuheap @acsmem
The @code{catclose} function can be used to free the resources
associated with a message catalog which previously was opened by a call
to @code{catopen}.  If the resources can be successfully freed the
function returns @code{0}.  Otherwise it returns @code{@minus{}1} and the
global variable @var{errno} is set.  Errors can occur if the catalog
descriptor @var{catalog_desc} is not valid in which case @var{errno} is
set to @code{EBADF}.
@end deftypefun


@node The message catalog files
@subsection  Format of the message catalog files

The only reasonable way to translate all the messages of a function and
store the result in a message catalog file which can be read by the
@code{catopen} function is to write all the message text to the
translator and let her/him translate them all.  I.e., we must have a
file with entries which associate the set/message tuple with a specific
translation.  This file format is specified in the X/Open standard and
is as follows:

@itemize @bullet
@item
Lines containing only whitespace characters or empty lines are ignored.

@item
Lines which contain as the first non-whitespace character a @code{$}
followed by a whitespace character are comment and are also ignored.

@item
If a line contains as the first non-whitespace characters the sequence
@code{$set} followed by a whitespace character an additional argument
is required to follow.  This argument can either be:

@itemize @minus
@item
a number.  In this case the value of this number determines the set
to which the following messages are added.

@item
an identifier consisting of alphanumeric characters plus the underscore
character.  In this case the set get automatically a number assigned.
This value is one added to the largest set number which so far appeared.

How to use the symbolic names is explained in section @ref{Common Usage}.

It is an error if a symbol name appears more than once.  All following
messages are placed in a set with this number.
@end itemize

@item
If a line contains as the first non-whitespace characters the sequence
@code{$delset} followed by a whitespace character an additional argument
is required to follow.  This argument can either be:

@itemize @minus
@item
a number.  In this case the value of this number determines the set
which will be deleted.

@item
an identifier consisting of alphanumeric characters plus the underscore
character.  This symbolic identifier must match a name for a set which
previously was defined.  It is an error if the name is unknown.
@end itemize

In both cases all messages in the specified set will be removed.  They
will not appear in the output.  But if this set is later again selected
with a @code{$set} command again messages could be added and these
messages will appear in the output.

@item
If a line contains after leading whitespaces the sequence
@code{$quote}, the quoting character used for this input file is
changed to the first non-whitespace character following
@code{$quote}.  If no non-whitespace character is present before the
line ends quoting is disabled.

By default no quoting character is used.  In this mode strings are
terminated with the first unescaped line break.  If there is a
@code{$quote} sequence present newline need not be escaped.  Instead a
string is terminated with the first unescaped appearance of the quote
character.

A common usage of this feature would be to set the quote character to
@code{"}.  Then any appearance of the @code{"} in the strings must
be escaped using the backslash (i.e., @code{\"} must be written).

@item
Any other line must start with a number or an alphanumeric identifier
(with the underscore character included).  The following characters
(starting after the first whitespace character) will form the string
which gets associated with the currently selected set and the message
number represented by the number and identifier respectively.

If the start of the line is a number the message number is obvious.  It
is an error if the same message number already appeared for this set.

If the leading token was an identifier the message number gets
automatically assigned.  The value is the current maximum message
number for this set plus one.  It is an error if the identifier was
already used for a message in this set.  It is OK to reuse the
identifier for a message in another thread.  How to use the symbolic
identifiers will be explained below (@pxref{Common Usage}).  There is
one limitation with the identifier: it must not be @code{Set}.  The
reason will be explained below.

The text of the messages can contain escape characters.  The usual bunch
of characters known from the @w{ISO C} language are recognized
(@code{\n}, @code{\t}, @code{\v}, @code{\b}, @code{\r}, @code{\f},
@code{\\}, and @code{\@var{nnn}}, where @var{nnn} is the octal coding of
a character code).
@end itemize

@strong{Important:} The handling of identifiers instead of numbers for
the set and messages is a GNU extension.  Systems strictly following the
X/Open specification do not have this feature.  An example for a message
catalog file is this:

@smallexample
$ This is a leading comment.
$quote "

$set SetOne
1 Message with ID 1.
two "   Message with ID \"two\", which gets the value 2 assigned"

$set SetTwo
$ Since the last set got the number 1 assigned this set has number 2.
4000 "The numbers can be arbitrary, they need not start at one."
@end smallexample

This small example shows various aspects:
@itemize @bullet
@item
Lines 1 and 9 are comments since they start with @code{$} followed by
a whitespace.
@item
The quoting character is set to @code{"}.  Otherwise the quotes in the
message definition would have to be omitted and in this case the
message with the identifier @code{two} would lose its leading whitespace.
@item
Mixing numbered messages with messages having symbolic names is no
problem and the numbering happens automatically.
@end itemize


While this file format is pretty easy it is not the best possible for
use in a running program.  The @code{catopen} function would have to
parse the file and handle syntactic errors gracefully.  This is not so
easy and the whole process is pretty slow.  Therefore the @code{catgets}
functions expect the data in another more compact and ready-to-use file
format.  There is a special program @code{gencat} which is explained in
detail in the next section.

Files in this other format are not human readable.  To be easy to use by
programs it is a binary file.  But the format is byte order independent
so translation files can be shared by systems of arbitrary architecture
(as long as they use @theglibc{}).

Details about the binary file format are not important to know since
these files are always created by the @code{gencat} program.  The
sources of @theglibc{} also provide the sources for the
@code{gencat} program and so the interested reader can look through
these source files to learn about the file format.


@node The gencat program
@subsection Generate Message Catalogs files

@cindex gencat
The @code{gencat} program is specified in the X/Open standard and the
GNU implementation follows this specification and so processes
all correctly formed input files.  Additionally some extension are
implemented which help to work in a more reasonable way with the
@code{catgets} functions.

The @code{gencat} program can be invoked in two ways:

@example
`gencat [@var{Option} @dots{}] [@var{Output-File} [@var{Input-File} @dots{}]]`
@end example

This is the interface defined in the X/Open standard.  If no
@var{Input-File} parameter is given, input will be read from standard
input.  Multiple input files will be read as if they were concatenated.
If @var{Output-File} is also missing, the output will be written to
standard output.  To provide the interface one is used to from other
programs a second interface is provided.

@smallexample
`gencat [@var{Option} @dots{}] -o @var{Output-File} [@var{Input-File} @dots{}]`
@end smallexample

The option @samp{-o} is used to specify the output file and all file
arguments are used as input files.

Beside this one can use @file{-} or @file{/dev/stdin} for
@var{Input-File} to denote the standard input.  Corresponding one can
use @file{-} and @file{/dev/stdout} for @var{Output-File} to denote
standard output.  Using @file{-} as a file name is allowed in X/Open
while using the device names is a GNU extension.

The @code{gencat} program works by concatenating all input files and
then @strong{merging} the resulting collection of message sets with a
possibly existing output file.  This is done by removing all messages
with set/message number tuples matching any of the generated messages
from the output file and then adding all the new messages.  To
regenerate a catalog file while ignoring the old contents therefore
requires removing the output file if it exists.  If the output is
written to standard output no merging takes place.

@noindent
The following table shows the options understood by the @code{gencat}
program.  The X/Open standard does not specify any options for the
program so all of these are GNU extensions.

@table @samp
@item -V
@itemx --version
Print the version information and exit.
@item -h
@itemx --help
Print a usage message listing all available options, then exit successfully.
@item --new
Do not merge the new messages from the input files with the old content
of the output file.  The old content of the output file is discarded.
@item -H
@itemx --header=name
This option is used to emit the symbolic names given to sets and
messages in the input files for use in the program.  Details about how
to use this are given in the next section.  The @var{name} parameter to
this option specifies the name of the output file.  It will contain a
number of C preprocessor @code{#define}s to associate a name with a
number.

Please note that the generated file only contains the symbols from the
input files.  If the output is merged with the previous content of the
output file the possibly existing symbols from the file(s) which
generated the old output files are not in the generated header file.
@end table


@node Common Usage
@subsection How to use the @code{catgets} interface

The @code{catgets} functions can be used in two different ways.  By
following slavishly the X/Open specs and not relying on the extension
and by using the GNU extensions.  We will take a look at the former
method first to understand the benefits of extensions.

@subsubsection Not using symbolic names

Since the X/Open format of the message catalog files does not allow
symbol names we have to work with numbers all the time.  When we start
writing a program we have to replace all appearances of translatable
strings with something like

@smallexample
catgets (catdesc, set, msg, "string")
@end smallexample

@noindent
@var{catgets} is retrieved from a call to @code{catopen} which is
normally done once at the program start.  The @code{"string"} is the
string we want to translate.  The problems start with the set and
message numbers.

In a bigger program several programmers usually work at the same time on
the program and so coordinating the number allocation is crucial.
Though no two different strings must be indexed by the same tuple of
numbers it is highly desirable to reuse the numbers for equal strings
with equal translations (please note that there might be strings which
are equal in one language but have different translations due to
difference contexts).

The allocation process can be relaxed a bit by different set numbers for
different parts of the program.  So the number of developers who have to
coordinate the allocation can be reduced.  But still lists must be keep
track of the allocation and errors can easily happen.  These errors
cannot be discovered by the compiler or the @code{catgets} functions.
Only the user of the program might see wrong messages printed.  In the
worst cases the messages are so irritating that they cannot be
recognized as wrong.  Think about the translations for @code{"true"} and
@code{"false"} being exchanged.  This could result in a disaster.


@subsubsection Using symbolic names

The problems mentioned in the last section derive from the fact that:

@enumerate
@item
the numbers are allocated once and due to the possibly frequent use of
them it is difficult to change a number later.
@item
the numbers do not allow guessing anything about the string and
therefore collisions can easily happen.
@end enumerate

By constantly using symbolic names and by providing a method which maps
the string content to a symbolic name (however this will happen) one can
prevent both problems above.  The cost of this is that the programmer
has to write a complete message catalog file while s/he is writing the
program itself.

This is necessary since the symbolic names must be mapped to numbers
before the program sources can be compiled.  In the last section it was
described how to generate a header containing the mapping of the names.
E.g., for the example message file given in the last section we could
call the @code{gencat} program as follows (assume @file{ex.msg} contains
the sources).

@smallexample
gencat -H ex.h -o ex.cat ex.msg
@end smallexample

@noindent
This generates a header file with the following content:

@smallexample
#define SetTwoSet 0x2   /* ex.msg:8 */

#define SetOneSet 0x1   /* ex.msg:4 */
#define SetOnetwo 0x2   /* ex.msg:6 */
@end smallexample

As can be seen the various symbols given in the source file are mangled
to generate unique identifiers and these identifiers get numbers
assigned.  Reading the source file and knowing about the rules will
allow to predict the content of the header file (it is deterministic)
but this is not necessary.  The @code{gencat} program can take care for
everything.  All the programmer has to do is to put the generated header
file in the dependency list of the source files of her/his project and
add a rule to regenerate the header if any of the input files change.

One word about the symbol mangling.  Every symbol consists of two parts:
the name of the message set plus the name of the message or the special
string @code{Set}.  So @code{SetOnetwo} means this macro can be used to
access the translation with identifier @code{two} in the message set
@code{SetOne}.

The other names denote the names of the message sets.  The special
string @code{Set} is used in the place of the message identifier.

If in the code the second string of the set @code{SetOne} is used the C
code should look like this:

@smallexample
catgets (catdesc, SetOneSet, SetOnetwo,
         "   Message with ID \"two\", which gets the value 2 assigned")
@end smallexample

Writing the function this way will allow to change the message number
and even the set number without requiring any change in the C source
code.  (The text of the string is normally not the same; this is only
for this example.)


@subsubsection How does to this allow to develop

To illustrate the usual way to work with the symbolic version numbers
here is a little example.  Assume we want to write the very complex and
famous greeting program.  We start by writing the code as usual:

@smallexample
#include <stdio.h>
int
main (void)
@{
  printf ("Hello, world!\n");
  return 0;
@}
@end smallexample

Now we want to internationalize the message and therefore replace the
message with whatever the user wants.

@smallexample
#include <nl_types.h>
#include <stdio.h>
#include "msgnrs.h"
int
main (void)
@{
  nl_catd catdesc = catopen ("hello.cat", NL_CAT_LOCALE);
  printf (catgets (catdesc, SetMainSet, SetMainHello,
                   "Hello, world!\n"));
  catclose (catdesc);
  return 0;
@}
@end smallexample

We see how the catalog object is opened and the returned descriptor used
in the other function calls.  It is not really necessary to check for
failure of any of the functions since even in these situations the
functions will behave reasonable.  They simply will be return a
translation.

What remains unspecified here are the constants @code{SetMainSet} and
@code{SetMainHello}.  These are the symbolic names describing the
message.  To get the actual definitions which match the information in
the catalog file we have to create the message catalog source file and
process it using the @code{gencat} program.

@smallexample
$ Messages for the famous greeting program.
$quote "

$set Main
Hello "Hallo, Welt!\n"
@end smallexample

Now we can start building the program (assume the message catalog source
file is named @file{hello.msg} and the program source file @file{hello.c}):

@smallexample
% gencat -H msgnrs.h -o hello.cat hello.msg
% cat msgnrs.h
#define MainSet 0x1     /* hello.msg:4 */
#define MainHello 0x1   /* hello.msg:5 */
% gcc -o hello hello.c -I.
% cp hello.cat /usr/share/locale/de/LC_MESSAGES
% echo $LC_ALL
de
% ./hello
Hallo, Welt!
%
@end smallexample

The call of the @code{gencat} program creates the missing header file
@file{msgnrs.h} as well as the message catalog binary.  The former is
used in the compilation of @file{hello.c} while the later is placed in a
directory in which the @code{catopen} function will try to locate it.
Please check the @code{LC_ALL} environment variable and the default path
for @code{catopen} presented in the description above.


@node The Uniforum approach
@section The Uniforum approach to Message Translation

Sun Microsystems tried to standardize a different approach to message
translation in the Uniforum group.  There never was a real standard
defined but still the interface was used in Sun's operating systems.
Since this approach fits better in the development process of free
software it is also used throughout the GNU project and the GNU
@file{gettext} package provides support for this outside @theglibc{}.

The code of the @file{libintl} from GNU @file{gettext} is the same as
the code in @theglibc{}.  So the documentation in the GNU
@file{gettext} manual is also valid for the functionality here.  The
following text will describe the library functions in detail.  But the
numerous helper programs are not described in this manual.  Instead
people should read the GNU @file{gettext} manual
(@pxref{Top,,GNU gettext utilities,gettext,Native Language Support Library and Tools}).
We will only give a short overview.

Though the @code{catgets} functions are available by default on more
systems the @code{gettext} interface is at least as portable as the
former.  The GNU @file{gettext} package can be used wherever the
functions are not available.


@menu
* Message catalogs with gettext::  The @code{gettext} family of functions.
* Helper programs for gettext::    Programs to handle message catalogs
                                    for @code{gettext}.
@end menu


@node Message catalogs with gettext
@subsection The @code{gettext} family of functions

The paradigms underlying the @code{gettext} approach to message
translations is different from that of the @code{catgets} functions the
basic functionally is equivalent.  There are functions of the following
categories:

@menu
* Translation with gettext::       What has to be done to translate a message.
* Locating gettext catalog::       How to determine which catalog to be used.
* Advanced gettext functions::     Additional functions for more complicated
                                    situations.
* Charset conversion in gettext::  How to specify the output character set
                                    @code{gettext} uses.
* GUI program problems::           How to use @code{gettext} in GUI programs.
* Using gettextized software::     The possibilities of the user to influence
                                    the way @code{gettext} works.
@end menu

@node Translation with gettext
@subsubsection What has to be done to translate a message?

The @code{gettext} functions have a very simple interface.  The most
basic function just takes the string which shall be translated as the
argument and it returns the translation.  This is fundamentally
different from the @code{catgets} approach where an extra key is
necessary and the original string is only used for the error case.

If the string which has to be translated is the only argument this of
course means the string itself is the key.  I.e., the translation will
be selected based on the original string.  The message catalogs must
therefore contain the original strings plus one translation for any such
string.  The task of the @code{gettext} function is to compare the
argument string with the available strings in the catalog and return the
appropriate translation.  Of course this process is optimized so that
this process is not more expensive than an access using an atomic key
like in @code{catgets}.

The @code{gettext} approach has some advantages but also some
disadvantages.  Please see the GNU @file{gettext} manual for a detailed
discussion of the pros and cons.

All the definitions and declarations for @code{gettext} can be found in
the @file{libintl.h} header file.  On systems where these functions are
not part of the C library they can be found in a separate library named
@file{libintl.a} (or accordingly different for shared libraries).

@comment libintl.h
@comment GNU
@deftypefun {char *} gettext (const char *@var{msgid})
@safety{@prelim{}@mtsafe{@mtsenv{}}@asunsafe{@asucorrupt{} @ascuheap{} @asulock{} @ascudlopen{}}@acunsafe{@acucorrupt{} @aculock{} @acsfd{} @acsmem{}}}
@c Wrapper for dcgettext.
The @code{gettext} function searches the currently selected message
catalogs for a string which is equal to @var{msgid}.  If there is such a
string available it is returned.  Otherwise the argument string
@var{msgid} is returned.

Please note that although the return value is @code{char *} the
returned string must not be changed.  This broken type results from the
history of the function and does not reflect the way the function should
be used.

Please note that above we wrote ``message catalogs'' (plural).  This is
a specialty of the GNU implementation of these functions and we will
say more about this when we talk about the ways message catalogs are
selected (@pxref{Locating gettext catalog}).

The @code{gettext} function does not modify the value of the global
@var{errno} variable.  This is necessary to make it possible to write
something like

@smallexample
  printf (gettext ("Operation failed: %m\n"));
@end smallexample

Here the @var{errno} value is used in the @code{printf} function while
processing the @code{%m} format element and if the @code{gettext}
function would change this value (it is called before @code{printf} is
called) we would get a wrong message.

So there is no easy way to detect a missing message catalog besides
comparing the argument string with the result.  But it is normally the
task of the user to react on missing catalogs.  The program cannot guess
when a message catalog is really necessary since for a user who speaks
the language the program was developed in, the message does not need any translation.
@end deftypefun

The remaining two functions to access the message catalog add some
functionality to select a message catalog which is not the default one.
This is important if parts of the program are developed independently.
Every part can have its own message catalog and all of them can be used
at the same time.  The C library itself is an example: internally it
uses the @code{gettext} functions but since it must not depend on a
currently selected default message catalog it must specify all ambiguous
information.

@comment libintl.h
@comment GNU
@deftypefun {char *} dgettext (const char *@var{domainname}, const char *@var{msgid})
@safety{@prelim{}@mtsafe{@mtsenv{}}@asunsafe{@asucorrupt{} @ascuheap{} @asulock{} @ascudlopen{}}@acunsafe{@acucorrupt{} @aculock{} @acsfd{} @acsmem{}}}
@c Wrapper for dcgettext.
The @code{dgettext} function acts just like the @code{gettext}
function.  It only takes an additional first argument @var{domainname}
which guides the selection of the message catalogs which are searched
for the translation.  If the @var{domainname} parameter is the null
pointer the @code{dgettext} function is exactly equivalent to
@code{gettext} since the default value for the domain name is used.

As for @code{gettext} the return value type is @code{char *} which is an
anachronism.  The returned string must never be modified.
@end deftypefun

@comment libintl.h
@comment GNU
@deftypefun {char *} dcgettext (const char *@var{domainname}, const char *@var{msgid}, int @var{category})
@safety{@prelim{}@mtsafe{@mtsenv{}}@asunsafe{@asucorrupt{} @ascuheap{} @asulock{} @ascudlopen{}}@acunsafe{@acucorrupt{} @aculock{} @acsfd{} @acsmem{}}}
@c dcgettext @mtsenv @asucorrupt @ascuheap @asulock @ascudlopen @acucorrupt @aculock @acsfd @acsmem
@c  dcigettext @mtsenv @asucorrupt @ascuheap @asulock @ascudlopen @acucorrupt @aculock @acsfd @acsmem
@c   libc_rwlock_rdlock @asulock @aculock
@c   current_locale_name ok [protected from @mtslocale]
@c   tfind ok
@c   libc_rwlock_unlock ok
@c   plural_lookup ok
@c    plural_eval ok
@c    rawmemchr ok
@c   DETERMINE_SECURE ok, nothing
@c   strcmp ok
@c   strlen ok
@c   getcwd @ascuheap @acsmem @acsfd
@c   strchr ok
@c   stpcpy ok
@c   category_to_name ok
@c   guess_category_value @mtsenv
@c    getenv @mtsenv
@c    current_locale_name dup ok [protected from @mtslocale by dcigettext]
@c    strcmp ok
@c   ENABLE_SECURE ok
@c   _nl_find_domain @mtsenv @asucorrupt @ascuheap @asulock @ascudlopen @acucorrupt @aculock @acsfd @acsmem
@c    libc_rwlock_rdlock dup @asulock @aculock
@c    _nl_make_l10nflist dup @ascuheap @acsmem
@c    libc_rwlock_unlock dup ok
@c    _nl_load_domain @mtsenv @asucorrupt @ascuheap @asulock @ascudlopen @acucorrupt @aculock @acsfd @acsmem
@c     libc_lock_lock_recursive @aculock
@c     libc_lock_unlock_recursive @aculock
@c     open->open_not_cancel_2 @acsfd
@c     fstat ok
@c     mmap dup @acsmem
@c     close->close_not_cancel_no_status @acsfd
@c     malloc dup @ascuheap @acsmem
@c     read->read_not_cancel ok
@c     munmap dup @acsmem
@c     W dup ok
@c     strlen dup ok
@c     get_sysdep_segment_value ok
@c     memcpy dup ok
@c     hash_string dup ok
@c     free dup @ascuheap @acsmem
@c     libc_rwlock_init ok
@c     _nl_find_msg dup @mtsenv @asucorrupt @ascuheap @asulock @ascudlopen @acucorrupt @aculock @acsfd @acsmem
@c     libc_rwlock_fini ok
@c     EXTRACT_PLURAL_EXPRESSION @ascuheap @acsmem
@c      strstr dup ok
@c      isspace ok
@c      strtoul ok
@c      PLURAL_PARSE @ascuheap @acsmem
@c       malloc dup @ascuheap @acsmem
@c       free dup @ascuheap @acsmem
@c      INIT_GERMANIC_PLURAL ok, nothing
@c        the pre-C99 variant is @acucorrupt [protected from @mtuinit by dcigettext]
@c    _nl_expand_alias dup @ascuheap @asulock @acsmem @acsfd @aculock
@c    _nl_explode_name dup @ascuheap @acsmem
@c    libc_rwlock_wrlock dup @asulock @aculock
@c    free dup @asulock @aculock @acsfd @acsmem
@c   _nl_find_msg @mtsenv @asucorrupt @ascuheap @asulock @ascudlopen @acucorrupt @aculock @acsfd @acsmem
@c    _nl_load_domain dup @mtsenv @asucorrupt @ascuheap @asulock @ascudlopen @acucorrupt @aculock @acsfd @acsmem
@c    strlen ok
@c    hash_string ok
@c    W ok
@c     SWAP ok
@c      bswap_32 ok
@c    strcmp ok
@c    get_output_charset @mtsenv @ascuheap @acsmem
@c     getenv dup @mtsenv
@c     strlen dup ok
@c     malloc dup @ascuheap @acsmem
@c     memcpy dup ok
@c    libc_rwlock_rdlock dup @asulock @aculock
@c    libc_rwlock_unlock dup ok
@c    libc_rwlock_wrlock dup @asulock @aculock
@c    realloc @ascuheap @acsmem
@c    strdup @ascuheap @acsmem
@c    strstr ok
@c    strcspn ok
@c    mempcpy dup ok
@c    norm_add_slashes dup ok
@c    gconv_open @asucorrupt @ascuheap @asulock @ascudlopen @acucorrupt @aculock @acsmem @acsfd
@c     [protected from @mtslocale by dcigettext locale lock]
@c    free dup @ascuheap @acsmem
@c    libc_lock_lock @asulock @aculock
@c    calloc @ascuheap @acsmem
@c    gconv dup @acucorrupt [protected from @mtsrace and @asucorrupt by lock]
@c    libc_lock_unlock ok
@c   malloc @ascuheap @acsmem
@c   mempcpy ok
@c   memcpy ok
@c   strcpy ok
@c   libc_rwlock_wrlock @asulock @aculock
@c   tsearch @ascuheap @acucorrupt @acsmem [protected from @mtsrace and @asucorrupt]
@c    transcmp ok
@c     strmp dup ok
@c   free @ascuheap @acsmem
The @code{dcgettext} adds another argument to those which
@code{dgettext} takes.  This argument @var{category} specifies the last
piece of information needed to localize the message catalog.  I.e., the
domain name and the locale category exactly specify which message
catalog has to be used (relative to a given directory, see below).

The @code{dgettext} function can be expressed in terms of
@code{dcgettext} by using

@smallexample
dcgettext (domain, string, LC_MESSAGES)
@end smallexample

@noindent
instead of

@smallexample
dgettext (domain, string)
@end smallexample

This also shows which values are expected for the third parameter.  One
has to use the available selectors for the categories available in
@file{locale.h}.  Normally the available values are @code{LC_CTYPE},
@code{LC_COLLATE}, @code{LC_MESSAGES}, @code{LC_MONETARY},
@code{LC_NUMERIC}, and @code{LC_TIME}.  Please note that @code{LC_ALL}
must not be used and even though the names might suggest this, there is
no relation to the environment variable of this name.

The @code{dcgettext} function is only implemented for compatibility with
other systems which have @code{gettext} functions.  There is not really
any situation where it is necessary (or useful) to use a different value
than @code{LC_MESSAGES} for the @var{category} parameter.  We are
dealing with messages here and any other choice can only be irritating.

As for @code{gettext} the return value type is @code{char *} which is an
anachronism.  The returned string must never be modified.
@end deftypefun

When using the three functions above in a program it is a frequent case
that the @var{msgid} argument is a constant string.  So it is worthwhile to
optimize this case.  Thinking shortly about this one will realize that
as long as no new message catalog is loaded the translation of a message
will not change.  This optimization is actually implemented by the
@code{gettext}, @code{dgettext} and @code{dcgettext} functions.


@node Locating gettext catalog
@subsubsection How to determine which catalog to be used

The functions to retrieve the translations for a given message have a
remarkable simple interface.  But to provide the user of the program
still the opportunity to select exactly the translation s/he wants and
also to provide the programmer the possibility to influence the way to
locate the search for catalogs files there is a quite complicated
underlying mechanism which controls all this.  The code is complicated
the use is easy.

Basically we have two different tasks to perform which can also be
performed by the @code{catgets} functions:

@enumerate
@item
Locate the set of message catalogs.  There are a number of files for
different languages which all belong to the package.  Usually they
are all stored in the filesystem below a certain directory.

There can be arbitrarily many packages installed and they can follow
different guidelines for the placement of their files.

@item
Relative to the location specified by the package the actual translation
files must be searched, based on the wishes of the user.  I.e., for each
language the user selects the program should be able to locate the
appropriate file.
@end enumerate

This is the functionality required by the specifications for
@code{gettext} and this is also what the @code{catgets} functions are
able to do.  But there are some problems unresolved:

@itemize @bullet
@item
The language to be used can be specified in several different ways.
There is no generally accepted standard for this and the user always
expects the program to understand what s/he means.  E.g., to select the
German translation one could write @code{de}, @code{german}, or
@code{deutsch} and the program should always react the same.

@item
Sometimes the specification of the user is too detailed.  If s/he, e.g.,
specifies @code{de_DE.ISO-8859-1} which means German, spoken in Germany,
coded using the @w{ISO 8859-1} character set there is the possibility
that a message catalog matching this exactly is not available.  But
there could be a catalog matching @code{de} and if the character set
used on the machine is always @w{ISO 8859-1} there is no reason why this
later message catalog should not be used.  (We call this @dfn{message
inheritance}.)

@item
If a catalog for a wanted language is not available it is not always the
second best choice to fall back on the language of the developer and
simply not translate any message.  Instead a user might be better able
to read the messages in another language and so the user of the program
should be able to define a precedence order of languages.
@end itemize

We can divide the configuration actions in two parts: the one is
performed by the programmer, the other by the user.  We will start with
the functions the programmer can use since the user configuration will
be based on this.

As the functions described in the last sections already mention separate
sets of messages can be selected by a @dfn{domain name}.  This is a
simple string which should be unique for each program part that uses a
separate domain.  It is possible to use in one program arbitrarily many
domains at the same time.  E.g., @theglibc{} itself uses a domain
named @code{libc} while the program using the C Library could use a
domain named @code{foo}.  The important point is that at any time
exactly one domain is active.  This is controlled with the following
function.

@comment libintl.h
@comment GNU
@deftypefun {char *} textdomain (const char *@var{domainname})
@safety{@prelim{}@mtsafe{}@asunsafe{@asulock{} @ascuheap{}}@acunsafe{@aculock{} @acsmem{}}}
@c textdomain @asulock @ascuheap @aculock @acsmem
@c  libc_rwlock_wrlock @asulock @aculock
@c  strcmp ok
@c  strdup @ascuheap @acsmem
@c  free @ascuheap @acsmem
@c  libc_rwlock_unlock ok
The @code{textdomain} function sets the default domain, which is used in
all future @code{gettext} calls, to @var{domainname}.  Please note that
@code{dgettext} and @code{dcgettext} calls are not influenced if the
@var{domainname} parameter of these functions is not the null pointer.

Before the first call to @code{textdomain} the default domain is
@code{messages}.  This is the name specified in the specification of
the @code{gettext} API.  This name is as good as any other name.  No
program should ever really use a domain with this name since this can
only lead to problems.

The function returns the value which is from now on taken as the default
domain.  If the system went out of memory the returned value is
@code{NULL} and the global variable @var{errno} is set to @code{ENOMEM}.
Despite the return value type being @code{char *} the return string must
not be changed.  It is allocated internally by the @code{textdomain}
function.

If the @var{domainname} parameter is the null pointer no new default
domain is set.  Instead the currently selected default domain is
returned.

If the @var{domainname} parameter is the empty string the default domain
is reset to its initial value, the domain with the name @code{messages}.
This possibility is questionable to use since the domain @code{messages}
really never should be used.
@end deftypefun

@comment libintl.h
@comment GNU
@deftypefun {char *} bindtextdomain (const char *@var{domainname}, const char *@var{dirname})
@safety{@prelim{}@mtsafe{}@asunsafe{@ascuheap{}}@acunsafe{@acsmem{}}}
@c bindtextdomain @ascuheap @acsmem
@c  set_binding_values @ascuheap @acsmem
@c   libc_rwlock_wrlock dup @asulock @aculock
@c   strcmp dup ok
@c   strdup dup @ascuheap @acsmem
@c   free dup @ascuheap @acsmem
@c   malloc dup @ascuheap @acsmem
The @code{bindtextdomain} function can be used to specify the directory
which contains the message catalogs for domain @var{domainname} for the
different languages.  To be correct, this is the directory where the
hierarchy of directories is expected.  Details are explained below.

For the programmer it is important to note that the translations which
come with the program have to be placed in a directory hierarchy starting
at, say, @file{/foo/bar}.  Then the program should make a
@code{bindtextdomain} call to bind the domain for the current program to
this directory.  So it is made sure the catalogs are found.  A correctly
running program does not depend on the user setting an environment
variable.

The @code{bindtextdomain} function can be used several times and if the
@var{domainname} argument is different the previously bound domains
will not be overwritten.

If the program which wish to use @code{bindtextdomain} at some point of
time use the @code{chdir} function to change the current working
directory it is important that the @var{dirname} strings ought to be an
absolute pathname.  Otherwise the addressed directory might vary with
the time.

If the @var{dirname} parameter is the null pointer @code{bindtextdomain}
returns the currently selected directory for the domain with the name
@var{domainname}.

The @code{bindtextdomain} function returns a pointer to a string
containing the name of the selected directory name.  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
@code{bindtextdomain} the return value is @code{NULL} and the global
variable @var{errno} is set accordingly.
@end deftypefun


@node Advanced gettext functions
@subsubsection Additional functions for more complicated situations

The functions of the @code{gettext} family described so far (and all the
@code{catgets} functions as well) have one problem in the real world
which has 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:

@smallexample
   printf ("%d file%s deleted", n, n == 1 ? "" : "s");
@end smallexample

@noindent
After the first complaints from people internationalizing the code people
either completely avoided formulations like this or used strings like
@code{"file(s)"}.  Both look unnatural and should be avoided.  First
tries to solve the problem correctly looked like this:

@smallexample
   if (n == 1)
     printf ("%d file deleted", n);
   else
     printf ("%d files deleted", n);
@end smallexample

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 uses are universal.  But the handling of plural
forms differs widely between the language families.  There are two
things we can differ between (and even inside language families);

@itemize @bullet
@item
The form how plural forms are build differs.  This is a problem with
language 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.

@item
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.
@end itemize

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 @code{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
@code{gettext} behavior).  In this case the rules for Germanic language
are 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 @theglibc{}
(as well as the GNU @code{gettext} package) is written as part of the
GNU package and the coding standards for the GNU project require programs
to be written in English, this solution nevertheless fulfills its
purpose.

@comment libintl.h
@comment GNU
@deftypefun {char *} ngettext (const char *@var{msgid1}, const char *@var{msgid2}, unsigned long int @var{n})
@safety{@prelim{}@mtsafe{@mtsenv{}}@asunsafe{@asucorrupt{} @ascuheap{} @asulock{} @ascudlopen{}}@acunsafe{@acucorrupt{} @aculock{} @acsfd{} @acsmem{}}}
@c Wrapper for dcngettext.
The @code{ngettext} function is similar to the @code{gettext} function
as it finds the message catalogs in the same way.  But it takes two
extra arguments.  The @var{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 @var{msgid2} parameter is the plural form.
The parameter @var{n} is used to determine the plural form.  If no
message catalog is found @var{msgid1} is returned if @code{n == 1},
otherwise @code{msgid2}.

An example for the use of this function is:

@smallexample
  printf (ngettext ("%d file removed", "%d files removed", n), n);
@end smallexample

Please note that the numeric value @var{n} has to be passed to the
@code{printf} function as well.  It is not sufficient to pass it only to
@code{ngettext}.
@end deftypefun

@comment libintl.h
@comment GNU
@deftypefun {char *} dngettext (const char *@var{domain}, const char *@var{msgid1}, const char *@var{msgid2}, unsigned long int @var{n})
@safety{@prelim{}@mtsafe{@mtsenv{}}@asunsafe{@asucorrupt{} @ascuheap{} @asulock{} @ascudlopen{}}@acunsafe{@acucorrupt{} @aculock{} @acsfd{} @acsmem{}}}
@c Wrapper for dcngettext.
The @code{dngettext} is similar to the @code{dgettext} function in the
way the message catalog is selected.  The difference is that it takes
two extra parameters to provide the correct plural form.  These two
parameters are handled in the same way @code{ngettext} handles them.
@end deftypefun

@comment libintl.h
@comment GNU
@deftypefun {char *} dcngettext (const char *@var{domain}, const char *@var{msgid1}, const char *@var{msgid2}, unsigned long int @var{n}, int @var{category})
@safety{@prelim{}@mtsafe{@mtsenv{}}@asunsafe{@asucorrupt{} @ascuheap{} @asulock{} @ascudlopen{}}@acunsafe{@acucorrupt{} @aculock{} @acsfd{} @acsmem{}}}
@c Wrapper for dcigettext.
The @code{dcngettext} is similar to the @code{dcgettext} function in the
way the message catalog is selected.  The difference is that it takes
two extra parameters to provide the correct plural form.  These two
parameters are handled in the same way @code{ngettext} handles them.
@end deftypefun

@subsubheading The problem of plural forms

A description of the problem can be found at the beginning of the last
section.  Now there is the question how to solve it.  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
details are explained in the GNU @code{gettext} manual.  Here only a
bit of information is provided.

The information about the plural form selection has to be stored in the
header entry (the one with the empty @code{msgid} string).  It looks
like this:

@smallexample
Plural-Forms: nplurals=2; plural=n == 1 ? 0 : 1;
@end smallexample

The @code{nplurals} value must be a decimal number which specifies how
many different plural forms exist for this language.  The string
following @code{plural} is an expression using the C language
syntax.  Exceptions are that no negative numbers are allowed, numbers
must be decimal, and the only variable allowed is @code{n}.  This
expression will be evaluated whenever one of the functions
@code{ngettext}, @code{dngettext}, or @code{dcngettext} is called.  The
numeric value passed to these functions is then substituted for all uses
of the variable @code{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 @code{nplurals}.

@noindent
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).@footnote{Additions are welcome.  Send appropriate information to
@email{bug-glibc-manual@@gnu.org}.}

@table @asis
@item 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:

@smallexample
Plural-Forms: nplurals=1; plural=0;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Finno-Ugric family
Hungarian
@item Asian family
Japanese, Korean
@item Turkic/Altaic family
Turkish
@end table

@item Two forms, singular used for one only
This is the form used in most existing programs since it is what English
uses.  A header entry would look like this:

@smallexample
Plural-Forms: nplurals=2; plural=n != 1;
@end smallexample

(Note: this uses the feature of C expressions that boolean expressions
have to value zero or one.)

@noindent
Languages with this property include:

@table @asis
@item Germanic family
Danish, Dutch, English, German, Norwegian, Swedish
@item Finno-Ugric family
Estonian, Finnish
@item Latin/Greek family
Greek
@item Semitic family
Hebrew
@item Romance family
Italian, Portuguese, Spanish
@item Artificial
Esperanto
@end table

@item Two forms, singular used for zero and one
Exceptional case in the language family.  The header entry would be:

@smallexample
Plural-Forms: nplurals=2; plural=n>1;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Romanic family
French, Brazilian Portuguese
@end table

@item Three forms, special case for zero
The header entry would be:

@smallexample
Plural-Forms: nplurals=3; plural=n%10==1 && n%100!=11 ? 0 : n != 0 ? 1 : 2;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Baltic family
Latvian
@end table

@item Three forms, special cases for one and two
The header entry would be:

@smallexample
Plural-Forms: nplurals=3; plural=n==1 ? 0 : n==2 ? 1 : 2;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Celtic
Gaeilge (Irish)
@end table

@item Three forms, special case for numbers ending in 1[2-9]
The header entry would look like this:

@smallexample
Plural-Forms: nplurals=3; \
    plural=n%10==1 && n%100!=11 ? 0 : \
           n%10>=2 && (n%100<10 || n%100>=20) ? 1 : 2;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Baltic family
Lithuanian
@end table

@item 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:

@smallexample
Plural-Forms: nplurals=3; \
    plural=n%100/10==1 ? 2 : n%10==1 ? 0 : (n+9)%10>3 ? 2 : 1;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Slavic family
Croatian, Czech, Russian, Ukrainian
@end table

@item Three forms, special cases for 1 and 2, 3, 4
The header entry would look like this:

@smallexample
Plural-Forms: nplurals=3; \
    plural=(n==1) ? 1 : (n>=2 && n<=4) ? 2 : 0;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Slavic family
Slovak
@end table

@item Three forms, special case for one and some numbers ending in 2, 3, or 4
The header entry would look like this:

@smallexample
Plural-Forms: nplurals=3; \
    plural=n==1 ? 0 : \
           n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Slavic family
Polish
@end table

@item Four forms, special case for one and all numbers ending in 02, 03, or 04
The header entry would look like this:

@smallexample
Plural-Forms: nplurals=4; \
    plural=n%100==1 ? 0 : n%100==2 ? 1 : n%100==3 || n%100==4 ? 2 : 3;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Slavic family
Slovenian
@end table
@end table


@node Charset conversion in gettext
@subsubsection How to specify the output character set @code{gettext} uses

@code{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 @code{nl_langinfo
(CODESET)}, which depends on the @code{LC_CTYPE} part of the current
locale.  But programs which store strings in a locale independent way
(e.g. UTF-8) can request that @code{gettext} and related functions
return the translations in that encoding, by use of the
@code{bind_textdomain_codeset} function.

Note that the @var{msgid} argument to @code{gettext} is not subject to
character set conversion.  Also, when @code{gettext} does not find a
translation for @var{msgid}, it returns @var{msgid} unchanged --
independently of the current output character set.  It is therefore
recommended that all @var{msgid}s be US-ASCII strings.

@comment libintl.h
@comment GNU
@deftypefun {char *} bind_textdomain_codeset (const char *@var{domainname}, const char *@var{codeset})
@safety{@prelim{}@mtsafe{}@asunsafe{@ascuheap{}}@acunsafe{@acsmem{}}}
@c bind_textdomain_codeset @ascuheap @acsmem
@c  set_binding_values dup @ascuheap @acsmem
The @code{bind_textdomain_codeset} function can be used to specify the
output character set for message catalogs for domain @var{domainname}.
The @var{codeset} argument must be a valid codeset name which can be used
for the @code{iconv_open} function, or a null pointer.

If the @var{codeset} parameter is the null pointer,
@code{bind_textdomain_codeset} returns the currently selected codeset
for the domain with the name @var{domainname}.  It returns @code{NULL} if
no codeset has yet been selected.

The @code{bind_textdomain_codeset} function can be used several times.
If used multiple times with the same @var{domainname} argument, the
later call overrides the settings made by the earlier one.

The @code{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
@code{bind_textdomain_codeset}, the return value is @code{NULL} and the
global variable @var{errno} is set accordingly.
@end deftypefun


@node GUI program problems
@subsubsection How to use @code{gettext} in GUI programs

One place where the @code{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 @code{gettext} approach is
wrong and instead @code{catgets} should be used which indeed does not
have this problem.  But there is a very simple and powerful method to
handle these kind of problems with the @code{gettext} functions.

@noindent
As an example consider the following fictional situation.  A GUI program
has a menu bar with the following entries:

@smallexample
+------------+------------+--------------------------------------+
| File       | Printer    |                                      |
+------------+------------+--------------------------------------+
| Open     | | Select   |
| New      | | Open     |
+----------+ | Connect  |
             +----------+
@end smallexample

To have the strings @code{File}, @code{Printer}, @code{Open},
@code{New}, @code{Select}, and @code{Connect} translated there has to be
at some point in the code a call to a function of the @code{gettext}
family.  But in two places the string passed into the function would be
@code{Open}.  The translations might not be the same and therefore we
are in the dilemma described above.

One solution to this problem is to artificially extend the strings
to make them unambiguous.  But what would the program do if no
translation is available?  The extended string is not what should be
printed.  So we should use a slightly modified version of the functions.

To extend the strings a uniform method should be used.  E.g., in the
example above, the strings could be chosen as

@smallexample
Menu|File
Menu|Printer
Menu|File|Open
Menu|File|New
Menu|Printer|Select
Menu|Printer|Open
Menu|Printer|Connect
@end smallexample

Now all the strings are different and if now instead of @code{gettext}
the following little wrapper function is used, everything works just
fine:

@cindex sgettext
@smallexample
  char *
  sgettext (const char *msgid)
  @{
    char *msgval = gettext (msgid);
    if (msgval == msgid)
      msgval = strrchr (msgid, '|') + 1;
    return msgval;
  @}
@end smallexample

What this little function does is to recognize the case when no
translation is available.  This can be done very efficiently by a
pointer comparison since the return value is the input value.  If there
is no translation we know that the input string is in the format we used
for the Menu entries and therefore contains a @code{|} character.  We
simply search for the last occurrence of this character and return a
pointer to the character following it.  That's it!

If one now consistently uses the extended string form and replaces
the @code{gettext} calls with calls to @code{sgettext} (this is normally
limited to very few places in the GUI implementation) then it is
possible to produce a program which can be internationalized.

With advanced compilers (such as GNU C) one can write the
@code{sgettext} functions as an inline function or as a macro like this:

@cindex sgettext
@smallexample
#define sgettext(msgid) \
  (@{ const char *__msgid = (msgid);            \
     char *__msgstr = gettext (__msgid);       \
     if (__msgval == __msgid)                  \
       __msgval = strrchr (__msgid, '|') + 1;  \
     __msgval; @})
@end smallexample

The other @code{gettext} functions (@code{dgettext}, @code{dcgettext}
and the @code{ngettext} equivalents) can and should have corresponding
functions as well which look almost identical, except for the parameters
and the call to the underlying function.

Now there is of course the question why such functions do not exist in
@theglibc{}?  There are two parts of the answer to this question.

@itemize @bullet
@item
They are easy to write and therefore can be provided by the project they
are used in.  This is not an answer by itself and must be seen together
with the second part which is:

@item
There is no way the C library can contain a version which can work
everywhere.  The problem is the selection of the character to separate
the prefix from the actual string in the extended string.  The
examples above used @code{|} which is a quite good choice because it
resembles a notation frequently used in this context and it also is a
character not often used in message strings.

But what if the character is used in message strings.  Or if the chose
character is not available in the character set on the machine one
compiles (e.g., @code{|} is not required to exist for @w{ISO C}; this is
why the @file{iso646.h} file exists in @w{ISO C} programming environments).
@end itemize

There is only one more comment to make left.  The wrapper function above
requires that the translations strings are not extended themselves.
This is only logical.  There is no need to disambiguate the strings
(since they are never used as keys for a search) and one also saves
quite some memory and disk space by doing this.


@node Using gettextized software
@subsubsection User influence on @code{gettext}

The last sections described what the programmer can do to
internationalize the messages of the program.  But it is finally up to
the user to select the message s/he wants to see.  S/He must understand
them.

The POSIX locale model uses the environment variables @code{LC_COLLATE},
@code{LC_CTYPE}, @code{LC_MESSAGES}, @code{LC_MONETARY}, @code{LC_NUMERIC},
and @code{LC_TIME} to select the locale which is to be used.  This way
the user can influence lots of functions.  As we mentioned above, the
@code{gettext} functions also take advantage of this.

To understand how this happens it is necessary to take a look at the
various components of the filename which gets computed to locate a
message catalog.  It is composed as follows:

@smallexample
@var{dir_name}/@var{locale}/LC_@var{category}/@var{domain_name}.mo
@end smallexample

The default value for @var{dir_name} is system specific.  It is computed
from the value given as the prefix while configuring the C library.
This value normally is @file{/usr} or @file{/}.  For the former the
complete @var{dir_name} is:

@smallexample
/usr/share/locale
@end smallexample

We can use @file{/usr/share} since the @file{.mo} files containing the
message catalogs are system independent, so all systems can use the same
files.  If the program executed the @code{bindtextdomain} function for
the message domain that is currently handled, the @code{dir_name}
component is exactly the value which was given to the function as
the second parameter.  I.e., @code{bindtextdomain} allows overwriting
the only system dependent and fixed value to make it possible to
address files anywhere in the filesystem.

The @var{category} is the name of the locale category which was selected
in the program code.  For @code{gettext} and @code{dgettext} this is
always @code{LC_MESSAGES}, for @code{dcgettext} this is selected by the
value of the third parameter.  As said above it should be avoided to
ever use a category other than @code{LC_MESSAGES}.

The @var{locale} component is computed based on the category used.  Just
like for the @code{setlocale} function here comes the user selection
into the play.  Some environment variables are examined in a fixed order
and the first environment variable set determines the return value of
the lookup process.  In detail, for the category @code{LC_xxx} the
following variables in this order are examined:

@table @code
@item LANGUAGE
@item LC_ALL
@item LC_xxx
@item LANG
@end table

This looks very familiar.  With the exception of the @code{LANGUAGE}
environment variable this is exactly the lookup order the
@code{setlocale} function uses.  But why introduce the @code{LANGUAGE}
variable?

The reason is that the syntax of the values these variables can have is
different to what is expected by the @code{setlocale} function.  If we
would set @code{LC_ALL} to a value following the extended syntax that
would mean the @code{setlocale} function will never be able to use the
value of this variable as well.  An additional variable removes this
problem plus we can select the language independently of the locale
setting which sometimes is useful.

While for the @code{LC_xxx} variables the value should consist of
exactly one specification of a locale the @code{LANGUAGE} variable's
value can consist of a colon separated list of locale names.  The
attentive reader will realize that this is the way we manage to
implement one of our additional demands above: we want to be able to
specify an ordered list of languages.

Back to the constructed filename we have only one component missing.
The @var{domain_name} part is the name which was either registered using
the @code{textdomain} function or which was given to @code{dgettext} or
@code{dcgettext} as the first parameter.  Now it becomes obvious that a
good choice for the domain name in the program code is a string which is
closely related to the program/package name.  E.g., for @theglibc{}
the domain name is @code{libc}.

@noindent
A limited piece of example code should show how the program is supposed
to work:

@smallexample
@{
  setlocale (LC_ALL, "");
  textdomain ("test-package");
  bindtextdomain ("test-package", "/usr/local/share/locale");
  puts (gettext ("Hello, world!"));
@}
@end smallexample

At the program start the default domain is @code{messages}, and the
default locale is "C".  The @code{setlocale} call sets the locale
according to the user's environment variables; remember that correct
functioning of @code{gettext} relies on the correct setting of the
@code{LC_MESSAGES} locale (for looking up the message catalog) and
of the @code{LC_CTYPE} locale (for the character set conversion).
The @code{textdomain} call changes the default domain to
@code{test-package}.  The @code{bindtextdomain} call specifies that
the message catalogs for the domain @code{test-package} can be found
below the directory @file{/usr/local/share/locale}.

If the user sets in her/his environment the variable @code{LANGUAGE}
to @code{de} the @code{gettext} function will try to use the
translations from the file

@smallexample
/usr/local/share/locale/de/LC_MESSAGES/test-package.mo
@end smallexample

From the above descriptions it should be clear which component of this
filename is determined by which source.

In the above example we assumed the @code{LANGUAGE} environment
variable to be @code{de}.  This might be an appropriate selection but what
happens if the user wants to use @code{LC_ALL} because of the wider
usability and here the required value is @code{de_DE.ISO-8859-1}?  We
already mentioned above that a situation like this is not infrequent.
E.g., a person might prefer reading a dialect and if this is not
available fall back on the standard language.

The @code{gettext} functions know about situations like this and can
handle them gracefully.  The functions recognize the format of the value
of the environment variable.  It can split the value is different pieces
and by leaving out the only or the other part it can construct new
values.  This happens of course in a predictable way.  To understand
this one must know the format of the environment variable value.  There
is one more or less standardized form, originally from the X/Open
specification:

@code{language[_territory[.codeset]][@@modifier]}

Less specific locale names will be stripped in the order of the
following list:

@enumerate
@item
@code{codeset}
@item
@code{normalized codeset}
@item
@code{territory}
@item
@code{modifier}
@end enumerate

The @code{language} field will never be dropped for obvious reasons.

The only new thing is the @code{normalized codeset} entry.  This is
another goodie which is introduced to help reduce the chaos which
derives from the inability of people to standardize the names of
character sets.  Instead of @w{ISO-8859-1} one can often see @w{8859-1},
@w{88591}, @w{iso8859-1}, or @w{iso_8859-1}.  The @code{normalized
codeset} value is generated from the user-provided character set name by
applying the following rules:

@enumerate
@item
Remove all characters besides numbers and letters.
@item
Fold letters to lowercase.
@item
If the same only contains digits prepend the string @code{"iso"}.
@end enumerate

@noindent
So all of the above names will be normalized to @code{iso88591}.  This
allows the program user much more freedom in choosing the locale name.

Even this extended functionality still does not help to solve the
problem that completely different names can be used to denote the same
locale (e.g., @code{de} and @code{german}).  To be of help in this
situation the locale implementation and also the @code{gettext}
functions know about aliases.

The file @file{/usr/share/locale/locale.alias} (replace @file{/usr} with
whatever prefix you used for configuring the C library) contains a
mapping of alternative names to more regular names.  The system manager
is free to add new entries to fill her/his own needs.  The selected
locale from the environment is compared with the entries in the first
column of this file ignoring the case.  If they match, the value of the
second column is used instead for the further handling.

In the description of the format of the environment variables we already
mentioned the character set as a factor in the selection of the message
catalog.  In fact, only catalogs which contain text written using the
character set of the system/program can be used (directly; there will
come a solution for this some day).  This means for the user that s/he
will always have to take care of this.  If in the collection of the
message catalogs there are files for the same language but coded using
different character sets the user has to be careful.


@node Helper programs for gettext
@subsection Programs to handle message catalogs for @code{gettext}

@Theglibc{} does not contain the source code for the programs to
handle message catalogs for the @code{gettext} functions.  As part of
the GNU project the GNU gettext package contains everything the
developer needs.  The functionality provided by the tools in this
package by far exceeds the abilities of the @code{gencat} program
described above for the @code{catgets} functions.

There is a program @code{msgfmt} which is the equivalent program to the
@code{gencat} program.  It generates from the human-readable and
-editable form of the message catalog a binary file which can be used by
the @code{gettext} functions.  But there are several more programs
available.

The @code{xgettext} program can be used to automatically extract the
translatable messages from a source file.  I.e., the programmer need not
take care of the translations and the list of messages which have to be
translated.  S/He will simply wrap the translatable string in calls to
@code{gettext} et.al and the rest will be done by @code{xgettext}.  This
program has a lot of options which help to customize the output or
help to understand the input better.

Other programs help to manage the development cycle when new messages appear
in the source files or when a new translation of the messages appears.
Here it should only be noted that using all the tools in GNU gettext it
is possible to @emph{completely} automate the handling of message
catalogs.  Besides marking the translatable strings in the source code and
generating the translations the developers do not have anything to do
themselves.