gettext.c

   1 /*
   2  * Copyright (c) 2010 Ævar Arnfjörð Bjarmason
   3  */
   4
   5 #include "git-compat-util.h"
   6 #include "gettext.h"
   7 #include "strbuf.h"
   8 #include "utf8.h"
   9
  10 #ifndef NO_GETTEXT
  11 #       include <locale.h>
  12 #       include <libintl.h>
  13 #       ifdef HAVE_LIBCHARSET_H
  14 #               include <libcharset.h>
  15 #       else
  16 #               include <langinfo.h>
  17 #               define locale_charset() nl_langinfo(CODESET)
  18 #       endif
  19 #endif
  20
  21 /*
  22  * Guess the user's preferred languages from the value in LANGUAGE environment
  23  * variable and LC_MESSAGES locale category if NO_GETTEXT is not defined.
  24  *
  25  * The result can be a colon-separated list like "ko:ja:en".
  26  */
  27 const char *get_preferred_languages(void)
  28 {
  29         const char *retval;
  30
  31         retval = getenv("LANGUAGE");
  32         if (retval && *retval)
  33                 return retval;
  34
  35 #ifndef NO_GETTEXT
  36         retval = setlocale(LC_MESSAGES, NULL);
  37         if (retval && *retval &&
  38                 strcmp(retval, "C") &&
  39                 strcmp(retval, "POSIX"))
  40                 return retval;
  41 #endif
  42
  43         return NULL;
  44 }
  45
  46 #ifdef GETTEXT_POISON
  47 int use_gettext_poison(void)
  48 {
  49         static int poison_requested = -1;
  50         if (poison_requested == -1)
  51                 poison_requested = getenv("GIT_GETTEXT_POISON") ? 1 : 0;
  52         return poison_requested;
  53 }
  54 #endif
  55
  56 #ifndef NO_GETTEXT
  57 static int test_vsnprintf(const char *fmt, ...)
  58 {
  59         char buf[26];
  60         int ret;
  61         va_list ap;
  62         va_start(ap, fmt);
  63         ret = vsnprintf(buf, sizeof(buf), fmt, ap);
  64         va_end(ap);
  65         return ret;
  66 }
  67
  68 static const char *charset;
  69 static void init_gettext_charset(const char *domain)
  70 {
  71         /*
  72            This trick arranges for messages to be emitted in the user's
  73            requested encoding, but avoids setting LC_CTYPE from the
  74            environment for the whole program.
  75
  76            This primarily done to avoid a bug in vsnprintf in the GNU C
  77            Library [1]. which triggered a "your vsnprintf is broken" error
  78            on Git's own repository when inspecting v0.99.6~1 under a UTF-8
  79            locale.
  80
  81            That commit contains a ISO-8859-1 encoded author name, which
  82            the locale aware vsnprintf(3) won't interpolate in the format
  83            argument, due to mismatch between the data encoding and the
  84            locale.
  85
  86            Even if it wasn't for that bug we wouldn't want to use LC_CTYPE at
  87            this point, because it'd require auditing all the code that uses C
  88            functions whose semantics are modified by LC_CTYPE.
  89
  90            But only setting LC_MESSAGES as we do creates a problem, since
  91            we declare the encoding of our PO files[2] the gettext
  92            implementation will try to recode it to the user's locale, but
  93            without LC_CTYPE it'll emit something like this on 'git init'
  94            under the Icelandic locale:
  95
  96                Bj? til t?ma Git lind ? /hlagh/.git/
  97
  98            Gettext knows about the encoding of our PO file, but we haven't
  99            told it about the user's encoding, so all the non-US-ASCII
 100            characters get encoded to question marks.
 101
 102            But we're in luck! We can set LC_CTYPE from the environment
 103            only while we call nl_langinfo and
 104            bind_textdomain_codeset. That suffices to tell gettext what
 105            encoding it should emit in, so it'll now say:
 106
 107                Bjó til tóma Git lind í /hlagh/.git/
 108
 109            And the equivalent ISO-8859-1 string will be emitted under a
 110            ISO-8859-1 locale.
 111
 112            With this change way we get the advantages of setting LC_CTYPE
 113            (talk to the user in his language/encoding), without the major
 114            drawbacks (changed semantics for C functions we rely on).
 115
 116            However foreign functions using other message catalogs that
 117            aren't using our neat trick will still have a problem, e.g. if
 118            we have to call perror(3):
 119
 120            #include <stdio.h>
 121            #include <locale.h>
 122            #include <errno.h>
 123
 124            int main(void)
 125            {
 126                    setlocale(LC_MESSAGES, "");
 127                    setlocale(LC_CTYPE, "C");
 128                    errno = ENODEV;
 129                    perror("test");
 130                    return 0;
 131            }
 132
 133            Running that will give you a message with question marks:
 134
 135            $ LANGUAGE= LANG=de_DE.utf8 ./test
 136            test: Kein passendes Ger?t gefunden
 137
 138            The vsnprintf bug has been fixed since glibc 2.17.
 139
 140            Then we could simply set LC_CTYPE from the environment, which would
 141            make things like the external perror(3) messages work.
 142
 143            See t/t0203-gettext-setlocale-sanity.sh's "gettext.c" tests for
 144            regression tests.
 145
 146            1. http://sourceware.org/bugzilla/show_bug.cgi?id=6530
 147            2. E.g. "Content-Type: text/plain; charset=UTF-8\n" in po/is.po
 148         */
 149         setlocale(LC_CTYPE, "");
 150         charset = locale_charset();
 151         bind_textdomain_codeset(domain, charset);
 152         /* the string is taken from v0.99.6~1 */
 153         if (test_vsnprintf("%.*s", 13, "David_K\345gedal") < 0)
 154                 setlocale(LC_CTYPE, "C");
 155 }
 156
 157 void git_setup_gettext(void)
 158 {
 159         const char *podir = getenv("GIT_TEXTDOMAINDIR");
 160
 161         if (!podir)
 162                 podir = GIT_LOCALE_PATH;
 163         bindtextdomain("git", podir);
 164         setlocale(LC_MESSAGES, "");
 165         setlocale(LC_TIME, "");
 166         init_gettext_charset("git");
 167         textdomain("git");
 168 }
 169
 170 /* return the number of columns of string 's' in current locale */
 171 int gettext_width(const char *s)
 172 {
 173         static int is_utf8 = -1;
 174         if (is_utf8 == -1)
 175                 is_utf8 = !strcmp(charset, "UTF-8");
 176
 177         return is_utf8 ? utf8_strwidth(s) : strlen(s);
 178 }
 179 #endif