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