* encoding.c (rb_locale_charmap): fallback to codepage if no

locale is found.  [ruby-core:21110]

* missing/langinfo.c (nl_langinfo_codeset): returns NULL if no
  locale is found.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@21511 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
This commit is contained in:
nobu 2009-01-14 14:57:34 +00:00
parent 3d829c8697
commit 22e711c4bf
3 changed files with 58 additions and 28 deletions

View File

@ -1,3 +1,11 @@
Wed Jan 14 23:57:28 2009 Nobuyoshi Nakada <nobu@ruby-lang.org>
* encoding.c (rb_locale_charmap): fallback to codepage if no
locale is found. [ruby-core:21110]
* missing/langinfo.c (nl_langinfo_codeset): returns NULL if no
locale is found.
Wed Jan 14 22:38:30 2009 Tanaka Akira <akr@fsij.org> Wed Jan 14 22:38:30 2009 Tanaka Akira <akr@fsij.org>
* gc.c (negative_size_allocation_error_with_gvl): abolish a warning. * gc.c (negative_size_allocation_error_with_gvl): abolish a warning.

View File

@ -1218,6 +1218,15 @@ rb_locale_charmap(VALUE klass)
{ {
#if defined NO_LOCALE_CHARMAP #if defined NO_LOCALE_CHARMAP
return rb_usascii_str_new2("ASCII-8BIT"); return rb_usascii_str_new2("ASCII-8BIT");
#elif defined __CYGWIN__
const char *nl_langinfo_codeset(void);
const char *codeset = nl_langinfo_codeset();
char cp[sizeof(int) * 3 + 4];
if (codeset) {
snprintf(cp, sizeof(cp), "CP%d", GetConsoleCP());
codeset = cp;
}
return rb_usascii_str_new2(codeset);
#elif defined HAVE_LANGINFO_H #elif defined HAVE_LANGINFO_H
char *codeset; char *codeset;
codeset = nl_langinfo(CODESET); codeset = nl_langinfo(CODESET);

View File

@ -1,3 +1,4 @@
/* -*- c-file-style: "gnu" -*- */
/* /*
* This is a quick-and-dirty emulator of the nl_langinfo(CODESET) * This is a quick-and-dirty emulator of the nl_langinfo(CODESET)
* function defined in the Single Unix Specification for those systems * function defined in the Single Unix Specification for those systems
@ -33,7 +34,16 @@
#include <stdlib.h> #include <stdlib.h>
#include <string.h> #include <string.h>
#if defined _WIN32 || defined __CYGWIN__
#include <windows.h>
int snprintf(char *, size_t, const char *, ...);
#if defined _WIN32
#define strncasecmp strnicmp
#endif
#endif
#ifdef HAVE_LANGINFO_H
#include "langinfo.h" #include "langinfo.h"
#endif
#define C_CODESET "US-ASCII" /* Return this as the encoding of the #define C_CODESET "US-ASCII" /* Return this as the encoding of the
* C/POSIX locale. Could as well one day * C/POSIX locale. Could as well one day
@ -44,12 +54,10 @@
static char buf[16]; static char buf[16];
static const char *nl_langinfo_0(nl_item item) const char *
nl_langinfo_codeset(void)
{ {
char *l, *p; const char *l, *p;
if (item != CODESET)
return NULL;
if (((l = getenv("LC_ALL")) && *l) || if (((l = getenv("LC_ALL")) && *l) ||
((l = getenv("LC_CTYPE")) && *l) || ((l = getenv("LC_CTYPE")) && *l) ||
@ -59,30 +67,29 @@ static const char *nl_langinfo_0(nl_item item)
return C_CODESET; return C_CODESET;
/* check for encoding name fragment */ /* check for encoding name fragment */
p = strchr(l, '.'); p = strchr(l, '.');
if (p++) { if (!p++) p = l;
if (strstart(p, "UTF")) if (strstart(p, "UTF"))
return "UTF-8"; return "UTF-8";
if (strstart(p, "8859-")) { if (strstart(p, "8859-")) {
if (digit(p[5])) {
p += 5;
memcpy(buf, "ISO-8859-\0\0", 12); memcpy(buf, "ISO-8859-\0\0", 12);
if (digit(p[5])) { buf[9] = *p++;
p += 5; if (digit(*p)) buf[10] = *p++;
buf[9] = *p++; return buf;
if (digit(*p)) buf[10] = *p++;
return buf;
}
} }
if (strstart(p, "KOI8-R")) return "KOI8-R";
if (strstart(p, "KOI8-U")) return "KOI8-U";
if (strstart(p, "620")) return "TIS-620";
if (strstart(p, "2312")) return "GB2312";
if (strstart(p, "HKSCS")) return "Big5HKSCS"; /* no MIME charset */
if (strstart(p, "BIG5")) return "Big5";
if (strstart(p, "GBK")) return "GBK"; /* no MIME charset */
if (strstart(p, "18030")) return "GB18030"; /* no MIME charset */
if (strstart(p, "Shift_JIS") || strstart(p, "SJIS")) return "Shift_JIS";
/* check for conclusive modifier */
if (strstart(p, "euro")) return "ISO-8859-15";
} }
if (strstart(p, "KOI8-R")) return "KOI8-R";
if (strstart(p, "KOI8-U")) return "KOI8-U";
if (strstart(p, "620")) return "TIS-620";
if (strstart(p, "2312")) return "GB2312";
if (strstart(p, "HKSCS")) return "Big5HKSCS"; /* no MIME charset */
if (strstart(p, "BIG5")) return "Big5";
if (strstart(p, "GBK")) return "GBK"; /* no MIME charset */
if (strstart(p, "18030")) return "GB18030"; /* no MIME charset */
if (strstart(p, "Shift_JIS") || strstart(p, "SJIS")) return "Windows-31J";
/* check for conclusive modifier */
if (strstart(p, "euro")) return "ISO-8859-15";
/* check for language (and perhaps country) codes */ /* check for language (and perhaps country) codes */
if (strstart(l, "zh_TW")) return "Big5"; if (strstart(l, "zh_TW")) return "Big5";
if (strstart(l, "zh_HK")) return "Big5HKSCS"; /* no MIME charset */ if (strstart(l, "zh_HK")) return "Big5HKSCS"; /* no MIME charset */
@ -106,15 +113,21 @@ static const char *nl_langinfo_0(nl_item item)
/* Send me further rules if you like, but don't forget that we are /* Send me further rules if you like, but don't forget that we are
* *only* interested in locale naming conventions on platforms * *only* interested in locale naming conventions on platforms
* that do not already provide an nl_langinfo(CODESET) implementation. */ * that do not already provide an nl_langinfo(CODESET) implementation. */
return "ISO-8859-1"; /* should perhaps be "UTF-8" instead */
} }
return C_CODESET; return NULL;
} }
#ifdef HAVE_LANGINFO_H
char *nl_langinfo(nl_item item) char *nl_langinfo(nl_item item)
{ {
return (char *)nl_langinfo_0(item); const char *codeset;
if (item != CODESET)
return NULL;
codeset = nl_langinfo_codeset();
if (!codeset) codeset = C_CODESET;
return (char *)codeset;
} }
#endif
/* For a demo, compile with "gcc -W -Wall -o langinfo -D TEST langinfo.c" */ /* For a demo, compile with "gcc -W -Wall -o langinfo -D TEST langinfo.c" */