common/loadinfo.c

/* Copyright (C) 1995-2014 Free Software Foundation, Inc.
   This file is part of the GNU C Library.
   Contributed by Ulrich Drepper <drepper@gnu.ai.mit.edu>, 1995.

   The GNU C Library is free software; you can redistribute it and/or
   modify it under the terms of the GNU Lesser General Public
   License as published by the Free Software Foundation; either
   version 2.1 of the License, or (at your option) any later version.

   The GNU C Library is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
   Lesser General Public License for more details.

   You should have received a copy of the GNU Lesser General Public
   License along with the GNU C Library;

/* Extracted from glibc's intl/i10nflist.c and intl/explodename.c.  */

#include "third_party/glibc_locales/common/loadinfo.h"

#include <ctype.h>
#include <locale.h>
#include <stdlib.h>
#include <string.h>

const char *google_nl_normalize_locale_name(const char *name) {
  /* If there is nothing to do, we'll return the name we got.  */
  const char *normalized_name = name;

  /* Determine if the name contains a codeset.  */
  const char *p = strchr(name, '.');
  if (p != NULL && p[1] != '@' && p[1] != '\0') {
    const char *rest = strchrnul(++p, '@');
    const char *normalized_codeset = google_nl_normalize_codeset(p, rest - p);
    if (normalized_codeset == NULL) return NULL;

    if (strncmp(normalized_codeset, p, rest - p) != 0 ||
        normalized_codeset[rest - p] != '\0') {
      /* There is a normalized codeset name that is different from
         what was specified; reconstruct a new locale name using it.  */
      size_t normlen = strlen(normalized_codeset);
      size_t restlen = strlen(rest) + 1;

      char *newname = malloc(p - name + normlen + restlen);
      if (newname == NULL) return NULL;

      memcpy(mempcpy(mempcpy(newname, name, p - name), normalized_codeset,
                     normlen),
             rest, restlen);
      normalized_name = newname;
    }
    free((char *)normalized_codeset);
  }

  return normalized_name;
}

const char *google_nl_normalize_codeset(const char *codeset, size_t name_len) {
  int len = 0;
  int only_digit = 1;
  char *retval;
  char *wp;
  size_t cnt;
  locale_t locale = newlocale(0, "C", NULL);

  for (cnt = 0; cnt < name_len; ++cnt)
    if (isalnum_l((unsigned char)codeset[cnt], locale)) {
      ++len;

      if (!isdigit_l((unsigned char)codeset[cnt], locale)) only_digit = 0;
    }

  retval = (char *)malloc((only_digit ? 3 : 0) + len + 1);

  if (retval != NULL) {
    wp = retval;
    if (only_digit) wp = stpcpy(wp, "iso");

    for (cnt = 0; cnt < name_len; ++cnt)
      if (isalpha_l((unsigned char)codeset[cnt], locale))
        *wp++ = tolower((unsigned char)codeset[cnt]);
      else if (isdigit_l((unsigned char)codeset[cnt], locale))
        *wp++ = codeset[cnt];

    *wp = '\0';
  }

  return (const char *)retval;
}

static char *google_nl_find_language(const char *name) {
  while (name[0] != '\0' && name[0] != '_' && name[0] != '@' && name[0] != '.')
    ++name;

  return (char *)name;
}

int google_nl_explode_name(char *name, const char **language,
                           const char **modifier, const char **territory,
                           const char **codeset,
                           const char **normalized_codeset) {
  char *cp;
  int mask;

  *modifier = NULL;
  *territory = NULL;
  *codeset = NULL;
  *normalized_codeset = NULL;

  /* Now we determine the single parts of the locale name.  First
     look for the language.  Termination symbols are `_', '.', and `@'.  */
  mask = 0;
  *language = cp = name;
  cp = google_nl_find_language(*language);

  if (*language == cp)
    /* This does not make sense: language has to be specified.  Use
       this entry as it is without exploding.  Perhaps it is an alias.  */
    return 0;

  if (cp[0] != '@') {
    if (cp[0] == '_') {
      /* Next is the territory.  */
      cp[0] = '\0';
      *territory = ++cp;

      while (cp[0] != '\0' && cp[0] != '.' && cp[0] != '@') ++cp;

      mask |= XPG_TERRITORY;
    }

    if (cp[0] == '.') {
      /* Next is the codeset.  */
      cp[0] = '\0';
      *codeset = ++cp;

      while (cp[0] != '\0' && cp[0] != '@') ++cp;

      mask |= XPG_CODESET;

      if (*codeset != cp && (*codeset)[0] != '\0') {
        *normalized_codeset =
            google_nl_normalize_codeset(*codeset, cp - *codeset);
        if (*normalized_codeset == NULL)
          return -1;
        else if (strcmp(*codeset, *normalized_codeset) == 0)
          free((char *)*normalized_codeset);
        else
          mask |= XPG_NORM_CODESET;
      }
    }
  }

  if (cp[0] == '@') {
    /* Next is the modifier.  */
    cp[0] = '\0';
    *modifier = ++cp;

    if (cp[0] != '\0') mask |= XPG_MODIFIER;
  }

  if (*territory != NULL && (*territory)[0] == '\0') mask &= ~XPG_TERRITORY;

  if (*codeset != NULL && (*codeset)[0] == '\0') mask &= ~XPG_CODESET;

  return mask;
}