gnupg/jnlib/w32-gettext.c

/* w32-gettext.c  - A simplified version of gettext for use under W32.
 * Copyright (C) 1995, 1996, 1997, 1999,
 *               2005, 2007 Free Software Foundation, Inc.
 *
 * This file is part of JNLIB.
 *
 * JNLIB 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 3 of
 * the License, or (at your option) any later version.
 *
 * JNLIB 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 this program; if not, see <http://www.gnu.org/licenses/>.
 */

/* 
   This is a simplified version of gettext written by Ulrich Drepper.
   It is used for the Win32 version of GnuPG becaise all the overhead
   of gettext is not needed and we have to do some special Win32
   stuff.  I decided that this is far easier than to tweak gettext for
   the special cases (I tried it but it is a lot of code). wk 15.09.99
 */

#include <config.h>
#ifdef USE_SIMPLE_GETTEXT
#if !defined (_WIN32) && !defined (__CYGWIN32__)
#error This module may only be build for Windows or Cygwin32
#endif

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#include <errno.h>
#include <sys/types.h>
#include <sys/stat.h>

#include "libjnlib-config.h"
#include "types.h"
#include "stringhelp.h"
#include "utf8conv.h"
#include "w32help.h"

#include "windows.h" /* For GetModuleFileName.  */

/* The magic number of the GNU message catalog format.	*/
#define MAGIC	      0x950412de
#define MAGIC_SWAPPED 0xde120495

/* Revision number of the currently used .mo (binary) file format.  */
#define MO_REVISION_NUMBER 0


/* Header for binary .mo file format.  */
struct mo_file_header
{
  /* The magic number.	*/
  u32 magic;
  /* The revision number of the file format.  */
  u32 revision;
  /* The number of strings pairs.  */
  u32 nstrings;
  /* Offset of table with start offsets of original strings.  */
  u32 orig_tab_offset;
  /* Offset of table with start offsets of translation strings.  */
  u32 trans_tab_offset;
  /* Size of hashing table.  */
  u32 hash_tab_size;
  /* Offset of first hashing entry.  */
  u32 hash_tab_offset;
};

struct string_desc
{
  /* Length of addressed string.  */
  u32 length;
  /* Offset of string in file.	*/
  u32 offset;
};


struct overflow_space_s
{
  struct overflow_space_s *next;
  u32 idx;
  char d[1];
};

struct loaded_domain
{
  char *data;
  int must_swap;
  u32 nstrings;
  char *mapped;  /* 0 = not yet mapped, 1 = mapped,
                    2 = mapped to
                    overflow space */
  struct overflow_space_s *overflow_space;
  struct string_desc *orig_tab;
  struct string_desc *trans_tab;
  u32 hash_size;
  u32 *hash_tab;
};


static struct loaded_domain *the_domain;

static __inline__ u32
do_swap_u32( u32 i )
{
  return (i << 24) | ((i & 0xff00) << 8) | ((i >> 8) & 0xff00) | (i >> 24);
}

#define SWAPIT(flag, data) ((flag) ? do_swap_u32(data) : (data) )


/* We assume to have `unsigned long int' value with at least 32 bits.  */
#define HASHWORDBITS 32

/* The so called `hashpjw' function by P.J. Weinberger
   [see Aho/Sethi/Ullman, COMPILERS: Principles, Techniques and Tools,
   1986, 1987 Bell Telephone Laboratories, Inc.]  */

static __inline__ ulong
hash_string (const char *str_param)
{
  unsigned long int hval, g;
  const char *str = str_param;
  
  hval = 0;
  while (*str != '\0')
    {
      hval <<= 4;
      hval += (unsigned long int) *str++;
      g = hval & ((unsigned long int) 0xf << (HASHWORDBITS - 4));
      if (g != 0)
	{
	  hval ^= g >> (HASHWORDBITS - 8);
	  hval ^= g;
	}
    }
  return hval;
}


static struct loaded_domain *
load_domain (const char *filename)
{
  FILE *fp;
  size_t size;
  struct stat st;
  struct mo_file_header *data = NULL;
  struct loaded_domain *domain = NULL;
  size_t to_read;
  char *read_ptr;
  
  fp = fopen( filename, "rb" );
  if (!fp)
    return NULL; /* Can't open the file.  */
  /* We need to know the size of the file.  */
  if (fstat( fileno(fp ), &st )
      || (size = (size_t)st.st_size) != st.st_size
      || size < sizeof (struct mo_file_header) ) 
    {
      fclose (fp);
      return NULL;
    }

  data = jnlib_malloc (size);
  if (!data)
    {
      fclose (fp);
      return NULL; /* Out of memory. */
    }

  to_read = size;
  read_ptr = (char *) data;
  do
    {
      long int nb;

      nb = fread (read_ptr, 1, to_read, fp);
      if (nb < to_read ) 
        {
          fclose (fp);
          jnlib_free (data);
          return NULL; /* Read error. */
	}
      read_ptr += nb;
      to_read -= nb;
    } 
  while (to_read > 0);
  fclose (fp);

  /* Using the magic number we test whether it is really a message
     catalog file.  */
  if (data->magic != MAGIC && data->magic != MAGIC_SWAPPED)
    {
      /* The magic number is wrong: not a message catalog file.  */
      jnlib_free (data);
      return NULL;
    }

  domain = jnlib_calloc (1, sizeof *domain);
  if (!domain) 
    {
      jnlib_free (data);
      return NULL;
    }
  domain->data = (char *) data;
  domain->must_swap = data->magic != MAGIC;

  /* Fill in the information about the available tables.  */
    switch (SWAPIT(domain->must_swap, data->revision))
      {
      case 0:
	domain->nstrings = SWAPIT(domain->must_swap, data->nstrings);
	domain->orig_tab = (struct string_desc *)
	  ((char *) data + SWAPIT(domain->must_swap, data->orig_tab_offset));
	domain->trans_tab = (struct string_desc *)
	  ((char *) data + SWAPIT(domain->must_swap, data->trans_tab_offset));
	domain->hash_size = SWAPIT(domain->must_swap, data->hash_tab_size);
	domain->hash_tab = (u32 *)
	  ((char *) data + SWAPIT(domain->must_swap, data->hash_tab_offset));
        break;
        
      default: /* This is an invalid revision.	*/
	jnlib_free( data );
	jnlib_free( domain );
	return NULL;
    }

    /* Allocate an array to keep track of code page mappings. */
    domain->mapped = jnlib_calloc (1, domain->nstrings);
    if (!domain->mapped)
      {
        jnlib_free (data);
        jnlib_free (domain);
        return NULL;
      }

    return domain;
}


/* Set the file used for translations.  Pass a NULL to disable
   translation.  A new filename may be set at anytime.  WARNING: After
   changing the filename you should not access any data retrieved by
   gettext().

   If REGKEY is not NULL, the function tries to selected the language
   the registry key "Lang" below that key.  If in addition the
   environment variable LANGUAGE has been set, that value will
   override a value set by the registry key.
 */
int
set_gettext_file ( const char *filename, const char *regkey )
{
  struct loaded_domain *domain = NULL;

  if ( filename && *filename )
    {
      if ( filename[0] == '/'
#ifdef HAVE_DRIVE_LETTERS
           || ( isalpha(filename[0])
                && filename[1] == ':'
                && (filename[2] == '/' || filename[2] == '\\') )
#endif
	   )
        {
          /* absolute path - use it as is */
          domain = load_domain( filename );
	}
      else if (regkey)  /* Standard.  */
        {
          char *instdir, *langid, *fname;
          char *p;
          int envvar_mode = 0;
          
        again:
          if (!envvar_mode && (p = getenv ("LANGUAGE")) && *p)
            {
              envvar_mode = 1;
              langid = jnlib_malloc (strlen (p)+1);
              if (!langid)
                return -1;
              strcpy (langid, p);
              /* We only make use of the first language given.  Strip
                 the rest.  */
              p = strchr (langid, ':');
              if (p)
                *p = 0;
              
              /* In the $LANGUAGE case we do not use the registered
                 installation directory but the one where the gpg
                 binary has been found.  */
              instdir = jnlib_malloc (MAX_PATH+5);
              if ( !instdir || !GetModuleFileName (NULL, instdir, MAX_PATH) )
                {
                  jnlib_free (langid);
                  jnlib_free (instdir);
                  return -1; /* Error getting the process' file name.  */
                }
              p = strrchr (instdir, DIRSEP_C);
              if (!p)
                {
                  jnlib_free (langid);
                  jnlib_free (instdir);
                  return -1; /* Invalid file name returned.  */
                }
              *p = 0;
            }
          else
            {
              instdir = read_w32_registry_string ("HKEY_LOCAL_MACHINE",
                                                  regkey,
                                                  "Install Directory");
              if (!instdir)
                return -1;
              langid = read_w32_registry_string (NULL, /* HKCU then HKLM */
                                                 regkey,
                                                 "Lang");
              if (!langid)
                {
                  jnlib_free (instdir);
                  return -1;
                }
            }
          
          /* Strip stuff after a dot in case the user tried to enter
             the entire locale syntacs as usual for POSIX.  */
          p = strchr (langid, '.');
          if (p)
            *p = 0;
          
          /* Build the key: "<instdir>/<domain>.nls/<langid>.mo" We
             use a directory below the installation directory with the
             domain included in case the software has been insalled
             with other software altogether at the same place.  */
          fname = jnlib_malloc (strlen (instdir) + 1 + strlen (filename) + 5
                                + strlen (langid) + 3 + 1);
          if (!fname)
            {
              jnlib_free (instdir);
              jnlib_free (langid);
              return -1;
            }
          strcpy (stpcpy (stpcpy (stpcpy (stpcpy ( stpcpy (fname,
                  instdir),"\\"), filename), ".nls\\"), langid), ".mo");
          jnlib_free (instdir);
          jnlib_free (langid);

          /* Better make sure that we don't mix forward and backward
             slashes.  It seems that some Windoze versions don't
             accept this. */
          for (p=fname; *p; p++) 
            {
              if (*p == '/')
                *p = '\\';
            }
          domain = load_domain (fname);
          jnlib_free(fname);

          if (!domain && envvar_mode == 1)
            {
              /* In case it failed, we try again using the registry
                 method. */
              envvar_mode++;
              goto again;
            }
	}
      

      if (!domain)
        return -1;
    }

  if ( the_domain )
    {
      struct overflow_space_s *os, *os2;

      jnlib_free ( the_domain->data );
      jnlib_free ( the_domain->mapped );
      for (os=the_domain->overflow_space; os; os = os2)
        {
          os2 = os->next;
          jnlib_free (os);
        }
      jnlib_free ( the_domain );
      the_domain = NULL;
    }
  the_domain = domain;
  return 0;
}


static const char*
get_string( struct loaded_domain *domain, u32 idx )
{
  struct overflow_space_s *os;
  char *p;

  p = domain->data + SWAPIT(domain->must_swap, domain->trans_tab[idx].offset);
  if (!domain->mapped[idx]) 
    {
      size_t plen, buflen;
      char *buf;

      domain->mapped[idx] = 1;

      plen = strlen (p);
      buf = utf8_to_native (p, plen, -1);
      buflen = strlen (buf);
      if (buflen <= plen)
        strcpy (p, buf);
      else
        {
          /* There is not enough space for the translation - store it
             in the overflow_space else and mark that in the mapped
             array.  Because we expect that this won't happen too
             often, we use a simple linked list.  */
          os = jnlib_malloc (sizeof *os + buflen);
          if (os)
            {
              os->idx = idx;
              strcpy (os->d, buf);
              os->next = domain->overflow_space;
              domain->overflow_space = os;
              p = os->d;
            }
          else
            p = "ERROR in GETTEXT MALLOC";
        }
      jnlib_free (buf);
    }
  else if (domain->mapped[idx] == 2) 
    { /* We need to get the string from the overflow_space. */
      for (os=domain->overflow_space; os; os = os->next)
        if (os->idx == idx)
          return (const char*)os->d;
      p = "ERROR in GETTEXT\n";
    }
  return (const char*)p;
}



const char *
gettext( const char *msgid )
{
  struct loaded_domain *domain;
  size_t act = 0;
  size_t top, bottom;
  
  if (!(domain = the_domain))
    goto not_found;
  
  /* Locate the MSGID and its translation.  */
  if (domain->hash_size > 2 && domain->hash_tab)
    {
      /* Use the hashing table.  */
      u32 len = strlen (msgid);
      u32 hash_val = hash_string (msgid);
      u32 idx = hash_val % domain->hash_size;
      u32 incr = 1 + (hash_val % (domain->hash_size - 2));
      u32 nstr = SWAPIT (domain->must_swap, domain->hash_tab[idx]);

      if ( !nstr ) /* Hash table entry is empty.  */
        goto not_found;
      
      if (SWAPIT(domain->must_swap,
                 domain->orig_tab[nstr - 1].length) == len
          && !strcmp (msgid,
                      domain->data + SWAPIT(domain->must_swap,
                                            domain->orig_tab[nstr-1].offset)))
        return get_string( domain, nstr - 1 );

      for (;;) 
        {
          if (idx >= domain->hash_size - incr)
            idx -= domain->hash_size - incr;
          else
            idx += incr;
          
          nstr = SWAPIT (domain->must_swap, domain->hash_tab[idx]);
          if (!nstr)
            goto not_found; /* Hash table entry is empty.  */

          if ( SWAPIT(domain->must_swap,
                      domain->orig_tab[nstr - 1].length) == len
               && !strcmp (msgid,
                           domain->data 
                           + SWAPIT(domain->must_swap,
                                    domain->orig_tab[nstr-1].offset)))
            return get_string( domain, nstr-1 );
	}
      /*NOTREACHED*/
    }

  /* Now we try the default method: binary search in the sorted array
     of messages.  */
  bottom = 0;
  top = domain->nstrings;
  while (bottom < top)
    {
      int cmp_val;
      
      act = (bottom + top) / 2;
      cmp_val = strcmp(msgid, domain->data
                       + SWAPIT(domain->must_swap,
                                domain->orig_tab[act].offset));
      if (cmp_val < 0)
        top = act;
      else if (cmp_val > 0)
        bottom = act + 1;
      else
        return get_string (domain, act);
    }
  
 not_found:
  return msgid;
}


const char *
ngettext (const char *msgid1, const char *msgid2, unsigned long int n)
{
  /* We use the simple Germanic plural rule. */
  return gettext (n==1? msgid1 : msgid2);
}



#if 0
       unsigned int cp1, cp2;

       cp1 = GetConsoleCP();
       cp2 = GetConsoleOutputCP();

       log_info("InputCP=%u  OutputCP=%u\n", cp1, cp2 );

       if( !SetConsoleOutputCP( 1252 ) )
            log_info("SetConsoleOutputCP failed: %s\n", w32_strerror (0));

       cp1 = GetConsoleCP();
       cp2 = GetConsoleOutputCP();
       log_info("InputCP=%u  OutputCP=%u after switch1\n", cp1, cp2 );
#endif

#endif /* USE_SIMPLE_GETTEXT */