lib/striconveha.h

   1 /* Character set conversion with error handling and autodetection.
   2    Copyright (C) 2002, 2005, 2007-2024 Free Software Foundation, Inc.
   3    Written by Bruno Haible.
   4
   5    This file is free software: you can redistribute it and/or modify
   6    it under the terms of the GNU Lesser General Public License as
   7    published by the Free Software Foundation; either version 2.1 of the
   8    License, or (at your option) any later version.
   9
  10    This file is distributed in the hope that it will be useful,
  11    but WITHOUT ANY WARRANTY; without even the implied warranty of
  12    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  13    GNU Lesser General Public License for more details.
  14
  15    You should have received a copy of the GNU Lesser General Public License
  16    along with this program.  If not, see <https://www.gnu.org/licenses/>.  */
  17
  18 #ifndef _STRICONVEHA_H
  19 #define _STRICONVEHA_H
  20
  21 /* This file uses _GL_ATTRIBUTE_MALLOC.  */
  22 #if !_GL_CONFIG_H_INCLUDED
  23  #error "Please include config.h first."
  24 #endif
  25
  26 #include <stdlib.h>
  27
  28 #include "iconveh.h"
  29
  30
  31 #ifdef __cplusplus
  32 extern "C" {
  33 #endif
  34
  35
  36 /* Convert an entire string from one encoding to another, using iconv.
  37    The original string is at [SRC,...,SRC+SRCLEN-1].
  38    The "from" encoding can also be a name defined for autodetection.
  39    If TRANSLITERATE is true, transliteration will attempted to avoid conversion
  40    errors, for iconv implementations that support this.  Usually you'll choose
  41    TRANSLITERATE = true if HANDLER != iconveh_error.
  42    If OFFSETS is not NULL, it should point to an array of SRCLEN integers; this
  43    array is filled with offsets into the result, i.e. the character starting
  44    at SRC[i] corresponds to the character starting at (*RESULTP)[OFFSETS[i]],
  45    and other offsets are set to (size_t)(-1).
  46    *RESULTP and *LENGTH should initially be a scratch buffer and its size,
  47    or *RESULTP can initially be NULL.
  48    May erase the contents of the memory at *RESULTP.
  49    Return value: 0 if successful, otherwise -1 and errno set.
  50    If successful: The resulting string is stored in *RESULTP and its length
  51    in *LENGTHP.  *RESULTP is set to a freshly allocated memory block, or is
  52    unchanged if no dynamic memory allocation was necessary.  */
  53 extern int
  54        mem_iconveha (const char *src, size_t srclen,
  55                      const char *from_codeset, const char *to_codeset,
  56                      bool transliterate,
  57                      enum iconv_ilseq_handler handler,
  58                      size_t *offsets,
  59                      char **resultp, size_t *lengthp);
  60
  61 /* Convert an entire string from one encoding to another, using iconv.
  62    The original string is the NUL-terminated string starting at SRC.
  63    Both the "from" and the "to" encoding must use a single NUL byte at the
  64    end of the string (i.e. not UCS-2, UCS-4, UTF-16, UTF-32).
  65    The "from" encoding can also be a name defined for autodetection.
  66    If TRANSLITERATE is true, transliteration will attempted to avoid conversion
  67    errors, for iconv implementations that support this.  Usually you'll choose
  68    TRANSLITERATE = true if HANDLER != iconveh_error.
  69    Allocate a malloced memory block for the result.
  70    Return value: the freshly allocated resulting NUL-terminated string if
  71    successful, otherwise NULL and errno set.  */
  72 extern char *
  73        str_iconveha (const char *src,
  74                      const char *from_codeset, const char *to_codeset,
  75                      bool transliterate,
  76                      enum iconv_ilseq_handler handler)
  77        _GL_ATTRIBUTE_MALLOC _GL_ATTRIBUTE_DEALLOC_FREE;
  78
  79
  80 /* In the above, FROM_CODESET can also be one of the following values:
  81       "autodetect_utf8"         supports ISO-8859-1 and UTF-8
  82       "autodetect_jp"           supports EUC-JP, ISO-2022-JP-2 and SHIFT_JIS
  83       "autodetect_kr"           supports EUC-KR and ISO-2022-KR
  84    More names can be defined for autodetection.  */
  85
  86 /* Registers an encoding name for autodetection.
  87    TRY_IN_ORDER is a NULL terminated list of encodings to be tried.
  88    Returns 0 upon success, or -1 (with errno set) in case of error.
  89    Particular errno values: ENOMEM.  */
  90 extern int
  91        uniconv_register_autodetect (const char *name,
  92                                     const char * const *try_in_order);
  93
  94
  95 #ifdef __cplusplus
  96 }
  97 #endif
  98
  99
 100 #endif /* _STRICONVEHA_H */