cannam@85: /*
cannam@85:  * libid3tag - ID3 tag manipulation library
cannam@85:  * Copyright (C) 2000-2004 Underbit Technologies, Inc.
cannam@85:  *
cannam@85:  * This program is free software; you can redistribute it and/or modify
cannam@85:  * it under the terms of the GNU General Public License as published by
cannam@85:  * the Free Software Foundation; either version 2 of the License, or
cannam@85:  * (at your option) any later version.
cannam@85:  *
cannam@85:  * This program is distributed in the hope that it will be useful,
cannam@85:  * but WITHOUT ANY WARRANTY; without even the implied warranty of
cannam@85:  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
cannam@85:  * GNU General Public License for more details.
cannam@85:  *
cannam@85:  * You should have received a copy of the GNU General Public License
cannam@85:  * along with this program; if not, write to the Free Software
cannam@85:  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
cannam@85:  *
cannam@85:  * $Id: utf8.c,v 1.9 2004/01/23 09:41:32 rob Exp $
cannam@85:  */
cannam@85: 
cannam@85: # ifdef HAVE_CONFIG_H
cannam@85: #  include "config.h"
cannam@85: # endif
cannam@85: 
cannam@85: # include "global.h"
cannam@85: 
cannam@85: # include <stdlib.h>
cannam@85: 
cannam@85: # include "id3tag.h"
cannam@85: # include "utf8.h"
cannam@85: # include "ucs4.h"
cannam@85: 
cannam@85: /*
cannam@85:  * NAME:	utf8->length()
cannam@85:  * DESCRIPTION:	return the number of ucs4 chars represented by a utf8 string
cannam@85:  */
cannam@85: id3_length_t id3_utf8_length(id3_utf8_t const *utf8)
cannam@85: {
cannam@85:   id3_length_t length = 0;
cannam@85: 
cannam@85:   while (*utf8) {
cannam@85:     if ((utf8[0] & 0x80) == 0x00)
cannam@85:       ++length;
cannam@85:     else if ((utf8[0] & 0xe0) == 0xc0 &&
cannam@85: 	     (utf8[1] & 0xc0) == 0x80) {
cannam@85:       if (((utf8[0] & 0x1fL) << 6) >= 0x00000080L) {
cannam@85: 	++length;
cannam@85: 	utf8 += 1;
cannam@85:       }
cannam@85:     }
cannam@85:     else if ((utf8[0] & 0xf0) == 0xe0 &&
cannam@85: 	     (utf8[1] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[2] & 0xc0) == 0x80) {
cannam@85:       if ((((utf8[0] & 0x0fL) << 12) |
cannam@85: 	   ((utf8[1] & 0x3fL) <<  6)) >= 0x00000800L) {
cannam@85: 	++length;
cannam@85: 	utf8 += 2;
cannam@85:       }
cannam@85:     }
cannam@85:     else if ((utf8[0] & 0xf8) == 0xf0 &&
cannam@85: 	     (utf8[1] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[2] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[3] & 0xc0) == 0x80) {
cannam@85:       if ((((utf8[0] & 0x07L) << 18) |
cannam@85: 	   ((utf8[1] & 0x3fL) << 12)) >= 0x00010000L) {
cannam@85: 	++length;
cannam@85: 	utf8 += 3;
cannam@85:       }
cannam@85:     }
cannam@85:     else if ((utf8[0] & 0xfc) == 0xf8 &&
cannam@85: 	     (utf8[1] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[2] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[3] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[4] & 0xc0) == 0x80) {
cannam@85:       if ((((utf8[0] & 0x03L) << 24) |
cannam@85: 	   ((utf8[0] & 0x3fL) << 18)) >= 0x00200000L) {
cannam@85: 	++length;
cannam@85: 	utf8 += 4;
cannam@85:       }
cannam@85:     }
cannam@85:     else if ((utf8[0] & 0xfe) == 0xfc &&
cannam@85: 	     (utf8[1] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[2] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[3] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[4] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[5] & 0xc0) == 0x80) {
cannam@85:       if ((((utf8[0] & 0x01L) << 30) |
cannam@85: 	   ((utf8[0] & 0x3fL) << 24)) >= 0x04000000L) {
cannam@85: 	++length;
cannam@85: 	utf8 += 5;
cannam@85:       }
cannam@85:     }
cannam@85: 
cannam@85:     ++utf8;
cannam@85:   }
cannam@85: 
cannam@85:   return length;
cannam@85: }
cannam@85: 
cannam@85: /*
cannam@85:  * NAME:	utf8->size()
cannam@85:  * DESCRIPTION:	return the encoding size of a utf8 string
cannam@85:  */
cannam@85: id3_length_t id3_utf8_size(id3_utf8_t const *utf8)
cannam@85: {
cannam@85:   id3_utf8_t const *ptr = utf8;
cannam@85: 
cannam@85:   while (*ptr)
cannam@85:     ++ptr;
cannam@85: 
cannam@85:   return ptr - utf8 + 1;
cannam@85: }
cannam@85: 
cannam@85: /*
cannam@85:  * NAME:	utf8->ucs4duplicate()
cannam@85:  * DESCRIPTION:	duplicate and decode a utf8 string into ucs4
cannam@85:  */
cannam@85: id3_ucs4_t *id3_utf8_ucs4duplicate(id3_utf8_t const *utf8)
cannam@85: {
cannam@85:   id3_ucs4_t *ucs4;
cannam@85: 
cannam@85:   ucs4 = malloc((id3_utf8_length(utf8) + 1) * sizeof(*ucs4));
cannam@85:   if (ucs4)
cannam@85:     id3_utf8_decode(utf8, ucs4);
cannam@85: 
cannam@85:   return release(ucs4);
cannam@85: }
cannam@85: 
cannam@85: /*
cannam@85:  * NAME:	utf8->decodechar()
cannam@85:  * DESCRIPTION:	decode a series of utf8 chars into a single ucs4 char
cannam@85:  */
cannam@85: id3_length_t id3_utf8_decodechar(id3_utf8_t const *utf8, id3_ucs4_t *ucs4)
cannam@85: {
cannam@85:   id3_utf8_t const *start = utf8;
cannam@85: 
cannam@85:   while (1) {
cannam@85:     if ((utf8[0] & 0x80) == 0x00) {
cannam@85:       *ucs4 = utf8[0];
cannam@85:       return utf8 - start + 1;
cannam@85:     }
cannam@85:     else if ((utf8[0] & 0xe0) == 0xc0 &&
cannam@85: 	     (utf8[1] & 0xc0) == 0x80) {
cannam@85:       *ucs4 =
cannam@85: 	((utf8[0] & 0x1fL) << 6) |
cannam@85: 	((utf8[1] & 0x3fL) << 0);
cannam@85:       if (*ucs4 >= 0x00000080L)
cannam@85: 	return utf8 - start + 2;
cannam@85:     }
cannam@85:     else if ((utf8[0] & 0xf0) == 0xe0 &&
cannam@85: 	     (utf8[1] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[2] & 0xc0) == 0x80) {
cannam@85:       *ucs4 =
cannam@85: 	((utf8[0] & 0x0fL) << 12) |
cannam@85: 	((utf8[1] & 0x3fL) <<  6) |
cannam@85: 	((utf8[2] & 0x3fL) <<  0);
cannam@85:       if (*ucs4 >= 0x00000800L)
cannam@85: 	return utf8 - start + 3;
cannam@85:     }
cannam@85:     else if ((utf8[0] & 0xf8) == 0xf0 &&
cannam@85: 	     (utf8[1] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[2] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[3] & 0xc0) == 0x80) {
cannam@85:       *ucs4 =
cannam@85: 	((utf8[0] & 0x07L) << 18) |
cannam@85: 	((utf8[1] & 0x3fL) << 12) |
cannam@85: 	((utf8[2] & 0x3fL) <<  6) |
cannam@85: 	((utf8[3] & 0x3fL) <<  0);
cannam@85:       if (*ucs4 >= 0x00010000L)
cannam@85: 	return utf8 - start + 4;
cannam@85:     }
cannam@85:     else if ((utf8[0] & 0xfc) == 0xf8 &&
cannam@85: 	     (utf8[1] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[2] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[3] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[4] & 0xc0) == 0x80) {
cannam@85:       *ucs4 =
cannam@85: 	((utf8[0] & 0x03L) << 24) |
cannam@85: 	((utf8[1] & 0x3fL) << 18) |
cannam@85: 	((utf8[2] & 0x3fL) << 12) |
cannam@85: 	((utf8[3] & 0x3fL) <<  6) |
cannam@85: 	((utf8[4] & 0x3fL) <<  0);
cannam@85:       if (*ucs4 >= 0x00200000L)
cannam@85: 	return utf8 - start + 5;
cannam@85:     }
cannam@85:     else if ((utf8[0] & 0xfe) == 0xfc &&
cannam@85: 	     (utf8[1] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[2] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[3] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[4] & 0xc0) == 0x80 &&
cannam@85: 	     (utf8[5] & 0xc0) == 0x80) {
cannam@85:       *ucs4 =
cannam@85: 	((utf8[0] & 0x01L) << 30) |
cannam@85: 	((utf8[1] & 0x3fL) << 24) |
cannam@85: 	((utf8[2] & 0x3fL) << 18) |
cannam@85: 	((utf8[3] & 0x3fL) << 12) |
cannam@85: 	((utf8[4] & 0x3fL) <<  6) |
cannam@85: 	((utf8[5] & 0x3fL) <<  0);
cannam@85:       if (*ucs4 >= 0x04000000L)
cannam@85: 	return utf8 - start + 6;
cannam@85:     }
cannam@85: 
cannam@85:     ++utf8;
cannam@85:   }
cannam@85: }
cannam@85: 
cannam@85: /*
cannam@85:  * NAME:	utf8->encodechar()
cannam@85:  * DESCRIPTION:	encode a single ucs4 char into a series of up to 6 utf8 chars
cannam@85:  */
cannam@85: id3_length_t id3_utf8_encodechar(id3_utf8_t *utf8, id3_ucs4_t ucs4)
cannam@85: {
cannam@85:   if (ucs4 <= 0x0000007fL) {
cannam@85:     utf8[0] = ucs4;
cannam@85: 
cannam@85:     return 1;
cannam@85:   }
cannam@85:   else if (ucs4 <= 0x000007ffL) {
cannam@85:     utf8[0] = 0xc0 | ((ucs4 >>  6) & 0x1f);
cannam@85:     utf8[1] = 0x80 | ((ucs4 >>  0) & 0x3f);
cannam@85: 
cannam@85:     return 2;
cannam@85:   }
cannam@85:   else if (ucs4 <= 0x0000ffffL) {
cannam@85:     utf8[0] = 0xe0 | ((ucs4 >> 12) & 0x0f);
cannam@85:     utf8[1] = 0x80 | ((ucs4 >>  6) & 0x3f);
cannam@85:     utf8[2] = 0x80 | ((ucs4 >>  0) & 0x3f);
cannam@85: 
cannam@85:     return 3;
cannam@85:   }
cannam@85:   else if (ucs4 <= 0x001fffffL) {
cannam@85:     utf8[0] = 0xf0 | ((ucs4 >> 18) & 0x07);
cannam@85:     utf8[1] = 0x80 | ((ucs4 >> 12) & 0x3f);
cannam@85:     utf8[2] = 0x80 | ((ucs4 >>  6) & 0x3f);
cannam@85:     utf8[3] = 0x80 | ((ucs4 >>  0) & 0x3f);
cannam@85: 
cannam@85:     return 4;
cannam@85:   }
cannam@85:   else if (ucs4 <= 0x03ffffffL) {
cannam@85:     utf8[0] = 0xf8 | ((ucs4 >> 24) & 0x03);
cannam@85:     utf8[1] = 0x80 | ((ucs4 >> 18) & 0x3f);
cannam@85:     utf8[2] = 0x80 | ((ucs4 >> 12) & 0x3f);
cannam@85:     utf8[3] = 0x80 | ((ucs4 >>  6) & 0x3f);
cannam@85:     utf8[4] = 0x80 | ((ucs4 >>  0) & 0x3f);
cannam@85: 
cannam@85:     return 5;
cannam@85:   }
cannam@85:   else if (ucs4 <= 0x7fffffffL) {
cannam@85:     utf8[0] = 0xfc | ((ucs4 >> 30) & 0x01);
cannam@85:     utf8[1] = 0x80 | ((ucs4 >> 24) & 0x3f);
cannam@85:     utf8[2] = 0x80 | ((ucs4 >> 18) & 0x3f);
cannam@85:     utf8[3] = 0x80 | ((ucs4 >> 12) & 0x3f);
cannam@85:     utf8[4] = 0x80 | ((ucs4 >>  6) & 0x3f);
cannam@85:     utf8[5] = 0x80 | ((ucs4 >>  0) & 0x3f);
cannam@85: 
cannam@85:     return 6;
cannam@85:   }
cannam@85: 
cannam@85:   /* default */
cannam@85: 
cannam@85:   return id3_utf8_encodechar(utf8, ID3_UCS4_REPLACEMENTCHAR);
cannam@85: }
cannam@85: 
cannam@85: /*
cannam@85:  * NAME:	utf8->decode()
cannam@85:  * DESCRIPTION:	decode a complete utf8 string into a ucs4 string
cannam@85:  */
cannam@85: void id3_utf8_decode(id3_utf8_t const *utf8, id3_ucs4_t *ucs4)
cannam@85: {
cannam@85:   do
cannam@85:     utf8 += id3_utf8_decodechar(utf8, ucs4);
cannam@85:   while (*ucs4++);
cannam@85: }
cannam@85: 
cannam@85: /*
cannam@85:  * NAME:	utf8->encode()
cannam@85:  * DESCRIPTION:	encode a complete ucs4 string into a utf8 string
cannam@85:  */
cannam@85: void id3_utf8_encode(id3_utf8_t *utf8, id3_ucs4_t const *ucs4)
cannam@85: {
cannam@85:   do
cannam@85:     utf8 += id3_utf8_encodechar(utf8, *ucs4);
cannam@85:   while (*ucs4++);
cannam@85: }
cannam@85: 
cannam@85: /*
cannam@85:  * NAME:	utf8->put()
cannam@85:  * DESCRIPTION:	serialize a single utf8 character
cannam@85:  */
cannam@85: id3_length_t id3_utf8_put(id3_byte_t **ptr, id3_utf8_t utf8)
cannam@85: {
cannam@85:   if (ptr)
cannam@85:     *(*ptr)++ = utf8;
cannam@85: 
cannam@85:   return 1;
cannam@85: }
cannam@85: 
cannam@85: /*
cannam@85:  * NAME:	utf8->get()
cannam@85:  * DESCRIPTION:	deserialize a single utf8 character
cannam@85:  */
cannam@85: id3_utf8_t id3_utf8_get(id3_byte_t const **ptr)
cannam@85: {
cannam@85:   return *(*ptr)++;
cannam@85: }
cannam@85: 
cannam@85: /*
cannam@85:  * NAME:	utf8->serialize()
cannam@85:  * DESCRIPTION:	serialize a ucs4 string using utf8 encoding
cannam@85:  */
cannam@85: id3_length_t id3_utf8_serialize(id3_byte_t **ptr, id3_ucs4_t const *ucs4,
cannam@85: 				int terminate)
cannam@85: {
cannam@85:   id3_length_t size = 0;
cannam@85:   id3_utf8_t utf8[6], *out;
cannam@85: 
cannam@85:   while (*ucs4) {
cannam@85:     switch (id3_utf8_encodechar(out = utf8, *ucs4++)) {
cannam@85:     case 6: size += id3_utf8_put(ptr, *out++);
cannam@85:     case 5: size += id3_utf8_put(ptr, *out++);
cannam@85:     case 4: size += id3_utf8_put(ptr, *out++);
cannam@85:     case 3: size += id3_utf8_put(ptr, *out++);
cannam@85:     case 2: size += id3_utf8_put(ptr, *out++);
cannam@85:     case 1: size += id3_utf8_put(ptr, *out++);
cannam@85:     case 0: break;
cannam@85:     }
cannam@85:   }
cannam@85: 
cannam@85:   if (terminate)
cannam@85:     size += id3_utf8_put(ptr, 0);
cannam@85: 
cannam@85:   return size;
cannam@85: }
cannam@85: 
cannam@85: /*
cannam@85:  * NAME:	utf8->deserialize()
cannam@85:  * DESCRIPTION:	deserialize a ucs4 string using utf8 encoding
cannam@85:  */
cannam@85: id3_ucs4_t *id3_utf8_deserialize(id3_byte_t const **ptr, id3_length_t length)
cannam@85: {
cannam@85:   id3_byte_t const *end;
cannam@85:   id3_utf8_t *utf8ptr, *utf8;
cannam@85:   id3_ucs4_t *ucs4;
cannam@85: 
cannam@85:   end = *ptr + length;
cannam@85: 
cannam@85:   utf8 = malloc((length + 1) * sizeof(*utf8));
cannam@85:   if (utf8 == 0)
cannam@85:     return 0;
cannam@85: 
cannam@85:   utf8ptr = utf8;
cannam@85:   while (end - *ptr > 0 && (*utf8ptr = id3_utf8_get(ptr)))
cannam@85:     ++utf8ptr;
cannam@85: 
cannam@85:   *utf8ptr = 0;
cannam@85: 
cannam@85:   ucs4 = malloc((id3_utf8_length(utf8) + 1) * sizeof(*ucs4));
cannam@85:   if (ucs4)
cannam@85:     id3_utf8_decode(utf8, ucs4);
cannam@85: 
cannam@85:   free(utf8);
cannam@85: 
cannam@85:   return ucs4;
cannam@85: }