X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=util%2Fcharmap.c;h=96a390ab14b8f230a87e33d0ee8a73fc4f2326aa;hb=1cc0116831e49bf70538639a8fc9b08b9524c683;hp=a80bbffdf20940b66a01ca7ca2bc4d3d4685347a;hpb=b6d668f5407b24e0c9a05f5a3ee4fa4c93377c5c;p=idzebra-moved-to-github.git diff --git a/util/charmap.c b/util/charmap.c index a80bbff..96a390a 100644 --- a/util/charmap.c +++ b/util/charmap.c @@ -1,11 +1,26 @@ -/* - * Copyright (C) 1996-2002, Index Data - * All rights reserved. - * Sebastian Hammer, Adam Dickmeiss - * - * $Id: charmap.c,v 1.24 2002-07-26 14:43:09 adam Exp $ - * - */ +/* $Id: charmap.c,v 1.29 2004-07-28 09:47:42 adam Exp $ + Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002,2003,2004 + Index Data Aps + +This file is part of the Zebra server. + +Zebra is free software; you can redistribute it and/or modify it under +the terms of the GNU General Public License as published by the Free +Software Foundation; either version 2, or (at your option) any later +version. + +Zebra 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 General Public License +for more details. + +You should have received a copy of the GNU General Public License +along with Zebra; see the file LICENSE.zebra. If not, write to the +Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA +02111-1307, USA. +*/ + + /* * Support module to handle character-conversions into and out of the @@ -16,17 +31,6 @@ #include #include -#if HAVE_ICONV_H -#include -#else -typedef int iconv_t; -static size_t iconv(iconv_t t, char **buf, size_t *inbytesleft, - char **outbuf, size_t *outbytesleft) -{ - return -1; -} -#endif - typedef unsigned ucs4_t; #include @@ -38,7 +42,8 @@ typedef unsigned ucs4_t; const char *CHR_UNKNOWN = "\001"; const char *CHR_SPACE = "\002"; -const char *CHR_BASE = "\003"; +const char *CHR_CUT = "\003"; +const char *CHR_BASE = "\005"; struct chrmaptab_info { @@ -240,6 +245,14 @@ unsigned char zebra_prim(char **s) return c; } +static int zebra_ucs4_strlen(ucs4_t *s) +{ + int i = 0; + while (*s++) + i++; + return i; +} + ucs4_t zebra_prim_w(ucs4_t **s) { ucs4_t c; @@ -259,13 +272,16 @@ ucs4_t zebra_prim_w(ucs4_t **s) case 't': c = '\t'; (*s)++; break; case 's': c = ' '; (*s)++; break; case 'x': - fmtstr[0] = (*s)[0]; - fmtstr[1] = (*s)[1]; - fmtstr[2] = (*s)[2]; - fmtstr[3] = 0; - sscanf(fmtstr, "x%2x", &i); - c = i; - *s += 3; break; + if (zebra_ucs4_strlen(*s) >= 3) + { + fmtstr[0] = (*s)[1]; + fmtstr[1] = (*s)[2]; + fmtstr[2] = 0; + sscanf(fmtstr, "%x", &i); + c = i; + *s += 3; + } + break; case '0': case '1': case '2': @@ -276,14 +292,30 @@ ucs4_t zebra_prim_w(ucs4_t **s) case '7': case '8': case '9': - fmtstr[0] = (*s)[0]; - fmtstr[1] = (*s)[1]; - fmtstr[2] = (*s)[2]; - fmtstr[3] = 0; - sscanf(fmtstr, "%3o", &i); - c = i; - *s += 3; + if (zebra_ucs4_strlen(*s) >= 3) + { + fmtstr[0] = (*s)[0]; + fmtstr[1] = (*s)[1]; + fmtstr[2] = (*s)[2]; + fmtstr[3] = 0; + sscanf(fmtstr, "%o", &i); + c = i; + *s += 3; + } break; + case 'L': + if (zebra_ucs4_strlen(*s) >= 5) + { + fmtstr[0] = (*s)[1]; + fmtstr[1] = (*s)[2]; + fmtstr[2] = (*s)[3]; + fmtstr[3] = (*s)[4]; + fmtstr[4] = 0; + sscanf(fmtstr, "%x", &i); + c = i; + *s += 5; + } + break; default: (*s)++; } @@ -323,6 +355,17 @@ static void fun_addspace(const char *s, void *data, int num) (char*) CHR_SPACE, 0); } +/* + * Callback function. + * Add a space-entry to the value space. + */ +static void fun_addcut(const char *s, void *data, int num) +{ + chrmaptab tab = (chrmaptab) data; + tab->input = set_map_string(tab->input, tab->nmem, s, strlen(s), + (char*) CHR_CUT, 0); +} + /* * Create a string containing the mapped characters provided. */ @@ -368,20 +411,22 @@ static void fun_add_qmap(const char *s, void *data, int num) logf (LOG_DEBUG, " %3d", (unsigned char) *s); } -static int scan_to_utf8 (iconv_t t, ucs4_t *from, size_t inlen, +static int scan_to_utf8 (yaz_iconv_t t, ucs4_t *from, size_t inlen, char *outbuf, size_t outbytesleft) { size_t inbytesleft = inlen * sizeof(ucs4_t); char *inbuf = (char*) from; size_t ret; - if (t == (iconv_t)(-1)) + if (t == 0) *outbuf++ = *from; /* ISO-8859-1 is OK here */ else { - ret = iconv (t, &inbuf, &inbytesleft, &outbuf, &outbytesleft); + ret = yaz_iconv (t, &inbuf, &inbytesleft, &outbuf, &outbytesleft); if (ret == (size_t) (-1)) { + yaz_log(LOG_LOG, "from: %2X %2X %2X %2X", + from[0], from[1], from[2], from[3]); yaz_log (LOG_WARN|LOG_ERRNO, "bad unicode sequence"); return -1; } @@ -391,7 +436,7 @@ static int scan_to_utf8 (iconv_t t, ucs4_t *from, size_t inlen, } static int scan_string(char *s_native, - iconv_t t_unicode, iconv_t t_utf8, + yaz_iconv_t t_unicode, yaz_iconv_t t_utf8, void (*fun)(const char *c, void *data, int num), void *data, int *num) { @@ -400,16 +445,17 @@ static int scan_string(char *s_native, ucs4_t arg[512]; ucs4_t *s0, *s = arg; ucs4_t c, begin, end; - size_t i, j; + size_t i; - if (t_unicode != (iconv_t)(-1)) + if (t_unicode != 0) { char *outbuf = (char *) arg; char *inbuf = s_native; size_t outbytesleft = sizeof(arg)-4; size_t inbytesleft = strlen(s_native); - size_t ret; - ret = iconv(t_unicode, &inbuf, &inbytesleft, &outbuf, &outbytesleft); + size_t ret; + ret = yaz_iconv(t_unicode, &inbuf, &inbytesleft, + &outbuf, &outbytesleft); if (ret == (size_t)(-1)) return -1; i = (outbuf - (char*) arg)/sizeof(ucs4_t); @@ -481,17 +527,16 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, int errors = 0; int argc, num = (int) *CHR_BASE, i; NMEM nmem; - iconv_t t_unicode = (iconv_t)(-1); - iconv_t t_utf8 = (iconv_t)(-1); + yaz_iconv_t t_unicode = 0; + yaz_iconv_t t_utf8 = 0; unsigned endian = 31; const char *ucs4_native = "UCS-4"; if (*(char*) &endian == 31) /* little endian? */ ucs4_native = "UCS-4LE"; -#if HAVE_ICONV_H - t_utf8 = iconv_open ("UTF-8", ucs4_native); -#endif + t_utf8 = yaz_iconv_open ("UTF-8", ucs4_native); + logf (LOG_DEBUG, "maptab %s open", name); if (!(f = yaz_fopen(tabpath, name, "r", tabroot))) { @@ -579,7 +624,7 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, { if (argc != 2) { - logf(LOG_FATAL, "Syntax error in charmap"); + logf(LOG_FATAL, "Syntax error in charmap for space"); ++errors; } if (scan_string(argv[1], t_unicode, t_utf8, @@ -589,6 +634,20 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, ++errors; } } + else if (!map_only && !yaz_matchstr(argv[0], "cut")) + { + if (argc != 2) + { + logf(LOG_FATAL, "Syntax error in charmap for cut"); + ++errors; + } + if (scan_string(argv[1], t_unicode, t_utf8, + fun_addcut, res, 0) < 0) + { + logf(LOG_FATAL, "Bad cut specification"); + ++errors; + } + } else if (!yaz_matchstr(argv[0], "map")) { chrwork buf; @@ -639,12 +698,27 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, } else if (!yaz_matchstr(argv[0], "encoding")) { -#if HAVE_ICONV_H - if (t_unicode != (iconv_t)(-1)) - iconv_close (t_unicode); - t_unicode = iconv_open (ucs4_native, argv[1]); -#else - logf (LOG_WARN, "Encoding ignored. iconv not installed"); + /* + * Fix me. When t_unicode==0 and use encoding directive in *.chr file the beheviour of the + * zebra need to comment next part of code. + */ + + /* Original code */ +#if 1 + if (t_unicode != 0) + yaz_iconv_close (t_unicode); + t_unicode = yaz_iconv_open (ucs4_native, argv[1]); +#endif + /* + * Fix me. It is additional staff for conversion of characters from local encoding + * of *.chr file to UTF-8 (internal encoding). + * NOTE: The derective encoding must be first directive in *.chr file. + */ + /* For whatever reason Oleg enabled this.. */ +#if 0 + if (t_utf8 != 0) + yaz_iconv_close(t_utf8); + t_utf8 = yaz_iconv_open ("UTF-8", argv[1]); #endif } else @@ -659,12 +733,10 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, res = 0; } logf (LOG_DEBUG, "maptab %s close %d errors", name, errors); -#if HAVE_ICONV_H - if (t_utf8 != (iconv_t)(-1)) - iconv_close(t_utf8); - if (t_unicode != (iconv_t)(-1)) - iconv_close(t_unicode); -#endif + if (t_utf8 != 0) + yaz_iconv_close(t_utf8); + if (t_unicode != 0) + yaz_iconv_close(t_unicode); return res; }