X-Git-Url: http://git.indexdata.com/?p=idzebra-moved-to-github.git;a=blobdiff_plain;f=util%2Fcharmap.c;h=f39f7069a6afbdf2d64775c52d995b78e73ba4bc;hp=a80bbffdf20940b66a01ca7ca2bc4d3d4685347a;hb=6c9fcd3b5d3108702fa1ffc92dab4ab6060f9a19;hpb=b6d668f5407b24e0c9a05f5a3ee4fa4c93377c5c diff --git a/util/charmap.c b/util/charmap.c index a80bbff..f39f706 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 1.34 2005-01-15 19:38:41 adam Exp $ + Copyright (C) 1995-2005 + 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 @@ -16,29 +31,21 @@ #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 #include +#include #define CHR_MAXSTR 1024 #define CHR_MAXEQUIV 32 +const unsigned char CHR_FIELD_BEGIN = '^'; + 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 { @@ -90,7 +97,7 @@ static chr_t_entry *set_map_string(chr_t_entry *root, NMEM nmem, root->target && root->target[0] && root->target[0][0] && strcmp (root->target[0], CHR_UNKNOWN)) { - yaz_log (LOG_WARN, "duplicate entry for charmap from '%s'", + yaz_log (YLOG_WARN, "duplicate entry for charmap from '%s'", from_0); } root->target = (unsigned char **) @@ -137,7 +144,7 @@ static chr_t_entry *find_entry(chr_t_entry *t, const char **from, int len) return t->target ? t : 0; } -static chr_t_entry *find_entry_x(chr_t_entry *t, const char **from, int *len) +static chr_t_entry *find_entry_x(chr_t_entry *t, const char **from, int *len, int first) { chr_t_entry *res; @@ -148,35 +155,49 @@ static chr_t_entry *find_entry_x(chr_t_entry *t, const char **from, int *len) from++; len++; } - if (*len > 0 && t->children && t->children[(unsigned char) **from]) + if (*len > 0 && t->children) { const char *old_from = *from; int old_len = *len; + + res = 0; + + if (first && t->children[CHR_FIELD_BEGIN]) + { + if ((res = find_entry_x(t->children[CHR_FIELD_BEGIN], from, len, 0)) && res != t->children[CHR_FIELD_BEGIN]) + return res; + else + res = 0; + /* otherwhise there was no match on beginning of field, move on */ + } - (*len)--; - (*from)++; - if ((res = find_entry_x(t->children[(unsigned char) *old_from], - from, len))) - return res; - /* no match */ - *len = old_len; - *from = old_from; + if (!res && t->children[(unsigned char) **from]) + { + (*len)--; + (*from)++; + if ((res = find_entry_x(t->children[(unsigned char) *old_from], + from, len, 0))) + return res; + /* no match */ + *len = old_len; + *from = old_from; + } } /* no children match. use ourselves, if we have a target */ return t->target ? t : 0; } -const char **chr_map_input_x(chrmaptab maptab, const char **from, int *len) +const char **chr_map_input_x(chrmaptab maptab, const char **from, int *len, int first) { chr_t_entry *t = maptab->input; chr_t_entry *res; - if (!(res = find_entry_x(t, from, len))) + if (!(res = find_entry_x(t, from, len, first))) abort(); return (const char **) (res->target); } -const char **chr_map_input(chrmaptab maptab, const char **from, int len) +const char **chr_map_input(chrmaptab maptab, const char **from, int len, int first) { chr_t_entry *t = maptab->input; chr_t_entry *res; @@ -184,7 +205,7 @@ const char **chr_map_input(chrmaptab maptab, const char **from, int len) len_tmp[0] = len; len_tmp[1] = -1; - if (!(res = find_entry_x(t, from, len_tmp))) + if (!(res = find_entry_x(t, from, len_tmp, first))) abort(); return (const char **) (res->target); } @@ -201,7 +222,7 @@ unsigned char zebra_prim(char **s) unsigned char c; unsigned int i = 0; - yaz_log (LOG_DEBUG, "prim %.3s", *s); + yaz_log (YLOG_DEBUG, "prim %.3s", *s); if (**s == '\\') { (*s)++; @@ -240,13 +261,21 @@ 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; ucs4_t i = 0; char fmtstr[8]; - yaz_log (LOG_DEBUG, "prim %.3s", (char *) *s); + yaz_log (YLOG_DEBUG, "prim_w %.3s", (char *) *s); if (**s == '\\') { (*s)++; @@ -259,13 +288,16 @@ 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 +308,30 @@ 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)++; } @@ -293,7 +341,7 @@ ucs4_t zebra_prim_w(ucs4_t **s) c = **s; ++(*s); } - yaz_log (LOG_DEBUG, "out %d", c); + yaz_log (YLOG_DEBUG, "out %d", c); return c; } @@ -323,6 +371,17 @@ 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. */ @@ -331,9 +390,9 @@ static void fun_mkstring(const char *s, void *data, int num) chrwork *arg = (chrwork *) data; const char **res, *p = s; - res = chr_map_input(arg->map, &s, strlen(s)); + res = chr_map_input(arg->map, &s, strlen(s), 0); if (*res == (char*) CHR_UNKNOWN) - logf(LOG_WARN, "Map: '%s' has no mapping", p); + yaz_log(YLOG_WARN, "Map: '%s' has no mapping", p); strncat(arg->string, *res, CHR_MAXSTR - strlen(arg->string)); arg->string[CHR_MAXSTR] = '\0'; } @@ -346,11 +405,11 @@ static void fun_add_map(const char *s, void *data, int num) chrwork *arg = (chrwork *) data; assert(arg->map->input); - logf (LOG_DEBUG, "set map %.*s", (int) strlen(s), s); + yaz_log (YLOG_DEBUG, "set map %.*s", (int) strlen(s), s); set_map_string(arg->map->input, arg->map->nmem, s, strlen(s), arg->string, 0); for (s = arg->string; *s; s++) - logf (LOG_DEBUG, " %3d", (unsigned char) *s); + yaz_log (YLOG_DEBUG, " %3d", (unsigned char) *s); } /* @@ -361,28 +420,30 @@ static void fun_add_qmap(const char *s, void *data, int num) chrwork *arg = (chrwork *) data; assert(arg->map->q_input); - logf (LOG_DEBUG, "set qmap %.*s", (int) strlen(s), s); + yaz_log (YLOG_DEBUG, "set qmap %.*s", (int) strlen(s), s); set_map_string(arg->map->q_input, arg->map->nmem, s, strlen(s), arg->string, 0); for (s = arg->string; *s; s++) - logf (LOG_DEBUG, " %3d", (unsigned char) *s); + yaz_log (YLOG_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_WARN|LOG_ERRNO, "bad unicode sequence"); + yaz_log(YLOG_LOG, "from: %2X %2X %2X %2X", + from[0], from[1], from[2], from[3]); + yaz_log (YLOG_WARN|YLOG_ERRNO, "bad unicode sequence"); return -1; } } @@ -391,25 +452,27 @@ 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) { char str[1024]; ucs4_t arg[512]; + ucs4_t arg_prim[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); @@ -431,14 +494,14 @@ static int scan_string(char *s_native, begin = zebra_prim_w(&s); if (*s != '-') { - logf(LOG_FATAL, "Bad range in char-map"); + yaz_log(YLOG_FATAL, "Bad range in char-map"); return -1; } s++; end = zebra_prim_w(&s); if (end <= begin) { - logf(LOG_FATAL, "Bad range in char-map"); + yaz_log(YLOG_FATAL, "Bad range in char-map"); return -1; } s++; @@ -452,11 +515,11 @@ static int scan_string(char *s_native, case '[': s++; abort(); break; case '(': ++s; - s0 = s; - while (*s != ')' || s[-1] == '\\') - s++; - *s = 0; - if (scan_to_utf8 (t_utf8, s0, s - s0, str, sizeof(str)-1)) + s0 = s; i = 0; + while (*s != ')' || s[-1] == '\\') + arg_prim[i++] = zebra_prim_w(&s); + arg_prim[i] = 0; + if (scan_to_utf8 (t_utf8, arg_prim, zebra_ucs4_strlen(arg_prim), str, sizeof(str)-1)) return -1; (*fun)(str, data, num ? (*num)++ : 0); s++; @@ -481,21 +544,20 @@ 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 - logf (LOG_DEBUG, "maptab %s open", name); + t_utf8 = yaz_iconv_open ("UTF-8", ucs4_native); + + yaz_log (YLOG_DEBUG, "maptab %s open", name); if (!(f = yaz_fopen(tabpath, name, "r", tabroot))) { - logf(LOG_WARN|LOG_ERRNO, "%s", name); + yaz_log(YLOG_WARN|YLOG_ERRNO, "%s", name); return 0; } nmem = nmem_create (); @@ -542,13 +604,13 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, { if (argc != 2) { - logf(LOG_FATAL, "Syntax error in charmap"); + yaz_log(YLOG_FATAL, "Syntax error in charmap"); ++errors; } if (scan_string(argv[1], t_unicode, t_utf8, fun_addentry, res, &num) < 0) { - logf(LOG_FATAL, "Bad value-set specification"); + yaz_log(YLOG_FATAL, "Bad value-set specification"); ++errors; } res->base_uppercase = num; @@ -560,18 +622,18 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, { if (!res->base_uppercase) { - logf(LOG_FATAL, "Uppercase directive with no lowercase set"); + yaz_log(YLOG_FATAL, "Uppercase directive with no lowercase set"); ++errors; } if (argc != 2) { - logf(LOG_FATAL, "Missing arg for uppercase directive"); + yaz_log(YLOG_FATAL, "Missing arg for uppercase directive"); ++errors; } if (scan_string(argv[1], t_unicode, t_utf8, fun_addentry, res, &num) < 0) { - logf(LOG_FATAL, "Bad value-set specification"); + yaz_log(YLOG_FATAL, "Bad value-set specification"); ++errors; } } @@ -579,13 +641,27 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, { if (argc != 2) { - logf(LOG_FATAL, "Syntax error in charmap"); + yaz_log(YLOG_FATAL, "Syntax error in charmap for space"); ++errors; } if (scan_string(argv[1], t_unicode, t_utf8, fun_addspace, res, 0) < 0) { - logf(LOG_FATAL, "Bad space specification"); + yaz_log(YLOG_FATAL, "Bad space specification"); + ++errors; + } + } + else if (!map_only && !yaz_matchstr(argv[0], "cut")) + { + if (argc != 2) + { + yaz_log(YLOG_FATAL, "Syntax error in charmap for cut"); + ++errors; + } + if (scan_string(argv[1], t_unicode, t_utf8, + fun_addcut, res, 0) < 0) + { + yaz_log(YLOG_FATAL, "Bad cut specification"); ++errors; } } @@ -595,7 +671,7 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, if (argc != 3) { - logf(LOG_FATAL, "charmap directive map requires 2 args"); + yaz_log(YLOG_FATAL, "charmap directive map requires 2 args"); ++errors; } buf.map = res; @@ -603,13 +679,13 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, if (scan_string(argv[2], t_unicode, t_utf8, fun_mkstring, &buf, 0) < 0) { - logf(LOG_FATAL, "Bad map target"); + yaz_log(YLOG_FATAL, "Bad map target"); ++errors; } if (scan_string(argv[1], t_unicode, t_utf8, fun_add_map, &buf, 0) < 0) { - logf(LOG_FATAL, "Bad map source"); + yaz_log(YLOG_FATAL, "Bad map source"); ++errors; } } @@ -619,7 +695,7 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, if (argc != 3) { - logf(LOG_FATAL, "charmap directive qmap requires 2 args"); + yaz_log(YLOG_FATAL, "charmap directive qmap requires 2 args"); ++errors; } buf.map = res; @@ -627,29 +703,44 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, if (scan_string(argv[2], t_unicode, t_utf8, fun_mkstring, &buf, 0) < 0) { - logf(LOG_FATAL, "Bad qmap target"); + yaz_log(YLOG_FATAL, "Bad qmap target"); ++errors; } if (scan_string(argv[1], t_unicode, t_utf8, fun_add_qmap, &buf, 0) < 0) { - logf(LOG_FATAL, "Bad qmap source"); + yaz_log(YLOG_FATAL, "Bad qmap source"); ++errors; } } 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 { - logf(LOG_WARN, "Syntax error at '%s' in %s", line, name); + yaz_log(YLOG_WARN, "Syntax error at '%s' in %s", line, name); } yaz_fclose(f); @@ -658,13 +749,11 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, chrmaptab_destroy(res); 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 + yaz_log (YLOG_DEBUG, "maptab %s close %d errors", name, errors); + if (t_utf8 != 0) + yaz_iconv_close(t_utf8); + if (t_unicode != 0) + yaz_iconv_close(t_unicode); return res; }