X-Git-Url: http://git.indexdata.com/?p=idzebra-moved-to-github.git;a=blobdiff_plain;f=util%2Fcharmap.c;h=7e4af55e64f6fd9e94ddfa197649d827a20e5a2a;hp=2919837f306ac081366745553dd2384f690a433c;hb=HEAD;hpb=f9e7e7c0cbaab295c911c44002eb7009dee03b7b diff --git a/util/charmap.c b/util/charmap.c index 2919837..7e4af55 100644 --- a/util/charmap.c +++ b/util/charmap.c @@ -1,8 +1,5 @@ -/* $Id: charmap.c,v 1.37 2005-06-14 12:42:49 adam Exp $ - Copyright (C) 1995-2005 - Index Data ApS - -This file is part of the Zebra server. +/* This file is part of the Zebra server. + Copyright (C) Index Data 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 @@ -15,18 +12,23 @@ 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. -*/ +along with this program; if not, write to the Free Software +Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA +*/ -/* +/** + * \file charmap.c + * \brief character conversions (.chr) + * * Support module to handle character-conversions into and out of the * Zebra dictionary. */ +#if HAVE_CONFIG_H +#include +#endif #include #include #include @@ -46,7 +48,7 @@ const unsigned char CHR_FIELD_BEGIN = '^'; const char *CHR_UNKNOWN = "\001"; const char *CHR_SPACE = "\002"; const char *CHR_CUT = "\003"; -const char *CHR_BASE = "\005"; +const char *CHR_BASE = "\005"; /* CHECK CHR_BASE_CHAR as well */ struct chrmaptab_info { @@ -69,7 +71,7 @@ struct chr_t_entry /* * General argument structure for callback functions (internal use only) */ -typedef struct chrwork +typedef struct chrwork { chrmaptab map; char string[CHR_MAXSTR+1]; @@ -101,14 +103,15 @@ static chr_t_entry *set_map_string(chr_t_entry *root, NMEM nmem, } if (!len) { - if (!root->target || !root->target[0] || strcmp(root->target[0], to)) + if (!root->target || !root->target[0] || + strcmp((const char *) root->target[0], to)) { - if (from_0 && + if (from_0 && root->target && root->target[0] && root->target[0][0] && - strcmp (root->target[0], CHR_UNKNOWN)) + strcmp((const char *) root->target[0], CHR_UNKNOWN)) { - yaz_log (YLOG_WARN, "duplicate entry for charmap from '%s'", - from_0); + yaz_log(YLOG_WARN, "duplicate entry for charmap from '%s'", + from_0); } root->target = (unsigned char **) nmem_malloc(nmem, sizeof(*root->target)*2); @@ -128,32 +131,13 @@ static chr_t_entry *set_map_string(chr_t_entry *root, NMEM nmem, root->children[i] = 0; } if (!(root->children[(unsigned char) *from] = - set_map_string(root->children[(unsigned char) *from], nmem, - from + 1, len - 1, to, from_0))) + set_map_string(root->children[(unsigned char) *from], nmem, + from + 1, len - 1, to, from_0))) return 0; } return root; } -static chr_t_entry *find_entry(chr_t_entry *t, const char **from, int len) -{ - chr_t_entry *res; - - if (len && t->children && t->children[(unsigned char) **from]) - { - const char *pos = *from; - - (*from)++; - if ((res = find_entry(t->children[(unsigned char) *pos], - from, len - 1))) - return res; - /* no match */ - *from = pos; - } - /* no children match. use ourselves, if we have a target */ - return t->target ? t : 0; -} - static chr_t_entry *find_entry_x(chr_t_entry *t, const char **from, int *len, int first) { chr_t_entry *res; @@ -179,8 +163,8 @@ static chr_t_entry *find_entry_x(chr_t_entry *t, const char **from, int *len, in else res = 0; /* otherwhise there was no match on beginning of field, move on */ - } - + } + if (!res && t->children[(unsigned char) **from]) { (*len)--; @@ -226,7 +210,7 @@ const char **chr_map_q_input(chrmaptab maptab, chr_t_entry *t = maptab->q_input; chr_t_entry *res; int len_tmp[2]; - + len_tmp[0] = len; len_tmp[1] = -1; if (!(res = find_entry_x(t, from, len_tmp, first))) @@ -237,8 +221,11 @@ const char **chr_map_q_input(chrmaptab maptab, const char *chr_map_output(chrmaptab maptab, const char **from, int len) { unsigned char c = ** (unsigned char **) from; - (*from)++; - return (const char*) maptab->output[c]; + const char *out = (const char*) maptab->output[c]; + + if (out) + (*from)++; + return out; } static int zebra_ucs4_strlen(ucs4_t *s) @@ -255,7 +242,6 @@ ucs4_t zebra_prim_w(ucs4_t **s) ucs4_t i = 0; char fmtstr[8]; - yaz_log (YLOG_DEBUG, "prim_w %.3s", (char *) *s); if (**s == '\\' && 1[*s]) { (*s)++; @@ -267,7 +253,7 @@ ucs4_t zebra_prim_w(ucs4_t **s) case 'n': c = '\n'; (*s)++; break; case 't': c = '\t'; (*s)++; break; case 's': c = ' '; (*s)++; break; - case 'x': + case 'x': if (zebra_ucs4_strlen(*s) >= 3) { fmtstr[0] = (*s)[1]; @@ -321,7 +307,7 @@ ucs4_t zebra_prim_w(ucs4_t **s) c = **s; ++(*s); } - yaz_log (YLOG_DEBUG, "out %d", c); + yaz_log(YLOG_DEBUG, "out %d", c); return c; } @@ -333,14 +319,14 @@ static void fun_addentry(const char *s, void *data, int num) { chrmaptab tab = (chrmaptab) data; char tmp[2]; - + tmp[0] = num; tmp[1] = '\0'; tab->input = set_map_string(tab->input, tab->nmem, s, strlen(s), tmp, 0); tab->output[num + tab->base_uppercase] = (unsigned char *) nmem_strdup(tab->nmem, s); } -/* +/* * Callback function. * Add a space-entry to the value space. */ @@ -351,7 +337,7 @@ 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. */ @@ -383,7 +369,7 @@ static void fun_mkstring(const char *s, void *data, int num) static void fun_add_equivalent_string(const char *s, void *data, int num) { chr_equiv_work *arg = (chr_equiv_work *) data; - + if (arg->no_eq == CHR_MAXEQUIV) return; arg->eq[arg->no_eq++] = nmem_strdup(arg->nmem, s); @@ -397,30 +383,34 @@ static void fun_add_map(const char *s, void *data, int num) chrwork *arg = (chrwork *) data; assert(arg->map->input); - yaz_log (YLOG_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++) - yaz_log (YLOG_DEBUG, " %3d", (unsigned char) *s); + yaz_log(YLOG_DEBUG, " %3d", (unsigned char) *s); } -static int scan_to_utf8 (yaz_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 == 0) *outbuf++ = *from; /* ISO-8859-1 is OK here */ else { - ret = yaz_iconv (t, &inbuf, &inbytesleft, &outbuf, &outbytesleft); + ret = yaz_iconv(t, &inbuf, &inbytesleft, &outbuf, &outbytesleft); + if (ret != (size_t) (-1)) + ret = yaz_iconv(t, 0, 0, &outbuf, &outbytesleft); + + if (ret == (size_t) (-1)) { 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"); + yaz_log(YLOG_WARN|YLOG_ERRNO, "bad unicode sequence"); return -1; } } @@ -437,7 +427,7 @@ static int scan_string(char *s_native, ucs4_t arg[512]; ucs4_t arg_prim[512]; - ucs4_t *s0, *s = arg; + ucs4_t *s = arg; ucs4_t c, begin, end; size_t i; @@ -447,15 +437,18 @@ static int scan_string(char *s_native, char *inbuf = s_native; size_t outbytesleft = sizeof(arg)-4; size_t inbytesleft = strlen(s_native); - size_t ret; + size_t ret; ret = yaz_iconv(t_unicode, &inbuf, &inbytesleft, &outbuf, &outbytesleft); + if (ret != (size_t)(-1)) + ret = yaz_iconv(t_unicode, 0, 0, &outbuf, &outbytesleft); + if (ret == (size_t)(-1)) return -1; i = (outbuf - (char*) arg)/sizeof(ucs4_t); } else - { + { for (i = 0; s_native[i]; i++) arg[i] = s_native[i] & 255; /* ISO-8859-1 conversion */ } @@ -484,25 +477,32 @@ static int scan_string(char *s_native, s++; for (c = begin; c <= end; c++) { - if (scan_to_utf8 (t_utf8, &c, 1, str, sizeof(str)-1)) + if (scan_to_utf8(t_utf8, &c, 1, str, sizeof(str)-1)) return -1; (*fun)(str, data, num ? (*num)++ : 0); } break; case '(': ++s; - s0 = s; i = 0; + i = 0; while (*s != ')' || s[-1] == '\\') + { + if (*s == '\0') + { + yaz_log(YLOG_FATAL, "Missing ) in charmap"); + return -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)) + 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++; break; default: c = zebra_prim_w(&s); - if (scan_to_utf8 (t_utf8, &c, 1, str, sizeof(str)-1)) + if (scan_to_utf8(t_utf8, &c, 1, str, sizeof(str)-1)) return -1; (*fun)(str, data, num ? (*num)++ : 0); } @@ -510,13 +510,14 @@ static int scan_string(char *s_native, return 0; } -chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, - const char *tabroot) +chrmaptab chrmaptab_create(const char *tabpath, const char *name, + const char *tabroot) { FILE *f; char line[512], *argv[50]; chrmaptab res; int lineno = 0; + int no_directives = 0; int errors = 0; int argc, num = (int) *CHR_BASE, i; NMEM nmem; @@ -525,18 +526,19 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, unsigned endian = 31; const char *ucs4_native = "UCS-4"; - if (*(char*) &endian == 31) /* little endian? */ - ucs4_native = "UCS-4LE"; - - t_utf8 = yaz_iconv_open ("UTF-8", ucs4_native); - - yaz_log (YLOG_DEBUG, "maptab %s open", name); + yaz_log(YLOG_DEBUG, "maptab %s open", name); if (!(f = yaz_fopen(tabpath, name, "r", tabroot))) { yaz_log(YLOG_WARN|YLOG_ERRNO, "%s", name); return 0; } - nmem = nmem_create (); + + if (*(char*) &endian == 31) /* little endian? */ + ucs4_native = "UCS-4LE"; + + t_utf8 = yaz_iconv_open("UTF-8", ucs4_native); + + nmem = nmem_create(); res = (chrmaptab) nmem_malloc(nmem, sizeof(*res)); res->nmem = nmem; res->input = (chr_t_entry *) nmem_malloc(res->nmem, sizeof(*res->input)); @@ -552,17 +554,9 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, nmem_malloc(res->nmem, sizeof(*res->input)); res->input->children[i]->children = 0; res->input->children[i]->target = (unsigned char **) - nmem_malloc (res->nmem, 2 * sizeof(unsigned char *)); + nmem_malloc(res->nmem, 2 * sizeof(unsigned char *)); res->input->children[i]->target[1] = 0; - if (map_only) - { - res->input->children[i]->target[0] = (unsigned char *) - nmem_malloc (res->nmem, 2 * sizeof(unsigned char)); - res->input->children[i]->target[0][0] = i; - res->input->children[i]->target[0][1] = 0; - } - else - res->input->children[i]->target[0] = (unsigned char*) CHR_UNKNOWN; + res->input->children[i]->target[0] = (unsigned char*) CHR_UNKNOWN; } res->q_input = (chr_t_entry *) nmem_malloc(res->nmem, sizeof(*res->q_input)); @@ -576,7 +570,9 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, res->base_uppercase = 0; while (!errors && (argc = readconf_line(f, &lineno, line, 512, argv, 50))) - if (!map_only && !yaz_matchstr(argv[0], "lowercase")) + { + no_directives++; + if (!yaz_matchstr(argv[0], "lowercase")) { if (argc != 2) { @@ -594,7 +590,7 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, res->output[(int) *CHR_UNKNOWN + num] = (unsigned char*) "@"; num = (int) *CHR_BASE; } - else if (!map_only && !yaz_matchstr(argv[0], "uppercase")) + else if (!yaz_matchstr(argv[0], "uppercase")) { if (!res->base_uppercase) { @@ -613,7 +609,7 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, ++errors; } } - else if (!map_only && !yaz_matchstr(argv[0], "space")) + else if (!yaz_matchstr(argv[0], "space")) { if (argc != 2) { @@ -627,7 +623,7 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, ++errors; } } - else if (!map_only && !yaz_matchstr(argv[0], "cut")) + else if (!yaz_matchstr(argv[0], "cut")) { if (argc != 2) { @@ -676,7 +672,7 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, } w.nmem = res->nmem; w.no_eq = 0; - if (scan_string(argv[1], t_unicode, t_utf8, + if (scan_string(argv[1], t_unicode, t_utf8, fun_add_equivalent_string, &w, 0) < 0) { yaz_log(YLOG_FATAL, "equivalent: invalid string"); @@ -720,41 +716,28 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, } else if (!yaz_matchstr(argv[0], "encoding")) { - /* - * 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 + yaz_iconv_close(t_unicode); + t_unicode = yaz_iconv_open(ucs4_native, argv[1]); } else { yaz_log(YLOG_WARN, "Syntax error at '%s' in %s", line, name); + errors++; } - + } yaz_fclose(f); + if (no_directives == 0) + { + yaz_log(YLOG_WARN, "No directives in '%s'", name); + errors++; + } if (errors) { chrmaptab_destroy(res); res = 0; } - yaz_log (YLOG_DEBUG, "maptab %s close %d errors", name, errors); + yaz_log(YLOG_DEBUG, "maptab %s num=%d close %d errors", name, num, errors); if (t_utf8 != 0) yaz_iconv_close(t_utf8); if (t_unicode != 0) @@ -765,7 +748,16 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, void chrmaptab_destroy(chrmaptab tab) { if (tab) - nmem_destroy (tab->nmem); + nmem_destroy(tab->nmem); } +/* + * Local variables: + * c-basic-offset: 4 + * c-file-style: "Stroustrup" + * indent-tabs-mode: nil + * End: + * vim: shiftwidth=4 tabstop=8 expandtab + */ +