X-Git-Url: http://git.indexdata.com/?p=idzebra-moved-to-github.git;a=blobdiff_plain;f=util%2Fcharmap.c;h=b36282f412f4275ab0fbf567355986d78387f546;hp=31642bcea4c4a67f445250ca0915105662cc7590;hb=7c3a0352f0492609a3b6b26b63a72b0b2d207aab;hpb=cad3d9cf6b923f0bd0adb6731db88e3ff6bac80a diff --git a/util/charmap.c b/util/charmap.c index 31642bc..b36282f 100644 --- a/util/charmap.c +++ b/util/charmap.c @@ -1,5 +1,5 @@ -/* $Id: charmap.c,v 1.27 2003-01-13 10:53:16 oleg Exp $ - Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002 +/* $Id: charmap.c,v 1.30 2004-09-14 14:38:08 quinn Exp $ + Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002,2003,2004 Index Data Aps This file is part of the Zebra server. @@ -40,9 +40,12 @@ typedef unsigned ucs4_t; #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 { @@ -141,7 +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; @@ -152,35 +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; @@ -188,7 +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); } @@ -244,13 +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 (LOG_DEBUG, "prim_w %.3s", (char *) *s); if (**s == '\\') { (*s)++; @@ -263,13 +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': @@ -280,14 +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)++; } @@ -327,6 +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. */ @@ -335,7 +390,7 @@ 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); strncat(arg->string, *res, CHR_MAXSTR - strlen(arg->string)); @@ -386,6 +441,8 @@ static int scan_to_utf8 (yaz_iconv_t t, ucs4_t *from, size_t inlen, 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; } @@ -402,6 +459,7 @@ static int scan_string(char *s_native, 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; @@ -457,11 +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++; @@ -583,7 +641,7 @@ 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, @@ -593,6 +651,20 @@ 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; @@ -648,20 +720,23 @@ chrmaptab chrmaptab_create(const char *tabpath, const char *name, int map_only, * 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 {