X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=util%2Fcharmap.c;h=60bc25401ad4f358814bc358b9ae64b1485ea5a6;hb=de4ada45ce039553f8322e620cc047d8cf4a065e;hp=79a8210d08022dad01bfdb072371ea8cc904db77;hpb=26b8ff30091b8d7c27cc38de45af595acb0445bf;p=idzebra-moved-to-github.git diff --git a/util/charmap.c b/util/charmap.c index 79a8210..60bc254 100644 --- a/util/charmap.c +++ b/util/charmap.c @@ -4,7 +4,28 @@ * Sebastian Hammer, Adam Dickmeiss * * $Log: charmap.c,v $ - * Revision 1.4 1996-06-03 16:32:13 quinn + * Revision 1.11 1997-09-05 09:52:32 adam + * Extra argument added to function chr_read_maptab (tab path). + * + * Revision 1.10 1997/07/01 13:01:08 adam + * Bug fix in routine find_entry: didn't take into account the len arg. + * + * Revision 1.9 1996/10/29 13:48:14 adam + * Updated to use zebrautl.h instead of alexutil.h. + * + * Revision 1.8 1996/10/18 12:39:23 adam + * Uses LOG_DEBUG instead of LOG_WARN for "Character map overlap". + * + * Revision 1.7 1996/06/06 12:08:56 quinn + * Fixed bug. + * + * Revision 1.6 1996/06/04 13:28:00 quinn + * More work on charmapping + * + * Revision 1.5 1996/06/04 08:32:15 quinn + * Moved default keymap to keychars.c + * + * Revision 1.4 1996/06/03 16:32:13 quinn * Temporary bug-fix * * Revision 1.3 1996/06/03 15:17:46 quinn @@ -26,18 +47,20 @@ #include #include +#include -#include +#include #include #include #include +#define CHR_MAXSTR 1024 +#define CHR_MAXEQUIV 32 + const char *CHR_UNKNOWN = "\001"; const char *CHR_SPACE = "\002"; const char *CHR_BASE = "\003"; -extern char *data1_tabpath; - /* * Character map trie node. */ @@ -46,7 +69,16 @@ struct chr_t_entry chr_t_entry **children; /* array of children */ unsigned char *target; /* target for this node, if any */ unsigned char *equiv; /* equivalent to, or sumthin */ -} t_entry; +}; + +/* + * General argument structure for callback functions (internal use only) + */ +typedef struct chrwork +{ + chrmaptab *map; + char string[CHR_MAXSTR+1]; +} chrwork; /* * Add an entry to the character map. @@ -66,7 +98,7 @@ static chr_t_entry *set_map_string(chr_t_entry *root, char *from, int len, (char*) root->target == CHR_UNKNOWN) root->target = (unsigned char *) xstrdup(to); else if ((char*) to != CHR_SPACE) - logf(LOG_WARN, "Character map overlap"); + logf(LOG_DEBUG, "Character map overlap"); } else { @@ -117,7 +149,7 @@ static chr_t_entry *find_entry(chr_t_entry *t, char **from, int len) { chr_t_entry *res; - if (t->children && t->children[(unsigned char) **from]) + if (len && t->children && t->children[(unsigned char) **from]) { char *pos = *from; @@ -128,28 +160,15 @@ static chr_t_entry *find_entry(chr_t_entry *t, char **from, int len) /* no match */ *from = pos; } - /* no children match. use ourselves */ - return t; + /* no children match. use ourselves, if we have a target */ + return t->target ? t : 0; } char **chr_map_input(chr_t_entry *t, char **from, int len) { - static char *buf[2] = {0, 0}, str[2] = {0, 0}; + static char *buf[2] = {0, 0}; chr_t_entry *res; - if (!t) /* no table loaded. Null mapping */ - { - if (isalnum(**from)) - { - str[0] = **from; - buf[0] = str; - } - else - buf[0] = (char*) CHR_SPACE; - (*from)++; - return buf; - } - /* no children match. use our target string */ if (!(res = find_entry(t, from, len))) abort(); buf[0] = (char *) res->target; @@ -246,6 +265,32 @@ static void fun_addspace(char *s, void *data, int num) tab->input = set_map_string(tab->input, s, strlen(s), (char*) CHR_SPACE); } +/* + * Create a string containing the mapped characters provided. + */ +static void fun_mkstring(char *s, void *data, int num) +{ + chrwork *arg = data; + char **res, *p = s; + + res = chr_map_input(arg->map->input, &s, strlen(s)); + if (*res == (char*) CHR_UNKNOWN) + logf(LOG_WARN, "Map: '%s' has no mapping", p); + strncat(arg->string, *res, CHR_MAXSTR - strlen(arg->string)); + arg->string[CHR_MAXSTR] = '\0'; +} + +/* + * Add a map to the string contained in the argument. + */ +static void fun_addmap(char *s, void *data, int num) +{ + chrwork *arg = data; + + assert(arg->map->input); + set_map_string(arg->map->input, s, strlen(s), arg->string); +} + static int scan_string(char *s, void (*fun)(char *c, void *data, int num), void *data, int *num) { @@ -303,14 +348,14 @@ static int scan_string(char *s, void (*fun)(char *c, void *data, int num), return 0; } -chrmaptab *chr_read_maptab(char *name) +chrmaptab *chr_read_maptab(const char *tabpath, const char *name) { FILE *f; char line[512], *argv[50]; chrmaptab *res = xmalloc(sizeof(*res)); int argc, num = (int) *CHR_BASE, i; - if (!(f = yaz_path_fopen(data1_tabpath, name, "r"))) + if (!(f = yaz_path_fopen(tabpath, name, "r"))) { logf(LOG_WARN|LOG_ERRNO, "%s", name); return 0; @@ -387,13 +432,38 @@ chrmaptab *chr_read_maptab(char *name) fclose(f); return 0; } - if (scan_string(argv[1], fun_addspace, res, 0)) + if (scan_string(argv[1], fun_addspace, res, 0) < 0) { logf(LOG_FATAL, "Bad space specification"); fclose(f); return 0; } } + else if (!yaz_matchstr(argv[0], "map")) + { + chrwork buf; + + if (argc != 3) + { + logf(LOG_FATAL, "charmap MAP directive requires 2 args"); + fclose(f); + return 0; + } + buf.map = res; + buf.string[0] = '\0'; + if (scan_string(argv[2], fun_mkstring, &buf, 0) < 0) + { + logf(LOG_FATAL, "Bad map target"); + fclose(f); + return 0; + } + if (scan_string(argv[1], fun_addmap, &buf, 0) < 0) + { + logf(LOG_FATAL, "Bad map source"); + fclose(f); + return 0; + } + } else { #if 0