X-Git-Url: http://git.indexdata.com/?p=idzebra-moved-to-github.git;a=blobdiff_plain;f=util%2Fzebramap.c;h=3fe7f06d4210f9826221df8c4841c9fb280a7e6c;hp=005793ba6b4dbeb00b024f27c54c530882d8fd98;hb=e199777080c6fa0963d51b7df1763fd5286ca9a4;hpb=8add234f71c852fd95ca3aef168e3563265c93b9 diff --git a/util/zebramap.c b/util/zebramap.c index 005793b..3fe7f06 100644 --- a/util/zebramap.c +++ b/util/zebramap.c @@ -1,23 +1,23 @@ -/* $Id: zebramap.c,v 1.39 2005-01-16 23:14:58 adam Exp $ - Copyright (C) 1995-2005 +/* $Id: zebramap.c,v 1.61 2007-10-31 16:56:15 adam Exp $ + Copyright (C) 1995-2007 Index Data ApS -This file is part of the Zebra server. + 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 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. + 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. + 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. */ #include @@ -25,212 +25,212 @@ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA #include #include +#include #include -#include +#include #define ZEBRA_MAP_TYPE_SORT 1 #define ZEBRA_MAP_TYPE_INDEX 2 +#define ZEBRA_MAP_TYPE_STATICRANK 3 #define ZEBRA_REPLACE_ANY 300 -struct zm_token { - int *token_from; - char *token_to; - int token_min; - struct zm_token *next; -}; - struct zebra_map { - unsigned reg_id; + const char *id; int completeness; int positioned; + int alwaysmatches; + int first_in_field; int type; union { struct { - int dummy; - } index; - struct { int entry_size; } sort; } u; chrmaptab maptab; const char *maptab_name; + zebra_maps_t zebra_maps; struct zebra_map *next; - struct zm_token *replace_tokens; }; -struct zebra_maps { +struct zebra_maps_s { char *tabpath; char *tabroot; NMEM nmem; - struct zebra_map *map_list; char temp_map_str[2]; const char *temp_map_ptr[2]; - struct zebra_map **lookup_array; - WRBUF wrbuf_1, wrbuf_2; + WRBUF wrbuf_1; + int no_maps; + zebra_map_t map_list; + zebra_map_t *last_map; }; -void zebra_maps_close (ZebraMaps zms) +void zebra_maps_close(zebra_maps_t zms) { struct zebra_map *zm = zms->map_list; while (zm) { if (zm->maptab) - chrmaptab_destroy (zm->maptab); + chrmaptab_destroy(zm->maptab); zm = zm->next; } - wrbuf_free (zms->wrbuf_1, 1); - wrbuf_free (zms->wrbuf_2, 1); - nmem_destroy (zms->nmem); - xfree (zms); + wrbuf_destroy(zms->wrbuf_1); + nmem_destroy(zms->nmem); + xfree(zms); +} + +zebra_map_t zebra_add_map(zebra_maps_t zms, const char *index_type, + int map_type) +{ + zebra_map_t zm = (zebra_map_t) nmem_malloc(zms->nmem, sizeof(*zm)); + + zm->zebra_maps = zms; + zm->id = nmem_strdup(zms->nmem, index_type); + zm->maptab_name = 0; + zm->maptab = 0; + zm->type = map_type; + zm->completeness = 0; + zm->positioned = 0; + zm->alwaysmatches = 0; + zm->first_in_field = 0; + + zm->next = 0; + *zms->last_map = zm; + zms->last_map = &zm->next; + + zms->no_maps++; + + return zm; } -static void zebra_map_read (ZebraMaps zms, const char *name) +ZEBRA_RES zebra_maps_read_file(zebra_maps_t zms, const char *fname) { FILE *f; char line[512]; char *argv[10]; int argc; int lineno = 0; - struct zebra_map **zm = 0, *zp; + int failures = 0; + zebra_map_t zm = 0; - if (!(f = yaz_fopen(zms->tabpath, name, "r", zms->tabroot))) + if (!(f = yaz_fopen(zms->tabpath, fname, "r", zms->tabroot))) { - yaz_log(YLOG_WARN|YLOG_ERRNO, "%s", name); - return ; + yaz_log(YLOG_ERRNO|YLOG_FATAL, "%s", fname); + return ZEBRA_FAIL; } while ((argc = readconf_line(f, &lineno, line, 512, argv, 10))) { - if (!yaz_matchstr (argv[0], "index") && argc == 2) + if (argc == 1) + { + yaz_log(YLOG_WARN, "%s:%d: Missing arguments for '%s'", + fname, lineno, argv[0]); + failures++; + break; + } + if (argc > 2) + { + yaz_log(YLOG_WARN, "%s:%d: Too many arguments for '%s'", + fname, lineno, argv[0]); + failures++; + break; + } + if (!yaz_matchstr(argv[0], "index")) { - if (!zm) - zm = &zms->map_list; - else - zm = &(*zm)->next; - *zm = (struct zebra_map *) nmem_malloc (zms->nmem, sizeof(**zm)); - (*zm)->reg_id = argv[1][0]; - (*zm)->maptab_name = NULL; - (*zm)->maptab = NULL; - (*zm)->type = ZEBRA_MAP_TYPE_INDEX; - (*zm)->completeness = 0; - (*zm)->positioned = 1; - (*zm)->replace_tokens = 0; + zm = zebra_add_map(zms, argv[1], ZEBRA_MAP_TYPE_INDEX); + zm->positioned = 1; } - else if (!yaz_matchstr (argv[0], "sort") && argc == 2) + else if (!yaz_matchstr(argv[0], "sort")) { - if (!zm) - zm = &zms->map_list; - else - zm = &(*zm)->next; - *zm = (struct zebra_map *) nmem_malloc (zms->nmem, sizeof(**zm)); - (*zm)->reg_id = argv[1][0]; - (*zm)->maptab_name = NULL; - (*zm)->type = ZEBRA_MAP_TYPE_SORT; - (*zm)->u.sort.entry_size = 80; - (*zm)->maptab = NULL; - (*zm)->completeness = 0; - (*zm)->positioned = 0; - (*zm)->replace_tokens = 0; + zm = zebra_add_map(zms, argv[1], ZEBRA_MAP_TYPE_SORT); + zm->u.sort.entry_size = 80; } - else if (zm && !yaz_matchstr (argv[0], "charmap") && argc == 2) + else if (!yaz_matchstr(argv[0], "staticrank")) { - (*zm)->maptab_name = nmem_strdup (zms->nmem, argv[1]); + zm = zebra_add_map(zms, argv[1], ZEBRA_MAP_TYPE_STATICRANK); + zm->completeness = 1; } - else if (zm && !yaz_matchstr (argv[0], "completeness") && argc == 2) + else if (!zm) + { + yaz_log(YLOG_WARN, "%s:%d: Missing sort/index before '%s'", + fname, lineno, argv[0]); + failures++; + } + else if (!yaz_matchstr(argv[0], "charmap") && argc == 2) { - (*zm)->completeness = atoi (argv[1]); + if (zm->type != ZEBRA_MAP_TYPE_STATICRANK) + zm->maptab_name = nmem_strdup(zms->nmem, argv[1]); + else + { + yaz_log(YLOG_WARN|YLOG_FATAL, "%s:%d: charmap for " + "staticrank is invalid", fname, lineno); + yaz_log(YLOG_LOG, "Type is %d", zm->type); + failures++; + } } - else if (zm && !yaz_matchstr (argv[0], "position") && argc == 2) + else if (!yaz_matchstr(argv[0], "completeness") && argc == 2) { - (*zm)->positioned = atoi (argv[1]); + zm->completeness = atoi(argv[1]); } - else if (zm && !yaz_matchstr (argv[0], "entrysize") && argc == 2) + else if (!yaz_matchstr(argv[0], "position") && argc == 2) + { + zm->positioned = atoi(argv[1]); + } + else if (!yaz_matchstr(argv[0], "alwaysmatches") && argc == 2) + { + if (zm->type != ZEBRA_MAP_TYPE_STATICRANK) + zm->alwaysmatches = atoi(argv[1]); + else + { + yaz_log(YLOG_WARN|YLOG_FATAL, "%s:%d: alwaysmatches for " + "staticrank is invalid", fname, lineno); + failures++; + } + } + else if (!yaz_matchstr(argv[0], "firstinfield") && argc == 2) + { + zm->first_in_field = atoi(argv[1]); + } + else if (!yaz_matchstr(argv[0], "entrysize") && argc == 2) { - if ((*zm)->type == ZEBRA_MAP_TYPE_SORT) - (*zm)->u.sort.entry_size = atoi (argv[1]); + if (zm->type == ZEBRA_MAP_TYPE_SORT) + zm->u.sort.entry_size = atoi(argv[1]); + else + { + yaz_log(YLOG_WARN, + "%s:%d: entrysize only valid in sort section", + fname, lineno); + failures++; + } + } - else if (zm && !yaz_matchstr (argv[0], "replace") && argc >= 2) + else { - struct zm_token *token = nmem_malloc (zms->nmem, sizeof(*token)); - token->next = (*zm)->replace_tokens; - (*zm)->replace_tokens = token; -#if 0 - yaz_log (YLOG_LOG, "replace %s", argv[1]); -#endif - token->token_from = 0; - if (argc >= 2) - { - char *cp = argv[1]; - int *dp = token->token_from = (int *) - nmem_malloc (zms->nmem, (1+strlen(cp))*sizeof(int)); - while (*cp) - if (*cp == '$') - { - *dp++ = ' '; - cp++; - } - else if (*cp == '.') - { - *dp++ = ZEBRA_REPLACE_ANY; - cp++; - } - else - { - *dp++ = zebra_prim(&cp); -#if 0 - yaz_log (YLOG_LOG, " char %2X %c", dp[-1], dp[-1]); -#endif - } - *dp = '\0'; - } - if (argc >= 3) - { - char *cp = argv[2]; - char *dp = token->token_to = - nmem_malloc (zms->nmem, strlen(cp)+1); - while (*cp) - if (*cp == '$') - { - *dp++ = ' '; - cp++; - } - else - *dp++ = zebra_prim(&cp); - *dp = '\0'; - } - else - token->token_to = 0; + yaz_log(YLOG_WARN, "%s:%d: Unrecognized directive '%s'", + fname, lineno, argv[0]); + failures++; } } - if (zm) - (*zm)->next = NULL; - yaz_fclose (f); + yaz_fclose(f); - for (zp = zms->map_list; zp; zp = zp->next) - zms->lookup_array[zp->reg_id] = zp; + if (failures) + return ZEBRA_FAIL; + return ZEBRA_OK; } -static void zms_map_handle (void *p, const char *name, const char *value) +zebra_maps_t zebra_maps_open(Res res, const char *base_path, + const char *profile_path) { - ZebraMaps zms = (ZebraMaps) p; - - zebra_map_read (zms, value); -} + zebra_maps_t zms = (zebra_maps_t) xmalloc(sizeof(*zms)); -ZebraMaps zebra_maps_open (Res res, const char *base) -{ - ZebraMaps zms = (ZebraMaps) xmalloc (sizeof(*zms)); - int i; - - zms->nmem = nmem_create (); - zms->tabpath = nmem_strdup (zms->nmem, - res_get_def (res, "profilePath", - DEFAULT_PROFILE_PATH)); + zms->nmem = nmem_create(); + zms->no_maps = 0; + zms->tabpath = profile_path ? nmem_strdup(zms->nmem, profile_path) : 0; zms->tabroot = 0; - if (base) - zms->tabroot = nmem_strdup (zms->nmem, base); - zms->map_list = NULL; + if (base_path) + zms->tabroot = nmem_strdup(zms->nmem, base_path); + zms->map_list = 0; + zms->last_map = &zms->map_list; zms->temp_map_str[0] = '\0'; zms->temp_map_str[1] = '\0'; @@ -238,201 +238,178 @@ ZebraMaps zebra_maps_open (Res res, const char *base) zms->temp_map_ptr[0] = zms->temp_map_str; zms->temp_map_ptr[1] = NULL; - zms->lookup_array = (struct zebra_map**) - nmem_malloc (zms->nmem, sizeof(*zms->lookup_array)*256); - for (i = 0; i<256; i++) - zms->lookup_array[i] = 0; - if (!res || !res_trav (res, "index", zms, zms_map_handle)) - zebra_map_read (zms, "default.idx"); - zms->wrbuf_1 = wrbuf_alloc(); - zms->wrbuf_2 = wrbuf_alloc(); return zms; } -struct zebra_map *zebra_map_get (ZebraMaps zms, unsigned reg_id) +zebra_map_t zebra_map_get(zebra_maps_t zms, const char *id) { - return zms->lookup_array[reg_id]; + zebra_map_t zm; + for (zm = zms->map_list; zm; zm = zm->next) + if (!strcmp(zm->id, id)) + break; + return zm; } -chrmaptab zebra_charmap_get (ZebraMaps zms, unsigned reg_id) +zebra_map_t zebra_map_get_or_add(zebra_maps_t zms, const char *id) { - struct zebra_map *zm = zebra_map_get (zms, reg_id); + struct zebra_map *zm = zebra_map_get(zms, id); if (!zm) { - zm = (struct zebra_map *) nmem_malloc (zms->nmem, sizeof(*zm)); - yaz_log (YLOG_WARN, "Unknown register type: %c", reg_id); - - zm->reg_id = reg_id; - zm->maptab_name = nmem_strdup (zms->nmem, "@"); - zm->maptab = NULL; - zm->type = ZEBRA_MAP_TYPE_INDEX; + zm = zebra_add_map(zms, id, ZEBRA_MAP_TYPE_INDEX); + + /* no reason to warn if no maps are installed at ALL + Note that zebra_add_maps increments no_maps .. + */ + if (zms->no_maps > 1) + yaz_log(YLOG_WARN, "Unknown register type: %s", id); + else + zms->no_maps = 0; + + zm->maptab_name = nmem_strdup(zms->nmem, "@"); zm->completeness = 0; - zm->next = zms->map_list; - zm->replace_tokens = 0; - zms->map_list = zm->next; - - zms->lookup_array[zm->reg_id & 255] = zm; + zm->positioned = 1; } + return zm; +} + +chrmaptab zebra_charmap_get(zebra_map_t zm) +{ if (!zm->maptab) { - if (!zm->maptab_name || !yaz_matchstr (zm->maptab_name, "@")) + if (!zm->maptab_name || !yaz_matchstr(zm->maptab_name, "@")) return NULL; - if (!(zm->maptab = chrmaptab_create (zms->tabpath, - zm->maptab_name, 0, - zms->tabroot))) + if (!(zm->maptab = chrmaptab_create(zm->zebra_maps->tabpath, + zm->maptab_name, + zm->zebra_maps->tabroot))) yaz_log(YLOG_WARN, "Failed to read character table %s", - zm->maptab_name); + zm->maptab_name); else yaz_log(YLOG_DEBUG, "Read character table %s", zm->maptab_name); } return zm->maptab; } -const char **zebra_maps_input (ZebraMaps zms, unsigned reg_id, - const char **from, int len, int first) +const char **zebra_maps_input(zebra_map_t zm, + const char **from, int len, int first) { - chrmaptab maptab; - - maptab = zebra_charmap_get (zms, reg_id); + chrmaptab maptab = zebra_charmap_get(zm); if (maptab) return chr_map_input(maptab, from, len, first); - zms->temp_map_str[0] = **from; + zm->zebra_maps->temp_map_str[0] = **from; + + (*from)++; + return zm->zebra_maps->temp_map_ptr; +} + +const char **zebra_maps_search(zebra_map_t zm, + const char **from, int len, int *q_map_match) +{ + chrmaptab maptab; + + *q_map_match = 0; + maptab = zebra_charmap_get(zm); + if (maptab) + { + const char **map; + map = chr_map_q_input(maptab, from, len, 0); + if (map && map[0]) + { + *q_map_match = 1; + return map; + } + map = chr_map_input(maptab, from, len, 0); + if (map) + return map; + } + zm->zebra_maps->temp_map_str[0] = **from; (*from)++; - return zms->temp_map_ptr; + return zm->zebra_maps->temp_map_ptr; } -const char *zebra_maps_output(ZebraMaps zms, unsigned reg_id, +const char *zebra_maps_output(zebra_map_t zm, const char **from) { - chrmaptab maptab = zebra_charmap_get (zms, reg_id); + chrmaptab maptab = zebra_charmap_get(zm); if (!maptab) return 0; - return chr_map_output (maptab, from, 1); + return chr_map_output(maptab, from, 1); } /* ------------------------------------ */ -typedef struct { - int type; - int major; - int minor; - Z_AttributeElement **attributeList; - int num_attributes; -} AttrType; +int zebra_maps_is_complete(zebra_map_t zm) +{ + if (zm) + return zm->completeness; + return 0; +} -static int attr_find (AttrType *src, oid_value *attributeSetP) +int zebra_maps_is_positioned(zebra_map_t zm) { - while (src->major < src->num_attributes) - { - Z_AttributeElement *element; - - element = src->attributeList[src->major]; - if (src->type == *element->attributeType) - { - switch (element->which) - { - case Z_AttributeValue_numeric: - ++(src->major); - if (element->attributeSet && attributeSetP) - { - oident *attrset; - - attrset = oid_getentbyoid (element->attributeSet); - *attributeSetP = attrset->value; - } - return *element->value.numeric; - break; - case Z_AttributeValue_complex: - if (src->minor >= element->value.complex->num_list || - element->value.complex->list[src->minor]->which != - Z_StringOrNumeric_numeric) - break; - ++(src->minor); - if (element->attributeSet && attributeSetP) - { - oident *attrset; - - attrset = oid_getentbyoid (element->attributeSet); - *attributeSetP = attrset->value; - } - return *element->value.complex->list[src->minor-1]->u.numeric; - default: - assert (0); - } - } - ++(src->major); - } - return -1; + if (zm) + return zm->positioned; + return 0; } -static void attr_init_APT (AttrType *src, Z_AttributesPlusTerm *zapt, int type) +int zebra_maps_is_index(zebra_map_t zm) { - src->attributeList = zapt->attributes->attributes; - src->num_attributes = zapt->attributes->num_attributes; - src->type = type; - src->major = 0; - src->minor = 0; + if (zm) + return zm->type == ZEBRA_MAP_TYPE_INDEX; + return 0; } -static void attr_init_AttrList (AttrType *src, Z_AttributeList *list, int type) +int zebra_maps_is_staticrank(zebra_map_t zm) { - src->attributeList = list->attributes; - src->num_attributes = list->num_attributes; - src->type = type; - src->major = 0; - src->minor = 0; + if (zm) + return zm->type == ZEBRA_MAP_TYPE_STATICRANK; + return 0; } - -/* ------------------------------------ */ - -int zebra_maps_is_complete (ZebraMaps zms, unsigned reg_id) -{ - struct zebra_map *zm = zebra_map_get (zms, reg_id); + +int zebra_maps_is_sort(zebra_map_t zm) +{ if (zm) - return zm->completeness; + return zm->type == ZEBRA_MAP_TYPE_SORT; return 0; } -int zebra_maps_is_positioned (ZebraMaps zms, unsigned reg_id) +int zebra_maps_is_alwaysmatches(zebra_map_t zm) { - struct zebra_map *zm = zebra_map_get (zms, reg_id); if (zm) - return zm->positioned; + return zm->alwaysmatches; return 0; } - -int zebra_maps_is_sort (ZebraMaps zms, unsigned reg_id) + +int zebra_maps_is_first_in_field(zebra_map_t zm) { - struct zebra_map *zm = zebra_map_get (zms, reg_id); if (zm) - return zm->type == ZEBRA_MAP_TYPE_SORT; + return zm->first_in_field; return 0; } -int zebra_maps_sort (ZebraMaps zms, Z_SortAttributes *sortAttributes, - int *numerical) +int zebra_maps_sort(zebra_maps_t zms, Z_SortAttributes *sortAttributes, + int *numerical) { AttrType use; AttrType structure; int structure_value; - attr_init_AttrList (&use, sortAttributes->list, 1); - attr_init_AttrList (&structure, sortAttributes->list, 4); + attr_init_AttrList(&use, sortAttributes->list, 1); + attr_init_AttrList(&structure, sortAttributes->list, 4); *numerical = 0; - structure_value = attr_find (&structure, 0); + structure_value = attr_find(&structure, 0); if (structure_value == 109) *numerical = 1; - return attr_find (&use, NULL); + return attr_find(&use, NULL); } -int zebra_maps_attr (ZebraMaps zms, Z_AttributesPlusTerm *zapt, - unsigned *reg_id, char **search_type, char *rank_type, - int *complete_flag, int *sort_flag) +int zebra_maps_attr(zebra_maps_t zms, Z_AttributesPlusTerm *zapt, + const char **index_type, char **search_type, char *rank_type, + int *complete_flag, int *sort_flag) { AttrType completeness; AttrType structure; @@ -442,49 +419,45 @@ int zebra_maps_attr (ZebraMaps zms, Z_AttributesPlusTerm *zapt, AttrType use; int completeness_value; int structure_value; + const char *structure_str = 0; int relation_value; int sort_relation_value; int weight_value; int use_value; - attr_init_APT (&structure, zapt, 4); - attr_init_APT (&completeness, zapt, 6); - attr_init_APT (&relation, zapt, 2); - attr_init_APT (&sort_relation, zapt, 7); - attr_init_APT (&weight, zapt, 9); - attr_init_APT (&use, zapt, 1); - - completeness_value = attr_find (&completeness, NULL); - structure_value = attr_find (&structure, NULL); - relation_value = attr_find (&relation, NULL); - sort_relation_value = attr_find (&sort_relation, NULL); - weight_value = attr_find (&weight, NULL); + attr_init_APT(&structure, zapt, 4); + attr_init_APT(&completeness, zapt, 6); + attr_init_APT(&relation, zapt, 2); + attr_init_APT(&sort_relation, zapt, 7); + attr_init_APT(&weight, zapt, 9); + attr_init_APT(&use, zapt, 1); + + completeness_value = attr_find(&completeness, NULL); + structure_value = attr_find_ex(&structure, NULL, &structure_str); + relation_value = attr_find(&relation, NULL); + sort_relation_value = attr_find(&sort_relation, NULL); + weight_value = attr_find(&weight, NULL); use_value = attr_find(&use, NULL); if (completeness_value == 2 || completeness_value == 3) *complete_flag = 1; else *complete_flag = 0; - *reg_id = 0; + *index_type = 0; - *sort_flag = (sort_relation_value > 0) ? 1 : 0; + *sort_flag =(sort_relation_value > 0) ? 1 : 0; *search_type = "phrase"; - strcpy (rank_type, "void"); + strcpy(rank_type, "void"); if (relation_value == 102) { if (weight_value == -1) weight_value = 34; - sprintf (rank_type, "rank,w=%d,u=%d", weight_value, use_value); - } - if (relation_value == 103) - { - *search_type = "always"; - return 0; + sprintf(rank_type, "rank,w=%d,u=%d", weight_value, use_value); } if (*complete_flag) - *reg_id = 'p'; + *index_type = "p"; else - *reg_id = 'w'; + *index_type = "w"; switch (structure_value) { case 6: /* word list */ @@ -504,136 +477,53 @@ int zebra_maps_attr (ZebraMaps zms, Z_AttributesPlusTerm *zapt, break; case 107: /* local-number */ *search_type = "local"; - *reg_id = 0; + *index_type = 0; break; case 109: /* numeric string */ - *reg_id = 'n'; + *index_type = "n"; *search_type = "numeric"; break; case 104: /* urx */ - *reg_id = 'u'; + *index_type = "u"; *search_type = "phrase"; break; case 3: /* key */ - *reg_id = '0'; + *index_type = "0"; *search_type = "phrase"; break; case 4: /* year */ - *reg_id = 'y'; + *index_type = "y"; *search_type = "phrase"; break; case 5: /* date */ - *reg_id = 'd'; + *index_type = "d"; *search_type = "phrase"; break; + case -2: + if (structure_str && *structure_str) + *index_type = structure_str; + else + return -1; + break; default: return -1; } return 0; } -int zebra_replace_sub(ZebraMaps zms, unsigned reg_id, const char *ex_list, - const char *input_str, int input_len, WRBUF wrbuf); - -WRBUF zebra_replace(ZebraMaps zms, unsigned reg_id, const char *ex_list, +WRBUF zebra_replace(zebra_map_t zm, const char *ex_list, const char *input_str, int input_len) { - struct zebra_map *zm = zebra_map_get (zms, reg_id); - - wrbuf_rewind(zms->wrbuf_1); - wrbuf_write(zms->wrbuf_1, input_str, input_len); - if (!zm || !zm->replace_tokens) - return zms->wrbuf_1; - -#if 0 - yaz_log (YLOG_LOG, "in:%.*s:", wrbuf_len(zms->wrbuf_1), - wrbuf_buf(zms->wrbuf_1)); -#endif - for (;;) - { - if (!zebra_replace_sub(zms, reg_id, ex_list, wrbuf_buf(zms->wrbuf_1), - wrbuf_len(zms->wrbuf_1), zms->wrbuf_2)) - return zms->wrbuf_2; - if (!zebra_replace_sub(zms, reg_id, ex_list, wrbuf_buf(zms->wrbuf_2), - wrbuf_len(zms->wrbuf_2), zms->wrbuf_1)) - return zms->wrbuf_1; - } - return 0; + wrbuf_rewind(zm->zebra_maps->wrbuf_1); + wrbuf_write(zm->zebra_maps->wrbuf_1, input_str, input_len); + return zm->zebra_maps->wrbuf_1; } -int zebra_replace_sub(ZebraMaps zms, unsigned reg_id, const char *ex_list, - const char *input_str, int input_len, WRBUF wrbuf) -{ - int i = -1; - int no_replaces = 0; - struct zebra_map *zm = zebra_map_get (zms, reg_id); - - wrbuf_rewind(wrbuf); - for (i = -1; i <= input_len; ) - { - struct zm_token *token; - char replace_string[128]; - int replace_out = 0; - int replace_in = 0; +/* + * Local variables: + * c-basic-offset: 4 + * indent-tabs-mode: nil + * End: + * vim: shiftwidth=4 tabstop=8 expandtab + */ - for (token = zm->replace_tokens; !replace_in && token; - token = token->next) - { - int j = 0; - int replace_done = 0; - replace_out = 0; - for (;; j++) - { - int c; - if (!token->token_from[j]) - { - replace_in = j; - break; - } - if (ex_list && strchr (ex_list, token->token_from[j])) - break; - if (i+j < 0 || j+i >= input_len) - c = ' '; - else - c = input_str[j+i] & 255; - if (token->token_from[j] == ZEBRA_REPLACE_ANY) - { - if (c == ' ') - break; - replace_string[replace_out++] = c; - } - else - { - if (c != token->token_from[j]) - { - break; - } - if (!replace_done) - { - const char *cp = token->token_to; - replace_done = 1; - for (; cp && *cp; cp++) - replace_string[replace_out++] = *cp; - } - } - } - } - if (!replace_in) - { - if (i >= 0 && i < input_len) - wrbuf_putc(wrbuf, input_str[i]); - i++; - } - else - { - no_replaces++; - if (replace_out) - wrbuf_write(wrbuf, replace_string, replace_out); - i += replace_in; - } - } -#if 0 - yaz_log (YLOG_LOG, "out:%.*s:", wrbuf_len(wrbuf), wrbuf_buf(wrbuf)); -#endif - return no_replaces; -}