X-Git-Url: http://git.indexdata.com/?p=idzebra-moved-to-github.git;a=blobdiff_plain;f=util%2Fzebramap.c;h=b075257e24094de70f3f1c9a8af9e6d4cb3747c9;hp=8fe80416df4d7569c04e39f9c568281c73e7952e;hb=c33ea56e3771c3b80ba66ef8fda3a09cad171ebb;hpb=988d4b71d4f60465198a51a0f43d166815541040 diff --git a/util/zebramap.c b/util/zebramap.c index 8fe8041..b075257 100644 --- a/util/zebramap.c +++ b/util/zebramap.c @@ -1,269 +1,492 @@ -/* - * Copyright (C) 1994-1997, Index Data I/S - * All rights reserved. - * Sebastian Hammer, Adam Dickmeiss - * - * $Log: zebramap.c,v $ - * Revision 1.3 1997-11-17 15:35:26 adam - * Bug fix. Relation=relevance wasn't observed. - * - * Revision 1.2 1997/10/31 12:39:30 adam - * Changed log message. - * - * Revision 1.1 1997/10/27 14:33:06 adam - * Moved towards generic character mapping depending on "structure" - * field in abstract syntax file. Fixed a few memory leaks. Fixed - * bug with negative integers when doing searches with relational - * operators. - * - */ +/* $Id: zebramap.c,v 1.53 2006-09-08 14:41:00 adam Exp $ + Copyright (C) 1995-2006 + 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. +*/ #include +#include #include -#include #include +#include +#include + #include +#define ZEBRA_MAP_TYPE_SORT 1 +#define ZEBRA_MAP_TYPE_INDEX 2 + +#define ZEBRA_REPLACE_ANY 300 + struct zebra_map { - int reg_type; + unsigned reg_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; struct zebra_map *next; }; struct zebra_maps { 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; + int no_maps; }; -void zebra_maps_close (ZebraMaps zms) +void zebra_maps_close(ZebraMaps zms) { struct zebra_map *zm = zms->map_list; while (zm) { - struct zebra_map *zm_next = zm->next; + if (zm->maptab) + chrmaptab_destroy(zm->maptab); + zm = zm->next; + } + wrbuf_free(zms->wrbuf_1, 1); + nmem_destroy(zms->nmem); + xfree(zms); +} + +ZEBRA_RES zebra_maps_read_file(ZebraMaps zms, const char *fname) +{ + FILE *f; + char line[512]; + char *argv[10]; + int argc; + int lineno = 0; + int failures = 0; + struct zebra_map **zm = 0, *zp; - chrmaptab_destroy (zm->maptab); - xfree (zm); - zm = zm_next; + if (!(f = yaz_fopen(zms->tabpath, fname, "r", zms->tabroot))) + { + yaz_log(YLOG_ERRNO|YLOG_FATAL, "%s", fname); + return ZEBRA_FAIL; + } + while ((argc = readconf_line(f, &lineno, line, 512, argv, 10))) + { + 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)->alwaysmatches = 0; + (*zm)->first_in_field = 0; + zms->no_maps++; + } + 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)->alwaysmatches = 0; + (*zm)->first_in_field = 0; + zms->no_maps++; + } + 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)->maptab_name = nmem_strdup(zms->nmem, argv[1]); + } + else if (!yaz_matchstr(argv[0], "completeness") && argc == 2) + { + (*zm)->completeness = atoi(argv[1]); + } + else if (!yaz_matchstr(argv[0], "position") && argc == 2) + { + (*zm)->positioned = atoi(argv[1]); + } + else if (!yaz_matchstr(argv[0], "alwaysmatches") && argc == 2) + { + (*zm)->alwaysmatches = atoi(argv[1]); + } + 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]); + } + else + { + yaz_log(YLOG_WARN, "%s:%d: Unrecognized directive '%s'", + fname, lineno, argv[0]); + failures++; + } } - nmem_destroy (zms->nmem); - xfree (zms); + if (zm) + (*zm)->next = NULL; + 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; } -ZebraMaps zebra_maps_open (const char *tabpath) +ZebraMaps zebra_maps_open(Res res, const char *base_path, + const char *profile_path) { - ZebraMaps zms = xmalloc (sizeof(*zms)); + ZebraMaps zms = (ZebraMaps) xmalloc(sizeof(*zms)); + int i; - zms->nmem = nmem_create (); - zms->tabpath = nmem_strdup (zms->nmem, tabpath); + zms->nmem = nmem_create(); + zms->no_maps = 0; + zms->tabpath = nmem_strdup(zms->nmem, profile_path); + zms->tabroot = 0; + if (base_path) + zms->tabroot = nmem_strdup(zms->nmem, base_path); zms->map_list = NULL; + + zms->temp_map_str[0] = '\0'; + zms->temp_map_str[1] = '\0'; + + 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); + zms->wrbuf_1 = wrbuf_alloc(); + + for (i = 0; i<256; i++) + zms->lookup_array[i] = 0; return zms; } -chrmaptab zebra_map_get (ZebraMaps zms, int reg_type) +struct zebra_map *zebra_map_get(ZebraMaps zms, unsigned reg_id) { - char name[512]; - struct zebra_map *zm; + assert(reg_id >= 0 && reg_id <= 255); + return zms->lookup_array[reg_id]; +} - for (zm = zms->map_list; zm; zm = zm->next) +chrmaptab zebra_charmap_get(ZebraMaps zms, unsigned reg_id) +{ + struct zebra_map *zm = zebra_map_get(zms, reg_id); + if (!zm) { - if (reg_type == zm->reg_type) - return zm->maptab; + zm = (struct zebra_map *) nmem_malloc(zms->nmem, sizeof(*zm)); + + /* no reason to warn if no maps are installed at ALL */ + if (zms->no_maps) + 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->completeness = 0; + zm->next = zms->map_list; + zms->map_list = zm->next; + + zms->lookup_array[zm->reg_id & 255] = zm; } - *name = '\0'; - switch (reg_type) + if (!zm->maptab) { - case 'w': - case 'p': - strcat (name, "string"); - break; - case 'n': - strcat (name, "numeric"); - break; - case 'u': - strcat (name, "urx"); - break; - default: - strcat (name, "null"); + if (!zm->maptab_name || !yaz_matchstr(zm->maptab_name, "@")) + return NULL; + if (!(zm->maptab = chrmaptab_create(zms->tabpath, + zm->maptab_name, + zms->tabroot))) + yaz_log(YLOG_WARN, "Failed to read character table %s", + zm->maptab_name); + else + yaz_log(YLOG_DEBUG, "Read character table %s", zm->maptab_name); } - strcat (name, ".chr"); - - zm = xmalloc (sizeof(*zm)); - zm->reg_type = reg_type; - zm->next = zms->map_list; - zms->map_list = zm; - if (!(zm->maptab = chrmaptab_create (zms->tabpath, name, 0))) - logf(LOG_WARN, "Failed to read character table %s", name); - else - logf(LOG_DEBUG, "Read character table %s", name); return zm->maptab; } -const char **zebra_maps_input (ZebraMaps zms, int reg_type, - const char **from, int len) +const char **zebra_maps_input(ZebraMaps zms, unsigned reg_id, + const char **from, int len, int first) { - static char str[2] = {0,0}; - static const char *buf[2] = {0,0}; chrmaptab maptab; - maptab = zebra_map_get (zms, reg_type); + maptab = zebra_charmap_get(zms, reg_id); if (maptab) - return chr_map_input(maptab, from, len); - - if (isalnum(**from)) - { - str[0] = isupper(**from) ? tolower(**from) : **from; - buf[0] = str; - } - else - buf[0] = (char*) CHR_SPACE; + return chr_map_input(maptab, from, len, first); + + zms->temp_map_str[0] = **from; + (*from)++; - return buf; + return zms->temp_map_ptr; } -const char *zebra_maps_output(ZebraMaps zms, int reg_type, const char **from) +const char **zebra_maps_search(ZebraMaps zms, unsigned reg_id, + const char **from, int len, int *q_map_match) { chrmaptab maptab; - unsigned char i = (unsigned char) **from; - static char buf[2] = {0,0}; - - maptab = zebra_map_get (zms, reg_type); + + *q_map_match = 0; + maptab = zebra_charmap_get(zms, reg_id); if (maptab) - return chr_map_output (maptab, from, 1); + { + 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; + } + zms->temp_map_str[0] = **from; + (*from)++; - buf[0] = i; - return buf; + return zms->temp_map_ptr; +} + +const char *zebra_maps_output(ZebraMaps zms, unsigned reg_id, + const char **from) +{ + chrmaptab maptab = zebra_charmap_get(zms, reg_id); + if (!maptab) + return 0; + return chr_map_output(maptab, from, 1); } /* ------------------------------------ */ -typedef struct { - int type; - int major; - int minor; - Z_AttributesPlusTerm *zapt; -} AttrType; +int zebra_maps_is_complete(ZebraMaps zms, unsigned reg_id) +{ + struct zebra_map *zm = zebra_map_get(zms, reg_id); + if (zm) + return zm->completeness; + return 0; +} -static int attr_find (AttrType *src, oid_value *attributeSetP) +int zebra_maps_is_positioned(ZebraMaps zms, unsigned reg_id) { - while (src->major < src->zapt->num_attributes) - { - Z_AttributeElement *element; - - element = src->zapt->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; + struct zebra_map *zm = zebra_map_get(zms, reg_id); + if (zm) + return zm->positioned; + return 0; } - -static void attr_init (AttrType *src, Z_AttributesPlusTerm *zapt, - int type) + +int zebra_maps_is_sort(ZebraMaps zms, unsigned reg_id) { - src->zapt = zapt; - src->type = type; - src->major = 0; - src->minor = 0; + struct zebra_map *zm = zebra_map_get(zms, reg_id); + if (zm) + return zm->type == ZEBRA_MAP_TYPE_SORT; + return 0; } -/* ------------------------------------ */ +int zebra_maps_is_alwaysmatches(ZebraMaps zms, unsigned reg_id) +{ + struct zebra_map *zm = zebra_map_get(zms, reg_id); + if (zm) + return zm->alwaysmatches; + return 0; +} -int zebra_maps_is_complete (ZebraMaps zms, int reg_type) +int zebra_maps_is_first_in_field(ZebraMaps zms, unsigned reg_id) { - if (reg_type == 'p') - return 1; + struct zebra_map *zm = zebra_map_get(zms, reg_id); + if (zm) + return zm->first_in_field; return 0; } -int zebra_maps_attr (ZebraMaps zms, Z_AttributesPlusTerm *zapt, - int *reg_type, char **search_type, int *complete_flag) +int zebra_maps_sort(ZebraMaps 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); + + *numerical = 0; + structure_value = attr_find(&structure, 0); + if (structure_value == 109) + *numerical = 1; + 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) { AttrType completeness; AttrType structure; AttrType relation; + AttrType sort_relation; + AttrType weight; + AttrType use; int completeness_value; int structure_value; int relation_value; + int sort_relation_value; + int weight_value; + int use_value; - attr_init (&structure, zapt, 4); - attr_init (&completeness, zapt, 6); - attr_init (&relation, zapt, 2); + 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); + 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); + use_value = attr_find(&use, NULL); if (completeness_value == 2 || completeness_value == 3) *complete_flag = 1; else *complete_flag = 0; - *reg_type = 0; + *reg_id = 0; + *sort_flag =(sort_relation_value > 0) ? 1 : 0; *search_type = "phrase"; + strcpy(rank_type, "void"); if (relation_value == 102) - *search_type = "ranked"; - + { + if (weight_value == -1) + weight_value = 34; + sprintf(rank_type, "rank,w=%d,u=%d", weight_value, use_value); + } + if (*complete_flag) + *reg_id = 'p'; + else + *reg_id = 'w'; switch (structure_value) { - case -1: - case 1: /* phrase */ - case 2: /* word */ - case 3: /* key */ case 6: /* word list */ + *search_type = "and-list"; + break; case 105: /* free-form-text */ + *search_type = "or-list"; + break; case 106: /* document-text */ + *search_type = "or-list"; + break; + case -1: + case 1: /* phrase */ + case 2: /* word */ case 108: /* string */ - if (*complete_flag) - *reg_type = 'p'; - else - *reg_type = 'w'; + *search_type = "phrase"; break; case 107: /* local-number */ *search_type = "local"; - *reg_type = 0; + *reg_id = 0; + break; case 109: /* numeric string */ - *reg_type = 'n'; + *reg_id = 'n'; + *search_type = "numeric"; break; case 104: /* urx */ - *reg_type = 'u'; + *reg_id = 'u'; + *search_type = "phrase"; break; + case 3: /* key */ + *reg_id = '0'; + *search_type = "phrase"; + break; + case 4: /* year */ + *reg_id = 'y'; + *search_type = "phrase"; + break; + case 5: /* date */ + *reg_id = 'd'; + *search_type = "phrase"; + break; default: return -1; } return 0; } + +WRBUF zebra_replace(ZebraMaps zms, unsigned reg_id, const char *ex_list, + const char *input_str, int input_len) +{ + wrbuf_rewind(zms->wrbuf_1); + wrbuf_write(zms->wrbuf_1, input_str, input_len); + return zms->wrbuf_1; +} + +/* + * Local variables: + * c-basic-offset: 4 + * indent-tabs-mode: nil + * End: + * vim: shiftwidth=4 tabstop=8 expandtab + */ +