From: Adam Dickmeiss Date: Fri, 19 May 2006 13:49:33 +0000 (+0000) Subject: Zebra uses string attributes for indexing internally. Using set+numeric X-Git-Tag: before.bug.529~114 X-Git-Url: http://jsfdemo.indexdata.com/cgi-bin?a=commitdiff_plain;h=6ba9698e88c0283e40fa5980a1a6b551fff2d597;p=idzebra-moved-to-github.git Zebra uses string attributes for indexing internally. Using set+numeric use attribute can still be used. But that is a search-only conversion which inspects .att-set files as indicated using attset-directives in zebra.cfg. attset references are no longer required, but when used they deserve as "check" for that the index names used are also present in .att. --- diff --git a/NEWS b/NEWS index 49bb6f8..1210790 100644 --- a/NEWS +++ b/NEWS @@ -1,3 +1,9 @@ +Zebra uses string attributes for indexing internally. Using set+numeric +use attribute can still be used. But that is a search-only conversion which +inspects .att-set files as indicated using attset-directives in zebra.cfg. +attset references are no longer required, but when used they deserve +as "check" for that the index names used are also present in .att. + Implemented bug #588: allrecords search. Using @attr 1=allrecords "" matches all records in a database. diff --git a/data1/d1_absyn.c b/data1/d1_absyn.c index 1df3e3f..a1bc53d 100644 --- a/data1/d1_absyn.c +++ b/data1/d1_absyn.c @@ -1,5 +1,5 @@ -/* $Id: d1_absyn.c,v 1.24 2006-05-10 08:13:17 adam Exp $ - Copyright (C) 1995-2005 +/* $Id: d1_absyn.c,v 1.25 2006-05-19 13:49:33 adam Exp $ + Copyright (C) 1995-2006 Index Data ApS This file is part of the Zebra server. @@ -28,6 +28,7 @@ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA #include #include #include +#include #include #include @@ -509,17 +510,11 @@ static const char * mk_xpath_regexp (data1_handle dh, const char *expr) return res; } -/* *ostrich* - - added arg xpelement... when called from xelm context, it's 1, saying - that ! means xpath, not element name as attribute name... - - pop, 2002-12-13 - */ -static int parse_termlists (data1_handle dh, data1_termlist ***tpp, - char *cp, const char *file, int lineno, - const char *element_name, data1_absyn *res, - int xpelement) +static int parse_termlists(data1_handle dh, data1_termlist ***tpp, + char *cp, const char *file, int lineno, + const char *element_name, data1_absyn *res, + int xpelement, + data1_attset *attset) { data1_termlist **tp = *tpp; while(1) @@ -570,27 +565,24 @@ static int parse_termlists (data1_handle dh, data1_termlist ***tpp, nmem_malloc(data1_nmem_get(dh), sizeof(**tp)); (*tp)->next = 0; -#if NATTR - (*tp)->index_name = nmem_strdup(data1_nmem_get(dh), element_name); - if (*attname == '!' && xpelement) - (*tp)->index_name = 0; -#else - if (!xpelement) { - if (*attname == '!') + if (*attname == '!') + { + if (!xpelement && element_name) strcpy(attname, element_name); - } - if (!((*tp)->att = data1_getattbyname(dh, res->attset, attname))) - { - if ((!xpelement) || (*attname != '!')) { - yaz_log(YLOG_WARN, - "%s:%d: Couldn't find att '%s' in attset", - file, lineno, attname); - return -1; - } else { - (*tp)->att = 0; + else if (xpelement) + strcpy(attname, ZEBRA_XPATH_CDATA); + } + if (attset) + { + if (!data1_getattbyname(dh, attset, attname)) + { + yaz_log(YLOG_WARN, "Index '%s' not found in attset(s)", + attname); } - } -#endif + } + + (*tp)->index_name = nmem_strdup(data1_nmem_get(dh), attname); + assert (*(*tp)->index_name != '!'); if (r == 2 && (source = strchr(structure, ':'))) *source++ = '\0'; /* cut off structure .. */ else @@ -704,6 +696,8 @@ data1_absyn *data1_read_absyn (data1_handle dh, const char *file, { data1_sub_elements *cur_elements = NULL; data1_xpelement *cur_xpelement = NULL; + data1_attset *attset_list = data1_empty_attset(dh); + data1_attset_child **attset_childp = &attset_list->children; data1_absyn *res = 0; FILE *f; @@ -712,7 +706,6 @@ data1_absyn *data1_read_absyn (data1_handle dh, const char *file, data1_maptab **maptabp; data1_marctab **marcp; data1_termlist *all = 0; - data1_attset_child **attset_childp; data1_tagset **tagset_childp; struct data1_systag **systagsp; int level = 0; @@ -737,9 +730,6 @@ data1_absyn *data1_read_absyn (data1_handle dh, const char *file, systagsp = &res->systags; tagset_childp = &res->tagset; - res->attset = data1_empty_attset (dh); - attset_childp = &res->attset->children; - res->varset = 0; res->esetnames = 0; esetpp = &res->esetnames; @@ -750,7 +740,7 @@ data1_absyn *data1_read_absyn (data1_handle dh, const char *file, res->sub_elements = NULL; res->main_elements = NULL; res->xp_elements = NULL; - + while (f && (argc = read_absyn_line(f, &lineno, line, 512, argv, 50))) { char *cmd = *argv; @@ -859,9 +849,8 @@ data1_absyn *data1_read_absyn (data1_handle dh, const char *file, p = termlists; if (*p != '-') { - assert (res->attset); - - if (parse_termlists (dh, &tp, p, file, lineno, name, res, 0)) + if (parse_termlists (dh, &tp, p, file, lineno, name, res, 0, + attset_list)) { fclose (f); return 0; @@ -949,10 +938,8 @@ data1_absyn *data1_read_absyn (data1_handle dh, const char *file, p = termlists; if (*p != '-') { - assert (res->attset); - if (parse_termlists (dh, &tp, p, file, lineno, - xpath_expr, res, 1)) + xpath_expr, res, 1, attset_list)) { fclose (f); return 0; @@ -1015,7 +1002,8 @@ data1_absyn *data1_read_absyn (data1_handle dh, const char *file, file, lineno); continue; } - if (parse_termlists (dh, &tp, argv[1], file, lineno, 0, res, 0)) + if (parse_termlists (dh, &tp, argv[1], file, lineno, 0, res, 0, + attset_list)) { fclose (f); return 0; @@ -1051,31 +1039,27 @@ data1_absyn *data1_read_absyn (data1_handle dh, const char *file, } else if (!strcmp(cmd, "attset")) { -#if NATTR - yaz_log(YLOG_WARN, "%s:%d: attset obsolete", file, lineno); -#else - char *name; - data1_attset *attset; - - if (argc != 2) - { - yaz_log(YLOG_WARN, "%s:%d: Bad # of args to attset", - file, lineno); - continue; - } - name = argv[1]; - if (!(attset = data1_get_attset (dh, name))) - { - yaz_log(YLOG_WARN, "%s:%d: Couldn't find attset %s", - file, lineno, name); - continue; - } - *attset_childp = (data1_attset_child *) - nmem_malloc (data1_nmem_get(dh), sizeof(**attset_childp)); - (*attset_childp)->child = attset; - (*attset_childp)->next = 0; - attset_childp = &(*attset_childp)->next; -#endif + char *name; + data1_attset *attset; + + if (argc != 2) + { + yaz_log(YLOG_WARN, "%s:%d: Bad # of args to attset", + file, lineno); + continue; + } + name = argv[1]; + if (!(attset = data1_get_attset (dh, name))) + { + yaz_log(YLOG_WARN, "%s:%d: Couldn't find attset %s", + file, lineno, name); + continue; + } + *attset_childp = (data1_attset_child *) + nmem_malloc (data1_nmem_get(dh), sizeof(**attset_childp)); + (*attset_childp)->child = attset; + (*attset_childp)->next = 0; + attset_childp = &(*attset_childp)->next; } else if (!strcmp(cmd, "tagset")) { diff --git a/data1/d1_attset.c b/data1/d1_attset.c index bcd3484..40e1b04 100644 --- a/data1/d1_attset.c +++ b/data1/d1_attset.c @@ -1,5 +1,5 @@ -/* $Id: d1_attset.c,v 1.8 2006-05-10 08:13:18 adam Exp $ - Copyright (C) 1995-2005 +/* $Id: d1_attset.c,v 1.9 2006-05-19 13:49:33 adam Exp $ + Copyright (C) 1995-2006 Index Data ApS This file is part of the Zebra server. @@ -27,21 +27,21 @@ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA #include #include -data1_att *data1_getattbyname(data1_handle dh, data1_attset *s, char *name) +data1_att *data1_getattbyname(data1_handle dh, data1_attset *s, const char *name) { data1_att *r; data1_attset_child *c; /* scan local set */ for (r = s->atts; r; r = r->next) - if (!data1_matchstr(r->name, name)) - return r; + if (!data1_matchstr(r->name, name)) + return r; for (c = s->children; c; c = c->next) { - assert (c->child); - /* scan included sets */ - if ((r = data1_getattbyname (dh, c->child, name))) - return r; + assert (c->child); + /* scan included sets */ + if ((r = data1_getattbyname (dh, c->child, name))) + return r; } return 0; } diff --git a/include/Makefile.am b/include/Makefile.am index 6bb6f49..1357c2a 100644 --- a/include/Makefile.am +++ b/include/Makefile.am @@ -1,6 +1,6 @@ -# $Id: Makefile.am,v 1.22 2005-03-30 09:25:23 adam Exp $ +# $Id: Makefile.am,v 1.23 2006-05-19 13:49:33 adam Exp $ noinst_HEADERS = bset.h charmap.h \ direntz.h passwddb.h dfa.h zebra_xpath.h d1_absyn.h \ -rset.h dfaset.h sortidx.h zebra-lock.h +rset.h dfaset.h sortidx.h zebra-lock.h attrfind.h SUBDIRS = idzebra diff --git a/include/attrfind.h b/include/attrfind.h new file mode 100644 index 0000000..cea9e4d --- /dev/null +++ b/include/attrfind.h @@ -0,0 +1,59 @@ +/* $Id: attrfind.h,v 1.1 2006-05-19 13:49:34 adam Exp $ + Copyright (C) 2005-2006 + 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. +*/ + +#ifndef ATTRFIND_H +#define ATTRFIND_H + +#include +#include +#include + +YAZ_BEGIN_CDECL + +typedef struct { + int type; + int major; + int minor; + Z_AttributeElement **attributeList; + int num_attributes; +} AttrType; + +void attr_init_APT(AttrType *src, Z_AttributesPlusTerm *zapt, int type); + +void attr_init_AttrList(AttrType *src, Z_AttributeList *list, int type); + +int attr_find_ex(AttrType *src, oid_value *attributeSetP, + const char **string_value); +int attr_find(AttrType *src, oid_value *attributeSetP); + + +YAZ_END_CDECL + +#endif +/* + * Local variables: + * c-basic-offset: 4 + * indent-tabs-mode: nil + * End: + * vim: shiftwidth=4 tabstop=8 expandtab + */ + diff --git a/include/d1_absyn.h b/include/d1_absyn.h index 00804f1..f7aa6c9 100644 --- a/include/d1_absyn.h +++ b/include/d1_absyn.h @@ -1,5 +1,5 @@ -/* $Id: d1_absyn.h,v 1.3 2006-05-10 08:13:18 adam Exp $ - Copyright (C) 1995-2005 +/* $Id: d1_absyn.h,v 1.4 2006-05-19 13:49:34 adam Exp $ + Copyright (C) 1995-2006 Index Data ApS This file is part of the Zebra server. @@ -46,7 +46,6 @@ struct data1_absyn char *name; oid_value reference; data1_tagset *tagset; - data1_attset *attset; data1_varset *varset; data1_esetname *esetnames; data1_maptab *maptabs; diff --git a/include/idzebra/data1.h b/include/idzebra/data1.h index eda8ea4..4151727 100644 --- a/include/idzebra/data1.h +++ b/include/idzebra/data1.h @@ -1,5 +1,5 @@ -/* $Id: data1.h,v 1.12 2006-05-10 08:13:20 adam Exp $ - Copyright (C) 1995-2005 +/* $Id: data1.h,v 1.13 2006-05-19 13:49:34 adam Exp $ + Copyright (C) 1995-2006 Index Data ApS This file is part of the Zebra server. @@ -88,7 +88,7 @@ struct data1_attset typedef struct data1_handle_info *data1_handle; YAZ_EXPORT data1_att *data1_getattbyname(data1_handle dh, data1_attset *s, - char *name); + const char *name); YAZ_EXPORT data1_attset *data1_read_attset(data1_handle dh, const char *file); YAZ_EXPORT data1_attset *data1_empty_attset(data1_handle dh); @@ -245,11 +245,7 @@ struct data1_tagset typedef struct data1_termlist { -#if NATTR char *index_name; -#else - data1_att *att; -#endif char *structure; char *source; struct data1_termlist *next; diff --git a/include/idzebra/recctrl.h b/include/idzebra/recctrl.h index fddb08a..4ad6157 100644 --- a/include/idzebra/recctrl.h +++ b/include/idzebra/recctrl.h @@ -1,4 +1,4 @@ -/* $Id: recctrl.h,v 1.19 2006-05-10 08:13:20 adam Exp $ +/* $Id: recctrl.h,v 1.20 2006-05-19 13:49:34 adam Exp $ Copyright (C) 1995-2006 Index Data ApS @@ -34,20 +34,23 @@ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA YAZ_BEGIN_CDECL -#if NATTR +/* 1 */ #define ZEBRA_XPATH_ELM_BEGIN "_XPATH_BEGIN" + +/* 2 */ #define ZEBRA_XPATH_ELM_END "_XPATH_END" + +/* 1016 */ #define ZEBRA_XPATH_CDATA "_XPATH_CDATA" -#define ZEBRA_XPATH_ATTR "_XPATH_ATTR" -#endif + +/* 3 */ +#define ZEBRA_XPATH_ATTR_NAME "_XPATH_ATTR_NAME" + +/* 1015 */ +#define ZEBRA_XPATH_ATTR_CDATA "_XPATH_ATTR_CDATA" /* single word entity */ typedef struct { -#if NATTR -#else - int attrSet; - int attrUse; -#endif unsigned index_type; const char *index_name; const char *term_buf; diff --git a/include/idzebra/util.h b/include/idzebra/util.h index b0f54f3..0744fea 100644 --- a/include/idzebra/util.h +++ b/include/idzebra/util.h @@ -1,4 +1,4 @@ -/* $Id: util.h,v 1.6 2006-05-10 08:13:20 adam Exp $ +/* $Id: util.h,v 1.7 2006-05-19 13:49:34 adam Exp $ Copyright (C) 1995-2005 Index Data ApS @@ -84,9 +84,6 @@ YAZ_END_CDECL #define CAST_ZINT_TO_INT(x) (int)(x) #define CAST_ZINT_TO_DOUBLE(x) (double)(x) -/* NATTR=1 for string-attribute architecture, =0 for set+use . */ -#define NATTR 0 - #endif /* * Local variables: diff --git a/index/attribute.c b/index/attribute.c index a637f99..10b1e6d 100644 --- a/index/attribute.c +++ b/index/attribute.c @@ -1,5 +1,5 @@ -/* $Id: attribute.c,v 1.22 2006-05-10 08:13:20 adam Exp $ - Copyright (C) 1995-2005 +/* $Id: attribute.c,v 1.23 2006-05-19 13:49:34 adam Exp $ + Copyright (C) 1995-2006 Index Data ApS This file is part of the Zebra server. @@ -23,30 +23,30 @@ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA #include #include +#include #include #include +#include #include "index.h" -static data1_att *getatt(data1_attset *p, int att, const char *sattr) +static data1_att *getatt(data1_attset *p, int att) { data1_att *a; data1_attset_child *c; /* scan local set */ for (a = p->atts; a; a = a->next) - if (sattr && !yaz_matchstr(sattr, a->name)) - return a; - else if (a->value == att) + if (a->value == att) return a; /* scan included sets */ for (c = p->children; c; c = c->next) - if ((a = getatt(c->child, att, sattr))) + if ((a = getatt(c->child, att))) return a; return 0; } -int att_getentbyatt(ZebraHandle zi, attent *res, oid_value set, int att, - const char *sattr) +static int att_getentbyatt(ZebraHandle zi, oid_value set, int att, + const char **name) { data1_att *r; data1_attset *p; @@ -57,18 +57,120 @@ int att_getentbyatt(ZebraHandle zi, attent *res, oid_value set, int att, p = data1_attset_search_id (zi->reg->dh, set); } if (!p) /* set undefined */ - { - if (sattr) - return -1; /* return bad string attribute */ - else - return -2; /* return bad set */ - } - if (!(r = getatt(p, att, sattr))) + return -2; + if (!(r = getatt(p, att))) return -1; - res->attset_ordinal = r->parent->reference; - res->local_attributes = r->locals; + *name = r->name; return 0; } + + +ZEBRA_RES zebra_attr_list_get_ord(ZebraHandle zh, + Z_AttributeList *attr_list, + int index_type, + oid_value curAttributeSet, + int *ord) +{ + int use_value = -1; + const char *use_string = 0; + AttrType use; + + attr_init_AttrList(&use, attr_list, 1); + use_value = attr_find_ex(&use, &curAttributeSet, &use_string); + + if (use_value < 0) + { + if (!use_string) + use_string = "any"; + } + else + { + /* we have a use attribute and attribute set */ + int r; + + r = att_getentbyatt(zh, curAttributeSet, use_value, &use_string); + if (r == -2) + { + zebra_setError_zint(zh, YAZ_BIB1_UNSUPP_ATTRIBUTE_SET, 0); + return ZEBRA_FAIL; + } + if (r == -1) + { + zebra_setError_zint(zh, YAZ_BIB1_UNSUPP_USE_ATTRIBUTE, use_value); + return ZEBRA_FAIL; + } + } + if (!use_string) + { + zebra_setError(zh, YAZ_BIB1_UNSUPP_USE_ATTRIBUTE, 0); + return ZEBRA_FAIL; + } + *ord = zebraExplain_lookup_attr_str(zh->reg->zei, index_type, use_string); + if (*ord == -1) + { + if (use_value < 0) + zebra_setError(zh, YAZ_BIB1_UNSUPP_USE_ATTRIBUTE, use_string); + else + zebra_setError_zint(zh, YAZ_BIB1_UNSUPP_USE_ATTRIBUTE, use_value); + return ZEBRA_FAIL; + } + return ZEBRA_OK; +} + +ZEBRA_RES zebra_apt_get_ord(ZebraHandle zh, + Z_AttributesPlusTerm *zapt, + int index_type, + const char *xpath_use, + oid_value curAttributeSet, + int *ord) +{ + if (!xpath_use) + return zebra_attr_list_get_ord(zh, zapt->attributes, + index_type, curAttributeSet, ord); + else + { + *ord = zebraExplain_lookup_attr_str(zh->reg->zei, index_type, + xpath_use); + if (*ord == -1) + { + yaz_log(YLOG_LOG, "zebra_apt_get_ord FAILED xpath=%s index_type=%c", + xpath_use, index_type); + zebra_setError(zh, YAZ_BIB1_UNSUPP_USE_ATTRIBUTE, 0); + return ZEBRA_FAIL; + } + else + { + yaz_log(YLOG_LOG, "zebra_apt_get_ord OK xpath=%s index_type=%c", + xpath_use, index_type); + } + return ZEBRA_OK; + } +} + +ZEBRA_RES zebra_sort_get_ord(ZebraHandle zh, + Z_SortAttributes *sortAttributes, + int *ord, + int *numerical) +{ + AttrType structure; + int structure_value; + attr_init_AttrList(&structure, sortAttributes->list, 4); + + *numerical = 0; + structure_value = attr_find(&structure, 0); + if (structure_value == 109) + *numerical = 1; + + if (zebra_attr_list_get_ord(zh, sortAttributes->list, + 's', VAL_BIB1, ord)== ZEBRA_OK) + return ZEBRA_OK; + if (zebra_attr_list_get_ord(zh, sortAttributes->list, + 'S', VAL_BIB1, ord)== ZEBRA_OK) + return ZEBRA_OK; + return ZEBRA_FAIL; +} + + /* * Local variables: * c-basic-offset: 4 diff --git a/index/extract.c b/index/extract.c index 86f868b..75bc9e0 100644 --- a/index/extract.c +++ b/index/extract.c @@ -1,4 +1,4 @@ -/* $Id: extract.c,v 1.214 2006-05-18 12:03:05 adam Exp $ +/* $Id: extract.c,v 1.215 2006-05-19 13:49:34 adam Exp $ Copyright (C) 1995-2006 Index Data ApS @@ -96,12 +96,7 @@ static void extract_init (struct recExtractCtrl *p, RecWord *w) { w->zebra_maps = p->zebra_maps; w->seqno = 1; -#if NATTR -#else - w->attrSet = VAL_BIB1; - w->attrUse = 1016; -#endif - w->index_name = 0; + w->index_name = "any"; w->index_type = 'w'; w->extractCtrl = p; w->record_id = 0; @@ -110,17 +105,22 @@ static void extract_init (struct recExtractCtrl *p, RecWord *w) static void searchRecordKey(ZebraHandle zh, zebra_rec_keys_t reckeys, - int attrSetS, int attrUseS, + const char *index_name, const char **ws, int ws_length) { int i; - int ch; + int ch = -1; for (i = 0; ireg->zei, - attrSetS, attrUseS); + if (ch < 0) + ch = zebraExplain_lookup_attr_str(zh->reg->zei, '0', index_name); + if (ch < 0) + ch = zebraExplain_lookup_attr_str(zh->reg->zei, 'p', index_name); + if (ch < 0) + ch = zebraExplain_lookup_attr_str(zh->reg->zei, 'w', index_name); + if (ch < 0) return ; @@ -236,9 +236,7 @@ static char *fileMatchStr (ZebraHandle zh, { const char *ws[32]; char attset_str[64], attname_str[64]; - data1_attset *attset; int i; - int attSet = 1, attUse = 1; int first = 1; for (s++; strchr(FILE_MATCH_BLANK, *s); s++) @@ -251,7 +249,9 @@ static char *fileMatchStr (ZebraHandle zh, for (; strchr(FILE_MATCH_BLANK, *s); s++) ; - if (*s == ',') + if (*s != ',') + strcpy(attname_str, attset_str); + else { for (s++; strchr(FILE_MATCH_BLANK, *s); s++) ; @@ -261,18 +261,8 @@ static char *fileMatchStr (ZebraHandle zh, attname_str[i++] = *s; attname_str[i] = '\0'; } - - if ((attset = data1_get_attset (zh->reg->dh, attset_str))) - { - data1_att *att; - attSet = attset->reference; - att = data1_getattbyname(zh->reg->dh, attset, attname_str); - if (att) - attUse = att->value; - else - attUse = atoi (attname_str); - } - searchRecordKey (zh, reckeys, attSet, attUse, ws, 32); + + searchRecordKey (zh, reckeys, attname_str, ws, 32); if (*s != ')') { @@ -1621,24 +1611,13 @@ void extract_add_index_string(RecWord *p, const char *str, int length) ZebraExplainInfo zei = zh->reg->zei; int ch; - if (p->index_name) - { - ch = zebraExplain_lookup_attr_str(zei, p->index_type, p->index_name); - if (ch < 0) - ch = zebraExplain_add_attr_str(zei, p->index_type, p->index_name); - } - else - { -#if NATTR - return; -#else - ch = zebraExplain_lookup_attr_su(zei, p->index_type, - p->attrSet, p->attrUse); - if (ch < 0) - ch = zebraExplain_add_attr_su(zei, p->index_type, - p->attrSet, p->attrUse); -#endif - } + if (!p->index_name) + return; + + ch = zebraExplain_lookup_attr_str(zei, p->index_type, p->index_name); + if (ch < 0) + ch = zebraExplain_add_attr_str(zei, p->index_type, p->index_name); + key.len = 4; key.mem[0] = ch; key.mem[1] = p->record_id; @@ -1673,24 +1652,12 @@ static void extract_add_sort_string(RecWord *p, const char *str, int length) ZebraExplainInfo zei = zh->reg->zei; int ch; - if (p->index_name) - { - ch = zebraExplain_lookup_attr_str(zei, p->index_type, p->index_name); - if (ch < 0) - ch = zebraExplain_add_attr_str(zei, p->index_type, p->index_name); - } - else - { -#if NATTR - return; -#else - ch = zebraExplain_lookup_attr_su(zei, p->index_type, - VAL_IDXPATH, p->attrUse); - if (ch < 0) - ch = zebraExplain_add_attr_su(zei, p->index_type, - VAL_IDXPATH, p->attrUse); -#endif - } + if (!p->index_name) + return; + + ch = zebraExplain_lookup_attr_str(zei, p->index_type, p->index_name); + if (ch < 0) + ch = zebraExplain_add_attr_str(zei, p->index_type, p->index_name); key.len = 4; key.mem[0] = ch; key.mem[1] = p->record_id; diff --git a/index/index.h b/index/index.h index 6ffa33e..d06d36d 100644 --- a/index/index.h +++ b/index/index.h @@ -1,4 +1,4 @@ -/* $Id: index.h,v 1.161 2006-05-18 12:03:05 adam Exp $ +/* $Id: index.h,v 1.162 2006-05-19 13:49:34 adam Exp $ Copyright (C) 1995-2005 Index Data ApS @@ -363,14 +363,13 @@ int zebra_server_lock (ZebraService zh, int lockCommit); void zebra_server_unlock (ZebraService zh, int commitPhase); int zebra_server_lock_get_state (ZebraService zh, time_t *timep); +#if 1 typedef struct attent { int attset_ordinal; data1_local_attribute *local_attributes; } attent; - -int att_getentbyatt(ZebraHandle zh, attent *res, oid_value set, int att, - const char *sattr); +#endif int zebra_record_fetch (ZebraHandle zh, SYSNO sysno, int score, zebra_snippets *hit_snippet, ODR stream, @@ -468,6 +467,24 @@ ZEBRA_RES zebra_get_hit_vector(ZebraHandle zh, const char *setname, zint sysno); void zebra_term_untrans(ZebraHandle zh, int reg_type, char *dst, const char *src); +ZEBRA_RES zebra_apt_get_ord(ZebraHandle zh, + Z_AttributesPlusTerm *zapt, + int index_type, + const char *xpath_use, + oid_value curAttributeSet, + int *ord); + +ZEBRA_RES zebra_attr_list_get_ord(ZebraHandle zh, + Z_AttributeList *attr_list, + int index_type, + oid_value curAttributeSet, + int *ord); + +ZEBRA_RES zebra_sort_get_ord(ZebraHandle zh, + Z_SortAttributes *sortAttributes, + int *ord, + int *numerical); + YAZ_END_CDECL #endif diff --git a/index/zebrash.c b/index/zebrash.c index e1f6ede..978ec6f 100644 --- a/index/zebrash.c +++ b/index/zebrash.c @@ -1,4 +1,4 @@ -/* $Id: zebrash.c,v 1.37 2006-05-10 08:13:23 adam Exp $ +/* $Id: zebrash.c,v 1.38 2006-05-19 13:49:34 adam Exp $ Copyright (C) 1995-2005 Index Data ApS @@ -365,7 +365,7 @@ static int cmd_record_insert( char *args[], WRBUF outbuff) 0); if (0==rc) { - wrbuf_printf(outbuff,"ok sysno=%d\n",sysno); + wrbuf_printf(outbuff,"ok sysno=" ZINT_FORMAT "\n",sysno); } return rc; } diff --git a/index/zinfo.c b/index/zinfo.c index f19a068..3d2aee8 100644 --- a/index/zinfo.c +++ b/index/zinfo.c @@ -1,4 +1,4 @@ -/* $Id: zinfo.c,v 1.63 2006-05-18 12:03:05 adam Exp $ +/* $Id: zinfo.c,v 1.64 2006-05-19 13:49:34 adam Exp $ Copyright (C) 1995-2006 Index Data ApS @@ -1491,7 +1491,7 @@ int zebraExplain_lookup_attr_str(ZebraExplainInfo zei, int index_type, *zsui; zsui = &(*zsui)->next) if ((*zsui)->info.index_type == index_type && (*zsui)->info.which == ZEB_SU_STR - && !strcmp((*zsui)->info.u.str, str)) + && !yaz_matchstr((*zsui)->info.u.str, str)) { struct zebSUInfoB *zsui_this = *zsui; @@ -1700,6 +1700,7 @@ int zebraExplain_add_attr_su(ZebraExplainInfo zei, int index_type, { struct zebSUInfoB *zsui = zebraExplain_add_sui_info(zei, index_type); + yaz_log(YLOG_WARN, "add_attr_su"); zebraExplain_addAttributeSet (zei, set); zsui->info.which = ZEB_SU_SET_USE; zsui->info.u.su.set = set; diff --git a/index/zrpn.c b/index/zrpn.c index f7ff740..f5da395 100644 --- a/index/zrpn.c +++ b/index/zrpn.c @@ -1,4 +1,4 @@ -/* $Id: zrpn.c,v 1.213 2006-05-17 17:46:45 adam Exp $ +/* $Id: zrpn.c,v 1.214 2006-05-19 13:49:34 adam Exp $ Copyright (C) 1995-2006 Index Data ApS @@ -33,7 +33,7 @@ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA #include #include "index.h" #include - +#include #include #include @@ -43,14 +43,6 @@ struct rpn_char_map_info int reg_type; }; -typedef struct -{ - int type; - int major; - int minor; - Z_AttributesPlusTerm *zapt; -} AttrType; - static int log_level_set = 0; static int log_level_rpn = 0; @@ -81,84 +73,6 @@ static void rpn_char_map_prepare(struct zebra_register *reg, int reg_type, dict_grep_cmap(reg->dict, map_info, rpn_char_map_handler); } -static int attr_find_ex(AttrType *src, oid_value *attributeSetP, - const char **string_value) -{ - int num_attributes; - - num_attributes = src->zapt->attributes->num_attributes; - while (src->major < num_attributes) - { - Z_AttributeElement *element; - - element = src->zapt->attributes->attributes[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) - break; - if (element->attributeSet && attributeSetP) - { - oident *attrset; - - attrset = oid_getentbyoid(element->attributeSet); - *attributeSetP = attrset->value; - } - if (element->value.complex->list[src->minor]->which == - Z_StringOrNumeric_numeric) - { - ++(src->minor); - return - *element->value.complex->list[src->minor-1]->u.numeric; - } - else if (element->value.complex->list[src->minor]->which == - Z_StringOrNumeric_string) - { - if (!string_value) - break; - ++(src->minor); - *string_value = - element->value.complex->list[src->minor-1]->u.string; - return -2; - } - else - break; - default: - assert(0); - } - } - ++(src->major); - } - return -1; -} - -static int attr_find(AttrType *src, oid_value *attributeSetP) -{ - return attr_find_ex(src, attributeSetP, 0); -} - -static void attr_init(AttrType *src, Z_AttributesPlusTerm *zapt, - int type) -{ - src->zapt = zapt; - src->type = type; - src->major = 0; - src->minor = 0; -} - #define TERM_COUNT struct grep_info { @@ -820,7 +734,7 @@ static int string_relation(ZebraHandle zh, Z_AttributesPlusTerm *zapt, char *term_tmp = term_dict + strlen(term_dict); char term_component[2*IT_MAX_WORD+20]; - attr_init(&relation, zapt, 2); + attr_init_APT(&relation, zapt, 2); relation_value = attr_find(&relation, NULL); *error_code = 0; @@ -994,7 +908,8 @@ static ZEBRA_RES string_term(ZebraHandle zh, Z_AttributesPlusTerm *zapt, struct grep_info *grep_info, int reg_type, int complete_flag, int num_bases, char **basenames, - char *term_dst, int xpath_use, + char *term_dst, + const char *xpath_use, struct ord_list **ol); static ZEBRA_RES term_limits_APT(ZebraHandle zh, @@ -1007,10 +922,10 @@ static ZEBRA_RES term_limits_APT(ZebraHandle zh, AttrType hits_limit_attr; int term_ref_id_int; - attr_init(&hits_limit_attr, zapt, 9); + attr_init_APT(&hits_limit_attr, zapt, 9); *hits_limit_value = attr_find(&hits_limit_attr, NULL); - attr_init(&term_ref_id_attr, zapt, 10); + attr_init_APT(&term_ref_id_attr, zapt, 10); term_ref_id_int = attr_find_ex(&term_ref_id_attr, NULL, term_ref_id_str); if (term_ref_id_int >= 0) { @@ -1052,7 +967,8 @@ static ZEBRA_RES term_trunc(ZebraHandle zh, int reg_type, int complete_flag, int num_bases, char **basenames, char *term_dst, - const char *rank_type, int xpath_use, + const char *rank_type, + const char *xpath_use, NMEM rset_nmem, RSET *rset, struct rset_key_control *kc) @@ -1091,16 +1007,14 @@ static ZEBRA_RES string_term(ZebraHandle zh, Z_AttributesPlusTerm *zapt, struct grep_info *grep_info, int reg_type, int complete_flag, int num_bases, char **basenames, - char *term_dst, int xpath_use, + char *term_dst, + const char *xpath_use, struct ord_list **ol) { char term_dict[2*IT_MAX_WORD+4000]; int j, r, base_no; AttrType truncation; int truncation_value; - AttrType use; - int use_value; - const char *use_string = 0; oid_value curAttributeSet = attributeSet; const char *termp; struct rpn_char_map_info rcmi; @@ -1111,26 +1025,25 @@ static ZEBRA_RES string_term(ZebraHandle zh, Z_AttributesPlusTerm *zapt, *ol = ord_list_create(stream); rpn_char_map_prepare (zh->reg, reg_type, &rcmi); - attr_init(&use, zapt, 1); - use_value = attr_find_ex(&use, &curAttributeSet, &use_string); - yaz_log(log_level_rpn, "string_term, use value %d", use_value); - attr_init(&truncation, zapt, 5); + attr_init_APT(&truncation, zapt, 5); truncation_value = attr_find(&truncation, NULL); yaz_log(log_level_rpn, "truncation value %d", truncation_value); - if (use_value == -1) /* no attribute - assumy "any" */ - use_value = 1016; for (base_no = 0; base_no < num_bases; base_no++) { int ord = -1; int attr_ok = 0; int regex_range = 0; int init_pos = 0; +#if 0 attent attp; data1_local_attribute id_xpath_attr; data1_local_attribute *local_attr; +#endif int max_pos, prefix_len = 0; int relation_error; + char ord_buf[32]; + int ord_len, i; termp = *term_sub; @@ -1140,111 +1053,29 @@ static ZEBRA_RES string_term(ZebraHandle zh, Z_AttributesPlusTerm *zapt, basenames[base_no]); return ZEBRA_FAIL; } - if (xpath_use > 0 && use_value == -2) - { - /* xpath mode and we have a string attribute */ - attp.local_attributes = &id_xpath_attr; - attp.attset_ordinal = VAL_IDXPATH; - id_xpath_attr.next = 0; - - use_value = xpath_use; /* xpath_use as use-attribute now */ - id_xpath_attr.local = use_value; - } - else if (curAttributeSet == VAL_IDXPATH && use_value >= 0) + + if (zebra_apt_get_ord(zh, zapt, reg_type, xpath_use, + curAttributeSet, &ord) + != ZEBRA_OK) { - /* X-Path attribute, use numeric value directly */ - attp.local_attributes = &id_xpath_attr; - attp.attset_ordinal = VAL_IDXPATH; - id_xpath_attr.next = 0; - id_xpath_attr.local = use_value; + break; } - else if (use_string && - (ord = zebraExplain_lookup_attr_str(zh->reg->zei, - reg_type, - use_string)) >= 0) - { - /* we have a match for a raw string attribute */ - char ord_buf[32]; - int i, ord_len; - - if (prefix_len) - term_dict[prefix_len++] = '|'; - else - term_dict[prefix_len++] = '('; - - ord_len = key_SU_encode (ord, ord_buf); - for (i = 0; inext) - { - char ord_buf[32]; - int i, ord_len; - - ord = zebraExplain_lookup_attr_su(zh->reg->zei, - reg_type, - attp.attset_ordinal, - local_attr->local); - if (ord < 0) - continue; - *ol = ord_list_append(stream, *ol, ord); - if (prefix_len) - term_dict[prefix_len++] = '|'; - else - term_dict[prefix_len++] = '('; - - ord_len = key_SU_encode (ord, ord_buf); - for (i = 0; i init_pos) - init_pos = ord_len; - } + } + if (ord_len > init_pos) + init_pos = ord_len; + bases_ok++; if (prefix_len) attr_ok = 1; @@ -1512,7 +1343,7 @@ static ZEBRA_RES grep_info_prepare(ZebraHandle zh, if (!zapt) return ZEBRA_OK; - attr_init(&termset, zapt, 8); + attr_init_APT(&termset, zapt, 8); termset_value_numeric = attr_find_ex(&termset, NULL, &termset_value_string); if (termset_value_numeric != -1) @@ -1561,7 +1392,8 @@ static ZEBRA_RES term_list_trunc(ZebraHandle zh, oid_value attributeSet, NMEM stream, int reg_type, int complete_flag, - const char *rank_type, int xpath_use, + const char *rank_type, + const char *xpath_use, int num_bases, char **basenames, NMEM rset_nmem, RSET **result_sets, int *num_result_sets, @@ -1627,7 +1459,8 @@ static ZEBRA_RES rpn_search_APT_phrase(ZebraHandle zh, oid_value attributeSet, NMEM stream, int reg_type, int complete_flag, - const char *rank_type, int xpath_use, + const char *rank_type, + const char *xpath_use, int num_bases, char **basenames, NMEM rset_nmem, RSET *rset, @@ -1665,7 +1498,7 @@ static ZEBRA_RES rpn_search_APT_or_list(ZebraHandle zh, NMEM stream, int reg_type, int complete_flag, const char *rank_type, - int xpath_use, + const char *xpath_use, int num_bases, char **basenames, NMEM rset_nmem, RSET *rset, @@ -1701,7 +1534,7 @@ static ZEBRA_RES rpn_search_APT_and_list(ZebraHandle zh, NMEM stream, int reg_type, int complete_flag, const char *rank_type, - int xpath_use, + const char *xpath_use, int num_bases, char **basenames, NMEM rset_nmem, RSET *rset, @@ -1748,7 +1581,7 @@ static int numeric_relation(ZebraHandle zh, Z_AttributesPlusTerm *zapt, char *term_tmp = term_dict + strlen(term_dict); *error_code = 0; - attr_init(&relation, zapt, 2); + attr_init_APT(&relation, zapt, 2); relation_value = attr_find(&relation, NULL); yaz_log(log_level_rpn, "numeric relation value=%d", relation_value); @@ -1799,13 +1632,12 @@ static ZEBRA_RES numeric_term(ZebraHandle zh, Z_AttributesPlusTerm *zapt, struct grep_info *grep_info, int reg_type, int complete_flag, int num_bases, char **basenames, - char *term_dst, int xpath_use, NMEM stream) + char *term_dst, + const char *xpath_use, + NMEM stream) { char term_dict[2*IT_MAX_WORD+2]; - int r, base_no; - AttrType use; - int use_value; - const char *use_string = 0; + int base_no; oid_value curAttributeSet = attributeSet; const char *termp; struct rpn_char_map_info rcmi; @@ -1813,93 +1645,40 @@ static ZEBRA_RES numeric_term(ZebraHandle zh, Z_AttributesPlusTerm *zapt, int bases_ok = 0; /* no of databases with OK attribute */ rpn_char_map_prepare (zh->reg, reg_type, &rcmi); - attr_init(&use, zapt, 1); - use_value = attr_find_ex(&use, &curAttributeSet, &use_string); - - if (use_value == -1) - use_value = 1016; for (base_no = 0; base_no < num_bases; base_no++) { - attent attp; - data1_local_attribute id_xpath_attr; - data1_local_attribute *local_attr; int max_pos, prefix_len = 0; int relation_error = 0; + int ord, ord_len, i; + char ord_buf[32]; termp = *term_sub; - if (use_value == -2) /* string attribute (assume IDXPATH/any) */ - { - use_value = xpath_use; - attp.local_attributes = &id_xpath_attr; - attp.attset_ordinal = VAL_IDXPATH; - id_xpath_attr.next = 0; - id_xpath_attr.local = use_value; - } - else if (curAttributeSet == VAL_IDXPATH) - { - attp.local_attributes = &id_xpath_attr; - attp.attset_ordinal = VAL_IDXPATH; - id_xpath_attr.next = 0; - id_xpath_attr.local = use_value; - } - else - { - if ((r = att_getentbyatt (zh, &attp, curAttributeSet, use_value, - use_string))) - { - yaz_log(YLOG_DEBUG, "att_getentbyatt fail. set=%d use=%d r=%d", - curAttributeSet, use_value, r); - if (r == -1) - { - if (use_string) - zebra_setError(zh, - YAZ_BIB1_UNSUPP_USE_ATTRIBUTE, - use_string); - else - zebra_setError_zint(zh, YAZ_BIB1_UNSUPP_USE_ATTRIBUTE, - use_value); - } - else - zebra_setError(zh, YAZ_BIB1_UNSUPP_ATTRIBUTE_SET, 0); - continue; - } - } + if (zebraExplain_curDatabase (zh->reg->zei, basenames[base_no])) { zebra_setError(zh, YAZ_BIB1_DATABASE_UNAVAILABLE, basenames[base_no]); return ZEBRA_FAIL; } - for (local_attr = attp.local_attributes; local_attr; - local_attr = local_attr->next) - { - int ord; - char ord_buf[32]; - int i, ord_len; - - ord = zebraExplain_lookup_attr_su(zh->reg->zei, - reg_type, - attp.attset_ordinal, - local_attr->local); - if (ord < 0) - continue; - if (prefix_len) - term_dict[prefix_len++] = '|'; - else - term_dict[prefix_len++] = '('; - ord_len = key_SU_encode (ord, ord_buf); - for (i = 0; ispecs) @@ -2119,7 +1899,7 @@ static int parse_xpath(ZebraHandle zh, Z_AttributesPlusTerm *zapt, AttrType use; const char *use_string = 0; - attr_init(&use, zapt, 1); + attr_init_APT(&use, zapt, 1); attr_find_ex(&use, &curAttributeSet, &use_string); if (!use_string || *use_string != '/') @@ -2131,7 +1911,8 @@ static int parse_xpath(ZebraHandle zh, Z_AttributesPlusTerm *zapt, static RSET xpath_trunc(ZebraHandle zh, NMEM stream, - int reg_type, const char *term, int use, + int reg_type, const char *term, + const char *xpath_use, oid_value curAttributeSet, NMEM rset_nmem, struct rset_key_control *kc) { @@ -2140,8 +1921,8 @@ static RSET xpath_trunc(ZebraHandle zh, NMEM stream, char term_dict[2048]; char ord_buf[32]; int prefix_len = 0; - int ord = zebraExplain_lookup_attr_su(zh->reg->zei, reg_type, - curAttributeSet, use); + int ord = zebraExplain_lookup_attr_str(zh->reg->zei, reg_type, + xpath_use); int ord_len, i, r, max_pos; int term_type = Z_Term_characterString; const char *flags = "void"; @@ -2300,7 +2081,7 @@ ZEBRA_RES rpn_search_xpath(ZebraHandle zh, } wrbuf_puts(wbuf, ""); rset_attr = xpath_trunc( - zh, stream, '0', wrbuf_buf(wbuf), 3, + zh, stream, '0', wrbuf_buf(wbuf), ZEBRA_XPATH_ATTR_NAME, curAttributeSet, rset_nmem, kc); wrbuf_free(wbuf, 1); } @@ -2313,10 +2094,16 @@ ZEBRA_RES rpn_search_xpath(ZebraHandle zh, if (strlen(xpath_rev)) { rset_start_tag = xpath_trunc(zh, stream, '0', - xpath_rev, 1, curAttributeSet, rset_nmem, kc); + xpath_rev, + ZEBRA_XPATH_ELM_BEGIN, + curAttributeSet, + rset_nmem, kc); rset_end_tag = xpath_trunc(zh, stream, '0', - xpath_rev, 2, curAttributeSet, rset_nmem, kc); + xpath_rev, + ZEBRA_XPATH_ELM_END, + curAttributeSet, + rset_nmem, kc); rset = rsbetween_create(rset_nmem, kc, kc->scope, rset_start_tag, rset, @@ -2345,7 +2132,7 @@ static ZEBRA_RES rpn_search_APT(ZebraHandle zh, Z_AttributesPlusTerm *zapt, int sort_flag; char termz[IT_MAX_WORD+1]; int xpath_len; - int xpath_use = 0; + const char *xpath_use = 0; struct xpath_location_step xpath[10]; if (!log_level_set) @@ -2371,9 +2158,10 @@ static ZEBRA_RES rpn_search_APT(ZebraHandle zh, Z_AttributesPlusTerm *zapt, xpath_len = parse_xpath(zh, zapt, attributeSet, xpath, 10, stream); if (xpath_len >= 0) { - xpath_use = 1016; /* searching for element by default */ if (xpath[xpath_len-1].part[0] == '@') - xpath_use = 1015; /* last step an attribute .. */ + xpath_use = ZEBRA_XPATH_ATTR_CDATA; /* last step is attribute */ + else + xpath_use = ZEBRA_XPATH_CDATA; /* searching for cdata */ } /* search using one of the various search type strategies @@ -2772,7 +2560,7 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, int errCode = 0; /* err code (if any is not OK) */ char *errString = 0; /* addinfo */ - unsigned reg_id; + unsigned index_type; char *search_type = NULL; char rank_type[128]; int complete_flag; @@ -2791,7 +2579,7 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, AttrType termset; int termset_value_numeric; const char *termset_value_string; - attr_init(&termset, zapt, 8); + attr_init_APT(&termset, zapt, 8); termset_value_numeric = attr_find_ex(&termset, NULL, &termset_value_string); if (termset_value_numeric != -1) @@ -2815,10 +2603,10 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, yaz_log(YLOG_DEBUG, "position = %d, num = %d set=%d", pos, num, attributeset); - attr_init(&use, zapt, 1); + attr_init_APT(&use, zapt, 1); use_value = attr_find_ex(&use, &attributeset, &use_string); - if (zebra_maps_attr(zh->reg->zebra_maps, zapt, ®_id, &search_type, + if (zebra_maps_attr(zh->reg->zebra_maps, zapt, &index_type, &search_type, rank_type, &complete_flag, &sort_flag)) { *num_entries = 0; @@ -2831,8 +2619,6 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, use_value = 1016; for (base_no = 0; base_no < num_bases && ord_no < 32; base_no++) { - data1_local_attribute *local_attr; - attent attp; int ord; if (zebraExplain_curDatabase (zh->reg->zei, basenames[base_no])) @@ -2843,51 +2629,12 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, return ZEBRA_FAIL; } - if (use_string && - (ord = zebraExplain_lookup_attr_str(zh->reg->zei, reg_id, - use_string)) >= 0) - { - /* we have a match for a raw string attribute */ - if (ord > 0) - ords[ord_no++] = ord; - attp.local_attributes = 0; /* no more attributes */ - } - else - { - int r; - - if ((r = att_getentbyatt (zh, &attp, attributeset, use_value, - use_string))) - { - yaz_log(YLOG_DEBUG, "att_getentbyatt fail. set=%d use=%d", - attributeset, use_value); - if (r == -1) - { - errCode = YAZ_BIB1_UNSUPP_USE_ATTRIBUTE; - if (use_string) - zebra_setError(zh, YAZ_BIB1_UNSUPP_USE_ATTRIBUTE, - use_string); - else - zebra_setError_zint(zh, YAZ_BIB1_UNSUPP_USE_ATTRIBUTE, - use_value); - } - else - { - zebra_setError(zh, YAZ_BIB1_UNSUPP_ATTRIBUTE_SET, 0); - } - continue; - } - } - bases_ok++; - for (local_attr = attp.local_attributes; local_attr && ord_no < 32; - local_attr = local_attr->next) - { - ord = zebraExplain_lookup_attr_su(zh->reg->zei, reg_id, - attp.attset_ordinal, - local_attr->local); - if (ord > 0) - ords[ord_no++] = ord; - } + if (zebra_apt_get_ord(zh, zapt, index_type, 0, attributeset, &ord) + != ZEBRA_OK) + { + break; + } + ords[ord_no++] = ord; } if (!bases_ok && errCode) { @@ -2924,7 +2671,7 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, struct scan_info *scan_info = scan_info_array + i; struct rpn_char_map_info rcmi; - rpn_char_map_prepare (zh->reg, reg_id, &rcmi); + rpn_char_map_prepare (zh->reg, index_type, &rcmi); scan_info->before = before; scan_info->after = after; @@ -2939,7 +2686,8 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, termz[prefix_len] = 0; strcpy(scan_info->prefix, termz); - if (trans_scan_term(zh, zapt, termz+prefix_len, reg_id) == ZEBRA_FAIL) + if (trans_scan_term(zh, zapt, termz+prefix_len, index_type) == + ZEBRA_FAIL) return ZEBRA_FAIL; dict_scan(zh->reg->dict, termz, &before_tmp, &after_tmp, @@ -2982,12 +2730,12 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, if (lo >= 0) { /* get result set for first term */ - zebra_term_untrans_iconv(zh, stream->mem, reg_id, + zebra_term_untrans_iconv(zh, stream->mem, index_type, &glist[lo].term, mterm); rset = rset_trunc(zh, &scan_info_array[j0].list[ptr[j0]].isam_p, 1, glist[lo].term, strlen(glist[lo].term), NULL, 0, zapt->term->which, rset_nmem, - kc, kc->scope, 0, reg_id, 0 /* hits_limit */, + kc, kc->scope, 0, index_type, 0 /* hits_limit */, 0 /* term_ref_id_str */); } ptr[j0]++; /* move index for this set .. */ @@ -3009,7 +2757,7 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, glist[lo].term, strlen(glist[lo].term), NULL, 0, zapt->term->which,rset_nmem, - kc, kc->scope, 0, reg_id, 0 /* hits_limit */, + kc, kc->scope, 0, index_type, 0 /* hits_limit */, 0 /* term_ref_id_str */ ); rset = rsmulti_or_create(rset_nmem, kc, kc->scope, 0 /* termid */, @@ -3076,14 +2824,14 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, if (j0 == -1) break; - zebra_term_untrans_iconv(zh, stream->mem, reg_id, + zebra_term_untrans_iconv(zh, stream->mem, index_type, &glist[lo].term, mterm); rset = rset_trunc (zh, &scan_info_array[j0].list[before-1-ptr[j0]].isam_p, 1, glist[lo].term, strlen(glist[lo].term), NULL, 0, zapt->term->which, rset_nmem, - kc, kc->scope, 0, reg_id, 0 /* hits_limit */, + kc, kc->scope, 0, index_type, 0 /* hits_limit */, 0 /* term_ref_id_str */); ptr[j0]++; @@ -3103,7 +2851,7 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, glist[lo].term, strlen(glist[lo].term), NULL, 0, zapt->term->which, rset_nmem, - kc, kc->scope, 0, reg_id, 0 /* hits_limit */, + kc, kc->scope, 0, index_type, 0 /* hits_limit */, 0 /* term_ref_id_str */); rset = rsmulti_or_create(rset_nmem, kc, kc->scope, 0 /* termid */, 2, rsets); diff --git a/index/zsets.c b/index/zsets.c index 65e5272..e915543 100644 --- a/index/zsets.c +++ b/index/zsets.c @@ -1,4 +1,4 @@ -/* $Id: zsets.c,v 1.102 2006-05-18 12:03:05 adam Exp $ +/* $Id: zsets.c,v 1.103 2006-05-19 13:49:35 adam Exp $ Copyright (C) 1995-2006 Index Data ApS @@ -802,7 +802,6 @@ ZEBRA_RES resultSetSortSingle(ZebraHandle zh, NMEM nmem, if (zh->m_staticrank) sysno_mem_index = 1; - assert(nmem); /* compiler shut up about unused param */ sset->sort_info->num_entries = 0; @@ -844,8 +843,17 @@ ZEBRA_RES resultSetSortSingle(ZebraHandle zh, NMEM nmem, case Z_SortKey_sortField: yaz_log(log_level_sort, "key %d is of type sortField", i+1); - zebra_setError(zh, YAZ_BIB1_CANNOT_SORT_ACCORDING_TO_SEQUENCE, 0); - return ZEBRA_FAIL; + sort_criteria[i].numerical = 0; + sort_criteria[i].ord = + zebraExplain_lookup_attr_str(zh->reg->zei, 's', + sk->u.sortField); + if (sort_criteria[i].ord == -1) + { + zebra_setError(zh, + YAZ_BIB1_CANNOT_SORT_ACCORDING_TO_SEQUENCE, 0); + return ZEBRA_FAIL; + } + break; case Z_SortKey_elementSpec: yaz_log(log_level_sort, "key %d is of type elementSpec", i+1); @@ -853,29 +861,11 @@ ZEBRA_RES resultSetSortSingle(ZebraHandle zh, NMEM nmem, return ZEBRA_FAIL; case Z_SortKey_sortAttributes: yaz_log(log_level_sort, "key %d is of type sortAttributes", i+1); - if (1) - { - int ord; - int use = zebra_maps_sort(zh->reg->zebra_maps, - sk->u.sortAttributes, - &sort_criteria[i].numerical); - yaz_log(log_level_sort, "use value = %d", use); - if (use == -1) - { - zebra_setError( - zh, YAZ_BIB1_USE_ATTRIBUTE_REQUIRED_BUT_NOT_SUPPLIED, 0); - return ZEBRA_FAIL; - } - ord = zebraExplain_lookup_attr_su_any_index(zh->reg->zei, - VAL_IDXPATH, use); - if (ord == -1) - { - zebra_setError( - zh, YAZ_BIB1_CANNOT_SORT_ACCORDING_TO_SEQUENCE, 0); - return ZEBRA_FAIL; - } - sort_criteria[i].ord = ord; - } + if (zebra_sort_get_ord(zh, sk->u.sortAttributes, + &sort_criteria[i].ord, + &sort_criteria[i].numerical) + != ZEBRA_OK) + return ZEBRA_FAIL; break; } } diff --git a/recctrl/recgrs.c b/recctrl/recgrs.c index 55bdaa5..f45f4d7 100644 --- a/recctrl/recgrs.c +++ b/recctrl/recgrs.c @@ -1,5 +1,5 @@ -/* $Id: recgrs.c,v 1.109 2006-05-10 08:13:28 adam Exp $ - Copyright (C) 1995-2005 +/* $Id: recgrs.c,v 1.110 2006-05-19 13:49:35 adam Exp $ + Copyright (C) 1995-2006 Index Data ApS This file is part of the Zebra server. @@ -471,33 +471,20 @@ static void index_xpath_attr (char *tag_path, char *name, char *value, char *structure, struct recExtractCtrl *p, RecWord *wrd) { -#if NATTR wrd->index_name = ZEBRA_XPATH_ELM_BEGIN; -#else - wrd->attrSet = VAL_IDXPATH; - wrd->attrUse = 1; -#endif wrd->index_type = '0'; wrd->term_buf = tag_path; wrd->term_len = strlen(tag_path); (*p->tokenAdd)(wrd); if (value) { -#if NATTR - wrd->index_name = ZEBRA_XPATH_ATTR; -#else - wrd->attrUse = 1015; -#endif + wrd->index_name = ZEBRA_XPATH_ATTR_CDATA; wrd->index_type = 'w'; wrd->term_buf = value; wrd->term_len = strlen(value); (*p->tokenAdd)(wrd); } -#if NATTR wrd->index_name = ZEBRA_XPATH_ELM_END; -#else - wrd->attrUse = 2; -#endif wrd->index_type = '0'; wrd->term_buf = tag_path; wrd->term_len = strlen(tag_path); @@ -533,12 +520,8 @@ static void mk_tag_path_full(char *tag_path_full, size_t max, data1_node *n) static void index_xpath(struct source_parser *sp, data1_node *n, struct recExtractCtrl *p, int level, RecWord *wrd, -#if NATTR char *xpath_index, int xpath_is_start -#else - int use -#endif ) { int i; @@ -546,19 +529,8 @@ static void index_xpath(struct source_parser *sp, data1_node *n, int termlist_only = 1; data1_termlist *tl; int xpdone = 0; -#if NATTR -#else - int xpath_is_start = 0; - if (use == 1) - xpath_is_start = 1; -#endif - -#if NATTR yaz_log(YLOG_DEBUG, "index_xpath level=%d xpath_index=%s", level, xpath_index); -#else - yaz_log(YLOG_DEBUG, "index_xpath level=%d use=%d", level, use); -#endif if ((!n->root->u.root.absyn) || (n->root->u.root.absyn->enable_xpath_indexing)) { termlist_only = 0; @@ -585,19 +557,10 @@ static void index_xpath(struct source_parser *sp, data1_node *n, memcpy (&wrd_tl, wrd, sizeof(*wrd)); if (tl->source) sp_parse(sp, n, &wrd_tl, tl->source); -#if NATTR if (!tl->index_name) -#else - if (!tl->att) -#endif { /* this is the ! case, so structure is for the xpath index */ -#if NATTR wrd_tl.index_name = xpath_index; -#else - wrd_tl.attrSet = VAL_IDXPATH; - wrd_tl.attrUse = use; -#endif if (p->flagShowRecords) { int i; @@ -615,25 +578,13 @@ static void index_xpath(struct source_parser *sp, data1_node *n, xpdone = 1; } else { /* this is just the old fashioned attribute based index */ -#if NATTR wrd_tl.index_name = tl->index_name; -#else - wrd_tl.attrSet = (int) (tl->att->parent->reference); - wrd_tl.attrUse = tl->att->locals->local; -#endif if (p->flagShowRecords) { int i; printf("%*sIdx: [%s]", (level + 1) * 4, "", tl->structure); -#if NATTR printf("%s %s", tl->index_name, tl->source); -#else - printf("%s:%s [%d] %s", - tl->att->parent->name, - tl->att->name, tl->att->value, - tl->source); -#endif printf (" XData:\""); for (i = 0; iflagShowRecords && !xpdone && !termlist_only) { -#if NATTR wrd->index_name = xpath_index; -#else - wrd->attrSet = VAL_IDXPATH; - wrd->attrUse = use; -#endif wrd->index_type = 'w'; (*p->tokenAdd)(wrd); } @@ -667,12 +613,7 @@ static void index_xpath(struct source_parser *sp, data1_node *n, wrd->index_type = '0'; wrd->term_buf = tag_path_full; wrd->term_len = strlen(tag_path_full); -#if NATTR wrd->index_name = xpath_index; -#else - wrd->attrSet = VAL_IDXPATH; - wrd->attrUse = use; -#endif if (p->flagShowRecords) { printf("%*s tag=", (level + 1) * 4, ""); @@ -698,11 +639,7 @@ static void index_xpath(struct source_parser *sp, data1_node *n, { for (; tl; tl = tl->next) { -#if NATTR if (!tl->index_name) -#else - if (!tl->att) -#endif do_xpindex = 1; } } @@ -742,13 +679,8 @@ static void index_xpath(struct source_parser *sp, data1_node *n, { for (; tl; tl = tl->next) { -#if NATTR if (!tl->index_name) do_xpindex = 1; -#else - if (!tl->att) - do_xpindex = 1; -#endif } } @@ -756,11 +688,7 @@ static void index_xpath(struct source_parser *sp, data1_node *n, /* attribute (no value) */ wrd->index_type = '0'; -#if NATTR - wrd->index_name = ZEBRA_XPATH_ATTR; -#else - wrd->attrUse = 3; -#endif + wrd->index_name = ZEBRA_XPATH_ATTR_NAME; wrd->term_buf = xp->name; wrd->term_len = strlen(xp->name); @@ -775,11 +703,7 @@ static void index_xpath(struct source_parser *sp, data1_node *n, strcat (comb, "="); strcat (comb, xp->value); -#if NATTR - wrd->index_name = ZEBRA_XPATH_ATTR; -#else - wrd->attrUse = 3; -#endif + wrd->index_name = ZEBRA_XPATH_ATTR_NAME; wrd->index_type = '0'; wrd->term_buf = comb; wrd->term_len = strlen(comb); @@ -805,11 +729,7 @@ static void index_xpath(struct source_parser *sp, data1_node *n, /* If there is a termlist given (=xelm directive) */ for (; tl; tl = tl->next) { -#if NATTR if (!tl->index_name) -#else - if (!tl->att) -#endif { /* add xpath index for the attribute */ index_xpath_attr (attr_tag_path_full, xp->name, @@ -820,13 +740,7 @@ static void index_xpath(struct source_parser *sp, data1_node *n, /* index attribute value (only path/@attr) */ if (xp->value) { -#if NATTR wrd->index_name = tl->index_name; -#else - wrd->attrSet = (int) - (tl->att->parent->reference); - wrd->attrUse = tl->att->locals->local; -#endif wrd->index_type = *tl->structure; wrd->term_buf = xp->value; wrd->term_len = strlen(xp->value); @@ -885,14 +799,7 @@ static void index_termlist (struct source_parser *sp, data1_node *par, int i; printf("%*sIdx: [%s]", (level + 1) * 4, "", tlist->structure); -#if NATTR printf("%s %s", tlist->index_name, tlist->source); -#else - printf("%s:%s [%d] %s", - tlist->att->parent->name, - tlist->att->name, tlist->att->value, - tlist->source); -#endif printf (" XData:\""); for (i = 0; iterm_len && i < 40; i++) fputc (wrd->term_buf[i], stdout); @@ -904,12 +811,7 @@ static void index_termlist (struct source_parser *sp, data1_node *par, else { wrd->index_type = *tlist->structure; -#if NATTR wrd->index_name = tlist->index_name; -#else - wrd->attrSet = (int) (tlist->att->parent->reference); - wrd->attrUse = tlist->att->locals->local; -#endif (*p->tokenAdd)(wrd); } } @@ -963,14 +865,9 @@ static int dumpkeys_r(struct source_parser *sp, { index_termlist(sp, n, n, p, level, wrd); /* index start tag */ -#if NATTR if (n->root->u.root.absyn) index_xpath(sp, n, p, level, wrd, ZEBRA_XPATH_ELM_BEGIN, 1 /* is start */); -#else - if (n->root->u.root.absyn) - index_xpath(sp, n, p, level, wrd, 1); -#endif } if (n->child) @@ -998,23 +895,15 @@ static int dumpkeys_r(struct source_parser *sp, if (par) index_termlist(sp, par, n, p, level, wrd); -#if NATTR index_xpath(sp, n, p, level, wrd, ZEBRA_XPATH_CDATA, 0 /* is start */); -#else - index_xpath(sp, n, p, level, wrd, 1016); -#endif } if (n->which == DATA1N_tag) { /* index end tag */ -#if NATTR index_xpath(sp, n, p, level, wrd, ZEBRA_XPATH_ELM_END, 0 /* is start */); -#else - index_xpath(sp, n, p, level, wrd, 2); -#endif } if (p->flagShowRecords && n->which == DATA1N_root) diff --git a/tab/gils.abs b/tab/gils.abs index da047a1..772e13f 100644 --- a/tab/gils.abs +++ b/tab/gils.abs @@ -1,7 +1,7 @@ # This is the abstract syntax (and most of the top-level profile info) # for GILS version 2. # -# $Id: gils.abs,v 1.2 2002-12-02 16:55:14 adam Exp $ +# $Id: gils.abs,v 1.3 2006-05-19 13:49:35 adam Exp $ name gils reference GILS-schema @@ -22,7 +22,7 @@ esetname F @ systag sysno none -#all Any +all Any elm (1,1) schemaIdentifier - elm (1,10) rank - diff --git a/test/api/Makefile.am b/test/api/Makefile.am index 9c56604..52a7d31 100644 --- a/test/api/Makefile.am +++ b/test/api/Makefile.am @@ -1,9 +1,9 @@ -# $Id: Makefile.am,v 1.35 2006-04-27 20:40:41 marc Exp $ +# $Id: Makefile.am,v 1.36 2006-05-19 13:49:35 adam Exp $ noinst_PROGRAMS = testclient testclient_SOURCES = testclient.c -simpletests = t0 t1 t2 t3 t4 t5 t6 t7 t8 t9 t10 t11 t12 t13 t14 t15 +simpletests = t0 t1 t2 t3 t4 t5 t6 t7 t8 t9 t11 t12 t13 t14 t15 safaritests = safari1 check_PROGRAMS = $(simpletests) $(safaritests) TESTS = $(check_PROGRAMS) @@ -24,7 +24,6 @@ t6_SOURCES = t6.c t7_SOURCES = t7.c t8_SOURCES = t8.c t9_SOURCES = t9.c rankingrecords.h -t10_SOURCES = t10.c rankingrecords.h t11_SOURCES = t11.c t12_SOURCES = t12.c t13_SOURCES = t13.c diff --git a/test/api/t10.c b/test/api/t10.c deleted file mode 100644 index 4429132..0000000 --- a/test/api/t10.c +++ /dev/null @@ -1,105 +0,0 @@ -/* $Id: t10.c,v 1.11 2006-05-10 08:13:35 adam Exp $ - Copyright (C) 1995-2005 - 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. -*/ - -/** t10.c - test zv-rank */ - -#include "testlib.h" -#include "rankingrecords.h" - -struct tst { - char *schema; - char *hit1; - int score1; - char *hit2; - int score2; - char *hit3; - int score3; -}; - -struct tst tests[] = { - {"ntc-atn", "first title", 0, "first title", 1000, "first title",1000 }, - {"ntc-ntn", "first title", 0, "first title", 1000, "first title",1000 }, - {"ntc-btn", "first title", 0, "first title", 1000, "first title",1000 }, - {"ntc-apn", "first title", 0, "first title", 1000, "first title",1000 }, - {"ntc-npn", "first title", 0, "first title", 1000, "first title",1000 }, - {"ntc-bpn", "first title", 0, "first title", 1000, "first title",1000 }, - {"atc-atn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"atc-ntn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"atc-btn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"atc-apn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"atc-npn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"atc-bpn", "first title", 0, "first title", 1000, "first title", 1000 }, - - {"npc-atn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"npc-ntn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"npc-btn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"npc-apn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"npc-npn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"npc-bpn", "first title", 0, "first title", 1000, "first title", 1000 }, - - {"apc-atn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"apc-ntn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"apc-btn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"apc-apn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"apc-npn", "first title", 0, "first title", 1000, "first title", 1000 }, - {"apc-bpn", "first title", 0, "first title", 1000, "first title", 1000 }, - {0,0,0,0,0,0,0}, -}; - -static void tst(int argc, char **argv) -{ - int i; - ZebraService zs = tl_start_up("zebrazv.cfg", argc, argv); - ZebraHandle zh = zebra_open(zs, 0); - - YAZ_CHECK(tl_init_data(zh, recs)); - zebra_close(zh); - for (i = 0; tests[i].schema; i++) - { - zh = zebra_open(zs, 0); - YAZ_CHECK(zh); - YAZ_CHECK(zebra_select_database(zh, "Default") == ZEBRA_OK); - - zebra_set_resource(zh, "zvrank.weighting-scheme", tests[i].schema); - yaz_log(YLOG_LOG,"============%d: %s ===========", i, tests[i].schema); - - YAZ_CHECK(tl_ranking_query(zh, "@attr 1=1016 @attr 2=102 the", - 3, tests[i].hit1, tests[i].score1)); - YAZ_CHECK(tl_ranking_query(zh, "@attr 1=1016 @attr 2=102 @or foo bar", - 3, tests[i].hit2, tests[i].score2)); - YAZ_CHECK(tl_ranking_query(zh, - "@attr 1=1016 @attr 2=102 @or @or the foo bar", - 3, tests[i].hit3, tests[i].score3)); - zebra_close(zh); - } - YAZ_CHECK(tl_close_down(0, zs)); -} - -TL_MAIN -/* - * Local variables: - * c-basic-offset: 4 - * indent-tabs-mode: nil - * End: - * vim: shiftwidth=4 tabstop=8 expandtab - */ - diff --git a/test/api/t5.c b/test/api/t5.c index 96ec8d3..df2eec6 100644 --- a/test/api/t5.c +++ b/test/api/t5.c @@ -1,4 +1,4 @@ -/* $Id: t5.c,v 1.16 2006-05-10 08:13:35 adam Exp $ +/* $Id: t5.c,v 1.17 2006-05-19 13:49:35 adam Exp $ Copyright (C) 1995-2005 Index Data ApS @@ -62,19 +62,19 @@ static void tst(int argc, char **argv) YAZ_CHECK(tl_query(zh, "@attr 1=4 @and x title", 2)); /* exl=0 distance=2 order=1 relation=2 (<=), known, unit=word */ - YAZ_CHECK(tl_query(zh, "@prox 0 2 1 2 k 2 my x", 2)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @prox 0 2 1 2 k 2 my x", 2)); /* exl=0 distance=2 order=1 relation=2 (<=), known, unit=word */ - YAZ_CHECK(tl_query(zh, "@prox 0 2 1 2 k 2 x my", 0)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @prox 0 2 1 2 k 2 x my", 0)); /* exl=0 distance=2 order=0 relation=2 (<=), known, unit=word */ - YAZ_CHECK(tl_query(zh, "@prox 0 2 0 2 k 2 x my", 2)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @prox 0 2 0 2 k 2 x my", 2)); /* exl=0 distance=2 order=0 relation=3 (=), known, unit=word */ - YAZ_CHECK(tl_query(zh, "@prox 0 2 1 3 k 2 my x", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @prox 0 2 1 3 k 2 my x", 1)); /* exl=1 distance=2 order=0 relation=3 (=), known, unit=word */ - YAZ_CHECK(tl_query(zh, "@prox 1 2 1 3 k 2 my x", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @prox 1 2 1 3 k 2 my x", 1)); /* provoke unsupported use attribute */ YAZ_CHECK(tl_query_x(zh, "@attr 1=999 @attr 4=1 x", 0, 114)); @@ -93,7 +93,7 @@ static void tst(int argc, char **argv) 0, 121)); /* provoke unsupported relation */ - YAZ_CHECK(tl_query_x(zh, "@attr 1=1016 @attr 2=6 x", 0, 117)); + YAZ_CHECK(tl_query_x(zh, "@attr 1=4 @attr 2=6 x", 0, 117)); YAZ_CHECK(tl_query_x(zh, "@attr 1=1016 @attr 2=6 @attr 4=109 x", 0, 114)); YAZ_CHECK(tl_close_down(zh, zs)); diff --git a/test/api/t8.c b/test/api/t8.c index 0c811a6..e1aa6ca 100644 --- a/test/api/t8.c +++ b/test/api/t8.c @@ -1,4 +1,4 @@ -/* $Id: t8.c,v 1.10 2006-05-10 08:13:35 adam Exp $ +/* $Id: t8.c,v 1.11 2006-05-19 13:49:35 adam Exp $ Copyright (C) 1995-2005 Index Data ApS @@ -64,7 +64,6 @@ static void tst(int argc, char **argv) /* couple of simple queries just to see that we have indexed the stuff */ YAZ_CHECK(tl_query(zh, "@attr 1=4 title", 2)); - YAZ_CHECK(tl_query(zh, "title", 2)); /* 1=2038: West-Bounding-Coordinate 2039: East: 2040: North: 2041 South*/ /* 4=109: numeric string */ diff --git a/test/api/t9.c b/test/api/t9.c index cfbae53..74eed98 100644 --- a/test/api/t9.c +++ b/test/api/t9.c @@ -1,5 +1,5 @@ -/* $Id: t9.c,v 1.9 2006-05-10 08:13:35 adam Exp $ - Copyright (C) 1995-2005 +/* $Id: t9.c,v 1.10 2006-05-19 13:49:35 adam Exp $ + Copyright (C) 1995-2006 Index Data ApS This file is part of the Zebra server. @@ -33,16 +33,16 @@ static void tst(int argc, char **argv) YAZ_CHECK(tl_init_data(zh, recs)); - YAZ_CHECK(tl_ranking_query(zh, "@attr 1=1016 @attr 2=102 the", - 3, "first title", 997 )); + YAZ_CHECK(tl_ranking_query(zh, "@attr 1=4 @attr 2=102 the", + 3, "first title", 1000 )); - YAZ_CHECK(tl_ranking_query(zh, "@attr 1=1016 @attr 2=102 foo", + YAZ_CHECK(tl_ranking_query(zh, "@attr 1=62 @attr 2=102 foo", 3, "second title", 850 )); /* get the record with the most significant hit, that is the 'bar' */ /* as that is the rarest of my search words */ YAZ_CHECK(tl_ranking_query(zh, "@attr 1=1016 @attr 2=102 @or @or the foo bar", - 3, "third title", 940 )); + 3, "third title", 813 )); YAZ_CHECK(tl_close_down(zh, zs)); } diff --git a/test/charmap/charmap1.c b/test/charmap/charmap1.c index 5783946..768b47a 100644 --- a/test/charmap/charmap1.c +++ b/test/charmap/charmap1.c @@ -1,4 +1,4 @@ -/* $Id: charmap1.c,v 1.10 2006-05-10 08:13:36 adam Exp $ +/* $Id: charmap1.c,v 1.11 2006-05-19 13:49:35 adam Exp $ Copyright (C) 1995-2005 Index Data ApS @@ -41,46 +41,46 @@ static void tst(int argc, char **argv) YAZ_CHECK(zebra_end_trans(zh) == ZEBRA_OK); zebra_commit(zh); - YAZ_CHECK(tl_query(zh, "@term string æ", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @term string æ", 1)); /* search for UNICODE 1E25 - letter h with dot below */ - YAZ_CHECK(tl_query(zh, "@term string ḥ", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @term string ḥ", 1)); /* search for UNICODE A ring */ - YAZ_CHECK(tl_query(zh, "@term string lÃ¥s", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @term string lÃ¥s", 1)); /* search for aa */ - YAZ_CHECK(tl_query(zh, "@term string laas", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @term string laas", 1)); /* search for aa regular-1 */ - YAZ_CHECK(tl_query(zh, "@attr 5=102 @term string lÃ¥s", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 5=102 @term string lÃ¥s", 1)); /* search for aa regular-2 */ - YAZ_CHECK(tl_query(zh, "@attr 5=103 @term string lÃ¥s", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 5=103 @term string lÃ¥s", 1)); /* search for aa trunc=104 */ - YAZ_CHECK(tl_query(zh, "@attr 5=104 @term string laas", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 5=104 @term string laas", 1)); /* search for aa trunc=105 */ - YAZ_CHECK(tl_query(zh, "@attr 5=104 @term string laas", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 5=104 @term string laas", 1)); /* search for aaa */ - YAZ_CHECK(tl_query(zh, "@term string laaas", 0)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @term string laaas", 0)); /* search ABC in title:0 . */ - YAZ_CHECK(tl_query(zh, "@attr 4=3 @attr 1=4 ABC", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 4=3 @attr 1=4 ABC", 1)); /* search DEF in title:0 . */ - YAZ_CHECK(tl_query(zh, "@attr 4=3 @attr 1=4 DEF", 0)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 4=3 @attr 1=4 DEF", 0)); /* search [ in title:0 . */ - YAZ_CHECK(tl_query(zh, "@attr 4=3 @attr 1=4 [", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 4=3 @attr 1=4 [", 1)); /* search \ in title:0 . */ - YAZ_CHECK(tl_query(zh, "@attr 4=3 @attr 1=4 \\\\\\\\", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 4=3 @attr 1=4 \\\\\\\\", 1)); /* search { in title:0 . */ - YAZ_CHECK(tl_query(zh, "@attr 4=3 @attr 1=4 \\{", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 4=3 @attr 1=4 \\{", 1)); YAZ_CHECK(tl_close_down(zh, zs)); } diff --git a/test/filters/grs.marc.c b/test/filters/grs.marc.c index bcd7f55..d115611 100644 --- a/test/filters/grs.marc.c +++ b/test/filters/grs.marc.c @@ -1,4 +1,4 @@ -/* $Id: grs.marc.c,v 1.4 2006-05-10 08:13:36 adam Exp $ +/* $Id: grs.marc.c,v 1.5 2006-05-19 13:49:35 adam Exp $ Copyright (C) 1995-2005 Index Data ApS @@ -42,9 +42,9 @@ static void tst(int argc, char **argv) YAZ_CHECK(zebra_end_trans(zh) == ZEBRA_OK); zebra_commit(zh); - YAZ_CHECK(tl_query(zh, "computer", 1)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 computer", 1)); - YAZ_CHECK(tl_query(zh, "computee", 0)); + YAZ_CHECK(tl_query(zh, "@attr 1=4 computee", 0)); YAZ_CHECK(tl_close_down(zh, zs)); } diff --git a/test/gils/test1.sh b/test/gils/test1.sh index 4eff517..13f7fd9 100755 --- a/test/gils/test1.sh +++ b/test/gils/test1.sh @@ -1,5 +1,5 @@ #!/bin/sh -# $Id: test1.sh,v 1.14 2006-04-27 10:52:26 marc Exp $ +# $Id: test1.sh,v 1.15 2006-05-19 13:49:36 adam Exp $ srcdir=${srcdir:-"."} @@ -44,7 +44,7 @@ grep "^Result count: 17$" log >/dev/null || exit 1 echo "search 2..." >>$LOG ../api/testclient unix:socket "@or utah the" > log || exit 1 -grep "^Result count: 40$" log >/dev/null || exit 1 +grep "^Result count: 41$" log >/dev/null || exit 1 echo "search 3..." >>$LOG ../api/testclient unix:socket "@attr 1=4 the" > log || exit 1 diff --git a/test/gils/test3.sh b/test/gils/test3.sh index fb5e07b..7350901 100755 --- a/test/gils/test3.sh +++ b/test/gils/test3.sh @@ -1,5 +1,5 @@ #!/bin/sh -# $Id: test3.sh,v 1.8 2006-04-27 10:52:26 marc Exp $ +# $Id: test3.sh,v 1.9 2006-05-19 13:49:36 adam Exp $ # Testing searches with lots of @and operators # in order to test the fast-forward operation of rsets @@ -37,7 +37,7 @@ echo "search A1..." >>$LOG ../api/testclient -c 17 unix:socket utah > log || exit 1 echo "search A2..." >>$LOG -../api/testclient -c 30 unix:socket the > log || exit 1 +../api/testclient -c 31 unix:socket the > log || exit 1 echo "search A3..." >>$LOG ../api/testclient -c 4 unix:socket deposits > log || exit 1 @@ -69,19 +69,19 @@ echo "search C6..." >>$LOG echo "search D1..." >>$LOG -../api/testclient -c 29 unix:socket "@and the of " > log || exit 1 +../api/testclient -c 30 unix:socket "@and the of " > log || exit 1 echo "search D2..." >>$LOG -../api/testclient -c 29 unix:socket "@and of the" > log || exit 1 +../api/testclient -c 30 unix:socket "@and of the" > log || exit 1 echo "search D3..." >>$LOG -../api/testclient -c 29 unix:socket "@and @and the of of" > log || exit 1 +../api/testclient -c 30 unix:socket "@and @and the of of" > log || exit 1 echo "search D4..." >>$LOG -../api/testclient -c 29 unix:socket "@and @and of the the" > log || exit 1 +../api/testclient -c 30 unix:socket "@and @and of the the" > log || exit 1 echo "search D5..." >>$LOG -../api/testclient -c 29 unix:socket "@and @and @and the of of the" > log || exit 1 +../api/testclient -c 30 unix:socket "@and @and @and the of of the" > log || exit 1 echo "search D6..." >>$LOG ../api/testclient -c 15 unix:socket '@and @and in the data' > log || exit 1 @@ -94,22 +94,22 @@ echo "search D8..." >>$LOG # This one failed at early fast-forwards echo "search E1..." >>$LOG -../api/testclient -c 40 unix:socket "@or the utah" > log || exit 1 +../api/testclient -c 41 unix:socket "@or the utah" > log || exit 1 echo "search E2..." >>$LOG -../api/testclient -c 40 unix:socket "@or utah the" > log || exit 1 +../api/testclient -c 41 unix:socket "@or utah the" > log || exit 1 echo "search E3..." >>$LOG ../api/testclient -c 42 unix:socket "@or deposits @or the utah" > log || exit 1 echo "search E4..." >>$LOG -../api/testclient -c 2 unix:socket "@and deposits @or the utah" > log || exit 1 +../api/testclient -c 3 unix:socket "@and deposits @or the utah" > log || exit 1 echo "search E5..." >>$LOG -../api/testclient -c 2 unix:socket "@and @or the utah deposits" > log || exit 1 +../api/testclient -c 3 unix:socket "@and @or the utah deposits" > log || exit 1 echo "search F1..." >>$LOG -../api/testclient -c 23 unix:socket "@not the utah " > log || exit 1 +../api/testclient -c 24 unix:socket "@not the utah " > log || exit 1 echo "search F2..." >>$LOG ../api/testclient -c 10 unix:socket "@not utah the " > log || exit 1 diff --git a/test/sort/sort1.c b/test/sort/sort1.c index 4e648c1..34419a6 100644 --- a/test/sort/sort1.c +++ b/test/sort/sort1.c @@ -1,4 +1,4 @@ -/* $Id: sort1.c,v 1.8 2006-05-10 08:13:39 adam Exp $ +/* $Id: sort1.c,v 1.9 2006-05-19 13:49:36 adam Exp $ Copyright (C) 1995-2005 Index Data ApS @@ -47,19 +47,19 @@ static void tst(int argc, char **argv) ids[1] = 2; ids[2] = 4; ids[3] = 5; - YAZ_CHECK(tl_sort(zh, "@or computer @attr 7=1 @attr 1=30 0", 4, ids)); + YAZ_CHECK(tl_sort(zh, "@or @attr 1=4 computer @attr 7=1 @attr 1=30 0", 4, ids)); ids[0] = 5; ids[1] = 4; ids[2] = 2; ids[3] = 3; - YAZ_CHECK(tl_sort(zh, "@or computer @attr 7=1 @attr 1=1021 0", 4, ids)); + YAZ_CHECK(tl_sort(zh, "@or @attr 1=4 computer @attr 7=1 @attr 1=1021 0", 4, ids)); ids[0] = 2; ids[1] = 5; ids[2] = 4; ids[3] = 3; - YAZ_CHECK(tl_sort(zh, "@or computer @attr 7=1 @attr 1=1021 @attr 4=109 0", 4, ids)); + YAZ_CHECK(tl_sort(zh, "@or @attr 1=4 computer @attr 7=1 @attr 1=1021 @attr 4=109 0", 4, ids)); YAZ_CHECK(tl_close_down(zh, zs)); } diff --git a/test/sort2/t1.c b/test/sort2/t1.c index 66f81c5..d2bd23b 100644 --- a/test/sort2/t1.c +++ b/test/sort2/t1.c @@ -1,4 +1,4 @@ -/* $Id: t1.c,v 1.6 2006-05-10 08:13:39 adam Exp $ +/* $Id: t1.c,v 1.7 2006-05-19 13:49:36 adam Exp $ Copyright (C) 1995-2005 Index Data ApS @@ -47,7 +47,7 @@ static void tst(int argc, char **argv) ids[1] = 5; ids[2] = 3; ids[3] = 4; - YAZ_CHECK(tl_sort(zh, "@or computer @attr 7=1 @attr 1=4 0", 4, ids)); + YAZ_CHECK(tl_sort(zh, "@or @attr 1=4 computer @attr 7=1 @attr 1=4 0", 4, ids)); YAZ_CHECK(tl_close_down(zh, zs)); } diff --git a/test/xpath/zebra.cfg b/test/xpath/zebra.cfg index 713cf95..ee9baf1 100644 --- a/test/xpath/zebra.cfg +++ b/test/xpath/zebra.cfg @@ -1,7 +1,8 @@ -# $Id: zebra.cfg,v 1.6 2004-12-15 13:07:07 adam Exp $ +# $Id: zebra.cfg,v 1.7 2006-05-19 13:49:38 adam Exp $ profilePath: ${srcdir:-.}:${srcdir:-.}/../../tab recordType: grs.sgml attset: bib1.att +attset: explain.att isam:b diff --git a/util/Makefile.am b/util/Makefile.am index 8ff62e1..4cba0d4 100644 --- a/util/Makefile.am +++ b/util/Makefile.am @@ -1,4 +1,4 @@ -## $Id: Makefile.am,v 1.17 2006-04-27 20:40:45 marc Exp $ +## $Id: Makefile.am,v 1.18 2006-05-19 13:49:38 adam Exp $ lib_LTLIBRARIES = libidzebra-util.la @@ -18,7 +18,7 @@ AM_CPPFLAGS = -I$(srcdir)/../include $(YAZINC) -DDEFAULT_PROFILE_PATH=\"$(pkgdat LDADD = libidzebra-util.la $(YAZLALIB) libidzebra_util_la_SOURCES = zint.c res.c charmap.c zebramap.c passwddb.c \ - zebra-lock.c dirent.c xpath.c atoi_zn.c snippet.c flock.c + zebra-lock.c dirent.c xpath.c atoi_zn.c snippet.c flock.c attrfind.c passtest_SOURCES = passtest.c diff --git a/util/attrfind.c b/util/attrfind.c new file mode 100644 index 0000000..80f9912 --- /dev/null +++ b/util/attrfind.c @@ -0,0 +1,122 @@ +/* $Id: attrfind.c,v 1.1 2006-05-19 13:49:38 adam Exp $ + Copyright (C) 2005-2006 + 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 + +void attr_init_APT(AttrType *src, Z_AttributesPlusTerm *zapt, int type) +{ + src->attributeList = zapt->attributes->attributes; + src->num_attributes = zapt->attributes->num_attributes; + src->type = type; + src->major = 0; + src->minor = 0; +} + +void attr_init_AttrList(AttrType *src, Z_AttributeList *list, int type) +{ + src->attributeList = list->attributes; + src->num_attributes = list->num_attributes; + src->type = type; + src->major = 0; + src->minor = 0; +} + +int attr_find_ex(AttrType *src, oid_value *attributeSetP, + const char **string_value) +{ + int num_attributes; + + num_attributes = src->num_attributes; + while (src->major < 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) + break; + if (element->attributeSet && attributeSetP) + { + oident *attrset; + + attrset = oid_getentbyoid(element->attributeSet); + *attributeSetP = attrset->value; + } + if (element->value.complex->list[src->minor]->which == + Z_StringOrNumeric_numeric) + { + ++(src->minor); + return + *element->value.complex->list[src->minor-1]->u.numeric; + } + else if (element->value.complex->list[src->minor]->which == + Z_StringOrNumeric_string) + { + if (!string_value) + break; + ++(src->minor); + *string_value = + element->value.complex->list[src->minor-1]->u.string; + return -2; + } + else + break; + default: + assert(0); + } + } + ++(src->major); + } + return -1; +} + +int attr_find(AttrType *src, oid_value *attributeSetP) +{ + return attr_find_ex(src, attributeSetP, 0); +} + + +/* + * Local variables: + * c-basic-offset: 4 + * indent-tabs-mode: nil + * End: + * vim: shiftwidth=4 tabstop=8 expandtab + */ + diff --git a/util/zebramap.c b/util/zebramap.c index d2ac8f9..6418560 100644 --- a/util/zebramap.c +++ b/util/zebramap.c @@ -1,5 +1,5 @@ -/* $Id: zebramap.c,v 1.48 2006-05-17 17:46:45 adam Exp $ - Copyright (C) 1995-2005 +/* $Id: zebramap.c,v 1.49 2006-05-19 13:49:38 adam Exp $ + Copyright (C) 1995-2006 Index Data ApS This file is part of the Zebra server. @@ -25,6 +25,7 @@ #include #include +#include #include #include @@ -275,78 +276,6 @@ const char *zebra_maps_output(ZebraMaps zms, unsigned reg_id, } -/* ------------------------------------ */ - -typedef struct { - int type; - int major; - int minor; - Z_AttributeElement **attributeList; - int num_attributes; -} AttrType; - -static int attr_find(AttrType *src, oid_value *attributeSetP) -{ - 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; -} - -static void attr_init_APT(AttrType *src, Z_AttributesPlusTerm *zapt, int type) -{ - src->attributeList = zapt->attributes->attributes; - src->num_attributes = zapt->attributes->num_attributes; - src->type = type; - src->major = 0; - src->minor = 0; -} - -static void attr_init_AttrList(AttrType *src, Z_AttributeList *list, int type) -{ - src->attributeList = list->attributes; - src->num_attributes = list->num_attributes; - src->type = type; - src->major = 0; - src->minor = 0; -} /* ------------------------------------ */ diff --git a/win/makefile b/win/makefile index 22d0802..8b95933 100644 --- a/win/makefile +++ b/win/makefile @@ -1,5 +1,5 @@ # Zebra makefile for MS NMAKE -# $Id: makefile,v 1.43 2006-03-29 08:38:45 adam Exp $ +# $Id: makefile,v 1.44 2006-05-19 13:49:38 adam Exp $ ########################################################### ############### Parameters @@ -329,6 +329,7 @@ ZEBRALIB_OBJS= \ $(MOD_XSLT_OBJS) \ $(OBJDIR)\atoi_zn.obj \ $(OBJDIR)\attribute.obj \ + $(OBJDIR)\attrfind.obj \ $(OBJDIR)\bfile.obj \ $(OBJDIR)\bset.obj \ $(OBJDIR)\cfile.obj \