X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=index%2Frpnscan.c;h=1d6f628ca38b9e5671e554dfbb3d1e99ab24b4f8;hb=ef21c46268e6bbdd5783099a822933ac78cf1a81;hp=79b12aa2fc85d2497371b6ff4d0f02bfa688b104;hpb=d9efeb647dc1dd2ff8045306605f2d90beabf3cd;p=idzebra-moved-to-github.git diff --git a/index/rpnscan.c b/index/rpnscan.c index 79b12aa..1d6f628 100644 --- a/index/rpnscan.c +++ b/index/rpnscan.c @@ -1,4 +1,4 @@ -/* $Id: rpnscan.c,v 1.20 2007-11-01 16:01:33 adam Exp $ +/* $Id: rpnscan.c,v 1.25 2008-01-24 16:17:03 adam Exp $ Copyright (C) 1995-2007 Index Data ApS @@ -45,14 +45,28 @@ Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA static ZEBRA_RES trans_scan_term(ZebraHandle zh, Z_AttributesPlusTerm *zapt, char *termz, zebra_map_t zm) { - char termz0[IT_MAX_WORD]; + char term_utf8[IT_MAX_WORD]; - if (zapt_term_to_utf8(zh, zapt, termz0) == ZEBRA_FAIL) + if (zapt_term_to_utf8(zh, zapt, term_utf8) == ZEBRA_FAIL) return ZEBRA_FAIL; /* error */ + else if (zebra_maps_is_icu(zm)) + { + const char *res_buf; + size_t res_len; + zebra_map_tokenize_start(zm, term_utf8, strlen(term_utf8)); + + if (zebra_map_tokenize_next(zm, &res_buf, &res_len, 0, 0)) + { + memcpy(termz, res_buf, res_len); + termz[res_len] = '\0'; + } + else + termz[0] = '\0'; + } else { const char **map; - const char *cp = (const char *) termz0; + const char *cp = (const char *) term_utf8; const char *cp_end = cp + strlen(cp); const char *src; int i = 0; @@ -79,31 +93,6 @@ static ZEBRA_RES trans_scan_term(ZebraHandle zh, Z_AttributesPlusTerm *zapt, return ZEBRA_OK; } -static void count_set(ZebraHandle zh, RSET rset, zint *count, zint approx_limit) -{ - zint psysno = 0; - struct it_key key; - RSFD rfd; - - yaz_log(YLOG_DEBUG, "count_set"); - - rset->hits_limit = approx_limit; - - *count = 0; - rfd = rset_open(rset, RSETF_READ); - while (rset_read(rfd, &key,0 /* never mind terms */)) - { - if (key.mem[0] != psysno) - { - psysno = key.mem[0]; - if (rfd->counted_items >= rset->hits_limit) - break; - } - } - rset_close(rfd); - *count = rset->hits_count; -} - static void get_first_snippet_from_rset(ZebraHandle zh, RSET rset, zebra_snippets *snippets, zint *sysno) @@ -238,13 +227,11 @@ static int scan_save_set(ZebraHandle zh, ODR stream, NMEM nmem, rset = rset_create_and(nmem, kc, kc->scope, 2, rsets); } /* count it */ - count_set(zh, rset, &count, approx_limit); + zebra_count_set(zh, rset, &count, approx_limit); if (pos != -1) { zint sysno; - int code = -1; - zebra_snippets *rec_snippets = zebra_snippets_create(); zebra_snippets *hit_snippets = zebra_snippets_create(); glist[pos].term = 0; @@ -252,22 +239,38 @@ static int scan_save_set(ZebraHandle zh, ODR stream, NMEM nmem, get_first_snippet_from_rset(zh, rset, hit_snippets, &sysno); if (sysno) - code = zebra_get_rec_snippets(zh, sysno, rec_snippets); - - if (code == 0) { - const struct zebra_snippet_word *w = - zebra_snippets_lookup(rec_snippets, hit_snippets); - if (w) + zebra_snippets *rec_snippets = zebra_snippets_create(); + int code = zebra_get_rec_snippets(zh, sysno, rec_snippets); + if (code == 0) { - glist[pos].display_term = odr_strdup(stream, w->term); + const struct zebra_snippet_word *w = + zebra_snippets_lookup(rec_snippets, hit_snippets); + if (w) + { + glist[pos].display_term = odr_strdup(stream, w->term); + } + else + { + yaz_log(YLOG_WARN, "zebra_snippets_lookup failed for pos=%d", pos); + } } + zebra_snippets_destroy(rec_snippets); + } + if (zebra_term_untrans_iconv(zh, stream->mem, index_type, + &glist[pos].term, term)) + { + /* failed.. use display_term instead (which could be 0) */ + glist[pos].term = glist[pos].display_term; } + if (!glist[pos].term) - zebra_term_untrans_iconv(zh, stream->mem, index_type, - &glist[pos].term, term); + { + yaz_log(YLOG_WARN, "Could not generate scan term for pos=%d", + pos); + glist[pos].term = "None"; + } glist[pos].occurrences = count; - zebra_snippets_destroy(rec_snippets); zebra_snippets_destroy(hit_snippets); } rset_delete(rset); @@ -290,7 +293,7 @@ static ZEBRA_RES rpn_scan_norm(ZebraHandle zh, ODR stream, NMEM nmem, { struct scan2_info_entry *ar = nmem_malloc(nmem, sizeof(*ar) * ord_no); struct rpn_char_map_info rcmi; - zebra_map_t zm = zebra_map_get(zh->reg->zebra_maps, index_type); + zebra_map_t zm = zebra_map_get_or_add(zh->reg->zebra_maps, index_type); int i, dif; int after_pos; int pos = 0; @@ -471,7 +474,7 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, const Odr_oid *attributeset, int num_bases, char **basenames, int *position, int *num_entries, ZebraScanEntry **list, - int *is_partial, RSET limit_set) + int *is_partial, const char *set_name) { int base_no; int ords[RPN_MAX_ORDS], ord_no = 0; @@ -484,6 +487,7 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, NMEM nmem; ZEBRA_RES res; struct rset_key_control *kc = 0; + RSET limit_set = 0; *list = 0; *is_partial = 0; @@ -491,37 +495,38 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, if (!attributeset) attributeset = yaz_oid_attset_bib_1; - if (!limit_set) /* no limit set given already */ + if (!set_name) { /* see if there is a @attr 8=set */ AttrType termset; int termset_value_numeric; - const char *termset_value_string; + const char *termset_value_string = 0; attr_init_APT(&termset, zapt, 8); termset_value_numeric = attr_find_ex(&termset, NULL, &termset_value_string); if (termset_value_numeric != -1) { - char resname[32]; - const char *termset_name = 0; - if (termset_value_numeric != -2) { + char resname[32]; sprintf(resname, "%d", termset_value_numeric); - termset_name = resname; + set_name = odr_strdup(stream, resname); } else - termset_name = termset_value_string; - - limit_set = resultSetRef(zh, termset_name); + set_name = odr_strdup(stream, termset_value_string); + } + } - if (!limit_set) - { - zebra_setError(zh, - YAZ_BIB1_SPECIFIED_RESULT_SET_DOES_NOT_EXIST, - termset_name); - return ZEBRA_FAIL; - } + if (set_name) + { + limit_set = resultSetRef(zh, set_name); + + if (!limit_set) + { + zebra_setError(zh, + YAZ_BIB1_SPECIFIED_RESULT_SET_DOES_NOT_EXIST, + set_name); + return ZEBRA_FAIL; } } @@ -540,11 +545,6 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, zebra_setError(zh, YAZ_BIB1_TOO_MANY_DATABASES_SPECIFIED, 0); return ZEBRA_FAIL; } - if (sort_flag) - { - return rpn_facet(zh, stream, zapt, attributeset, position, num_entries, - list, is_partial, limit_set); - } for (base_no = 0; base_no < num_bases; base_no++) { int ord; @@ -575,8 +575,8 @@ ZEBRA_RES rpn_scan(ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt, kc = zebra_key_control_create(zh); res = rpn_scan_norm(zh, stream, nmem, kc, zapt, position, num_entries, - list, - is_partial, limit_set, index_type, ord_no, ords); + list, is_partial, limit_set, + index_type, ord_no, ords); nmem_destroy(nmem); (*kc->dec)(kc); return res;