X-Git-Url: http://git.indexdata.com/?p=idzebra-moved-to-github.git;a=blobdiff_plain;f=index%2Fextract.c;h=0d648a6d7de890fbfde525d6f41f7fbf35b54a06;hp=dc951f389c543978130446eabdfe3fd4d39ffc54;hb=40869f1460c8b3804904ec207b18c5607f82de6e;hpb=fc458befb0a78a1de6629f6873d9a467be1393c1 diff --git a/index/extract.c b/index/extract.c index dc951f3..0d648a6 100644 --- a/index/extract.c +++ b/index/extract.c @@ -1,8 +1,5 @@ -/* $Id: extract.c,v 1.276 2007-12-20 11:21:29 adam Exp $ - Copyright (C) 1995-2007 - Index Data ApS - -This file is part of the Zebra server. +/* This file is part of the Zebra server. + Copyright (C) 1995-2008 Index Data 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 @@ -187,7 +184,7 @@ static void snippet_add_complete_field(RecWord *p, int ord, } if (!i) return; - if (last && start != last) + if (last && start != last && zebra_maps_is_index(zm)) zebra_snippets_appendn(h->snippets, p->seqno, 0, ord, start, last - start); } @@ -222,7 +219,7 @@ static void snippet_add_incomplete_field(RecWord *p, int ord, zebra_map_t zm) } if (!map) break; - if (start != last) + if (start != last && zebra_maps_is_index(zm)) { zebra_snippets_appendn(h->snippets, p->seqno, 1, ord, start, last - start); @@ -256,7 +253,7 @@ static void snippet_add_incomplete_field(RecWord *p, int ord, zebra_map_t zm) p->seqno++; } } - if (start != last) + if (start != last && zebra_maps_is_index(zm)) zebra_snippets_appendn(h->snippets, p->seqno, 0, ord, start, last - start); start = last; @@ -279,8 +276,9 @@ static void snippet_add_icu(RecWord *p, int ord, zebra_map_t zm) while (zebra_map_tokenize_next(zm, &res_buf, &res_len, &display_buf, &display_len)) { - zebra_snippets_appendn(h->snippets, p->seqno, 0, ord, - display_buf, display_len); + if (zebra_maps_is_index(zm)) + zebra_snippets_appendn(h->snippets, p->seqno, 0, ord, + display_buf, display_len); p->seqno++; } } @@ -291,7 +289,7 @@ static void snippet_token_add(RecWord *p) ZebraHandle zh = h->zh; zebra_map_t zm = zebra_map_get(zh->reg->zebra_maps, p->index_type); - if (zm && zebra_maps_is_index(zm)) + if (zm) { ZebraExplainInfo zei = zh->reg->zei; int ch = zebraExplain_lookup_attr_str( @@ -1614,7 +1612,15 @@ static void extract_add_string(RecWord *p, zebra_map_t zm, if (!p->index_name) return; + if (log_level_details) + { + WRBUF w = wrbuf_alloc(); + + wrbuf_write_escaped(w, string, length); + yaz_log(log_level_details, "extract_add_string: %s", wrbuf_cstr(w)); + wrbuf_destroy(w); + } if (zebra_maps_is_index(zm)) { extract_add_index_string(p, zinfo_index_category_index, @@ -1861,8 +1867,56 @@ void extract_flush_sort_keys(ZebraHandle zh, zint sysno, const char *str; struct it_key key_in; +#define USE_SORT_ENT 1 +#if USE_SORT_ENT + NMEM nmem = nmem_create(); + struct sort_add_ent { + int ord; + int cmd; + struct sort_add_ent *next; + struct zebra_sort_ent sort_ent; + }; + struct sort_add_ent *sort_ent_list = 0; +#endif zebra_sort_sysno(si, sysno); +#if USE_SORT_ENT + while (zebra_rec_keys_read(reckeys, &str, &slen, &key_in)) + { + int ord = CAST_ZINT_TO_INT(key_in.mem[0]); + + struct sort_add_ent **e = &sort_ent_list; + while (*e && (*e)->ord != ord) + e = &(*e)->next; + if (!*e) + { + *e = nmem_malloc(nmem, sizeof(**e)); + (*e)->next = 0; + (*e)->sort_ent.wrbuf = wrbuf_alloc(); + (*e)->sort_ent.num = 0; + (*e)->ord = ord; + (*e)->cmd = cmd; + } + + wrbuf_write((*e)->sort_ent.wrbuf, str, slen); + wrbuf_putc((*e)->sort_ent.wrbuf, '\0'); + (*e)->sort_ent.num++; + } + if (sort_ent_list) + { + struct sort_add_ent *e = sort_ent_list; + for (; e; e = e->next) + { + zebra_sort_type(si, e->ord); + if (e->cmd == 1) + zebra_sort_add_ent(si, &e->sort_ent); + else + zebra_sort_delete(si); + wrbuf_destroy(e->sort_ent.wrbuf); + } + } + nmem_destroy(nmem); +#else while (zebra_rec_keys_read(reckeys, &str, &slen, &key_in)) { int ord = CAST_ZINT_TO_INT(key_in.mem[0]); @@ -1873,6 +1927,7 @@ void extract_flush_sort_keys(ZebraHandle zh, zint sysno, else zebra_sort_delete(si); } +#endif } }