X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=index%2Fextract.c;h=c70e6a1ee26b25b37476e21f23e0f53442ca11ba;hb=290a65e5f45f2c5c774489e4175beaec23f5ea23;hp=0eec47e898066ea36e1652fcf0cf49dc487785d2;hpb=4b903c542156253ebb2a4f004b528fe9e3af5212;p=idzebra-moved-to-github.git diff --git a/index/extract.c b/index/extract.c index 0eec47e..c70e6a1 100644 --- a/index/extract.c +++ b/index/extract.c @@ -1,4 +1,4 @@ -/* $Id: extract.c,v 1.264 2007-10-29 13:43:57 adam Exp $ +/* $Id: extract.c,v 1.267 2007-10-31 16:56:14 adam Exp $ Copyright (C) 1995-2007 Index Data ApS @@ -67,9 +67,6 @@ static void zebra_init_log_level(void) } } -static void extract_flush_record_keys(ZebraHandle zh, zint sysno, - int cmd, zebra_rec_keys_t reckeys, - zint staticrank); static void extract_flush_sort_keys(ZebraHandle zh, zint sysno, int cmd, zebra_rec_keys_t skp); static void extract_schema_add(struct recExtractCtrl *p, Odr_oid *oid); @@ -100,14 +97,6 @@ static void logRecord(ZebraHandle zh) static void init_extractCtrl(ZebraHandle zh, struct recExtractCtrl *ctrl) { - int i; - for (i = 0; i<256; i++) - { - if (zebra_maps_is_positioned(zh->reg->zebra_maps, i)) - ctrl->seqno[i] = 1; - else - ctrl->seqno[i] = 0; - } ctrl->flagShowRecords = !zh->m_flag_rw; } @@ -122,7 +111,7 @@ static void extract_init(struct recExtractCtrl *p, RecWord *w) { w->seqno = 1; w->index_name = "any"; - w->index_type = 'w'; + w->index_type = "w"; w->extractCtrl = p; w->record_id = 0; w->section_id = 0; @@ -135,10 +124,10 @@ struct snip_rec_info { }; -static void snippet_add_complete_field(RecWord *p, int ord) +static void snippet_add_complete_field(RecWord *p, int ord, + zebra_map_t zm) { struct snip_rec_info *h = p->extractCtrl->handle; - ZebraHandle zh = h->zh; const char *b = p->term_buf; char buf[IT_MAX_WORD+1]; @@ -148,7 +137,7 @@ static void snippet_add_complete_field(RecWord *p, int ord) const char *last = 0; if (remain > 0) - map = zebra_maps_input(zh->reg->zebra_maps, p->index_type, &b, remain, 1); + map = zebra_maps_input(zm, &b, remain, 1); while (remain > 0 && i < IT_MAX_WORD) { @@ -162,8 +151,7 @@ static void snippet_add_complete_field(RecWord *p, int ord) { int first = i ? 0 : 1; /* first position */ - map = zebra_maps_input(zh->reg->zebra_maps, p->index_type, - &b, remain, first); + map = zebra_maps_input(zm, &b, remain, first); } else map = 0; @@ -192,8 +180,7 @@ static void snippet_add_complete_field(RecWord *p, int ord) remain = p->term_len - (b - p->term_buf); if (remain > 0) { - map = zebra_maps_input(zh->reg->zebra_maps, p->index_type, &b, - remain, 0); + map = zebra_maps_input(zm, &b, remain, 0); } else map = 0; @@ -206,10 +193,9 @@ static void snippet_add_complete_field(RecWord *p, int ord) start, last - start); } -static void snippet_add_incomplete_field(RecWord *p, int ord) +static void snippet_add_incomplete_field(RecWord *p, int ord, zebra_map_t zm) { struct snip_rec_info *h = p->extractCtrl->handle; - ZebraHandle zh = h->zh; const char *b = p->term_buf; int remain = p->term_len; int first = 1; @@ -218,7 +204,7 @@ static void snippet_add_incomplete_field(RecWord *p, int ord) const char *last = b; if (remain > 0) - map = zebra_maps_input(zh->reg->zebra_maps, p->index_type, &b, remain, 0); + map = zebra_maps_input(zm, &b, remain, 0); while (map) { @@ -231,8 +217,7 @@ static void snippet_add_incomplete_field(RecWord *p, int ord) remain = p->term_len - (b - p->term_buf); last = b; if (remain > 0) - map = zebra_maps_input(zh->reg->zebra_maps, p->index_type, &b, - remain, 0); + map = zebra_maps_input(zm, &b, remain, 0); else map = 0; } @@ -256,7 +241,7 @@ static void snippet_add_incomplete_field(RecWord *p, int ord) remain = p->term_len - (b - p->term_buf); last = b; if (remain > 0) - map = zebra_maps_input(zh->reg->zebra_maps, p->index_type, &b, remain, 0); + map = zebra_maps_input(zm, &b, remain, 0); else map = 0; } @@ -266,7 +251,7 @@ static void snippet_add_incomplete_field(RecWord *p, int ord) if (first) { first = 0; - if (zebra_maps_is_first_in_field(zh->reg->zebra_maps, p->index_type)) + if (zebra_maps_is_first_in_field(zm)) { /* first in field marker */ p->seqno++; @@ -285,17 +270,18 @@ static void snippet_token_add(RecWord *p) { struct snip_rec_info *h = p->extractCtrl->handle; ZebraHandle zh = h->zh; + zebra_map_t zm = zebra_map_get(zh->reg->zebra_maps, p->index_type); - if (zebra_maps_is_index(zh->reg->zebra_maps, p->index_type)) + if (zm && zebra_maps_is_index(zm)) { ZebraExplainInfo zei = zh->reg->zei; int ch = zebraExplain_lookup_attr_str( zei, zinfo_index_category_index, p->index_type, p->index_name); - if(zebra_maps_is_complete (h->zh->reg->zebra_maps, p->index_type)) - snippet_add_complete_field(p, ch); + if (zebra_maps_is_complete(zm)) + snippet_add_complete_field(p, ch, zm); else - snippet_add_incomplete_field(p, ch); + snippet_add_incomplete_field(p, ch, zm); } } @@ -351,11 +337,11 @@ static void searchRecordKey(ZebraHandle zh, ws[i] = NULL; if (ch < 0) - ch = zebraExplain_lookup_attr_str(zh->reg->zei, cat, '0', index_name); + ch = zebraExplain_lookup_attr_str(zh->reg->zei, cat, "0", index_name); if (ch < 0) - ch = zebraExplain_lookup_attr_str(zh->reg->zei, cat, 'p', index_name); + ch = zebraExplain_lookup_attr_str(zh->reg->zei, cat, "p", index_name); if (ch < 0) - ch = zebraExplain_lookup_attr_str(zh->reg->zei, cat, 'w', index_name); + ch = zebraExplain_lookup_attr_str(zh->reg->zei, cat, "w", index_name); if (ch < 0) return ; @@ -540,7 +526,7 @@ static void all_matches_add(struct recExtractCtrl *ctrl) RecWord word; extract_init(ctrl, &word); word.index_name = "_ALLRECORDS"; - word.index_type = 'w'; + word.index_type = "w"; word.seqno = 1; extract_add_index_string(&word, zinfo_index_category_alwaysmatches, "", 0); @@ -1272,7 +1258,8 @@ void extract_rec_keys_log(ZebraHandle zh, int is_insert, char keystr[200]; /* room for zints to print */ char *dst_term = 0; int ord = CAST_ZINT_TO_INT(key.mem[0]); - int index_type, i; + const char *index_type; + int i; const char *string_index; zebraExplain_lookup_ord(zh->reg->zei, ord, &index_type, @@ -1306,12 +1293,12 @@ void extract_rec_keys_log(ZebraHandle zh, int is_insert, { sprintf(dst_buf + strlen(dst_buf), " %d", str[i] & 0xff); } - yaz_log(level, "%s%c %s %s", keystr, index_type, + yaz_log(level, "%s%s %s %s", keystr, index_type, string_index, dst_buf); } else - yaz_log(level, "%s%c %s \"%s\"", keystr, index_type, + yaz_log(level, "%s%s %s \"%s\"", keystr, index_type, string_index, dst_term); nmem_reset(nmem); @@ -1445,47 +1432,6 @@ void extract_flush_record_keys2(ZebraHandle zh, zint sysno, yaz_log(log_level_extract, "normal=%d optimized=%d", normal, optimized); } -void extract_flush_record_keys(ZebraHandle zh, zint sysno, int cmd, - zebra_rec_keys_t reckeys, - zint staticrank) -{ - ZebraExplainInfo zei = zh->reg->zei; - - extract_rec_keys_adjust(zh, cmd, reckeys); - - if (log_level_details) - { - yaz_log(log_level_details, "Keys for record " ZINT_FORMAT " %s", - sysno, cmd ? "insert" : "delete"); - extract_rec_keys_log(zh, cmd, reckeys, log_level_details); - } - - if (!zh->reg->key_block) - { - int mem = 1024*1024 * atoi( res_get_def( zh->res, "memmax", "8")); - const char *key_tmp_dir = res_get_def(zh->res, "keyTmpDir", "."); - int use_threads = atoi(res_get_def(zh->res, "threads", "1")); - zh->reg->key_block = key_block_create(mem, key_tmp_dir, use_threads); - } - zebraExplain_recordCountIncrement(zei, cmd ? 1 : -1); - -#if 0 - yaz_log(YLOG_LOG, "sysno=" ZINT_FORMAT " cmd=%d", sysno, cmd); - print_rec_keys(zh, reckeys); -#endif - if (zebra_rec_keys_rewind(reckeys)) - { - size_t slen; - const char *str; - struct it_key key_in; - while(zebra_rec_keys_read(reckeys, &str, &slen, &key_in)) - { - key_block_write(zh->reg->key_block, sysno, - &key_in, cmd, str, slen, - staticrank, zh->m_staticrank); - } - } -} ZEBRA_RES zebra_rec_keys_to_snippets(ZebraHandle zh, zebra_rec_keys_t reckeys, @@ -1502,7 +1448,7 @@ ZEBRA_RES zebra_rec_keys_to_snippets(ZebraHandle zh, char *dst_term = 0; int ord; zint seqno; - int index_type; + const char *index_type; assert(key.len <= IT_KEY_LEVEL_MAX && key.len > 2); seqno = key.mem[key.len-1]; @@ -1533,7 +1479,7 @@ void print_rec_keys(ZebraHandle zh, zebra_rec_keys_t reckeys) { char dst_buf[IT_MAX_WORD]; zint seqno; - int index_type; + const char *index_type; int ord = CAST_ZINT_TO_INT(key.mem[0]); const char *db = 0; assert(key.len <= IT_KEY_LEVEL_MAX && key.len > 2); @@ -1607,19 +1553,19 @@ static void extract_add_staticrank_string(RecWord *p, ctrl->staticrank = atozint(valz); } -static void extract_add_string(RecWord *p, const char *string, int length) +static void extract_add_string(RecWord *p, zebra_map_t zm, + const char *string, int length) { - ZebraHandle zh = p->extractCtrl->handle; assert(length > 0); if (!p->index_name) return; - if (zebra_maps_is_index(zh->reg->zebra_maps, p->index_type)) + if (zebra_maps_is_index(zm)) { extract_add_index_string(p, zinfo_index_category_index, string, length); - if (zebra_maps_is_alwaysmatches(zh->reg->zebra_maps, p->index_type)) + if (zebra_maps_is_alwaysmatches(zm)) { RecWord word; memcpy(&word, p, sizeof(word)); @@ -1629,26 +1575,25 @@ static void extract_add_string(RecWord *p, const char *string, int length) &word, zinfo_index_category_alwaysmatches, "", 0); } } - else if (zebra_maps_is_sort(zh->reg->zebra_maps, p->index_type)) + else if (zebra_maps_is_sort(zm)) { extract_add_sort_string(p, string, length); } - else if (zebra_maps_is_staticrank(zh->reg->zebra_maps, p->index_type)) + else if (zebra_maps_is_staticrank(zm)) { extract_add_staticrank_string(p, string, length); } } -static void extract_add_incomplete_field(RecWord *p) +static void extract_add_incomplete_field(RecWord *p, zebra_map_t zm) { - ZebraHandle zh = p->extractCtrl->handle; const char *b = p->term_buf; int remain = p->term_len; int first = 1; const char **map = 0; if (remain > 0) - map = zebra_maps_input(zh->reg->zebra_maps, p->index_type, &b, remain, 0); + map = zebra_maps_input(zm, &b, remain, 0); while (map) { @@ -1660,8 +1605,7 @@ static void extract_add_incomplete_field(RecWord *p) { remain = p->term_len - (b - p->term_buf); if (remain > 0) - map = zebra_maps_input(zh->reg->zebra_maps, p->index_type, &b, - remain, 0); + map = zebra_maps_input(zm, &b, remain, 0); else map = 0; } @@ -1676,7 +1620,7 @@ static void extract_add_incomplete_field(RecWord *p) buf[i++] = *(cp++); remain = p->term_len - (b - p->term_buf); if (remain > 0) - map = zebra_maps_input(zh->reg->zebra_maps, p->index_type, &b, remain, 0); + map = zebra_maps_input(zm, &b, remain, 0); else map = 0; } @@ -1686,28 +1630,27 @@ static void extract_add_incomplete_field(RecWord *p) if (first) { first = 0; - if (zebra_maps_is_first_in_field(zh->reg->zebra_maps, p->index_type)) + if (zebra_maps_is_first_in_field(zm)) { /* first in field marker */ - extract_add_string(p, FIRST_IN_FIELD_STR, FIRST_IN_FIELD_LEN); + extract_add_string(p, zm, FIRST_IN_FIELD_STR, FIRST_IN_FIELD_LEN); p->seqno++; } } - extract_add_string(p, buf, i); + extract_add_string(p, zm, buf, i); p->seqno++; } } -static void extract_add_complete_field(RecWord *p) +static void extract_add_complete_field(RecWord *p, zebra_map_t zm) { - ZebraHandle zh = p->extractCtrl->handle; const char *b = p->term_buf; char buf[IT_MAX_WORD+1]; const char **map = 0; int i = 0, remain = p->term_len; if (remain > 0) - map = zebra_maps_input(zh->reg->zebra_maps, p->index_type, &b, remain, 1); + map = zebra_maps_input(zm, &b, remain, 1); while (remain > 0 && i < IT_MAX_WORD) { @@ -1718,7 +1661,7 @@ static void extract_add_complete_field(RecWord *p) if (remain > 0) { int first = i ? 0 : 1; /* first position */ - map = zebra_maps_input(zh->reg->zebra_maps, p->index_type, &b, remain, first); + map = zebra_maps_input(zm, &b, remain, first); } else map = 0; @@ -1746,8 +1689,7 @@ static void extract_add_complete_field(RecWord *p) remain = p->term_len - (b - p->term_buf); if (remain > 0) { - map = zebra_maps_input(zh->reg->zebra_maps, p->index_type, &b, - remain, 0); + map = zebra_maps_input(zm, &b, remain, 0); } else map = 0; @@ -1755,7 +1697,7 @@ static void extract_add_complete_field(RecWord *p) } if (!i) return; - extract_add_string(p, buf, i); + extract_add_string(p, zm, buf, i); } static void extract_token_add2_index(ZebraHandle zh, zebra_index_type_t type, @@ -1792,12 +1734,8 @@ static void extract_token_add2_index(ZebraHandle zh, zebra_index_type_t type, static void extract_token_add2(RecWord *p) { - zebra_index_type_t type; ZebraHandle zh = p->extractCtrl->handle; - char type_tmp[2]; - type_tmp[0] = p->index_type; - type_tmp[1] = '\0'; - type = zebra_index_type_get(zh->reg->index_types, type_tmp); + zebra_index_type_t type = zebra_index_type_get(zh->reg->index_types, p->index_type); if (type) { if (zebra_index_type_is_index(type)) @@ -1830,25 +1768,25 @@ static void extract_token_add2(RecWord *p) static void extract_token_add(RecWord *p) { ZebraHandle zh = p->extractCtrl->handle; + zebra_map_t zm = zebra_map_get_or_add(zh->reg->zebra_maps, p->index_type); WRBUF wrbuf; if (log_level_details) { yaz_log(log_level_details, "extract_token_add " - "type=%c index=%s seqno=" ZINT_FORMAT " s=%.*s", + "type=%s index=%s seqno=" ZINT_FORMAT " s=%.*s", p->index_type, p->index_name, p->seqno, p->term_len, p->term_buf); } - if ((wrbuf = zebra_replace(zh->reg->zebra_maps, p->index_type, 0, - p->term_buf, p->term_len))) + if ((wrbuf = zebra_replace(zm, 0, p->term_buf, p->term_len))) { p->term_buf = wrbuf_buf(wrbuf); p->term_len = wrbuf_len(wrbuf); } - if (zebra_maps_is_complete(zh->reg->zebra_maps, p->index_type)) - extract_add_complete_field(p); + if (zebra_maps_is_complete(zm)) + extract_add_complete_field(p, zm); else - extract_add_incomplete_field(p); + extract_add_incomplete_field(p, zm); } static void extract_set_store_data_cb(struct recExtractCtrl *p,