X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=index%2Fextract.c;h=3a6f8c2f766214c186df1ed6c1659aea8781a2a1;hb=1bf3a26f6bb1fc5090047457bf5fd7ae87bc2aae;hp=45fde5d7c362d8c7af3b93eb8e59a592033c7274;hpb=52824e358151328181178232866be5766dcfce6c;p=idzebra-moved-to-github.git diff --git a/index/extract.c b/index/extract.c index 45fde5d..3a6f8c2 100644 --- a/index/extract.c +++ b/index/extract.c @@ -1,4 +1,4 @@ -/* $Id: extract.c,v 1.199 2005-11-10 11:25:47 adam Exp $ +/* $Id: extract.c,v 1.202 2006-02-09 08:31:02 adam Exp $ Copyright (C) 1995-2005 Index Data ApS @@ -32,6 +32,7 @@ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA #include #include "index.h" +#include "orddict.h" #include #include @@ -206,6 +207,8 @@ static void file_end (void *handle, off_t offset) } } +#define FILE_MATCH_BLANK "\t " + static char *fileMatchStr (ZebraHandle zh, zebra_rec_keys_t reckeys, const char *fname, const char *spec) @@ -216,8 +219,8 @@ static char *fileMatchStr (ZebraHandle zh, while (1) { - while (*s == ' ' || *s == '\t') - s++; + for (; *s && strchr(FILE_MATCH_BLANK, *s); s++) + ; if (!*s) break; if (*s == '(') @@ -226,21 +229,26 @@ static char *fileMatchStr (ZebraHandle zh, char attset_str[64], attname_str[64]; data1_attset *attset; int i; - char matchFlag[32]; int attSet = 1, attUse = 1; int first = 1; - - s++; - for (i = 0; *s && *s != ',' && *s != ')'; s++) - if (i < 63) + + for (s++; strchr(FILE_MATCH_BLANK, *s); s++) + ; + for (i = 0; *s && *s != ',' && *s != ')' && + !strchr(FILE_MATCH_BLANK, *s); s++) + if (i+1 < sizeof(attset_str)) attset_str[i++] = *s; attset_str[i] = '\0'; - + + for (; strchr(FILE_MATCH_BLANK, *s); s++) + ; if (*s == ',') { - s++; - for (i = 0; *s && *s != ')'; s++) - if (i < 63) + for (s++; strchr(FILE_MATCH_BLANK, *s); s++) + ; + for (i = 0; *s && *s != ')' && + !strchr(FILE_MATCH_BLANK, *s); s++) + if (i+1 < sizeof(attname_str)) attname_str[i++] = *s; attname_str[i] = '\0'; } @@ -257,12 +265,7 @@ static char *fileMatchStr (ZebraHandle zh, } searchRecordKey (zh, reckeys, attSet, attUse, ws, 32); - if (*s == ')') - { - for (i = 0; i<32; i++) - matchFlag[i] = 1; - } - else + if (*s != ')') { yaz_log (YLOG_WARN, "Missing ) in match criteria %s in group %s", spec, zh->m_group ? zh->m_group : "none"); @@ -271,7 +274,7 @@ static char *fileMatchStr (ZebraHandle zh, s++; for (i = 0; i<32; i++) - if (matchFlag[i] && ws[i]) + if (ws[i]) { if (first) { @@ -294,12 +297,12 @@ static char *fileMatchStr (ZebraHandle zh, char special[64]; const char *spec_src = NULL; const char *s1 = ++s; - while (*s1 && *s1 != ' ' && *s1 != '\t') + while (*s1 && !strchr(FILE_MATCH_BLANK, *s1)) s1++; spec_len = s1 - s; - if (spec_len > 63) - spec_len = 63; + if (spec_len > sizeof(special)-1) + spec_len = sizeof(special)-1; memcpy (special, s, spec_len); special[spec_len] = '\0'; s = s1; @@ -329,7 +332,7 @@ static char *fileMatchStr (ZebraHandle zh, while (*s && *s != stopMarker) { - if (i < 63) + if (i+1 < sizeof(tmpString)) tmpString[i++] = *s++; } if (*s) @@ -490,7 +493,9 @@ static int file_extract_record(ZebraHandle zh, } if (matchStr) { - char *rinfo = dict_lookup (zh->reg->matchDict, matchStr); + int db_ord = zebraExplain_get_database_ord(zh->reg->zei); + char *rinfo = dict_lookup_ord(zh->reg->matchDict, db_ord, + matchStr); if (rinfo) { assert(*rinfo == sizeof(*sysno)); @@ -533,7 +538,9 @@ static int file_extract_record(ZebraHandle zh, if (matchStr) { - dict_insert (zh->reg->matchDict, matchStr, sizeof(*sysno), sysno); + int db_ord = zebraExplain_get_database_ord(zh->reg->zei); + dict_insert_ord(zh->reg->matchDict, db_ord, matchStr, + sizeof(*sysno), sysno); } #if NATTR extract_flushSortKeys (zh, *sysno, 1, zh->reg->sortKeys); @@ -595,7 +602,10 @@ static int file_extract_record(ZebraHandle zh, zh->m_record_type, fname, recordOffset); zh->records_deleted++; if (matchStr) - dict_delete (zh->reg->matchDict, matchStr); + { + int db_ord = zebraExplain_get_database_ord(zh->reg->zei); + dict_delete_ord(zh->reg->matchDict, db_ord, matchStr); + } rec_del (zh->reg->records, &rec); } rec_rm (&rec); @@ -845,6 +855,7 @@ ZEBRA_RES buffer_extract_record(ZebraHandle zh, int force_update, int allow_update) { + SYSNO sysno0 = 0; RecordAttr *recordAttr; struct recExtractCtrl extractCtrl; int r; @@ -942,14 +953,14 @@ ZEBRA_RES buffer_extract_record(ZebraHandle zh, yaz_log (YLOG_WARN, "extract error: no such filter"); return ZEBRA_FAIL; } - /* match criteria */ - matchStr = NULL; if (extractCtrl.match_criteria[0]) match_criteria = extractCtrl.match_criteria; - if (! *sysno) { - char *rinfo; + if (!sysno) { + + sysno = &sysno0; + if (match_criteria && *match_criteria) { matchStr = match_criteria; } else { @@ -963,8 +974,11 @@ ZEBRA_RES buffer_extract_record(ZebraHandle zh, } } } - if (matchStr) { - rinfo = dict_lookup (zh->reg->matchDict, matchStr); + if (matchStr) + { + int db_ord = zebraExplain_get_database_ord(zh->reg->zei); + char *rinfo = dict_lookup_ord(zh->reg->matchDict, db_ord, + matchStr); if (rinfo) { assert(*rinfo == sizeof(*sysno)); @@ -1002,8 +1016,9 @@ ZEBRA_RES buffer_extract_record(ZebraHandle zh, if (matchStr) { - dict_insert (zh->reg->matchDict, matchStr, - sizeof(*sysno), sysno); + int db_ord = zebraExplain_get_database_ord(zh->reg->zei); + dict_insert_ord(zh->reg->matchDict, db_ord, matchStr, + sizeof(*sysno), sysno); } #if NATTR extract_flushSortKeys (zh, *sysno, 1, zh->reg->sortKeys); @@ -1079,7 +1094,10 @@ ZEBRA_RES buffer_extract_record(ZebraHandle zh, pr_fname, (long) recordOffset); zh->records_deleted++; if (matchStr) - dict_delete (zh->reg->matchDict, matchStr); + { + int db_ord = zebraExplain_get_database_ord(zh->reg->zei); + dict_delete_ord(zh->reg->matchDict, db_ord, matchStr); + } rec_del (zh->reg->records, &rec); } rec_rm (&rec);