X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=index%2Fextract.c;h=fb13623d24eb3832bcf2ae7bac455b7fdd065195;hb=eb6f5a00932fd50cb95aca3c1dd5039116254f4b;hp=107b4cae0f11320402951e472a8d7b6c33650083;hpb=cef013b96c9508d5fd8552e7fd05962029b37e6f;p=idzebra-moved-to-github.git diff --git a/index/extract.c b/index/extract.c index 107b4ca..fb13623 100644 --- a/index/extract.c +++ b/index/extract.c @@ -1,4 +1,4 @@ -/* $Id: extract.c,v 1.162 2004-09-15 08:13:51 adam Exp $ +/* $Id: extract.c,v 1.170 2004-12-21 21:55:58 adam Exp $ Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002,2003,2004 Index Data Aps @@ -72,7 +72,7 @@ static void logRecord (ZebraHandle zh) ++zh->records_processed; if (!(zh->records_processed % 1000)) { - logf (LOG_LOG, "Records: "ZINT_FORMAT" i/u/d " + yaz_log (YLOG_LOG, "Records: "ZINT_FORMAT" i/u/d " ZINT_FORMAT"/"ZINT_FORMAT"/"ZINT_FORMAT, zh->records_processed, zh->records_inserted, zh->records_updated, zh->records_deleted); @@ -87,6 +87,8 @@ static void extract_init (struct recExtractCtrl *p, RecWord *w) w->attrUse = 1016; w->reg_type = 'w'; w->extractCtrl = p; + w->record_id = 0; + w->section_id = 0; } static const char **searchRecordKey (ZebraHandle zh, @@ -113,9 +115,9 @@ static const char **searchRecordKey (ZebraHandle zh, iscz1_decode(decode_handle, &dst, &src); assert(key.len < 4 && key.len > 2); - attrSet = (int) key.mem[0]; - attrUse = (int) key.mem[1]; - seqno = (int) key.mem[2]; + attrSet = (int) key.mem[0] >> 16; + attrUse = (int) key.mem[0] & 65535; + seqno = (int) key.mem[key.len-1]; if (attrUseS == attrUse && attrSetS == attrSet) { @@ -282,7 +284,7 @@ static char *fileMatchStr (ZebraHandle zh, } else { - logf (LOG_WARN, "Missing ) in match criteria %s in group %s", + yaz_log (YLOG_WARN, "Missing ) in match criteria %s in group %s", spec, zh->m_group ? zh->m_group : "none"); return NULL; } @@ -301,7 +303,7 @@ static char *fileMatchStr (ZebraHandle zh, } if (first) { - logf (LOG_WARN, "Record didn't contain match" + yaz_log (YLOG_WARN, "Record didn't contain match" " fields in (%s,%s)", attset_str, attname_str); return NULL; } @@ -358,7 +360,7 @@ static char *fileMatchStr (ZebraHandle zh, } else { - logf (LOG_WARN, "Syntax error in match criteria %s in group %s", + yaz_log (YLOG_WARN, "Syntax error in match criteria %s in group %s", spec, zh->m_group ? zh->m_group : "none"); return NULL; } @@ -366,7 +368,7 @@ static char *fileMatchStr (ZebraHandle zh, } if (dst == dstBuf) { - logf (LOG_WARN, "No match criteria for record %s in group %s", + yaz_log (YLOG_WARN, "No match criteria for record %s in group %s", fname, zh->m_group ? zh->m_group : "none"); return NULL; } @@ -394,19 +396,18 @@ static int file_extract_record(ZebraHandle zh, { RecordAttr *recordAttr; int r; - const char *matchStr; + const char *matchStr = 0; SYSNO sysnotmp; Record rec; off_t recordOffset = 0; RecType recType; void *clientData; - char subType[128]; if (!(recType = - recType_byName (zh->reg->recTypes, zh->m_record_type, subType, + recType_byName (zh->reg->recTypes, zh->res, zh->m_record_type, &clientData))) { - logf (LOG_WARN, "No such record type: %s", zh->m_record_type); + yaz_log (YLOG_WARN, "No such record type: %s", zh->m_record_type); return 0; } @@ -436,11 +437,11 @@ static int file_extract_record(ZebraHandle zh, extractCtrl.tellf = file_tell; extractCtrl.endf = file_end; extractCtrl.fh = fi; - extractCtrl.subType = subType; extractCtrl.init = extract_init; extractCtrl.tokenAdd = extract_token_add; extractCtrl.schemaAdd = extract_schema_add; extractCtrl.dh = zh->reg->dh; + extractCtrl.match_criteria[0] = '\0'; extractCtrl.handle = zh; for (i = 0; i<256; i++) { @@ -472,7 +473,7 @@ static int file_extract_record(ZebraHandle zh, if (zh->m_flag_rw && zh->records_processed < zh->m_file_verbose_limit) { - logf (LOG_WARN, "fail %s %s " PRINTF_OFF_T, zh->m_record_type, + yaz_log (YLOG_WARN, "fail %s %s " PRINTF_OFF_T, zh->m_record_type, fname, recordOffset); } return 0; @@ -483,7 +484,7 @@ static int file_extract_record(ZebraHandle zh, if (zh->m_flag_rw && zh->records_processed < zh->m_file_verbose_limit) { - logf (LOG_WARN, "no filter for %s %s " + yaz_log (YLOG_WARN, "no filter for %s %s " PRINTF_OFF_T, zh->m_record_type, fname, recordOffset); } @@ -496,40 +497,41 @@ static int file_extract_record(ZebraHandle zh, if (!zh->m_flag_rw) return 1; - logf (LOG_WARN, "empty %s %s " PRINTF_OFF_T, zh->m_record_type, - fname, recordOffset); + if (zh->records_processed < zh->m_file_verbose_limit) + yaz_log (YLOG_WARN, "empty %s %s " PRINTF_OFF_T, zh->m_record_type, + fname, recordOffset); return 1; } + if (extractCtrl.match_criteria[0]) + matchStr = extractCtrl.match_criteria; } /* perform match if sysno not known and if match criteria is specified */ - - matchStr = NULL; if (!sysno) { sysnotmp = 0; sysno = &sysnotmp; - if (zh->m_record_id && *zh->m_record_id) + + if (matchStr == 0 && zh->m_record_id && *zh->m_record_id) { - char *rinfo; matchStr = fileMatchStr (zh, &zh->reg->keys, fname, zh->m_record_id); - if (matchStr) - { - rinfo = dict_lookup (zh->reg->matchDict, matchStr); - if (rinfo) - { - assert(*rinfo == sizeof(*sysno)); - memcpy (sysno, rinfo+1, sizeof(*sysno)); - } - } - else - { - logf (LOG_WARN, "Bad match criteria"); - return 0; - } - } + if (!matchStr) + { + yaz_log(YLOG_WARN, "Bad match criteria"); + return 0; + } + } + if (matchStr) + { + char *rinfo = dict_lookup (zh->reg->matchDict, matchStr); + if (rinfo) + { + assert(*rinfo == sizeof(*sysno)); + memcpy (sysno, rinfo+1, sizeof(*sysno)); + } + } } if (! *sysno) @@ -537,13 +539,13 @@ static int file_extract_record(ZebraHandle zh, /* new record */ if (deleteFlag) { - logf (LOG_LOG, "delete %s %s " PRINTF_OFF_T, zh->m_record_type, + yaz_log (YLOG_LOG, "delete %s %s " PRINTF_OFF_T, zh->m_record_type, fname, recordOffset); - logf (LOG_WARN, "cannot delete record above (seems new)"); + yaz_log (YLOG_WARN, "cannot delete record above (seems new)"); return 1; } if (zh->records_processed < zh->m_file_verbose_limit) - logf (LOG_LOG, "add %s %s " PRINTF_OFF_T, zh->m_record_type, + yaz_log (YLOG_LOG, "add %s %s " PRINTF_OFF_T, zh->m_record_type, fname, recordOffset); rec = rec_new (zh->reg->records); @@ -574,8 +576,8 @@ static int file_extract_record(ZebraHandle zh, if (!force_update && recordAttr->runNumber == zebraExplain_runNumberIncrement (zh->reg->zei, 0)) { - yaz_log (LOG_LOG, "run number = %d", recordAttr->runNumber); - yaz_log (LOG_LOG, "skipped %s %s " PRINTF_OFF_T, + yaz_log (YLOG_LOG, "run number = %d", recordAttr->runNumber); + yaz_log (YLOG_LOG, "skipped %s %s " PRINTF_OFF_T, zh->m_record_type, fname, recordOffset); extract_flushSortKeys (zh, *sysno, -1, &zh->reg->sortKeys); rec_rm (&rec); @@ -595,14 +597,14 @@ static int file_extract_record(ZebraHandle zh, /* record going to be deleted */ if (!delkeys.buf_used) { - logf (LOG_LOG, "delete %s %s " PRINTF_OFF_T, + yaz_log (YLOG_LOG, "delete %s %s " PRINTF_OFF_T, zh->m_record_type, fname, recordOffset); - logf (LOG_WARN, "cannot delete file above, storeKeys false"); + yaz_log (YLOG_WARN, "cannot delete file above, storeKeys false"); } else { if (zh->records_processed < zh->m_file_verbose_limit) - logf (LOG_LOG, "delete %s %s " PRINTF_OFF_T, + yaz_log (YLOG_LOG, "delete %s %s " PRINTF_OFF_T, zh->m_record_type, fname, recordOffset); zh->records_deleted++; if (matchStr) @@ -618,14 +620,14 @@ static int file_extract_record(ZebraHandle zh, /* record going to be updated */ if (!delkeys.buf_used) { - logf (LOG_LOG, "update %s %s " PRINTF_OFF_T, + yaz_log (YLOG_LOG, "update %s %s " PRINTF_OFF_T, zh->m_record_type, fname, recordOffset); - logf (LOG_WARN, "cannot update file above, storeKeys false"); + yaz_log (YLOG_WARN, "cannot update file above, storeKeys false"); } else { if (zh->records_processed < zh->m_file_verbose_limit) - logf (LOG_LOG, "update %s %s " PRINTF_OFF_T, + yaz_log (YLOG_LOG, "update %s %s " PRINTF_OFF_T, zh->m_record_type, fname, recordOffset); extract_flushSortKeys (zh, *sysno, 1, &zh->reg->sortKeys); extract_flushRecordKeys (zh, *sysno, 1, &zh->reg->keys); @@ -688,14 +690,14 @@ static int file_extract_record(ZebraHandle zh, xmalloc (recordAttr->recordSize); if (lseek (fi->fd, recordOffset, SEEK_SET) < 0) { - logf (LOG_ERRNO|LOG_FATAL, "seek to " PRINTF_OFF_T " in %s", + yaz_log (YLOG_ERRNO|YLOG_FATAL, "seek to " PRINTF_OFF_T " in %s", recordOffset, fname); exit (1); } if (read (fi->fd, rec->info[recInfo_storeData], recordAttr->recordSize) < recordAttr->recordSize) { - logf (LOG_ERRNO|LOG_FATAL, "read %d bytes of %s", + yaz_log (YLOG_ERRNO|YLOG_FATAL, "read %d bytes of %s", recordAttr->recordSize, fname); exit (1); } @@ -734,7 +736,7 @@ int fileExtract (ZebraHandle zh, SYSNO *sysno, const char *fname, else sprintf (gprefix, "%s.", zh->m_group); - logf (LOG_DEBUG, "fileExtract %s", fname); + yaz_log (YLOG_DEBUG, "fileExtract %s", fname); /* determine file extension */ *ext = '\0'; @@ -756,7 +758,7 @@ int fileExtract (ZebraHandle zh, SYSNO *sysno, const char *fname, if (!zh->m_record_type) { if (zh->records_processed < zh->m_file_verbose_limit) - logf (LOG_LOG, "? %s", fname); + yaz_log (YLOG_LOG, "? %s", fname); return 0; } /* determine match criteria */ @@ -784,7 +786,7 @@ int fileExtract (ZebraHandle zh, SYSNO *sysno, const char *fname, if ((fd = open (full_rep, O_BINARY|O_RDONLY)) == -1) { - logf (LOG_WARN|LOG_ERRNO, "open %s", full_rep); + yaz_log (YLOG_WARN|YLOG_ERRNO, "open %s", full_rep); zh->m_record_type = original_record_type; return 0; } @@ -824,7 +826,6 @@ int buffer_extract_record (ZebraHandle zh, int i, r; const char *matchStr = 0; RecType recType = NULL; - char subType[1024]; void *clientData; Record rec; long recordOffset = 0; @@ -860,26 +861,25 @@ int buffer_extract_record (ZebraHandle zh, } if (recordType && *recordType) { - logf (LOG_DEBUG, "Record type explicitly specified: %s", recordType); - recType = recType_byName (zh->reg->recTypes, recordType, subType, + yaz_log (YLOG_DEBUG, "Record type explicitly specified: %s", recordType); + recType = recType_byName (zh->reg->recTypes, zh->res, recordType, &clientData); } else { if (!(zh->m_record_type)) { - logf (LOG_WARN, "No such record type defined"); + yaz_log (YLOG_WARN, "No such record type defined"); return 0; } - logf (LOG_DEBUG, "Get record type from rgroup: %s",zh->m_record_type); - recType = recType_byName (zh->reg->recTypes, zh->m_record_type, subType, - &clientData); + yaz_log (YLOG_DEBUG, "Get record type from rgroup: %s",zh->m_record_type); + recType = recType_byName (zh->reg->recTypes, zh->res, + zh->m_record_type, &clientData); recordType = zh->m_record_type; } if (!recType) { - logf (LOG_WARN, "No such record type: %s", zh->m_record_type); + yaz_log (YLOG_WARN, "No such record type: %s", zh->m_record_type); return 0; } - extractCtrl.subType = subType; extractCtrl.init = extract_init; extractCtrl.tokenAdd = extract_token_add; extractCtrl.schemaAdd = extract_schema_add; @@ -887,6 +887,7 @@ int buffer_extract_record (ZebraHandle zh, extractCtrl.handle = zh; extractCtrl.zebra_maps = zh->reg->zebra_maps; extractCtrl.flagShowRecords = 0; + extractCtrl.match_criteria[0] = '\0'; for (i = 0; i<256; i++) { if (zebra_maps_is_positioned(zh->reg->zebra_maps, i)) @@ -902,13 +903,13 @@ int buffer_extract_record (ZebraHandle zh, else if (r == RECCTRL_EXTRACT_ERROR_GENERIC) { /* error occured during extraction ... */ - yaz_log (LOG_WARN, "extract error: generic"); + yaz_log (YLOG_WARN, "extract error: generic"); return 0; } else if (r == RECCTRL_EXTRACT_ERROR_NO_SUCH_FILTER) { /* error occured during extraction ... */ - yaz_log (LOG_WARN, "extract error: no such filter"); + yaz_log (YLOG_WARN, "extract error: no such filter"); return 0; } if (zh->reg->keys.buf_used == 0) @@ -917,13 +918,16 @@ int buffer_extract_record (ZebraHandle zh, is probably empty - unless flagShowRecords is in use */ if (test_mode) return 1; - logf (LOG_WARN, "No keys generated for record"); - logf (LOG_WARN, " The file is probably empty"); + yaz_log (YLOG_WARN, "No keys generated for record"); + yaz_log (YLOG_WARN, " The file is probably empty"); return 1; } /* match criteria */ matchStr = NULL; + if (extractCtrl.match_criteria[0]) + match_criteria = extractCtrl.match_criteria; + if (! *sysno) { char *rinfo; if (match_criteria && *match_criteria) { @@ -934,7 +938,7 @@ int buffer_extract_record (ZebraHandle zh, zh->m_record_id); if (!matchStr) { - logf (LOG_WARN, "Bad match criteria (recordID)"); + yaz_log (YLOG_WARN, "Bad match criteria (recordID)"); return 1; } } @@ -954,12 +958,12 @@ int buffer_extract_record (ZebraHandle zh, /* new record */ if (delete_flag) { - logf (LOG_LOG, "delete %s %s %ld", recordType, + yaz_log (YLOG_LOG, "delete %s %s %ld", recordType, pr_fname, (long) recordOffset); - logf (LOG_WARN, "cannot delete record above (seems new)"); + yaz_log (YLOG_WARN, "cannot delete record above (seems new)"); return 1; } - logf (LOG_LOG, "add %s %s %ld", recordType, pr_fname, + yaz_log (YLOG_LOG, "add %s %s %ld", recordType, pr_fname, (long) recordOffset); rec = rec_new (zh->reg->records); @@ -984,7 +988,7 @@ int buffer_extract_record (ZebraHandle zh, struct sortKeys sortKeys; if (!allow_update) { - logf (LOG_LOG, "skipped %s %s %ld", + yaz_log (YLOG_LOG, "skipped %s %s %ld", recordType, pr_fname, (long) recordOffset); logRecord(zh); return -1; @@ -999,7 +1003,7 @@ int buffer_extract_record (ZebraHandle zh, if (recordAttr->runNumber == zebraExplain_runNumberIncrement (zh->reg->zei, 0)) { - logf (LOG_LOG, "skipped %s %s %ld", recordType, + yaz_log (YLOG_LOG, "skipped %s %s %ld", recordType, pr_fname, (long) recordOffset); extract_flushSortKeys (zh, *sysno, -1, &zh->reg->sortKeys); rec_rm (&rec); @@ -1021,13 +1025,13 @@ int buffer_extract_record (ZebraHandle zh, /* record going to be deleted */ if (!delkeys.buf_used) { - logf (LOG_LOG, "delete %s %s %ld", recordType, + yaz_log (YLOG_LOG, "delete %s %s %ld", recordType, pr_fname, (long) recordOffset); - logf (LOG_WARN, "cannot delete file above, storeKeys false"); + yaz_log (YLOG_WARN, "cannot delete file above, storeKeys false"); } else { - logf (LOG_LOG, "delete %s %s %ld", recordType, + yaz_log (YLOG_LOG, "delete %s %s %ld", recordType, pr_fname, (long) recordOffset); zh->records_deleted++; if (matchStr) @@ -1043,13 +1047,13 @@ int buffer_extract_record (ZebraHandle zh, /* record going to be updated */ if (!delkeys.buf_used) { - logf (LOG_LOG, "update %s %s %ld", recordType, + yaz_log (YLOG_LOG, "update %s %s %ld", recordType, pr_fname, (long) recordOffset); - logf (LOG_WARN, "cannot update file above, storeKeys false"); + yaz_log (YLOG_WARN, "cannot update file above, storeKeys false"); } else { - logf (LOG_LOG, "update %s %s %ld", recordType, + yaz_log (YLOG_LOG, "update %s %s %ld", recordType, pr_fname, (long) recordOffset); extract_flushSortKeys (zh, *sysno, 1, &zh->reg->sortKeys); extract_flushRecordKeys (zh, *sysno, 1, &zh->reg->keys); @@ -1162,6 +1166,7 @@ int explain_extract (void *handle, Record rec, data1_node *n) extractCtrl.seqno[i] = 0; extractCtrl.zebra_maps = zh->reg->zebra_maps; extractCtrl.flagShowRecords = 0; + extractCtrl.match_criteria[0] = '\0'; extractCtrl.handle = handle; if (n) @@ -1212,7 +1217,7 @@ void extract_flushRecordKeys (ZebraHandle zh, SYSNO sysno, int mem= 1024*1024* atoi( res_get_def( zh->res, "memmax", "8")); if (mem <= 0) { - logf(LOG_WARN, "Invalid memory setting, using default 8 MB"); + yaz_log(YLOG_WARN, "Invalid memory setting, using default 8 MB"); mem= 1024*1024*8; } /* FIXME: That "8" should be in a default settings include */ @@ -1233,10 +1238,10 @@ void extract_flushRecordKeys (ZebraHandle zh, SYSNO sysno, int attrSet, attrUse; iscz1_decode(decode_handle, &dst, &src); - assert(key.len < 4 && key.len > 2); + assert(key.len == 4); - attrSet = (int) key.mem[0]; - attrUse = (int) key.mem[1]; /* sequence in mem[2] */ + attrSet = (int) key.mem[0] >> 16; + attrUse = (int) key.mem[0] & 65535; if (zh->reg->key_buf_used + 1024 > (zh->reg->ptr_top -zh->reg->ptr_i)*sizeof(char*)) @@ -1260,10 +1265,14 @@ void extract_flushRecordKeys (ZebraHandle zh, SYSNO sysno, ((char*)(zh->reg->key_buf))[(zh->reg->key_buf_used)++] = '\0'; ((char*)(zh->reg->key_buf))[(zh->reg->key_buf_used)++] = cmd; - key.len = 2; - key.mem[0] = sysno; - key.mem[1] = key.mem[2]; /* sequence .. */ - + key.len = 3; + if (key.mem[1]) /* filter specify record ID */ + key.mem[0] = key.mem[1]; + else + key.mem[0] = sysno; + key.mem[1] = key.mem[2]; /* section_id */ + key.mem[2] = key.mem[3]; /* sequence .. */ + memcpy ((char*)zh->reg->key_buf + zh->reg->key_buf_used, &key, sizeof(key)); (zh->reg->key_buf_used) += sizeof(key); @@ -1289,19 +1298,19 @@ void extract_flushWriteKeys (ZebraHandle zh, int final) #endif if (!zh->reg->key_buf || ptr_i <= 0) { - logf (LOG_DEBUG, " nothing to flush section=%d buf=%p i=%d", + yaz_log (YLOG_DEBUG, " nothing to flush section=%d buf=%p i=%d", zh->reg->key_file_no, zh->reg->key_buf, ptr_i); - logf (LOG_DEBUG, " buf=%p ", + yaz_log (YLOG_DEBUG, " buf=%p ", zh->reg->key_buf); - logf (LOG_DEBUG, " ptr=%d ",zh->reg->ptr_i); - logf (LOG_DEBUG, " reg=%p ",zh->reg); + yaz_log (YLOG_DEBUG, " ptr=%d ",zh->reg->ptr_i); + yaz_log (YLOG_DEBUG, " reg=%p ",zh->reg); return; } (zh->reg->key_file_no)++; - logf (LOG_LOG, "sorting section %d", (zh->reg->key_file_no)); - logf (LOG_DEBUG, " sort_buff at %p n=%d", + yaz_log (YLOG_LOG, "sorting section %d", (zh->reg->key_file_no)); + yaz_log (YLOG_DEBUG, " sort_buff at %p n=%d", zh->reg->key_buf + zh->reg->ptr_top - ptr_i,ptr_i); #if !SORT_EXTRA qsort (zh->reg->key_buf + zh->reg->ptr_top - ptr_i, ptr_i, @@ -1315,7 +1324,7 @@ void extract_flushWriteKeys (ZebraHandle zh, int final) temp_policy=toupper(res_get_def(zh->res,"tempfiles","auto")[0]); if (temp_policy != 'Y' && temp_policy != 'N' && temp_policy != 'A') { - logf (LOG_WARN, "Illegal tempfiles setting '%c'. using 'Auto' ", + yaz_log (YLOG_WARN, "Illegal tempfiles setting '%c'. using 'Auto' ", temp_policy); temp_policy='A'; } @@ -1337,10 +1346,10 @@ void extract_flushWriteKeys (ZebraHandle zh, int final) if (!(outf = fopen (out_fname, "wb"))) { - logf (LOG_FATAL|LOG_ERRNO, "fopen %s", out_fname); + yaz_log (YLOG_FATAL|YLOG_ERRNO, "fopen %s", out_fname); exit (1); } - logf (LOG_LOG, "writing section %d", zh->reg->key_file_no); + yaz_log (YLOG_LOG, "writing section %d", zh->reg->key_file_no); prevcp = cp = (zh->reg->key_buf)[zh->reg->ptr_top - ptr_i]; encode_key_init (&encode_info); @@ -1366,10 +1375,10 @@ void extract_flushWriteKeys (ZebraHandle zh, int final) if (!(outf = fopen (out_fname, "wb"))) { - logf (LOG_FATAL|LOG_ERRNO, "fopen %s", out_fname); + yaz_log (YLOG_FATAL|YLOG_ERRNO, "fopen %s", out_fname); exit (1); } - logf (LOG_LOG, "writing section %d", key_file_no); + yaz_log (YLOG_LOG, "writing section %d", key_file_no); i = ptr_i; prevcp = key_buf[ptr_top-i]; while (1) @@ -1377,7 +1386,7 @@ void extract_flushWriteKeys (ZebraHandle zh, int final) { key_y_len = strlen(prevcp)+1; #if 0 - logf (LOG_LOG, "key_y_len: %2d %02x %02x %s", + yaz_log (YLOG_LOG, "key_y_len: %2d %02x %02x %s", key_y_len, prevcp[0], prevcp[1], 2+prevcp); #endif qsort (key_buf + ptr_top-ptr_i, ptr_i - i, @@ -1399,21 +1408,21 @@ void extract_flushWriteKeys (ZebraHandle zh, int final) #endif if (fclose (outf)) { - logf (LOG_FATAL|LOG_ERRNO, "fclose %s", out_fname); + yaz_log (YLOG_FATAL|YLOG_ERRNO, "fclose %s", out_fname); exit (1); } - logf (LOG_LOG, "finished section %d", zh->reg->key_file_no); + yaz_log (YLOG_LOG, "finished section %d", zh->reg->key_file_no); zh->reg->ptr_i = 0; zh->reg->key_buf_used = 0; } -void extract_add_index_string (RecWord *p, const char *str, int length) +void extract_add_it_key (ZebraHandle zh, + int reg_type, + const char *str, int slen, struct it_key *key) { char *dst; - ZebraHandle zh = p->extractCtrl->handle; struct recKeys *keys = &zh->reg->keys; - struct it_key key; - const char *src = (char*) &key; + const char *src = (char*) key; if (keys->buf_used+1024 > keys->buf_max) { @@ -1427,24 +1436,33 @@ void extract_add_index_string (RecWord *p, const char *str, int length) } dst = keys->buf + keys->buf_used; - key.len = 3; - key.mem[0] = p->attrSet; - key.mem[1] = p->attrUse; - key.mem[2] = p->seqno; + iscz1_encode(keys->codec_handle, &dst, &src); + + *dst++ = reg_type; + memcpy (dst, str, slen); + dst += slen; + *dst++ = '\0'; + keys->buf_used = dst - keys->buf; +} + +void extract_add_index_string (RecWord *p, const char *str, int length) +{ + struct it_key key; + key.len = 4; + key.mem[0] = p->attrSet * 65536 + p->attrUse; + key.mem[1] = p->record_id; + key.mem[2] = p->section_id; + key.mem[3] = p->seqno; #if 0 /* just for debugging .. */ - yaz_log(LOG_LOG, "set=%d use=%d seqno=%d", p->attrSet, p->attrUse, - p->seqno); + yaz_log(YLOG_LOG, "add: set=%d use=%d " + "record_id=%lld section_id=%lld seqno=%lld", + p->attrSet, p->attrUse, p->record_id, p->section_id, p->seqno); #endif - iscz1_encode(keys->codec_handle, &dst, &src); - - *dst++ = p->reg_type; - memcpy (dst, str, length); - dst += length; - *dst++ = '\0'; - keys->buf_used = dst - keys->buf; + extract_add_it_key(p->extractCtrl->handle, p->reg_type, str, + length, &key); } static void extract_add_sort_string (RecWord *p, const char *str, @@ -1498,6 +1516,8 @@ static void extract_add_incomplete_field (RecWord *p) const char *b = p->string; int remain = p->length; const char **map = 0; + + yaz_log(YLOG_DEBUG, "Incomplete field, w='%.*s'", p->length, p->string); if (remain > 0) map = zebra_maps_input(p->zebra_maps, p->reg_type, &b, remain, 0); @@ -1546,7 +1566,7 @@ static void extract_add_complete_field (RecWord *p) int i = 0, remain = p->length; int first; /* first position */ -yaz_log(LOG_DEBUG, "Complete field, w='%s'", p->string); + yaz_log(YLOG_DEBUG, "Complete field, w='%.*s'", p->length, p->string); if (remain > 0) map = zebra_maps_input (p->zebra_maps, p->reg_type, &b, remain, 1); @@ -1582,7 +1602,7 @@ yaz_log(LOG_DEBUG, "Complete field, w='%s'", p->string); { if (i >= IT_MAX_WORD) break; - yaz_log(LOG_DEBUG, "Adding string to index '%d'", **map); + yaz_log(YLOG_DEBUG, "Adding string to index '%d'", **map); while (i < IT_MAX_WORD && *cp) buf[i++] = *(cp++); } @@ -1605,7 +1625,7 @@ void extract_token_add (RecWord *p) { WRBUF wrbuf; #if 0 - yaz_log (LOG_LOG, "token_add " + yaz_log (YLOG_LOG, "token_add " "reg_type=%c attrSet=%d attrUse=%d seqno=%d s=%.*s", p->reg_type, p->attrSet, p->attrUse, p->seqno, p->length, p->string); @@ -1689,7 +1709,7 @@ void encode_key_write (char *k, struct encode_info *i, FILE *outf) *bp0 = (*k * 128) + bp - bp0 - 1; /* length and insert/delete combined */ if (fwrite (i->buf, bp - i->buf, 1, outf) != 1) { - logf (LOG_FATAL|LOG_ERRNO, "fwrite"); + yaz_log (YLOG_FATAL|YLOG_ERRNO, "fwrite"); exit (1); } } @@ -1758,7 +1778,7 @@ void encode_key_write (char *k, struct encode_info *i, FILE *outf) i->cmd = i->prevcmd; if (fwrite (i->buf, bp - i->buf, 1, outf) != 1) { - logf (LOG_FATAL|LOG_ERRNO, "fwrite"); + yaz_log (YLOG_FATAL|YLOG_ERRNO, "fwrite"); exit (1); } i->keylen=0; /* ok, it's written, forget it */ @@ -1792,7 +1812,7 @@ void encode_key_flush (struct encode_info *i, FILE *outf) i->cmd = i->prevcmd; if (fwrite (i->buf, bp - i->buf, 1, outf) != 1) { - logf (LOG_FATAL|LOG_ERRNO, "fwrite"); + yaz_log (YLOG_FATAL|YLOG_ERRNO, "fwrite"); exit (1); } i->keylen=0; /* ok, it's written, forget it */