X-Git-Url: http://git.indexdata.com/?p=idzebra-moved-to-github.git;a=blobdiff_plain;f=index%2Fsafari.c;h=be8c567764233a4659e0590fc1f172cfcb1a0b5b;hp=4210a9288f1b0c8405cafd15d8141eaf1f79d275;hb=97a7adeb9e5059463f039495cc01cfa448463a27;hpb=1b5ef8265837240930862dddd2d7ef963a0cc211 diff --git a/index/safari.c b/index/safari.c index 4210a92..be8c567 100644 --- a/index/safari.c +++ b/index/safari.c @@ -1,8 +1,5 @@ -/* $Id: safari.c,v 1.7 2007-04-16 21:54:37 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 @@ -118,6 +115,7 @@ static int filter_extract(void *clientData, struct recExtractCtrl *p) struct filter_info *tinfo = clientData; char line[512]; RecWord recWord; + int ret = RECCTRL_EXTRACT_OK; struct fi_info *fi = fi_open(p); #if 0 @@ -127,50 +125,62 @@ static int filter_extract(void *clientData, struct recExtractCtrl *p) (*p->init)(p, &recWord); if (!fi_gets(fi, line, sizeof(line)-1)) - return RECCTRL_EXTRACT_ERROR_GENERIC; - sscanf(line, "%255s", p->match_criteria); - - recWord.index_type = '0'; - while (fi_gets(fi, line, sizeof(line)-1)) + ret = RECCTRL_EXTRACT_EOF; + else { - int nor = 0; - char field[40]; - char *cp; + sscanf(line, "%255s", p->match_criteria); + while (fi_gets(fi, line, sizeof(line)-1)) + { + int nor = 0; + char field[40]; + const char *cp = line; + char type_cstr[2]; #if 0 - yaz_log(YLOG_LOG, "safari line: %s", line); + yaz_log(YLOG_LOG, "safari line: %s", line); #endif - if (tinfo->segments) - { - if (sscanf(line, ZINT_FORMAT " " ZINT_FORMAT " " ZINT_FORMAT - ZINT_FORMAT " %39s %n", - &recWord.record_id, &recWord.section_id, - &recWord.segment, - &recWord.seqno, - field, &nor) < 5) + type_cstr[1] = '\0'; + if (*cp >= '0' && *cp <= '9') + type_cstr[0] = '0'; /* the default is 0 (raw) */ + else + type_cstr[0] = *cp++; /* type given */ + type_cstr[1] = '\0'; + + recWord.index_type = type_cstr; + if (tinfo->segments) { - yaz_log(YLOG_WARN, "Bad safari record line: %s", line); - return RECCTRL_EXTRACT_ERROR_GENERIC; + if (sscanf(cp, ZINT_FORMAT " " ZINT_FORMAT " " ZINT_FORMAT + ZINT_FORMAT " %39s %n", + &recWord.record_id, &recWord.section_id, + &recWord.segment, + &recWord.seqno, + field, &nor) < 5) + { + yaz_log(YLOG_WARN, "Bad safari record line: %s", line); + ret = RECCTRL_EXTRACT_ERROR_GENERIC; + break; + } } - } - else - { - if (sscanf(line, ZINT_FORMAT " " ZINT_FORMAT " " ZINT_FORMAT " %39s %n", - &recWord.record_id, &recWord.section_id, &recWord.seqno, - field, &nor) < 4) + else { - yaz_log(YLOG_WARN, "Bad safari record line: %s", line); - return RECCTRL_EXTRACT_ERROR_GENERIC; + if (sscanf(cp, ZINT_FORMAT " " ZINT_FORMAT " " ZINT_FORMAT " %39s %n", + &recWord.record_id, &recWord.section_id, &recWord.seqno, + field, &nor) < 4) + { + yaz_log(YLOG_WARN, "Bad safari record line: %s", line); + ret = RECCTRL_EXTRACT_ERROR_GENERIC; + break; + } } + for (cp = cp + nor; *cp == ' '; cp++) + ; + recWord.index_name = field; + recWord.term_buf = cp; + recWord.term_len = strlen(cp); + (*p->tokenAdd)(&recWord); } - for (cp = line + nor; *cp == ' '; cp++) - ; - recWord.index_name = field; - recWord.term_buf = cp; - recWord.term_len = strlen(cp); - (*p->tokenAdd)(&recWord); } fi_close(fi); - return RECCTRL_EXTRACT_OK; + return ret; } static int filter_retrieve (void *clientData, struct recRetrieveCtrl *p)