tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Bad concatenation generated absurd page numbers for document with several multiple paeg breaks

Browse Source
This commit is contained in:
Jean-Francois Dockes 2012-09-19 14:04:20 +02:00
parent ff1bf58ae0
commit 1b5136539f
1 changed files with 8 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
src/rcldb/rcldb.cpp
View File

@ -307,6 +307,7 @@ double Db::Native::qualityTerms(Xapian::docid docid,
// Return the positions list for the page break term // Return the positions list for the page break term
bool Db::Native::getPagePositions(Xapian::docid docid, vector<int>& vpos) bool Db::Native::getPagePositions(Xapian::docid docid, vector<int>& vpos)
{ {
vpos.clear();
// Need to retrieve the document record to check for multiple page breaks // Need to retrieve the document record to check for multiple page breaks
// that we store there for lack of better place // that we store there for lack of better place
map<int, int> mbreaksmap; map<int, int> mbreaksmap;
@ -637,9 +638,11 @@ vector<string> Db::Native::makeAbstract(Xapian::docid docid, Query *query)
continue; continue;
if (chunk.empty() && !vpbreaks.empty()) { if (chunk.empty() && !vpbreaks.empty()) {
int pnum = getPageNumberForPosition(vpbreaks, it->first); int pnum = getPageNumberForPosition(vpbreaks, it->first);
ostringstream ss; if (pnum > 0) {
ss << pnum; ostringstream ss;
chunk += string(" [p ") + ss.str() + "] "; ss << pnum;
chunk += string(" [p ") + ss.str() + "] ";
}
} }
Utf8Iter uit(it->second); Utf8Iter uit(it->second);
bool newcjk = false; bool newcjk = false;
@ -1463,6 +1466,8 @@ bool Db::addOrUpdate(const string &udi, const string &parent_udi,
if (!tpidx.m_pageincrvec.empty()) { if (!tpidx.m_pageincrvec.empty()) {
ostringstream multibreaks; ostringstream multibreaks;
for (unsigned int i = 0; i < tpidx.m_pageincrvec.size(); i++) { for (unsigned int i = 0; i < tpidx.m_pageincrvec.size(); i++) {
if (i != 0)
multibreaks << ",";
multibreaks << tpidx.m_pageincrvec[i].first << "," << multibreaks << tpidx.m_pageincrvec[i].first << "," <<
tpidx.m_pageincrvec[i].second; tpidx.m_pageincrvec[i].second;
} }