Skip to content

Commit

Permalink
tuning bm25 parameters, setting sort
Browse files Browse the repository at this point in the history
  • Loading branch information
maneeshpm committed Feb 13, 2021
1 parent 86c62d4 commit 95fe273
Show file tree
Hide file tree
Showing 2 changed files with 30 additions and 0 deletions.
4 changes: 4 additions & 0 deletions src/search.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -353,6 +353,7 @@ Search::iterator Search::begin() const {
}
Xapian::Query query;
try {
// flags = Xapian::QueryParser::FLAG_PHRASE;
query = queryParser->parse_query(this->query, flags, prefix);
} catch (Xapian::QueryParserError& e) {
estimated_matches_number = 0;
Expand All @@ -379,6 +380,9 @@ Search::iterator Search::begin() const {
}
}

// Tune BM25 to reduce wdf parameter and set sort order
enquire.set_weighting_scheme(Xapian::BM25Weight(0.001,0,1,1,0.5));
enquire.set_sort_by_relevance_then_value(valuesmap["titles"], true);
enquire.set_query(query);

#if WITH_LEV
Expand Down
26 changes: 26 additions & 0 deletions src/writer/creator.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -170,7 +170,23 @@ namespace zim

auto dirent = data->createItemDirent(item.get());
data->addItemData(dirent, item->getContentProvider(), compressContent);
<<<<<<< HEAD
data->handle(dirent, item);
=======

#if defined(ENABLE_XAPIAN)
if (item->getMimeType() == "text/html" && !item->getTitle().empty()) {
data->nbIndexItems++;
std::string ss = item->getTitle();
ss.insert(0, "zzzz ");
data->titleIndexer.indexTitle(item->getPath(), ss);
std::cout<<"indexing.: "<<ss<<std::endl;
if(m_withIndex) {
data->taskList.pushToQueue(new IndexTask(item));
}
}
#endif
>>>>>>> tuning bm25 parameters, setting sort

if (data->dirents.size()%1000 == 0) {
TPROGRESS();
Expand Down Expand Up @@ -199,8 +215,18 @@ namespace zim
TPROGRESS();
}

<<<<<<< HEAD
data->handle(dirent);
}
=======
#if defined(ENABLE_XAPIAN)
if (!title.empty()) {
data->titleIndexer.indexTitle(path, title);
std::cout<<"indexing: "<<title<<std::endl;
}
#endif
}
>>>>>>> tuning bm25 parameters, setting sort

void Creator::finishZimCreation()
{
Expand Down

0 comments on commit 95fe273

Please sign in to comment.