void TextIndexMergerTestCase::makeMergedAnswer(TermMap& mergedAnswer)
{
for (size_t j = 0; j < m_nNumEntries; ++j)
{
TermMap& ans = m_mergeEntries[j]->answer;
for (TermMap::const_iterator it2 = ans.begin();
it2 != ans.end(); ++it2)
{
for (size_t l = 0; l < it2->second.size(); ++l)
{
if (m_mergeEntries[j]->pDocIdRemapper)
{
docid_t docId = it2->second[l].first;
if (!m_mergeEntries[j]->pDocFilter->test((size_t)docId))
{
docId = (*(m_mergeEntries[j]->pDocIdRemapper))[docId];
TermMap::iterator it = mergedAnswer.find(it2->first);
if (it == mergedAnswer.end())
{
DocVector docVect;
mergedAnswer.insert(make_pair(it2->first, docVect));
it = mergedAnswer.find(it2->first);
}
it->second.push_back(make_pair(docId + m_mergeEntries[j]->newBaseDocId,
it2->second[l].second));
}
}
else
{
TermMap::iterator it = mergedAnswer.find(it2->first);
if (it == mergedAnswer.end())
{
DocVector docVect;
mergedAnswer.insert(make_pair(it2->first, docVect));
it = mergedAnswer.find(it2->first);
}
it->second.push_back(make_pair(it2->second[l].first +
m_mergeEntries[j]->newBaseDocId,
it2->second[l].second));
}
// FX_TRACE("==term: %llu, doc: %d, first pos: %d", it2->first, it2->second[l].first,
// it2->second[l].second[0]);
}
}
}
}
void TextIndexMergerTestCase::checkMergedResult(TermIteratorPtr& pTermIt, TermMap& answer)
{
CPPUNIT_ASSERT(pTermIt);
CPPUNIT_ASSERT_EQUAL((int64_t)answer.size(), pTermIt->size());
size_t termCount = 0;
while (pTermIt->hasNext())
{
TermIterator::TermEntry entry = pTermIt->next();
const TextTermIterator::TermType* pTerm =
dynamic_cast<const TextTermIterator::TermType*>(entry.term);
CPPUNIT_ASSERT(pTerm != NULL);
uint64_t hash = pTerm->getValue();
FX_TRACE("Term hash: %llu", hash);
TermMap::const_iterator it = answer.find(hash);
CPPUNIT_ASSERT(it != answer.end());
CPPUNIT_ASSERT_EQUAL(it->first, hash);
const TermMeta& termMeta = entry.postingIterator->getTermMeta();
size_t df = it->second.size();
CPPUNIT_ASSERT_EQUAL((df_t)df, termMeta.getDocFreq());
ctf_t ctf = 0;
for (size_t i = 0; i < df; ++i)
{
docid_t docId = entry.postingIterator->skipTo(it->second[i].first);
CPPUNIT_ASSERT_EQUAL(it->second[i].first, docId);
tf_t tf = it->second[i].second.size();
CPPUNIT_ASSERT_EQUAL(tf, entry.postingIterator->freq());
for (size_t j = 0; j < (size_t)tf; ++j)
{
loc_t posExp = it->second[i].second[j];
loc_t pos = entry.postingIterator->skipToPosition(posExp);
CPPUNIT_ASSERT_EQUAL(posExp, pos);
}
ctf += it->second[i].second.size();
}
CPPUNIT_ASSERT_EQUAL(ctf, termMeta.getCTF());
termCount++;
}
CPPUNIT_ASSERT_EQUAL(answer.size(), termCount);
}
void TextIndexMergerTestCase::printPositngAnswer(const TermMap& answer)
{
for (TermMap::const_iterator it = answer.begin();
it != answer.end(); ++it)
{
std::cout << "Anser term hash: " << it->first << std::endl;
std::cout << "anser doc list: " << std::endl;
for (size_t i = 0; i < it->second.size(); ++i)
{
std::cout << it->second[i].first << ", ";
}
std::cout << "end anser doc list: " << std::endl;
ctf_t ctf = 0;
std::cout << "anser tf list: " << std::endl;
for (size_t i = 0; i < it->second.size(); ++i)
{
ctf += it->second[i].second.size();
std::cout << it->second[i].second.size() << ", ";
}
std::cout << "end anser tf list: " << std::endl;
std::cout << "anser pos list: " << std::endl;
for (size_t i = 0; i < it->second.size(); ++i)
{
for (size_t j = 0; j < it->second[i].second.size(); ++j)
{
std::cout << it->second[i].second[j] << ", ";
}
}
std::cout << "end anser pos list: " << std::endl;
std::cout << "Anser term meta: df: " << it->second.size()
<< ", ctf: " << ctf << std::endl;
}
}
