/*!
*\~english
* Gets current document prefix
*\~russian
* Выбирает префикс текущего документа.
*\~
*\return \~english prexix
* \~russian префикс \~
*/
QString
aDocJournal::getPrefix()
{
QString pref = "";
qulonglong dUid = docId();
if ( dUid )
{
QSqlQuery q = db->db()->exec(QString("SELECT pnum FROM a_journ WHERE idd=%1").arg(dUid));
if ( q.first() ) pref = q.value(0).toString();
}
return pref;
}
/*!
*\~english
* Gets current document.
*\~russian
* Возвращает текущий документ. Полученный документ необходимо удалить после использования.
* \code
* aDocument *doc = sysJournal->CurrentDocument();
* //какие-то действия с документом
* delete doc;
* doc=0;
* \endcode
*\~
*\return \~english current document
* \~russian текущий документ \~
*/
aDocument*
aDocJournal::CurrentDocument()
{
aCfgItem i = md->find( docType() );
if(!i.isNull())
{
aDocument *d = new aDocument( i, db );
if(!d->select( docId() ))
{
return d;
}
else
{
delete d;
return 0;
}
}
else
{
return 0;
}
}
LimaStatusCode SpecificEntitiesXmlLogger::process(
AnalysisContent& analysis) const
{
SELOGINIT;
LDEBUG << "SpecificEntitiesXmlLogger::process";
TimeUtils::updateCurrentTime();
AnnotationData* annotationData = static_cast< AnnotationData* >(analysis.getData("AnnotationData"));
if (annotationData == 0) {
SELOGINIT;
LERROR << "no annotationData ! abort";
return MISSING_DATA;
}
LinguisticAnalysisStructure::AnalysisGraph* graphp = static_cast<LinguisticAnalysisStructure::AnalysisGraph*>(analysis.getData(m_graph));
if (graphp == 0) {
SELOGINIT;
LERROR << "no graph "<< m_graph <<" ! abort";
return MISSING_DATA;
}
const LinguisticAnalysisStructure::AnalysisGraph& graph = *graphp;
LinguisticGraph* lingGraph = const_cast<LinguisticGraph*>(graph.getGraph());
VertexTokenPropertyMap tokenMap = get(vertex_token, *lingGraph);
LinguisticMetaData* metadata=static_cast<LinguisticMetaData*>(analysis.getData("LinguisticMetaData"));
if (metadata == 0) {
SELOGINIT;
LERROR << "no LinguisticMetaData ! abort";
return MISSING_DATA;
}
DumperStream* dstream=initialize(analysis);
ostream& out=dstream->out();
uint64_t offset(0);
try {
offset=atoi(metadata->getMetaData("StartOffset").c_str());
}
catch (LinguisticProcessingException& ) {
// do nothing: not set in analyzeText (only in analyzeXmlDocuments)
}
uint64_t offsetIndexingNode(0);
try {
offsetIndexingNode=atoi(metadata->getMetaData("StartOffsetIndexingNode").c_str());
}
catch (LinguisticProcessingException& ) {
// do nothing: not set in analyzeText (only in analyzeXmlDocuments)
}
std::string docId("");
try {
docId=metadata->getMetaData("DocId");
}
catch (LinguisticProcessingException& ) {
// do nothing: not set in analyzeText (only in analyzeXmlDocuments)
}
if (m_compactFormat) {
out << "<entities docid=\"" << docId
<< "\" offsetNode=\"" << offsetIndexingNode
<< "\" offset=\"" << offset
<< "\">" << endl;
}
else {
out << "<specific_entities>" << endl;
}
// SELOGINIT;
if (m_followGraph) {
// instead of looking to all annotations, follow the graph (in
// morphological graph, some vertices are not related to main graph:
// idiomatic expressions parts and named entity parts)
// -> this will not include nested entities
AnalysisGraph* tokenList=static_cast<AnalysisGraph*>(analysis.getData(m_graph));
if (tokenList==0) {
LERROR << "graph " << m_graph << " has not been produced: check pipeline";
return MISSING_DATA;
}
LinguisticGraph* graph=tokenList->getGraph();
//const FsaStringsPool& sp=Common::MediaticData::MediaticData::single().stringsPool(m_language);
std::queue<LinguisticGraphVertex> toVisit;
std::set<LinguisticGraphVertex> visited;
toVisit.push(tokenList->firstVertex());
LinguisticGraphOutEdgeIt outItr,outItrEnd;
while (!toVisit.empty()) {
LinguisticGraphVertex v=toVisit.front();
toVisit.pop();
if (v == tokenList->lastVertex()) {
continue;
}
for (boost::tie(outItr,outItrEnd)=out_edges(v,*graph); outItr!=outItrEnd; outItr++)
{
LinguisticGraphVertex next=target(*outItr,*graph);
if (visited.find(next)==visited.end())
{
visited.insert(next);
toVisit.push(next);
}
}
const SpecificEntityAnnotation* annot=getSpecificEntityAnnotation(v,annotationData);
if (annot != 0) {
outputEntity(out,v,annot,tokenMap,offset);
}
}
}
else {
// take all annotations
AnnotationGraphVertexIt itv, itv_end;
boost::tie(itv, itv_end) = vertices(annotationData->getGraph());
for (; itv != itv_end; itv++)
{
// LDEBUG << "SpecificEntitiesXmlLogger on annotation vertex " << *itv;
if (annotationData->hasAnnotation(*itv,Common::Misc::utf8stdstring2limastring("SpecificEntity")))
{
// LDEBUG << " it has SpecificEntityAnnotation";
const SpecificEntityAnnotation* annot = 0;
try
{
annot = annotationData->annotation(*itv,Common::Misc::utf8stdstring2limastring("SpecificEntity"))
.pointerValue<SpecificEntityAnnotation>();
}
catch (const boost::bad_any_cast& )
{
SELOGINIT;
LERROR << "This annotation is not a SpecificEntity; SE not logged";
continue;
}
// recuperer l'id du vertex morph cree
LinguisticGraphVertex v;
if (!annotationData->hasIntAnnotation(*itv,Common::Misc::utf8stdstring2limastring(m_graph)))
{
// SELOGINIT;
// LDEBUG << *itv << " has no " << m_graph << " annotation. Skeeping it.";
continue;
}
v = annotationData->intAnnotation(*itv,Common::Misc::utf8stdstring2limastring(m_graph));
outputEntity(out,v,annot,tokenMap,offset);
}
}
}
// LDEBUG << " all vertices done";
if (m_compactFormat) {
out << "</entities>" << endl;
}
else {
out << "</specific_entities>" << endl;
}
delete dstream;
TimeUtils::logElapsedTime("SpecificEntitiesXmlLogger");
return SUCCESS_ID;
}
LimaStatusCode SemanticRelationsXmlLogger::
process(AnalysisContent& analysis) const
{
TimeUtils::updateCurrentTime();
SEMLOGINIT;
LERROR << "SemanticRelationsXmlLogger" << LENDL;
AnnotationData* annotationData = static_cast< AnnotationData* >(analysis.getData("AnnotationData"));
const LinguisticAnalysisStructure::AnalysisGraph& graph =
*(static_cast<LinguisticAnalysisStructure::AnalysisGraph*>(analysis.getData(m_graph)));
LinguisticGraph* lingGraph = const_cast<LinguisticGraph*>(graph.getGraph());
VertexTokenPropertyMap tokenMap = get(vertex_token, *lingGraph);
LinguisticMetaData* metadata=static_cast<LinguisticMetaData*>(analysis.getData("LinguisticMetaData"));
if (metadata == 0) {
SEMLOGINIT;
LERROR << "no LinguisticMetaData ! abort" << LENDL;
return MISSING_DATA;
}
ofstream out;
if (!openLogFile(out,metadata->getMetaData("FileName"))) {
SEMLOGINIT;
LERROR << "Can't open log file " << LENDL;
return UNKNOWN_ERROR;
}
uint64_t offset(0);
try {
offset=atoi(metadata->getMetaData("StartOffset").c_str());
}
catch (LinguisticProcessingException& e) {
// do nothing: not set in analyzeText (only in analyzeXmlDocuments)
}
uint64_t offsetIndexingNode(0);
try {
offsetIndexingNode=atoi(metadata->getMetaData("StartOffsetIndexingNode").c_str());
}
catch (LinguisticProcessingException& e) {
// do nothing: not set in analyzeText (only in analyzeXmlDocuments)
}
std::string docId("");
try {
docId=metadata->getMetaData("DocId");
}
catch (LinguisticProcessingException& e) {
// do nothing: not set in analyzeText (only in analyzeXmlDocuments)
}
out << "<relations docid=\"" << docId
<< "\" offsetNode=\"" << offsetIndexingNode
<< "\">" << endl;
// LDEBUG << "SemanticRelationsXmlLogger on graph " << m_graph << LENDL;
//look at all vertices for annotations
AnnotationGraphVertexIt itv, itv_end;
boost::tie(itv, itv_end) = vertices(annotationData->getGraph());
for (; itv != itv_end; itv++)
{
LDEBUG << "SemanticRelationsXmlLogger on annotation vertex " << *itv << LENDL;
if (annotationData->hasAnnotation(*itv,Common::Misc::utf8stdstring2limastring("SemanticAnnotation")))
{
// LDEBUG << " it has SemanticRelationAnnotation" << LENDL;
const SemanticAnnotation* annot = 0;
try
{
annot = annotationData->annotation(*itv,Common::Misc::utf8stdstring2limastring("SemanticAnnotation"))
.pointerValue<SemanticAnnotation>();
}
catch (const boost::bad_any_cast& e)
{
SEMLOGINIT;
LERROR << "This annotation is not a SemanticRelation" << LENDL;
continue;
}
// output
out << "<annotation type=\"" << annot->getType() << "\">" << endl
<< vertexStringForSemanticAnnotation("vertex",*itv,tokenMap,annotationData,offset)
<< "</annotation>" << endl;
}
}
// look at all edges for relations
AnnotationGraphEdgeIt it,it_end;
const AnnotationGraph& annotGraph=annotationData->getGraph();
boost::tie(it, it_end) = edges(annotGraph);
for (; it != it_end; it++) {
LDEBUG << "SemanticRelationsXmlLogger on annotation edge "
<< source(*it,annotGraph) << "->" << target(*it,annotationData->getGraph()) << LENDL;
if (annotationData->hasAnnotation(*it,Common::Misc::utf8stdstring2limastring("SemanticRelation")))
{
SEMLOGINIT;
LDEBUG << "found semantic relation" << LENDL;
const SemanticRelationAnnotation* annot = 0;
try
{
annot = annotationData->annotation(*it,Common::Misc::utf8stdstring2limastring("SemanticRelation"))
.pointerValue<SemanticRelationAnnotation>();
}
catch (const boost::bad_any_cast& e)
{
SEMLOGINIT;
LERROR << "This annotation is not a SemanticAnnotation" << LENDL;
continue;
}
//output
out << "<relation type=\"" << annot->type() << "\">" << endl
<< vertexStringForSemanticAnnotation("source",source(*it,annotGraph),tokenMap,annotationData,offset)
<< vertexStringForSemanticAnnotation("target",target(*it,annotGraph),tokenMap,annotationData,offset)
<< "</relation>" << endl;
}
}
// LDEBUG << " all vertices done" << LENDL;
out << "</relations>" << endl;
out.close();
TimeUtils::logElapsedTime("SemanticRelationsXmlLogger");
return SUCCESS_ID;
}
LimaStatusCode SegmentationDataXmlLogger::process(
AnalysisContent& analysis) const
{
LOGINIT("LP::Segmentation");
LDEBUG << "SegmentationDataXmlLogger::process" << LENDL;
TimeUtils::updateCurrentTime();
LinguisticMetaData* metadata=static_cast<LinguisticMetaData*>(analysis.getData("LinguisticMetaData"));
if (metadata == 0) {
LOGINIT("LP::Segmentation");
LERROR << "no LinguisticMetaData ! abort" << LENDL;
return MISSING_DATA;
}
// open output file
ofstream out;
if (!openLogFile(out,metadata->getMetaData("FileName"))) {
LOGINIT("LP::Segmentation");
LERROR << "Can't open log file '" << metadata->getMetaData("FileName") << "'" << LENDL;
return UNKNOWN_ERROR;
}
// get metadata (useful for XML documents)
// uint64_t offset(0);
// uint64_t offsetIndexingNode(0);
std::string docId("");
try {
// offset=atoi(metadata->getMetaData("StartOffset").c_str());
// offsetIndexingNode=atoi(metadata->getMetaData("StartOffsetIndexingNode").c_str());
docId=metadata->getMetaData("DocId");
}
catch (LinguisticProcessingException& ) {
// do nothing: not set in analyzeText (only in analyzeXmlDocuments)
}
// log
out << "<segmentation>" << endl;
const AnalysisData* data =analysis.getData(m_data);
if (data!=0) {
const SegmentationData* segData=static_cast<const SegmentationData*>(data);
const vector<Segment> seg=segData->getSegments();
for (vector<Segment>::const_iterator it=seg.begin(), it_end=seg.end(); it!=it_end; it++) {
out
<< "<segment>"
<< "<pos>" << (*it).getPosBegin() << "</pos>"
<< "<len>" << (*it).getLength() << "</len>"
<< "<type>" << (*it).getType() << "</type>"
<< "</segment>"
<< endl;
}
}
else {
LOGINIT("LP::Segmentation");
LDEBUG << "no SegmentationData of name " << m_data << LENDL;
}
out << "</segmentation>" << endl;
out.close();
TimeUtils::logElapsedTime("SegmentationDataXmlLogger");
return SUCCESS_ID;
}
/*!
*\~english
* Gets current document type.
*\~russian
* Получение типа текущего документа.
*\~
*\return \~english database document type. \~russian тип документа в базе данных.\~
*/
int
aDocJournal::docType()
{
return db->uidType( docId() );
}