private void buildDocument() throws IOException {
_docId = _documentDescriptor;
String currentDocId = DocumentIdOperations.documentDescriptorToId(_documentDescriptor);
int lastSnippetId = 0;
DocumentHolder docHolder = new DocumentHolder(_indexReader, _docId);
DocForReport reportDocument = new DocForReport(currentDocId);
for (MatchDescriptor match : _searchResult.matches) {
// there are ~ 100 matches in a regular report page, and it's OK to traverse them all each
// time building
// a new document node
DocumentDescriptor docDescriptor = match.doc;
if (!(docDescriptor.equals(_docId))) {
continue;
}
if (!_searchType.equals("all-documents") && !docDescriptor.equals(_docId)) {
continue;
}
// searching for the specific snippet
if (_searchType.equals("snippet") && lastSnippetId != _sId) {
++lastSnippetId;
continue;
}
addMatchToDocument(match, reportDocument, docHolder);
++lastSnippetId;
}
createDocumentNode(_documentDescriptor);
_rootNode.setAttribute("snippets", String.valueOf(reportDocument.snippets.size()));
for (Snippet snippet : reportDocument.snippets) {
Element snippetNode = createSnippetNode(snippet);
_rootNode.appendChild(snippetNode);
}
}
protected void readQueryParameters(final Map<String, String> inCgiConfig) {
_docId = null;
_sId = -1;
if (inCgiConfig.containsKey("docid")) {
_docId = DocumentIdOperations.idToDocumentDescriptor(inCgiConfig.get("docid"));
}
if (inCgiConfig.containsKey("sid")) {
_sId = Integer.parseInt(inCgiConfig.get("sid"));
}
}
void createDocumentNode(final DocumentDescriptor inDescriptor) throws IOException {
try {
_document = DocumentBuilderFactory.newInstance().newDocumentBuilder().newDocument();
_rootNode = _document.createElement("document");
} catch (ParserConfigurationException e) {
e.printStackTrace();
System.exit(1);
}
AtomicReader segmentReader = _indexReader.leaves().get(inDescriptor.segmentNumber).reader();
_rootNode.setAttribute("id", DocumentIdOperations.documentDescriptorToId(inDescriptor));
// TODO: implement the proper way of building a title from the production report
_rootNode.setAttribute("title", buildDocumentTitle(segmentReader, inDescriptor));
_rootNode.setAttribute("path", "ruscorpora.ru");
_rootNode.setAttribute(
"tagging", Attributes.getDocAttribute(segmentReader, inDescriptor.localId, "tagging"));
_rootNode.setAttribute("snippets", "0");
Element attributesNode = _document.createElement("attributes");
_rootNode.appendChild(attributesNode);
FieldInfos fields = segmentReader.getFieldInfos();
for (int fieldIndex = 0; fieldIndex != fields.size(); ++fieldIndex) {
FieldInfo field = fields.fieldInfo(fieldIndex);
// TODO: understand why field may turn into null
if (field == null) {
continue;
}
String name = field.name;
if (Attributes.ATTRIBUTES.contains(name)
|| Attributes.ATTRIBUTES_FOR_REPORT.contains(name)
|| Attributes.ATTRIBUTES_FOR_WORD_INFO.contains(name)
|| !field.hasDocValues()) {
// it's a word attribute
continue;
}
Element attrNode = _document.createElement("attr");
attrNode.setAttribute("name", name);
attrNode.setAttribute(
"value", Attributes.getDocAttribute(segmentReader, inDescriptor.localId, name));
attributesNode.appendChild(attrNode);
}
}