//-------------------------------------------------------------------------
/// <summary>
/// Parses the tree from the StAX stream reader, capturing references.
/// <para>
/// The reader should be created using the factory returned from <seealso cref="#xmlInputFactory()"/>.
/// </para>
/// <para>
/// This method supports capturing attribute references, such as an id/href pair.
/// Wherever the parser finds an attribute with the specified name, the element is added
/// to the specified map. Note that the map is mutated.
///
/// </para>
/// </summary>
/// <param name="reader"> the StAX stream reader, positioned at or before the element to be parsed </param>
/// <param name="refAttr"> the attribute name that should be parsed as a reference, null if not applicable </param>
/// <param name="refs"> the mutable map of references to update, null if not applicable </param>
/// <returns> the parsed element </returns>
/// <exception cref="IllegalArgumentException"> if the input cannot be parsed </exception>
private static XmlElement parse(XMLStreamReader reader, string refAttr, IDictionary <string, XmlElement> refs)
{
try
{
// parse start element
string elementName = parseElementName(reader);
ImmutableMap <string, string> attrs = parseAttributes(reader);
// parse children or content
ImmutableList.Builder <XmlElement> childBuilder = ImmutableList.builder();
string content = "";
int @event = reader.next();
while (@event != XMLStreamConstants.END_ELEMENT)
{
switch (@event)
{
// parse child when start element found
case XMLStreamConstants.START_ELEMENT:
childBuilder.add(parse(reader, refAttr, refs));
break;
// append content when characters found
// since XMLStreamReader has IS_COALESCING=true means there should only be one content call
case XMLStreamConstants.CHARACTERS:
case XMLStreamConstants.CDATA:
content += reader.Text;
break;
default:
break;
}
@event = reader.next();
}
ImmutableList <XmlElement> children = childBuilder.build();
XmlElement parsed = children.Empty ? XmlElement.ofContent(elementName, attrs, content) : XmlElement.ofChildren(elementName, attrs, children);
string @ref = attrs.get(refAttr);
if (!string.ReferenceEquals(@ref, null))
{
refs[@ref] = parsed;
}
return(parsed);
}
catch (XMLStreamException ex)
{
throw new System.ArgumentException(ex);
}
}
// find the start element and parses the name
//JAVA TO C# CONVERTER WARNING: Method 'throws' clauses are not available in .NET:
//ORIGINAL LINE: private static String parseElementName(javax.xml.stream.XMLStreamReader reader) throws javax.xml.stream.XMLStreamException
private static string parseElementName(XMLStreamReader reader)
{
int @event = reader.EventType;
while (@event != XMLStreamConstants.START_ELEMENT)
{
@event = reader.next();
}
return(reader.LocalName);
}
//JAVA TO C# CONVERTER WARNING: Method 'throws' clauses are not available in .NET:
//ORIGINAL LINE: public boolean readSentence(org.maltparser.core.syntaxgraph.TokenStructure syntaxGraph) throws org.maltparser.core.exception.MaltChainedException
public virtual bool readSentence(ITokenStructure syntaxGraph)
{
if (syntaxGraph == null || !(syntaxGraph is PhraseStructure))
{
return(false);
}
syntaxGraph.Clear();
//JAVA TO C# CONVERTER WARNING: The original Java variable was marked 'final':
//ORIGINAL LINE: final org.maltparser.core.syntaxgraph.PhraseStructure phraseStructure = (org.maltparser.core.syntaxgraph.PhraseStructure)syntaxGraph;
PhraseStructure phraseStructure = (PhraseStructure)syntaxGraph;
PhraseStructureNode parent = null;
PhraseStructureNode child = null;
// if (header == null) {
// header = new TigerXMLHeader(syntaxGraph.getSymbolTables());
// }
try
{
while (true)
{
int @event = reader.next();
if (@event == XMLStreamConstants.START_ELEMENT)
{
if (reader.LocalName.length() == 0)
{
continue;
}
if (reader.LocalName.charAt(0) == 'e')
{
// e -> edge, edgelabel
if (reader.LocalName.length() == 4)
{ //edge
int childid = -1;
int indexSep = reader.getAttributeValue(null, "idref").IndexOf('_');
try
{
if (indexSep != -1)
{
childid = int.Parse(reader.getAttributeValue(null, "idref").substring(indexSep + 1));
}
else
{
childid = int.Parse(reader.getAttributeValue(null, "idref"));
}
if (childid == -1)
{
throw new SyntaxGraphException("The tiger reader couldn't recognize the idref attribute '" + reader.getAttributeValue(null, "idref") + "' of the edge element. ");
}
}
catch (System.FormatException)
{
throw new SyntaxGraphException("The tiger reader couldn't recognize the idref attribute '" + reader.getAttributeValue(null, "idref") + "' of the edge element. ");
}
if (childid < START_ID_OF_NONTERMINALS)
{
child = phraseStructure.GetTokenNode(childid);
}
else
{
child = phraseStructure.getNonTerminalNode(childid - START_ID_OF_NONTERMINALS + 1);
}
Edge.Edge e = phraseStructure.addPhraseStructureEdge(parent, child);
SortedDictionary <string, SymbolTable> inputTables = dataFormatInstance.getPhraseStructureEdgeLabelSymbolTables(phraseStructure.SymbolTables);
foreach (string name in inputTables.Keys)
{
e.addLabel(inputTables[name], reader.getAttributeValue(null, name.ToLower()));
}
}
else if (reader.LocalName.Equals("edgelabel"))
{ // edgelabel
// domain = Domain.EL;
}
}
else if (reader.LocalName.charAt(0) == 'n')
{
// n -> nt, nonterminals, name
if (reader.LocalName.length() == 2)
{ // nt
//JAVA TO C# CONVERTER WARNING: The original Java variable was marked 'final':
//ORIGINAL LINE: final String id = reader.getAttributeValue(null, "id");
string id = reader.getAttributeValue(null, "id");
if (graphRootID.Length == id.Length && graphRootID.ToString().Equals(id))
{
parent = phraseStructure.PhraseStructureRoot;
}
else
{
int index = id.IndexOf('_');
if (index != -1)
{
parent = phraseStructure.addNonTerminalNode(int.Parse(id.Substring(index + 1)) - START_ID_OF_NONTERMINALS + 1);
}
}
SortedDictionary <string, SymbolTable> inputTables = dataFormatInstance.getPhraseStructureNodeLabelSymbolTables(phraseStructure.SymbolTables);
foreach (string name in inputTables.Keys)
{
parent.addLabel(inputTables[name], reader.getAttributeValue(null, name.ToLower()));
}
}
else if (reader.LocalName.Equals("name"))
{ // name
// elementContent.setLength(0);
// collectChar = true;
}
}
else if (reader.LocalName.charAt(0) == 't')
{
// t -> t, terminals
if (reader.LocalName.length() == 1)
{ // t
SortedDictionary <string, SymbolTable> inputTables = dataFormatInstance.getInputSymbolTables(phraseStructure.SymbolTables);
child = syntaxGraph.AddTokenNode();
foreach (string name in inputTables.Keys)
{
child.addLabel(inputTables[name], reader.getAttributeValue(null, name.ToLower()));
}
}
}
else if (reader.LocalName.charAt(0) == 's')
{
// s -> subcorpus, secedge, s, secedgelabel
if (reader.LocalName.length() == 1)
{ // s
string id = reader.getAttributeValue(null, "id");
bool indexable = false;
int index = -1;
if (!ReferenceEquals(id, null) && id.Length > 0)
{
for (int i = 0, n = id.Length; i < n; i++)
{
if (char.IsDigit(id[i]))
{
if (index == -1)
{
index = i;
}
indexable = true;
}
}
}
if (indexable)
{
phraseStructure.SentenceID = int.Parse(id.Substring(index));
}
else
{
phraseStructure.SentenceID = sentenceCount + 1;
}
}
}
else if (reader.LocalName.charAt(0) == 'v')
{
// v -> variable, value
// if (reader.getLocalName().equals("value")) {
// valueName.setLength(0);
// valueName.append(reader.getAttributeValue(null, "name"));
// elementContent.setLength(0);
// collectChar = true;
// }
}
else
{
// a -> annotation, author
// b -> body
// c -> corpus
// d -> date, description,
// f -> feature, format
// g -> graph
// h -> head, history
// m -> matches, match
if (reader.LocalName.Equals("graph"))
{
graphRootID.Length = 0;
graphRootID.Append(reader.getAttributeValue(null, "root"));
}
else if (reader.LocalName.Equals("corpus"))
{
// header.setCorpusID(reader.getAttributeValue(null, "id"));
// header.setCorpusID(reader.getAttributeValue(null, "version"));
}
else if (reader.LocalName.Equals("feature"))
{
// if (header != null) {
// currentFeatureName.setLength(0);
// currentFeatureName.append(reader.getAttributeValue(null, "name"));
// header.addFeature(reader.getAttributeValue(null, "name"), reader.getAttributeValue(null, "domain"));
// }
// domain = Domain.valueOf(reader.getAttributeValue(null, "domain"));
}
else if (reader.LocalName.Equals("secedgelabel"))
{
// domain = Domain.SEL;
}
else if (reader.LocalName.Equals("author"))
{
// elementContent.setLength(0);
// collectChar = true;
}
else if (reader.LocalName.Equals("date"))
{
// elementContent.setLength(0);
// collectChar = true;
}
else if (reader.LocalName.Equals("description"))
{
// elementContent.setLength(0);
// collectChar = true;
}
else if (reader.LocalName.Equals("format"))
{
// elementContent.setLength(0);
// collectChar = true;
}
else if (reader.LocalName.Equals("history"))
{
// elementContent.setLength(0);
// collectChar = true;
}
}
}
else if (@event == XMLStreamConstants.END_ELEMENT)
{
if (reader.LocalName.length() == 0)
{
continue;
}
if (reader.LocalName.charAt(0) == 'e')
{
// e -> edge, edgelabel
}
else if (reader.LocalName.charAt(0) == 'n')
{
// n -> nt, nonterminals, name
if (reader.LocalName.Equals("nt"))
{
ntid.Length = 0;
}
else if (reader.LocalName.Equals("nonterminals"))
{
if (phraseStructure.NTokenNode() == 1 && phraseStructure.nNonTerminals() == 0 && ((NonTerminalNode)phraseStructure.PhraseStructureRoot).nChildren() == 0)
{
Edge.Edge e = phraseStructure.addPhraseStructureEdge(phraseStructure.PhraseStructureRoot, phraseStructure.GetTokenNode(1));
SortedDictionary <string, SymbolTable> inputTables = dataFormatInstance.getPhraseStructureEdgeLabelSymbolTables(phraseStructure.SymbolTables);
foreach (string name in inputTables.Keys)
{
e.addLabel(inputTables[name], "--");
}
}
}
// else if (reader.getLocalName().equals("name")) {
// if (header != null) {
// header.setMetaName(elementContent.toString());
// }
// collectChar = false;
// }
}
else if (reader.LocalName.charAt(0) == 't')
{
// t -> t, terminals
}
else if (reader.LocalName.charAt(0) == 's')
{
// s -> subcorpus, secedge, s, secedgelabel
if (reader.LocalName.Equals("s"))
{
if (syntaxGraph.HasTokens())
{
sentenceCount++;
}
if (syntaxGraph is MappablePhraseStructureGraph)
{
((MappablePhraseStructureGraph)syntaxGraph).Mapping.updateDependenyGraph(((MappablePhraseStructureGraph)syntaxGraph), ((PhraseStructure)syntaxGraph).PhraseStructureRoot);
}
return(true);
}
}
else if (reader.LocalName.charAt(0) == 'v')
{
// v -> variable, value
// if (reader.getLocalName().equals("value")) {
// if (header != null) {
// if (domain == Domain.T || domain == Domain.NT || domain == Domain.FREC) {
// header.addFeatureValue(currentFeatureName.toString(), valueName.toString(), elementContent.toString());
// } else if (domain == Domain.EL) {
// header.addEdgeLabelValue(valueName.toString(), elementContent.toString());
// } else if (domain == Domain.SEL) {
// header.addSecEdgeLabelValue(valueName.toString(), elementContent.toString());
// }
// }
// collectChar = false;
// }
}
else
{
// a -> annotation, author
// b -> body
// c -> corpus
// d -> date, description,
// f -> feature, format
// g -> graph
// h -> head, history
// m -> matches, match
if (reader.LocalName.Equals("body"))
{
//sentence = dataStructures.getSentence();
//phraseTree = dataStructures.getInPhraseTree();
//sentence.clear();
//phraseTree.clear();
//dataStructures.setLastProcessObject(true);
}
else if (reader.LocalName.Equals("author"))
{
// if (header != null) {
// header.setMetaAuthor(elementContent.toString());
// }
// collectChar = false;
}
else if (reader.LocalName.Equals("date"))
{
// if (header != null) {
// header.setMetaInDate(elementContent.toString());
// }
// collectChar = false;
}
else if (reader.LocalName.Equals("description"))
{
// if (header != null) {
// header.setMetaDescription(elementContent.toString());
// }
// collectChar = false;
}
else if (reader.LocalName.Equals("format"))
{
// if (header != null) {
// header.setMetaFormat(elementContent.toString());
// }
// collectChar = false;
}
else if (reader.LocalName.Equals("history"))
{
// if (header != null) {
// header.setMetaHistory(elementContent.toString());
// }
// collectChar = false;
} /* else if (reader.getLocalName().equals("annotation")) {
* if (header != null) {
* System.out.println(header.toTigerXML());
* }
* collectChar = false;
* } */
}
}
else if (@event == XMLStreamConstants.END_DOCUMENT)
{
if (syntaxGraph.HasTokens())
{
sentenceCount++;
}
if (cIterations < nIterations)
{
cIterations++;
reopen();
return(true);
}
return(false);
}
else if (@event == XMLStreamConstants.CHARACTERS)
{
// if (collectChar) {
// char[] ch = reader.getTextCharacters();
// final int size = reader.getTextStart()+reader.getTextLength();
// for (int i = reader.getTextStart(); i < size; i++) {
// elementContent.append(ch[i]);
// }
// }
}
}
}
catch (XMLStreamException e)
{
throw new DataFormatException("", e);
}
}