public virtual void Annotate(Annotation annotation)
{
if (annotation.ContainsKey(typeof(CoreAnnotations.SentencesAnnotation)))
{
// parse a tree for each sentence
foreach (ICoreMap sentence in annotation.Get(typeof(CoreAnnotations.SentencesAnnotation)))
{
IList <CoreLabel> words = sentence.Get(typeof(CoreAnnotations.TokensAnnotation));
if (Verbose)
{
log.Info("Parsing: " + words);
}
int maxSentenceLength = parser.GetMaxSentenceLength();
// generate the constituent tree
Tree tree;
// initialized below
if (maxSentenceLength <= 0 || words.Count < maxSentenceLength)
{
tree = parser.GetBestParse(words);
}
else
{
tree = ParserUtils.XTree(words);
}
IList <Tree> trees = Generics.NewArrayList(1);
trees.Add(tree);
ParserAnnotatorUtils.FillInParseAnnotations(Verbose, BuildGraphs, gsf, sentence, trees, GrammaticalStructure.Extras.None);
}
}
else
{
throw new Exception("unable to find sentences in: " + annotation);
}
}
private void FinishSentence(ICoreMap sentence, IList <Tree> trees)
{
if (treeMap != null)
{
IList <Tree> mappedTrees = Generics.NewLinkedList();
foreach (Tree tree in trees)
{
Tree mappedTree = treeMap.Apply(tree);
mappedTrees.Add(mappedTree);
}
trees = mappedTrees;
}
ParserAnnotatorUtils.FillInParseAnnotations(Verbose, BuildGraphs, gsf, sentence, trees, extraDependencies);
if (saveBinaryTrees)
{
TreeBinarizer binarizer = TreeBinarizer.SimpleTreeBinarizer(parser.GetTLPParams().HeadFinder(), parser.TreebankLanguagePack());
Tree binarized = binarizer.TransformTree(trees[0]);
Edu.Stanford.Nlp.Trees.Trees.ConvertToCoreLabels(binarized);
sentence.Set(typeof(TreeCoreAnnotations.BinarizedTreeAnnotation), binarized);
}
// for some reason in some corner cases nodes aren't having sentenceIndex set
// do a pass and make sure all nodes have sentenceIndex set
SemanticGraph sg = sentence.Get(typeof(SemanticGraphCoreAnnotations.CollapsedDependenciesAnnotation));
if (sg != null)
{
foreach (IndexedWord iw in sg.VertexSet())
{
if (iw.Get(typeof(CoreAnnotations.SentenceIndexAnnotation)) == null && sentence.Get(typeof(CoreAnnotations.SentenceIndexAnnotation)) != null)
{
iw.SetSentIndex(sentence.Get(typeof(CoreAnnotations.SentenceIndexAnnotation)));
}
}
}
}