/// <summary>
/// demoDP demonstrates turning a file into tokens and then parse
/// trees.
/// </summary>
/// <remarks>
/// demoDP demonstrates turning a file into tokens and then parse
/// trees. Note that the trees are printed by calling pennPrint on
/// the Tree object. It is also possible to pass a PrintWriter to
/// pennPrint if you want to capture the output.
/// This code will work with any supported language.
/// </remarks>
public static void DemoDP(LexicalizedParser lp, string filename)
{
// This option shows loading, sentence-segmenting and tokenizing
// a file using DocumentPreprocessor.
ITreebankLanguagePack tlp = lp.TreebankLanguagePack();
// a PennTreebankLanguagePack for English
IGrammaticalStructureFactory gsf = null;
if (tlp.SupportsGrammaticalStructures())
{
gsf = tlp.GrammaticalStructureFactory();
}
// You could also create a tokenizer here (as below) and pass it
// to DocumentPreprocessor
foreach (IList <IHasWord> sentence in new DocumentPreprocessor(filename))
{
Tree parse = lp.Apply(sentence);
parse.PennPrint();
System.Console.Out.WriteLine();
if (gsf != null)
{
GrammaticalStructure gs = gsf.NewGrammaticalStructure(parse);
ICollection tdl = gs.TypedDependenciesCCprocessed();
System.Console.Out.WriteLine(tdl);
System.Console.Out.WriteLine();
}
}
}
/// <summary>
/// demoAPI demonstrates other ways of calling the parser with
/// already tokenized text, or in some cases, raw text that needs to
/// be tokenized as a single sentence.
/// </summary>
/// <remarks>
/// demoAPI demonstrates other ways of calling the parser with
/// already tokenized text, or in some cases, raw text that needs to
/// be tokenized as a single sentence. Output is handled with a
/// TreePrint object. Note that the options used when creating the
/// TreePrint can determine what results to print out. Once again,
/// one can capture the output by passing a PrintWriter to
/// TreePrint.printTree. This code is for English.
/// </remarks>
public static void DemoAPI(LexicalizedParser lp)
{
// This option shows parsing a list of correctly tokenized words
string[] sent = new string[] { "This", "is", "an", "easy", "sentence", "." };
IList <CoreLabel> rawWords = SentenceUtils.ToCoreLabelList(sent);
Tree parse = lp.Apply(rawWords);
parse.PennPrint();
System.Console.Out.WriteLine();
// This option shows loading and using an explicit tokenizer
string sent2 = "This is another sentence.";
ITokenizerFactory <CoreLabel> tokenizerFactory = PTBTokenizer.Factory(new CoreLabelTokenFactory(), string.Empty);
ITokenizer <CoreLabel> tok = tokenizerFactory.GetTokenizer(new StringReader(sent2));
IList <CoreLabel> rawWords2 = tok.Tokenize();
parse = lp.Apply(rawWords2);
ITreebankLanguagePack tlp = lp.TreebankLanguagePack();
// PennTreebankLanguagePack for English
IGrammaticalStructureFactory gsf = tlp.GrammaticalStructureFactory();
GrammaticalStructure gs = gsf.NewGrammaticalStructure(parse);
IList <TypedDependency> tdl = gs.TypedDependenciesCCprocessed();
System.Console.Out.WriteLine(tdl);
System.Console.Out.WriteLine();
// You can also use a TreePrint object to print trees and dependencies
TreePrint tp = new TreePrint("penn,typedDependenciesCollapsed");
tp.PrintTree(parse);
}
/// <summary>This example shows a few more ways of providing input to a parser.</summary>
/// <remarks>
/// This example shows a few more ways of providing input to a parser.
/// Usage: ParserDemo2 [grammar [textFile]]
/// </remarks>
/// <exception cref="System.IO.IOException"/>
public static void Main(string[] args)
{
string grammar = args.Length > 0 ? args[0] : "edu/stanford/nlp/models/lexparser/englishPCFG.ser.gz";
string[] options = new string[] { "-maxLength", "80", "-retainTmpSubcategories" };
LexicalizedParser lp = ((LexicalizedParser)LexicalizedParser.LoadModel(grammar, options));
ITreebankLanguagePack tlp = lp.GetOp().Langpack();
IGrammaticalStructureFactory gsf = tlp.GrammaticalStructureFactory();
IEnumerable <IList <IHasWord> > sentences;
if (args.Length > 1)
{
DocumentPreprocessor dp = new DocumentPreprocessor(args[1]);
IList <IList <IHasWord> > tmp = new List <IList <IHasWord> >();
foreach (IList <IHasWord> sentence in dp)
{
tmp.Add(sentence);
}
sentences = tmp;
}
else
{
// Showing tokenization and parsing in code a couple of different ways.
string[] sent = new string[] { "This", "is", "an", "easy", "sentence", "." };
IList <IHasWord> sentence = new List <IHasWord>();
foreach (string word in sent)
{
sentence.Add(new Word(word));
}
string sent2 = ("This is a slightly longer and more complex " + "sentence requiring tokenization.");
// Use the default tokenizer for this TreebankLanguagePack
ITokenizer <IHasWord> toke = tlp.GetTokenizerFactory().GetTokenizer(new StringReader(sent2));
IList <IHasWord> sentence2 = toke.Tokenize();
string[] sent3 = new string[] { "It", "can", "can", "it", "." };
string[] tag3 = new string[] { "PRP", "MD", "VB", "PRP", "." };
// Parser gets second "can" wrong without help
IList <TaggedWord> sentence3 = new List <TaggedWord>();
for (int i = 0; i < sent3.Length; i++)
{
sentence3.Add(new TaggedWord(sent3[i], tag3[i]));
}
Tree parse = lp.Parse(sentence3);
parse.PennPrint();
IList <IList <IHasWord> > tmp = new List <IList <IHasWord> >();
tmp.Add(sentence);
tmp.Add(sentence2);
tmp.Add(sentence3);
sentences = tmp;
}
foreach (IList <IHasWord> sentence_1 in sentences)
{
Tree parse = lp.Parse(sentence_1);
parse.PennPrint();
System.Console.Out.WriteLine();
GrammaticalStructure gs = gsf.NewGrammaticalStructure(parse);
IList <TypedDependency> tdl = gs.TypedDependenciesCCprocessed();
System.Console.Out.WriteLine(tdl);
System.Console.Out.WriteLine();
System.Console.Out.WriteLine("The words of the sentence:");
foreach (ILabel lab in parse.Yield())
{
if (lab is CoreLabel)
{
System.Console.Out.WriteLine(((CoreLabel)lab).ToString(CoreLabel.OutputFormat.ValueMap));
}
else
{
System.Console.Out.WriteLine(lab);
}
}
System.Console.Out.WriteLine();
System.Console.Out.WriteLine(parse.TaggedYield());
System.Console.Out.WriteLine();
}
// This method turns the String into a single sentence using the
// default tokenizer for the TreebankLanguagePack.
string sent3_1 = "This is one last test!";
lp.Parse(sent3_1).PennPrint();
}
// TODO: these booleans would be more readable as enums similar to Mode.
// Then the arguments would make more sense
public static SemanticGraph MakeFromTree(GrammaticalStructure gs, SemanticGraphFactory.Mode mode, GrammaticalStructure.Extras includeExtras, IPredicate <TypedDependency> filter)
{
ICollection <TypedDependency> deps;
switch (mode)
{
case SemanticGraphFactory.Mode.Enhanced:
{
deps = gs.TypedDependenciesEnhanced();
break;
}
case SemanticGraphFactory.Mode.EnhancedPlusPlus:
{
deps = gs.TypedDependenciesEnhancedPlusPlus();
break;
}
case SemanticGraphFactory.Mode.CollapsedTree:
{
deps = gs.TypedDependenciesCollapsedTree();
break;
}
case SemanticGraphFactory.Mode.Collapsed:
{
deps = gs.TypedDependenciesCollapsed(includeExtras);
break;
}
case SemanticGraphFactory.Mode.Ccprocessed:
{
deps = gs.TypedDependenciesCCprocessed(includeExtras);
break;
}
case SemanticGraphFactory.Mode.Basic:
{
deps = gs.TypedDependencies(includeExtras);
break;
}
default:
{
throw new ArgumentException("Unknown mode " + mode);
}
}
if (filter != null)
{
IList <TypedDependency> depsFiltered = Generics.NewArrayList();
foreach (TypedDependency td in deps)
{
if (filter.Test(td))
{
depsFiltered.Add(td);
}
}
deps = depsFiltered;
}
// there used to be an if clause that filtered out the case of empty
// dependencies. However, I could not understand (or replicate) the error
// it alluded to, and it led to empty dependency graphs for very short fragments,
// which meant they were ignored by the RTE system. Changed. (pado)
// See also the SemanticGraph constructor.
//log.info(deps.toString());
return(new SemanticGraph(deps));
}
