/// <summary>Converts a tree to the Morfette training format.</summary>
private static string TreeToMorfette(Tree tree)
{
StringBuilder sb = new StringBuilder();
IList <ILabel> yield = tree.Yield();
IList <ILabel> tagYield = tree.PreTerminalYield();
System.Diagnostics.Debug.Assert(yield.Count == tagYield.Count);
int listLen = yield.Count;
for (int i = 0; i < listLen; ++i)
{
CoreLabel token = (CoreLabel)yield[i];
CoreLabel tag = (CoreLabel)tagYield[i];
string morphStr = token.OriginalText();
if (morphStr == null || morphStr.Equals(string.Empty))
{
morphStr = tag.Value();
}
string lemma = token.Lemma();
if (lemma == null || lemma.Equals(string.Empty))
{
lemma = token.Value();
}
sb.Append(string.Format("%s %s %s%n", token.Value(), lemma, morphStr));
}
return(sb.ToString());
}
/// <param name="args"/>
public static void Main(string[] args)
{
if (args.Length != 1)
{
System.Console.Error.Printf("Usage: java %s tree_file%n", typeof(TreeToMorfette).FullName);
System.Environment.Exit(-1);
}
string treeFile = args[0];
ITreeReaderFactory trf = new FrenchTreeReaderFactory();
try
{
ITreeReader tr = trf.NewTreeReader(new BufferedReader(new InputStreamReader(new FileInputStream(treeFile), "UTF-8")));
for (Tree tree1; (tree1 = tr.ReadTree()) != null;)
{
IList <ILabel> pretermYield = tree1.PreTerminalYield();
IList <ILabel> yield = tree1.Yield();
int yieldLen = yield.Count;
for (int i = 0; i < yieldLen; ++i)
{
CoreLabel rawToken = (CoreLabel)yield[i];
string word = rawToken.Value();
string morphStr = rawToken.OriginalText();
Pair <string, string> lemmaMorph = MorphoFeatureSpecification.SplitMorphString(word, morphStr);
string lemma = lemmaMorph.First();
string morph = lemmaMorph.Second();
if (morph == null || morph.Equals(string.Empty) || morph.Equals("XXX"))
{
morph = ((CoreLabel)pretermYield[i]).Value();
}
System.Console.Out.Printf("%s %s %s%n", word, lemma, morph);
}
System.Console.Out.WriteLine();
}
tr.Close();
}
catch (UnsupportedEncodingException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (FileNotFoundException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (IOException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
}
private static void ReplacePOSTag(Tree t, MorphoFeatureSpecification morpho)
{
if (!t.IsPreTerminal())
{
throw new ArgumentException("Can only operate on preterminals");
}
if (!(t.Label() is CoreLabel))
{
throw new ArgumentException("Only operates on CoreLabels");
}
CoreLabel label = (CoreLabel)t.Label();
Tree child = t.Children()[0];
if (!(child.Label() is CoreLabel))
{
throw new ArgumentException("Only operates on CoreLabels");
}
CoreLabel childLabel = (CoreLabel)child.Label();
// Morphological Analysis
string morphStr = childLabel.OriginalText();
if (morphStr == null || morphStr.Equals(string.Empty))
{
morphStr = label.Value();
// POS subcategory
string subCat = childLabel.Category();
if (subCat != null && subCat != string.Empty)
{
morphStr += "-" + subCat + "--";
}
else
{
morphStr += "---";
}
}
MorphoFeatures feats = morpho.StrToFeatures(morphStr);
if (feats.GetAltTag() != null && !feats.GetAltTag().Equals(string.Empty))
{
label.SetValue(feats.GetAltTag());
label.SetTag(feats.GetAltTag());
}
}
public static void MungeLeaves(Tree tree, bool lemmasAsLeaves, bool addMorphoToLeaves)
{
IList <ILabel> labels = tree.Yield();
foreach (ILabel label in labels)
{
++nTokens;
if (!(label is CoreLabel))
{
throw new ArgumentException("Only works with CoreLabels trees");
}
CoreLabel coreLabel = (CoreLabel)label;
string lemma = coreLabel.Lemma();
//PTB escaping since we're going to put this in the leaf
if (lemma == null)
{
// No lemma, so just add the surface form
lemma = coreLabel.Word();
}
else
{
if (lemma.Equals("("))
{
lemma = "-LRB-";
}
else
{
if (lemma.Equals(")"))
{
lemma = "-RRB-";
}
}
}
if (lemmasAsLeaves)
{
string escapedLemma = lemma;
coreLabel.SetWord(escapedLemma);
coreLabel.SetValue(escapedLemma);
coreLabel.SetLemma(lemma);
}
if (addMorphoToLeaves)
{
string morphStr = coreLabel.OriginalText();
if (morphStr == null || morphStr.Equals(string.Empty))
{
morphStr = MorphoFeatureSpecification.NoAnalysis;
}
else
{
++nMorphAnalyses;
}
// Normalize punctuation analyses
if (morphStr.StartsWith("PONCT"))
{
morphStr = "PUNC";
}
string newLeaf = string.Format("%s%s%s%s%s", coreLabel.Value(), MorphoFeatureSpecification.MorphoMark, lemma, MorphoFeatureSpecification.LemmaMark, morphStr);
coreLabel.SetValue(newLeaf);
coreLabel.SetWord(newLeaf);
}
}
}
