/// <summary>
/// ReadObject wraps the loop to read each property in an object and call the corresponding
/// setter to set it.
/// </summary>
/// <remarks>
/// This works for classes only, as the instance can't be passed by ref.
/// Structs can serialize as arrays or directly implement the loop to decode themselves.
/// Setters take a reference to the instance so that the Dictionary can be static per type,
/// which is critical for acceptable performance.
/// </remarks>
/// <typeparam name="T">Type being deserialized</typeparam>
/// <param name="reader">ITreeReader being read from</param>
/// <param name="instance">T instance being initialized</param>
/// <param name="setters">Dictionary of setter per field name</param>
/// <param name="throwOnUnknown">Throw if property name not in setters found</param>
public static void ReadObject <T>(this ITreeReader reader, T instance, Dictionary <string, Setter <T> > setters, bool throwOnUnknown = true) where T : ITreeSerializable
{
// Ensure object state reset before Read
instance.Clear();
reader.Expect(TreeToken.StartObject);
reader.Read();
while (reader.TokenType == TreeToken.PropertyName)
{
string propertyName = reader.ReadAsString();
reader.Read();
if (setters.TryGetValue(propertyName, out Setter <T> setter))
{
setter(reader, instance);
reader.Read();
}
else
{
if (throwOnUnknown)
{
throw new IOException($"Found unknown {typeof(T).Name} property, \"{propertyName}\", expected one of \"{String.Join("; ", setters.Keys)}\" at {reader.Position:n0} using {reader.GetType().Name}.");
}
else
{
reader.Skip();
}
}
}
reader.Expect(TreeToken.EndObject);
// EndObject must be left for caller to handle
}
/// <summary>Processes a single file containing AnCora XML trees.</summary>
/// <remarks>
/// Processes a single file containing AnCora XML trees. Returns MWE statistics for the trees in
/// the file and the actual parsed trees.
/// </remarks>
private static Pair <TwoDimensionalCounter <string, string>, IList <Tree> > ProcessTreeFile(File file, SpanishXMLTreeReaderFactory trf, string encoding)
{
TwoDimensionalCounter <string, string> tagger = new TwoDimensionalCounter <string, string>();
try
{
Reader @in = new BufferedReader(new InputStreamReader(new FileInputStream(file), encoding));
ITreeReader tr = trf.NewTreeReader(file.GetPath(), @in);
IList <Tree> trees = new List <Tree>();
Tree t;
Tree splitPoint;
while ((t = tr.ReadTree()) != null)
{
do
{
// We may need to split the current tree into multiple parts.
// (If not, a call to `split` with a `null` split-point is a
// no-op
splitPoint = FindSplitPoint(t);
Pair <Tree, Tree> split = Split(t, splitPoint);
Tree toAdd = split.First();
t = split.Second();
trees.Add(toAdd);
UpdateTagger(tagger, toAdd);
}while (splitPoint != null);
}
tr.Close();
return(new Pair <TwoDimensionalCounter <string, string>, IList <Tree> >(tagger, trees));
}
catch (IOException e)
{
Sharpen.Runtime.PrintStackTrace(e);
return(null);
}
}
/// <summary>
/// Read existing items in an existing Dictionary instance.
/// Used with Dictionaries of specific things which may or may not be present in the file, like Table.Columns.
/// </summary>
/// <typeparam name="T">Type of values in Dictionary</typeparam>
/// <param name="reader">ITreeReader to read from</param>
/// <param name="dictionary">Dictionary containing items to read</param>
/// <param name="throwOnUnknown">True to throw for property name not in Dictionary, false to quietly skip over it</param>
public static void ReadDictionaryItems <T>(this ITreeReader reader, Dictionary <string, T> dictionary, bool throwOnUnknown = true) where T : ITreeSerializable
{
reader.Expect(TreeToken.StartObject);
reader.Read();
while (reader.TokenType == TreeToken.PropertyName)
{
string itemName = reader.ReadAsString();
reader.Read();
if (dictionary.TryGetValue(itemName, out T item))
{
item.Read(reader);
reader.Read();
}
else
{
if (throwOnUnknown)
{
throw new IOException($"Found unknown {typeof(T).Name} property \"{itemName}\", expected one of \"{String.Join("; ", dictionary.Keys)}\" at {reader.Position:n0} using {reader.GetType().Name}.");
}
else
{
reader.Skip();
}
}
}
reader.Expect(TreeToken.EndObject);
}
public static Dictionary <string, T> ReadStringDictionary <T>(this ITreeReader reader, Func <T> ctor) where T : ITreeSerializable
{
if (reader.TokenType == TreeToken.Null)
{
return(null);
}
Dictionary <string, T> result = new Dictionary <string, T>();
reader.Expect(TreeToken.StartObject);
reader.Read();
while (reader.TokenType == TreeToken.PropertyName)
{
string key = reader.ReadAsString();
reader.Read();
T value = ctor();
value.Read(reader);
result[key] = value;
reader.Read();
}
reader.Expect(TreeToken.EndObject);
return(result);
}
/// <exception cref="System.IO.IOException"/>
public virtual void Process()
{
SpanishXMLTreeReaderFactory trf = new SpanishXMLTreeReaderFactory();
Tree t;
foreach (File file in fileList)
{
Reader @in = new BufferedReader(new InputStreamReader(new FileInputStream(file), AncoraEncoding));
ITreeReader tr = trf.NewTreeReader(@in);
// Tree reading will implicitly perform tree normalization for us
while ((t = tr.ReadTree()) != null)
{
// Update tagger with this tree
IList <CoreLabel> yield = t.TaggedLabeledYield();
foreach (CoreLabel leafLabel in yield)
{
if (leafLabel.Tag().Equals(SpanishTreeNormalizer.MwTag))
{
continue;
}
unigramTagger.IncrementCount(leafLabel.Word(), leafLabel.Tag());
}
}
}
}
public static void Load(this ITreeSerializable item, Stream stream, TreeFormat format, TreeSerializationSettings settings = null)
{
using (ITreeReader reader = TreeSerializer.Reader(format, stream, settings))
{
item.Read(reader);
}
}
public void Read(ITreeReader reader)
{
Clear();
// Read Tables, skipping unknown tables if Settings.Strict == false
reader.ReadDictionaryItems(Tables, throwOnUnknown: reader.Settings.Strict);
}
private static void ResolveDummyTags(File treeFile, TwoDimensionalCounter <string, string> pretermLabel, TwoDimensionalCounter <string, string> unigramTagger)
{
try
{
BufferedReader br = new BufferedReader(new InputStreamReader(new FileInputStream(treeFile), "UTF-8"));
ITreeReaderFactory trf = new FrenchTreeReaderFactory();
ITreeReader tr = trf.NewTreeReader(br);
PrintWriter pw = new PrintWriter(new TextWriter(new FileOutputStream(new File(treeFile + ".fixed")), false, "UTF-8"));
int nTrees = 0;
for (Tree t; (t = tr.ReadTree()) != null; nTrees++)
{
TraverseAndFix(t, pretermLabel, unigramTagger);
pw.Println(t.ToString());
}
pw.Close();
tr.Close();
System.Console.Out.WriteLine("Processed " + nTrees + " trees");
}
catch (UnsupportedEncodingException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (FileNotFoundException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (IOException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
}
/// <param name="args">File to run on</param>
public static void Main(string[] args)
{
if (args.Length < 1)
{
System.Console.Out.Printf("Usage: java %s tree_file%n", typeof(Edu.Stanford.Nlp.Trees.International.Negra.NegraPennTreeReaderFactory).FullName);
return;
}
ITreebankLanguagePack tlp = new NegraPennLanguagePack();
ITreeReaderFactory trf = new Edu.Stanford.Nlp.Trees.International.Negra.NegraPennTreeReaderFactory(2, false, false, tlp);
try
{
ITreeReader tr = trf.NewTreeReader(IOUtils.ReaderFromString(args[0], tlp.GetEncoding()));
for (Tree t; (t = tr.ReadTree()) != null;)
{
t.PennPrint();
}
tr.Close();
}
catch (UnsupportedEncodingException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (FileNotFoundException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (IOException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
}
/// <summary>Load a collection of parse trees from a Reader.</summary>
/// <remarks>
/// Load a collection of parse trees from a Reader.
/// Each tree may optionally be encased in parens to allow for Penn
/// Treebank style trees.
/// </remarks>
/// <param name="r">
/// The reader to read trees from. (If you want it buffered,
/// you should already have buffered it!)
/// </param>
/// <param name="id">
/// An ID for where these files come from (arbitrary, but
/// something like a filename. Can be <code>null</code> for none.
/// </param>
public void Load(Reader r, string id)
{
try
{
// could throw an IO exception?
ITreeReader tr = TreeReaderFactory().NewTreeReader(r);
int sentIndex = 0;
for (Tree pt; (pt = tr.ReadTree()) != null;)
{
if (pt.Label() is IHasIndex)
{
// so we can trace where this tree came from
IHasIndex hi = (IHasIndex)pt.Label();
if (id != null)
{
hi.SetDocID(id);
}
hi.SetSentIndex(sentIndex);
}
parseTrees.Add(pt);
sentIndex++;
}
}
catch (IOException e)
{
log.Info("load IO Exception: " + e);
}
}
public void Read(ITreeReader reader)
{
Array = reader.ReadBlockArray <T>();
Index = 0;
Count = Array.Length;
IsExpandable = false;
}
public static Dictionary <int, T> ReadIntDictionary <T>(this ITreeReader reader, Func <T> ctor) where T : ITreeSerializable
{
if (reader.TokenType == TreeToken.Null)
{
return(null);
}
Dictionary <int, T> result = new Dictionary <int, T>();
reader.Expect(TreeToken.StartArray);
reader.Read();
int[] keys = reader.ReadBlockArray <int>();
reader.Read();
reader.Expect(TreeToken.StartArray);
reader.Read();
for (int i = 0; i < keys.Length; ++i)
{
int key = keys[i];
T value = ctor();
value.Read(reader);
result[key] = value;
reader.Read();
}
reader.Expect(TreeToken.EndArray);
reader.Read();
reader.Expect(TreeToken.EndArray);
return(result);
}
/// <summary>
/// Verify the current token is a required value, otherwise throw a good exception.
/// </summary>
/// <param name="reader">ITreeReader to check</param>
/// <param name="expected">Current TreeToken expected</param>
public static void Expect(this ITreeReader reader, TreeToken expected)
{
if (reader.TokenType != expected)
{
throw new IOException($"{reader.GetType().Name} expected \"{expected}\" but found \"{reader.TokenType}\" at {reader.Position:n0}");
}
}
public TreeDiagnosticsReader(ITreeReader inner)
{
Inner = inner;
LastPropertyName = "<Database>";
Path = new Stack <TreeDiagnostics>();
Tree = Open();
}
/// <param name="args"/>
public static void Main(string[] args)
{
if (args.Length != 1)
{
System.Console.Error.Printf("Usage: java %s tree_file%n", typeof(SplitMaker).FullName);
System.Environment.Exit(-1);
}
ITreebankLanguagePack tlp = new HebrewTreebankLanguagePack();
string inputFile = args[0];
File treeFile = new File(inputFile);
try
{
ITreeReaderFactory trf = new HebrewTreeReaderFactory();
BufferedReader br = new BufferedReader(new InputStreamReader(new FileInputStream(treeFile), tlp.GetEncoding()));
ITreeReader tr = trf.NewTreeReader(br);
PrintWriter pwDev = new PrintWriter(new TextWriter(new FileOutputStream(inputFile + ".clean.dev"), false, tlp.GetEncoding()));
PrintWriter pwTrain = new PrintWriter(new TextWriter(new FileOutputStream(inputFile + ".clean.train"), false, tlp.GetEncoding()));
PrintWriter pwTest = new PrintWriter(new TextWriter(new FileOutputStream(inputFile + ".clean.test"), false, tlp.GetEncoding()));
int numTrees = 0;
for (Tree t; ((t = tr.ReadTree()) != null); numTrees++)
{
if (numTrees < 483)
{
pwDev.Println(t.ToString());
}
else
{
if (numTrees >= 483 && numTrees < 5724)
{
pwTrain.Println(t.ToString());
}
else
{
pwTest.Println(t.ToString());
}
}
}
tr.Close();
pwDev.Close();
pwTrain.Close();
pwTest.Close();
System.Console.Error.Printf("Processed %d trees.%n", numTrees);
}
catch (UnsupportedEncodingException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (FileNotFoundException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (IOException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
}
/// <summary>For debugging.</summary>
/// <param name="args"/>
public static void Main(string[] args)
{
if (args.Length < 1)
{
System.Console.Error.Printf("Usage: java %s tree_file(s)%n%n", typeof(Edu.Stanford.Nlp.Trees.International.French.FrenchXMLTreeReader).FullName);
System.Environment.Exit(-1);
}
IList <File> fileList = new List <File>();
foreach (string arg in args)
{
fileList.Add(new File(arg));
}
ITreeReaderFactory trf = new FrenchXMLTreeReaderFactory(false);
int totalTrees = 0;
ICollection <string> morphAnalyses = Generics.NewHashSet();
try
{
foreach (File file in fileList)
{
ITreeReader tr = trf.NewTreeReader(new BufferedReader(new InputStreamReader(new FileInputStream(file), "UTF-8")));
Tree t;
int numTrees;
string canonicalFileName = Sharpen.Runtime.Substring(file.GetName(), 0, file.GetName().LastIndexOf('.'));
for (numTrees = 0; (t = tr.ReadTree()) != null; numTrees++)
{
string ftbID = ((CoreLabel)t.Label()).Get(typeof(CoreAnnotations.SentenceIDAnnotation));
System.Console.Out.Printf("%s-%s\t%s%n", canonicalFileName, ftbID, t.ToString());
IList <ILabel> leaves = t.Yield();
foreach (ILabel label in leaves)
{
if (label is CoreLabel)
{
morphAnalyses.Add(((CoreLabel)label).OriginalText());
}
}
}
tr.Close();
System.Console.Error.Printf("%s: %d trees%n", file.GetName(), numTrees);
totalTrees += numTrees;
}
//wsg2011: Print out the observed morphological analyses
// for(String analysis : morphAnalyses)
// log.info(analysis);
System.Console.Error.Printf("%nRead %d trees%n", totalTrees);
}
catch (FileNotFoundException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (IOException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
}
public Sawyer(AddressQualityChecker addressQualityChecker, IGeocodeManager geocodeManager, ILocationCacheFactory locationCacheFactory, ILogger <Sawyer> logger, TreeParser treeParser, ITreeReader treeReader, ITreeWriter treeWriter)
{
this.addressQualityChecker = addressQualityChecker;
this.geocodeManager = geocodeManager;
this.locationCache = locationCacheFactory.Create();
this.logger = logger;
this.treeParser = treeParser;
this.treeReader = treeReader;
this.treeWriter = treeWriter;
}
public void Read(ITreeReader reader)
{
reader.ReadObject(this, setters);
if (_valueEndInPage != null)
{
Count = _valueEndInPage.Length;
_lastNonEmptyIndex = Count - 1;
}
}
public static void VerifySkip <T>(T value, TreeFormat format) where T : ITreeSerializable
{
// Test serialization details
using (MemoryStream stream = new MemoryStream())
{
TreeSerializationSettings settings = new TreeSerializationSettings()
{
LeaveStreamOpen = true
};
using (ITreeWriter writer = Writer(format, stream, settings))
{
value.Write(writer);
}
long bytesWritten = stream.Position;
// Read tokens individually and verify 'None' returned at end
stream.Seek(0, SeekOrigin.Begin);
using (ITreeReader reader = Reader(format, stream, settings))
{
while (reader.Read())
{
// Verify each token type is coming back properly (no reading random bytes)
Assert.True((byte)reader.TokenType <= (byte)TreeToken.BlockArray);
}
Assert.Equal(TreeToken.None, reader.TokenType);
Assert.Equal(bytesWritten, stream.Position);
}
// Verify Skip once skips everything (each ITreeSerializable must be one value or one root array or object
stream.Seek(0, SeekOrigin.Begin);
using (ITreeReader reader = Reader(format, stream, settings))
{
reader.Skip();
Assert.Equal(TreeToken.None, reader.TokenType);
Assert.Equal(bytesWritten, stream.Position);
}
// For objects, verify each property can be skipped correctly
// Each Skip should read the value, so that the next token is the next PropertyName
stream.Seek(0, SeekOrigin.Begin);
using (ITreeReader reader = Reader(format, stream, settings))
{
if (reader.TokenType == TreeToken.StartObject)
{
Empty empty = new Empty();
empty.Read(reader);
Assert.Equal(bytesWritten, stream.Position);
}
}
}
}
//Delete sentence-initial punctuation
//Delete sentence final punctuation that is preceded by punctuation (first time)
//Delete sentence final punctuation that is preceded by punctuation (second time)
//Convert remaining sentence-final punctuation to either . if it is not [.!?]
//Delete medial, sentence-final punctuation
//Now move the sentence-final mark under SENT
//For those trees that lack a sentence-final punc, add one.
//Finally, delete these punctuation marks, which I can't seem to kill otherwise...
//A bad MWADV tree in the training set
// Not sure why this got a label of X. Similar trees suggest it
// should be A instead
// This also seems to be mislabeled
/// <param name="args"/>
public static void Main(string[] args)
{
if (args.Length != 1)
{
log.Info("Usage: java " + typeof(Edu.Stanford.Nlp.International.French.Pipeline.FTBCorrector).FullName + " filename\n");
System.Environment.Exit(-1);
}
ITreeTransformer tt = new Edu.Stanford.Nlp.International.French.Pipeline.FTBCorrector();
File f = new File(args[0]);
try
{
//These bad trees in the Candito training set should be thrown out:
// (ROOT (SENT (" ") (. .)))
// (ROOT (SENT (. .)))
TregexPattern pBadTree = TregexPattern.Compile("@SENT <: @PUNC");
TregexPattern pBadTree2 = TregexPattern.Compile("@SENT <1 @PUNC <2 @PUNC !<3 __");
BufferedReader br = new BufferedReader(new InputStreamReader(new FileInputStream(f), "UTF-8"));
ITreeReaderFactory trf = new FrenchTreeReaderFactory();
ITreeReader tr = trf.NewTreeReader(br);
int nTrees = 0;
for (Tree t; (t = tr.ReadTree()) != null; nTrees++)
{
TregexMatcher m = pBadTree.Matcher(t);
TregexMatcher m2 = pBadTree2.Matcher(t);
if (m.Find() || m2.Find())
{
log.Info("Discarding tree: " + t.ToString());
}
else
{
Tree fixedT = tt.TransformTree(t);
System.Console.Out.WriteLine(fixedT.ToString());
}
}
tr.Close();
System.Console.Error.Printf("Wrote %d trees%n", nTrees);
}
catch (UnsupportedEncodingException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (FileNotFoundException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (IOException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (TregexParseException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
}
/// <summary>
/// Deserialize record from XML using short class name without namespace
/// for the root XML element.
/// </summary>
public static void ParseXml(this Key obj, string xmlString)
{
ITreeReader reader = new XmlTreeReader(xmlString);
// Root node of serialized XML must be the same as mapped class name without namespace
var mappedFullName = obj.GetType().Name;
ITreeReader recordNodes = reader.ReadElement(mappedFullName);
// Deserialize from XML nodes inside the root node
obj.DeserializeFrom(recordNodes);
}
/// <param name="args"/>
public static void Main(string[] args)
{
if (args.Length != 1)
{
System.Console.Error.Printf("Usage: java %s file%n", typeof(MWEPreprocessor).FullName);
System.Environment.Exit(-1);
}
File treeFile = new File(args[0]);
TwoDimensionalCounter <string, string> labelTerm = new TwoDimensionalCounter <string, string>();
TwoDimensionalCounter <string, string> termLabel = new TwoDimensionalCounter <string, string>();
TwoDimensionalCounter <string, string> labelPreterm = new TwoDimensionalCounter <string, string>();
TwoDimensionalCounter <string, string> pretermLabel = new TwoDimensionalCounter <string, string>();
TwoDimensionalCounter <string, string> unigramTagger = new TwoDimensionalCounter <string, string>();
try
{
BufferedReader br = new BufferedReader(new InputStreamReader(new FileInputStream(treeFile), "UTF-8"));
ITreeReaderFactory trf = new FrenchTreeReaderFactory();
ITreeReader tr = trf.NewTreeReader(br);
for (Tree t; (t = tr.ReadTree()) != null;)
{
CountMWEStatistics(t, unigramTagger, labelPreterm, pretermLabel, labelTerm, termLabel);
}
tr.Close();
//Closes the underlying reader
System.Console.Out.WriteLine("Generating {MWE Type -> Terminal}");
PrintCounter(labelTerm, "label_term.csv");
System.Console.Out.WriteLine("Generating {Terminal -> MWE Type}");
PrintCounter(termLabel, "term_label.csv");
System.Console.Out.WriteLine("Generating {MWE Type -> POS sequence}");
PrintCounter(labelPreterm, "label_pos.csv");
System.Console.Out.WriteLine("Generating {POS sequence -> MWE Type}");
PrintCounter(pretermLabel, "pos_label.csv");
System.Console.Out.WriteLine("Resolving DUMMY tags");
ResolveDummyTags(treeFile, pretermLabel, unigramTagger);
System.Console.Out.WriteLine("#Unknown Word Types: " + MWEPreprocessor.ManualUWModel.nUnknownWordTypes);
System.Console.Out.WriteLine("#Missing POS: " + nMissingPOS);
System.Console.Out.WriteLine("#Missing Phrasal: " + nMissingPhrasal);
System.Console.Out.WriteLine("Done!");
}
catch (UnsupportedEncodingException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (FileNotFoundException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (IOException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
}
public void Read(ITreeReader reader)
{
_chapters = reader.ReadList <ArraySliceChapter <T> >(() => new ArraySliceChapter <T>());
int chapterCount = _chapters?.Count ?? 0;
if (chapterCount > 0)
{
// Infer count; N - 1 full chapters and the actual row count from the last
_count = ArraySliceChapter <T> .ChapterRowCount * (chapterCount - 1) + _chapters[chapterCount - 1].Count;
}
}
/// <param name="args"/>
public static void Main(string[] args)
{
if (args.Length != 2)
{
System.Console.Error.Printf("Usage: java %s tree_file morfette_tnt_file%n", typeof(MungeTreesWithMorfetteAnalyses).FullName);
System.Environment.Exit(-1);
}
string treeFile = args[0];
string morfetteFile = args[1];
ITreeReaderFactory trf = new FrenchTreeReaderFactory();
try
{
ITreeReader tr = trf.NewTreeReader(new BufferedReader(new InputStreamReader(new FileInputStream(treeFile), "UTF-8")));
IEnumerator <IList <CoreLabel> > morfetteItr = new MungeTreesWithMorfetteAnalyses.MorfetteFileIterator(morfetteFile);
for (Tree tree; (tree = tr.ReadTree()) != null && morfetteItr.MoveNext();)
{
IList <CoreLabel> analysis = morfetteItr.Current;
IList <ILabel> yield = tree.Yield();
System.Diagnostics.Debug.Assert(analysis.Count == yield.Count);
int yieldLen = yield.Count;
for (int i = 0; i < yieldLen; ++i)
{
CoreLabel tokenAnalysis = analysis[i];
ILabel token = yield[i];
string lemma = GetLemma(token.Value(), tokenAnalysis.Lemma());
string newLeaf = string.Format("%s%s%s%s%s", token.Value(), MorphoFeatureSpecification.MorphoMark, lemma, MorphoFeatureSpecification.LemmaMark, tokenAnalysis.Tag());
((CoreLabel)token).SetValue(newLeaf);
}
System.Console.Out.WriteLine(tree.ToString());
}
if (tr.ReadTree() != null || morfetteItr.MoveNext())
{
log.Info("WARNING: Uneven input files!");
}
tr.Close();
}
catch (UnsupportedEncodingException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (FileNotFoundException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (IOException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
}
/// <summary>
/// Read a single element containing atomic value (returns empty string if not found).
/// Error message if more than one element with the specified name is present.
/// </summary>
public string ReadValueElement(string elementName)
{
ITreeReader elementReader = ReadElement(elementName);
// Return value if element is present, otherwise null
if (elementReader != null)
{
return(elementReader.ReadValue());
}
else
{
return(String.Empty);
}
}
/// <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);
}
}
/// <param name="args"/>
public static void Main(string[] args)
{
Properties options = StringUtils.ArgsToProperties(args, argOptionDefs);
if (!options.Contains(string.Empty) || options.Contains("help"))
{
log.Info(Usage());
return;
}
bool retainNER = PropertiesUtils.GetBool(options, "ner", false);
bool normalize = PropertiesUtils.GetBool(options, "normalize", true);
File treeFile = new File(options.GetProperty(string.Empty));
TwoDimensionalCounter <string, string> labelTerm = new TwoDimensionalCounter <string, string>();
TwoDimensionalCounter <string, string> termLabel = new TwoDimensionalCounter <string, string>();
TwoDimensionalCounter <string, string> labelPreterm = new TwoDimensionalCounter <string, string>();
TwoDimensionalCounter <string, string> pretermLabel = new TwoDimensionalCounter <string, string>();
TwoDimensionalCounter <string, string> unigramTagger = new TwoDimensionalCounter <string, string>();
try
{
BufferedReader br = new BufferedReader(new InputStreamReader(new FileInputStream(treeFile), "UTF-8"));
ITreeReaderFactory trf = new SpanishTreeReaderFactory();
ITreeReader tr = trf.NewTreeReader(br);
for (Tree t; (t = tr.ReadTree()) != null;)
{
UpdateTagger(unigramTagger, t);
}
tr.Close();
//Closes the underlying reader
System.Console.Out.WriteLine("Resolving DUMMY tags");
ResolveDummyTags(treeFile, unigramTagger, retainNER, normalize ? new SpanishTreeNormalizer(true, false, false) : null);
System.Console.Out.WriteLine("#Unknown Word Types: " + MultiWordPreprocessor.ManualUWModel.nUnknownWordTypes);
System.Console.Out.WriteLine(string.Format("#Missing POS: %d (fixed: %d, %.2f%%)", nMissingPOS, nFixedPOS, (double)nFixedPOS / nMissingPOS * 100));
System.Console.Out.WriteLine(string.Format("#Missing Phrasal: %d (fixed: %d, %.2f%%)", nMissingPhrasal, nFixedPhrasal, (double)nFixedPhrasal / nMissingPhrasal * 100));
System.Console.Out.WriteLine("Done!");
}
catch (UnsupportedEncodingException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (FileNotFoundException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
catch (IOException e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
}
public void Read(ITreeReader reader)
{
reader.ReadObject(this, setters);
if (IsNull.Count == 0 && Values.Count > 0)
{
// Only wrote values means all values are non-null
IsNull[Values.Count - 1] = false;
IsNull.SetAll(false);
}
else if (IsNull.Count > 0 && Values.Count == 0)
{
// Only wrote nulls means all values are null
Values[IsNull.Count - 1] = ArraySlice <byte> .Empty;
}
}
public virtual ITreeReader NewTreeReader(Reader @in)
{
ITreeReader tr = null;
if (noNormalization)
{
tr = new PennTreeReader(@in, new LabeledScoredTreeFactory(), new TreeNormalizer(), new ArabicTreebankTokenizer(@in));
}
else
{
tr = new PennTreeReader(@in, new LabeledScoredTreeFactory(), new ArabicTreeNormalizer(retainNPTmp, retainPRD, changeNoLabels, retainNPSbj, retainPPClr), new ArabicTreebankTokenizer(@in));
}
if (filterX)
{
tr = new FilteringTreeReader(tr, new ArabicTreeReaderFactory.XFilter());
}
return(tr);
}
public void Read(ITreeReader reader)
{
reader.ReadObject(this, setters);
if (IsNull != null)
{
if (IsNull.Count == 0 && Values.Count > 0)
{
// Only wrote values means all values are non-null
IsNull[Values.Count - 1] = false;
IsNull.SetAll(false);
}
else if (IsNull.Count > 0 && Values.Count == 0)
{
// Only wrote nulls means all values are null
Values[IsNull.Count - 1] = default(T);
}
}
}
