private void Init(System.IO.TextReader input, HebMorph.StreamLemmatizer _lemmatizer,
HebMorph.LemmaFilters.LemmaFilterBase _lemmaFilter, bool AlwaysSaveMarkedOriginal)
{
termAtt = AddAttribute <ITermAttribute>();
offsetAtt = AddAttribute<IOffsetAttribute>();
posIncrAtt = AddAttribute<IPositionIncrementAttribute>();
typeAtt = AddAttribute <ITypeAttribute>();
//payAtt = (PayloadAttribute)AddAttribute(typeof(PayloadAttribute));
this.input = input;
this._streamLemmatizer = _lemmatizer;
this._streamLemmatizer.SetStream(input);
this.alwaysSaveMarkedOriginal = AlwaysSaveMarkedOriginal;
this.lemmaFilter = _lemmaFilter;
}
private void Init(System.IO.TextReader input, HebMorph.StreamLemmatizer _lemmatizer,
HebMorph.LemmaFilters.LemmaFilterBase _lemmaFilter, bool AlwaysSaveMarkedOriginal)
{
termAtt = AddAttribute <ITermAttribute>();
offsetAtt = AddAttribute <IOffsetAttribute>();
posIncrAtt = AddAttribute <IPositionIncrementAttribute>();
typeAtt = AddAttribute <ITypeAttribute>();
//payAtt = (PayloadAttribute)AddAttribute(typeof(PayloadAttribute));
this.input = input;
this._streamLemmatizer = _lemmatizer;
this._streamLemmatizer.SetStream(input);
this.alwaysSaveMarkedOriginal = AlwaysSaveMarkedOriginal;
this.lemmaFilter = _lemmaFilter;
}
public void Run(string reportPath)
{
radix = null;
radix = new DictRadix<CoverageData>();
ReportProgress(0, "Initializing hspell...", true);
lemmatizer = new HebMorph.StreamLemmatizer(HSpellPath, true, false) {TolerateWhenLemmatizingStream = false};
corpusReader.OnDocument += GotDocument;
corpusReader.OnProgress += ReportProgress;
corpusReader.AbortReading = false;
corpusReader.Read();
if (!WasAbortSet && !string.IsNullOrEmpty(reportPath))
{
SaveReport(reportPath);
}
ReportProgress(100, "Finalizing...", false);
}
public MorphAnalyzer(string HSpellDataFilesPath)
: base()
{
hebMorphLemmatizer = new HebMorph.StreamLemmatizer();
hebMorphLemmatizer.InitFromHSpellFolder(HSpellDataFilesPath, true, false);
}
public MorphAnalyzer(HebMorph.StreamLemmatizer hml)
: base()
{
hebMorphLemmatizer = hml;
}
public StreamLemmasFilter(System.IO.TextReader input, HebMorph.StreamLemmatizer _lemmatizer,
HebMorph.LemmaFilters.LemmaFilterBase _lemmaFilter)
//: base(input) <- converts to CharStream, and causes issues due to a call to ReadToEnd in ctor
{
Init(input, _lemmatizer, _lemmaFilter, false);
}
public StreamLemmasFilter(System.IO.TextReader input, HebMorph.StreamLemmatizer _lemmatizer,
HebMorph.LemmaFilters.LemmaFilterBase _lemmaFilter, bool AlwaysSaveMarkedOriginal)
//: base(input) <- converts to CharStream, and causes issues due to a call to ReadToEnd in ctor
{
Init(input, _lemmatizer, _lemmaFilter, AlwaysSaveMarkedOriginal);
}
public MorphAnalyzer(MorphAnalyzer other)
: base()
{
hebMorphLemmatizer = other.hebMorphLemmatizer;
}
