unsafe public void Run(List <Word> words
#if DEBUG
, bool applyMorphoAmbiguityPreProcess
#endif
)
{
var wordMorphology = default(WordMorphology);
for (int i = 0, wordsLength = words.Count; i < wordsLength; i++)
{
var word = words[i];
if (word.posTaggerExtraWordType != PosTaggerExtraWordType.__DEFAULT__)
{
_wordMorphoAmbiguities.Add(_wordMorphoAmbiguityFactory.Create(word, i));
continue;
}
switch (word.posTaggerInputType)
{
case PosTaggerInputType.O: // = "O"; // Другой
case PosTaggerInputType.AllLat: // - только латиница: нет строчных и точек;
case PosTaggerInputType.FstC: // - первая заглавная, не содержит пробелов;
{
if (word.valueUpper == null)
{
_wordMorphoAmbiguities.Add(_wordMorphoAmbiguityFactory.Create(word, i));
continue;
}
var mode = GetWordFormMorphologyMode(word, i);
wordMorphology = _morphoAnalyzer.GetWordMorphology_NoToUpper(word.valueUpper, mode);
}
break;
case PosTaggerInputType.Num: // – содержит хотя бы одну цифру и не содержит букв;
{
if (word.posTaggerLastValueUpperInNumeralChain == null)
{
_wordMorphoAmbiguities.Add(_wordMorphoAmbiguityFactory.Create(word, i));
continue;
}
var mode = GetWordFormMorphologyMode(word, i);
wordMorphology = _morphoAnalyzer.GetWordMorphology_4LastValueUpperInNumeralChain(
word.posTaggerLastValueUpperInNumeralChain, mode);
}
break;
default:
_wordMorphoAmbiguities.Add(_wordMorphoAmbiguityFactory.Create(word, i));
continue;
}
#region post process MorphoAnalyze result
if (wordMorphology.HasWordFormMorphologies)
{
var wfms = default(WordFormMorphology[]);
//Если данное слово имеет только одну часть речи, прописанную в морфословаре, то использовать ее вместо определённой с помощью PoS-tagger.
if (wordMorphology.IsSinglePartOfSpeech)
{
CorrectPosTaggerOutputType(word, wordMorphology.PartOfSpeech);
wfms = wordMorphology.WordFormMorphologies.ToArray();
_wordMorphoAmbiguities.Add(_wordMorphoAmbiguityFactory.Create(word, i, wfms));
}
else
{
#region clause #1
//для данного слова в морфословаре определено несколько частей речи.
//ищем среди морфоинформации по слову морфоинформацию по части речи от pos-tagger'а, если она есть - берем её
//вот эта хуйня из-за двойной трансляции
//{PosTaggerOutputType::AdjectivePronoun => PartOfSpeechEnum::Adjective, PartOfSpeechEnum::Pronoun}
// &
//{PosTaggerOutputType::AdverbialPronoun => PartOfSpeechEnum::Adverb, PartOfSpeechEnum::Pronoun}
var partOfSpeech = default(PartOfSpeechEnum?);
switch (word.posTaggerOutputType)
{
case PosTaggerOutputType.AdjectivePronoun:
{
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Adjective, PartOfSpeechEnum.Pronoun);
if (wfms != null)
{
_wordMorphoAmbiguities.Add(_wordMorphoAmbiguityFactory.Create(word, i, wfms));
continue;
}
}
break;
case PosTaggerOutputType.AdverbialPronoun:
{
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Adverb, PartOfSpeechEnum.Pronoun);
if (wfms != null)
{
_wordMorphoAmbiguities.Add(_wordMorphoAmbiguityFactory.Create(word, i, wfms));
continue;
}
}
break;
default:
{
partOfSpeech = ToPartOfSpeech(word.posTaggerOutputType);
if (partOfSpeech.HasValue)
{
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, partOfSpeech.Value);
if (wfms != null)
{
_wordMorphoAmbiguities.Add(_wordMorphoAmbiguityFactory.Create(word, i, wfms));
continue;
}
}
}
break;
}
#endregion
#region clause #2
//При этом для данного слова в морфословаре определено несколько частей речи.
//В данном случае в первую очередь м.б. соответствия (слева выход PoS-tagger, справа морфословарь, последовательность пунктов неважна):
switch (word.posTaggerOutputType)
{
/*2.3. Participle = Adjective */
case PosTaggerOutputType.Participle:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Adjective);
break;
default:
#region clause #3
{
if (partOfSpeech.HasValue)
{
switch (partOfSpeech.Value)
{
/*3.1. Pronoun = Noun*/
case PartOfSpeechEnum.Pronoun:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Noun);
break;
/*3.2. Noun = Pronoun */
case PartOfSpeechEnum.Noun:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Pronoun);
break;
/*3.3. Conjunction = Particle */
case PartOfSpeechEnum.Conjunction:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Particle);
break;
/*3.4. Particle = Conjunction*/
case PartOfSpeechEnum.Particle:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Conjunction);
break;
/*3.5. Numeral = Noun, Adjective */
case PartOfSpeechEnum.Numeral:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Noun, PartOfSpeechEnum.Adjective);
break;
/*3.6. Adjective = Verb, Adverb*/
case PartOfSpeechEnum.Adjective:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Verb, PartOfSpeechEnum.Adverb);
break;
/*3.7. Adverb = Adjective */
case PartOfSpeechEnum.Adverb:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Adjective);
break;
}
}
}
#endregion
break;
}
/*Если таковых соответствий не нашлось, то берется первая из выдачи морфословаря часть речи.*/
if (wfms == null)
{
var _partOfSpeech = wordMorphology.WordFormMorphologies[0].PartOfSpeech;
word.posTaggerOutputType = ToPosTaggerOutputType(_partOfSpeech);
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, _partOfSpeech);
}
_wordMorphoAmbiguities.Add(_wordMorphoAmbiguityFactory.Create(word, i, wfms));
#endregion
}
}
else
{
_wordMorphoAmbiguities.Add(_wordMorphoAmbiguityFactory.Create(word, i));
}
#endregion
}
#if DEBUG
if (applyMorphoAmbiguityPreProcess)
{
#endif
_morphoAmbiguityPreProcessor.Run(_wordMorphoAmbiguities);
#if DEBUG
}
#endif
_morphoAmbiguityResolver.Resolve(_wordMorphoAmbiguities);
_wordMorphoAmbiguities.Clear();
}
unsafe public void Run(List <word_t> words
#if DEBUG
, bool applyMorphoAmbiguityPreProcess
#endif
)
{
//-apply morpho-analysis-//
var wordMorphology = default(WordMorphology_t);
for (int i = 0, wordsLength = words.Count; i < wordsLength; i++)
{
var word = words[i];
#region [.sikp something & make Morpho-Analyze.]
if (word.posTaggerExtraWordType != PosTaggerExtraWordType.__DEFAULT__)
{
_WordMorphoAmbiguities.Add(_WordMorphoAmbiguityFactory.Create(word, i));
continue;
}
switch (word.posTaggerInputType)
{
case PosTaggerInputType.O: // = "O"; // Другой
case PosTaggerInputType.AllLat: // - только латиница: нет строчных и точек;
case PosTaggerInputType.FstC: // - первая заглавная, не содержит пробелов;
#region
{
if (word.valueUpper == null)
{
_WordMorphoAmbiguities.Add(_WordMorphoAmbiguityFactory.Create(word, i));
continue;
}
#if DEBUG
var wordFirstCharIsUpper = (*(_CTM + word.valueOriginal[0]) & CharType.IsUpper) == CharType.IsUpper;
Debug.Assert(wordFirstCharIsUpper == word.posTaggerFirstCharIsUpper, "(wordFirstCharIsUpper != word.posTaggerFirstCharIsUpper)");
#endif
var mode = GetWordFormMorphologyMode(word, i);
wordMorphology = _MorphoAnalyzer.GetWordMorphology_NoToUpper(word.valueUpper, mode);
//---!!!---DONT-FREE-USED-IN-FURTHER---// word.valueUpper = null; //free resource
}
#endregion
break;
case PosTaggerInputType.Num: // – содержит хотя бы одну цифру и не содержит букв;
#region
{
//---Debug.Assert( word.valueUpper == null, "word.valueUpper != null" );---//
if (word.posTaggerLastValueUpperInNumeralChain == null)
{
_WordMorphoAmbiguities.Add(_WordMorphoAmbiguityFactory.Create(word, i));
continue;
}
var mode = GetWordFormMorphologyMode(word, i);
wordMorphology = _MorphoAnalyzer.GetWordMorphology_4LastValueUpperInNumeralChain(
word.posTaggerLastValueUpperInNumeralChain, mode);
#region [.commented.]
/*
* fixed ( char* ptr = word.posTaggerLastValueOriginalInNumeralChain )
* {
* //-get-first-is-upper-from-valueOriginal-//
* //---var wordFirstCharIsUpper = (*(_CTM + *ptr) & CharType.IsUpper) == CharType.IsUpper;
* //---var wordFormMorphologyMode = GetWordFormMorphologyMode( word, wordFirstCharIsUpper, i );
* var mode = GetWordFormMorphologyMode( word, i );
* var length = word.posTaggerLastValueOriginalInNumeralChain.Length;
* wordMorphology = _MorphoAnalyzer.GetWordMorphology_4LastValueOriginalInNumeralChain( ptr, length, mode );
* }
*/
#endregion
//---!!!---DONT-FREE-USED-IN-FURTHER---// word.posTaggerLastValueOriginalInNumeralChain = null; //free resource
}
#endregion
break;
default:
#region [.who's skip.]
//CompPh – составные (имеющие хотя бы один пробел);
//Col – двоеточие.
//Com – запятая;
//Dush – тире;
//OneCP – первая заглавная с точкой;
#endregion
_WordMorphoAmbiguities.Add(_WordMorphoAmbiguityFactory.Create(word, i));
continue;
}
#endregion
#region [.post-process Morpho-Analyze-result.]
if (wordMorphology.HasWordFormMorphologies)
{
var wfms = default(WordFormMorphology_t[]);
//Если данное слово имеет только одну часть речи, прописанную в морфословаре, то использовать ее вместо определённой с помощью PoS-tagger.
if (wordMorphology.IsSinglePartOfSpeech)
{
CorrectPosTaggerOutputType(word, wordMorphology.PartOfSpeech);
wfms = wordMorphology.WordFormMorphologies.ToArray();
_WordMorphoAmbiguities.Add(_WordMorphoAmbiguityFactory.Create(word, i, wfms));
#if DEBUG
word.morphologies = wfms;
#endif
}
else
{
#region [.clause #1.]
//для данного слова в морфословаре определено несколько частей речи.
//ищем среди морфоинформации по слову морфоинформацию по части речи от pos-tagger'а, если она есть - берем её
//вот эта хуйня из-за двойной трансляции
//{PosTaggerOutputType::AdjectivePronoun => PartOfSpeechEnum::Adjective, PartOfSpeechEnum::Pronoun}
// &
//{PosTaggerOutputType::AdverbialPronoun => PartOfSpeechEnum::Adverb, PartOfSpeechEnum::Pronoun}
var partOfSpeech = default(PartOfSpeechEnum?);
switch (word.posTaggerOutputType)
{
case PosTaggerOutputType.AdjectivePronoun:
#region
{
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Adjective, PartOfSpeechEnum.Pronoun);
if (wfms != null)
{
_WordMorphoAmbiguities.Add(_WordMorphoAmbiguityFactory.Create(word, i, wfms));
#if DEBUG
word.morphologies = wfms;
#endif
continue;
}
}
#endregion
break;
case PosTaggerOutputType.AdverbialPronoun:
#region
{
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Adverb, PartOfSpeechEnum.Pronoun);
if (wfms != null)
{
_WordMorphoAmbiguities.Add(_WordMorphoAmbiguityFactory.Create(word, i, wfms));
#if DEBUG
word.morphologies = wfms;
#endif
continue;
}
}
#endregion
break;
default:
#region
{
partOfSpeech = ToPartOfSpeech(word.posTaggerOutputType);
if (partOfSpeech.HasValue)
{
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, partOfSpeech.Value);
if (wfms != null)
{
_WordMorphoAmbiguities.Add(_WordMorphoAmbiguityFactory.Create(word, i, wfms));
#if DEBUG
word.morphologies = wfms;
#endif
continue;
}
}
}
#endregion
break;
}
#endregion
#region [.clause #2.]
//При этом для данного слова в морфословаре определено несколько частей речи.
//В данном случае в первую очередь м.б. соответствия (слева выход PoS-tagger, справа морфословарь, последовательность пунктов неважна):
switch (word.posTaggerOutputType)
{
#region [.commented. previous.]
/*
* /*2.1. AdjectivePronoun = Adjective* /
* case PosTaggerOutputType.AdjectivePronoun:
* wfms = TryGetByPosTaggerOutputType( ref wordMorphology, PartOfSpeechEnum.Adjective );
#if DEBUG
* word.morphologies = wfms;
#endif
* break;
*
* /*2.2. AdverbialPronoun = Adverb* /
* case PosTaggerOutputType.AdverbialPronoun:
* wfms = TryGetByPosTaggerOutputType( ref wordMorphology, PartOfSpeechEnum.Adverb );
#if DEBUG
* word.morphologies = wfms;
#endif
* break;
*/
#endregion
/*2.3. Participle = Adjective */
case PosTaggerOutputType.Participle:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Adjective);
#if DEBUG
word.morphologies = wfms;
#endif
break;
default:
#region [.clause #3.]
{
if (partOfSpeech.HasValue)
{
switch (partOfSpeech.Value)
{
/*3.1. Pronoun = Noun*/
case PartOfSpeechEnum.Pronoun:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Noun);
#if DEBUG
word.morphologies = wfms;
#endif
break;
/*3.2. Noun = Pronoun */
case PartOfSpeechEnum.Noun:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Pronoun);
#if DEBUG
word.morphologies = wfms;
#endif
break;
/*3.3. Conjunction = Particle */
case PartOfSpeechEnum.Conjunction:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Particle);
#if DEBUG
word.morphologies = wfms;
#endif
break;
/*3.4. Particle = Conjunction*/
case PartOfSpeechEnum.Particle:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Conjunction);
#if DEBUG
word.morphologies = wfms;
#endif
break;
/*3.5. Numeral = Noun, Adjective */
case PartOfSpeechEnum.Numeral:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Noun, PartOfSpeechEnum.Adjective);
#if DEBUG
word.morphologies = wfms;
#endif
break;
/*3.6. Adjective = Verb, Adverb*/
case PartOfSpeechEnum.Adjective:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Verb, PartOfSpeechEnum.Adverb);
#if DEBUG
word.morphologies = wfms;
#endif
break;
/*3.7. Adverb = Adjective */
case PartOfSpeechEnum.Adverb:
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, PartOfSpeechEnum.Adjective);
#if DEBUG
word.morphologies = wfms;
#endif
break;
}
}
}
#endregion
break;
}
/*Если таковых соответствий не нашлось, то берется первая из выдачи морфословаря часть речи.*/
if (wfms == null)
{
var _partOfSpeech = wordMorphology.WordFormMorphologies[0].PartOfSpeech;
word.posTaggerOutputType = ToPosTaggerOutputType(_partOfSpeech);
wfms = TryGetByPosTaggerOutputType(ref wordMorphology, _partOfSpeech);
#if DEBUG
word.morphologies = wfms;
#endif
}
_WordMorphoAmbiguities.Add(_WordMorphoAmbiguityFactory.Create(word, i, wfms));
#endregion
}
}
else
{
_WordMorphoAmbiguities.Add(_WordMorphoAmbiguityFactory.Create(word, i));
}
#endregion
}
//-pre-process morpho-ambiguity-//
#if DEBUG
if (applyMorphoAmbiguityPreProcess)
{
#endif
_MorphoAmbiguityPreProcessor.Run(_WordMorphoAmbiguities);
#if DEBUG
}
#endif
//-resolve morpho-ambiguity-//
_MorphoAmbiguityResolver.Resolve(_WordMorphoAmbiguities);
//-clear-//
_WordMorphoAmbiguities.Clear();
}
