public MorphAutomat(Lemmatizer lemmatizer,InternalMorphLanguage language, char annotChar)
: base(lemmatizer,language, annotChar) {
}
private static int InitAlphabet(InternalMorphLanguage language, int[] pCode2Alphabet, int[] pAlphabet2Code, char annotChar) {
if (char.IsUpper(annotChar)) {
throw new MorphException("annotChar is not upper");
}
var AdditionalEnglishChars = "'1234567890";
var AdditionalGermanChars = "";
var AlphabetSize = 0;
for (var i = 0; i < Constants.AlphabetSize; i++) {
var ch = Convert.ToChar(i);
if (Lang.is_upper_alpha((byte)i, language)
|| (ch == '-')
|| (ch == annotChar)
|| ((language == InternalMorphLanguage.morphEnglish)
&& (AdditionalEnglishChars.IndexOf(ch) >= 0)
)
|| ((language == InternalMorphLanguage.morphGerman)
&& (AdditionalGermanChars.IndexOf(ch) >= 0)
)
|| ((language == InternalMorphLanguage.morphURL)
&& Lang.is_alpha((byte)i, language)
)) {
pCode2Alphabet[AlphabetSize] = i;
pAlphabet2Code[i] = AlphabetSize;
AlphabetSize++;
} else {
pAlphabet2Code[i] = -1;
}
}
if (AlphabetSize > Constants.MaxAlphabetSize) {
throw new MorphException("Error! The ABC is too large");
}
return AlphabetSize;
}
public byte TransferReverseVowelNoToCharNo(string form, byte accentCharNo, InternalMorphLanguage language, int codePage)
{
if (accentCharNo == Constants.UnknownAccent)
{
return(Constants.UnknownAccent);
}
if (accentCharNo >= form.Length)
{
throw new MorphException("AccentCharNo >= form.Length");
}
var countOfVowels = -1;
var i = form.Length - 1;
if (i >= Constants.UnknownAccent)
{
throw new MorphException("i >= UnknownAccent");
}
for (; i >= 0; i--)
{
if (Lang.is_lower_vowel(GetByte(form[i], codePage), language) ||
Lang.is_upper_vowel(GetByte(form[i], codePage), language)
)
{
countOfVowels++;
}
if (countOfVowels == accentCharNo)
{
return((byte)i);
}
}
return(Constants.UnknownAccent);
}
internal static bool is_upper_consonant(byte x, InternalMorphLanguage Langua)
{
if (!is_upper_alpha(x, Langua))
{
return(false);
}
return(!is_upper_vowel(x, Langua));
}
public Lemmatizer(InternalMorphLanguage language)
: base(language) {
_loaded = false;
_useStatistic = false;
_maximalPrediction = false;
_allowRussianJo = false;
_predict = new PredictBase(this,language);
InitAutomat(new MorphAutomat(this,language, Constants.MorphAnnotChar));
}
public Lemmatizer(InternalMorphLanguage language)
: base(language)
{
_loaded = false;
_useStatistic = false;
_maximalPrediction = false;
_allowRussianJo = false;
_predict = new PredictBase(this, language);
InitAutomat(new MorphAutomat(this, language, Constants.MorphAnnotChar));
}
public ABCEncoder(Lemmatizer lemmatizer,InternalMorphLanguage language, char annotChar) {
_lemmatizer = lemmatizer;
_language = language;
_annotChar=annotChar;
_alphabetSize = InitAlphabet(language, _code2Alphabet, _alphabet2Code, _annotChar);
_alphabetSizeWithoutAnnotator = InitAlphabet(language,_code2AlphabetWithoutAnnotator,_alphabet2CodeWithoutAnnotator,(char)257/* non-exeting symbol */);
if (_alphabetSizeWithoutAnnotator + 1 != _alphabetSize) {
throw new MorphException("_alphabetSizeWithoutAnnotator + 1 != _alphabetSize");
}
}
internal static bool is_lower_vowel(byte x, InternalMorphLanguage Langua)
{
switch (Langua)
{
case InternalMorphLanguage.morphRussian: return(is_russian_lower_vowel(x));
case InternalMorphLanguage.morphEnglish: return(is_english_lower_vowel(x));
case InternalMorphLanguage.morphGerman: return(is_german_lower_vowel(x));
}
return(false);
}
public ABCEncoder(Lemmatizer lemmatizer, InternalMorphLanguage language, char annotChar)
{
_lemmatizer = lemmatizer;
_language = language;
_annotChar = annotChar;
_alphabetSize = InitAlphabet(language, _code2Alphabet, _alphabet2Code, _annotChar);
_alphabetSizeWithoutAnnotator = InitAlphabet(language, _code2AlphabetWithoutAnnotator, _alphabet2CodeWithoutAnnotator, (char)257 /* non-exeting symbol */);
if (_alphabetSizeWithoutAnnotator + 1 != _alphabetSize)
{
throw new MorphException("_alphabetSizeWithoutAnnotator + 1 != _alphabetSize");
}
}
internal static bool is_upper_alpha(byte x, InternalMorphLanguage Langua)
{
switch (Langua)
{
case InternalMorphLanguage.morphRussian: return(is_russian_upper(x));
case InternalMorphLanguage.morphEnglish: return(is_english_upper(x));
case InternalMorphLanguage.morphGerman: return(is_german_upper(x));
case InternalMorphLanguage.morphGeneric: return(is_generic_upper(x));
case InternalMorphLanguage.morphURL: return(false);
}
return(false);
}
internal static bool is_alpha(byte x, InternalMorphLanguage Langua)
{
switch (Langua)
{
case InternalMorphLanguage.morphRussian: return(is_russian_alpha(x));
case InternalMorphLanguage.morphEnglish: return(is_english_alpha(x));
case InternalMorphLanguage.morphGerman: return(is_german_alpha(x));
case InternalMorphLanguage.morphGeneric: return(is_generic_alpha(x));
case InternalMorphLanguage.morphURL: return(is_URL_alpha(x));
}
throw new MorphException("unknown char x");
}
public static string GetStringByLanguage(InternalMorphLanguage langua)
{
switch (langua)
{
case InternalMorphLanguage.morphRussian: return("Russian");
case InternalMorphLanguage.morphEnglish: return("English");
case InternalMorphLanguage.morphGerman: return("German");
case InternalMorphLanguage.morphGeneric: return("Generic");
case InternalMorphLanguage.morphURL: return("URL_ABC");
default: return("unk");
}
}
private static int InitAlphabet(InternalMorphLanguage language, int[] pCode2Alphabet, int[] pAlphabet2Code, char annotChar)
{
if (char.IsUpper(annotChar))
{
throw new MorphException("annotChar is not upper");
}
var AdditionalEnglishChars = "'1234567890";
var AdditionalGermanChars = "";
var AlphabetSize = 0;
for (var i = 0; i < Constants.AlphabetSize; i++)
{
var ch = Convert.ToChar(i);
if (Lang.is_upper_alpha((byte)i, language) ||
(ch == '-') ||
(ch == annotChar) ||
((language == InternalMorphLanguage.morphEnglish) &&
(AdditionalEnglishChars.IndexOf(ch) >= 0)
) ||
((language == InternalMorphLanguage.morphGerman) &&
(AdditionalGermanChars.IndexOf(ch) >= 0)
) ||
((language == InternalMorphLanguage.morphURL) &&
Lang.is_alpha((byte)i, language)
))
{
pCode2Alphabet[AlphabetSize] = i;
pAlphabet2Code[i] = AlphabetSize;
AlphabetSize++;
}
else
{
pAlphabet2Code[i] = -1;
}
}
if (AlphabetSize > Constants.MaxAlphabetSize)
{
throw new MorphException("Error! The ABC is too large");
}
return(AlphabetSize);
}
internal static bool is_upper_alpha(byte x, InternalMorphLanguage Langua) {
switch (Langua) {
case InternalMorphLanguage.morphRussian: return is_russian_upper(x);
case InternalMorphLanguage.morphEnglish: return is_english_upper(x);
case InternalMorphLanguage.morphGerman: return is_german_upper(x);
case InternalMorphLanguage.morphGeneric: return is_generic_upper(x);
case InternalMorphLanguage.morphURL: return false;
}
return false;
}
protected MorphDict(InternalMorphLanguage language)
{
_formAutomat = null;
_comparer = new InternalComparer(_bases);
}
public PredictBase(Lemmatizer lemmatizer, InternalMorphLanguage lang)
{
_suffixAutomat = new MorphAutomat(lemmatizer, lang, Constants.MorphAnnotChar);
}
public MorphAutomat(Lemmatizer lemmatizer, InternalMorphLanguage language, char annotChar)
: base(lemmatizer, language, annotChar)
{
_tools = new Tools();
}
public PredictBase(Lemmatizer lemmatizer,InternalMorphLanguage lang) {
_suffixAutomat = new MorphAutomat(lemmatizer,lang, Constants.MorphAnnotChar);
}
internal static bool is_alpha(byte x, InternalMorphLanguage Langua) {
switch (Langua) {
case InternalMorphLanguage.morphRussian: return is_russian_alpha(x);
case InternalMorphLanguage.morphEnglish: return is_english_alpha(x);
case InternalMorphLanguage.morphGerman: return is_german_alpha(x);
case InternalMorphLanguage.morphGeneric: return is_generic_alpha(x);
case InternalMorphLanguage.morphURL: return is_URL_alpha(x);
}
throw new MorphException("unknown char x");
}
internal static bool is_lower_vowel(byte x, InternalMorphLanguage Langua) {
switch (Langua) {
case InternalMorphLanguage.morphRussian: return is_russian_lower_vowel(x);
case InternalMorphLanguage.morphEnglish: return is_english_lower_vowel(x);
case InternalMorphLanguage.morphGerman: return is_german_lower_vowel(x);
}
return false;
}
internal static bool is_upper_consonant(byte x, InternalMorphLanguage Langua) {
if (!is_upper_alpha(x, Langua)) return false;
return !is_upper_vowel(x, Langua);
}
