void EnchantSpellingAlternativesPrivate::setEnchantSpellingAlternatives(
Token* token,
MorphoSyntacticData* tokenData,
FsaStringsPool& sp)
{
// try to find simple Uncapitalization
MORPHOLOGINIT;
// FIXME Conditions below could be process unit parameters
const LimaString& tokenStr=token->stringForm();
if (token->status().getAlphaCapital() == T_CAPITAL
|| token->status().getAlphaCapital() == T_CAPITAL_1ST
|| token->status().getAlphaCapital() == T_CAPITAL_SMALL
|| token->status().isAlphaConcatAbbrev()
|| token->status().isAlphaHyphen()
|| token->status().isAlphaPossessive()
|| tokenStr.toUpper() == tokenStr)
{
return;
}
std::vector<std::string> suggestions = m_enchantDictionary->suggest(tokenStr.toUtf8().constData());
for (std::vector<std::string>::const_iterator it = suggestions.begin(); it != suggestions.end();it++)
{
LimaString correction = LimaString::fromUtf8((*it).c_str());
// FIXME Conditions below could be process unit parameters
if ( correction.size() > 1 && correction != tokenStr )
{
DictionaryEntry* entry = new DictionaryEntry(m_dictionary->getEntry(correction));
MorphoSyntacticDataHandler lingInfosHandler(*tokenData, SPELLING_ALTERNATIVE);
if (!entry->isEmpty())
{
LINFO << "EnchantSpellingAlternativesPrivate::setEnchantSpellingAlternatives correcting" << tokenStr << "into" << correction;
// add orthographic alternative to Token;
StringsPoolIndex idx=sp[correction];
token->addOrthographicAlternatives(idx);
if (entry->hasLingInfos())
{
entry->parseLingInfos(&lingInfosHandler);
}
}
else
{
delete entry;
}
}
}
}
void OrthographicAlternatives::createAlternative(
Token* srcToken,
MorphoSyntacticData* tokenData,
LimaString& str,
AnalysisDict::AbstractAnalysisDictionary* dictionary,
StringsPool& sp)
{
MORPHOLOGINIT;
LDEBUG << "OrthographicAlternatives::createAlternative" << str;
DictionaryEntry* dicoEntry = new DictionaryEntry(dictionary->getEntry(str));
if (!dicoEntry->isEmpty())
{
// add orthographic alternative to Token;
StringsPoolIndex infl=sp[str];
Token* altToken=new Token(infl,str,srcToken->position(),srcToken->length(),new TStatus(*(srcToken->status())));
altToken->setDictionaryEntry(dicoEntry);
srcToken->addOrthographicAlternative(altToken);
tokenData->appendLingInfo(infl,dicoEntry,ORTHOGRAPHIC_ALTERNATIVE,sp);
// if entry has other accented forms,
// keep them ("PARIS" -> "paris" -> "Paris")
if (dicoEntry->hasAccented())
{
dicoEntry->reset();
Lima::LimaString alternativeStr = dicoEntry->nextAccented();
while (alternativeStr.size() != 0)
{
// give it its simple word entry into dictionary
DictionaryEntry* altDicoEntry = new DictionaryEntry(dictionary->getEntry(alternativeStr));
StringsPoolIndex infl2=sp[alternativeStr];
tokenData->appendLingInfo(infl2,altDicoEntry,ORTHOGRAPHIC_ALTERNATIVE,sp);
// add orthographic alternative to Token
Token* altToken2=new Token(infl2,alternativeStr,srcToken->position(),srcToken->length(),new TStatus(*(srcToken->status())));
altToken2->setDictionaryEntry(altDicoEntry);
srcToken->addOrthographicAlternative(altToken2);
alternativeStr = dicoEntry->nextAccented();
}
}
} else {
delete dicoEntry;
}
}