/**
* For a string of 1 or more surface words, return all of the lexical entries for each word as a
* list of sign hashes. Tokenization is performed using the configured tokenizer.
*
* @param w the words in string format
* @return a list of sign hashes
* @exception LexException thrown if word not found
*/
public List<SignHash> getEntriesFromWords(String s) throws LexException {
List<SignHash> entries = new ArrayList<SignHash>();
List<Word> words = tokenizer.tokenize(s);
for (Iterator<Word> it = words.iterator(); it.hasNext(); ) {
Word w = it.next();
SignHash signs = getSignsFromWord(w);
if (signs.size() == 0) {
throw new LexException("Word not in lexicon: \"" + w + "\"");
}
entries.add(signs);
}
return entries;
}
/**
* For a given word, return all of its surface word's lexical entries. If the word is not listed
* in the lexicon, the tokenizer is consulted to see if it is a special token (date, time, etc.);
* otherwise an exception is thrown. If the word has coarticulations, all applicable
* coarticulation entries are applied to the base word, in an arbitrary order.
*
* @param w the word
* @return a sign hash
* @exception LexException thrown if word not found
*/
public SignHash getSignsFromWord(Word w) throws LexException {
// reduce word to its core, removing coart attrs if any
Word surfaceWord = Word.createSurfaceWord(w);
Word coreWord =
(surfaceWord.attrsIntersect(_coartAttrs))
? Word.createCoreSurfaceWord(surfaceWord, _coartAttrs)
: surfaceWord;
// lookup core word
SignHash result = getSignsFromWord(coreWord, null, null, null);
if (result.size() == 0) {
throw new LexException(coreWord + " not found in lexicon");
}
// return signs if no coart attrs
if (coreWord == surfaceWord) return result;
// otherwise apply coarts for word
applyCoarts(surfaceWord, result);
return result;
}
