def get_words_with_tag(language, words, pos_tag, tagger, separator='_'):
"""
Returns the words that are tagged with the given tag.
>>> get_words_with_tag('en', ['cat', 'little'], 'NN', 'pos_tag_stanford')
[('cat', 'NN')]
>>> get_words_with_tag('fr', [(u'jouer_VB'), (u'aimer_VB'), \\
... (u'peut-être_RB')], 'VB', 'pos_tag_melt')
[(u'jouer', u'VB'), (u'aimer', u'VB')]
:language: language id, currently en or fr
:param words: list of words, optionally tagged <word>_<tag>
:param tag: POS-tag
:param tagger: method to use for tagging
:param separator: the character that separates the word and the tag
in the word list
:return: list of words of the given part of speech
"""
selected = []
if len(words) == 0:
return selected
else:
if words[0].find(separator) == -1:
if language == 'en':
tagged = tag_en.quick_pos_tag(words, tagger=tagger)
if language == 'fr':
tagged = tag_fr.quick_pos_tag(words, tagger=tagger)
if language == 'fi':
tagged = tag_fi.quick_pos_tag(words)
else:
tagged = helpers.strings_to_tuples(words, separator)
tagged = helpers.capitalize_nnp(tagged)
for token in tagged:
if len(token) > 1:
if token[1] == pos_tag:
selected.append(token)
return selected
def get_words_with_tag(language, words, pos_tag, tagger, separator='_'):
"""
Returns the words that are tagged with the given tag.
>>> get_words_with_tag('en', ['cat', 'little'], 'NN', 'pos_tag_stanford')
[('cat', 'NN')]
>>> get_words_with_tag('fr', [(u'jouer_VB'), (u'aimer_VB'), \\
... (u'peut-être_RB')], 'VB', 'pos_tag_melt')
[(u'jouer', u'VB'), (u'aimer', u'VB')]
:language: language id, currently en or fr
:param words: list of words, optionally tagged <word>_<tag>
:param tag: POS-tag
:param tagger: method to use for tagging
:param separator: the character that separates the word and the tag
in the word list
:return: list of words of the given part of speech
"""
selected = []
if len(words) == 0:
return selected
else:
if words[0].find(separator) == -1:
if language == 'en':
tagged = tag_en.quick_pos_tag(words, tagger=tagger)
if language == 'fr':
tagged = tag_fr.quick_pos_tag(words, tagger=tagger)
if language == 'fi':
tagged = tag_fi.quick_pos_tag(words)
else:
tagged = helpers.strings_to_tuples(words, separator)
tagged = helpers.capitalize_nnp(tagged)
for token in tagged:
if len(token) > 1:
if token[1] == pos_tag:
selected.append(token)
return selected
