def _combined_analyses(self, word_dediac, prefix_analyses, stem_analyses,
suffix_analyses):
combined = deque()
for p in itertools.product(prefix_analyses, stem_analyses):
prefix_cat = p[0][0]
prefix_feats = p[0][1]
stem_cat = p[1][0]
stem_feats = p[1][1]
if stem_cat in self._db.prefix_stem_compat[prefix_cat]:
for suffix_cat, suffix_feats in suffix_analyses:
if ((stem_cat not in self._db.stem_suffix_compat) or
(prefix_cat not in self._db.prefix_suffix_compat)
or (suffix_cat
not in self._db.stem_suffix_compat[stem_cat])
or
(suffix_cat
not in self._db.prefix_suffix_compat[prefix_cat])):
continue
merged = merge_features(self._db, prefix_feats, stem_feats,
suffix_feats)
merged['stem'] = stem_feats['diac']
merged['stemcat'] = stem_cat
merged_dediac = dediac_ar(merged['diac'])
if word_dediac.replace(u'\u0640', '') != merged_dediac:
merged['source'] = 'spvar'
combined.append(merged)
return combined
def _combined_backoff_analyses(self,
stem,
word_dediac,
prefix_analyses,
stem_analyses,
suffix_analyses):
combined = deque()
for p in itertools.product(prefix_analyses, stem_analyses):
prefix_cat = p[0][0]
prefix_feats = p[0][1]
stem_cat = p[1][0]
stem_feats = copy.copy(p[1][1])
if stem_cat in self._db.prefix_stem_compat[prefix_cat]:
for suffix_cat, suffix_feats in suffix_analyses:
if ((suffix_cat not in
self._db.stem_suffix_compat[stem_cat]) or
(prefix_cat not in self._db.prefix_suffix_compat or
suffix_cat not in
self._db.prefix_suffix_compat[prefix_cat])):
continue
if (self._backoff_action == 'PROP' and
'NOUN_PROP' not in stem_feats['bw']):
continue
stem_feats['bw'] = _NOAN_RE.sub(stem, stem_feats['bw'])
stem_feats['diac'] = _NOAN_RE.sub(stem, stem_feats['diac'])
stem_feats['lex'] = _NOAN_RE.sub(stem, stem_feats['lex'])
stem_feats['caphi'] = simple_ar_to_caphi(stem)
merged = merge_features(self._db, prefix_feats, stem_feats,
suffix_feats)
merged['stem'] = stem_feats['diac']
merged['stemcat'] = stem_cat
merged['source'] = 'backoff'
merged['gloss'] = stem_feats['gloss']
combined.append(merged)
return combined
def generate(self, lemma, feats):
"""Generate surface forms and their associated analyses for a given
lemma and a given set of (possibly underspecified) features.
The surface form is accessed through the `diac` feature.
Args:
lemma (:obj:`str`): Lemma to generate from.
feats (:obj:`dict`): Dictionary of features. Must contain 'pos'
feature.
See :doc:`/reference/camel_morphology_features` for
more information on features and their values.
Returns:
:obj:`list` of :obj:`dict`: List of generated analyses.
See :doc:`/reference/camel_morphology_features` for more
information on features and their values.
Raises:
:obj:`~camel_tools.morphology.errors.InvalidGeneratorFeature`: If
a feature is given that is not defined in database.
:obj:`~camel_tools.morphology.errors.InvalidGeneratorFeatureValue`:
If an invalid value is given to a feature or if 'pos' feature
is not defined.
"""
lemma = strip_lex(lemma)
if lemma not in self._db.lemma_hash:
return []
for feat in feats:
if feat not in self._db.defines:
raise InvalidGeneratorFeature(feat)
elif (self._db.defines[feat] is not None and
feats[feat] not in self._db.defines[feat]):
raise InvalidGeneratorFeatureValue(feat, feats[feat])
if 'pos' not in feats or feats['pos'] not in self._db.defines['pos']:
raise InvalidGeneratorFeatureValue('pos', feats.get('pos', None))
feats = copy.copy(feats)
default = self._db.defaults[feats['pos']]
default_feat_set = frozenset(default.keys())
feat_set = frozenset(feats.keys())
if not feat_set.issubset(default_feat_set):
return []
# Set default values for undefined feats
for feat in ['prc0', 'prc1', 'prc2', 'prc3', 'enc0', 'enc1', 'enc2']:
if feat not in feats and feat in default:
feats[feat] = default[feat]
stem_feats_list = self._db.lemma_hash[lemma]
analyses = collections.deque()
for stem_feats in stem_feats_list:
if 'vox' in feats and stem_feats['vox'] != feats['vox']:
continue
if 'rat' in feats and stem_feats['rat'] != feats['rat']:
continue
if 'pos' in feats and stem_feats['pos'] != feats['pos']:
continue
ignore_stem = False
for feat in ['prc0', 'prc1', 'prc2', 'prc3', 'enc0', 'enc1', 'enc2']:
if feat not in feats:
continue
if (feat in stem_feats and
stem_feats[feat] != '0' and
feats[feat] != stem_feats[feat]):
ignore_stem = True
break
if ignore_stem:
continue
prefix_cats = self._db.stem_prefix_compat[stem_feats['stemcat']]
suffix_cats = self._db.stem_suffix_compat[stem_feats['stemcat']]
for prefix_cat in prefix_cats:
if prefix_cat not in self._db.prefix_cat_hash:
continue
prefix_feats_list = self._db.prefix_cat_hash[prefix_cat]
for prefix_feats in prefix_feats_list:
ignore_prefix = False
for feat in ['prc0', 'prc1', 'prc2', 'prc3']:
if feat not in feats:
continue
if ((feats[feat] != '0' and
feat not in prefix_feats and
stem_feats.get(feat, '0') != feats[feat]) or
(feat in prefix_feats and
feats[feat] != prefix_feats[feat])):
ignore_prefix = True
break
if ignore_prefix:
continue
for suffix_cat in suffix_cats:
if suffix_cat not in self._db.suffix_cat_hash:
continue
suffix_feats_list = (
self._db.suffix_cat_hash[suffix_cat])
for suffix_feats in suffix_feats_list:
if ((prefix_cat not in
self._db.prefix_suffix_compat) or
(suffix_cat not in
self._db.prefix_suffix_compat[prefix_cat])):
continue
ignore_suffix = False
for feat in ['enc0', 'enc1', 'enc2']:
if feat not in feats:
continue
if ((feats[feat] != '0' and
feat not in suffix_feats and
stem_feats.get(feat, '0') != feats[feat])
or (feat in suffix_feats and
feats[feat] != suffix_feats[feat])):
ignore_suffix = True
break
if ignore_suffix:
continue
merged = merge_features(self._db, prefix_feats,
stem_feats, suffix_feats)
ignore_analysis = False
for feat in feats.keys():
if (feat in merged and
merged[feat] != feats[feat]):
ignore_analysis = True
break
if not ignore_analysis:
analyses.append(merged)
return list(analyses)
