def write_sa_verb_and_pattern_of_particle_and_morphs_oneline(i, oneline):
map_sa_verb_and_pattern_of_particle_and_morphs = {}
for i_d, dst in enumerate(oneline):
if not NLP43.is_verb(dst.morphs()[0]):
continue
for i_s in dst.srcs():
chunk = oneline[i_s].morphs()
if len(chunk) < 2:
continue
last_morph_of_src = chunk[-1]
second_last_morph_of_src = chunk[-2]
if not (is_particle_wo(last_morph_of_src) and is_noun_connected_sa(second_last_morph_of_src)):
continue
key_base = NLP42.concat_morphs(chunk) + dst.morphs()[0].base()
for i_src_of_s in dst.srcs():
if i_src_of_s == i_s:
continue
c = oneline[i_src_of_s].morphs()
c = omit_punctual_from_morphs(c)
if len(c) < 1:
continue
last_morph_of_i_src_of_s = c[-1]
if NLP45.is_particle(last_morph_of_i_src_of_s):
if key_base in map_sa_verb_and_pattern_of_particle_and_morphs:
d = map_sa_verb_and_pattern_of_particle_and_morphs[key_base]
d['particles'].append(copy.deepcopy(last_morph_of_i_src_of_s.surface()))
d['morphs'].append(copy.deepcopy(c))
else:
new_dict = dict(particles = copy.deepcopy([last_morph_of_i_src_of_s.surface()]), morphs = copy.deepcopy([c]))
map_sa_verb_and_pattern_of_particle_and_morphs[key_base] = new_dict
write_verb_and_pattern_of_particle_and_morphs_result(i, map_sa_verb_and_pattern_of_particle_and_morphs)
def replace_noun_morph_from_chunk_with(chunk, what):
temp_morphs = chunk.morphs()
new_morphs = list(map(lambda x: Morph(what, what, x.pos(), x.pos1()) if NLP43.is_noun(x) else copy.deepcopy(x), temp_morphs))
for m in chunk.morphs():
pass
new_chunk = Chunk(new_morphs, chunk.dst())
for s in chunk.srcs():
new_chunk.append_srcs(s)
return new_chunk
def write_verb_and_pattern_of_particle_oneline(i, oneline):
map_verb_and_pattern_of_particle = {}
for i_d, dst in enumerate(oneline):
if not NLP43.is_verb(dst.morphs()[0]):
continue
particles = []
for i_s in dst.srcs():
last_morph_of_src = oneline[i_s].morphs()[-1]
if is_particle(last_morph_of_src):
particles.append(last_morph_of_src.surface())
key = dst.morphs()[0].base()
if key in map_verb_and_pattern_of_particle:
map_verb_and_pattern_of_particle[key] += copy.deepcopy(particles)
else:
map_verb_and_pattern_of_particle[key] = copy.deepcopy(particles)
write_verb_and_pattern_of_particle_result(i, map_verb_and_pattern_of_particle)
def write_verb_and_pattern_of_particle_and_morphs_oneline(i, oneline):
map_verb_and_pattern_of_particle_and_morphs = {}
for i_d, dst in enumerate(oneline):
if not NLP43.is_verb(dst.morphs()[0]):
continue
particles = []
morphs = []
for i_s in dst.srcs():
last_morph_of_src = oneline[i_s].morphs()[-1]
if NLP45.is_particle(last_morph_of_src):
particles.append(last_morph_of_src.surface())
morphs.append(oneline[i_s].morphs())
key_base = dst.morphs()[0].base()
if key_base in map_verb_and_pattern_of_particle_and_morphs:
d = map_verb_and_pattern_of_particle_and_morphs[key_base]
d['particles'] += copy.deepcopy(particles)
d['morphs'] += copy.deepcopy(morphs)
else:
new_dict = dict(particles = copy.deepcopy(particles), morphs = copy.deepcopy(morphs))
map_verb_and_pattern_of_particle_and_morphs[key_base] = new_dict
write_verb_and_pattern_of_particle_and_morphs_result(i, map_verb_and_pattern_of_particle_and_morphs)
def get_noun_index(i, chunks):
noun_index = []
for index, chunk in enumerate(chunks):
if NLP43.has_chunk_noun(chunk) and chunk.dst() != -1:
noun_index.append(index)
return noun_index
def is_noun_connected_sa(m):
return NLP43.is_noun(m) and m.pos1().startswith('サ変接続')
