def __init__(self, fnname, params_str, body_str):
assert isinstance(params_str, str)
self._fnname = fnname
self._body_str = body_str
if DEBUGGING:
for i in range(len(body_str)):
if body_str[i] == "`":
WarningMsg(
"Looks like you've used a backquote in a macro definition. As of now, this probably means your code will only run in a js interpreter that supports template strings (es6)."
)
possibly_typed_params_list_chunks = split_by_top_level_commas(
Chunk.fromSingleLine(params_str))
possibly_typed_params_list = list(
map(lambda x: x.asSingleLine(), possibly_typed_params_list_chunks))
self._numRestParams = 1 if len(
possibly_typed_params_list
) > 0 and "..." in possibly_typed_params_list[-1] else 0
self._numOptionalParams = sum(
map(lambda x: 1 if ("?" in x) else 0, possibly_typed_params_list))
self._numParams = len(possibly_typed_params_list)
self._numRegParams = self._numParams - self._numOptionalParams - self._numRestParams
self._params = list(
map(MacroDefn._parse_typed_param, possibly_typed_params_list))
self._param2ind = dict()
for i in range(len(self._params)):
self._param2ind[self._params[i]] = i
self._params_re = re.compile("|".join(self._params))
self._param_occurrences = self._findParamOccurrences()
if len(self._param_occurrences) == 0:
self._no_subst = True
self._fillTemplate()
line_info = line.split()
idx = int(line_info[1])
dst = int(re.search(r'(.*?)D', line_info[2]).group(1))
if (dst in srcs_dict) and (dst != -1):
srcs_dict[dst].append(idx)
else:
srcs_dict[dst] = [idx]
if idx != 0:
chunks.append(chunk)
else:
chunks = []
chunk = Chunk(idx, dst)
elif line == "EOS":
chunks.append(chunk)
for k, v in srcs_dict.items():
chunks[k].update_srsc(v)
sents.append(chunks)
srcs_dict.clear()
else:
morph = Morph(line)
chunk.update_morph(morph)
for sent in sents:
for m in sent:
srcs_list = []
sentence_list = []
sentence_list_temp = [] # リセット
morph_list = []
next
chunk_len = len(dst_list)
for i in range(chunk_len):
if str(i) in dst_list:
srcs_list.append([
str(s) for s, x in enumerate(dst_list) if x == str(i)
])
else:
srcs_list.append(["-1"])
sentence = "".join(sentence_list)
if len(sentence) > 0:
all_sent.append(
Chunk(sentence_list, morph_list, dst_list, srcs_list))
one_sent = []
dst_list = []
one_morph = []
one_chank = []
srcs_list = []
sentence_list = []
sentence_list_temp = []
morph_list = []
with open("data/all_sent.pkl", "wb") as f:
pickle.dump(all_sent, f)