yoxem/pseudo-chinese
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Fix some indent.

Browse source
This commit is contained in:
k2font 2020-04-14 20:18:21 +09:00
parent 23f4e3a045
commit dd1ca71957
1 changed files with 43 additions and 47 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

90
pseudo-chinese.py
View file

@ -10,19 +10,17 @@ BASE_URL = "https://api.ce-cotoha.com/api/dev/nlp/"
def auth(client_id, client_secret): def auth(client_id, client_secret):
token_url = "https://api.ce-cotoha.com/v1/oauth/accesstokens" token_url = "https://api.ce-cotoha.com/v1/oauth/accesstokens"
headers = { headers = {
"Content-Type": "application/json", "Content-Type": "application/json",
"charset": "UTF-8" "charset": "UTF-8"
} }
data = { data = {
"grantType": "client_credentials", "grantType": "client_credentials",
"clientId": client_id, "clientId": client_id,
"clientSecret": client_secret "clientSecret": client_secret
} }
r = requests.post(token_url, r = requests.post(token_url,headers=headers,data=json.dumps(data))
headers=headers,
data=json.dumps(data))
return r.json()["access_token"] return r.json()["access_token"]
@ -33,26 +31,24 @@ def parse(sentence, access_token):
base_url = BASE_URL base_url = BASE_URL
headers = { headers = {
"Content-Type": "application/json", "Content-Type": "application/json",
"charset": "UTF-8", "charset": "UTF-8",
"Authorization": "Bearer {}".format(access_token) "Authorization": "Bearer {}".format(access_token)
} }
data = { data = {
"sentence": sentence, "sentence": sentence,
"type": "default" "type": "default"
} }
r = requests.post(base_url + "v1/parse", r = requests.post(base_url + "v1/parse",headers=headers,data=json.dumps(data))
headers=headers,
data=json.dumps(data))
return r.json() return r.json()
# ひらがなを削除する関数 # ひらがなを削除する関数
# Function to delete hiragana. # Function to delete hiragana.
# 删除平假名的功能 # 删除平假名的功能
def hira_to_blank(str): def hira_to_blank(str):
return "".join(["" if ("ぁ" <= ch <= "ん") else ch for ch in str]) return "".join(["" if ("ぁ" <= ch <= "ん") else ch for ch in str])
if __name__ == "__main__": if __name__ == "__main__":
envjson = open('env.json', 'r') envjson = open('env.json', 'r')
@ -70,39 +66,39 @@ if __name__ == "__main__":
print(parse_document) print(parse_document)
result_list = list() result_list = list()
for chunks in parse_document['result']: for chunks in parse_document['result']:
for token in chunks["tokens"]: for token in chunks["tokens"]:
# 形態素解析結果に置き換えルールを適用する # 形態素解析結果に置き換えルールを適用する
if (token["pos"] != "連用助詞" if (token["pos"] != "連用助詞"
and token["pos"] != "引用助詞" and token["pos"] != "引用助詞"
and token["pos"] != "終助詞" and token["pos"] != "終助詞"
and token["pos"] != "接続接尾辞" and token["pos"] != "接続接尾辞"
and token["pos"] != "動詞活用語尾"): and token["pos"] != "動詞活用語尾"):
if token["pos"] == "動詞接尾辞" and '終止' in token["features"]: if token["pos"] == "動詞接尾辞" and '終止' in token["features"]:
if ("する" in token["lemma"]) or ("ます" in token["lemma"]): if ("する" in token["lemma"]) or ("ます" in token["lemma"]):
prime = "也" prime = "也"
elif "たい" in token["lemma"]: elif "たい" in token["lemma"]:
prime = "希望" prime = "希望"
elif token['lemma'] != 'ない': elif token['lemma'] != 'ない':
prime = "了" prime = "了"
else: else:
prime = "実行" prime = "実行"
else: else:
prime = token["form"] prime = token["form"]
if token['lemma'] == '私': if token['lemma'] == '私':
prime = '我' prime = '我'
if (token['lemma'] == '君' or token['lemma'] == 'あなた' or token['lemma'] == 'お前'): if (token['lemma'] == '君' or token['lemma'] == 'あなた' or token['lemma'] == 'お前'):
prime = '你' prime = '你'
if len(token["features"]) != 0: if len(token["features"]) != 0:
if "SURU" in token["features"][0] : if "SURU" in token["features"][0] :
prime = "実行" prime = "実行"
elif "連体" in token['features'][0]: elif "連体" in token['features'][0]:
prime = "的" prime = "的"
elif "疑問符" in token["features"][0]: elif "疑問符" in token["features"][0]:
prime = "如何?" prime = "如何?"
result_list.append(hira_to_blank(prime)) result_list.append(hira_to_blank(prime))
print(''.join(result_list)) print(''.join(result_list))