def save_to_csv(self, filename, data):
"""Save this data on csv file by prefecture"""
row = []
address = zenhan.z2h(data['address'], zenhan.ALL)
# remove the zip code
address = re.sub(r'%s\d+-\d+' % u'〒', '', address).strip()
row.append(data['name'])
row.append(data['name_kata'])
row.append(address)
row.append(u'\n'.join(data['routes'] or u''))
row.append(data['prefecture'])
row.append(data['area'])
row.append(zenhan.z2h(data['phone'], zenhan.ALL))
row.append(data['working_hours'])
row.append(data['holydays'])
row.append(data['shop_url'])
row.append(data['credit_cards_comment'])
row.append(u'・'.join(data['credit_cards'] or u''))
row.append(data['seats'])
row.append(data['stylist'])
row.append(data['parking'])
row.append(unicode(data['cut_price']))
row.append(data['page_url'])
CsvWriter.write_to_csv(filename, row, firs_row=self.first_row)
def normalize(self,text):
#アルファベット：全角=>半角
text = zenhan.z2h(text,mode=1)
#数字：全角=>半角
text = zenhan.z2h(text,mode=2)
#カタカナ：半角=>全角
text = zenhan.h2z(text,mode=4)
return text
def delete_aft(line):
text = zenhan.z2h(line, mode=1) #アルファベット（全角→半角）
text = zenhan.z2h(text, mode=2) #数字（全角→半角）
text = zenhan.h2z(text, mode=4) #カタカナ（半角→全角）
text = re.sub(
r'[\u0000-\u002F\u003A-\u0040\u005B-\u0060\u007B-\u3004\u3007-\u303F\u3099-\u30A0\u30FB\u30FD-\u4DFF\uA000-\uE0FFF]',
"", text) #その他文字列削除
return text
def delete_symbol(line):
text = zenhan.z2h(line, mode=1) #アルファベット（全角→半角）
text = zenhan.z2h(text, mode=2) #数字（全角→半角）
text = zenhan.h2z(text, mode=4) #カタカナ（半角→全角）
symbol = re.sub(r'[\u0000-\uE0FFF]', "", text) #unicode非対応の文字
text = re.sub(
r'[\u0000-\u002F\u003A-\u0040\u005B-\u0060\u007B-\u3004\u3007-\u303F\u3099-\u30A0\u30FB\u30FD-\u4DFF\uA000-\uE0FFF]',
"", text) #その他文字列削除
"""unicode非対応の文字の削除"""
if not symbol == "":
text = re.sub("[%s]" % symbol, "", text)
return text
def conversion_data_format(input_data_dict):
"""変換処理を行いリストに格納し、それを返す処理。
"""
name = input_data_dict['name'].strip()
gender = GENDER_MODIFIED_MAP[input_data_dict['gender']]
birthday = input_data_dict['birthday']
email = input_data_dict['email']
tel = zenhan.z2h(input_data_dict['tel']).replace('ー', '-')
zip_code = input_data_dict['post_code'].replace('ー', '')
address = input_data_dict['address']
item_sample = SAMPLE_PRODUCT_MAP[input_data_dict['item_num']]
reception_datetime = parser.parse(input_data_dict['reception_date'])
str_reception_datetime = reception_datetime.strftime('%Y/%m/%d %H:%M:%S')
conversion_format = [
name, gender, birthday, email, tel, zip_code, address, item_sample,
str_reception_datetime
]
return conversion_format
def normalize(ingredient):
ingredient = ingredient.strip()
for SURROUND in SURROUNDS:
ingredient = SURROUND.sub(lambda s: '', ingredient)
ingredient = OPTIONAL_START.sub(lambda s: '', ingredient)
match = UNCLOSED_PAREN.match(ingredient)
if match:
ingredient = match.groups()[0]
ingredient = zenhan.z2h(ingredient, mode=1) # ascii
ingredient = zenhan.h2z(ingredient, mode=4) # kana
# convert all katakana to hiragana
ingredient = hiragana(ingredient)
match = STARTS_WITH_ALPHA.match(ingredient)
if match and not ingredient.startswith('S&B'):
ingredient = match.groups()[0]
for SPECIAL_SYMBOL in SPECIAL_SYMBOLS:
ingredient = SPECIAL_SYMBOL.sub(lambda s: '', ingredient)
ingredients = SPLIT.split(ingredient)
ingredients = map(lambda ingr: ENDS_WITH.sub(lambda s: '', ingr), ingredients)
ingredients = map(lambda ingr: ingr.strip(), ingredients)
ingredients = filter(lambda ingr: ingr, ingredients)
for ingredient in ingredients:
yield ingredient
def load(self, f):
for wseq in super(WikiEdaTree, self).load(f):
eposlist = []
for i, word in enumerate(wseq.word_list):
if "misc" in word and len(word["misc"]) > 0:
if len(word["misc"]) < 2: # space B/I
raise FormattingException("malformed annotation: %s" % "".join(word["misc"]))
word["_wpadding"] = word["misc"].pop(0)
stype = word["misc"].pop(0)
if stype not in ("B", "I"):
raise FormattingException("malformed annotation: %s" % "".join(word["misc"]))
if stype == "B":
word["stype"] = self.WIKI_B
word["misc"].pop(0)
word["entity"] = "".join(word["misc"])
del word["misc"]
eposlist.append(i)
else:
if i <= 0 \
or "stype" not in wseq.word_list[i - 1] \
or wseq.word_list[i - 1]["stype"] not in (self.WIKI_B, self.WIKI_I):
raise FormattingException("malformed annotation: I-without-B: %s" % word["wid"])
word["stype"] = self.WIKI_I
self.WIKI_I
else:
word["stype"] = self.WIKI_O
for epos in eposlist:
mention_orig = wseq.word_list[epos]["surface"]
for i in xrange(epos + 1, len(wseq.word_list)):
if wseq.word_list[i]["stype"] == self.WIKI_I:
mention_orig += wseq.word_list[i]["surface"]
else:
break
wseq.word_list[epos]["mention"] = z2h(mention_orig, mode=3)
yield wseq
def clean_text(text):
# del_n = re.compile('\n')
# text = del_n.sub('',text)
text = text.lower()
text = unicodedata.normalize('NFKC', text)
text = zenhan.z2h(text, zenhan.ASCII | zenhan.DIGIT)
return text
def zenNum2hanNum(strings):
"""
全角数字を半角数字に変換する
その他の文字はそのまま
"""
strings = MultiBytes.convert2unicode(strings)
return zenhan.z2h(strings, mode=2)
def clean_text(text):
# del_n = re.compile('\n')
# text = del_n.sub('',text)
text = text.lower()
text = unicodedata.normalize('NFKC', text)
text = zenhan.z2h(text,zenhan.ASCII|zenhan.DIGIT)
return text
def setEffect(string, material):
effect = p.sub("", string)
effect = effect.replace('－', '-')
effect = zenhan.z2h(effect, 3)
effect = effect.replace(material, "")
effect = effect.replace("。", "。\n")
return effect
def normalize(word):
word = zenhan.z2h(word.lower(), zenhan.ASCII).strip()
for t_word in _trim_words:
if word.startswith(t_word):
return word.lstrip(t_word)
else:
return word
def test_zenhan():
logging.info("=========================================")
logging.info("= zenhan =")
logging.info("=========================================")
test_cases = get_test_cases()
for tc in test_cases:
title = tc['title']
body = tc['body']
logging.info("ひらがな（全角） to カタカナ（全角） for %s" % title)
logging.info("Not implemented")
logging.info("カタカナ（全角） to ひらがな（全角） for %s" % title)
logging.info("Not implemented")
logging.info("ひらがな（全角） to カタカナ（半角） for %s" % title)
logging.info("Not implemented")
logging.info("半角 to 全角 for %s" % title)
calc_time(zenhan.h2z, body, zenhan.ASCII|zenhan.KANA|zenhan.DIGIT)
logging.debug("result: %s" % zenhan.h2z(body, zenhan.ASCII|zenhan.KANA|zenhan.DIGIT))
logging.info("全角 to 半角 for %s" % title)
calc_time(zenhan.z2h, body, zenhan.ASCII|zenhan.KANA|zenhan.DIGIT)
logging.debug("result: %s" % zenhan.z2h(body, zenhan.ASCII|zenhan.KANA|zenhan.DIGIT))
def setMaterial(string):
string = zenhan.z2h(string.replace('－', '-'), 3)
if '<br' in string:
string = p.sub("", string[:string.index('<br')])
else:
string = p.sub("", string)
return string
def delete_twitter(line):
text = zenhan.z2h(line, mode=1) #アルファベット（全角→半角）
text = zenhan.z2h(text, mode=2) #数字（全角→半角）
text = zenhan.h2z(text, mode=4) #カタカナ（半角→全角）
symbol = re.sub(r'[\u0000-\uE0FFF]', "", text) #unicode非対応の文字
text = re.sub(r'https?://[\w/:%#\$&\?\(\)~\.=\+\-…]+', "", text) #URL
text = re.sub(r'@[\w/:%#\$&\?\(\)~\.=\+\-…]+', "", text) #ユーザ名
text = re.sub(r'#[\w/:%#\$&\?\(\)~\.=\+\-…]+', "", text) #ハッシュタグ
text = re.sub(
r'[\u0000-\u002F\u003A-\u0040\u005B-\u0060\u007B-\u3004\u3007-\u303F\u3099-\u30A0\u30FB\u30FD-\u4DFF\uA000-\uE0FFF]',
"", text) #その他文字列削除
"""unicode非対応の文字の削除"""
if not symbol == "":
text = re.sub("[%s]" % symbol, "", text)
return text
def get_tweet(auth, g):
url = "https://api.twitter.com/1.1/statuses/mentions_timeline.json"
tweets = requests.get(url,
auth=auth,
params={
"count": "200",
"since_id": g.last_mention
}).json()
if len(tweets) > 0:
g.last_mention = tweets[0]['id_str']
for tweet in tweets:
try:
got_tweet = tweet['text']
patternScreenName = r"@[a-zA-Z0-9/:%#\$&\?\(\)~\.=\+\-_]+"
patternUrl = r"https?://[a-zA-Z0-9/:%#\$&\?\(\)~\.=\+\-_]+"
got_tweet = re.sub(patternScreenName, "", got_tweet)
got_tweet = re.sub(patternUrl, "", got_tweet)
got_tweet = re.sub(r'[\r|\t]', '', got_tweet)
got_tweet = got_tweet.replace('\n', '')
print('kitayo:' + got_tweet)
utterLine = conv.parser(zh.z2h(got_tweet).lower())
utterLineR = utterLine[::-1]
text = "@" + str(
tweet['user']['screen_name']) + " " + conv.conversation(
utterLineR, conv.model, conv.dictionary, conv.id2wd)
if any([text.find(ng) != -1 for ng in ng_words]):
text = "@" + str(tweet['user']['screen_name']) + str(
get_ng_word())
put_tweet(auth, text, tweet['id'], g)
except:
print("wakarazu")
text = "@" + str(tweet['user']['screen_name']) + str(
get_unknown_word())
put_tweet(auth, text, tweet['id'], g)
def copy_tmp_to_forpdf(filename, gakki):
dt = datetime.today()
try:
wb = px.load_workbook(filename)
ws1 = wb['ForPDF']
ws2 = wb['temp']
kogi_bango = ""
for i in range(66, 71):
for j1 in range(2, 17, 2): # Kougi-bango #write course number
r1 = round(3 * j1 / 2 + 2)
kogi_bango = ws2[cell(c=chr(i), i=j1)].value
if kogi_bango is None:
pass
elif isinstance(kogi_bango, str):
kogi_bango0 = kogi_bango.split(" ", 1)[0].strip()
ws1[cell(c=chr(i), i=r1)].value = ""
ws1[cell(c=chr(i), i=r1)].value =\
zh.z2h(text=kogi_bango0, mode=7)
elif isinstance(kogi_bango, int):
kogi_bango0 = "{:06d}".format(kogi_bango)
ws1[cell(c=chr(i), i=r1)].value = ""
ws1[cell(c=chr(i), i=r1)].value =\
zh.z2h(text=kogi_bango0, mode=7)
for j2 in range(3, 18, 2): # write course title
r2 = round(((3 * j2 + 7) / 2) - 1)
kogi_me = ws2[cell(c=chr(i), i=j2)].value
if kogi_me is not None:
kogi_me = kogi_me.replace("－", "-").\
replace("英語コミュニケーション", "EC")
ws1[cell(c=chr(i), i=r2)].value = ""
ws1[cell(c=chr(i), i=r2)].value =\
zh.z2h(text=kogi_me, mode=3)
ws1["F2"].value = dt.strftime("%Y/%m/%d")
if gakki != "":
ws1["D1"].value = "Q{}".format(gakki)
wb.save(filename)
print("Successfully completed")
return True
except PermissionError:
print("The file was not closed.")
return False
def norm(s):
s = s.split("※",1)[0]
s = s.replace(" ", " ")
s = s.replace("－", "-")
s = zenhan.z2h(s, mode=7)
s = zenhan.h2z(s, mode=4)
s = s.strip()
return s
def import_file(cls, filename, kind):
"""Store data from csv files.
filename: string
kind: hotel/restaurant
"""
import csv
import progressbar
import time
from ghost_spider.elastic import LatteHotelEs, LatteRestaurantEs
to_class = None
if kind == 'hotel':
to_class = LatteHotelEs
elif kind == 'restaurant':
to_class = LatteRestaurantEs
else:
raise NotImplementedError()
csvfile = open(filename, 'rb')
fieldnames = cls.fieldnames
reader = csv.DictReader(csvfile, fieldnames)
try:
to_class.DEBUG = False
next(reader) # skip the title line
rows = list(reader)
total = len(rows)
progress = progressbar.AnimatedProgressBar(end=total, width=100)
bulk = ""
count_lines = 0
for line, row in enumerate(rows):
progress += 1
progress.show_progress()
data = {}
for k, v in row.iteritems():
if v:
if not isinstance(v, (list, tuple)):
data.update({k: v.decode('utf-8')})
data["name_low"] = data["name"].lower()
data["name_cleaned"] = to_class.analyze(data["name"].lower(), 'baseform_analyzer')
data["name_cleaned"] = zenhan.z2h(data["name_cleaned"], zenhan.ASCII)
data["url"] = data["url"].lower()
data["kind"] = data["kind"].split('|') if data.get('kind') else []
bulk += to_class.bulk_data(data, action="create")
count_lines += 1
if (count_lines % 200) == 0:
to_class.send(bulk)
bulk = ""
if bulk:
to_class.send(bulk)
progress.show_progress()
print " "
finally:
if csvfile:
csvfile.close()
def conv(txt, unic=False):
kZ = unicode(txt)
kZ = zenhan.z2h(kZ)
kZ = kZ.lower()
kZ = zenhan.h2z(kZ)
if unic:
return kZ
kZ = kZ.encode('utf8')
return kZ
def wakati(str):
words = []
for line in mecab.parse(zenhan.z2h(str, mode=3).lower()).split("\n"):
cols = line.split("\t")
if len(cols) >= 2:
c = cols[1].split(",")
if not c[0] in ["助詞", "助動詞", "副詞", "記号"] and not c[1] in ["非自立", "代名詞"]:
words.append(cols[0])
return words
def run(self, edit):
for region in self.view.sel():
select_texts = self.view.substr(region)
if select_texts != "":
zen2han_text = zenhan.h2z(select_texts,zenhan.KANA)
han2zen_text = zenhan.z2h(select_texts,zenhan.KANA)
if select_texts != zen2han_text:
self.view.replace(edit, region, zen2han_text)
elif select_texts != han2zen_text:
self.view.replace(edit, region, han2zen_text)
def run(self, edit):
for region in self.view.sel():
select_texts = self.view.substr(region)
if select_texts != "":
zen2han_text = zenhan.h2z(select_texts, zenhan.KANA)
han2zen_text = zenhan.z2h(select_texts, zenhan.KANA)
if select_texts != zen2han_text:
self.view.replace(edit, region, zen2han_text)
elif select_texts != han2zen_text:
self.view.replace(edit, region, han2zen_text)
def zenhan_search(self, statement, numOfResult):
han_statement = zenhan.z2h(statement)
zen_statement = zenhan.h2z(statement)
han_list = self.tokenizer.split_query(han_statement)
zen_list = self.tokenizer.split_query(zen_statement)
if han_statement != zen_statement:
to_search = han_list + zen_list
else:
to_search = self.tokenizer.split_query(statement)
return self._search(to_search, numOfResult)
def zenhan_search(self, statement, numOfResult):
han_statement = zenhan.z2h(statement)
zen_statement = zenhan.h2z(statement)
han_list = self.tokenizer.split_query(han_statement)
zen_list = self.tokenizer.split_query(zen_statement)
if han_statement != zen_statement:
to_search = han_list + zen_list
else:
to_search = self.tokenizer.split_query(statement)
return self._search(to_search, numOfResult)
def save_for_production(cls, filename, data):
"""Save this data on csv file by prefecture"""
row = []
address = zenhan.z2h(data['address'], zenhan.ALL)
# remove the zip code
address = re.sub(r'%s\d+-\d+' % u'〒', '', address).strip()
row.append(data['name'])
row.append(data['name_kata'])
row.append(address)
row.append(data['parent_url_key'])
row.append(zenhan.z2h(data['phone'], zenhan.ALL))
# hotel_kind = u'ホテル'
# if data.get('kind') and data.get('kind') in LocationHotelSelectors.REPLACE_HOTEL:
# hotel_kind = data.get('kind')
# else:
# for genre in data['genre']:
# if genre in LocationHotelSelectors.REPLACE_HOTEL:
# hotel_kind = LocationHotelSelectors.REPLACE_HOTEL[genre]
# break
row.append(data.get('kind') or '')
CsvWriter.write_to_csv(filename, row, firs_row=cls.production_first_row)
def normalize(data):
NBSP = b"\xC2\xA0".decode("UTF-8")
data = unicodedata.normalize("NFKC", zenhan.z2h(zenhan.h2z(data.replace(NBSP, ""))))
# 0x2010 -- 0x2015
dashesU8 = [b'\xe2\x80\x90', b'\xe2\x80\x91', b'\xe2\x80\x92', b'\xe2\x80\x93', b'\xe2\x80\x94', b'\xe2\x80\x95']
dashes = "".join([s.decode("UTF-8") for s in dashesU8])
digits = re.match("^[0-9\\+\\-{0}]+$".format(dashes), data)
if digits:
for d in dashes:
data = data.replace(d, "-")
return data
def normalizeText(string):
patternUrl = r"https?://[a-zA-Z0-9/:%#\$&\?\(\)~\.=\+\-_]+"
patternScreenName = r"@[a-zA-Z0-9/:%#\$&\?\(\)~\.=\+\-_]+"
patternHashtag = r"#[a-zA-Z0-9/:%#\$&\?\(\)~\.=\+\-_]+"
rep1str = re.sub(patternUrl, "", string)
rep2str = re.sub(patternScreenName, "", rep1str)
rep3str = re.sub(patternHashtag, "", rep2str)
rep4str = re.sub("(^(\s)*)|((\s)*$)", "", rep3str)
rep5str = rep4str.replace("\n", "")
rep6str = zh.z2h(rep5str).lower()
return rep6str
def extract_data_from_html(filename):
# html to table
url_filename = return_urlfilename(filename)
html = urlopen(url_filename)
bsObj = BeautifulSoup(html, "html.parser")
table = bsObj.findAll("table")
# table to update_date
updated_date = strptime(z2h(table[1].get_text().strip(), 2), "%Y年%m月%d日")
# table to gpa_list
gpa_table = table[4]
gpa_rows = gpa_table.findAll("tr")
gpa_list = []
gpa_items = [
"course_title", "lecturer", "year_completed", "grade_points", "grade",
"credits", "gp"
]
gpa_rows.pop(0)
for gpa_row in gpa_rows:
tmp_dict = {item: None for item in gpa_items}
is_append_list = True
for i, cell in enumerate(gpa_row.findAll(['td', 'th'])):
tmp_celltext =\
z2h(cell.get_text().strip().replace("\u3000", " "), 3)
if i == 3 and tmp_celltext == "":
is_append_list = False
continue
tmp_dict[gpa_items[i]] = tmp_celltext
if is_append_list is True:
# tmp_dict["year_completed"] = []
tmp_dict["grade_points"] = int(tmp_dict["grade_points"])
tmp_dict["credits"] = float(tmp_dict["credits"])
tmp_dict["gp"] = float(tmp_dict["gp"])
gpa_list.append(tmp_dict)
return updated_date, gpa_list
def save_to_csv(cls, filename, data):
"""Save this data on csv file by prefecture"""
row = []
address = zenhan.z2h(data['address'], zenhan.ASCII)
# remove the zip code
address = re.sub(r'%s\d+-\d+' % u'〒', '', address).strip()
row.append(data['name'])
row.append(data['name_kata'])
row.append(address)
row.append(data['prefecture'])
row.append(data['area'])
row.append(zenhan.z2h(data['phone'], zenhan.ALL))
row.append(data['kind'])
row.append(data.get('latte_url') or u'')
row.append(data['page_url'])
row.append(data['id'])
CsvWriter.write_to_csv(filename, row, firs_row=cls.first_row)
def tokenize(text):
'''
とりあえず形態素解析して名詞だけ取り出す感じにしてる
Extract alphabet as lower, hankaku, space-trimed
'''
node = mecab.parseToNode(text)
while node:
if node.feature.split(',')[0] == '名詞':
try:
yield zenhan.z2h(node.surface.lower().strip())
except:
yield '0'
node = node.next
def search(browser):
#検索値入力
query = input("商品名入力: ")
#amazonHPへ
browser.get('https://www.amazon.co.jp/')
#検索値入力
search = browser.find_element_by_id('twotabsearchtextbox')
search.send_keys(query)
#検索ボタン押す
search_btn = browser.find_element_by_class_name('nav-input')
search_btn.click()
time.sleep(2)
#商品一覧取得
items = browser.find_elements_by_class_name('s-result-item')
time.sleep(2)
for num, item in enumerate(items):
try:
item_name = item.find_element_by_tag_name('h2').text
print(num, ':', item_name)
print('-' * 20)
except:
pass
#商品の選択
select_num = str(input('商品番号入力: '))
select_item_num = 'result_' + zenhan.z2h(select_num)
time.sleep(1)
select_item_html = browser.find_element_by_id(select_item_num)
#その商品の詳細ページURL取得
select_item = select_item_html.find_element_by_class_name('a-link-normal')
select_item_url = select_item.get_attribute('href')
#商品名取得
select_item_name = select_item_html.find_element_by_tag_name('h2').text
browser.get(select_item_url)
time.sleep(1)
return select_item_name
def month_date_end_search(line):
"""月表記のある予定終了の日付を検出し,intとintで返す."""
zen_tilde = '～'
# 全角スペース
zen_space = ' '
# 全角0
zen_zero = '０'
nichi = '日'
tsuki = '月'
dollar = '$'
# 全角スペースを0に置き換えることで無理やり対応
line = line.replace(zen_space, zen_zero)
line = line.replace(zen_tilde, zen_zero)
index_month = line.find(tsuki)
# 日が一桁の場合の対策
line = line.replace(tsuki, zen_zero, 1)
# 二度目のnichiの位置を検出
index_second_date = line.find(nichi, index_month + 1)
# 日と曜日の位置関係から誤表記を訂正
index_second_dollar = line.find(dollar, index_month + 1)
if index_second_date + 1 != index_second_dollar:
index_second_date = index_second_dollar
# 月, 日を返す
return int(zenhan.z2h(line[index_month - 2:index_month])), int(zenhan.z2h(line[index_second_date - 2:index_second_date]))
def date_start_search(line):
"""予定開始の日付を検出し,strで返す."""
# 全角スペース
zen_space = ' '
# 全角0
zen_zero = '０'
nichi = '日'
dollar = '$'
# 全角スペースを0に置き換えることで無理やり対応
line = line.replace(zen_space, zen_zero)
index = line.find(nichi)
# 日と曜日の位置関係から誤表記を訂正
index_first_dollar = line.find(dollar, index + 1)
if index + 1 != index_first_dollar:
index = index_first_dollar
# ex. １ → ０１
#if line[index - 1] == zen_space:
# line[index - 1] = zen_zero
return zenhan.z2h(line[index - 2:index])
def getCardURL(name):
url = "https://ocg.xpg.jp/search/search.fcgi?Name=" + urllib.parse.quote(
name.encode('Shift_JIS')) + "&Mode=0"
try:
fp = requests.get(url)
soup = BeautifulSoup(fp.content, "html.parser")
fp.close
time.sleep(1)
texts = soup.find_all("a", href=re.compile("/c/+"))
for text in texts:
name_tmp = p.sub("", str(text))
name_tmp = zenhan.z2h(name_tmp.replace('－', '-'), 3)
if '【' in name_tmp:
name_tmp = name_tmp[:name_tmp.index("【")]
if name == name_tmp:
url_text = text.get("href")
return url_text
except urllib.error.HTTPError:
time.sleep(1)
return False
def date_end_search(line):
"""月表記のない予定終了の日付を検出し,strで返す."""
zen_tilde = '～'
# 全角スペース
zen_space = ' '
# 全角0
zen_zero = '０'
nichi = '日'
dollar = '$'
# 全角スペースと全角チルダを0に置き換えることで無理やり対応
line = line.replace(zen_space, zen_zero)
line = line.replace(zen_tilde, zen_zero)
index_first_date = line.find(nichi)
# 二度目のnichiの位置を検出
index_second_date = line.find(nichi, index_first_date + 1)
# 日と曜日の位置関係から誤表記を訂正
index_second_dollar = line.find(dollar, index_first_date + 2)
if index_second_date + 1 != index_second_dollar:
index_second_date = index_second_dollar
return zenhan.z2h(line[index_second_date - 2:index_second_date])
def changeClassroom(self, classroom):
if match(r".*,.*", classroom):
cr = classroom
elif classroom == "工学部１号館情報実習室１（CAE室）":
cr = "工1-CAE室"
elif match(r"一般教育棟.*", classroom):
cr = classroom.replace("一般教育棟", "").replace("教室", "")
elif match(r"工学部.*", classroom):
cr = classroom.replace('工学部', "工").replace("号館第", "-").replace(
"号館", "-").replace("講義室", "")
elif match(r"情報実習室.*", classroom):
cr = classroom.replace("情報実習室", "情")
elif match(r"理学部.*", classroom):
cr = classroom.replace("理学部", "理").replace("号館第", "-").replace(
"号館", "-").replace("講義室", "")
else:
cr = classroom
cr = cr.replace(" ", "")
return zh.z2h(text=cr, mode=3)
def scp_number(msg):
msg = zenhan.z2h(msg.casefold()).replace("-", "").replace("scp", "")
number = re.sub("\\D", "", msg)
if number is (None and ""):
return None
brt = msg.replace(number, "")
if brt == "":
brt = "en"
if brt not in BRANCHS: # 要改良
reply = get_country_from_code(brt)
return reply
try:
dictionary = pd.read_csv(currentpath + "/data/scps.csv", index_col=0)
except FileNotFoundError as e:
print(e)
result = dictionary.query('branches in @brt')
result = result.query('url.str.contains(@number)', engine='python')
result = result[0:1].values.tolist()
result = itertools.chain(*result)
result = list(result)
if len(result) == 0 or number is re.sub("\\D", "", result[0]):
if len(number) > 4:
return None
if "en" in brt:
return("scp-" + str(number) + "はまだ存在しません")
else:
return("scp-" + str(number) + "-" + str(brt) + "はまだ存在しません")
return(result)
def setName(string):
name = zenhan.z2h(p.sub("", string).replace('－', '-'), 3)
if '】' in name:
name = name[:name.index('【')]
return name
def med_facility(tdfk, facid):
tdfk = zenhan.z2h(tdfk, mode=7, ignore=()).zfill(2)
facid = zenhan.z2h(facid, mode=7, ignore=()).zfill(7)
return 'M' + tdfk + facid
def get_recipe(url, dish):
#print "**********"
#print dish
request = urllib2.Request(url)
response = urllib2.urlopen(request)
response1=response.read()
soup = BeautifulSoup(response1,"html.parser")
recipe =""
for p in soup.findAll('p',text=False):
if p.text.find("人")!=-1 and p.text.find("材料")!=-1:
recipe = p.text
temps=recipe.split("\n")
elements =dict()
amount=0
people=0
# print recipe
for temp in temps:
# calculate number of people (done)
if temp.find("材料")!=-1:
people=float(re.search("[0-9]",zenhan.z2h(temp,2)).group())
# get each element for one man (done)
elif temp!="":
element=temp.replace("●","").replace("○","").replace("〇","").replace("◎","").lstrip(" ")
if temp.find("…")!=-1:
element= element.split("…")
else:
element= element.split(None,1)
# print element
if people!=0:
# print people
# convert all string to hankaku
if len(element) >= 2:
han_element = zenhan.z2h(element[1],2)
else:
# print element
break
# march unit
#print element[1]
#print han_element
unit=re.search("[^0-9\/～ ]+",han_element).group(0)
#print zenhan.z2h(element[1],2)
string_amount=re.search("[0-9\/ ]+",han_element.replace(unit,""))
#print string_amount
if string_amount!=None:
amount= float(sum(Fraction(s) for s in string_amount.group(0).split()))/people
else:
amount =0
else:
print "people=0"
#print element[0]+"\t"+str(amount)+"\t"+unit
elements.update({element[0]:[amount,zenhan.h2z(unit,4)]})
return elements
def normalize(text):
return zenhan.z2h(text, mode=zenhan.DIGIT | zenhan.ASCII)
def normalize_text(text):
text = text.strip()
text = zenhan.z2h(text, mode=7)
for a, b in normalize_replace_map:
text = text.replace(a, b)
return text
def test_z2h_digit_only(self):
converted = zenhan.z2h(self.original, zenhan.DIGIT)
self.assertEqual(converted, u("ﾟabcＤＥﾞＦ123456ｱｶﾞｻダナバビﾌﾟﾍﾟﾟ"))
def zenAlphaNum2hanAlphaNum(strings):
"""
全角英数字を半角英数字に変換する
"""
strings = MultiBytes.convert2unicode(strings)
return zenhan.z2h(strings, mode=3)
def test_z2h_all(self):
converted = zenhan.z2h(self.original, zenhan.ALL)
self.assertEqual(converted, u("ﾟabcDEﾞF123456ｱｶﾞｻﾀﾞﾅﾊﾞﾋﾞﾌﾟﾍﾟﾟ"))
self.assertEqual(converted,
zenhan.z2h(self.original,
zenhan.ASCII|zenhan.DIGIT|zenhan.KANA))
def test_z2h_ascii_and_digit(self):
converted = zenhan.z2h(self.original, zenhan.ASCII|zenhan.DIGIT)
self.assertEqual(converted, u("ﾟabcDEﾞF123456ｱｶﾞｻダナバビﾌﾟﾍﾟﾟ"))
i = 0
for p1 in tmp1.find_all("td", {"class": "a-center tb-color001"}):
for p2 in p1.find_all("span"):
# 銘柄コード
#print(p2.attrs['id'])
data['code'][i] = p2.attrs['id']
i += 1
for i, p1 in enumerate(
tmp1.find_all(attrs={
"rowspan": "2",
"class": "a-center tb-color001 w-space"
})):
#上場日
data['jojodate'][i] = zenhan.z2h(p1.contents[0].string).strip()
i = 0
for p1 in tmp1.find_all(attrs={"rowspan": "2", "class": "a-left tb-color001"}):
#会社名
data['name'][i] = p1.find("a").text.replace('*', '').strip()
i += 1
i = 0
for p1 in tmp1.find_all("tr"):
for p2 in p1.find_all("td", {"class": "a-center tb-color001"}):
for p3 in p2:
if ("第一部" in p3 or "第二部" in p3 or "マザーズ" in p3
or "JQスタンダード" in p3):
def test_z2h_ascii_and_kana(self):
converted = zenhan.z2h(self.original, zenhan.ASCII|zenhan.KANA)
self.assertEqual(converted, u("ﾟabcDEﾞF123４５６ｱｶﾞｻﾀﾞﾅﾊﾞﾋﾞﾌﾟﾍﾟﾟ"))
def zenkaku_to_hankaku(text):
'''全角文字を半角文字に変換します。'''
return zenhan.z2h(text, mode=7)
def setPendulumScale(scale):
scale = int(zenhan.z2h(scale[scale.index('赤') + 1:]))
return scale
def test_z2h_kana_only(self):
converted = zenhan.z2h(self.original, zenhan.KANA)
self.assertEqual(converted, u("ﾟabcＤＥﾞＦ123４５６ｱｶﾞｻﾀﾞﾅﾊﾞﾋﾞﾌﾟﾍﾟﾟ"))
def test_z2h_digit_and_kana(self):
converted = zenhan.z2h(self.original, zenhan.DIGIT|zenhan.KANA)
self.assertEqual(converted, u("ﾟabcＤＥﾞＦ123456ｱｶﾞｻﾀﾞﾅﾊﾞﾋﾞﾌﾟﾍﾟﾟ"))
def messaging_service():
logger.info('messaging_service()')
strreq = request.data
try:
strreq = strreq.decode()
except AttributeError:
pass
jsonreq = json.loads(strreq)
if debuglog:
logger.info(str(jsonreq))
reply_token = ''
message = ''
strMessage = ''
user_id = ''
etype = ''
postbackdata = ''
timestamp = datetime.now()
for e in jsonreq['events']:
etype = e['type']
reply_token = e['replyToken']
user_id = e['source']['userId']
if etype == 'message':
message = e['message']
strMessage = message['text']
timestamp = int(e['timestamp'])
elif etype == 'follow':
timestamp = int(e['timestamp'])
elif etype == 'postback':
logger.info(str(e))
postbackdata = e['postback']['data']
timestamp = int(e['timestamp'])
date = datetime.now()
send_line_bot_log(user_id, f"{user_id},{strMessage}," + date.strftime('%Y/%m/%d %H:%M:%S'))
strIntent = ""
_i = " "
_q = " "
_a = " "
_n = " "
#followevent
if etype == 'follow' or strMessage == u'クイズ連携':
send_line_bot_log(user_id, f"{user_id},follow message," + date.strftime('%Y/%m/%d %H:%M:%S'))
mess = []
mess.append("言えまてんBotです！フォローありがとう！")
mess.append({
'type': 'template',
'altText': '確認',
'template' : {
'type': 'confirm',
'text': 'Clovaはお持ちですか？言えまてんクイズからのメッセージをこちらに送信しても大丈夫ですか？',
'actions': [
{
'type': 'postback',
'label': 'はい',
'data': 'res=yes'
},
{
'type': 'postback',
'label': 'いいえ',
'data': 'res=no'
}
]
}
})
send_line_reply(reply_token, mess)
if etype == 'postback':
if postbackdata == 'res=yes':
text = "ありがとう！言えまてんクイズの使い方を知りたい時は「使い方」と言ってみてね！"
send_line_reply(reply_token, text)
insert(user_id, "follow", "follow", "ok" ,_i,_q,_a,_n)
else:
text = "残念です！言えまてんクイズの使い方を知りたい時は「使い方」と言ってみてね！"
send_line_reply(reply_token, text)
insert(user_id, "follow", "follow", "ng" ,_i,_q,_a,_n)
return True
intent = getIntent(user_id, "reply")
if 'date' in intent:
postDate = intent['date']
if datetime.strptime(postDate , '%Y/%m/%d %H:%M:%S') > datetime.now() - timedelta(hours=8):
strIntent = intent['intent']
_i = intent['induction']
_q = intent['quiz']
_a = intent['answer']
if re.compile("こんにちは|Hello|こんばんは|おはよう").search(message['text']):
text = "こんにちは！言えまてんボットです。よろしくね。「使い方」というと説明するよ！"
send_line_reply(reply_token, text)
elif re.compile("^(使い方|つかいかた|Help|ヘルプ)").search(message['text']):
mess = u'言えまてんクイズの使い方です。最初にひとつの言葉を10回繰り返して言ってもらいます。\n'
mess += u'次に、その言葉に少し関係のある問題を出すので答えを考えてね。問題は全部で' + str(len(quiz)) + 'つありますよ。\n'
mess += u'答えがあっていると正解！です。もう一度問題をやるか聞かれたら「はい」か「いいえ」と答えてね。\n'
mess += u'「問題1」、「問題1の答え」のようにBotに言うと問題についてお答えします。問題を思いついた人は「応募」と言ってみてね。\n'
send_line_reply(reply_token,mess)
elif strIntent == "post1" and len(message['text']) > 1:
_i = message['text']
insert(user_id, "reply", "post2", message,_i,_q,_a,_n)
mess = []
mess.append( "「" + message['text'] + "」ですね。わかりました。")
mess.append( "次に問題を教えてください。")
send_line_reply(reply_token,mess)
elif strIntent == "post2" and len(message['text']) > 5:
_q = message['text']
insert(user_id, "reply", "post3", message,_i,_q,_a,_n)
mess = []
mess.append( "問題は「" + message['text'] + "」ですね。わかりました。")
mess.append( "次は答えを教えてください。")
send_line_reply(reply_token,mess)
elif strIntent == "post3" and len(message['text']) > 0:
_a = message['text']
insert(user_id, "reply", "post4", message,_i,_q,_a,_n)
mess = []
mess.append( "答えは「" + message['text'] + "」ですね。わかりました。")
mess.append( "最後にニックネームを教えて！もし採用されたら問題の解説の時に紹介するね。内緒にしたい時は「匿名」と答えてね。")
send_line_reply(reply_token,mess)
elif strIntent == "post4" and len(message['text']) > 0:
_n = message['text']
insert(user_id, "reply", "finish", message,_i,_q,_a,_n)
mess = []
mess.append( "「" + message['text'] + "」さん、応募ありがとう！")
mess.append( "参考にするね！")
send_line_reply(reply_token,mess)
send_sns(str(jsonreq), _i, _q, _a, _n)
elif re.compile("^(応募|おうぼ|投稿)").search(message['text']):
insert(user_id, "reply", "post1", message,_i,_q,_a,_n)
mess = []
mess.append( "言えまてんクイズです。新しい問題を応募してます。面白い問題を考えた人は、1.10回繰り返す言葉(キリンとか)、2.問題、3.答えの3つを教えてね。")
mess.append( "では、10回繰り返して言ってもらうフレーズを教えてください。")
send_line_reply(reply_token, mess)
elif re.compile("^問題[0-9]{1,2}$").search(zenhan.z2h(message['text'])):
match = re.compile("[0-9]{1,2}").search(zenhan.z2h(message['text']))
num = int(match.group())
if num > 0:
if len(quiz) >= num:
text = f"問題{num}: {quiz[num]['q']}"
send_line_reply(reply_token,text)
else:
text = f"問題{num}がみつかりません。"
send_line_reply(reply_token,text)
elif re.compile("^問題[0-9]{1,2}の答").search(zenhan.z2h(message['text'])):
match = re.compile("[0-9]{1,2}").search(zenhan.z2h(message['text']))
num = int(match.group())
if num > 0:
if len(quiz) >= num:
mess = []
mess.append(f"問題{num}の答え=> {quiz[num]['a'][0]}")
mess.append(f"問題{num}の解説=> {quiz[num]['i']}")
send_line_reply(reply_token,mess)
else:
text = f"問題{num}がみつかりません。"
send_line_reply(reply_token,text)
return True
def normalize(data):
NBSP = b"\xC2\xA0".decode("UTF-8")
return unicodedata.normalize("NFKC", zenhan.z2h(zenhan.h2z(data.replace(NBSP, ""))))
def test_z2h_ascii_only(self):
converted = zenhan.z2h(self.original, zenhan.ASCII)
self.assertEqual(converted, u("ﾟabcDEﾞF123４５６ｱｶﾞｻダナバビﾌﾟﾍﾟﾟ"))