def ranking_today2():
df = get_speadsheet_data()
rank = 0
ranking = "昨日の新型コロナウイルス感染者数ランキング\n"
is_same = False
current_num = -1
yesterday = str((datetime.today() - timedelta(days=1)).date())
for city, num in df.loc[yesterday].sort_values(
ascending=False).to_dict().items():
if current_num == num:
is_same = True
else:
ranking += "\n"
rank += 1
is_same = False
if not is_same:
if count_twitter(ranking) >= 230:
ranking += f"(以下 https://narumi-midori.net/twitter_aichi_covid19/{yesterday}.html)"
break
else:
ranking += f"{rank}位 {num}人 {city}"
current_num = num
else:
if count_twitter(ranking) >= 230:
ranking += f"(以下 https://narumi-midori.net/twitter_aichi_covid19/{yesterday}.html)"
break
else:
ranking += f", {city}"
post(ranking)
return ranking
def post_nagoya():
if os.path.isfile("nagoya_lock.zip"):
pass
else:
load_url = 'https://www.city.nagoya.jp/kenkofukushi/page/0000126920.html'
html = requests.get(load_url)
soup = BeautifulSoup(html.content, "html.parser")
nagoya_h3 = soup.find("h3")
date = datetime.strptime(
nagoya_h3.text.split("令和3年")[1].split("現在")[0], '%m月%d日')
# date = datetime.strptime(nagoya_h3.text.split("令和2年")
# [1].split("現在")[0], '%m月%d日')
today = datetime.today()
if (today.month == date.month) & (today.day == date.day):
is_today = True
else:
is_today = False
article_text = nagoya_h3.next_element.next_element.find("p").text
article_url = load_url
num_today = int(re.sub("\\D", "", article_text))
# is_today = True
if is_today:
df0 = pandas.read_pickle("database.zip")
num_last_week = get_number_by_delta(df0, -7, region="名古屋市")
youbi = get_day_of_week_jp(today)
header = f'[速報]名古屋市の本日の新型コロナウイルスの新規感染者数は{num_today}人(先週の{youbi}に比べて{num_today-num_last_week:+}人)でした。詳細は公式サイトを参照 > {article_url}'
post(header)
data_for_save = pandas.DataFrame([{
'本日': num_today,
'先週': num_last_week
}],
index=['名古屋市'])
data_for_save.to_pickle("nagoya_lock.zip")
time.sleep(5)
print("名古屋市更新しました", datetime.today())
def post() -> None:
from twitter_post import post
medical = get_medical_number(get_medical_data())
non_medical = get_non_medical_number(get_open_data())
headline = generate_headline_first_second(medical, non_medical)
last_number = get_last_total_number()
current_numer = extract_total_number(headline)
if current_numer > last_number:
post(headline)
else:
print("The number is the same as the last number.")
def post_toyohashi():
if os.path.isfile("toyohashi_lock.zip"):
pass
else:
load_url = 'https://www.city.toyohashi.lg.jp/41805.htm'
html = requests.get(load_url)
soup = BeautifulSoup(html.content, "html.parser")
toyohashi_new = soup.find(class_='Item_normal')
# toyohashi_header = "豊橋市が新型コロナウイルス情報を更新しました > "
article_text = toyohashi_new.text.replace("\n", "").replace("\xa0", "")
article_url = load_url
today = datetime.today()
nums = re.findall(r"\d+", article_text)
nums = [
_num.translate(
str.maketrans(
{chr(0xFF01 + i): chr(0x21 + i)
for i in range(94)})) for _num in nums
]
date = datetime.strptime(f"{nums[1]}月{nums[2]}日", "%m月%d日")
if (today.month == date.month) & (today.day == date.day):
is_today = True
else:
is_today = False
# is_today = True
if is_today:
num_today = int(nums[3])
df0 = pandas.read_pickle("database.zip")
num_last_week = get_number_by_delta(df0, -7, region="豊橋市")
youbi = get_day_of_week_jp(today)
header = f'[速報]豊橋市の本日の新型コロナウイルスの新規感染者数は{num_today}人(先週の{youbi}に比べて{num_today-num_last_week:+}人)でした。詳細は公式サイトを参照 > {article_url}'
post(header)
data_for_save = pandas.DataFrame([{
'本日': num_today,
'先週': num_last_week
}],
index=['豊橋市'])
data_for_save.to_pickle("toyohashi_lock.zip")
time.sleep(5)
print("豊橋市更新しました", datetime.today())
def post_zentai():
is_toyohashi_done = os.path.isfile("toyohashi_lock.zip")
is_toyota_done = os.path.isfile("toyota_lock.zip")
is_okazaki_done = os.path.isfile("okazaki_lock.zip")
is_nagoya_done = os.path.isfile("nagoya_lock.zip")
is_aichi_done = os.path.isfile("aichi_lock.zip")
is_ichinomiya_done = os.path.isfile("ichinomiya_lock.zip")
is_dones = [
is_aichi_done, is_nagoya_done, is_okazaki_done, is_toyohashi_done,
is_toyota_done, is_ichinomiya_done
]
if (all(is_dones)) and (not os.path.isfile("zentai.lock")):
df_toyohashi = pandas.read_pickle("toyohashi_lock.zip")
df_toyota = pandas.read_pickle("toyota_lock.zip")
df_okazaki = pandas.read_pickle("okazaki_lock.zip")
df_nagoya = pandas.read_pickle("nagoya_lock.zip")
df_aichi = pandas.read_pickle("aichi_lock.zip")
df_ichinomiya = pandas.read_pickle("ichinomiya_lock.zip")
df_today = pandas.concat([
df_toyohashi, df_aichi, df_nagoya, df_toyota, df_okazaki,
df_ichinomiya
])
num_today = df_today['本日'].sum()
num_last_week = df_today['先週'].sum()
youbi = get_day_of_week_jp(datetime.today() - timedelta(hours=6))
article_url = 'https://www.pref.aichi.jp/site/covid19-aichi/'
header = f'[速報]本日の愛知県全体の新型コロナウイルスの新規感染者数は{num_today}人(先週の{youbi}に比べて{num_today-num_last_week:+}人)でした。詳細は公式サイトを参照 > {article_url}'
# print(header)
post(header)
df_today.to_pickle(
os.path.join(
"data",
f"{str(datetime.today()-timedelta(hours=6)).split()[0]}_from_sum.zip"
))
with open("zentai.lock", "w", encoding="utf-8") as f:
f.write("")
time.sleep(5)
print("愛知県全体更新しました", datetime.today())
def post_aichi():
if os.path.isfile("aichi_lock.zip"):
pass
else:
d_atom = feedparser.parse(
'https://www.pref.aichi.jp/rss/10/site-758.xml')
is_today = False
today = datetime.today() - timedelta(hours=6)
for entry in d_atom['entries']:
_day = datetime.strptime(entry['updated'],
"%Y-%m-%dT%H:%M:%S+09:00")
# print(_day.month, today.month, _day.day, today.day,
# entry['title'], '新型コロナウイルス感染症患者の発生について')
if (_day.month == today.month) and (_day.day == today.day) and (
entry['title'] == '新型コロナウイルス感染症患者の発生について'):
article_url = entry['id']
is_today = True
break
if is_today:
load_url = article_url
# load_url = 'https://www.pref.aichi.jp/site/covid19-aichi/pressrelease-ncov201208.html'
html = requests.get(load_url)
soup = BeautifulSoup(html.content, "html.parser")
article_text = soup.find(class_="mol_textblock").text
nums = re.findall(r"\d+", article_text)
num_today = int(nums[0])
df0 = pandas.read_pickle("database.zip")
num_last_week = get_number_by_delta(df0, -7, region="愛知県")
youbi = get_day_of_week_jp(today)
header = f'[速報]愛知県管轄自治体（名古屋市・豊橋市・豊田市・岡崎市・一宮市を除く愛知県）の本日の新型コロナウイルスの新規感染者数は{num_today}人(先週の{youbi}に比べて{num_today-num_last_week:+}人)でした。詳細は公式サイトを参照 > {article_url}'
post(header)
data_for_save = pandas.DataFrame([{
'本日': num_today,
'先週': num_last_week
}],
index=['愛知県'])
data_for_save.to_pickle("aichi_lock.zip")
time.sleep(5)
print("愛知県更新しました", datetime.today())
def ranking_week2():
data = get_speadsheet_data()
rank = 0
ranking = "昨日まで直近1週間の新型コロナウイルス感染者数ランキング\n"
is_same = False
current_num = -1
yesterday = str((datetime.today() - timedelta(days=1)).date())
df = pandas.DataFrame([])
indices = data.index.sort_values(ascending=False)
for num in range(len(indices) - 6):
df1 = data.loc[indices[num:num + 7], :].sum().to_frame().transpose()
df1.index = [indices[num]]
df = pandas.concat([df, df1])
for city, num in df.loc[yesterday].loc[yesterday].sort_values(
ascending=False).to_dict().items():
if current_num == num:
is_same = True
else:
ranking += "\n"
rank += 1
is_same = False
if not is_same:
if count_twitter(ranking) >= 230:
ranking += f"(以下 https://narumi-midori.net/twitter_aichi_covid19/{yesterday}_week.html)"
break
else:
ranking += f"{rank}位 {num}人 {city}"
current_num = num
else:
if count_twitter(ranking) >= 230:
ranking += f"(以下 https://narumi-midori.net/twitter_aichi_covid19/{yesterday}_week.html)"
break
else:
ranking += f", {city}"
post(ranking)
return ranking
def ranking_week():
if not (os.path.isfile("ranking_week.lock")):
this_mo = pandas.read_pickle("database.zip")
# this_mo = pandas.read_pickle(f"{os.path.splitext(pdf_name)[0]}.zip")
this_week = this_mo[this_mo["発表日"] >= datetime.today() -
timedelta(days=8) - timedelta(hours=6)]
pd_week = pandas.DataFrame(
collections.Counter(this_week["住居地"]).most_common())
pd_week[0] = [_.replace("⻄", "西") for _ in pd_week[0]]
ranking_text = "昨日まで直近1週間の新型コロナウイルス感染者数ランキング\n"
rank = 0
num_prior = 0
yesterday = datetime.today() - timedelta(days=1)
for city, num in zip(pd_week[0], pd_week[1]):
if num == num_prior:
# if parse_tweet(ranking_text).weightedLength > 258:
if parse_tweet(ranking_text).weightedLength > 223:
ranking_text += f"(以下 https://narumi-midori.net/twitter_aichi_covid19/{str(yesterday.date())}_week.html)"
break
else:
ranking_text += f", {city}"
else:
# if parse_tweet(ranking_text).weightedLength > 252:
if parse_tweet(ranking_text).weightedLength > 227:
ranking_text += f"(以下 https://narumi-midori.net/twitter_aichi_covid19/{str(yesterday.date())}_week.html)"
break
else:
rank += 1
ranking_text += f"\n{rank}位 {num}人: {city}"
num_prior = num * 1
# print(ranking_text, parse_tweet(ranking_text).weightedLength)
post(ranking_text)
# print(header)
with open("ranking_week.lock", "w", encoding="utf-8") as f:
f.write("")
def post_toyota():
# if os.path.isfile("toyota_lock.zip"):
# pass
# else:
load_url = 'https://www.city.toyota.aichi.jp/kurashi/kenkou/eisei/1039225.html'
html = requests.get(load_url)
soup = BeautifulSoup(html.content, "html.parser")
toyota_new = soup.find(class_="objectlink")
today = f"{datetime.today().month}月{datetime.today().day}日"
article_text = toyota_new.find("li").text
article_url = urljoin(load_url, toyota_new.find("li").find('a')['href'])
today = datetime.today()
nums = re.findall(r"\d+", article_text)
# print(nums)
date = datetime.strptime(f"{nums[0]}月{nums[1]}日", "%m月%d日")
if (today.month == date.month) & (today.day == date.day):
is_today = True
else:
is_today = False
is_zero = False
if not is_today:
load_url2 = "https://www.city.toyota.aichi.jp/kurashi/kenkou/eisei/1037578.html"
html = requests.get(load_url2)
soup = BeautifulSoup(html.content, "html.parser")
toyota_new = soup.find("h2")
article_text = toyota_new.next_element.next_element.next_element.text
nums = re.findall(r"\d+", article_text)
is_zero = "いません" in article_text
date = datetime.strptime(f"{nums[0]}月{nums[1]}日", "%m月%d日")
if (today.month == date.month) & (today.day == date.day):
is_today = True
else:
is_today = False
# ex1 = "市内在住者（3人）が新型コロナウイルスに感染したことが判明しました。（1248～1250例目）"
if is_today:
if len(nums) == 3:
num_today = 1
elif is_zero:
num_today = 0
else:
num_today = int(nums[3]) - int(nums[2]) + 1
# num_today = int(nums[2])
df0 = pandas.read_pickle("database.zip")
num_last_week = get_number_by_delta(df0, -7, region="豊田市")
youbi = get_day_of_week_jp(today)
if not os.path.isfile("toyota_lock.zip"):
header = f'[速報]豊田市の本日の新型コロナウイルスの新規感染者数は{num_today}人(先週の{youbi}に比べて{num_today-num_last_week:+}人)でした。詳細は公式サイトを参照 > {article_url}'
elif int(pandas.read_pickle("toyota_lock.zip").loc["豊田市",
"本日"]) < num_today:
header = f'[更新]豊田市の本日の新型コロナウイルスの新規感染者数は{num_today}人(先週の{youbi}に比べて{num_today-num_last_week:+}人)に更新されました。詳細は公式サイトを参照 > {article_url}'
else:
header = None
if (header is not None) and (num_today > 0):
data_for_save = pandas.DataFrame([{
'本日': num_today,
'先週': num_last_week
}],
index=['豊田市'])
data_for_save.to_pickle("toyota_lock.zip")
post(header)
time.sleep(5)
print("豊田市更新しました", datetime.today())
if h3.find("a") is not None:
article_url = urljoin(load_url,
h3.find("a")['href'].replace("./", ""))
else:
article_url = load_url
# print(article_url)
youbi = get_day_of_week_jp(today)
if not os.path.isfile("okazaki_lock.zip"):
header = f'[速報]岡崎市の本日の新型コロナウイルスの新規感染者数は{num_today}人(先週の{youbi}に比べて{num_today-num_last_week:+}人)でした。詳細は公式サイトを参照 > {article_url}'
elif int(pandas.read_pickle("okazaki_lock.zip").loc["岡崎市",
"本日"]) < num_today:
header = f'[更新]岡崎市の本日の新型コロナウイルスの新規感染者数は{num_today}人(先週の{youbi}に比べて{num_today-num_last_week:+}人)に更新されました。詳細は公式サイトを参照 > {article_url}'
else:
header = None
if header is not None:
post(header)
# print(header)
data_for_save = pandas.DataFrame([{
'本日': num_today,
'先週': num_last_week
}],
index=['岡崎市'])
data_for_save.to_pickle("okazaki_lock.zip")
time.sleep(5)
print("岡崎市更新しました", datetime.today())
def post_toyohashi():
if os.path.isfile("toyohashi_lock.zip"):
pass
else:
# compile then match
repatter = re.compile(pattern)
result = repatter.match(text)
return int(result.group(1))
def get_last_total_number() -> int:
from twitter_post import get_posts
timelines = get_posts()
text_line_text = get_last_post(timelines)
return extract_total_number(text_line_text)
if __name__ == "__main__":
post()
# post2_vaccination()
# medical = get_medical_number(get_medical_data())
# non_medical = get_non_medical_number(get_open_data())
# print(generate_headline_first_second(medical, non_medical))
# a = get_vaccination_number_from_open_data_df("summary_by_prefecture.csv")
# print(a["count_first_or_mid_general"])
# print(a["count_second_or_full_general"])
# print(get_medical_number("IRYO-kenbetsu-vaccination_data.xlsx"))
# df_m = get_df("IRYO-kenbetsu-vaccination_data.xlsx", "医療従事者接種回数")
# print(generate_headline())
# print(get_open_data())
# print(get_vaccination_number())
def ranking_today():
if not (os.path.isfile("ranking_today.lock")):
press_url = "https://www.pref.aichi.jp/site/covid19-aichi/index-2.html"
html = requests.get(press_url)
soup = BeautifulSoup(html.content, "html.parser")
# デプロイ前にタイムデルタを取る
# today = (datetime.today()).strftime('%Y年%-m月%-d日')
today = (datetime.today() - timedelta(days=1) -
timedelta(hours=6)).strftime('%Y年%-m月%-d日')
url_flake = ""
for li in soup.find(class_="list_ccc").find_all("li"):
if (today in li.text) & ("感染者の発生" in li.text) & ("愛知県職員における"
not in li.text):
url_flake = li.find("a")["href"]
if url_flake != "":
today_url = urljoin(multi_dirname(press_url, 3), url_flake)
html = requests.get(today_url)
soup = BeautifulSoup(html.content, "html.parser")
pdf_url = urljoin(
multi_dirname(press_url, 3),
soup.find(class_="detail_free").find("a")["href"])
pdf_file_path = os.path.join(
"data",
f"{str(datetime.today()-timedelta(hours=6)).split()[0]}_aichi.pdf"
)
urlretrieve(pdf_url, pdf_file_path)
tbls = camelot.read_pdf(pdf_file_path, pages='1-end')
dfs = []
for table in tbls:
df = table.df
dfs.append(df)
df_all = pandas.concat(dfs)
df_all.columns = df_all.iloc[0, :]
df_all = df_all[df_all["年代"] != "年代"]
# デプロイ前にタイムデルタを消す
# _name = str(datetime.today()).split()[0]
# _name = str(datetime.today() - timedelta(days=1)).split()[0]
# df_zentai = pandas.read_pickle(
# os.path.join("data", f"{_name}_from_sum.zip"))
df_zentai = get_last_numbers_from_posts(get_posts(tweet_number=30),
day_before=1)
df_zentai.pop("愛知県管轄")
df_zentai = pandas.DataFrame.from_dict(df_zentai, orient="index")
df_zentai.columns = ["本日"]
aichi_kobetsu = pandas.DataFrame(
collections.Counter(df_all["居住地"]).most_common())
aichi_kobetsu = aichi_kobetsu.set_index(0)
aichi_kobetsu.columns = ["本日"]
aichi_total = pandas.concat([
aichi_kobetsu, df_zentai[df_zentai.index != "愛知県"]
]).sort_values("本日", ascending=False)
ranking_text = "昨日の新型コロナウイルス感染者数ランキング\n"
rank = 0
num_prior = 0
yesterday = datetime.today() - timedelta(days=1)
for city, num in zip(aichi_total.index, aichi_total["本日"]):
if num == num_prior:
if parse_tweet(ranking_text).weightedLength > 233:
# if parse_tweet(ranking_text).weightedLength > 258:
ranking_text += f"(以下 https://narumi-midori.net/twitter_aichi_covid19/{str(yesterday.date())}.html)"
break
else:
ranking_text += f", {city}"
else:
if parse_tweet(ranking_text).weightedLength > 226:
# if parse_tweet(ranking_text).weightedLength > 251:
ranking_text += f"(以下 https://narumi-midori.net/twitter_aichi_covid19/{str(yesterday.date())}.html)"
break
else:
rank += 1
ranking_text += f"\n{rank}位 {num}人: {city}"
num_prior = num * 1
# print(ranking_text)
post(ranking_text)
with open("ranking_today.lock", "w", encoding="utf-8") as f:
f.write("")
def post_cities():
from twitter_post import get_posts, post
numbers_from_tweets = get_last_numbers_from_posts(get_posts())
info = pre_post("岡崎市", get_okazaki_info)
# print(info)
if (info["is_postable"]) & (numbers_from_tweets[info["city"]] <
info["number_today"]):
# print(info["headline"])
post(info["headline"])
else:
print(info["city"], info["is_postable"],
numbers_from_tweets[info["city"]], info["number_today"])
info = pre_post("豊橋市", get_toyohashi_info)
# print(info)
if (info["is_postable"]) & (numbers_from_tweets[info["city"]] <
info["number_today"]):
# print(info["headline"])
post(info["headline"])
else:
print(info["city"], info["is_postable"],
numbers_from_tweets[info["city"]], info["number_today"])
info = pre_post("豊田市", get_toyota_info, engine_number=2)
# print(info)
if (info["is_postable"]) & (numbers_from_tweets[info["city"]] <
info["number_today"]):
# print(info["headline"])
post(info["headline"])
else:
print(info["city"], info["is_postable"],
numbers_from_tweets[info["city"]], info["number_today"])
info = pre_post("一宮市", get_ichinomiya_info)
# print(info)
if (info["is_postable"]) & (numbers_from_tweets[info["city"]] <
info["number_today"]):
# print(info["headline"])
post(info["headline"])
else:
print(info["city"], info["is_postable"],
numbers_from_tweets[info["city"]], info["number_today"])
info = pre_post("名古屋市", get_nagoya_info)
# print(info)
if (info["is_postable"]) & (numbers_from_tweets[info["city"]] <
info["number_today"]):
# print(info["headline"])
post(info["headline"])
else:
print(info["city"], info["is_postable"],
numbers_from_tweets[info["city"]], info["number_today"])
info = pre_post("愛知県管轄自治体（名古屋市・豊橋市・豊田市・岡崎市・一宮市を除く愛知県）", get_aichi_ken_info)
# print(info)
if (info["is_postable"]) & (numbers_from_tweets[info["city"]] <
info["number_today"]):
# print(info["headline"])
post(info["headline"])
else:
print(info["city"], info["is_postable"],
numbers_from_tweets[info["city"]], info["number_today"])
# info_list = [info_okazaki, info_toyohashi, info_toyota,
# info_ichinomiya, info_nagoya, info_aichi_ken]
# for info in info_list:
# print(f"----------{info['city']}---------")
# print(numbers_from_tweets[info["city"]])
# print(info)
# if (info["is_postable"]) & (numbers_from_tweets[info["city"]] < info["number_today"]):
# print(info["headline"])
# else:
# print("Not postable", info['city'])
time.sleep(20)
numbers_from_tweets = get_last_numbers_from_posts(get_posts())
info = pre_post_zentai(get_zentai_info, numbers_from_tweets)
print("-------------全体-------------")
# post(info_zentai["headline"])
if (info["is_postable"]) & (numbers_from_tweets[info["city"]] <
info["number_today"]):
# print(info["headline"])
post(info["headline"])
else:
print(info["city"], info["is_postable"],
numbers_from_tweets[info["city"]], info["number_today"])