Girls Girls ガルガルのツイートから撮可レギュレーションをパースする

glgl_regulation.py

import twitter
import json
import re
from dataclasses import dataclass


TWITTER_CONSUMER_KEY = '*'
TWITTER_CONSUMER_SECRET = '*'
TWITTER_ACCESS_TOKEN_KEY = '*'
TWITTER_ACCESS_TOKEN_SECRET = '*'
SCREEN_NAME = "Girls_Girlsinfo"
api = twitter.Api(consumer_key=TWITTER_CONSUMER_KEY,
                  consumer_secret=TWITTER_CONSUMER_SECRET,
                  access_token_key=TWITTER_ACCESS_TOKEN_KEY,
                  access_token_secret=TWITTER_ACCESS_TOKEN_SECRET,
                  )


@dataclass
class glgl_regulation():
    group: str
    event: str
    dt: str
    place: str
    regulation_comment: str


def get_tweets():
    timeline = api.GetUserTimeline(screen_name=SCREEN_NAME)
    earliest_tweet = min(timeline, key=lambda x: x.id).id
    while True:
        tweets = api.GetUserTimeline(screen_name=SCREEN_NAME,max_id=earliest_tweet,count=200)
        new_earliest = min(tweets, key=lambda x: x.id).id

        if not tweets or new_earliest == earliest_tweet:
            break
        else:
            earliest_tweet = new_earliest
            timeline += tweets

        return timeline


def parse_glgl_regulation(text: str) -> glgl_regulation:
    line_splited_tweet = text.split('\n')
    line_splited_tweet = list(filter(None, line_splited_tweet))
    group_line = []
    place_line = []
    regulation_line = []
    dt = line_splited_tweet[0]
    event = line_splited_tweet[1]
    for splited in line_splited_tweet:
        if "さんです" in splited:
            group_match = re.search('(.*)さんです.*', splited)
            group_name = group_match.group(1)
            group_line.append(group_name)
        if "in " in splited:
            place_match = re.search('in (.*)', splited)
            place_txt = place_match.group(1)
            place_line.append(place_txt)
        if "撮影" in splited or "静止画" in splited:
            regu_match = re.search('(.*)です.*', splited)
            regu_txt = regu_match.group(1)
            regulation_line.append(regu_txt)
    if len(group_line) != 1:
        print(group_line)
    if len(place_line) != 1:
        print(place_line)
    if len(regulation_line) != 1:
        regulation_line = [" ".join(regulation_line)]
    regulation = regulation_line[0]
    return glgl_regulation(group_line[0], event, dt, place_line[0], regulation_line[0])


if __name__ == "__main__":
    timeline = get_tweets()
    regulations = []
    for tweet in timeline:
        if "撮影" in tweet.text or "静止画" in tweet.text:
            regu = parse_glgl_regulation(tweet.text)
            regulations.append(regu)
    regulations = sorted(regulations, key=lambda x: x.group)
    print('| グループ名 | レギュレーション | 日時 | 場所 |')
    print('|---|---|---|---|')
    for r in regulations:
        print('| ', r.group, ' | ', r.regulation_comment, ' | ', r.dt, ' | 'k r.place, ' |')