feedToots.py

import feedparser
from datetime import datetime, timedelta, timezone
from dateutil import parser

import requests
from bs4 import BeautifulSoup
import re

import urllib.parse

from mastodon import Mastodon

mastodon = Mastodon(
    access_token = '----',
    api_base_url = 'https://mastodon.tetaneutral.net'
)

pastHour = datetime.now(timezone.utc) - timedelta(hours=1)
d = feedparser.parse('https://www.theguardian.com/international/rss')

for entry in d.entries:
    if parser.parse(entry.published) > pastHour:
        r = requests.get(entry.link)
        page = r.text
        soup = BeautifulSoup(page, 'lxml')
        paras = soup.find_all('p')
        top_fake = False
        top_real = False
        for para in paras:
            text = para.get_text()
            if len(text) > 300 and len(text) < 400:
                resp = requests.get(url="http://localhost:8080/?" +  urllib.parse.quote(text))
                data = resp.json()
                if top_real == False or top_real['real_probability'] < data['real_probability']:
                    top_real = data
                    top_real['text'] = text
                if top_fake == False or top_fake['fake_probability'] < data['fake_probability']:
                    top_fake = data
                    top_fake['text'] = text

        if top_fake != False and top_fake['fake_probability'] > 0.9:
            print(top_fake)
            mastodon.toot(top_fake['text'] + " - source: " + entry.link + " - #gpt2Detector #probablyGenerated #score0" + str(round(top_fake['fake_probability']*100)) + " #cdl #capitoledulibre")