Attrap_pref44.py

import os
import datetime
import logging

from bs4 import BeautifulSoup
from urllib.parse import unquote

from Attrap import Attrap

logger = logging.getLogger(__name__)


class Attrap_pref44(Attrap):

    # Config
    hostname = 'https://www.loire-atlantique.gouv.fr'
    raa_page = f'{hostname}/Publications/Recueil-des-actes-administratifs-RAA-en-Loire-Atlantique'
    user_agent = 'Mozilla/5.0 (Windows NT 10.0; rv:109.0) Gecko/20100101 Firefox/115.0'
    full_name = 'Préfecture de la Loire-Atlantique'
    short_code = 'pref44'
    timezone = 'Europe/Paris'

    def __init__(self, data_dir):
        super().__init__(data_dir, self.user_agent)
        self.set_sleep_time(30)

    def get_raa(self, keywords):
        pages_to_parse = []

        # Parfois un RAA est mal catégorisé et se retrouve sur la page racine, donc on la parse
        pages_to_parse.append(self.raa_page)

        # On détermine quelles pages d'année parser
        year_pages_to_parse = []
        page_content = self.get_page(self.raa_page, 'get').content
        year_pages = self.get_sub_pages(
            page_content,
            '.fr-card.fr-card--sm.fr-card--grey.fr-enlarge-link div.fr-card__body div.fr-card__content h2.fr-card__title a',
            self.hostname,
            False
        )
        for year_page in year_pages:
            year = 9999
            try:
                year = int(year_page['name'].strip())
                if year is None:
                    year = 9999
            except Exception as exc:
                logger.warning(f"Impossible de deviner l\'année de la page {year_page['name']}")
                year = 9999

            if year >= self.not_before.year:
                year_pages_to_parse.append(year_page['url'])

                # Parfois un RAA est mal catégorisé et se retrouve sur la page de l'année, donc on la parse
                pages_to_parse.append(year_page['url'])

        # Pour chaque année, on cherche les sous-pages de mois
        month_pages_to_parse = []
        for year_page in year_pages_to_parse:
            page_content = self.get_page(year_page, 'get').content
            month_pages = self.get_sub_pages(
                page_content,
                '.fr-card.fr-card--sm.fr-card--grey.fr-enlarge-link div.fr-card__body div.fr-card__content h2.fr-card__title a',
                self.hostname,
                False
            )[::-1]

            for month_page in month_pages:
                pages_to_parse.append(month_page['url'])

        # On parse les pages sélectionnées
        elements = self.get_raa_with_pager(
            pages_to_parse,
            "ul.fr-pagination__list li a.fr-pagination__link.fr-pagination__link--next.fr-pagination__link--lg-label",
            self.hostname
        )[::-1]

        self.parse_raa(elements, keywords)
        self.mailer()

    def get_raa_elements(self, page_content):
        elements = []

        # On récupère chaque carte avec un RAA
        for card in BeautifulSoup(page_content, 'html.parser').select('div.fr-card.fr-card--horizontal div.fr-card__body div.fr-card__content'):
            # On récupère le lien
            links = card.select('h2.fr-card__title a.fr-card__link.menu-item-link')
            # On récupère la date
            dates_raw = card.select('div.fr-card__end p.fr-card__detail')

            # Si on a toutes les infos, on continue
            if links and links[0] and dates_raw and dates_raw[0]:
                a = links[0]
                date_raw = dates_raw[0]

                if a.get('href') and a['href'].endswith('.pdf'):
                    if a['href'].startswith('/'):
                        url = f"{self.hostname}{a['href']}"
                    else:
                        url = a['href']

                    url = unquote(url)
                    name = a.get_text().strip()
                    date = datetime.datetime.strptime(date_raw.get_text().replace('Publié le', '').strip(), '%d/%m/%Y')

                    raa = Attrap.RAA(url, date, name, timezone=self.timezone)
                    elements.append(raa)
        return elements