RAAspotter_pref81.py

import os
import datetime

from bs4 import BeautifulSoup
from urllib.parse import unquote

from RAAspotter import RAAspotter


class RAAspotter_pref81(RAAspotter):

    # Config
    __HOST = 'https://www.tarn.gouv.fr'
    __RAA_PAGE = {
        'default': f'{__HOST}/Publications/RAA-Recueil-des-Actes-Administratifs/RAA',
        '2024': f'{__HOST}/Publications/RAA-Recueil-des-Actes-Administratifs/RAA/2024',
        '2023': f'{__HOST}/Publications/RAA-Recueil-des-Actes-Administratifs/RAA/2023',
        '2022': f'{__HOST}/Publications/RAA-Recueil-des-Actes-Administratifs/RAA/2022',
        '2021': f'{__HOST}/Publications/RAA-Recueil-des-Actes-Administratifs/RAA/2021',
        '2020': f'{__HOST}/Publications/RAA-Recueil-des-Actes-Administratifs/RAA/2020',
        '2019': f'{__HOST}/Publications/RAA-Recueil-des-Actes-Administratifs/RAA/2019',
    }
    __USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/115.0'
    full_name = 'Préfecture du Tarn'
    short_code = 'pref81'

    def __init__(self, data_dir):
        super().__init__(data_dir, self.__USER_AGENT)
        self.enable_tor(10)

    def get_raa(self, keywords):
        self.print_output('RAAspotter_pref81')
        self.print_output(f'Termes recherchés: {keywords}')
        self.print_output('')

        pages_to_parse = []
        if self.not_before.year <= 2024:
            pages_to_parse.append(self.__RAA_PAGE['2024'])
        if self.not_before.year <= 2023:
            pages_to_parse.append(self.__RAA_PAGE['2023'])
        if self.not_before.year <= 2022:
            pages_to_parse.append(self.__RAA_PAGE['2022'])
        if self.not_before.year <= 2021:
            pages_to_parse.append(self.__RAA_PAGE['2021'])
        if self.not_before.year <= 2020:
            pages_to_parse.append(self.__RAA_PAGE['2020'])
        if self.not_before.year <= 2019:
            pages_to_parse.append(self.__RAA_PAGE['2019'])

        sub_pages_to_parse = [self.__RAA_PAGE['default']]

        # Pour chaque année, on cherche les sous-pages de mois
        for raa_page in pages_to_parse:
            page_content = self.get_page(raa_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.__HOST,
                False
            )[::-1]

            # On regarde aussi si sur la page de l'année il n'y aurait pas un
            # RAA mal catégorisé
            for page_to_parse in self.find_raa_card(raa_page):
                sub_pages_to_parse.append(page_to_parse)

            # Pour chaque mois, on cherche les pages des RAA
            for month_page in month_pages:
                year = RAAspotter.guess_date(month_page['name'], '(.*)').year
                for page_to_parse in self.find_raa_card(month_page['url'], year):
                    sub_pages_to_parse.append(page_to_parse)
                # On ajoute aussi la page des mois à parser au cas où il y ait
                # eu une redirection vers un RAA
                sub_pages_to_parse.append(month_page['url'])

        # On parse les pages contenant des RAA
        for page in sub_pages_to_parse:
            page_content = self.get_page(page, 'get').content
            raa_elements = self.get_raa_elements(page_content)
            self.parse_raa(raa_elements, keywords.split(','))

        self.mailer()

    def find_raa_card(self, page, year=None):
        pages = []
        card_pages = self.get_sub_pages_with_pager(
            page,
            'div.fr-card__body div.fr-card__content h2.fr-card__title a.fr-card__link',
            'ul.fr-pagination__list li a.fr-pagination__link.fr-pagination__link--next',
            'div.fr-card__body div.fr-card__content div.fr-card__end p.fr-card__detail',
            self.__HOST
        )[::-1]
        for card_page in card_pages:
            # On filtre les pages de RAA ne correspondant pas à la période analysée
            guessed_date = datetime.datetime.strptime(card_page['details'].replace('Publié le ', '').strip(), '%d/%m/%Y')
            if guessed_date >= self.not_before:
                pages.append(card_page['url'])
        return pages

    def get_raa_elements(self, page_content):
        elements = []
        # On charge le parser
        soup = BeautifulSoup(page_content, 'html.parser')

        # On récupère chaque balise a
        for a in soup.select('div.fr-downloads-group.fr-downloads-group--bordered ul li a'):
            if a.get('href') and a['href'].endswith('.pdf'):
                if a['href'].startswith('/'):
                    url = f"{self.__HOST}{a['href']}"
                else:
                    url = a['href']

                url = unquote(url)
                name = a.find('span').previous_sibling.replace('Télécharger ', '').strip()
                date = datetime.datetime.strptime(a.find('span').get_text().split(' - ')[-1].strip(), '%d/%m/%Y')
                filename = url.split('/')[-1]

                raa = RAAspotter.RAA(url, date, name, filename)
                elements.append(raa)
        return elements