Pref01: Added support for Prefecture 01

This implements support for prefecture 01. Since I'm developing on Mac OS, the OCR part could not be verified - but the pdfs are downloaded as expected and look the same as in the other prefecture. So it probably.

The implementation is fairly straight-forward. Only particularity is that there may be several years in a single subpage. To ensure compatibility with the minimal date, we use the lowest number occurring in the string.

Attrap_pref01.py

+import os
+import re
+import datetime
+
+from bs4 import BeautifulSoup
+from urllib.parse import unquote
+
+from Attrap import Attrap
+
+
+class Attrap_pref01(Attrap):
+
+    # Config
+    hostname = 'https://www.ain.gouv.fr'
+    raa_page = f'{hostname}/Publications/Recueil-des-actes-administratifs-RAA'
+    user_agent = 'Mozilla/5.0 (Windows NT 10.0; rv:109.0) Gecko/20100101 Firefox/115.0'
+    full_name = 'Ain'
+    short_code = 'pref01'
+    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):
+        # La préfecture de Haute-Corse organise son site avec une page dédiée à l'année N, une autre dédiée à l'année N-1,
+        # puis les années d'avant sont regroupées ensemble sur deux pages. On doit donc parser les pages jusqu'à ce qu'on ne
+        # tombe plus sur des cartes d'années.
+
+        pages_to_parse = []
+
+        page_content = self.get_page(self.raa_page, 'get').content
+
+        # iterate over years
+        for card in self.get_sub_pages(
+            page_content,
+            'div.fr-card__body div.fr-card__content h2.fr-card__title a',
+            self.hostname,
+            False
+        ):
+            # Page names contain year(s) - keep if smallest year is bigger than "not_before"
+            if min(self.get_all_numbers_from_string(card["name"])) >= self.not_before.year:
+                pages_to_parse.append(card['url'])
+
+        # Pour chaque année, on cherche les sous-pages de mois
+        month_pages_to_parse = []
+        for year_page in pages_to_parse:
+            for month_page in self.get_sub_pages_with_pager(
+                year_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',
+                None,
+                self.hostname
+            ):
+                if min(self.get_all_numbers_from_string(month_page["name"])) >= self.not_before.month:
+                    month_pages_to_parse.append(month_page['url'])
+
+        # On parse les pages contenant des RAA
+        elements = []
+        for page in month_pages_to_parse:
+            page_content = self.get_page(page, 'get').content
+            for element in self.get_raa_elements(page_content):
+                elements.append(element)
+
+        self.parse_raa(elements[::-1], keywords)
+        self.mailer()
+
+    def get_all_numbers_from_string(self, string):
+        return [float(e) for e in re.findall(r'\d+', string)]
+
+    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.hostname}{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')
+
+                raa = Attrap.RAA(url, date, name, timezone=self.timezone)
+                elements.append(raa)
+        return elements