pref73: détecte l'URL de l'année voulue automatiquement

Attrap_pref73.py

 import os
+import re
 import datetime
 
 from bs4 import BeautifulSoup
@@ -11,16 +12,9 @@ class Attrap_pref73(Attrap):
 
     # Config
     hostname = 'https://www.savoie.gouv.fr'
-    raa_page = {
-        '2024': f'{hostname}/Publications/Recueils-hebdomadaires-et-speciaux-des-actes-administratifs/2024',
-        '2023': f'{hostname}/Publications/Recueils-hebdomadaires-et-speciaux-des-actes-administratifs/2023',
-        '2022': f'{hostname}/Publications/Recueils-hebdomadaires-et-speciaux-des-actes-administratifs/2022',
-        '2021': f'{hostname}/Publications/Recueils-hebdomadaires-et-speciaux-des-actes-administratifs/2021',
-        '2020': f'{hostname}/Publications/Recueils-hebdomadaires-et-speciaux-des-actes-administratifs/2020',
-        '2019': f'{hostname}/Publications/Recueils-hebdomadaires-et-speciaux-des-actes-administratifs/2019',
-    }
+    raa_page = f'{hostname}/Publications/Recueils-hebdomadaires-et-speciaux-des-actes-administratifs'
     user_agent = 'Mozilla/5.0 (Windows NT 10.0; rv:109.0) Gecko/20100101 Firefox/115.0'
-    full_name = 'Préfecture de la Savoie'
+    full_name = 'Savoie'
     short_code = 'pref73'
     timezone = 'Europe/Paris'
 
@@ -28,39 +22,53 @@ class Attrap_pref73(Attrap):
         super().__init__(data_dir, self.user_agent)
         self.set_sleep_time(30)
 
-    def get_raa(self, keywords):
-        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'])
+        self.page_urls_to_parse = [self.raa_page]
+        self.elements = []
 
-        # On parse les pages contenant des RAA
-        elements = []
-        print(pages_to_parse)
-        for page in pages_to_parse:
-            page_content = self.get_page(page, 'get').content
-            for element in self.get_raa_elements(page_content):
-                elements.append(element)
+    def get_raa(self, keywords):
+        while not self.page_urls_to_parse == []:
+            page_url = self.page_urls_to_parse[-1]
+            page_content = self.get_page(page_url, 'get').content  # On récupère le HTML de la page
+            self.get_year_pages(page_content)  # On cherche les cartes d'années
+            self.get_elements_pages(page_url)  # On cherche les cartes d'éléments
+            for element in self.get_raa_elements(page_content):  # On cherche les éléments
+                self.elements.append(element)
+            self.page_urls_to_parse.remove(page_url)  # On supprime la page de la liste de celles à parser
 
-        self.parse_raa(elements, keywords)
+        self.parse_raa(self.elements[::-1], keywords)
         self.mailer()
 
+    def get_year_pages(self, page_content):
+        for card in self.get_sub_pages(
+            page_content,
+            'div.fr-card.fr-card--sm.fr-card--grey.fr-enlarge-link div.fr-card__body div.fr-card__content h2.fr-card__title a',  # Pages d'années
+            self.hostname,
+            False
+        ):
+            date = Attrap.guess_date(card['name'].strip(), '([0-9]{4})').replace(day=1, month=1)
+            if date.year >= self.not_before.year:
+                self.page_urls_to_parse.append(card['url'])
+
+    def get_elements_pages(self, page_url):
+        print(page_url)
+        for card in self.get_sub_pages_with_pager(
+            page_url,
+            'div.fr-card.fr-card--horizontal.fr-card--sm.fr-enlarge-link div.fr-card__body div.fr-card__content h2.fr-card__title a.fr-card__link',  # Carte avec un élément (mois ou arrêté seul)
+            'ul.fr-pagination__list li a.fr-pagination__link.fr-pagination__link--next.fr-pagination__link--lg-label',  # Pager
+            'div.fr-card.fr-card--horizontal.fr-card--sm.fr-enlarge-link div.fr-card__body div.fr-card__content div.fr-card__end p.fr-card__detail',  # Détails (avec la date de publication)
+            self.hostname
+        ):
+            date = datetime.datetime.strptime(card['details'].replace('Publié le ', '').strip(), '%d/%m/%Y')
+            if date >= self.not_before:
+                self.page_urls_to_parse.append(card['url'])
+
     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('a.fr-link.fr-link--download'):
+        for a in soup.select('div.fr-downloads-group.fr-downloads-group--bordered ul li a,div a.fr-link.fr-link--download'):
             if a.get('href') and a['href'].endswith('.pdf'):
                 if a['href'].startswith('/'):
                     url = f"{self.hostname}{a['href']}"
@@ -68,10 +76,8 @@ class Attrap_pref73(Attrap):
                     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')
+                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)