prefdpt: les regex des cartes sont configurables

Attrap_pref01.py

@@ -9,4 +9,4 @@ class Attrap_pref01(Attrap_prefdpt):
     full_name = 'Préfecture de l\'Ain'
     short_code = 'pref01'
     timezone = 'Europe/Paris'
-    year_regex = '(?:Recueil|Recueils) (?:des actes administratifs)(?:[ -])*([0-9]{4})'
+    Attrap_prefdpt.grey_card['year'] = '(?:Recueil|Recueils) (?:des actes administratifs)(?:[ -])*([0-9]{4})'

Attrap_pref02.py

@@ -9,4 +9,4 @@ class Attrap_pref02(Attrap_prefdpt):
     full_name = 'Préfecture de l\'Aisne'
     short_code = 'pref02'
     timezone = 'Europe/Paris'
-    year_regex = 'RAA [Aa]nnée ([0-9]{4})'
+    Attrap_prefdpt.grey_card['year'] = 'RAA [Aa]nnée ([0-9]{4})'

Attrap_pref11.py

@@ -9,4 +9,4 @@ class Attrap_pref11(Attrap_prefdpt):
     full_name = 'Préfecture de l\'Aude'
     short_code = 'pref11'
     timezone = 'Europe/Paris'
-    year_regex = 'Année *([0-9]{4})'
+    Attrap_prefdpt.grey_card['year'] = 'Année *([0-9]{4})'

Attrap_pref29.py

@@ -9,4 +9,4 @@ class Attrap_pref29(Attrap_prefdpt):
     full_name = 'Préfecture du Finistère'
     short_code = 'pref29'
     timezone = 'Europe/Paris'
-    year_regex = '(?:Recueils publiés en ).*([0-9]{4})'
+    Attrap_prefdpt.grey_card['year'] = '(?:Recueils publiés en ).*([0-9]{4})'

Attrap_pref49.py

@@ -9,4 +9,4 @@ class Attrap_pref49(Attrap_prefdpt):
     full_name = 'Préfecture de Maine-et-Loire'
     short_code = 'pref49'
     timezone = 'Europe/Paris'
-    year_regex = '.*([0-9]{4})'
+    Attrap_prefdpt.grey_card['year'] = '.*([0-9]{4})'

Attrap_pref50.py

@@ -9,4 +9,4 @@ class Attrap_pref50(Attrap_prefdpt):
     full_name = 'Préfecture de la Manche'
     short_code = 'pref50'
     timezone = 'Europe/Paris'
-    year_regex = '([0-9]{4})'
+    Attrap_prefdpt.grey_card['year'] = '([0-9]{4})'

Attrap_pref73.py

@@ -9,4 +9,4 @@ class Attrap_pref73(Attrap_prefdpt):
     full_name = 'Préfecture de Savoie'
     short_code = 'pref73'
     timezone = 'Europe/Paris'
-    year_regex = '([0-9]{4})'
+    Attrap_prefdpt.grey_card['year'] = '([0-9]{4})'

Attrap_pref87.py

@@ -9,4 +9,4 @@ class Attrap_pref87(Attrap_prefdpt):
     full_name = 'Préfecture de la Haute-Vienne'
     short_code = 'pref87'
     timezone = 'Europe/Paris'
-    year_regex = '([0-9]{4})'
+    Attrap_prefdpt.grey_card['year'] = '([0-9]{4})'

Attrap_prefdpt.py

@@ -11,6 +11,21 @@ class Attrap_prefdpt(Attrap):
     # Config
     user_agent = 'Mozilla/5.0 (Windows NT 10.0; rv:109.0) Gecko/20100101 Firefox/115.0'
     year_regex = '([0-9]{4})'
+    grey_card = {
+        'year': None,
+        'month': None
+    }
+    white_card = {
+        'year': None,
+        'month': None
+    }
+
+    grey_card_css_path = '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'  # Chemin CSS des cartes grises
+    white_card_title_css_path = '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'  # Chemin CSS du titre des cartes blanches
+    white_card_details_css_path = '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'  # Chemin CSS du détail des cartes blanches
+    white_card_pager_css_path = 'ul.fr-pagination__list li a.fr-pagination__link.fr-pagination__link--next.fr-pagination__link--lg-label'  # Chemin CSS du pager des cartes blanches
+
+    element_css_path = 'div.fr-downloads-group.fr-downloads-group--bordered ul li a,div a.fr-link.fr-link--download'  # Chemin CSS d'un RAA
 
     def __init__(self, data_dir):
         super().__init__(data_dir, self.user_agent)
@@ -23,36 +38,52 @@ class Attrap_prefdpt(Attrap):
         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.parse_grey_cards(page_content)  # On parse les cartes grises
+            self.parse_white_cards(page_url)  # On parse les cartes blanches
+            for element in self.get_raa_elements(page_content):  # On cherche les RAA
                 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(self.elements[::-1], keywords)
         self.mailer()
 
-    def get_year_pages(self, page_content):
+    def parse_grey_cards(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.grey_card_css_path,
             self.hostname,
             False
         ):
-            date = Attrap.guess_date(card['name'].strip(), self.year_regex).replace(day=1, month=1)
-            if date.year >= self.not_before.year:
-                self.page_urls_to_parse.append(card['url'])
+            if self.grey_card['year']:
+                date = Attrap.guess_date(card['name'].strip(), self.grey_card['year']).replace(day=1, month=1)
+                if date >= self.not_before.replace(day=1, month=1) and card['url'] not in self.page_urls_to_parse:
+                    self.page_urls_to_parse.append(card['url'])
+            if self.grey_card['month']:
+                date = Attrap.guess_date(card['name'].strip(), self.grey_card['month']).replace(day=1)
+                if date >= self.not_before.replace(day=1) and card['url'] not in self.page_urls_to_parse:
+                    self.page_urls_to_parse.append(card['url'])
 
-    def get_elements_pages(self, page_url):
+    def parse_white_cards(self, 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.white_card_title_css_path,  # Carte avec un élément (mois ou arrêté seul)
+            self.white_card_pager_css_path,  # Pager
+            self.white_card_details_css_path,  # Détails (avec la date de publication)
             self.hostname
         ):
+            # Pour les cartes blanches, on parse le titre si un regex a été spécifié
+            if self.white_card['year']:
+                date = Attrap.guess_date(card['name'].strip(), self.white_card['year']).replace(day=1, month=1)
+                if date >= self.not_before.replace(day=1, month=1) and card['url'] not in self.page_urls_to_parse:
+                    self.page_urls_to_parse.append(card['url'])
+            if self.white_card['month']:
+                date = Attrap.guess_date(card['name'].strip(), self.white_card['month']).replace(day=1)
+                if date >= self.not_before.replace(day=1) and card['url'] not in self.page_urls_to_parse:
+                    self.page_urls_to_parse.append(card['url'])
+
+            # On regarde aussi la date dans les détails de la carte
             date = datetime.datetime.strptime(card['details'].replace('Publié le ', '').strip(), '%d/%m/%Y')
-            if date >= self.not_before:
+            if date >= self.not_before and card['url'] not in self.page_urls_to_parse:
                 self.page_urls_to_parse.append(card['url'])
 
     def get_raa_elements(self, page_content):
@@ -61,8 +92,8 @@ class Attrap_prefdpt(Attrap):
         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,div a.fr-link.fr-link--download'):
-            if a.get('href') and a['href'].endswith('.pdf'):
+        for a in soup.select(self.element_css_path):
+            if a.get('href'):
                 if a['href'].startswith('/'):
                     url = f"{self.hostname}{a['href']}"
                 else: