From c6e6b26af3581a2d54694f5e7ab66c5779e8355d Mon Sep 17 00:00:00 2001
From: Bastien Le Querrec <blq@laquadrature.net>
Date: Tue, 7 Jan 2025 12:13:54 +0100
Subject: [PATCH] pref35: utilise prefdpt
---
Attrap_pref35.py | 81 ++++++++----------------------------------------
1 file changed, 13 insertions(+), 68 deletions(-)
diff --git a/Attrap_pref35.py b/Attrap_pref35.py
index fa8fbcb..202bbf0 100644
--- a/Attrap_pref35.py
+++ b/Attrap_pref35.py
@@ -1,79 +1,24 @@
-import os
-import datetime
+from Attrap_prefdpt import Attrap_prefdpt
-from bs4 import BeautifulSoup
-from urllib.parse import unquote
-from Attrap import Attrap
+class Attrap_pref35(Attrap_prefdpt):
-
-class Attrap_pref35(Attrap):
-
- # Config
+ # Configuration de la préfecture
hostname = 'https://www.ille-et-vilaine.gouv.fr'
raa_page = f'{hostname}/Publications/Recueil-des-actes-administratifs'
- raa_page_archives = f'{hostname}/Publications/Recueil-des-actes-administratifs/Archives-des-recueils-des-actes-administratifs'
- user_agent = 'Mozilla/5.0 (Windows NT 10.0; rv:109.0) Gecko/20100101 Firefox/115.0'
full_name = 'Préfecture d\'Ille-et-Vilaine'
short_code = 'pref35'
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):
- year_pages_to_parse = []
+ # Configuration des widgets à analyser
+ year_regex = 'Recueil des actes administratifs ([0-9]{4})'
+ Attrap_prefdpt.white_card['regex']['year'] = year_regex
- # La page de l'année en cours est normalement listée sur raa_page
- year_pages = self.get_sub_pages_with_pager(
- self.raa_page,
- 'div.fr-card.fr-card--horizontal.fr-card--sm.fr-enlarge-link.fr-mb-3w 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.fr-pagination__link--lg-label',
- None,
- self.hostname
+ # On ajoute un widget de menu déroulant
+ Attrap_prefdpt.select_widgets.append(
+ Attrap_prefdpt.DptSelectWidget(
+ 'menu_deroulant',
+ regex=year_regex,
+ css_path='div.fr-select-group select#Archives-des-RAA-liste-docs.fr-select'
)
- for year_page in year_pages:
- year = Attrap.guess_date(year_page['name'], '.*([0-9]{4})').year
- if year >= self.not_before.year:
- year_pages_to_parse.append(year_page['url'])
-
- # Les URL des pages des années précédentes sont dans un menu déroulant de raa_page_archives
- page_content = self.get_page(self.raa_page_archives, 'get').content
- for option in BeautifulSoup(page_content, 'html.parser').select('div.fr-select-group select.fr-select option'):
- if not option['value'] == '':
- year = Attrap.guess_date(option.get_text().strip(), '.*([0-9]{4})').year
- if year >= self.not_before.year:
- url = option['value']
- year_pages_to_parse.append(f'{self.hostname}/{url}')
-
- elements = []
- for raa_page in year_pages_to_parse:
- page_content = self.get_page(raa_page, 'get').content
- for element in self.get_raa_elements(page_content):
- elements.append(element)
-
- self.parse_raa(elements, keywords)
- self.mailer()
-
- def get_raa_elements(self, page_content):
- elements = []
- # On charge le parser
- soup = BeautifulSoup(page_content, 'html.parser')
-
- # Pour chaque balise a, on regarde si c'est un PDF, et si oui on le
- # parse
- for a in soup.find_all('a', href=True, class_='fr-link--download'):
- if 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
+ )
--
GitLab