Comparer les révisions

d9f7df28 · d9f7df28 · d9f7df28 · d9f7df28 · d9f7df28 · d9f7df28
--- a/RAAspotter_pref09.py
+++ b/RAAspotter_pref09.py
-import os
-import datetime
-
-from bs4 import BeautifulSoup
-from urllib.parse import unquote
-
-from RAAspotter import RAAspotter
-
-
-class RAAspotter_pref09(RAAspotter):
-
-    # Config
-    __HOST = 'https://www.ariege.gouv.fr'
-    __RAA_PAGE = f'{__HOST}/Publications/Recueil-des-actes-administratifs' \
-        '/Recueils-des-Actes-Administratifs-de-l-Ariege-a-partir-du-28' \
-        '-avril-2015'
-    __USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64; rv:109.0) ' \
-        'Gecko/20100101 Firefox/115.0'
-    full_name = 'Préfecture de l\'Ariège'
-    short_code = 'pref09'
-
-    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_pref09')
-        self.print_output(f'Termes recherchés: {keywords}')
-        self.print_output('')
-
-        pages_to_parse = []
-
-        # Les RAA de l'Ariège sont éparpillés sur des sous-pages par mois.
-        # Donc on parse la page principale à la recherche des sous-pages.
-        sub_pages = self.get_sub_pages_with_pager(
-            self.__RAA_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]
-
-        # On filtre par date les sous-pages pour limiter les requêtes
-        for sub_page in sub_pages:
-            guessed_date = datetime.datetime.strptime(
-                sub_page['details'].replace('Publié le ', '').strip(),
-                '%d/%m/%Y'
-            )
-            guessed_date.replace(day=1)
-            if guessed_date >= self.not_before:
-                pages_to_parse.append(sub_page['url'])
-
-        # On parse les pages contenant des RAA
-        for page in 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 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
--- a/RAAspotter_pref34.py
+++ b/RAAspotter_pref34.py
-import os
-import datetime
-
-from bs4 import BeautifulSoup
-from urllib.parse import unquote
-
-from RAAspotter import RAAspotter
-
-
-class RAAspotter_pref34(RAAspotter):
-
-    # Config
-    __HOST = 'https://www.herault.gouv.fr'
-    __RAA_PAGE = {
-        '2024': f'{__HOST}/Publications/Recueils-des-actes-administratifs'
-                '/Recueil-des-actes-administratifs-2024',
-        '2023': f'{__HOST}/Publications/Recueils-des-actes-administratifs'
-                '/Recueil-des-actes-administratifs-2023',
-        '2022': f'{__HOST}/Publications/Recueils-des-actes-administratifs'
-                '/Recueil-des-actes-administratifs-2022',
-        '2021': f'{__HOST}/Publications/Recueils-des-actes-administratifs'
-                '/Recueil-des-actes-administratifs-2021',
-        '2020': f'{__HOST}/Publications/Recueils-des-actes-administratifs'
-                '/Recueil-des-actes-administratifs-2020',
-        '2019': f'{__HOST}/Publications/Recueils-des-actes-administratifs'
-                '/Archives/Recueil-des-actes-administratifs-2019'
-    }
-    __USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 ' \
-        '(KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36'
-    full_name = 'Préfecture de l\'Hérault'
-    short_code = 'pref34'
-
-    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_pref34')
-        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'])
-
-        for raa_page in pages_to_parse:
-            page_content = self.get_page(raa_page, 'get').content
-            raa_elements = self.get_raa_elements(page_content)
-            self.parse_raa(raa_elements, keywords.split(','))
-
-        self.mailer()
-
-    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'):
-            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
--- a/RAAspotter_pref35.py
+++ b/RAAspotter_pref35.py
-import os
-import datetime
-
-from bs4 import BeautifulSoup
-from urllib.parse import unquote
-
-from RAAspotter import RAAspotter
-
-
-class RAAspotter_pref35(RAAspotter):
-
-    # Config
-    __HOST = 'https://www.ille-et-vilaine.gouv.fr'
-    __RAA_PAGE = [
-        f'{__HOST}/Publications/Recueil-des-actes-administratifs/Recueil-des-'
-        'actes-administratifs-2024',
-        f'{__HOST}/Publications/Recueil-des-actes-administratifs/Archives-'
-        'des-recueils-des-actes-administratifs/Recueil-des-actes-'
-        'administratifs-2023',
-        f'{__HOST}/Publications/Recueil-des-actes-administratifs/Archives-'
-        'des-recueils-des-actes-administratifs/Recueil-des-actes-'
-        'administratifs-2022',
-        f'{__HOST}/Publications/Recueil-des-actes-administratifs/Archives-'
-        'des-recueils-des-actes-administratifs/Recueil-des-actes-'
-        'administratifs-2021',
-        f'{__HOST}/Publications/Recueil-des-actes-administratifs/Archives-'
-        'des-recueils-des-actes-administratifs/Recueil-des-actes-'
-        'administratifs-2020',
-        f'{__HOST}/Publications/Recueil-des-actes-administratifs/Archives-'
-        'des-recueils-des-actes-administratifs/Recueil-des-actes-'
-        'administratifs-2019'
-    ]
-    __USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 ' \
-        '(KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36'
-    full_name = 'Préfecture d\'Ille-et-Vilaine'
-    short_code = 'pref35'
-
-    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_pref35')
-        self.print_output(f'Termes recherchés: {keywords}')
-        self.print_output('')
-
-        for raa_page in self.__RAA_PAGE:
-            page_content = self.get_page(raa_page, 'get').content
-            raa_elements = self.get_raa_elements(page_content)
-            self.parse_raa(raa_elements, keywords.split(','))
-
-        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.__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
--- a/RAAspotter_pref38.py
+++ b/RAAspotter_pref38.py
-import os
-import datetime
-import logging
-
-from bs4 import BeautifulSoup
-from urllib.parse import unquote
-
-from RAAspotter import RAAspotter
-
-logger = logging.getLogger(__name__)
-
-
-class RAAspotter_pref38(RAAspotter):
-
-    # Config
-    __HOST = 'https://www.isere.gouv.fr'
-    __RAA_PAGE = {
-        '2024': f'{__HOST}/Publications/RAA-Recueil-des-actes-administratifs'
-        '/Recueils-des-Actes-Administratifs-de-la-prefecture-de-l-Isere-2024',
-        '2023': f'{__HOST}/Publications/RAA-Recueil-des-actes-administratifs'
-        '/Recueils-des-Actes-Administratifs-de-la-prefecture-de-l-Isere-2023',
-        '2022': f'{__HOST}/Publications/RAA-Recueil-des-actes-administratifs'
-        '/Archives/Recueils-des-Actes-Administratifs-de-la-prefecture-de-l-'
-        'Isere-2022',
-        '2021': f'{__HOST}/Publications/RAA-Recueil-des-actes-administratifs'
-        '/Archives/Archives-des-recueils-des-actes-administratifs-de-la-'
-        'prefecture-de-l-Isere-2021/Recueils-des-Actes-Administratifs-de-la-'
-        'prefecture-de-l-Isere-2021',
-        '2020': f'{__HOST}/Publications/RAA-Recueil-des-actes-administratifs'
-        '/Archives/Archives-des-recueils-des-actes-administratifs-de-la-'
-        'prefecture-de-l-Isere-2020/Recueils-des-Actes-Administratifs-de-la-'
-        'Prefecture-de-l-Isere-2020',
-        '2019': f'{__HOST}/Publications/RAA-Recueil-des-actes-administratifs'
-        '/Archives/Archives-des-Recueils-des-Actes-Administratifs-de-la-'
-        'prefecture-de-l-Isere-2019/Archives-des-Recueils-des-Actes-'
-        'Administratifs-de-la-prefecture-de-l-Isere-2019'
-    }
-    __USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64; rv:109.0) ' \
-        'Gecko/20100101 Firefox/115.0'
-    full_name = 'Préfecture de l\'Isère'
-    short_code = 'pref38'
-
-    def __init__(self, data_dir):
-        super().__init__(data_dir, self.__USER_AGENT)
-        self.enable_tor(20)
-
-    def get_raa(self, keywords):
-        self.print_output('RAAspotter_pref38')
-        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'])
-
-        for raa_page in pages_to_parse:
-            page_content = self.get_page(raa_page, 'get').content
-            raa_elements = self.get_raa_elements(page_content, raa_page)
-            self.parse_raa(raa_elements, keywords.split(','))
-        self.mailer()
-
-    def get_raa_elements(self, page_content, raa_page):
-        elements = []
-        # On charge le parser
-        soup = BeautifulSoup(page_content, 'html.parser')
-
-        # On récupère le select qui contient la liste des RAA
-        select_list = soup.select('select#-liste-docs')[0]
-        # On analyse chaque résultat
-        for option in select_list.find_all('option'):
-            if not option['value'] == "":
-                # On estime la date à partir du nom de fichier
-                guessed_date = RAAspotter.guess_date(
-                    option['title'],
-                    '.* n°[ 0-9]* du ([0-9]*(?:er)? [a-zéû]* [0-9]*)'
-                )
-
-                # Si la date estimée correspond à la plage d'analyse, on
-                # demande au serveur les détails du RAA
-                if guessed_date >= self.not_before:
-                    page_content = self.get_page(
-                        raa_page,
-                        'post',
-                        {
-                            '-liste-docs': option['value']
-                        }
-                    ).content
-
-                    # On parse la page de détails pour obtenir les propriétés
-                    # du RAA
-                    soup = BeautifulSoup(page_content, 'html.parser')
-                    a = soup.select(
-                        'div.liste_deroulante a.fr-link.fr-link--download'
-                    )[0]
-
-                    # Si la page contient une balise a qui renvoie vers un pdf,
-                    # c'est qu'on a obtenu les détails du RAA demandé, donc
-                    # on le parse
-                    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
--- a/RAAspotter_pref59.py
+++ b/RAAspotter_pref59.py
-import os
-import datetime
-import dateparser
-import logging
-
-from bs4 import BeautifulSoup
-from urllib.parse import unquote
-
-from RAAspotter import RAAspotter
-
-logger = logging.getLogger(__name__)
-
-
-class RAAspotter_pref59(RAAspotter):
-
-    # Config
-    __HOST = 'https://www.nord.gouv.fr'
-    __RAA_PAGE = {
-        '2024': f'{__HOST}/Publications/Recueils-des-actes-administratifs'
-        '/RAA-du-departement-du-Nord/2024',
-        '2023': f'{__HOST}/Publications/Recueils-des-actes-administratifs'
-        '/RAA-du-departement-du-Nord/2023',
-        '2022': f'{__HOST}/Publications/Recueils-des-actes-administratifs'
-        '/RAA-du-departement-du-Nord/2022',
-        '2021': f'{__HOST}/Publications/Recueils-des-actes-administratifs'
-        '/RAA-du-departement-du-Nord/2021',
-        '2020': f'{__HOST}/Publications/Recueils-des-actes-administratifs'
-        '/RAA-du-departement-du-Nord/2020',
-        '2019': f'{__HOST}/Publications/Recueils-des-actes-administratifs'
-        '/RAA-du-departement-du-Nord/2019'
-    }
-    __USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64; rv:109.0) ' \
-        'Gecko/20100101 Firefox/115.0'
-    full_name = 'Préfecture du Nord'
-    short_code = 'pref59'
-
-    def __init__(self, data_dir):
-        super().__init__(data_dir, self.__USER_AGENT)
-        self.enable_tor(20)
-
-    def get_raa(self, keywords):
-        self.print_output('RAAspotter_pref59')
-        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'])
-
-        for raa_page in pages_to_parse:
-            page_content = self.get_page(raa_page, 'get').content
-            sub_pages = self.get_sub_pages(
-                page_content,
-                "div.fr-card__body div.fr-card__content h2.fr-card__title a",
-                self.__HOST,
-                True
-            )
-            for sub_page in sub_pages[::-1]:
-                sub_page_content = self.get_page(
-                    sub_page['url'],
-                    'get'
-                ).content
-                sub_raa_elements = self.get_raa_elements(sub_page_content)
-                self.parse_raa(sub_raa_elements, keywords.split(','))
-        self.mailer()
-
-    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'):
-            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
--- a/RAAspotter_pref62.py
+++ b/RAAspotter_pref62.py
-import os
-import datetime
-
-from bs4 import BeautifulSoup
-from urllib.parse import unquote
-
-from RAAspotter import RAAspotter
-
-
-class RAAspotter_pref62(RAAspotter):
-
-    # Config
-    __HOST = 'https://www.pas-de-calais.gouv.fr'
-    __RAA_PAGE = {
-        '2024': [
-            f'{__HOST}/Publications/Recueil-des-actes-administratifs'
-            '/2024-Recueils-des-actes-administratifs'
-        ],
-        '2023': [
-            f'{__HOST}/Publications/Recueil-des-actes-administratifs'
-            '/2023-Recueils-des-actes-administratifs',
-            f'{__HOST}/Publications/Recueil-des-actes-administratifs'
-            '/2023-Recueils-speciaux-des-actes-administratifs'
-        ],
-        '2022': [
-            f'{__HOST}/Publications/Recueil-des-actes-administratifs'
-            '/2022-Recueils-des-Actes-Administratifs',
-            f'{__HOST}/Publications/Recueil-des-actes-administratifs'
-            '/2022-Recueils-Speciaux-des-Actes-Administratifs'
-        ],
-        '2021': [
-            f'{__HOST}/Publications/Recueil-des-actes-administratifs'
-            '/2021-Recueils-des-actes-administratifs',
-            f'{__HOST}/Publications/Recueil-des-actes-administratifs'
-            '/2021-Recueils-speciaux-des-actes-administratifs'
-        ],
-        '2020': [
-            f'{__HOST}/Publications/Recueil-des-actes-administratifs'
-            '/2020-Recueils-des-actes-administratifs',
-            f'{__HOST}/Publications/Recueil-des-actes-administratifs'
-            '/2020-Recueils-speciaux-des-actes-administratifs'
-        ],
-        '2019': [
-            f'{__HOST}/Publications/Recueil-des-actes-administratifs'
-            '/2019-Recueil-des-actes-administratifs',
-            f'{__HOST}/Publications/Recueil-des-actes-administratifs'
-            '/2019-Recueils-speciaux-des-actes-administratifs'
-        ]
-    }
-    __USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64; rv:109.0) ' \
-        'Gecko/20100101 Firefox/115.0'
-    full_name = 'Préfecture du Pas-de-Calais'
-    short_code = 'pref62'
-
-    def __init__(self, data_dir):
-        super().__init__(data_dir, self.__USER_AGENT)
-        self.enable_tor(20)
-
-    def get_raa(self, keywords):
-        self.print_output('RAAspotter_pref62')
-        self.print_output(f'Termes recherchés: {keywords}')
-        self.print_output('')
-
-        pages_to_parse = []
-        if self.not_before.year <= 2024:
-            for page in self.__RAA_PAGE['2024']:
-                pages_to_parse.append(page)
-        if self.not_before.year <= 2023:
-            for page in self.__RAA_PAGE['2023']:
-                pages_to_parse.append(page)
-        if self.not_before.year <= 2022:
-            for page in self.__RAA_PAGE['2022']:
-                pages_to_parse.append(page)
-        if self.not_before.year <= 2021:
-            for page in self.__RAA_PAGE['2021']:
-                pages_to_parse.append(page)
-        if self.not_before.year <= 2020:
-            for page in self.__RAA_PAGE['2020']:
-                pages_to_parse.append(page)
-        if self.not_before.year <= 2019:
-            for page in self.__RAA_PAGE['2019']:
-                pages_to_parse.append(page)
-
-        for raa_page in pages_to_parse:
-            page_content = self.get_page(
-                raa_page,
-                'get'
-            ).content
-            raa_elements = self.get_raa_elements(page_content)
-            self.parse_raa(raa_elements, keywords.split(','))
-        self.mailer()
-
-    def get_raa_elements(self, page_content):
-        elements = []
-        # On charge le parser
-        soup = BeautifulSoup(page_content, 'html.parser')
-
-        # On récupère le div qui contient la liste des RAA
-        cards = soup.select(
-            'div.fr-downloads-group.fr-downloads-group--bordered'
-        )[0]
-        # On analyse chaque balise a dans ce div
-        for a in cards.find_all('a', href=True):
-            if 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[::-1]
--- a/RAAspotter_pref65.py
+++ b/RAAspotter_pref65.py
-import os
-import datetime
-
-from bs4 import BeautifulSoup
-from urllib.parse import unquote
-
-from RAAspotter import RAAspotter
-
-
-class RAAspotter_pref65(RAAspotter):
-
-    # Config
-    __HOST = 'https://www.hautes-pyrenees.gouv.fr'
-    __RAA_PAGE = {
-        '2024': f'{__HOST}/Publications/Recueil-d-actes-administratifs'
-        '/RAA-2024',
-        '2023': f'{__HOST}/Publications/Recueil-d-actes-administratifs'
-        '/RAA-2023',
-        '2022': f'{__HOST}/Publications/Recueil-d-actes-administratifs'
-        '/RAA-2022',
-        '2021': f'{__HOST}/Publications/Recueil-d-actes-administratifs'
-        '/RAA-2021',
-        '2020': f'{__HOST}/Publications/Recueil-d-actes-administratifs'
-        '/RAA-2020',
-        '2019': f'{__HOST}/Publications/Recueil-d-actes-administratifs'
-        '/RAA-2019'
-    }
-    __USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64; rv:109.0) ' \
-        'Gecko/20100101 Firefox/115.0'
-    full_name = 'Préfecture des Hautes-Pyrénées'
-    short_code = 'pref65'
-
-    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_pref65')
-        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'])
-
-        for raa_page in pages_to_parse:
-            page_content = self.get_page(raa_page, 'get').content
-            raa_elements = self.get_raa_elements(page_content)
-            self.parse_raa(raa_elements, keywords.split(','))
-
-        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.select('a.fr-link.fr-link--download'):
-            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
--- a/RAAspotter_pref81.py
+++ b/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
--- a/RAAspotter_pref83.py
+++ b/RAAspotter_pref83.py
-import os
-import datetime
-
-from bs4 import BeautifulSoup
-from urllib.parse import unquote
-
-from RAAspotter import RAAspotter
-
-
-class RAAspotter_pref83(RAAspotter):
-
-    # Config
-    __HOST = 'https://www.var.gouv.fr'
-    __RAA_PAGE = {
-        '2024': f'{__HOST}/Publications/RAA-Recueil-des-actes-administratifs'
-        '/Recueil-des-actes-administratifs-2024',
-        '2023': f'{__HOST}/Publications/RAA-Recueil-des-actes-administratifs'
-        '/Recueil-des-actes-administratifs-2023',
-        '2022': f'{__HOST}/Publications/RAA-Recueil-des-actes-administratifs'
-        '/Recueil-des-actes-administratifs-2022',
-        '2021': f'{__HOST}/Publications/RAA-Recueil-des-actes-administratifs'
-        '/Recueil-des-actes-administratifs-2021',
-        '2020': f'{__HOST}/Publications/RAA-Recueil-des-actes-administratifs'
-        '/Recueil-des-actes-administratifs-2020',
-        '2019': f'{__HOST}/Publications/RAA-Recueil-des-actes-administratifs'
-        '/Recueil-des-actes-administratifs-2019'
-    }
-    __USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64; rv:109.0) ' \
-        'Gecko/20100101 Firefox/115.0'
-    full_name = 'Préfecture du Var'
-    short_code = 'pref83'
-
-    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_pref83')
-        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 = []
-
-        # Pour chaque année, on cherche les sous-pages de mois
-        for raa_page in pages_to_parse:
-            sub_pages_to_parse.append(raa_page)
-            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]
-            for month_page in month_pages:
-                sub_pages_to_parse.append(month_page['url'])
-
-        # On parse les pages contenant des RAA
-        elements = self.get_raa_with_pager(
-            sub_pages_to_parse[::-1],
-            ".fr-pagination__link.fr-pagination__link--next",
-            self.__HOST
-        )
-        self.parse_raa(elements, keywords.split(','))
-
-        self.mailer()
-
-    def get_raa_elements(self, page_content):
-        elements = []
-        # On charge le parser
-        soup = BeautifulSoup(page_content, 'html.parser')
-
-        # On récupère chaque section contenant un RAA
-        cards = soup.select(
-            'div.fr-card__body div.fr-card__content '
-            'h2.fr-card__title a.fr-card__link.menu-item-link'
-        )
-        for a in cards:
-            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.get_text().strip()
-                date = datetime.datetime.strptime(
-                    a['title'].split(' - ')[-1].strip(),
-                    '%d/%m/%Y'
-                )
-                filename = url.split('/')[-1]
-
-                raa = RAAspotter.RAA(url, date, name, filename)
-                elements.append(raa)
-        return elements
--- a/README.md
+++ b/README.md
--- a/cli.py
+++ b/cli.py
--- a/misc/download-all-from-s3.sh
+++ b/misc/download-all-from-s3.sh
--- a/misc/download-from-s3.sh
+++ b/misc/download-from-s3.sh
+#!/usr/bin/env bash
+
+set -e
+
+pref="${1}"
+s3_key="${2}"
+s3_secret="${3}"
+s3_host="${4}"
+s3_bucket="${5}"
+dest="${6}"
+
+if test -z "$pref" || test -z "$s3_key" || test -z "$s3_secret" || test -z "$s3_host" || test -z "$s3_bucket" || test -z "$dest"; then
+	echo "Usage: ${0} <pref> <s3_key> <s3_secret> <s3_host> <s3_bucket> <dest>"
+	exit 1
+fi
+
+dest=$(realpath "${dest}")
+
+mkdir -p "${dest}/"
+cd "${dest}/"
+
+file="${pref}.zip"
+
+echo "Downloading ${pref}..."
+
+ressource="/${s3_bucket}/${file}"
+content_type="application/octet-stream"
+date=$(date --utc -R)
+
+signature=$(echo -en "GET\n\n${content_type}\n${date}\n${ressource}" | openssl sha1 -hmac "${s3_secret}" -binary | base64)
+
+curl -X GET \
+	 --silent \
+	 -H "Date: ${date}" \
+	 -H "Content-Type: ${content_type}" \
+	 -H "Authorization: AWS ${s3_key}:${signature}" \
+	 "${s3_host}${ressource}" \
+	 -o "${file}"
+
+unzip -o "${file}" > /dev/null
+rm "${file}"
--- a/misc/fix-pref75-prefidf-url.py
+++ b/misc/fix-pref75-prefidf-url.py
--- a/misc/update_metadata_format.py
+++ b/misc/update_metadata_format.py
--- a/misc/upload-all-to-s3.sh
+++ b/misc/upload-all-to-s3.sh
--- a/misc/upload-to-s3.sh
+++ b/misc/upload-to-s3.sh
--- a/requirements.txt
+++ b/requirements.txt
 beautifulsoup4>=4.12.3
 dateparser>=1.2.0
+ftfy>=6.2.0
 Mastodon.py>=1.8.1
-pdfminer.six>=20231228
+ocrmypdf<16.4.0
 pycodestyle>=2.11.1
+pypdf>=4.2.0
+pytz>=2024.2
 PyVirtualDisplay>=3.0
 requests>=2.31.0
 selenium>=4.19.0
-stem>=1.8.2
No results found