From dc4f709eeafe496b512c476ed71e3f24e2f2408e Mon Sep 17 00:00:00 2001
From: Aurel <aurel.pere@ikmail.com>
Date: Sun, 9 Jun 2024 17:49:08 +0200
Subject: [PATCH] update with print and download at the end, still testing
---
Attrap.py | 27 ++++++++++++++++-----
Attrap_pref12.py | 61 ++++++++++++++++++++++++++++++------------------
Attrap_pref40.py | 6 +++--
Attrap_pref47.py | 6 +++--
4 files changed, 67 insertions(+), 33 deletions(-)
diff --git a/Attrap.py b/Attrap.py
index 7977e5d..91f2c4c 100644
--- a/Attrap.py
+++ b/Attrap.py
@@ -392,7 +392,7 @@ class Attrap:
if page.status_code == 429:
logger.warning('Erreur 429 Too Many Requests reçue, temporisation...')
self.tor_get_new_id()
- time.sleep(55)
+ time.sleep(60)
return self.get_page(url, method, data)
if self.tor_enabled:
@@ -415,16 +415,31 @@ class Attrap:
self.user_agent = user_agent
self.session.headers.update({'User-Agent': self.user_agent})
- def download_file(self, raa):
+ def download_file(self, raa, overwrite=True):
try:
os.makedirs(
os.path.dirname(f'{self.data_dir}/raa/{raa.get_sha256()}.pdf'),
exist_ok=True
)
- file = self.get_page(raa.url, 'get')
- f = open(f'{self.data_dir}/raa/{raa.get_sha256()}.pdf', 'wb')
- f.write(file.content)
- f.close()
+ # content-length 114 erronée sur test fichier
+ # response = requests.head(raa.url)
+ # size = response.headers.get('content-length')#length 9549465
+ # if size is not None:
+ # print(size)
+
+ if overwrite and os.path.isfile(f'{self.data_dir}/raa/{raa.get_sha256()}.pdf'):
+ print(f'file already present on disk, overwriting {self.data_dir}/raa/{raa.get_sha256()}.pdf for {raa}')
+ #if size!=os.path.getsize(f'{self.data_dir}/raa/{raa.get_sha256()}.pdf')
+ file = self.get_page(raa.url, 'get')
+ f = open(f'{self.data_dir}/raa/{raa.get_sha256()}.pdf', 'wb')
+ f.write(file.content)
+ f.close()
+ elif not os.path.isfile(f'{self.data_dir}/raa/{raa.get_sha256()}.pdf'):
+ print(f'file not present on disk, downloading {self.data_dir}/raa/{raa.get_sha256()}.pdf for {raa}')
+ file = self.get_page(raa.url, 'get')
+ f = open(f'{self.data_dir}/raa/{raa.get_sha256()}.pdf', 'wb')
+ f.write(file.content)
+ f.close()
except (requests.exceptions.ConnectionError,
requests.exceptions.ChunkedEncodingError):
logger.warning(f'ATTENTION: la connexion a été interrompue pendant le téléchargement de {raa.url}, nouvelle tentative...')
diff --git a/Attrap_pref12.py b/Attrap_pref12.py
index 4d4d48f..7c1fc06 100644
--- a/Attrap_pref12.py
+++ b/Attrap_pref12.py
@@ -1,13 +1,13 @@
import os
import datetime
-
+import time
from bs4 import BeautifulSoup
from urllib.parse import unquote
import re
from Attrap import Attrap
-class Attrap_pref46(Attrap):
+class Attrap_pref12(Attrap):
# Config
__HOST = 'https://www.aveyron.gouv.fr'
@@ -19,15 +19,13 @@ class Attrap_pref46(Attrap):
def __init__(self, data_dir):
super().__init__(data_dir, self.__USER_AGENT)
self.enable_tor(10)
- self.not_before = datetime.datetime(2020, 1, 1)# avant 2020 page "archives"
+ self.not_before = datetime.datetime(2020, 1, 1) # avant 2020 page "archives"
#supercharge pour classes css avec whitespaces
def get_sub_pages_with_pager(self, page, sub_page_element, pager_element, details_element, host):
pages = []
page_content = self.get_page(page, 'get').content
-
# On initialise le parser
soup = BeautifulSoup(page_content, 'html.parser')
-
# On recherche les sous-pages
sub_pages = soup.select(sub_page_element)
sub_pages_details = None
@@ -46,7 +44,6 @@ class Attrap_pref46(Attrap):
page['details'] = sub_pages_details[i].get_text().strip()
pages.append(page)
i = i + 1
-
# On recherche un pager, et si on le trouve on le suit
# modif ici, le parametre pager_element ne doit contenir
# que le contenu de la classe
@@ -62,36 +59,51 @@ class Attrap_pref46(Attrap):
host
):
pages.append(sub_page)
-
return pages
-
def get_raa(self, keywords):
elements = []
page_content = self.get_page(self.__RAA_PAGE, 'get').content
soup = BeautifulSoup(page_content, 'html.parser')
#selection des grey cards
+ print(f"not_before={self.not_before.year}")
for a in soup.select('div.fr-card--grey div.fr-card__body div.fr-card__content h2.fr-card__title a'):
- #regular
- if Attrap.guess_date(f'{a.get_text().strip()}', '([0-9]{4}).*').year >= self.not_before.year and "Archives" not in f'{a.get_text().strip()}':
- page_content = self.get_page(f"{self.__HOST}{a['href']}", 'get').content
- for sub_page in self.get_sub_pages_with_pager(f"{self.__HOST}{a['href']}", 'div.fr-card__body div.fr-card__content h2.fr-card__title a', 'fr-pagination__link fr-pagination__link--next fr-pagination__link--lg-label', None, self.__HOST):
- sub_page_content = self.get_page(sub_page['url'], 'get').content
- for element in self.get_raa_elements(sub_page_content):
- elements.append(element)
- #archives
- elif self.not_before.year<2021 and "Archives" in f'{a.get_text().strip()}':
+ #archives links
+ if self.not_before.year<2021 and "Archives" in f'{a.get_text().strip()}':
+ print(f"""########################\n
+archives links\n
+{a.get_text().strip()}
+###########################""")
page_content = self.get_page(f"{self.__HOST}{a['href']}", 'get').content
for sub_page in self.get_sub_pages(page_content,
'div.fr-card__body div.fr-card__content h2.fr-card__title a',
self.__HOST,
True):
+ print(f"retrieving {sub_page['url']}")
sub_page_content = self.get_page(sub_page['url'], 'get').content
- for a in sub_page_content.select('div.fr-card__body div.fr-card__content h2.fr-card__title a'):
+ subsoup = BeautifulSoup(sub_page_content, 'html.parser')
+ for a in subsoup.select('div.fr-card__body div.fr-card__content h2.fr-card__title a'):
sub_sub_page_content = self.get_page(a['url'], 'get').content
for element in self.get_raa_elements(sub_sub_page_content):
+ print(f"appending {element}")
elements.append(element)
+ #regular links
+ if Attrap.guess_date(f'{a.get_text().strip()}', '([0-9]{4}).*').year >= self.not_before.year and "Archives" not in f'{a.get_text().strip()}':
+ print(f"""########################\n
+regular links\n
+{a.get_text().strip()}
+###########################""")
+ page_content = self.get_page(f"{self.__HOST}{a['href']}", 'get').content
+ for sub_page in self.get_sub_pages_with_pager(f"{self.__HOST}{a['href']}", 'div.fr-card__body div.fr-card__content h2.fr-card__title a', 'fr-pagination__link fr-pagination__link--next fr-pagination__link--lg-label', None, self.__HOST):
+ print(f"retrieving {sub_page['url']}")
+ sub_page_content = self.get_page(sub_page['url'], 'get').content
+ for element in self.get_raa_elements(sub_page_content):
+ print(f"appending {element}")
+ elements.append(element)
#selection des "spécials"
for div in soup.select("div.fr-card.fr-card--horizontal.fr-card--sm.fr-enlarge-link.fr-mb-3w"):
+ print("""########################\n
+specials links\n
+###########################""")
for a in div.select("div.fr-card__body div.fr-card__content h2.fr-card__title a"):
print(a)
search_pattern=re.search('(?<=Publié le).*',f'{a.parent.parent.get_text()}')
@@ -104,7 +116,12 @@ class Attrap_pref46(Attrap):
True):
sub_page_content = self.get_page(sub_page['url'], 'get').content
for element in self.get_raa_elements(sub_page_content):
+ print(f"appending {element}")
elements.append(element)
+ for raa in elements:
+ print(f"downloading {raa}")
+ self.download_file(raa,overwrite=False)
+ time.sleep(14)
#bug sur ocrmypdf sur mon ubuntu 20.04 (test avec arch prochainement)
#sur --invalidate-digital-signatures bien que dans la doc
#ici https://ocrmypdf.readthedocs.io/en/latest/pdfsecurity.html
@@ -118,21 +135,19 @@ class Attrap_pref46(Attrap):
# Pour chaque balise a, on regarde si c'est un PDF, et si oui on le
# parse
- print(soup.find_all("a",{"id":'class="fr-link'}))
- print(len(soup.find_all("a",{"id":'class="fr-link'})))
+ #print(soup.find_all("a",{"id":'class="fr-link'}))
+ #print(len(soup.find_all("a",{"id":'class="fr-link'})))
for a in soup.find_all("a",{"id":'class="fr-link'}):
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')
raa = Attrap.RAA(url, date, name)
- self.download_file(raa)
elements.append(raa)
print(elements)
return elements
-
+Attrap_pref12('test').get_raa('algorithmes')
diff --git a/Attrap_pref40.py b/Attrap_pref40.py
index 51f882b..4876fc3 100644
--- a/Attrap_pref40.py
+++ b/Attrap_pref40.py
@@ -69,6 +69,7 @@ class Attrap_pref40(Attrap):
elements = []
page_content = self.get_page(self.__RAA_PAGE, 'get').content
soup = BeautifulSoup(page_content, 'html.parser')
+ print(f"not_before={self.not_before.year}")
for a in soup.select('div.fr-card__body div.fr-card__content h2.fr-card__title a'):
# Annee archivees
if self.not_before.year<2016 and "antérieures" in a.get_text().strip():
@@ -87,7 +88,9 @@ class Attrap_pref40(Attrap):
sub_page_content = self.get_page(sub_page['url'], 'get').content
for element in self.get_raa_elements(sub_page_content):
elements.append(element)
-
+ for raa in elements:
+ print(f'downloading {raa}')
+ self.download_file(raa, overwriting=False)
#bug sur ocrmypdf sur mon ubuntu 20.04 (test avec arch prochainement)
#sur --invalidate-digital-signatures bien que dans la doc
#ici https://ocrmypdf.readthedocs.io/en/latest/pdfsecurity.html
@@ -112,7 +115,6 @@ class Attrap_pref40(Attrap):
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)
- self.download_file(raa)
elements.append(raa)
print(elements)
return elements
diff --git a/Attrap_pref47.py b/Attrap_pref47.py
index c066f3b..a163094 100644
--- a/Attrap_pref47.py
+++ b/Attrap_pref47.py
@@ -71,6 +71,7 @@ class Attrap_pref47(Attrap):
elements = []
page_content = self.get_page(self.__RAA_PAGE, 'get').content
soup = BeautifulSoup(page_content, 'html.parser')
+ print(f"not_before={self.not_before.year}")
for a in soup.select('div.fr-card__body div.fr-card__content h2.fr-card__title a'):
# Annees sans pager
if Attrap.guess_date(f'{self.__HOST}{a.get_text().strip()}', '([0-9]{4}).*').year >= self.not_before.year:
@@ -89,7 +90,9 @@ class Attrap_pref47(Attrap):
# sub_page_content = self.get_page(sub_page['url'], 'get').content
# for element in self.get_raa_elements(sub_page_content):
# elements.append(element)
-
+ for raa in elements:
+ print(f"downloading {raa}")
+ self.download_file(raa, overwriting=False)
#bug sur ocrmypdf sur mon ubuntu 20.04 (test avec arch prochainement)
#sur --invalidate-digital-signatures bien que dans la doc
#ici https://ocrmypdf.readthedocs.io/en/latest/pdfsecurity.html
@@ -116,7 +119,6 @@ class Attrap_pref47(Attrap):
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)
- self.download_file(raa)
elements.append(raa)
print(elements)
return elements
--
GitLab