81 lines
3.3 KiB
Python
Executable File
81 lines
3.3 KiB
Python
Executable File
#!/usr/bin/env python3
|
|
|
|
import warnings
|
|
warnings.filterwarnings("ignore", category=DeprecationWarning)
|
|
|
|
import re
|
|
import requests
|
|
import logging
|
|
|
|
from mdf import retronews
|
|
from argparse import ArgumentParser
|
|
|
|
|
|
|
|
if __name__ == '__main__':
|
|
parser = ArgumentParser()
|
|
parser.add_argument('--url', type=str, required=True)
|
|
parser.add_argument('--output', type=str, required=True,
|
|
help='output directory')
|
|
parser.add_argument('--merge-threads', default=retronews.TILE_MERGING_POOL_SIZE, type=int)
|
|
parser.add_argument('--fetch-threads', default=retronews.PAGE_FETCHING_POOL_SIZE, type=int)
|
|
parser.add_argument('--continue-prev', action='store_true',
|
|
help='keep scrapping backwards in time')
|
|
parser.add_argument('--continue-next', action='store_true',
|
|
help='keep scrapping forwards in time')
|
|
parser.add_argument('--only-fetch', action='store_true',
|
|
help='only fetch magazine tiles and exit, do not merge anything')
|
|
parser.add_argument('--force-overwrite', action='store_true',
|
|
help='if file yyyy-mm-dd.pdf already exists, delete it and start over')
|
|
parser.add_argument('--force-probe', action='store_true',
|
|
help='force all pages to use the \'probe\' method')
|
|
parser.add_argument('--fetch-probe-pages', nargs='+', type=int,
|
|
help='force some pages to use the \'probe\' method, when count of vertical and horizontal tiles is unknown')
|
|
|
|
args = parser.parse_args()
|
|
|
|
with_continuation = args.continue_prev or args.continue_next
|
|
if args.fetch_probe_pages and with_continuation:
|
|
raise RuntimeError('--fetch-probe-pages cannot be used together with --continue-* options, it\'s a one time hack')
|
|
if args.only_fetch and with_continuation:
|
|
raise RuntimeError('--only-fetch cannot be used together with --continue-* options, it\'s a one time hack')
|
|
|
|
TILE_MERGING_POOL_SIZE = args.merge_threads
|
|
PAGE_FETCHING_POOL_SIZE = args.fetch_threads
|
|
|
|
url = args.url
|
|
while True:
|
|
print(f'grabbing {url}...')
|
|
if not retronews.grab_magazine(url,
|
|
output_root=args.output,
|
|
probe_pages=args.fetch_probe_pages,
|
|
probe_all=args.force_probe,
|
|
only_fetch=args.only_fetch,
|
|
force_overwrite=args.force_overwrite):
|
|
logging.error('failed to grab')
|
|
break
|
|
|
|
if not args.continue_prev and not args.continue_next:
|
|
break
|
|
|
|
r = requests.get(url)
|
|
|
|
try:
|
|
next_url = None
|
|
if args.continue_next:
|
|
next_url = re.search(r'<a class="float-right pt-4 text-secondary" href="([^"]+)">SUIVANT', r.text, re.S).groups()[0]
|
|
elif args.continue_prev:
|
|
next_url = re.search(r'<a class="float-left pt-4 text-secondary" href="([^"]+)"><i class="fa fa-chevron-left">\s+</i>\s+PRÉCÉDENT</a>', r.text, re.S).groups()[0]
|
|
|
|
if not next_url:
|
|
break
|
|
|
|
if next_url.startswith('/'):
|
|
next_url = f'https://www.retronews.fr{next_url}'
|
|
|
|
url = next_url
|
|
|
|
except:
|
|
print('error: failed to find previous link! exiting')
|
|
break
|