yt_dlp/extractor/tf1.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     parse_iso8601,
  10     try_get,
  11 )
  12
  13
  14 class TF1IE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?tf1\.fr/[^/]+/(?P<program_slug>[^/]+)/videos/(?P<id>[^/?&#]+)\.html'
  16     _TESTS = [{
  17         'url': 'https://www.tf1.fr/tmc/quotidien-avec-yann-barthes/videos/quotidien-premiere-partie-11-juin-2019.html',
  18         'info_dict': {
  19             'id': '13641379',
  20             'ext': 'mp4',
  21             'title': 'md5:f392bc52245dc5ad43771650c96fb620',
  22             'description': 'md5:a02cdb217141fb2d469d6216339b052f',
  23             'upload_date': '20190611',
  24             'timestamp': 1560273989,
  25             'duration': 1738,
  26             'series': 'Quotidien avec Yann Barthès',
  27             'tags': ['intégrale', 'quotidien', 'Replay'],
  28         },
  29         'params': {
  30             # Sometimes wat serves the whole file with the --test option
  31             'skip_download': True,
  32         },
  33     }, {
  34         'url': 'http://www.tf1.fr/tf1/koh-lanta/videos/replay-koh-lanta-22-mai-2015.html',
  35         'only_matching': True,
  36     }, {
  37         'url': 'http://www.tf1.fr/hd1/documentaire/videos/mylene-farmer-d-une-icone.html',
  38         'only_matching': True,
  39     }]
  40
  41     def _real_extract(self, url):
  42         program_slug, slug = self._match_valid_url(url).groups()
  43         video = self._download_json(
  44             'https://www.tf1.fr/graphql/web', slug, query={
  45                 'id': '9b80783950b85247541dd1d851f9cc7fa36574af015621f853ab111a679ce26f',
  46                 'variables': json.dumps({
  47                     'programSlug': program_slug,
  48                     'slug': slug,
  49                 })
  50             })['data']['videoBySlug']
  51         wat_id = video['streamId']
  52
  53         tags = []
  54         for tag in (video.get('tags') or []):
  55             label = tag.get('label')
  56             if not label:
  57                 continue
  58             tags.append(label)
  59
  60         decoration = video.get('decoration') or {}
  61
  62         thumbnails = []
  63         for source in (try_get(decoration, lambda x: x['image']['sources'], list) or []):
  64             source_url = source.get('url')
  65             if not source_url:
  66                 continue
  67             thumbnails.append({
  68                 'url': source_url,
  69                 'width': int_or_none(source.get('width')),
  70             })
  71
  72         return {
  73             '_type': 'url_transparent',
  74             'id': wat_id,
  75             'url': 'wat:' + wat_id,
  76             'title': video.get('title'),
  77             'thumbnails': thumbnails,
  78             'description': decoration.get('description'),
  79             'timestamp': parse_iso8601(video.get('date')),
  80             'duration': int_or_none(try_get(video, lambda x: x['publicPlayingInfos']['duration'])),
  81             'tags': tags,
  82             'series': decoration.get('programLabel'),
  83             'season_number': int_or_none(video.get('season')),
  84             'episode_number': int_or_none(video.get('episode')),
  85         }