youtube_dl/extractor/tf1.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     int_or_none,
  10     parse_iso8601,
  11     try_get,
  12 )
  13
  14
  15 class TF1IE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?tf1\.fr/[^/]+/(?P<program_slug>[^/]+)/videos/(?P<id>[^/?&#]+)\.html'
  17     _TESTS = [{
  18         'url': 'https://www.tf1.fr/tmc/quotidien-avec-yann-barthes/videos/quotidien-premiere-partie-11-juin-2019.html',
  19         'info_dict': {
  20             'id': '13641379',
  21             'ext': 'mp4',
  22             'title': 'md5:f392bc52245dc5ad43771650c96fb620',
  23             'description': 'md5:a02cdb217141fb2d469d6216339b052f',
  24             'upload_date': '20190611',
  25             'timestamp': 1560273989,
  26             'duration': 1738,
  27             'series': 'Quotidien avec Yann Barthès',
  28             'tags': ['intégrale', 'quotidien', 'Replay'],
  29         },
  30         'params': {
  31             # Sometimes wat serves the whole file with the --test option
  32             'skip_download': True,
  33             'format': 'bestvideo',
  34         },
  35     }, {
  36         'url': 'http://www.tf1.fr/tf1/koh-lanta/videos/replay-koh-lanta-22-mai-2015.html',
  37         'only_matching': True,
  38     }, {
  39         'url': 'http://www.tf1.fr/hd1/documentaire/videos/mylene-farmer-d-une-icone.html',
  40         'only_matching': True,
  41     }]
  42
  43     def _real_extract(self, url):
  44         program_slug, slug = re.match(self._VALID_URL, url).groups()
  45         video = self._download_json(
  46             'https://www.tf1.fr/graphql/web', slug, query={
  47                 'id': '9b80783950b85247541dd1d851f9cc7fa36574af015621f853ab111a679ce26f',
  48                 'variables': json.dumps({
  49                     'programSlug': program_slug,
  50                     'slug': slug,
  51                 })
  52             })['data']['videoBySlug']
  53         wat_id = video['streamId']
  54
  55         tags = []
  56         for tag in (video.get('tags') or []):
  57             label = tag.get('label')
  58             if not label:
  59                 continue
  60             tags.append(label)
  61
  62         decoration = video.get('decoration') or {}
  63
  64         thumbnails = []
  65         for source in (try_get(decoration, lambda x: x['image']['sources'], list) or []):
  66             source_url = source.get('url')
  67             if not source_url:
  68                 continue
  69             thumbnails.append({
  70                 'url': source_url,
  71                 'width': int_or_none(source.get('width')),
  72             })
  73
  74         return {
  75             '_type': 'url_transparent',
  76             'id': wat_id,
  77             'url': 'wat:' + wat_id,
  78             'title': video.get('title'),
  79             'thumbnails': thumbnails,
  80             'description': decoration.get('description'),
  81             'timestamp': parse_iso8601(video.get('date')),
  82             'duration': int_or_none(try_get(video, lambda x: x['publicPlayingInfos']['duration'])),
  83             'tags': tags,
  84             'series': decoration.get('programLabel'),
  85             'season_number': int_or_none(video.get('season')),
  86             'episode_number': int_or_none(video.get('episode')),
  87         }