aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/extractor/arte.py
blob: 3a34d1ecc67e590c568d10131b557b1d3022fe4d (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
# encoding: utf-8
from __future__ import unicode_literals

import re

from .common import InfoExtractor
from ..utils import (
    ExtractorError,
    find_xpath_attr,
    unified_strdate,
    determine_ext,
    get_element_by_id,
    compat_str,
    get_element_by_attribute,
)

# There are different sources of video in arte.tv, the extraction process 
# is different for each one. The videos usually expire in 7 days, so we can't
# add tests.


class ArteTvIE(InfoExtractor):
    _VALID_URL = r'http://videos\.arte\.tv/(?P<lang>fr|de)/.*-(?P<id>.*?)\.html'
    IE_NAME = 'arte.tv'

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        lang = mobj.group('lang')
        video_id = mobj.group('id')

        ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
        ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
        ref_xml_doc = self._download_xml(
            ref_xml_url, video_id, note='Downloading metadata')
        config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
        config_xml_url = config_node.attrib['ref']
        config = self._download_xml(
            config_xml_url, video_id, note='Downloading configuration')

        formats = [{
            'forma_id': q.attrib['quality'],
            # The playpath starts at 'mp4:', if we don't manually
            # split the url, rtmpdump will incorrectly parse them
            'url': q.text.split('mp4:', 1)[0],
            'play_path': 'mp4:' + q.text.split('mp4:', 1)[1],
            'ext': 'flv',
            'quality': 2 if q.attrib['quality'] == 'hd' else 1,
        } for q in config.findall('./urls/url')]
        self._sort_formats(formats)

        title = config.find('.//name').text
        thumbnail = config.find('.//firstThumbnailUrl').text
        return {
            'id': video_id,
            'title': title,
            'thumbnail': thumbnail,
            'formats': formats,
        }


class ArteTVPlus7IE(InfoExtractor):
    IE_NAME = 'arte.tv:+7'
    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'

    @classmethod
    def _extract_url_info(cls, url):
        mobj = re.match(cls._VALID_URL, url)
        lang = mobj.group('lang')
        # This is not a real id, it can be for example AJT for the news
        # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
        video_id = mobj.group('id')
        return video_id, lang

    def _real_extract(self, url):
        video_id, lang = self._extract_url_info(url)
        webpage = self._download_webpage(url, video_id)
        return self._extract_from_webpage(webpage, video_id, lang)

    def _extract_from_webpage(self, webpage, video_id, lang):
        json_url = self._html_search_regex(
            [r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'],
            webpage, 'json vp url')
        return self._extract_from_json_url(json_url, video_id, lang)

    def _extract_from_json_url(self, json_url, video_id, lang):
        info = self._download_json(json_url, video_id)
        player_info = info['videoJsonPlayer']

        upload_date_str = player_info.get('shootingDate')
        if not upload_date_str:
            upload_date_str = player_info.get('VDA', '').split(' ')[0]

        info_dict = {
            'id': player_info['VID'],
            'title': player_info['VTI'],
            'description': player_info.get('VDE'),
            'upload_date': unified_strdate(upload_date_str),
            'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
        }

        all_formats = player_info['VSR'].values()
        # Some formats use the m3u8 protocol
        all_formats = list(filter(lambda f: f.get('videoFormat') != 'M3U8', all_formats))
        def _match_lang(f):
            if f.get('versionCode') is None:
                return True
            # Return true if that format is in the language of the url
            if lang == 'fr':
                l = 'F'
            elif lang == 'de':
                l = 'A'
            else:
                l = lang
            regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l]
            return any(re.match(r, f['versionCode']) for r in regexes)
        # Some formats may not be in the same language as the url
        # TODO: Might want not to drop videos that does not match requested language
        # but to process those formats with lower precedence
        formats = filter(_match_lang, all_formats)
        formats = list(formats)  # in python3 filter returns an iterator
        if not formats:
            # Some videos are only available in the 'Originalversion'
            # they aren't tagged as being in French or German
            # Sometimes there are neither videos of requested lang code
            # nor original version videos available
            # For such cases we just take all_formats as is
            formats = all_formats
            if not formats:
                raise ExtractorError('The formats list is empty')

        if re.match(r'[A-Z]Q', formats[0]['quality']) is not None:
            def sort_key(f):
                return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
        else:
            def sort_key(f):
                versionCode = f.get('versionCode')
                if versionCode is None:
                    versionCode = ''
                return (
                    # Sort first by quality
                    int(f.get('height', -1)),
                    int(f.get('bitrate', -1)),
                    # The original version with subtitles has lower relevance
                    re.match(r'VO-ST(F|A)', versionCode) is None,
                    # The version with sourds/mal subtitles has also lower relevance
                    re.match(r'VO?(F|A)-STM\1', versionCode) is None,
                    # Prefer http downloads over m3u8
                    0 if f['url'].endswith('m3u8') else 1,
                )
        formats = sorted(formats, key=sort_key)
        def _format(format_info):
            quality = ''
            height = format_info.get('height')
            if height is not None:
                quality = compat_str(height)
            bitrate = format_info.get('bitrate')
            if bitrate is not None:
                quality += '-%d' % bitrate
            if format_info.get('versionCode') is not None:
                format_id = '%s-%s' % (quality, format_info['versionCode'])
            else:
                format_id = quality
            media_type = format_info.get('mediaType')
            if media_type is not None:
                format_id += '-%s' % media_type
            info = {
                'format_id': format_id,
                'format_note': format_info.get('versionLibelle'),
                'width': format_info.get('width'),
                'height': height,
            }
            if format_info['mediaType'] == 'rtmp':
                info['url'] = format_info['streamer']
                info['play_path'] = 'mp4:' + format_info['url']
                info['ext'] = 'flv'
            else:
                info['url'] = format_info['url']
                info['ext'] = determine_ext(info['url'])
            return info
        info_dict['formats'] = [_format(f) for f in formats]

        return info_dict


# It also uses the arte_vp_url url from the webpage to extract the information
class ArteTVCreativeIE(ArteTVPlus7IE):
    IE_NAME = 'arte.tv:creative'
    _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/(?:magazine?/)?(?P<id>[^?#]+)'

    _TESTS = [{
        'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
        'info_dict': {
            'id': '72176',
            'ext': 'mp4',
            'title': 'Folge 2 - Corporate Design',
            'upload_date': '20131004',
        },
    }, {
        'url': 'http://creative.arte.tv/fr/Monty-Python-Reunion',
        'info_dict': {
            'id': '160676',
            'ext': 'mp4',
            'title': 'Monty Python live (mostly)',
            'description': 'Événement ! Quarante-cinq ans après leurs premiers succès, les légendaires Monty Python remontent sur scène.\n',
            'upload_date': '20140805',
        }
    }]


class ArteTVFutureIE(ArteTVPlus7IE):
    IE_NAME = 'arte.tv:future'
    _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(thema|sujet)/.*?#article-anchor-(?P<id>\d+)'

    _TEST = {
        'url': 'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
        'info_dict': {
            'id': '5201',
            'ext': 'mp4',
            'title': 'Les champignons au secours de la planète',
            'upload_date': '20131101',
        },
    }

    def _real_extract(self, url):
        anchor_id, lang = self._extract_url_info(url)
        webpage = self._download_webpage(url, anchor_id)
        row = get_element_by_id(anchor_id, webpage)
        return self._extract_from_webpage(row, anchor_id, lang)


class ArteTVDDCIE(ArteTVPlus7IE):
    IE_NAME = 'arte.tv:ddc'
    _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'

    def _real_extract(self, url):
        video_id, lang = self._extract_url_info(url)
        if lang == 'folge':
            lang = 'de'
        elif lang == 'emission':
            lang = 'fr'
        webpage = self._download_webpage(url, video_id)
        scriptElement = get_element_by_attribute('class', 'visu_video_block', webpage)
        script_url = self._html_search_regex(r'src="(.*?)"', scriptElement, 'script url')
        javascriptPlayerGenerator = self._download_webpage(script_url, video_id, 'Download javascript player generator')
        json_url = self._search_regex(r"json_url=(.*)&rendering_place.*", javascriptPlayerGenerator, 'json url')
        return self._extract_from_json_url(json_url, video_id, lang)


class ArteTVConcertIE(ArteTVPlus7IE):
    IE_NAME = 'arte.tv:concert'
    _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>de|fr)/(?P<id>.+)'

    _TEST = {
        'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
        'md5': '9ea035b7bd69696b67aa2ccaaa218161',
        'info_dict': {
            'id': '186',
            'ext': 'mp4',
            'title': 'The Notwist im Pariser Konzertclub "Divan du Monde"',
            'upload_date': '20140128',
            'description': 'md5:486eb08f991552ade77439fe6d82c305',
        },
    }


class ArteTVEmbedIE(ArteTVPlus7IE):
    IE_NAME = 'arte.tv:embed'
    _VALID_URL = r'''(?x)
        http://www\.arte\.tv
        /playerv2/embed\.php\?json_url=
        (?P<json_url>
            http://arte\.tv/papi/tvguide/videos/stream/player/
            (?P<lang>[^/]+)/(?P<id>[^/]+)[^&]*
        )
    '''

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        video_id = mobj.group('id')
        lang = mobj.group('lang')
        json_url = mobj.group('json_url')
        return self._extract_from_json_url(json_url, video_id, lang)