]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/vice.py
debian/README.source: Add my preference with respect to patches.
[youtubedl] / youtube_dl / extractor / vice.py
index 8a00c8fee17ee84ae0d8d0e1e7360ca67befc8b0..bcc28693a4545f9260f5e7942bcac4942faac380 100644 (file)
@@ -7,6 +7,7 @@ import hashlib
 import json
 
 from .adobepass import AdobePassIE
+from .youtube import YoutubeIE
 from .common import InfoExtractor
 from ..compat import compat_HTTPError
 from ..utils import (
@@ -20,7 +21,7 @@ from ..utils import (
 
 
 class ViceBaseIE(AdobePassIE):
-    def _extract_preplay_video(self, url, webpage):
+    def _extract_preplay_video(self, url, locale, webpage):
         watch_hub_data = extract_attributes(self._search_regex(
             r'(?s)(<watch-hub\s*.+?</watch-hub>)', webpage, 'watch hub'))
         video_id = watch_hub_data['vms-id']
@@ -32,7 +33,8 @@ class ViceBaseIE(AdobePassIE):
             resource = self._get_mvpd_resource(
                 'VICELAND', title, video_id,
                 watch_hub_data.get('video-rating'))
-            query['tvetoken'] = self._extract_mvpd_auth(url, video_id, 'VICELAND', resource)
+            query['tvetoken'] = self._extract_mvpd_auth(
+                url, video_id, 'VICELAND', resource)
 
         # signature generation algorithm is reverse engineered from signatureGenerator in
         # webpack:///../shared/~/vice-player/dist/js/vice-player.js in
@@ -45,11 +47,14 @@ class ViceBaseIE(AdobePassIE):
 
         try:
             host = 'www.viceland' if is_locked else self._PREPLAY_HOST
-            preplay = self._download_json('https://%s.com/en_us/preplay/%s' % (host, video_id), video_id, query=query)
+            preplay = self._download_json(
+                'https://%s.com/%s/preplay/%s' % (host, locale, video_id),
+                video_id, query=query)
         except ExtractorError as e:
             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
                 error = json.loads(e.cause.read().decode())
-                raise ExtractorError('%s said: %s' % (self.IE_NAME, error['details']), expected=True)
+                raise ExtractorError('%s said: %s' % (
+                    self.IE_NAME, error['details']), expected=True)
             raise
 
         video_data = preplay['video']
@@ -70,10 +75,10 @@ class ViceBaseIE(AdobePassIE):
             'url': uplynk_preplay_url,
             'id': video_id,
             'title': title,
-            'description': base.get('body'),
+            'description': base.get('body') or base.get('display_body'),
             'thumbnail': watch_hub_data.get('cover-image') or watch_hub_data.get('thumbnail'),
-            'duration': parse_duration(video_data.get('video_duration') or watch_hub_data.get('video-duration')),
-            'timestamp': int_or_none(video_data.get('created_at')),
+            'duration': int_or_none(video_data.get('video_duration')) or parse_duration(watch_hub_data.get('video-duration')),
+            'timestamp': int_or_none(video_data.get('created_at'), 1000),
             'age_limit': parse_age_limit(video_data.get('video_rating')),
             'series': video_data.get('show_title') or watch_hub_data.get('show-title'),
             'episode_number': int_or_none(episode.get('episode_number') or watch_hub_data.get('episode')),
@@ -88,41 +93,30 @@ class ViceBaseIE(AdobePassIE):
 
 
 class ViceIE(ViceBaseIE):
-    _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)?videos?/(?P<id>[^/?#&]+)'
+    IE_NAME = 'vice'
+    _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:(?P<locale>[^/]+)/)?videos?/(?P<id>[^/?#&]+)'
 
     _TESTS = [{
-        'url': 'http://www.vice.com/video/cowboy-capitalists-part-1',
-        'md5': 'e9d77741f9e42ba583e683cd170660f7',
+        'url': 'https://news.vice.com/video/experimenting-on-animals-inside-the-monkey-lab',
+        'md5': '7d3ae2f9ba5f196cdd9f9efd43657ac2',
         'info_dict': {
-            'id': '43cW1mYzpia9IlestBjVpd23Yu3afAfp',
+            'id': 'N2bzkydjraWDGwnt8jAttCF6Y0PDv4Zj',
             'ext': 'flv',
-            'title': 'VICE_COWBOYCAPITALISTS_PART01_v1_VICE_WM_1080p.mov',
-            'duration': 725.983,
+            'title': 'Monkey Labs of Holland',
+            'description': 'md5:92b3c7dcbfe477f772dd4afa496c9149',
         },
         'add_ie': ['Ooyala'],
-    }, {
-        'url': 'http://www.vice.com/video/how-to-hack-a-car',
-        'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
-        'info_dict': {
-            'id': '3jstaBeXgAs',
-            'ext': 'mp4',
-            'title': 'How to Hack a Car: Phreaked Out (Episode 2)',
-            'description': 'md5:ee95453f7ff495db8efe14ae8bf56f30',
-            'uploader_id': 'MotherboardTV',
-            'uploader': 'Motherboard',
-            'upload_date': '20140529',
-        },
-        'add_ie': ['Youtube'],
     }, {
         'url': 'https://video.vice.com/en_us/video/the-signal-from-tolva/5816510690b70e6c5fd39a56',
-        'md5': '',
         'info_dict': {
             'id': '5816510690b70e6c5fd39a56',
             'ext': 'mp4',
             'uploader': 'Waypoint',
             'title': 'The Signal From Tölva',
+            'description': 'md5:3927e3c79f9e8094606a2b3c5b5e55d5',
             'uploader_id': '57f7d621e05ca860fa9ccaf9',
-            'timestamp': 1477941983938,
+            'timestamp': 1477941983,
+            'upload_date': '20161031',
         },
         'params': {
             # m3u8 download
@@ -130,19 +124,31 @@ class ViceIE(ViceBaseIE):
         },
         'add_ie': ['UplynkPreplay'],
     }, {
-        'url': 'https://news.vice.com/video/experimenting-on-animals-inside-the-monkey-lab',
-        'only_matching': True,
-    }, {
-        'url': 'http://www.vice.com/ru/video/big-night-out-ibiza-clive-martin-229',
-        'only_matching': True,
+        'url': 'https://video.vice.com/alps/video/ulfs-wien-beruchtigste-grafitti-crew-part-1/581b12b60a0e1f4c0fb6ea2f',
+        'info_dict': {
+            'id': '581b12b60a0e1f4c0fb6ea2f',
+            'ext': 'mp4',
+            'title': 'ULFs - Wien berüchtigste Grafitti Crew - Part 1',
+            'description': '<p>Zwischen Hinterzimmer-Tattoos und U-Bahnschächten erzählen uns die Ulfs, wie es ist, "süchtig nach Sachbeschädigung" zu sein.</p>',
+            'uploader': 'VICE',
+            'uploader_id': '57a204088cb727dec794c67b',
+            'timestamp': 1485368119,
+            'upload_date': '20170125',
+            'age_limit': 14,
+        },
+        'params': {
+            # AES-encrypted m3u8
+            'skip_download': True,
+        },
+        'add_ie': ['UplynkPreplay'],
     }, {
-        'url': 'https://munchies.vice.com/en/videos/watch-the-trailer-for-our-new-series-the-pizza-show',
+        'url': 'https://video.vice.com/en_us/video/pizza-show-trailer/56d8c9a54d286ed92f7f30e4',
         'only_matching': True,
     }]
     _PREPLAY_HOST = 'video.vice'
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        locale, video_id = re.match(self._VALID_URL, url).groups()
         webpage, urlh = self._download_webpage_handle(url, video_id)
         embed_code = self._search_regex(
             r'embedCode=([^&\'"]+)', webpage,
@@ -153,10 +159,11 @@ class ViceIE(ViceBaseIE):
             r'data-youtube-id="([^"]+)"', webpage, 'youtube id', default=None)
         if youtube_id:
             return self.url_result(youtube_id, 'Youtube')
-        return self._extract_preplay_video(urlh.geturl(), webpage)
+        return self._extract_preplay_video(urlh.geturl(), locale, webpage)
 
 
 class ViceShowIE(InfoExtractor):
+    IE_NAME = 'vice:show'
     _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)?show/(?P<id>[^/?#&]+)'
 
     _TEST = {
@@ -183,6 +190,84 @@ class ViceShowIE(InfoExtractor):
             r'<title>(.+?)</title>', webpage, 'title', default=None)
         if title:
             title = re.sub(r'(.+)\s*\|\s*.+$', r'\1', title).strip()
-        description = self._html_search_meta('description', webpage, 'description')
+        description = self._html_search_meta(
+            'description', webpage, 'description')
 
         return self.playlist_result(entries, show_id, title, description)
+
+
+class ViceArticleIE(InfoExtractor):
+    IE_NAME = 'vice:article'
+    _VALID_URL = r'https://www\.vice\.com/[^/]+/article/(?P<id>[^?#]+)'
+
+    _TESTS = [{
+        'url': 'https://www.vice.com/en_us/article/on-set-with-the-woman-making-mormon-porn-in-utah',
+        'info_dict': {
+            'id': '58dc0a3dee202d2a0ccfcbd8',
+            'ext': 'mp4',
+            'title': 'Mormon War on Porn ',
+            'description': 'md5:ad396a2481e7f8afb5ed486878421090',
+            'uploader': 'VICE',
+            'uploader_id': '57a204088cb727dec794c693',
+            'timestamp': 1489160690,
+            'upload_date': '20170310',
+        },
+        'params': {
+            # AES-encrypted m3u8
+            'skip_download': True,
+        },
+        'add_ie': ['UplynkPreplay'],
+    }, {
+        'url': 'https://www.vice.com/en_us/article/how-to-hack-a-car',
+        'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
+        'info_dict': {
+            'id': '3jstaBeXgAs',
+            'ext': 'mp4',
+            'title': 'How to Hack a Car: Phreaked Out (Episode 2)',
+            'description': 'md5:ee95453f7ff495db8efe14ae8bf56f30',
+            'uploader_id': 'MotherboardTV',
+            'uploader': 'Motherboard',
+            'upload_date': '20140529',
+        },
+        'add_ie': ['Youtube'],
+    }, {
+        'url': 'https://www.vice.com/en_us/article/cowboy-capitalists-part-1',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.vice.com/ru/article/big-night-out-ibiza-clive-martin-229',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        prefetch_data = self._parse_json(self._search_regex(
+            r'window\.__PREFETCH_DATA\s*=\s*({.*});',
+            webpage, 'prefetch data'), display_id)
+        body = prefetch_data['body']
+
+        def _url_res(video_url, ie_key):
+            return {
+                '_type': 'url_transparent',
+                'url': video_url,
+                'display_id': display_id,
+                'ie_key': ie_key,
+            }
+
+        embed_code = self._search_regex(
+            r'embedCode=([^&\'"]+)', body,
+            'ooyala embed code', default=None)
+        if embed_code:
+            return _url_res('ooyala:%s' % embed_code, 'Ooyala')
+
+        youtube_url = YoutubeIE._extract_url(body)
+        if youtube_url:
+            return _url_res(youtube_url, YoutubeIE.ie_key())
+
+        video_url = self._html_search_regex(
+            r'data-video-url="([^"]+)"',
+            prefetch_data['embed_code'], 'video URL')
+
+        return _url_res(video_url, ViceIE.ie_key())