Merge adc6e01964 into 1036478d13

* WEB URLs are, MWEB not
* resolves #33017

youtube_dl/extractor/extractors.py

@@ -645,6 +645,10 @@ from .livestream import (
 )
 from .lnkgo import LnkGoIE
 from .localnews8 import LocalNews8IE
+from .loom import (
+    LoomIE,
+    LoomFolderIE
+)
 from .lovehomeporn import LoveHomePornIE
 from .lrt import LRTIE
 from .lynda import (

youtube_dl/extractor/loom.py

@@ -0,0 +1,194 @@
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+
+from ..compat import (
+    compat_urllib_parse_unquote,
+    compat_urllib_parse_urlencode,
+    compat_urllib_request
+)
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    try_get,
+    unified_timestamp,
+    url_or_none
+)
+
+
+class LoomBaseInfoIE(InfoExtractor):
+    _BASE_URL = 'https://www.loom.com/'
+
+
+class LoomIE(LoomBaseInfoIE):
+    _VALID_URL = r'https?://(?:www\.)?loom\.com/share/(?!folder)(?P<id>[a-zA-Z0-9]+)'
+    _TESTS = [
+        {
+            'url': 'https://www.loom.com/share/31b41727a5b24dacb6c1417a565b2ebf',
+            'md5': '8b94361aabff2075141dc60bd6d35453',
+            'info_dict': {
+                'id': '31b41727a5b24dacb6c1417a565b2ebf',
+                'ext': 'mp4',
+                'title': 'How to resize your camera bubble',
+                'uploader': 'Allie Hitchcock',
+                'upload_date': '20201007',
+                'timestamp': 1602089241
+            }
+        },
+        {
+            'url': 'https://www.loom.com/share/7e5168ec3b0744cab5e08a340cc7e086',
+            'md5': '47dd14aa1d8054c249b68ca57ad9963f',
+            'info_dict': {
+                'id': '7e5168ec3b0744cab5e08a340cc7e086',
+                'ext': 'mp4',
+                'title': 'How to flip your camera ',
+                'uploader': 'Matthew Flores',
+                'upload_date': '20200423',
+                'timestamp': 1587646164
+            }
+        },
+        {
+            'url': 'https://www.loom.com/share/6670e3eba3c84dc09ada8306c7138075',
+            'md5': 'bfad8181ed49d6252b10dfdeb46c535e',
+            'info_dict': {
+                'id': '6670e3eba3c84dc09ada8306c7138075',
+                'ext': 'mp4',
+                'title': 'How to record your first video on Loom',
+                'uploader': 'Allie Hitchcock',
+                'upload_date': '20201118',
+                'timestamp': 1605729404
+            }
+        }
+    ]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        info_json = self._html_search_regex(
+            r'window.loomSSRVideo = (.+?);',
+            webpage,
+            'info')
+        info = self._parse_json(info_json, 'json', js_to_json)
+
+        formats = []
+        for type in ['transcoded-url', 'raw-url']:
+            json_doc = self._download_json(
+                self._BASE_URL + 'api/campaigns/sessions/' + video_id + '/' + type,
+                video_id, data={})
+            url = url_or_none(json_doc.get('url'))
+            part_credentials = json_doc.get('part_credentials')
+
+            ext = self._search_regex(
+                r'\.([a-zA-Z0-9]+)\?',
+                url, 'ext', default=None)
+            if ext != 'm3u8':
+                formats.append({
+                    'url': url,
+                    'ext': ext,
+                    'format_id': type,
+                    'width': int_or_none(try_get(info, lambda x: x['video_properties']['width'])),
+                    'height': int_or_none(try_get(info, lambda x: x['video_properties']['height']))
+                })
+            else:
+                credentials = compat_urllib_parse_urlencode(part_credentials)
+                m3u8_formats = self._extract_m3u8_formats(url, video_id)
+                for item in m3u8_formats:
+                    item['protocol'] = 'm3u8_native'
+                    item['url'] += '?' + credentials
+                    item['ext'] = 'mp4'
+                    item['format_id'] = 'hls-' + str(item.get('height', 0))
+                    item['extra_param_to_segment_url'] = credentials
+                for i in range(len(m3u8_formats)):
+                    formats.insert(
+                        (-1, len(formats))[i == len(m3u8_formats) - 1],
+                        m3u8_formats[i])
+
+        return {
+            'id': info.get('id') or video_id,
+            'title': info.get('name'),
+            'formats': formats,
+            'thumbnails': [
+                {
+                    'id': key,
+                    'url': url_or_none(self._BASE_URL + value)
+                } for key, value in info.get('thumbnails').items()
+            ],
+            'description': info.get('description'),
+            'uploader': info.get('owner_full_name'),
+            'timestamp': unified_timestamp(info.get('createdAt'))
+        }
+
+
+class LoomFolderIE(LoomBaseInfoIE):
+    _VALID_URL = r'https?://(?:www\.)?loom\.com/share/folder/(?P<id>.+)/?'
+    _TESTS = [
+        {
+            'url': 'https://www.loom.com/share/folder/997db4db046f43e5912f10dc5f817b5c/List%20A-%20a%2C%20i%2C%20o',
+            'info_dict': {
+                'id': '9a8a87f6b6f546d9a400c8e7575ff7f2',
+                'title': 'List A- a, i, o'
+            },
+            'playlist_mincount': 12
+        },
+        {
+            'url': 'https://www.loom.com/share/folder/997db4db046f43e5912f10dc5f817b5c',
+            'info_dict': {
+                'id': '997db4db046f43e5912f10dc5f817b5c',
+                'title': 'Blending Lessons '
+            },
+            'playlist_mincount': 16
+        }
+    ]
+
+    def _get_real_folder_id(self, path):
+        subfolders = re.match(
+            r'^([a-zA-Z0-9]+)(?:\/(.+))*$',
+            compat_urllib_parse_unquote(path))
+        folder_names = subfolders.groups()[1:]
+        parent_folder_id = subfolders.group(1)
+        if(folder_names[0] is None):
+            return path
+
+        # Fetch folder id
+        request = compat_urllib_request.Request(
+            self._BASE_URL + 'v1/folders/by_name',
+            json.dumps({
+                'folder_names': folder_names,
+                'parent_folder_id': parent_folder_id
+            }).encode('utf-8'))
+        json_doc = self._download_json(request, parent_folder_id)
+
+        return try_get(json_doc, lambda x: x['current_folder']['id'])
+
+    def _get_folder_info(self, folder_id):
+        json_doc = self._download_json(url_or_none(self._BASE_URL + 'v1/folders/' + folder_id), folder_id)
+        videos = []
+
+        # Recursive call for subfolder
+        for folder in json_doc.get('folders'):
+            subfolder_info = self._get_folder_info(folder.get('id'))
+            videos.extend(subfolder_info.get('entries'))
+        videos.extend([val.get('id') for val in json_doc.get('videos')])
+
+        return {
+            'id': folder_id,
+            'title': json_doc.get('name'),
+            'description': json_doc.get('description'),
+            'entries': videos
+        }
+
+    def _real_extract(self, url):
+        folder_id = self._match_id(url)
+        folder_id = self._get_real_folder_id(folder_id)
+        folder_info = self._get_folder_info(folder_id)
+        folder_info['_type'] = 'playlist'
+
+        for i in range(len(folder_info['entries'])):
+            video_id = folder_info['entries'][i]
+            folder_info['entries'][i] = self.url_result(url_or_none(self._BASE_URL + 'share/' + video_id), 'Loom', video_id)
+
+        return folder_info

youtube_dl/extractor/youtube.py

@@ -9,6 +9,7 @@ import json
 import os.path
 import random
 import re
+import string
 import time
 import traceback
 
@@ -67,6 +68,7 @@ from ..utils import (
 
 class YoutubeBaseInfoExtractor(InfoExtractor):
     """Provide base functions for Youtube extractors"""
+
     _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
     _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
 
@@ -138,7 +140,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                 [2, 1, None, 1,
                  'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
                  None, [], 4],
-                1, [None, None, []], None, None, None, True
+                1, [None, None, []], None, None, None, True,
             ],
             username,
         ]
@@ -160,7 +162,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             None, 1, None, [1, None, None, None, [password, None, True]],
             [
                 None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
-                1, [None, None, []], None, None, None, True
+                1, [None, None, []], None, None, None, True,
             ]]
 
         challenge_results = req(
@@ -213,7 +215,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                     user_hash, None, 2, None,
                     [
                         9, None, None, None, None, None, None, None,
-                        [None, tfa_code, True, 2]
+                        [None, tfa_code, True, 2],
                     ]]
 
                 tfa_results = req(
@@ -284,7 +286,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'client': {
                 'clientName': 'WEB',
                 'clientVersion': '2.20201021.03.00',
-            }
+            },
         },
     }
 
@@ -385,7 +387,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                 'client': {
                     'clientName': 'WEB',
                     'clientVersion': '2.20201021.03.00',
-                }
+                },
             },
             'query': query,
         }
@@ -462,7 +464,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             #       (HTML, videodetails, metadata, renderers)
             'name': ('content', 'author', (('ownerChannelName', None), 'title'), ['text']),
             'url': ('href', 'ownerProfileUrl', 'vanityChannelUrl',
-                    ['navigationEndpoint', 'browseEndpoint', 'canonicalBaseUrl'])
+                    ['navigationEndpoint', 'browseEndpoint', 'canonicalBaseUrl']),
         }
         if any((videodetails, metadata, renderers)):
             result = (
@@ -671,7 +673,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
                 'description': '',
                 'uploader': '8KVIDEO',
-                'title': 'UHDTV TEST 8K VIDEO.mp4'
+                'title': 'UHDTV TEST 8K VIDEO.mp4',
             },
             'params': {
                 'youtube_include_dash_manifest': True,
@@ -711,7 +713,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@theamazingatheist',
                 'title': 'Burning Everyone\'s Koran',
                 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms \r\n\r\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
-            }
+            },
         },
         # Age-gated videos
         {
@@ -839,7 +841,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             },
             'expected_warnings': [
                 'DASH manifest missing',
-            ]
+            ],
         },
         # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
         {
@@ -1820,8 +1822,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
         # cpn generation algorithm is reverse engineered from base.js.
         # In fact it works even with dummy cpn.
-        CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
-        cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
+        CPN_ALPHABET = string.ascii_letters + string.digits + '-_'
+        cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(16))
 
         # more consistent results setting it to right before the end
         qs = parse_qs(playback_url)
@@ -1881,8 +1883,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
         if mobj is None:
             raise ExtractorError('Invalid URL: %s' % url)
-        video_id = mobj.group(2)
-        return video_id
+        return mobj.group(2)
 
     def _extract_chapters_from_json(self, data, video_id, duration):
         chapters_list = try_get(
@@ -2035,7 +2036,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             headers = {
                 'X-YouTube-Client-Name': '85',
                 'X-YouTube-Client-Version': '2.0',
-                'Origin': 'https://www.youtube.com'
+                'Origin': 'https://www.youtube.com',
             }
 
             video_info = self._call_api('player', query, video_id, fatal=False, headers=headers)
@@ -2064,8 +2065,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             return ''.join([r['text'] for r in runs if isinstance(r.get('text'), compat_str)])
 
         search_meta = (
-            lambda x: self._html_search_meta(x, webpage, default=None)) \
-            if webpage else lambda x: None
+            (lambda x: self._html_search_meta(x, webpage, default=None))
+            if webpage else lambda _: None)
 
         video_details = player_response.get('videoDetails') or {}
         microformat = try_get(
@@ -2137,7 +2138,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         def build_fragments(f):
             return LazyList({
                 'url': update_url_query(f['url'], {
-                    'range': '{0}-{1}'.format(range_start, min(range_start + CHUNK_SIZE - 1, f['filesize']))
+                    'range': '{0}-{1}'.format(range_start, min(range_start + CHUNK_SIZE - 1, f['filesize'])),
                 })
             } for range_start in range(0, f['filesize'], CHUNK_SIZE))
 
@@ -2236,7 +2237,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     'protocol': 'http_dash_segments',
                     'fragments': build_fragments(dct),
                 } if dct['filesize'] else {
-                    'downloader_options': {'http_chunk_size': CHUNK_SIZE}  # No longer useful?
+                    'downloader_options': {'http_chunk_size': CHUNK_SIZE},  # No longer useful?
                 })
 
             formats.append(dct)
@@ -2414,9 +2415,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'is_live': is_live,
         }
 
-        pctr = try_get(
+        pctr = traverse_obj(
             player_response,
-            lambda x: x['captions']['playerCaptionsTracklistRenderer'], dict)
+            ('captions', 'playerCaptionsTracklistRenderer', T(dict)))
         if pctr:
             def process_language(container, base_url, lang_code, query):
                 lang_subs = []
@@ -2430,31 +2431,34 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     })
                 container[lang_code] = lang_subs
 
-            subtitles = {}
-            for caption_track in (pctr.get('captionTracks') or []):
-                base_url = caption_track.get('baseUrl')
-                if not base_url:
-                    continue
-                if caption_track.get('kind') != 'asr':
-                    lang_code = caption_track.get('languageCode')
-                    if not lang_code:
+            def process_subtitles():
+                subtitles = {}
+                for caption_track in traverse_obj(pctr, (
+                        'captionTracks', lambda _, v: v.get('baseUrl'))):
+                    base_url = self._yt_urljoin(caption_track['baseUrl'])
+                    if not base_url:
                         continue
-                    process_language(
-                        subtitles, base_url, lang_code, {})
-                    continue
-                automatic_captions = {}
-                for translation_language in (pctr.get('translationLanguages') or []):
-                    translation_language_code = translation_language.get('languageCode')
-                    if not translation_language_code:
+                    if caption_track.get('kind') != 'asr':
+                        lang_code = caption_track.get('languageCode')
+                        if not lang_code:
+                            continue
+                        process_language(
+                            subtitles, base_url, lang_code, {})
                         continue
-                    process_language(
-                        automatic_captions, base_url, translation_language_code,
-                        {'tlang': translation_language_code})
-                info['automatic_captions'] = automatic_captions
-            info['subtitles'] = subtitles
+                    automatic_captions = {}
+                    for translation_language in traverse_obj(pctr, (
+                            'translationLanguages', lambda _, v: v.get('languageCode'))):
+                        translation_language_code = translation_language['languageCode']
+                        process_language(
+                            automatic_captions, base_url, translation_language_code,
+                            {'tlang': translation_language_code})
+                    info['automatic_captions'] = automatic_captions
+                info['subtitles'] = subtitles
+
+            process_subtitles()
 
         parsed_url = compat_urllib_parse_urlparse(url)
-        for component in [parsed_url.fragment, parsed_url.query]:
+        for component in (parsed_url.fragment, parsed_url.query):
             query = compat_parse_qs(component)
             for k, v in query.items():
                 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
@@ -2684,7 +2688,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'title': 'Super Cooper Shorts - Shorts',
             'uploader': 'Super Cooper Shorts',
             'uploader_id': '@SuperCooperShorts',
-        }
+        },
     }, {
         # Channel that does not have a Shorts tab. Test should just download videos on Home tab instead
         'url': 'https://www.youtube.com/@emergencyawesome/shorts',
@@ -2738,7 +2742,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'description': 'md5:609399d937ea957b0f53cbffb747a14c',
             'uploader': 'ThirstForScience',
             'uploader_id': '@ThirstForScience',
-        }
+        },
     }, {
         'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
         'only_matching': True,
@@ -3037,7 +3041,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'uploader': '3Blue1Brown',
             'uploader_id': '@3blue1brown',
             'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
-        }
+        },
     }]
 
     @classmethod
@@ -3335,7 +3339,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'client': {
                 'clientName': 'WEB',
                 'clientVersion': client_version,
-            }
+            },
         }
         visitor_data = try_get(context, lambda x: x['client']['visitorData'], compat_str)
 
@@ -3354,7 +3358,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                 headers['x-goog-visitor-id'] = visitor_data
             data['continuation'] = continuation['continuation']
             data['clickTracking'] = {
-                'clickTrackingParams': continuation['itct']
+                'clickTrackingParams': continuation['itct'],
             }
             count = 0
             retries = 3
@@ -3613,7 +3617,7 @@ class YoutubePlaylistIE(InfoExtractor):
             'uploader': 'milan',
             'uploader_id': '@milan5503',
             'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
-        }
+        },
     }, {
         'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
         'playlist_mincount': 455,
@@ -3623,7 +3627,7 @@ class YoutubePlaylistIE(InfoExtractor):
             'uploader': 'LBK',
             'uploader_id': '@music_king',
             'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
-        }
+        },
     }, {
         'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
         'only_matching': True,
@@ -3734,7 +3738,7 @@ class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor):
         'info_dict': {
             'id': 'youtube-dl test video',
             'title': 'youtube-dl test video',
-        }
+        },
     }]
 
     def _get_n_results(self, query, n):
@@ -3754,7 +3758,7 @@ class YoutubeSearchDateIE(YoutubeSearchIE):
         'info_dict': {
             'id': 'youtube-dl test video',
             'title': 'youtube-dl test video',
-        }
+        },
     }]
 
 
@@ -3769,7 +3773,7 @@ class YoutubeSearchURLIE(YoutubeBaseInfoExtractor):
             'id': 'youtube-dl test video',
             'title': 'youtube-dl test video',
         },
-        'params': {'playlistend': 5}
+        'params': {'playlistend': 5},
     }, {
         'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
         'only_matching': True,
@@ -3785,6 +3789,7 @@ class YoutubeSearchURLIE(YoutubeBaseInfoExtractor):
 class YoutubeFeedsInfoExtractor(YoutubeTabIE):
     """
     Base class for feed extractors
+
     Subclasses must define the _FEED_NAME property.
     """
     _LOGIN_REQUIRED = True