mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2025-12-13 19:55:25 +00:00
Code Issues Projects Releases Wiki Activity

[ie/loom] Fix extractor (#15236)

Browse Source 
Closes #15141
Authored by: bashonly
This commit is contained in:
bashonly
2025-12-05 17:18:02 -06:00
committed by GitHub
parent 7ec6b9bc40
commit 36b29bb353
1 changed files with 42 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

66
yt_dlp/extractor/loom.py
View File

@@ -8,12 +8,10 @@ from ..utils import (
ExtractorError, ExtractorError,
determine_ext, determine_ext,
filter_dict, filter_dict,
get_first,
int_or_none, int_or_none,
parse_iso8601, parse_iso8601,
update_url, update_url,
url_or_none, url_or_none,
variadic,
) )
from ..utils.traversal import traverse_obj from ..utils.traversal import traverse_obj
@@ -51,7 +49,7 @@ class LoomIE(InfoExtractor):
}, { }, {
# m3u8 raw-url, mp4 transcoded-url, cdn url == raw-url, vtt sub and json subs # m3u8 raw-url, mp4 transcoded-url, cdn url == raw-url, vtt sub and json subs
'url': 'https://www.loom.com/share/9458bcbf79784162aa62ffb8dd66201b', 'url': 'https://www.loom.com/share/9458bcbf79784162aa62ffb8dd66201b',
'md5': '51737ec002969dd28344db4d60b9cbbb', 'md5': '7b6bfdef8181c4ffc376e18919a4dcc2',
'info_dict': { 'info_dict': {
'id': '9458bcbf79784162aa62ffb8dd66201b', 'id': '9458bcbf79784162aa62ffb8dd66201b',
'ext': 'mp4', 'ext': 'mp4',
@@ -71,12 +69,13 @@ class LoomIE(InfoExtractor):
'ext': 'webm', 'ext': 'webm',
'title': 'OMFG clown', 'title': 'OMFG clown',
'description': 'md5:285c5ee9d62aa087b7e3271b08796815', 'description': 'md5:285c5ee9d62aa087b7e3271b08796815',
'uploader': 'MrPumkin B', 'uploader': 'Brailey Bragg',
'upload_date': '20210924', 'upload_date': '20210924',
'timestamp': 1632519618, 'timestamp': 1632519618,
'duration': 210, 'duration': 210,
}, },
'params': {'skip_download': 'dash'}, 'params': {'skip_download': 'dash'},
'expected_warnings': ['Failed to parse JSON'], # transcoded-url no longer available
}, { }, {
# password-protected # password-protected
'url': 'https://www.loom.com/share/50e26e8aeb7940189dff5630f95ce1f4', 'url': 'https://www.loom.com/share/50e26e8aeb7940189dff5630f95ce1f4',
@@ -91,10 +90,11 @@ class LoomIE(InfoExtractor):
'duration': 35, 'duration': 35,
}, },
'params': {'videopassword': 'seniorinfants2'}, 'params': {'videopassword': 'seniorinfants2'},
'expected_warnings': ['Failed to parse JSON'], # transcoded-url no longer available
}, { }, {
# embed, transcoded-url endpoint sends empty JSON response, split video and audio HLS formats # embed, transcoded-url endpoint sends empty JSON response, split video and audio HLS formats
'url': 'https://www.loom.com/embed/ddcf1c1ad21f451ea7468b1e33917e4e', 'url': 'https://www.loom.com/embed/ddcf1c1ad21f451ea7468b1e33917e4e',
'md5': 'b321d261656848c184a94e3b93eae28d', 'md5': 'f983a0f02f24331738b2f43aecb05256',
'info_dict': { 'info_dict': {
'id': 'ddcf1c1ad21f451ea7468b1e33917e4e', 'id': 'ddcf1c1ad21f451ea7468b1e33917e4e',
'ext': 'mp4', 'ext': 'mp4',
@@ -119,11 +119,12 @@ class LoomIE(InfoExtractor):
'duration': 247, 'duration': 247,
'timestamp': 1676274030, 'timestamp': 1676274030,
}, },
'skip': '404 Not Found',
}] }]
_GRAPHQL_VARIABLES = { _GRAPHQL_VARIABLES = {
'GetVideoSource': { 'GetVideoSource': {
'acceptableMimes': ['DASH', 'M3U8', 'MP4'], 'acceptableMimes': ['DASH', 'M3U8', 'MP4', 'WEBM'],
}, },
} }
_GRAPHQL_QUERIES = { _GRAPHQL_QUERIES = {
@@ -192,6 +193,12 @@ class LoomIE(InfoExtractor):
id id
nullableRawCdnUrl(acceptableMimes: $acceptableMimes, password: $password) { nullableRawCdnUrl(acceptableMimes: $acceptableMimes, password: $password) {
url url
credentials {
Policy
Signature
KeyPairId
__typename
}
__typename __typename
} }
__typename __typename
@@ -240,9 +247,9 @@ class LoomIE(InfoExtractor):
} }
}\n'''), }\n'''),
} }
_APOLLO_GRAPHQL_VERSION = '0a1856c' _APOLLO_GRAPHQL_VERSION = '45a5bd4'
def _call_graphql_api(self, operations, video_id, note=None, errnote=None): def _call_graphql_api(self, operation_name, video_id, note=None, errnote=None, fatal=True):
password = self.get_param('videopassword') password = self.get_param('videopassword')
return self._download_json( return self._download_json(
'https://www.loom.com/graphql', video_id, note or 'Downloading GraphQL JSON', 'https://www.loom.com/graphql', video_id, note or 'Downloading GraphQL JSON',
@@ -252,7 +259,9 @@ class LoomIE(InfoExtractor):
'x-loom-request-source': f'loom_web_{self._APOLLO_GRAPHQL_VERSION}', 'x-loom-request-source': f'loom_web_{self._APOLLO_GRAPHQL_VERSION}',
'apollographql-client-name': 'web', 'apollographql-client-name': 'web',
'apollographql-client-version': self._APOLLO_GRAPHQL_VERSION, 'apollographql-client-version': self._APOLLO_GRAPHQL_VERSION,
}, data=json.dumps([{ 'graphql-operation-name': operation_name,
'Origin': 'https://www.loom.com',
}, data=json.dumps({
'operationName': operation_name, 'operationName': operation_name,
'variables': { 'variables': {
'videoId': video_id, 'videoId': video_id,
@@ -260,7 +269,7 @@ class LoomIE(InfoExtractor):
**self._GRAPHQL_VARIABLES.get(operation_name, {}), **self._GRAPHQL_VARIABLES.get(operation_name, {}),
}, },
'query': self._GRAPHQL_QUERIES[operation_name], 'query': self._GRAPHQL_QUERIES[operation_name],
} for operation_name in variadic(operations)], separators=(',', ':')).encode()) }, separators=(',', ':')).encode(), fatal=fatal)
def _call_url_api(self, endpoint, video_id): def _call_url_api(self, endpoint, video_id):
response = self._download_json( response = self._download_json(
@@ -275,7 +284,7 @@ class LoomIE(InfoExtractor):
}, separators=(',', ':')).encode()) }, separators=(',', ':')).encode())
return traverse_obj(response, ('url', {url_or_none})) return traverse_obj(response, ('url', {url_or_none}))
def _extract_formats(self, video_id, metadata, gql_data): def _extract_formats(self, video_id, metadata, video_data):
formats = [] formats = []
video_properties = traverse_obj(metadata, ('video_properties', { video_properties = traverse_obj(metadata, ('video_properties', {
'width': ('width', {int_or_none}), 'width': ('width', {int_or_none}),
@@ -330,7 +339,7 @@ class LoomIE(InfoExtractor):
transcoded_url = self._call_url_api('transcoded-url', video_id) transcoded_url = self._call_url_api('transcoded-url', video_id)
formats.extend(get_formats(transcoded_url, 'transcoded', quality=-1)) # transcoded quality formats.extend(get_formats(transcoded_url, 'transcoded', quality=-1)) # transcoded quality
cdn_url = get_first(gql_data, ('data', 'getVideo', 'nullableRawCdnUrl', 'url', {url_or_none})) cdn_url = traverse_obj(video_data, ('data', 'getVideo', 'nullableRawCdnUrl', 'url', {url_or_none}))
# cdn_url is usually a dupe, but the raw-url/transcoded-url endpoints could return errors # cdn_url is usually a dupe, but the raw-url/transcoded-url endpoints could return errors
valid_urls = [update_url(url, query=None) for url in (raw_url, transcoded_url) if url] valid_urls = [update_url(url, query=None) for url in (raw_url, transcoded_url) if url]
if cdn_url and update_url(cdn_url, query=None) not in valid_urls: if cdn_url and update_url(cdn_url, query=None) not in valid_urls:
@@ -338,10 +347,21 @@ class LoomIE(InfoExtractor):
return formats return formats
def _get_subtitles(self, video_id):
subs_data = self._call_graphql_api(
'FetchVideoTranscript', video_id, 'Downloading GraphQL subtitles JSON', fatal=False)
return filter_dict({
'en': traverse_obj(subs_data, (
'data', 'fetchVideoTranscript',
('source_url', 'captions_source_url'), {
'url': {url_or_none},
})) or None,
})
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
metadata = get_first( metadata = traverse_obj(
self._call_graphql_api('GetVideoSSR', video_id, 'Downloading GraphQL metadata JSON'), self._call_graphql_api('GetVideoSSR', video_id, 'Downloading GraphQL metadata JSON', fatal=False),
('data', 'getVideo', {dict})) or {} ('data', 'getVideo', {dict})) or {}
if metadata.get('__typename') == 'VideoPasswordMissingOrIncorrect': if metadata.get('__typename') == 'VideoPasswordMissingOrIncorrect':
@@ -350,22 +370,19 @@ class LoomIE(InfoExtractor):
'This video is password-protected, use the --video-password option', expected=True) 'This video is password-protected, use the --video-password option', expected=True)
raise ExtractorError('Invalid video password', expected=True) raise ExtractorError('Invalid video password', expected=True)
gql_data = self._call_graphql_api(['FetchChapters', 'FetchVideoTranscript', 'GetVideoSource'], video_id) video_data = self._call_graphql_api(
'GetVideoSource', video_id, 'Downloading GraphQL video JSON')
chapter_data = self._call_graphql_api(
'FetchChapters', video_id, 'Downloading GraphQL chapters JSON', fatal=False)
duration = traverse_obj(metadata, ('video_properties', 'duration', {int_or_none})) duration = traverse_obj(metadata, ('video_properties', 'duration', {int_or_none}))
return { return {
'id': video_id, 'id': video_id,
'duration': duration, 'duration': duration,
'chapters': self._extract_chapters_from_description( 'chapters': self._extract_chapters_from_description(
get_first(gql_data, ('data', 'fetchVideoChapters', 'content', {str})), duration) or None, traverse_obj(chapter_data, ('data', 'fetchVideoChapters', 'content', {str})), duration) or None,
'formats': self._extract_formats(video_id, metadata, gql_data), 'formats': self._extract_formats(video_id, metadata, video_data),
'subtitles': filter_dict({ 'subtitles': self.extract_subtitles(video_id),
'en': traverse_obj(gql_data, (
..., 'data', 'fetchVideoTranscript',
('source_url', 'captions_source_url'), {
'url': {url_or_none},
})) or None,
}),
**traverse_obj(metadata, { **traverse_obj(metadata, {
'title': ('name', {str}), 'title': ('name', {str}),
'description': ('description', {str}), 'description': ('description', {str}),
@@ -376,6 +393,7 @@ class LoomIE(InfoExtractor):
class LoomFolderIE(InfoExtractor): class LoomFolderIE(InfoExtractor):
_WORKING = False
IE_NAME = 'loom:folder' IE_NAME = 'loom:folder'
_VALID_URL = r'https?://(?:www\.)?loom\.com/share/folder/(?P<id>[\da-f]{32})' _VALID_URL = r'https?://(?:www\.)?loom\.com/share/folder/(?P<id>[\da-f]{32})'
_TESTS = [{ _TESTS = [{