mirror of
https://github.com/yt-dlp/yt-dlp.git
synced 2024-09-10 00:58:17 +02:00
[facebook] Remove SWF params so that 1080P are detected
Closes #11073 In the provided link, SWF params give up to 720P, and VideoConfig gives 1080P for both best and bestvideo. I guess all Facebook videos supports HTML5 now, so I remove the old detection for SWF params
This commit is contained in:
parent
e1a0b3b81c
commit
d9ee2e5cf6
@ -1,6 +1,7 @@
|
|||||||
version <unreleased>
|
version <unreleased>
|
||||||
|
|
||||||
Extractors
|
Extractors
|
||||||
|
* [facebook] Improve 1080P video detection (#11073)
|
||||||
* [imgur] Recognize /r/ URLs (#11071)
|
* [imgur] Recognize /r/ URLs (#11071)
|
||||||
* [openload] Fix extraction (#10408)
|
* [openload] Fix extraction (#10408)
|
||||||
* [adultswim] Fix extraction (#10979)
|
* [adultswim] Fix extraction (#10979)
|
||||||
|
@ -1,6 +1,5 @@
|
|||||||
from __future__ import unicode_literals
|
from __future__ import unicode_literals
|
||||||
|
|
||||||
import json
|
|
||||||
import re
|
import re
|
||||||
import socket
|
import socket
|
||||||
|
|
||||||
@ -100,7 +99,8 @@ class FacebookIE(InfoExtractor):
|
|||||||
'ext': 'mp4',
|
'ext': 'mp4',
|
||||||
'title': '"What are you doing running in the snow?"',
|
'title': '"What are you doing running in the snow?"',
|
||||||
'uploader': 'FailArmy',
|
'uploader': 'FailArmy',
|
||||||
}
|
},
|
||||||
|
'skip': 'Video gone',
|
||||||
}, {
|
}, {
|
||||||
'url': 'https://m.facebook.com/story.php?story_fbid=1035862816472149&id=116132035111903',
|
'url': 'https://m.facebook.com/story.php?story_fbid=1035862816472149&id=116132035111903',
|
||||||
'md5': '1deb90b6ac27f7efcf6d747c8a27f5e3',
|
'md5': '1deb90b6ac27f7efcf6d747c8a27f5e3',
|
||||||
@ -110,6 +110,7 @@ class FacebookIE(InfoExtractor):
|
|||||||
'title': 'What the Flock Is Going On In New Zealand Credit: ViralHog',
|
'title': 'What the Flock Is Going On In New Zealand Credit: ViralHog',
|
||||||
'uploader': 'S. Saint',
|
'uploader': 'S. Saint',
|
||||||
},
|
},
|
||||||
|
'skip': 'Video gone',
|
||||||
}, {
|
}, {
|
||||||
'note': 'swf params escaped',
|
'note': 'swf params escaped',
|
||||||
'url': 'https://www.facebook.com/barackobama/posts/10153664894881749',
|
'url': 'https://www.facebook.com/barackobama/posts/10153664894881749',
|
||||||
@ -119,6 +120,18 @@ class FacebookIE(InfoExtractor):
|
|||||||
'ext': 'mp4',
|
'ext': 'mp4',
|
||||||
'title': 'Facebook video #10153664894881749',
|
'title': 'Facebook video #10153664894881749',
|
||||||
},
|
},
|
||||||
|
}, {
|
||||||
|
# have 1080P, but only up to 720p in swf params
|
||||||
|
'url': 'https://www.facebook.com/cnn/videos/10155529876156509/',
|
||||||
|
'md5': '0d9813160b146b3bc8744e006027fcc6',
|
||||||
|
'info_dict': {
|
||||||
|
'id': '10155529876156509',
|
||||||
|
'ext': 'mp4',
|
||||||
|
'title': 'Holocaust survivor becomes US citizen',
|
||||||
|
'timestamp': 1477818095,
|
||||||
|
'upload_date': '20161030',
|
||||||
|
'uploader': 'CNN',
|
||||||
|
},
|
||||||
}, {
|
}, {
|
||||||
'url': 'https://www.facebook.com/video.php?v=10204634152394104',
|
'url': 'https://www.facebook.com/video.php?v=10204634152394104',
|
||||||
'only_matching': True,
|
'only_matching': True,
|
||||||
@ -227,43 +240,13 @@ def _extract_from_url(self, url, video_id, fatal_if_no_video=True):
|
|||||||
|
|
||||||
video_data = None
|
video_data = None
|
||||||
|
|
||||||
BEFORE = '{swf.addParam(param[0], param[1]);});'
|
server_js_data = self._parse_json(self._search_regex(
|
||||||
AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
|
r'handleServerJS\(({.+})(?:\);|,")', webpage, 'server js data', default='{}'), video_id)
|
||||||
PATTERN = re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER)
|
for item in server_js_data.get('instances', []):
|
||||||
|
if item[1][0] == 'VideoConfig':
|
||||||
for m in re.findall(PATTERN, webpage):
|
video_data = item[2][0]['videoData']
|
||||||
swf_params = m.replace('\\\\', '\\').replace('\\"', '"')
|
|
||||||
data = dict(json.loads(swf_params))
|
|
||||||
params_raw = compat_urllib_parse_unquote(data['params'])
|
|
||||||
video_data_candidate = json.loads(params_raw)['video_data']
|
|
||||||
for _, f in video_data_candidate.items():
|
|
||||||
if not f:
|
|
||||||
continue
|
|
||||||
if isinstance(f, dict):
|
|
||||||
f = [f]
|
|
||||||
if not isinstance(f, list):
|
|
||||||
continue
|
|
||||||
if f[0].get('video_id') == video_id:
|
|
||||||
video_data = video_data_candidate
|
|
||||||
break
|
|
||||||
if video_data:
|
|
||||||
break
|
break
|
||||||
|
|
||||||
def video_data_list2dict(video_data):
|
|
||||||
ret = {}
|
|
||||||
for item in video_data:
|
|
||||||
format_id = item['stream_type']
|
|
||||||
ret.setdefault(format_id, []).append(item)
|
|
||||||
return ret
|
|
||||||
|
|
||||||
if not video_data:
|
|
||||||
server_js_data = self._parse_json(self._search_regex(
|
|
||||||
r'handleServerJS\(({.+})(?:\);|,")', webpage, 'server js data', default='{}'), video_id)
|
|
||||||
for item in server_js_data.get('instances', []):
|
|
||||||
if item[1][0] == 'VideoConfig':
|
|
||||||
video_data = video_data_list2dict(item[2][0]['videoData'])
|
|
||||||
break
|
|
||||||
|
|
||||||
if not video_data:
|
if not video_data:
|
||||||
if not fatal_if_no_video:
|
if not fatal_if_no_video:
|
||||||
return webpage, False
|
return webpage, False
|
||||||
@ -276,7 +259,8 @@ def video_data_list2dict(video_data):
|
|||||||
raise ExtractorError('Cannot parse data')
|
raise ExtractorError('Cannot parse data')
|
||||||
|
|
||||||
formats = []
|
formats = []
|
||||||
for format_id, f in video_data.items():
|
for f in video_data:
|
||||||
|
format_id = f['stream_type']
|
||||||
if f and isinstance(f, dict):
|
if f and isinstance(f, dict):
|
||||||
f = [f]
|
f = [f]
|
||||||
if not f or not isinstance(f, list):
|
if not f or not isinstance(f, list):
|
||||||
|
Loading…
Reference in New Issue
Block a user