mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2025-09-03 08:35:32 +00:00
Code Issues Projects Releases Wiki Activity

[extractors] Use new framework for existing embeds (#4307)

Browse Source 
`Brightcove` is difficult to migrate because it's subclasses may depend
on the signature of the current functions. So it is left as-is for now

Note: Tests have not been migrated
This commit is contained in:
pukkandan
2022-08-01 06:53:25 +05:30
parent 1e8fe57e5c
commit bfd973ece3
138 changed files with 499 additions and 1909 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
yt_dlp/extractor/eagleplatform.py
View File

@ -1,3 +1,4 @@
import functools
import re
from .common import InfoExtractor
@ -5,6 +6,7 @@ from ..compat import compat_HTTPError
from ..utils import (
ExtractorError,
int_or_none,
smuggle_url,
unsmuggle_url,
url_or_none,
)
@ -18,6 +20,7 @@ class EaglePlatformIE(InfoExtractor):
)
(?P<id>\d+)
'''
_EMBED_REGEX = [r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1']
_TESTS = [{
# http://lenta.ru/news/2015/03/06/navalny/
'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201',
@ -52,14 +55,14 @@ class EaglePlatformIE(InfoExtractor):
'only_matching': True,
}]
@staticmethod
def _extract_url(webpage):
# Regular iframe embedding
mobj = re.search(
r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1',
webpage)
if mobj is not None:
return mobj.group('url')
@classmethod
def _extract_embed_urls(cls, url, webpage):
add_referer = functools.partial(smuggle_url, data={'referrer': url})
res = tuple(super()._extract_embed_urls(url, webpage))
if res:
return map(add_referer, res)
PLAYER_JS_RE = r'''
<script[^>]+
src=(?P<qjs>["\'])(?:https?:)?//(?P<host>(?:(?!(?P=qjs)).)+\.media\.eagleplatform\.com)/player/player\.js(?P=qjs)
@ -74,7 +77,7 @@ class EaglePlatformIE(InfoExtractor):
data-id=["\'](?P<id>\d+)
''' % PLAYER_JS_RE, webpage)
if mobj is not None:
return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
return [add_referer('eagleplatform:%(host)s:%(id)s' % mobj.groupdict())]
# Generalization of "Javascript code usage", "Combined usage" and
# "Usage without attaching to DOM" embeddings (see
# http://dultonmedia.github.io/eplayer/)
@ -95,7 +98,7 @@ class EaglePlatformIE(InfoExtractor):
</script>
''' % PLAYER_JS_RE, webpage)
if mobj is not None:
return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
return [add_referer('eagleplatform:%(host)s:%(id)s' % mobj.groupdict())]
@staticmethod
def _handle_error(response):
@ -201,3 +204,14 @@ class EaglePlatformIE(InfoExtractor):
'age_limit': age_limit,
'formats': formats,
}
class ClipYouEmbedIE(InfoExtractor):
_VALID_URL = False
@classmethod
def _extract_embed_urls(cls, url, webpage):
mobj = re.search(
r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
if mobj is not None:
yield smuggle_url('eagleplatform:%(host)s:%(id)s' % mobj.groupdict(), {'referrer': url})