Mirrors/yt-dlp
2
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-15 07:42:42 +00:00
Code Issues Projects Releases Packages Wiki Activity

[pornhub] Fix extraction and add support for m3u8 formats (closes #22749, closes #23082)

Browse source
This commit is contained in:
Sergey M․ 2019-12-31 23:29:06 +07:00
parent 0606808746
commit f41347260c
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D
1 changed files with 37 additions and 10 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

47
youtube_dl/extractor/pornhub.py
View file

@ -227,12 +227,13 @@ class PornHubIE(PornHubBaseIE):
else: else:
thumbnail, duration = [None] * 2 thumbnail, duration = [None] * 2
if not video_urls: def extract_js_vars(webpage, pattern, fatal=True):
tv_webpage = dl_webpage('tv')
assignments = self._search_regex( assignments = self._search_regex(
r'(var.+?mediastring.+?)</script>', tv_webpage, pattern, webpage, 'encoded url', fatal=fatal)
'encoded url').split(';') if not assignments:
return {}
assignments = assignments.split(';')
js_vars = {} js_vars = {}
@ -254,11 +255,31 @@ class PornHubIE(PornHubBaseIE):
assn = re.sub(r'var\s+', '', assn) assn = re.sub(r'var\s+', '', assn)
vname, value = assn.split('=', 1) vname, value = assn.split('=', 1)
js_vars[vname] = parse_js_value(value) js_vars[vname] = parse_js_value(value)
return js_vars
video_url = js_vars['mediastring'] def add_video_url(video_url):
if video_url not in video_urls_set: v_url = url_or_none(video_url)
video_urls.append((video_url, None)) if not v_url:
video_urls_set.add(video_url) return
if v_url in video_urls_set:
return
video_urls.append((v_url, None))
video_urls_set.add(v_url)
if not video_urls:
FORMAT_PREFIXES = ('media', 'quality')
js_vars = extract_js_vars(
webpage, r'(var\s+(?:%s)_.+)' % '|'.join(FORMAT_PREFIXES),
fatal=False)
if js_vars:
for key, format_url in js_vars.items():
if any(key.startswith(p) for p in FORMAT_PREFIXES):
add_video_url(format_url)
if not video_urls:
js_vars = extract_js_vars(
dl_webpage('tv'), r'(var.+?mediastring.+?)</script>')
add_video_url(js_vars['mediastring'])
for mobj in re.finditer( for mobj in re.finditer(
r'<a[^>]+\bclass=["\']downloadBtn\b[^>]+\bhref=(["\'])(?P<url>(?:(?!\1).)+)\1', r'<a[^>]+\bclass=["\']downloadBtn\b[^>]+\bhref=(["\'])(?P<url>(?:(?!\1).)+)\1',
@ -276,10 +297,16 @@ class PornHubIE(PornHubBaseIE):
r'/(\d{6}/\d{2})/', video_url, 'upload data', default=None) r'/(\d{6}/\d{2})/', video_url, 'upload data', default=None)
if upload_date: if upload_date:
upload_date = upload_date.replace('/', '') upload_date = upload_date.replace('/', '')
if determine_ext(video_url) == 'mpd': ext = determine_ext(video_url)
if ext == 'mpd':
formats.extend(self._extract_mpd_formats( formats.extend(self._extract_mpd_formats(
video_url, video_id, mpd_id='dash', fatal=False)) video_url, video_id, mpd_id='dash', fatal=False))
continue continue
elif ext == 'm3u8':
formats.extend(self._extract_m3u8_formats(
video_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False))
continue
tbr = None tbr = None
mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', video_url) mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', video_url)
if mobj: if mobj: