def tudou_download(url, **kwargs):
if 'video.tudou.com' in url:
vid = match1(url, r'.*?video.tudou.com/v/([\w=]+)')
else:
page = get_content(url)
video_info = json.loads(
match1(page, r'window.__INITIAL_STATE__=\s*(.+?);</script>'))
vid = video_info['videoDesc']['detail']['videoid']
youku_download_by_vid(vid, **kwargs)
def entry(self, **kwargs):
# tencent player
tc_flashvars = re.search(
r'"bili-cid=\d+&bili-aid=\d+&vid=([^"]+)"', self.page
)
if tc_flashvars:
tc_flashvars = tc_flashvars.group(1)
if tc_flashvars is not None:
self.out = True
qq_download_by_vid(
tc_flashvars, self.title, output_dir=kwargs['output_dir'],
merge=kwargs['merge'], info_only=kwargs['info_only']
)
return
has_plist = re.search(r'<option', self.page)
if has_plist and r1('index_(\d+).html', self.url) is None:
log.w(
'This page contains a playlist. (use --playlist to download '
'all videos.)'
)
try:
cid = re.search(r'cid=(\d+)', self.page).group(1)
except Exception:
cid = re.search(r'"cid":(\d+)', self.page).group(1)
if cid is not None:
self.download_by_vid(
cid, re.search('bangumi', self.url) is not None, **kwargs
)
else:
# flashvars?
flashvars = re.search(r'flashvars="([^"]+)"', self.page).group(1)
if flashvars is None:
raise Exception('Unsupported page {}'.format(self.url))
param = flashvars.split('&')[0]
t, cid = param.split('=')
t = t.strip()
cid = cid.strip()
if t == 'vid':
sina_download_by_vid(
cid, self.title, output_dir=kwargs['output_dir'],
merge=kwargs['merge'], info_only=kwargs['info_only']
)
elif t == 'ykid':
youku_download_by_vid(
cid, self.title, output_dir=kwargs['output_dir'],
merge=kwargs['merge'], info_only=kwargs['info_only']
)
else:
raise NotImplementedError(
'Unknown flashvars {}'.format(flashvars)
)
return
def acfun_download_by_vid(vid,
title,
output_dir='.',
merge=True,
info_only=False,
**kwargs):
"""str, str, str, bool, bool -> None
Download Acfun video by vid.
Call Acfun API, decide which site to use, and pass the job to its
extractor.
"""
# first call the main parasing API
info = json.loads(
get_content(
'http://www.acfun.cn/video/getVideo.aspx?id={}'.format(vid)))
sourceType = info['sourceType']
# decide sourceId to know which extractor to use
if 'sourceId' in info:
sourceId = info['sourceId']
# danmakuId = info['danmakuId']
# call extractor decided by sourceId
if sourceType == 'sina':
sina_download_by_vid(sourceId,
title,
output_dir=output_dir,
merge=merge,
info_only=info_only)
elif sourceType == 'youku':
youku_download_by_vid(sourceId,
title=title,
output_dir=output_dir,
merge=merge,
info_only=info_only,
**kwargs)
elif sourceType == 'qq':
qq_download_by_vid(sourceId,
title,
output_dir=output_dir,
merge=merge,
info_only=info_only)
elif sourceType == 'letv':
letvcloud_download_by_vu(sourceId,
'2d8c027396',
title,
output_dir=output_dir,
merge=merge,
info_only=info_only)
elif sourceType == 'zhuzhan':
# As in Jul.28.2016, Acfun is using embsig to anti hotlink so we need
# to pass this In Mar. 2017 there is a dedicated ``acfun_proxy'' in
# youku cloud player old code removed
url = 'http://www.acfun.cn/v/ac{}'.format(vid)
yk_streams = youku_acfun_proxy(info['sourceId'], info['encode'], url)
seq = ['mp4hd3', 'mp4hd2', 'mp4hd', 'flvhd']
for t in seq:
if yk_streams.get(t):
preferred = yk_streams[t]
break
# total_size in the json could be incorrect(F.I. 0)
size = 0
for url in preferred[0]:
_, _, seg_size = url_info(url)
size += seg_size
# fallback to flvhd is not quite possible
print_info(site_info, title, 'mp4', size)
if not info_only:
download_urls(preferred[0],
title,
'mp4',
size,
output_dir=output_dir,
merge=merge)
else:
raise NotImplementedError(sourceType)
if not info_only and not dry_run:
if not kwargs['caption']:
print('Skipping danmaku.')
return
try:
title = get_filename(title)
print('Downloading %s ...\n' % (title + '.cmt.json'))
cmt = get_srt_json(vid)
with open(os.path.join(output_dir, title + '.cmt.json'),
'w',
encoding='utf-8') as x:
x.write(cmt)
except Exception:
pass
def embed_download(url, output_dir='.', merge=True, info_only=False, **kwargs):
content = get_content(url)
found = False
title = match1(content, '<title>([^<>]+)</title>')
vids = matchall(content, youku_embed_patterns)
for vid in set(vids):
found = True
youku_download_by_vid(
vid, title=title, output_dir=output_dir, merge=merge,
info_only=info_only
)
vids = matchall(content, yinyuetai_embed_patterns)
for vid in vids:
found = True
yinyuetai_download_by_id(
vid, title=title, output_dir=output_dir, merge=merge,
info_only=info_only
)
vids = matchall(content, iqiyi_embed_patterns)
for vid in vids:
found = True
iqiyi_download_by_vid(
(vid[1], vid[0]), title=title, output_dir=output_dir, merge=merge,
info_only=info_only
)
urls = matchall(content, netease_embed_patterns)
for url in urls:
found = True
netease_download(
url, output_dir=output_dir, merge=merge, info_only=info_only
)
urls = matchall(content, vimeo_embed_patters)
for url in urls:
found = True
vimeo_download_by_id(
url, title=title, output_dir=output_dir, merge=merge,
info_only=info_only, referer=url
)
urls = matchall(content, dailymotion_embed_patterns)
for url in urls:
found = True
dailymotion_download(
url, output_dir=output_dir, merge=merge, info_only=info_only
)
aids = matchall(content, bilibili_embed_patterns)
for aid in aids:
found = True
url = 'http://www.bilibili.com/video/av{}/'.format(aid)
bilibili_download(
url, output_dir=output_dir, merge=merge, info_only=info_only
)
iqiyi_urls = matchall(content, iqiyi_patterns)
for url in iqiyi_urls:
found = True
iqiyi.download(
url, output_dir=output_dir, merge=merge, info_only=info_only,
**kwargs
)
bokecc_metas = matchall(content, bokecc_patterns)
for meta in bokecc_metas:
found = True
bokecc.bokecc_download_by_id(
meta[1], output_dir=output_dir, merge=merge, info_only=info_only,
**kwargs
)
if found:
return True
# Try harder, check all iframes
if 'recur_lv' not in kwargs or kwargs['recur_lv'] < recur_limit:
r = kwargs.get('recur_lv')
if r is None:
r = 1
else:
r += 1
iframes = matchall(content, [r'<iframe.+?src=(?:\"|\')(.+?)(?:\"|\')'])
for iframe in iframes:
if not iframe.startswith('http'):
src = urllib.parse.urljoin(url, iframe)
else:
src = iframe
found = embed_download(
src, output_dir=output_dir, merge=merge, info_only=info_only,
recur_lv=r, **kwargs
)
if found:
return True
if not found and 'recur_lv' not in kwargs:
raise NotImplementedError(url)
else:
return found