def scrapeWebPageForVideoLinks(data):
logger.info("")
links = {}
fmt_value = {
5: "240p h263 flv",
6: "240p h263 flv",
18: "360p h264 mp4",
22: "720p h264 mp4",
26: "???",
33: "???",
34: "360p h264 flv",
35: "480p h264 flv",
36: "3gpp",
37: "1080p h264 mp4",
38: "4K h264 mp4",
43: "360p vp8 webm",
44: "480p vp8 webm",
45: "720p vp8 webm",
46: "1080p vp8 webm",
59: "480p h264 mp4",
78: "480p h264 mp4",
82: "360p h264 3D",
83: "480p h264 3D",
84: "720p h264 3D",
85: "1080p h264 3D",
100: "360p vp8 3D",
101: "480p vp8 3D",
102: "720p vp8 3D"
}
video_urls=[]
flashvars = extractFlashVars(data)
if not flashvars.has_key(u"url_encoded_fmt_stream_map"):
return links
if flashvars.has_key(u"ttsurl"):
logger.info("ttsurl="+flashvars[u"ttsurl"])
if flashvars.has_key('hlsvp'):
url = flashvars[u"hlsvp"]
video_urls.append( [ "(LIVE .m3u8) [youtube]" , url ])
return video_urls
js_signature = ""
data_flashvars = flashvars[u"url_encoded_fmt_stream_map"].split(u",")
for url_desc in data_flashvars:
url_desc_map = cgi.parse_qs(url_desc)
logger.info(u"url_map: " + repr(url_desc_map))
if not (url_desc_map.has_key(u"url") or url_desc_map.has_key(u"stream")):
continue
try:
key = int(url_desc_map[u"itag"][0])
if not fmt_value.get(key):
continue
url = u""
if url_desc_map.has_key(u"url"):
url = urllib.unquote(url_desc_map[u"url"][0])
elif url_desc_map.has_key(u"conn") and url_desc_map.has_key(u"stream"):
url = urllib.unquote(url_desc_map[u"conn"][0])
if url.rfind("/") < len(url) -1:
url = url + "/"
url = url + urllib.unquote(url_desc_map[u"stream"][0])
elif url_desc_map.has_key(u"stream") and not url_desc_map.has_key(u"conn"):
url = urllib.unquote(url_desc_map[u"stream"][0])
if url_desc_map.has_key(u"sig"):
url = url + u"&signature=" + url_desc_map[u"sig"][0]
elif url_desc_map.has_key(u"s"):
sig = url_desc_map[u"s"][0]
if not js_signature:
urljs = scrapertools.find_single_match(data, '"assets":.*?"js":\s*"([^"]+)"')
urljs = urljs.replace("\\", "")
if urljs:
data_js = scrapertools.downloadpage("http:"+urljs)
from jsinterpreter import JSInterpreter
funcname = scrapertools.find_single_match(data_js, '\.sig\|\|([A-z0-9$]+)\(')
jsi = JSInterpreter(data_js)
js_signature = jsi.extract_function(funcname)
signature = js_signature([sig])
url += u"&signature=" + signature
# Se encodean las comas para que no falle en método built-in
url = url.replace(",", "%2C")
video_urls.append( [ "("+fmt_value[key]+") [youtube]" , url ])
except:
import traceback
logger.info(traceback.format_exc())
return video_urls
def extract_videos(video_id):
fmt_value = {
5: "240p h263 flv",
6: "240p h263 flv",
18: "360p h264 mp4",
22: "720p h264 mp4",
26: "???",
33: "???",
34: "360p h264 flv",
35: "480p h264 flv",
36: "3gpp",
37: "1080p h264 mp4",
38: "4K h264 mp4",
43: "360p vp8 webm",
44: "480p vp8 webm",
45: "720p vp8 webm",
46: "1080p vp8 webm",
59: "480p h264 mp4",
78: "480p h264 mp4",
82: "360p h264 3D",
83: "480p h264 3D",
84: "720p h264 3D",
85: "1080p h264 3D",
100: "360p vp8 3D",
101: "480p vp8 3D",
102: "720p vp8 3D"
}
url = 'http://www.youtube.com/get_video_info?video_id=%s&eurl=https://youtube.googleapis.com/v/%s&ssl_stream=1' % \
(video_id, video_id)
data = httptools.downloadpage(url).data
video_urls = []
params = dict(urlparse.parse_qsl(data))
if params.get('hlsvp'):
video_urls.append(["(LIVE .m3u8) [youtube]", params['hlsvp']])
return video_urls
if config.is_xbmc():
import xbmc
xbmc_version = int(
xbmc.getInfoLabel("System.BuildVersion").split(".", 1)[0])
if xbmc_version > 16 and xbmc.getCondVisibility('System.HasAddon(inputstream.adaptive)') \
and params.get('dashmpd'):
if params.get('use_cipher_signature', '') != 'True':
video_urls.append(
['mpd HD [youtube]', params['dashmpd'], 0, '', True])
js_signature = ""
youtube_page_data = httptools.downloadpage(
"http://www.youtube.com/watch?v=%s" % video_id).data
params = extract_flashvars(youtube_page_data)
if params.get('url_encoded_fmt_stream_map'):
data_flashvars = params["url_encoded_fmt_stream_map"].split(",")
for url_desc in data_flashvars:
url_desc_map = dict(urlparse.parse_qsl(url_desc))
if not url_desc_map.get("url") and not url_desc_map.get("stream"):
continue
try:
key = int(url_desc_map["itag"])
if not fmt_value.get(key):
continue
if url_desc_map.get("url"):
url = urllib.unquote(url_desc_map["url"])
elif url_desc_map.get("conn") and url_desc_map.get("stream"):
url = urllib.unquote(url_desc_map["conn"])
if url.rfind("/") < len(url) - 1:
url += "/"
url += urllib.unquote(url_desc_map["stream"])
elif url_desc_map.get(
"stream") and not url_desc_map.get("conn"):
url = urllib.unquote(url_desc_map["stream"])
if url_desc_map.get("sig"):
url += "&signature=" + url_desc_map["sig"]
elif url_desc_map.get("s"):
sig = url_desc_map["s"]
if not js_signature:
urljs = scrapertools.find_single_match(
youtube_page_data, '"assets":.*?"js":\s*"([^"]+)"')
urljs = urljs.replace("\\", "")
if urljs:
if not re.search(r'https?://', urljs):
urljs = urlparse.urljoin(
"https://www.youtube.com", urljs)
data_js = httptools.downloadpage(urljs).data
from jsinterpreter import JSInterpreter
funcname = scrapertools.find_single_match(
data_js, '\.sig\|\|([A-z0-9$]+)\(')
if not funcname:
funcname = scrapertools.find_single_match(
data_js, '["\']signature["\']\s*,\s*'
'([A-z0-9$]+)\(')
jsi = JSInterpreter(data_js)
js_signature = jsi.extract_function(funcname)
signature = js_signature([sig])
url += "&signature=" + signature
url = url.replace(",", "%2C")
video_urls.append(["(" + fmt_value[key] + ") [youtube]", url])
except:
import traceback
logger.info(traceback.format_exc())
return video_urls
def get_video_url(item):
logger.trace()
itemlist = []
video_id = scrapertools.find_single_match(item.url, 'v=([A-z0-9_-]{11})')
url = 'http://www.youtube.com/get_video_info?video_id=%s&eurl=https://youtube.googleapis.com/v/%s&ssl_stream=1' % (
video_id, video_id)
data = httptools.downloadpage(url).data
params = dict(urlparse.parse_qsl(data))
if params.get('hlsvp'):
itemlist.append(Video(type='Live', url=params['hlsvp']))
return itemlist
if params.get('dashmpd') and params.get('use_cipher_signature',
'') != 'True':
itemlist.append(Video(type='MPD', url=params['dashmpd'], mpd=True))
js_signature = ""
youtube_page_data = httptools.downloadpage(
"http://www.youtube.com/watch?v=%s" % video_id).data
params = extract_flashvars(youtube_page_data)
if params.get('url_encoded_fmt_stream_map'):
data_flashvars = params["url_encoded_fmt_stream_map"].split(",")
for url_desc in data_flashvars:
url_desc_map = dict(urlparse.parse_qsl(url_desc))
if not url_desc_map.get("url") and not url_desc_map.get("stream"):
continue
try:
key = int(url_desc_map["itag"])
if not fmt_value.get(key):
continue
if url_desc_map.get("url"):
url = urllib.unquote(url_desc_map["url"])
elif url_desc_map.get("conn") and url_desc_map.get("stream"):
url = urllib.unquote(url_desc_map["conn"])
if url.rfind("/") < len(url) - 1:
url += "/"
url += urllib.unquote(url_desc_map["stream"])
elif url_desc_map.get(
"stream") and not url_desc_map.get("conn"):
url = urllib.unquote(url_desc_map["stream"])
if url_desc_map.get("sig"):
url += "&signature=" + url_desc_map["sig"]
elif url_desc_map.get("s"):
sig = url_desc_map["s"]
if not js_signature:
urljs = scrapertools.find_single_match(
youtube_page_data, '"assets":.*?"js":\s*"([^"]+)"')
urljs = urljs.replace("\\", "")
if urljs:
if not re.search(r'https?://', urljs):
urljs = urlparse.urljoin(
"https://www.youtube.com", urljs)
data_js = httptools.downloadpage(urljs).data
from jsinterpreter import JSInterpreter
funcname = scrapertools.find_single_match(
data_js, '\.sig\|\|([A-z0-9$]+)\(')
if not funcname:
funcname = scrapertools.find_single_match(
data_js, '["\']signature["\']\s*,\s*'
'([A-z0-9$]+)\(')
jsi = JSInterpreter(data_js)
js_signature = jsi.extract_function(funcname)
signature = js_signature([sig])
url += "&signature=" + signature
url = url.replace(",", "%2C")
itemlist.append(
Video(type=fmt_value[key][1],
res=fmt_value[key][0],
url=url))
except:
logger.error()
return itemlist
