def _get_streams(self):
res = http.get(self.url)
match = _player_js.search(res.text)
if match:
player_js = match.group(0)
self.logger.info("Found player js {0}", player_js)
else:
self.logger.info(
"Didn't find player js. Probably this page doesn't contain a video"
)
return
res = http.get(player_js)
jsonp_start = res.text.find('(') + 1
jsonp_end = res.text.rfind(')')
if jsonp_start <= 0 or jsonp_end <= 0:
self.logger.info(
"Couldn't extract json metadata from player.js: {0}",
player_js)
return
json_s = res.text[jsonp_start:jsonp_end]
stream_metadata = json.loads(json_s)
return HDSStream.parse_manifest(
self.session,
stream_metadata['mediaResource']['dflt']['videoURL']).items()
def _get_streams(self):
# Discover root
match = _url_re.search(self.url)
root = match.group(1)
# Download main URL
res = http.get(self.url)
# Find playlist
match = _playlist_re.search(res.text)
playlist_url = root + match.group(1) + "d"
# Download playlist
res = http.get(playlist_url)
# Find manifest
match = _manifest_re.search(res.text)
manifest_url = match.group(1)
# Find SWF
match = _swf_re.search(res.text)
swf_url = match.group(1)
streams = {}
streams.update(
HDSStream.parse_manifest(self.session, manifest_url,
pvswf=swf_url))
return streams
def _get_vod_stream(self):
vod_url = self.url
if vod_url.endswith('/'):
vod_url = vod_url[:-1]
json_url = '{0}.securevideo.json'.format(vod_url)
res = http.get(json_url)
match = _json_re.search(res.text)
if not match:
return
data = parse_json(match.group(1))
res = http.get(API_VOD.format(data['clientid'], data['mzid']))
data = http.json(res, schema=_stream_schema)
for d in data['targetUrls']:
if d['type'] == 'HDS':
hds_url = d['url']
for s in HDSStream.parse_manifest(self.session,
hds_url).items():
yield s
if d['type'] == 'HLS':
hls_url = d['url']
for s in HLSStream.parse_variant_playlist(
self.session, hls_url).items():
yield s
def _get_streams(self):
data_url = http.get(self.url, schema=self._player_url_schema)
if data_url:
res = http.get(urljoin(self.url, data_url))
stream_info = http.xml(res, schema=self._livestream_schema)
for stream in stream_info:
url = stream["url"]
try:
if ".m3u8" in url:
for s in HLSStream.parse_variant_playlist(
self.session, url, name_key="bitrate").items():
yield s
elif ".f4m" in url:
for s in HDSStream.parse_manifest(
self.session,
url,
pvswf=self.swf_url,
is_akamai=True).items():
yield s
elif ".mp4" in url:
yield "{0}k".format(stream["bitrate"]), HTTPStream(
self.session, url)
except IOError as err:
self.logger.warning("Error parsing stream: {0}", err)
def _get_vod_streams(self, params):
manifest_url = params.get("autoURL")
if not manifest_url:
return
res = http.get(manifest_url)
if res.headers.get("Content-Type") == "application/f4m+xml":
streams = HDSStream.parse_manifest(self.session, res.url)
# TODO: Replace with "yield from" when dropping Python 2.
for __ in streams.items():
yield __
elif res.headers.get(
"Content-Type") == "application/vnd.apple.mpegurl":
streams = HLSStream.parse_variant_playlist(self.session, res.url)
# TODO: Replace with "yield from" when dropping Python 2.
for __ in streams.items():
yield __
else:
manifest = http.json(res, schema=_vod_manifest_schema)
for params in manifest["alternates"]:
name = "{0}p".format(params["height"])
stream = self._create_flv_playlist(params["template"])
yield name, stream
failovers = params.get("failover", [])
for failover in failovers:
stream = self._create_flv_playlist(failover)
yield name, stream
def _get_streams(self):
url, params = parse_url_params(self.url)
urlnoproto = self._url_re.match(url).group(2)
urlnoproto = update_scheme("http://", urlnoproto)
return HDSStream.parse_manifest(self.session, urlnoproto, **params)
def _resolve_playlist(self, res, playlist_all):
""" yield for _resolve_res
Args:
res: Content from self._res_text
playlist_all: List of streams
Returns:
yield every stream
"""
# m_base is used for .f4m files that doesn't have a base_url
m_base = self._stream_base_re.search(res)
if m_base:
stream_base = m_base.group("base")
else:
stream_base = ""
playlist_list = self._make_url_list(playlist_all, self.url,
stream_base)
self.logger.debug("Found URL: {0}".format(", ".join(playlist_list)))
endswith_blacklist = (".mp3", ".mp4", ".vtt")
for url in playlist_list:
if ".m3u8" in url and not url.endswith(endswith_blacklist):
try:
streams = HLSStream.parse_variant_playlist(
self.session, url, headers=self.headers).items()
if not streams:
yield "live", HLSStream(self.session,
url,
headers=self.headers)
for s in streams:
yield s
except Exception:
self.logger.error("Skipping hls_url: {0}".format(url))
elif ".f4m" in url and not url.endswith(endswith_blacklist):
try:
for s in HDSStream.parse_manifest(
self.session, url, headers=self.headers).items():
yield s
except Exception:
self.logger.error("Skipping hds_url: {0}".format(url))
elif self.list_in_item(url, [
".mp3", ".mp4"
]) and not self.list_in_item(url, [".f4m", ".m3u8", ".mpd"]):
try:
name = "live"
m = self._httpstream_bitrate_re.search(url)
if m:
name = "{0}k".format(m.group("bitrate"))
yield name, HTTPStream(self.session,
url,
headers=self.headers)
except Exception:
self.logger.error("Skipping http_url: {0}".format(url))
elif ".mpd" in url and not url.endswith(endswith_blacklist):
try:
self.logger.info("Found mpd: {0}".format(url))
except Exception:
self.logger.error("Skipping mpd_url: {0}".format(url))
def _get_streams(self):
# Get video ID and channel from URL
match = self._url_re.match(self.url)
video_id = match.group('video_id')
if video_id is None:
# Retrieve URL page and search for video ID
res = http.get(self.url)
match = self._video_id_re.search(res.text)
if match is None:
return
video_id = match.group('video_id')
res = http.get(self.API_URL.format(video_id))
videos = http.json(res, schema=self._api_schema)
parsed = []
headers = {'User-Agent': self._user_agent}
# Some videos may be also available on Dailymotion (especially on CNews)
if videos['ID_DM'] != '':
for stream in self.session.streams('https://www.dailymotion.com/video/' + videos['ID_DM']).items():
yield stream
for quality, video_url in list(videos['MEDIA']['VIDEOS'].items()):
# Ignore empty URLs
if video_url == '':
continue
# Ignore duplicate video URLs
if video_url in parsed:
continue
parsed.append(video_url)
try:
# HDS streams don't seem to work for live videos
if '.f4m' in video_url and 'LIVE' not in videos['TYPE']:
for stream in HDSStream.parse_manifest(self.session,
video_url,
params={'hdcore': self.HDCORE_VERSION},
headers=headers).items():
yield stream
elif '.m3u8' in video_url:
for stream in HLSStream.parse_variant_playlist(self.session,
video_url,
headers=headers).items():
yield stream
elif '.mp4' in video_url:
# Get bitrate from video filename
match = self._mp4_bitrate_re.match(video_url)
if match is not None:
bitrate = match.group('bitrate')
else:
bitrate = quality
yield bitrate, HTTPStream(self.session,
video_url,
params={'secret': self.SECRET},
headers=headers)
except IOError as err:
if '403 Client Error' in str(err):
self.logger.error('Failed to access stream, may be due to geo-restriction')
def _get_streams(self):
res = http.get(self.url)
match = _meta_xmlurl_id_re.search(res.text)
if not match:
return
xml_info_url = STREAMS_INFO_URL.format(match.group(1))
video_info_res = http.get(xml_info_url)
parsed_info = http.xml(video_info_res)
live_el = parsed_info.find("live")
live = live_el is not None and live_el.text == "1"
streams = {}
hdsurl_el = parsed_info.find("hdsurl")
if hdsurl_el is not None and hdsurl_el.text is not None:
hdsurl = hdsurl_el.text
streams.update(HDSStream.parse_manifest(self.session, hdsurl))
if live:
vurls_el = parsed_info.find("vurls")
if vurls_el is not None:
for i, vurl_el in enumerate(vurls_el):
bitrate = vurl_el.get("bitrate")
name = bitrate + "k" if bitrate is not None else "rtmp{0}".format(
i)
params = {
"rtmp": vurl_el.text,
}
streams[name] = RTMPStream(self.session, params)
parsed_urls = set()
mobileurls_el = parsed_info.find("mobileurls")
if mobileurls_el is not None:
for mobileurl_el in mobileurls_el:
text = mobileurl_el.text
if not text:
continue
if text in parsed_urls:
continue
parsed_urls.add(text)
url = urlparse(text)
if url[0] == "http" and url[2].endswith("m3u8"):
streams.update(
HLSStream.parse_variant_playlist(self.session, text))
return streams
def _get_smil_streams(self, info):
res = http.get(info["_stream"])
smil = http.xml(res, "SMIL config", schema=_smil_schema)
for video in smil["videos"]:
url = "{0}/{1}{2}".format(smil["base"], video, HDCORE_PARAMETER)
streams = HDSStream.parse_manifest(
self.session,
url,
pvswf=SWF_URL,
is_akamai=smil["cdn"] == "akamai")
for stream in streams.items():
yield stream
def _get_streams(self):
res = http.get(self.url)
match = self.player_re.search(res.text)
if match:
channel_id = match.group(1)
stream_api = self.stream_api.format(
id=channel_id,
time=int(time.time()),
hash=''.join(
random.choice("abcdef0123456789") for x in range(28)))
res = http.get(stream_api)
f4m_url = self.manifest_re.search(res.text)
if f4m_url:
return HDSStream.parse_manifest(self.session, f4m_url.group(1))
def mediaselector(self, vpid):
for platform in self.platforms:
url = self.api_url.format(vpid=vpid,
vpid_hash=self._hash_vpid(vpid),
platform=platform)
self.logger.debug("Info API request: {0}", url)
stream_urls = http.get(url, schema=self.mediaselector_schema)
for media in stream_urls:
for connection in media["connection"]:
if connection.get("transferFormat") == "hds":
for s in HDSStream.parse_manifest(
self.session, connection["href"]).items():
yield s
if connection.get("transferFormat") == "hls":
for s in HLSStream.parse_variant_playlist(
self.session, connection["href"]).items():
yield s
def _resolve_playlist(self, res, playlist_all):
""" yield for _resolve_res
Args:
res: Content from self._res_text
playlist_all: List of streams
Returns:
yield every stream
"""
for url in playlist_all:
parsed_url = urlparse(url)
if parsed_url.path.endswith((".m3u8")):
try:
streams = HLSStream.parse_variant_playlist(self.session, url, headers=self.headers).items()
if not streams:
yield "live", HLSStream(self.session, url, headers=self.headers)
for s in streams:
yield s
except Exception as e:
self.logger.error("Skipping hls_url - {0}".format(str(e)))
self.help_info_e(e)
elif parsed_url.path.endswith((".f4m")):
try:
for s in HDSStream.parse_manifest(self.session, url, headers=self.headers).items():
yield s
except Exception as e:
self.logger.error("Skipping hds_url - {0}".format(str(e)))
self.help_info_e(e)
elif parsed_url.path.endswith((".mp3", ".mp4")):
try:
name = "live"
m = self._httpstream_bitrate_re.search(url)
if m:
name = "{0}k".format(m.group("bitrate"))
yield name, HTTPStream(self.session, url, headers=self.headers)
except Exception as e:
self.logger.error("Skipping http_url - {0}".format(str(e)))
self.help_info_e(e)
elif parsed_url.path.endswith((".mpd")):
try:
self.logger.info("Found mpd: {0}".format(url))
except Exception as e:
self.logger.error("Skipping mpd_url - {0}".format(str(e)))
self.help_info_e(e)
def _get_hds_streams(self, channel):
channel = self.hds_channel_remap.get(channel,
"{0}live".format(channel))
manifest_url = http.get(self.api_url.format(channel),
params={
"getURL": 1
},
headers={
"User-Agent": useragents.FIREFOX
}).text
for s in HDSStream.parse_manifest(self.session,
manifest_url,
pvswf=self.swf_url,
headers={
"User-Agent": useragents.FIREFOX
}).items():
yield s
def _create_stream(self, stream, language):
stream_name = "{0}p".format(stream["height"])
stream_type = stream["mediaType"]
stream_url = stream["url"]
stream_language = stream["versionShortLibelle"]
if language == "de":
language = ["DE", "VOST-DE", "VA", "VOA", "Dt. Live", "OV", "OmU"]
elif language == "en":
language = ["ANG", "VOST-ANG"]
elif language == "es":
language = ["ESP", "VOST-ESP"]
elif language == "fr":
language = [
"FR", "VOST-FR", "VF", "VOF", "Frz. Live", "VO", "ST mal"
]
elif language == "pl":
language = ["POL", "VOST-POL"]
if stream_language in language:
if stream_type in ("hls", "mp4"):
if urlparse(stream_url).path.endswith("m3u8"):
try:
streams = HLSStream.parse_variant_playlist(
self.session, stream_url)
for stream in streams.items():
yield stream
except IOError as err:
self.logger.error("Failed to extract HLS streams: {0}",
err)
else:
yield stream_name, HTTPStream(self.session, stream_url)
elif stream_type == "f4m":
try:
streams = HDSStream.parse_manifest(self.session,
stream_url)
for stream in streams.items():
yield stream
except IOError as err:
self.logger.error("Failed to extract HDS streams: {0}",
err)
def _get_streams(self):
m_url = self._url_re.match(self.url)
if not m_url:
return
video_id = m_url.group("id")
appname = self.appname_map[m_url.group("host")]
apiname = self.apiname_map[appname]
headers = {
"User-Agent": useragents.FIREFOX,
"Referer": self.url
}
res = http.get(self.api_url.format(apiname, video_id, appname), headers=headers)
data = http.json(res, schema=self._video_schema)
title = data.get("title")
streamurls = data.get("streamUrls")
if title:
self.stream_title = title
if streamurls:
hls_url = streamurls.get("hls")
if hls_url:
self.logger.debug("HLS URL: {0}".format(hls_url))
streams = HLSStream.parse_variant_playlist(self.session, hls_url, headers=headers).items()
if not streams:
yield "live", HLSStream(self.session, hls_url, headers=headers)
for s in streams:
yield s
hds_url = streamurls.get("hds")
if hds_url:
self.logger.debug("HDS URL: {0}".format(hds_url))
for s in HDSStream.parse_manifest(self.session, hds_url, headers=headers).items():
yield s
mp4_url = streamurls.get("mp4")
if mp4_url:
self.logger.debug("MP4 URL: {0}".format(mp4_url))
name = "live"
yield name, HTTPStream(self.session, mp4_url, headers=headers)
def _get_streams(self):
hls_urls = []
hds_urls = []
http.headers.update({'User-Agent': useragents.FIREFOX})
match = self._url_re.match(self.url)
if match is None:
return
video_id = match.group('id')
self.logger.debug('video_id: {0}'.format(video_id))
res = http.get(self.api_play.format(video_id, self.url))
data = http.json(res, schema=self._video_schema)
live_data = data.get('live_streams')
vod_data = data.get('video_balancer')
if live_data:
self.logger.debug('Found live_data')
for d in live_data['hls']:
hls_urls.append(d['url'])
for e in live_data['hds']:
hds_urls.append(e['url'])
elif vod_data:
self.logger.debug('Found vod_data')
hls_urls.append(vod_data['m3u8'])
hds_urls.append(vod_data['default'])
else:
self.logger.error(
'restricted access to this video for your region')
raise NoStreamsError(self.url)
for hls_url in hls_urls:
self.logger.debug('HLS URL: {0}'.format(hls_url))
for s in HLSStream.parse_variant_playlist(self.session,
hls_url).items():
yield s
for hds_url in hds_urls:
self.logger.debug('HDS URL: {0}'.format(hds_url))
for s in HDSStream.parse_manifest(self.session, hds_url).items():
yield s
def _resolve_playlist(self, playlist_all):
""" create streams
Args:
playlist_all: List of stream urls
Returns:
all streams
"""
http.headers.update({"Referer": self.url})
for url in playlist_all:
parsed_url = urlparse(url)
if parsed_url.path.endswith((".m3u8")):
try:
streams = HLSStream.parse_variant_playlist(
self.session, url).items()
if not streams:
yield "live", HLSStream(self.session, url)
for s in streams:
yield s
except Exception as e:
self.logger.error("Skipping hls_url - {0}".format(str(e)))
elif parsed_url.path.endswith((".f4m")):
try:
for s in HDSStream.parse_manifest(self.session,
url).items():
yield s
except Exception as e:
self.logger.error("Skipping hds_url - {0}".format(str(e)))
elif parsed_url.path.endswith((".mp3", ".mp4")):
try:
name = "live"
m = self._httpstream_bitrate_re.search(url)
if m:
name = "{0}k".format(m.group("bitrate"))
yield name, HTTPStream(self.session, url)
except Exception as e:
self.logger.error("Skipping http_url - {0}".format(str(e)))
elif parsed_url.path.endswith((".mpd")):
try:
self.logger.info("Found mpd: {0}".format(url))
except Exception as e:
self.logger.error("Skipping mpd_url - {0}".format(str(e)))
def _get_streams(self):
match = _url_re.match(self.url)
video_id = match.group("video_id")
res = http.get(ASSET_URL.format(video_id))
assets = http.xml(res, schema=_asset_schema)
streams = {}
for asset in assets:
base = asset["base"]
url = asset["url"]
if urlparse(url).path.endswith(".f4m"):
streams.update(
HDSStream.parse_manifest(self.session, url, pvswf=SWF_URL))
elif base.startswith("rtmp"):
name = "{0}k".format(asset["bitrate"])
params = {"rtmp": asset["base"], "playpath": url, "live": True}
streams[name] = RTMPStream(self.session, params)
return streams
def _get_live_streams(self, params, swf_url):
for key, quality in QUALITY_MAP.items():
key_url = "{0}URL".format(key)
url = params.get(key_url)
if not url:
continue
try:
res = http.get(url, exception=IOError)
except IOError:
continue
if quality == "hds":
self.logger.debug('PLAYLIST URL: {0}'.format(res.url))
try:
streams = HDSStream.parse_manifest(self.session, res.url)
except BaseException:
streams = HLSStream.parse_variant_playlist(
self.session, res.url)
for name, stream in streams.items():
if key == "source":
name += "+"
yield name, stream
elif res.text.startswith("rtmp"):
match = _rtmp_re.match(res.text)
if not match:
continue
stream = RTMPStream(
self.session, {
"rtmp": match.group("host"),
"app": match.group("app"),
"playpath": match.group("playpath"),
"swfVfy": swf_url,
"live": True
})
yield quality, stream
def _get_hds_streams(self, channel):
channel = self.hds_channel_remap.get(channel,
"{0}live".format(channel))
manifest_url = http.get(self.api_url.format(channel),
params={
"getURL": 1
},
headers={
"User-Agent": useragents.FIREFOX
}).text
self.logger.debug("OLD HDS URL: {0}".format(manifest_url))
manifest_url = filter_urlquery(manifest_url, ["hdnea"], True)
self.logger.debug("NEW HDS URL: {0}".format(manifest_url))
for s in HDSStream.parse_manifest(self.session,
manifest_url,
pvswf=self.swf_url,
headers={
"User-Agent": useragents.FIREFOX
}).items():
yield s
def _get_streams(self):
match = self._player_embed_re.match(self.url)
if match:
channel_id = match.group("channel_id")
else:
try:
res = http.get(self.url)
except Exception:
raise NoStreamsError(self.url)
channel_id = self._find_channel_id(res.text)
if not channel_id:
return
self.logger.info("Channel ID: {0}".format(channel_id))
res = http.get(self.PLAYER_EMBED_URL.format(channel_id))
items = self._find_stream_id(res.text)
if not items:
return
a = b = False
for stream in items:
if stream["abr"] == "hls":
try:
for s in HLSStream.parse_variant_playlist(
self.session, stream["streamUrl"]).items():
yield s
except IOError:
a = True
elif stream["abr"] == "hds":
try:
for s in HDSStream.parse_manifest(
self.session, stream["streamUrl"]).items():
yield s
except IOError:
b = True
if a and b:
self.logger.warning(
"Could not open the stream, perhaps the channel is offline")
def _get_streams(self):
args = dict(parse_qsl(urlparse(self.url).query))
if "k" in args:
self.logger.debug("Loading channel: {k}", **args)
res = http.get(self.url)
stream_data_m = self.stream_data_re.search(res.text)
if stream_data_m:
script_vars = b64decode(stream_data_m.group(1)).decode("utf8")
url_m = self.m3u8_re.search(script_vars)
hls_url = url_m and url_m.group("url")
if hls_url:
for s in HLSStream.parse_variant_playlist(
self.session, hls_url).items():
yield s
f4m_m = self.f4mm_re.search(script_vars)
f4m_url = f4m_m and f4m_m.group("url")
if f4m_url:
for n, s in HDSStream.parse_manifest(
self.session, f4m_url).items():
yield n, s
def _get_streams(self):
match = self._url_re.match(self.url)
video_id = match.group('video_id')
if video_id is not None:
# VOD
res = http.get(self.VOD_API_URL.format(video_id))
stream_data = http.json(res, schema=self._vod_api_schema)
# Check whether video format is expired
current_date = datetime.strptime(stream_data['current_date'], '%Y-%m-%dT%H:%M:%S.%f')
valid_start = datetime.strptime(stream_data['shows']['valid_start'], '%Y-%m-%dT%H:%M:%S')
valid_end = datetime.strptime(stream_data['shows']['valid_end'], '%Y-%m-%dT%H:%M:%S')
if current_date < valid_start or current_date > valid_end:
self.logger.error('Failed to access stream, may be due to expired content')
return
streams = stream_data['shows']['media:group']
else:
# Live
channel_id = match.group('channel_id')
# Get live streams for desktop
res = http.get(self.LIVE_API_URL, params={'channelid': channel_id})
streams = http.xml(res, schema=self._live_api_schema)
# Get HLS streams for Iphone
res = http.get(self.LIVE_API_URL, params={'channelid': channel_id, 'platform': 'iphone'})
stream = http.json(res, schema=self._live_api_iphone_schema)
if stream != 'none':
streams.append(stream)
for stream in streams:
if '.f4m' in stream:
for s in HDSStream.parse_manifest(self.session, stream).items():
yield s
if '.m3u8' in stream:
for s in HLSStream.parse_variant_playlist(self.session, stream).items():
yield s
def _get_streams(self):
match = self._url_re.match(self.url)
channel = match.group('channel')
res = http.get(self.FORMATS_URL.format(channel))
streams = http.json(res, schema=self._formats_schema)['streams']
if streams == []:
self.logger.error(
'Channel may be geo-restricted, not directly provided by PlayTV or not freely available'
)
return
for language in streams:
for protocol, bitrates in list(streams[language].items()):
# - Ignore non-supported protocols (RTSP, DASH)
# - Ignore deprecated Flash (RTMPE/HDS) streams (PlayTV doesn't provide anymore a Flash player)
if protocol in ['rtsp', 'flash', 'dash', 'hds']:
continue
for bitrate in bitrates['bitrates']:
if bitrate['value'] == 0:
continue
api_url = self.API_URL.format(channel, protocol, language,
bitrate['value'])
res = http.get(api_url)
video_url = http.json(res, schema=self._api_schema)['url']
bs = '{0}k'.format(bitrate['value'])
if protocol == 'hls':
for _, stream in HLSStream.parse_variant_playlist(
self.session, video_url).items():
yield bs, stream
elif protocol == 'hds':
for _, stream in HDSStream.parse_manifest(
self.session, video_url).items():
yield bs, stream
def _get_streams(self):
if '/news/videos/' in self.url:
# VOD
streams = self._get_vod_streams()
else:
# Live
streams = self._get_live_streams()
for video_url in streams:
if '.f4m' in video_url:
for stream in HDSStream.parse_manifest(self.session,
video_url).items():
yield stream
elif '.m3u8' in video_url:
for stream in HLSStream.parse_variant_playlist(
self.session, video_url).items():
yield stream
if '.mp4' in video_url:
match = self._mp4_bitrate_re.match(video_url)
if match is not None:
bitrate = match.group('bitrate') + 'k'
else:
bitrate = 'vod'
yield bitrate, HTTPStream(self.session, video_url)
def _get_streams(self):
# Retrieve geolocation data
res = http.get(self.GEO_URL)
geo = http.json(res, schema=self._geo_schema)
country_code = geo['reponse']['geo_info']['country_code']
# Retrieve URL page and search for video ID
res = http.get(self.url)
if 'france.tv' in self.url:
match = self._pluzz_video_id_re.search(res.text)
elif 'ludo.fr' in self.url or 'zouzous.fr' in self.url:
match = self._jeunesse_video_id_re.search(res.text)
elif 'france3-regions.francetvinfo.fr' in self.url:
match = self._f3_regions_video_id_re.search(res.text)
elif 'sport.francetvinfo.fr' in self.url:
match = self._sport_video_id_re.search(res.text)
if match is None:
return
video_id = match.group('video_id')
# Retrieve SWF player URL
swf_url = None
res = http.get(self.PLAYER_GENERATOR_URL)
player_url = update_scheme(
self.url,
http.json(res, schema=self._player_schema)['result'])
res = http.get(player_url)
match = self._swf_re.search(res.text)
if match is not None:
swf_url = update_scheme(self.url, match.group(0))
res = http.get(self.API_URL.format(video_id))
videos = http.json(res, schema=self._api_schema)
now = time.time()
offline = False
geolocked = False
drm = False
expired = False
streams = []
for video in videos['videos']:
video_url = video['url']
# Check whether video format is available
if video['statut'] != 'ONLINE':
offline = offline or True
continue
# Check whether video format is geo-locked
if video['geoblocage'] is not None and country_code not in video[
'geoblocage']:
geolocked = geolocked or True
continue
# Check whether video is DRM-protected
if video['drm']:
drm = drm or True
continue
# Check whether video format is expired
available = False
for interval in video['plages_ouverture']:
available = (interval['debut'] or 0) <= now <= (interval['fin']
or sys.maxsize)
if available:
break
if not available:
expired = expired or True
continue
# TODO: add DASH streams once supported
if '.mpd' in video_url:
continue
if '.f4m' in video_url or 'france.tv' in self.url:
res = http.get(self.TOKEN_URL.format(video_url))
video_url = res.text
if '.f4m' in video_url and swf_url is not None:
for bitrate, stream in HDSStream.parse_manifest(
self.session, video_url, is_akamai=True,
pvswf=swf_url).items():
# HDS videos with data in their manifest fragment token
# doesn't seem to be supported by HDSStream. Ignore such
# stream (but HDS stream having only the hdntl parameter in
# their manifest token will be provided)
pvtoken = stream.request_params['params'].get(
'pvtoken', '')
match = self._hds_pv_data_re.search(pvtoken)
if match is None:
streams.append((bitrate, stream))
elif '.m3u8' in video_url:
for stream in HLSStream.parse_variant_playlist(
self.session, video_url).items():
streams.append(stream)
# HBB TV streams are not provided anymore by France Televisions
elif '.mp4' in video_url and '/hbbtv/' not in video_url:
match = self._mp4_bitrate_re.match(video_url)
if match is not None:
bitrate = match.group('bitrate')
else:
# Fallback bitrate (seems all France Televisions MP4 videos
# seem have such bitrate)
bitrate = '1500k'
streams.append((bitrate, HTTPStream(self.session, video_url)))
if self.get_option("mux_subtitles") and videos['subtitles'] != []:
substreams = {}
for subtitle in videos['subtitles']:
# TTML subtitles are available but not supported by FFmpeg
if subtitle['format'] == 'ttml':
continue
substreams[subtitle['type']] = HTTPStream(
self.session, subtitle['url'])
for quality, stream in streams:
yield quality, MuxedStream(self.session,
stream,
subtitles=substreams)
else:
for stream in streams:
yield stream
if offline:
self.logger.error(
'Failed to access stream, may be due to offline content')
if geolocked:
self.logger.error(
'Failed to access stream, may be due to geo-restricted content'
)
if drm:
self.logger.error(
'Failed to access stream, may be due to DRM-protected content')
if expired:
self.logger.error(
'Failed to access stream, may be due to expired content')
def _get_streams(self):
res = http.get(self.url)
match = self._player_js.search(res.text)
if match:
player_js = match.group(0)
self.logger.info("Found player js {0}", player_js)
else:
self.logger.info(
"Didn't find player js. Probably this page doesn't contain a video"
)
return
res = http.get(player_js)
m = self._data_re.search(res.text)
if not m:
self.logger.info(
"Couldn't extract json metadata from player.js: {0}",
player_js)
return
stream_metadata = json.loads(m.group("data"))
is_video = stream_metadata["mediaType"] in ["live", "vod"]
is_audio = stream_metadata["mediaType"] == "aod"
media_version = tuple([
int(d)
for d in stream_metadata["mediaVersion"].split("-")[0].split(".")
])
if is_video or is_audio:
media_url = stream_metadata["mediaResource"]["dflt"][
"videoURL" if is_video else "audioURL"]
media_url_alt = stream_metadata["mediaResource"]["alt"][
"videoURL" if is_video else "audioURL"]
media_name = "audio" if is_audio else "vod"
if media_version >= (1, 2, 0):
media_format = stream_metadata["mediaResource"]["dflt"][
"mediaFormat"]
media_format_alt = stream_metadata["mediaResource"]["alt"][
"mediaFormat"]
else:
media_format = stream_metadata["mediaFormat"]
media_format_alt = media_url_alt[-4:]
stream_url = {
"url": media_url,
"format": media_format,
"name": media_name
}
stream_url_alt = {
"url": media_url_alt,
"format": media_format_alt,
"name": media_name
}
for stream in [stream_url, stream_url_alt]:
url = update_scheme("http://", stream["url"])
try:
if stream["format"] in ["hds", ".f4m"]:
for s in HDSStream.parse_manifest(
self.session, url, is_akamai=True).items():
yield s
elif stream["format"] in ["hls", "m3u8"]:
streams = HLSStream.parse_variant_playlist(
self.session, url).items()
if not streams:
yield "live", HLSStream(self.session, url)
for s in streams:
yield s
elif stream["format"] in ["mp3", "mp4", ".mp3", ".mp4"]:
yield stream["name"], HTTPStream(self.session, url)
except IOError as err:
self.logger.error("Failed to extract {0} streams: {1}",
stream["format"], err)
def _get_hds_streams(self, info):
# Needs the hdcore parameter added
url = info["_stream"] + HDCORE_PARAMETER
return HDSStream.parse_manifest(self.session, url,
pvswf=SWF_URL).items()
def _get_streams(self):
"""
Find all the streams for the ITV url
:return: Mapping of quality to stream
"""
soap_message = self._soap_request()
headers = {
'Content-Length': '{0:d}'.format(len(soap_message)),
'Content-Type': 'text/xml; charset=utf-8',
'Host': 'secure-mercury.itv.com',
'Origin': 'http://www.itv.com',
'Referer':
'http://www.itv.com/Mercury/Mercury_VideoPlayer.swf?v=null',
'SOAPAction': "http://tempuri.org/PlaylistService/GetPlaylist",
'User-Agent': ITV_PLAYER_USER_AGENT,
"X-Requested-With": "ShockwaveFlash/16.0.0.305"
}
res = http.post(
"https://secure-mercury.itv.com/PlaylistService.svc?wsdl",
headers=headers,
data=soap_message)
# Parse XML
xmldoc = http.xml(res)
# Check that geo region has been accepted
faultcode = xmldoc.find('.//faultcode')
if faultcode is not None:
if 'InvalidGeoRegion' in faultcode.text:
self.logger.error('Failed to retrieve playlist data '
'(invalid geo region)')
return None
# Look for <MediaFiles> tag (RTMP streams)
mediafiles = xmldoc.find('.//VideoEntries//MediaFiles')
# Look for <ManifestFile> tag (HDS streams)
manifestfile = xmldoc.find('.//VideoEntries//ManifestFile')
# No streams
if not mediafiles and not manifestfile:
return None
streams = {}
# Proxy not needed for media retrieval (Note: probably better to use flag)
# for x in ('http', 'https'):
# if x in http.proxies:
# http.proxies.pop(x);
# Parse RTMP streams
if mediafiles:
rtmp = mediafiles.attrib['base']
for mediafile in mediafiles.findall("MediaFile"):
playpath = mediafile.find("URL").text
rtmp_url = urlparse(rtmp)
app = (rtmp_url.path[1:] + '?' + rtmp_url.query).rstrip('?')
live = app == "live"
params = dict(
rtmp="{u.scheme}://{u.netloc}{u.path}".format(u=rtmp_url),
app=app.rstrip('?'),
playpath=playpath,
swfVfy=LIVE_SWF_URL if live else ONDEMAND_SWF_URL,
timeout=10)
if live:
params['live'] = True
bitrate = int(mediafile.attrib['bitrate']) / 1000
quality = "{0:d}k".format(int(bitrate))
streams[quality] = RTMPStream(self.session, params)
# Parse HDS streams
if manifestfile:
url = manifestfile.find('URL').text
if urlparse(url).path.endswith('f4m'):
streams.update(
HDSStream.parse_manifest(self.session,
url,
pvswf=LIVE_SWF_URL))
return streams