Skip to content

Instantly share code, notes, and snippets.

@saxicek

saxicek/ct.py

Last active January 31, 2016 16:22
Show Gist options
  • Star 2 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save saxicek/753008560a978a052e3e to your computer and use it in GitHub Desktop.
Save saxicek/753008560a978a052e3e to your computer and use it in GitHub Desktop.
Download ZIP
iVysilani RTMP -> HLS
Raw
ct.py
#!/usr/bin/env python3
#
# částečně inspirováno ctsream od petr_p
__author__ = "Jakub Lužný"
__desc__ = "ČT (iVysílání)"
__url__ = r"https?://www\.ceskatelevize\.cz/(porady|ivysilani)/.+"
import re,os.path, urllib.request, urllib.parse, json, http.cookiejar, logging
import xml.etree.ElementTree as ElementTree
from urllib.parse import urlparse, unquote
log = logging.getLogger()
urlopen = urllib.request.urlopen
def flatten(obj, prefix = ''):
out = []
# print(prefix)
if type(obj) == dict:
for key in obj:
out+= flatten(obj[key], prefix+"[{}]".format(key) )
elif type(obj) == list:
for i in range(0, len(obj)):
out+= flatten(obj[i], prefix+'[{}]'.format(i) )
else:
out.append( (prefix, obj) )
return out
def srt_time(time):
time = int(time)
sec = time / 1000
msec = time % 1000
hour = sec / 3600
sec = sec % 3600
min = sec / 60
sec = sec % 60
return "{:02}:{:02}:{:02},{:03}".format(int(hour), int(min), int(sec), msec)
def txt_to_srt(txt):
subs = re.findall('\s*(\d+); (\d+) (\d+)\n(.+?)\n\n', txt, re.DOTALL)
srt = ''
for s in subs:
srt += "{}\n{} --> {}\n{}\n\n".format(s[0], srt_time(s[1]), srt_time(s[2]), s[3] )
return srt
class CtEngine:
def __init__(self, url):
url = url.replace('/porady/', '/ivysilani/').replace('/video/', '')
self.jar = http.cookiejar.CookieJar()
self.opener = urllib.request.build_opener(urllib.request.HTTPCookieProcessor(self.jar))
self.opener.addheaders = [
('User-agent', 'Mozilla/5.0'),
('x-addr', '127.0.0.1'),
('Referer', url)
]
urllib.request.install_opener(self.opener)
self.b_page = urlopen(url).read() # .decode('utf-8')
#get playlist URL first
data = re.findall(b"getPlaylistUrl\((.+?]), request", self.b_page)[0]
data = data.decode('utf-8')
data = json.loads(data)
data = data[0]
data = {
'playlist[0][type]' : data['type'],
'playlist[0][id]' : data['id'],
'requestUrl' : urlparse(url).path,
'requestSource' : 'iVysilani'
}
data = urllib.parse.urlencode( data, 'utf-8')
header = {
"Content-type": "application/x-www-form-urlencoded"
}
req = urllib.request.Request('http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist', bytes(data, 'utf-8'), header )
data = json.loads(urlopen(req).read().decode('utf-8'))
#-----------------
#print('====> Playlist URL: ' + data['url'])
urlobj = urlparse(data['url'])
urlhost = urlobj.netloc
urlpath = urllib.parse.urlunparse(('', '', urlobj.path, urlobj.params, urlobj.query, urlobj.fragment))
#print('===> Playlist Host= ' + urlhost + ' Path=' + urlpath)
### Get Playlists
conn = http.client.HTTPConnection(urlhost)
headers = {
'Connection': 'keep-alive',
#'Referer': 'http://imgct.ceskatelevize.cz/global/swf/player/player.swf?version=1.45.15a',
'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1700.107 Safari/537.36',
'Accept-Encoding': 'identity',
#'Accept-Encoding': 'gzip,deflate,sdch',
'Accept-Language': 'en-US,en;q=0.8,cs;q=0.6'
}
conn.request('GET', urlpath, '', headers)
res = conn.getresponse()
httpdata = res.read()
conn.close()
#print '====> PLAYLIST PAGE START'
#print httpdata
#print '====> PLAYLIST PAGE END'
### Read client config
jsondata = json.loads(httpdata.decode())
m3u_url = jsondata["playlist"][0]["streamUrls"]["main"]
#print('====> M3U URL: ' + m3u_url)
urlobj = urlparse(m3u_url)
urlhost = urlobj.netloc
urlpath = urllib.parse.urlunparse(('', '', urlobj.path, urlobj.params, urlobj.query, urlobj.fragment))
#print('===> M3U Host= ' + urlhost + ' Path=' + urlpath)
### Get M3u
conn = http.client.HTTPConnection(urlhost)
headers = {
'Connection': 'close',
'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1700.107 Safari/537.36',
}
conn.request('GET', urlpath, '', headers)
res = conn.getresponse()
m3u_url = res.getheader("Location")
conn.close()
#print('====> M3U URL: ' + m3u_url)
urlobj = urlparse(m3u_url)
urlhost = urlobj.netloc
urlpath = urllib.parse.urlunparse(('', '', urlobj.path, urlobj.params, urlobj.query, urlobj.fragment))
#print('===> M3U Host= ' + urlhost + ' Path=' + urlpath)
### Get M3u
conn = http.client.HTTPConnection(urlhost)
headers = {
'Connection': 'close',
'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1700.107 Safari/537.36',
}
conn.request('GET', urlpath, '', headers)
res = conn.getresponse()
httpdata = res.read()
conn.close()
# Add links
stream_urls = re.findall(r"^http://.*.m3u8$", httpdata.decode(), re.MULTILINE)
print('720p stream url=' + stream_urls[3])
print('ffmpeg -i ' + stream_urls[3] + ' -c copy ceske.stoleti.S01E0.mkv')
self.videos = [{'720p': stream_urls[3]}]
#-----------------
# url = urllib.parse.unquote(data['url'])
#
# self.playlist = urlopen(url).read().decode('utf-8')
# self.getMovie()
# self.videos = self.movie.findall('video')
#
# # setridime podle kvality: z popisku nechame jenom cislo
# # kvuli audio-description verzi (label AD) pridame 0, abychom to mohli tridit jako cisla
# self.videos = sorted(self.videos, key=lambda k: int(re.sub(r"\D", "", k.get('label')+"0")), reverse=True)
#
# if len(self.videos) == 0:
# raise ValueError('Není k dispozici žádná kvalita videa.')
def getMovie(self):
xml = ElementTree.fromstring(self.playlist)
for e in xml.findall('smilRoot/body/switchItem'):
i = e.get('id')
if not ('AD' in i or 'BO' in i):
self.movie = e
break
self.subtitles = None
for e in xml.findall('metaDataRoot/Playlist/PlaylistItem'):
if e.get('id') == self.movie.get('id'):
s = e.find('SubtitlesURL')
if s is not None:
self.subtitles = s.text
def qualities(self):
return [( v.get('label'), v.get('label') ) for v in self.videos] + ([('srt', 'Titulky')] if self.subtitles is not None else [] )
def movies(self):
return [ ('0', re.findall(b'<title>(.+?) &mdash;', self.b_page)[0].decode('utf-8')) ]
def get_video(self, quality):
for video in self.videos:
if video.get('label') == quality:
log.info('Vybraná kvalita: {}'.format(quality))
return video
raise ValueError('Není k dispozici zadaná kvalita videa.')
def download(self, quality, movie):
if quality == 'srt':
return self.download_srt()
if quality:
video = self.get_video(quality)
else:
video = self.videos[0]
log.info('Automaticky vybraná kvalita: {}'.format(video.get('label')) )
base = self.movie.get('base')
src = video.get('src')
filename = os.path.basename( src)[:-3] + 'flv'
parsedurl = urlparse(base)
app = parsedurl.path[1:] + '?' + parsedurl.query
# rtmpdump --live kvůli restartům - viz http://www.abclinuxu.cz/blog/pb/2011/5/televize-9-ctstream-3#18
return ('rtmp', filename, { 'url': base, 'playpath': src, 'app' : app, 'rtmpdump_args' : '--live'} )
def download_srt(self):
if self.subtitles is None:
raise ValueError('Titulky nejsou k dispozici.')
txt = urllib.request.urlopen(self.subtitles).read().decode('utf8')
srt = txt_to_srt(txt)
return ('text', 'subtitles.srt', srt.encode('cp1250') )
Raw
run.cmd
python tv-dl http://www.ceskatelevize.cz/porady/10362011008-ceske-stoleti/21251212024-kulka-pro-heydricha-1941/
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment