-#!/usr/bin/env python3
+#!/usr/bin/env python3.4
# -*- coding: utf-8 -*-
#
# (C) Copyright 2010 Mikael Frykholm <mikael@frykholm.com>
import socket
import feedparser
from datetime import datetime, timezone
+from pathlib import Path
+
class Video(dict):
def __init__(self, *args, **kwargs):
self.update(dict(*args, **kwargs)) # use the free update to set keys
else:
flashvars = requests.get("http://www.svtplay.se/%s"%video_player.attrs['data-json-href']+"?output=json").json()
video['duration'] = video_player.attrs.get('data-length',0)
- if not video['title']:
+ if not 'title' in video:
video['title'] = soup.find('meta',{'property':'og:title'}).attrs['content'].replace('|','_').replace('/','_')
if not 'genre' in video:
if soup.find(text='Kategori:'):
video['genre'] = 'Ingen Genre'
if 'dynamicStreams' in flashvars:
video['url'] = flashvars['dynamicStreams'][0].split('url:')[1].split('.mp4,')[0] +'.mp4'
- filename = video['title']+".mp4"
+ filename = Path(video['title']).with_suffix(".mp4")
print(Popen(["rtmpdump","-o"+filename,"-r", url], stdout=PIPE).communicate()[0])
if 'pathflv' in flashvars:
rtmp = flashvars['pathflv'][0]
- filename = video['title']+".flv"
+ filename = Path(video['title']).with_suffix(".flv")
print(Popen(["mplayer","-dumpstream","-dumpfile",filename, rtmp], stdout=PIPE).communicate()[0])
if not 'timestamp' in video:
if soup.find_all(datetime=True):
xmldate_str = soup.find_all(datetime=True)[0].attrs['datetime']
- video['timestamp'] = datetime(*feedparser._parse_date_w3dtf(xmldate_str)[:6]) #naive in utc
- video['timestamp'] = video['timestamp'].replace(tzinfo=timezone.utc).astimezone(tz=None) #convert to local time
+ if xmldate_str:
+ video['timestamp'] = datetime(*feedparser._parse_date_w3dtf(xmldate_str)[:6]) #naive in utc
+ video['timestamp'] = video['timestamp'].replace(tzinfo=timezone.utc).astimezone(tz=None) #convert to local time
if 'video' in flashvars:
for reference in flashvars['video']['videoReferences']:
if 'm3u8' in reference['url']:
video['url']=reference['url']
- video['filename'] = video['title']+'.ts'
+ video['filename'] = Path(video['title']).with_suffix('.ts')
if 'statistics' in flashvars:
video['category'] = flashvars['statistics']['category']
- download_from_playlist(video)
+ if not download_from_playlist(video):
+ return False
if not 'url' in video:
print("Could not find any streams")
return False
return video
def download_from_playlist(video):
- playlist = parse_playlist(requests.get(video['url']).text)
+ params = requests.utils.urlparse(video['url']).query
+ print(params)
+ if 'cc1=' in params: #'cc1=name=Svenska~default=yes~forced=no~uri=http://media.svt.se/download/mcc/wp3/undertexter-wsrt/1134047/1134047-025A/C(sv)/index.m3u8~lang=sv'
+ video['subs'] = [dict([k.split('=') for k in params.split('cc1=')[1].split('~')])] #make a dict from the paramstring
+ try:
+ req = requests.get(video['url']).text
+ except:
+ print("Error reading, skipping file")
+ print(sys.exc_info()[1])
+ return False
+ if 'subs' in video:
+ try:
+ segments = [item for item in requests.get(video['subs'][0]['uri']).text.split('\n') if 'vtt' in item]
+ except:
+ print("Error reading, skipping subtitle")
+ print(sys.exc_info()[1])
+ segments = [] #ugly FIXME
+ video['subs'][0]['download'] = []
+ for segment in segments:
+ if not segment.startswith('http'):
+ segment = "{}/{}".format(os.path.dirname(video['subs'][0]['uri']), segment)
+ try:
+ video['subs'][0]['download'].append(requests.get(segment).text)
+ except:
+ print("Error reading, skipping subtitle")
+ print(sys.exc_info()[1])
+ break
+ playlist = parse_playlist(req)
if not playlist:
return
videourl = sorted(playlist, key=lambda k: int(k['BANDWIDTH']))[-1]['url']
videourl = "{}/{}".format(os.path.dirname(video['url']), videourl)
segments, metadata = parse_segment_playlist(videourl)
if "EXT-X-KEY" in metadata:
- key = requests.get(metadata["EXT-X-KEY"]['URI'].strip('"')).text
+ try:
+ key = requests.get(metadata["EXT-X-KEY"]['URI'].strip('"')).text
+ except:
+ print("Error reading, skipping file")
+ print(sys.exc_info()[1])
+ return False
decrypt=True
else:
decrypt=False
- with open("%s"%video['filename'],"wb") as ofile:
+ with video['filename'].open("wb") as ofile:
segment=0
size = 0
for url in segments:
- ufile = requests.get(url, stream=True).raw
- print("\r{0:.2f} MB".format(size/1024/1024),end="")
+ try:
+ ufile = requests.get(url, stream=True).raw
+ except:
+ print("Error reading, skipping file")
+ print(sys.exc_info()[1])
+ return False
+ print("\r{0:.2f} MB".format(size/1024/1024), end="")
sys.stdout.flush()
if decrypt:
iv=struct.pack("IIII",segment,0,0,0)
- decryptor = AES.new(key, AES.MODE_CBC, iv)
+ try:
+ decryptor = AES.new(key, AES.MODE_CBC, iv) #ValueError: AES key must be either 16, 24, or 32 bytes long
+ except(ValueError) as e:
+ print("Error using decryption key. Skipping")
+ print(e)
+ return False
while(True):
try:
buf = ufile.read(4096)
- except (socket.error, TypeError) as e:
- print("Error reading, skipping file")
- print(e)
- return
+ except:
+ print("Error reading, skipping file") #FIXME mark file as failed
+ print(sys.exc_info()[1])
+ return False
if not buf:
break
if decrypt:
segment += 1
if 'thumb-url' in video:
- video['thumb'] = requests.get(video['thumb-url'],stream=True).raw
+ try:
+ video['thumb'] = requests.get(video['thumb-url'],stream=True).raw
+ except:
+ print("Error reading thumbnail") #FIXME mark file as failed
+ print(sys.exc_info()[1])
+
+ return True
def parse_playlist(playlist):
if not playlist.startswith("#EXTM3U"):
page_num += 1
def remux(video, xml=None):
- basename = video['filename'].split('.ts')[0]
if 'genre' in video:
if not os.path.exists(video['genre']):
os.mkdir(video['genre'])
- video['path'] = os.path.join(video['genre'],basename+'.mkv')
+ video['path'] = Path(video['genre'] / video['filename']).with_suffix('.mkv')
else:
- video['path'] = basename+'.mkv'
- command = ["mkvmerge","-o",video['path'], '--title',video['title']]
+ video['path'] = video['filename'].with_suffix('.mkv')
+ command = ["mkvmerge","-o",str(video['path']), '--title',video['title']]
if xml:
- with open(basename+'.xml','w') as f:
+ with video['filename'].with_suffix('.xml').open('w') as f:
f.write(xml)
- command.extend(['--global-tags',basename+'.xml'])
+ command.extend(['--global-tags',str(video['filename'].with_suffix('.xml'))])
if 'thumb' in video:
with open('thumbnail.jpg','wb') as f: #FIXME use title instead for many downloaders
f.write(video['thumb'].read())
command.extend(['--attachment-description', "Thumbnail",
'--attachment-mime-type', 'image/jpeg',
'--attach-file', 'thumbnail.jpg'])
- command.append(video['filename'])
+ # if 'subs' in video:
+ # for sub in video['subs']:
+ # if 'download' in sub:
+ # with open("{}.vtt".format(sub['lang']),'wb') as f:
+ # f.write(bytes("".join(sub['download']),'utf-8')) #FIXME
+ # command.extend(['--language 0:{} {}.vtt'.format(sub['lang'],sub['lang'])])
+
+ command.append(str(video['filename']))
print(Popen(command, stdout=PIPE).communicate()[0])
- for fname in (video['filename'], basename+'.xml','thumbnail.jpg'):
+ for fname in (video['filename'], video['filename'].with_suffix('.xml'),Path('thumbnail.jpg')):
try:
- os.unlink(fname)
+ fname.unlink()
except:
pass
if 'timestamp' in video:
- os.utime(video['path'], times=(video['timestamp'].timestamp(),video['timestamp'].timestamp()))
+ try:
+ os.utime(str(video['path']), times=(video['timestamp'].timestamp(),video['timestamp'].timestamp()))
+ except FileNotFoundError as e:
+ print(e)
def mkv_metadata(video):
if args.no_act:
continue
open(os.path.join('.seen',video['title']),'w').close() #touch
- video = scrape_player_page(video)
+ ret = scrape_player_page(video)
+ if not ret:
+ if not os.path.exists('.failed'):
+ os.mkdir('.failed')
+ open(os.path.join('.failed',video['title']),'w').close() #touch
+ continue
+ video = ret
if args.no_remux:
continue
xml = mkv_metadata(video)
if not args.no_act:
video = scrape_player_page({'url':args.url})
if not args.no_remux:
- remux({'title':e.title})
+ remux(video)
print(("Downloaded {}".format(args.url)))
projects / svtplaydump.git / blobdiff