Add support for using rtmpdump

This commit is contained in:
Ricardo Garcia 2010-01-03 13:12:11 +01:00
parent a692ca7c49
commit 0487b407a1

View file

@ -13,10 +13,12 @@ import os.path
import re import re
import socket import socket
import string import string
import subprocess
import sys import sys
import time import time
import urllib import urllib
import urllib2 import urllib2
import urlparse
std_headers = { std_headers = {
'User-Agent': 'Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.1.2) Gecko/20090729 Firefox/3.5.2', 'User-Agent': 'Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.1.2) Gecko/20090729 Firefox/3.5.2',
@ -390,7 +392,36 @@ class FileDownloader(object):
if info is None: if info is None:
break break
def _download_with_rtmpdump(self, filename, url):
self.report_destination(filename)
# Check for rtmpdump first
try:
subprocess.call(['rtmpdump', '-h'], stdout=(file(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
except (OSError, IOError):
self.trouble(u'ERROR: RTMP download detected but "rtmpdump" could not be run')
return False
# Download using rtmpdump. rtmpdump returns exit code 2 when
# the connection was interrumpted and resuming appears to be
# possible. This is part of rtmpdump's normal usage, AFAIK.
retval = subprocess.call(['rtmpdump', '-q', '-r', url, '-o', filename] + [[], ['-e']][self.params.get('continuedl', False)])
while retval == 2:
self.to_stdout(u'\r[rtmpdump] %s bytes' % os.path.getsize(filename), skip_eol=True)
time.sleep(2.0) # This seems to be needed
retval = subprocess.call(['rtmpdump', '-q', '-e', '-r', url, '-o', filename])
if retval == 0:
self.to_stdout(u'\r[rtmpdump] %s bytes' % os.path.getsize(filename))
return True
else:
self.trouble('ERROR: rtmpdump exited with code %d' % retval)
return False
def _do_download(self, filename, url): def _do_download(self, filename, url):
# Attempt to download using rtmpdump
if url.startswith('rtmp'):
return self._download_with_rtmpdump(filename, url)
stream = None stream = None
open_mode = 'wb' open_mode = 'wb'
basic_request = urllib2.Request(url, None, std_headers) basic_request = urllib2.Request(url, None, std_headers)
@ -596,6 +627,10 @@ class YoutubeIE(InfoExtractor):
"""Report extracted video URL.""" """Report extracted video URL."""
self._downloader.to_stdout(u'[youtube] %s: Format %s not available' % (video_id, format)) self._downloader.to_stdout(u'[youtube] %s: Format %s not available' % (video_id, format))
def report_rtmp_download(self):
"""Indicate the download will use the RTMP protocol."""
self._downloader.to_stdout(u'[youtube] RTMP download detected')
def _real_initialize(self): def _real_initialize(self):
if self._downloader is None: if self._downloader is None:
return return
@ -694,43 +729,45 @@ class YoutubeIE(InfoExtractor):
try: try:
self.report_video_info_webpage_download(video_id) self.report_video_info_webpage_download(video_id)
video_info_webpage = urllib2.urlopen(request).read() video_info_webpage = urllib2.urlopen(request).read()
video_info = urlparse.parse_qs(video_info_webpage)
except (urllib2.URLError, httplib.HTTPException, socket.error), err: except (urllib2.URLError, httplib.HTTPException, socket.error), err:
self._downloader.trouble(u'ERROR: unable to download video info webpage: %s' % str(err)) self._downloader.trouble(u'ERROR: unable to download video info webpage: %s' % str(err))
return return
self.report_information_extraction(video_id) self.report_information_extraction(video_id)
# "t" param # "t" param
mobj = re.search(r'(?m)&token=([^&]+)(?:&|$)', video_info_webpage) if 'token' not in video_info:
if mobj is None:
# Attempt to see if YouTube has issued an error message # Attempt to see if YouTube has issued an error message
mobj = re.search(r'(?m)&reason=([^&]+)(?:&|$)', video_info_webpage) if 'reason' not in video_info:
if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract "t" parameter for unknown reason') self._downloader.trouble(u'ERROR: unable to extract "t" parameter for unknown reason')
stream = open('reportme-ydl-%s.dat' % time.time(), 'wb') stream = open('reportme-ydl-%s.dat' % time.time(), 'wb')
stream.write(video_info_webpage) stream.write(video_info_webpage)
stream.close() stream.close()
else: else:
reason = urllib.unquote_plus(mobj.group(1)) reason = urllib.unquote_plus(video_info['reason'][0])
self._downloader.trouble(u'ERROR: YouTube said: %s' % reason.decode('utf-8')) self._downloader.trouble(u'ERROR: YouTube said: %s' % reason.decode('utf-8'))
return return
token = urllib.unquote(mobj.group(1)) token = urllib.unquote_plus(video_info['token'][0])
video_real_url = 'http://www.youtube.com/get_video?video_id=%s&t=%s&eurl=&el=detailpage&ps=default&gl=US&hl=en' % (video_id, token) video_real_url = 'http://www.youtube.com/get_video?video_id=%s&t=%s&eurl=&el=detailpage&ps=default&gl=US&hl=en' % (video_id, token)
if format_param is not None: if format_param is not None:
video_real_url = '%s&fmt=%s' % (video_real_url, format_param) video_real_url = '%s&fmt=%s' % (video_real_url, format_param)
# Check possible RTMP download
if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
self.report_rtmp_download()
video_real_url = video_info['conn'][0]
# uploader # uploader
mobj = re.search(r'(?m)&author=([^&]+)(?:&|$)', video_info_webpage) if 'author' not in video_info:
if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract uploader nickname') self._downloader.trouble(u'ERROR: unable to extract uploader nickname')
return return
video_uploader = urllib.unquote(mobj.group(1)) video_uploader = urllib.unquote_plus(video_info['author'][0])
# title # title
mobj = re.search(r'(?m)&title=([^&]*)(?:&|$)', video_info_webpage) if 'title' not in video_info:
if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract video title') self._downloader.trouble(u'ERROR: unable to extract video title')
return return
video_title = urllib.unquote_plus(mobj.group(1)) video_title = urllib.unquote_plus(video_info['title'][0])
video_title = video_title.decode('utf-8') video_title = video_title.decode('utf-8')
video_title = re.sub(ur'(?u)&(.+?);', self.htmlentity_transform, video_title) video_title = re.sub(ur'(?u)&(.+?);', self.htmlentity_transform, video_title)
video_title = video_title.replace(os.sep, u'%') video_title = video_title.replace(os.sep, u'%')