'Gergely Imreh',
'Rogério Brito',
'Philipp Hagemeister',
+ 'Sören Schulze',
)
__license__ = 'Public Domain'
-__version__ = '2011.08.28-phihag'
+__version__ = '2011.09.09-phihag'
UPDATE_URL = 'https://raw.github.com/phihag/youtube-dl/master/youtube-dl'
except ImportError:
pass # Handled below
+try:
+ import xml.etree.ElementTree
+except ImportError: # Python<2.5
+ pass # Not officially supported, but let it slip
+
std_headers = {
'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:5.0.1) Gecko/20100101 Firefox/5.0.1',
'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
yield pref
return yield_preferredencoding().next()
+
def htmlentity_transform(matchobj):
"""Transforms an HTML entity to a Unicode character.
# Unknown entity in name, return its literal representation
return (u'&%s;' % entity)
+
def sanitize_title(utitle):
"""Sanitizes a video title so it could be used as part of a filename."""
utitle = re.sub(ur'(?u)&(.+?);', htmlentity_transform, utitle)
return utitle.replace(unicode(os.sep), u'%')
+
def sanitize_open(filename, open_mode):
"""Try to open the given filename, and slightly tweak it if this fails.
stream = open(filename, open_mode)
return (stream, filename)
+
def timeconvert(timestr):
- """Convert RFC 2822 defined time string into system timestamp"""
- timestamp = None
- timetuple = email.utils.parsedate_tz(timestr)
- if timetuple is not None:
- timestamp = email.utils.mktime_tz(timetuple)
- return timestamp
+ """Convert RFC 2822 defined time string into system timestamp"""
+ timestamp = None
+ timetuple = email.utils.parsedate_tz(timestr)
+ if timetuple is not None:
+ timestamp = email.utils.mktime_tz(timetuple)
+ return timestamp
+
class DownloadError(Exception):
"""Download Error exception.
"""
pass
+
class SameFileError(Exception):
"""Same File exception.
"""
pass
+
class PostProcessingError(Exception):
"""Post Processing exception.
"""
pass
+
class UnavailableVideoError(Exception):
"""Unavailable Format exception.
"""
pass
+
class ContentTooShortError(Exception):
"""Content Too Short exception.
self.downloaded = downloaded
self.expected = expected
+
class YoutubeDLHandler(urllib2.HTTPHandler):
"""Handler for HTTP requests and responses.
a particular request, the original request in the program code only has
to include the HTTP header "Youtubedl-No-Compression", which will be
removed before making the real request.
-
+
Part of this code was copied from:
- http://techknack.net/python-urllib2-handlers/
-
+ http://techknack.net/python-urllib2-handlers/
+
Andrew Rowls, the author of that code, agreed to release it to the
public domain.
"""
return zlib.decompress(data, -zlib.MAX_WBITS)
except zlib.error:
return zlib.decompress(data)
-
+
@staticmethod
def addinfourl_wrapper(stream, headers, url, code):
if hasattr(urllib2.addinfourl, 'getcode'):
ret = urllib2.addinfourl(stream, headers, url)
ret.code = code
return ret
-
+
def http_request(self, req):
for h in std_headers:
if h in req.headers:
resp.msg = old_resp.msg
return resp
+
class FileDownloader(object):
"""File Downloader class.
self.params = params
@staticmethod
- def pmkdir(filename):
- """Create directory components in filename. Similar to Unix "mkdir -p"."""
- components = filename.split(os.sep)
- aggregate = [os.sep.join(components[0:x]) for x in xrange(1, len(components))]
- aggregate = ['%s%s' % (x, os.sep) for x in aggregate] # Finish names with separator
- for dir in aggregate:
- if not os.path.exists(dir):
- os.mkdir(dir)
-
- @staticmethod
def format_bytes(bytes):
if bytes is None:
return 'N/A'
else:
exponent = long(math.log(bytes, 1024.0))
suffix = 'bkMGTPEZY'[exponent]
- converted = float(bytes) / float(1024**exponent)
+ converted = float(bytes) / float(1024 ** exponent)
return '%.2f%s' % (converted, suffix)
@staticmethod
os.rename(old_filename, new_filename)
except (IOError, OSError), err:
self.trouble(u'ERROR: unable to rename file')
-
+
def try_utime(self, filename, last_modified_hdr):
"""Try to set the last-modified time of the given file."""
if last_modified_hdr is None:
if filetime is None:
return
try:
- os.utime(filename,(time.time(), filetime))
+ os.utime(filename, (time.time(), filetime))
except:
pass
return
try:
- self.pmkdir(filename)
+ dn = os.path.dirname(filename)
+ if dn != '' and not os.path.exists(dn):
+ os.makedirs(dn)
except (OSError, IOError), err:
- self.trouble(u'ERROR: unable to create directories: %s' % str(err))
+ self.trouble(u'ERROR: unable to create directory ' + unicode(err))
return
if self.params.get('writedescription', False):
finally:
descfile.close()
except (OSError, IOError):
- self.trouble(u'ERROR: Cannot write description file: %s' % str(descfn))
+ self.trouble(u'ERROR: Cannot write description file ' + descfn)
return
if self.params.get('writeinfojson', False):
finally:
infof.close()
except (OSError, IOError):
- self.trouble(u'ERROR: Cannot write metadata to JSON file: %s' % str(infofn))
+ self.trouble(u'ERROR: Cannot write metadata to JSON file ' + infofn)
return
try:
# 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.
- basic_args = ['rtmpdump', '-q'] + [[], ['-W', player_url]][player_url is not None] + ['-r', url, '-o', tmpfilename]
+ basic_args = ['rtmpdump'] + [[], ['-W', player_url]][player_url is not None] + ['-r', url, '-o', tmpfilename]
retval = subprocess.call(basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)])
while retval == 2 or retval == 1:
prevsize = os.path.getsize(tmpfilename)
# Request parameters in case of being able to resume
if self.params.get('continuedl', False) and resume_len != 0:
self.report_resuming_byte(resume_len)
- request.add_header('Range','bytes=%d-' % resume_len)
+ request.add_header('Range', 'bytes=%d-' % resume_len)
open_mode = 'ab'
count = 0
else:
# Examine the reported length
if (content_length is not None and
- (resume_len - 100 < long(content_length) < resume_len + 100)):
+ (resume_len - 100 < long(content_length) < resume_len + 100)):
# The file had already been fully downloaded.
# Explanation to the above condition: in issue #175 it was revealed that
# YouTube sometimes adds or removes a few bytes from the end of the file,
if stream is None:
try:
(stream, tmpfilename) = sanitize_open(tmpfilename, open_mode)
+ assert stream is not None
filename = self.undo_temp_name(tmpfilename)
self.report_destination(filename)
except (OSError, IOError), err:
# Apply rate limit
self.slow_down(start, byte_counter - resume_len)
+ if stream is None:
+ self.trouble(u'\nERROR: Did not get any data blocks')
+ return False
stream.close()
self.report_finish()
if data_len is not None and byte_counter != data_len:
return True
+
class InfoExtractor(object):
"""Information Extractor class.
"""Real extraction process. Redefine in subclasses."""
pass
+
class YoutubeIE(InfoExtractor):
"""Information extractor for youtube.com."""
self.report_video_info_webpage_download(video_id)
for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
video_info_url = ('http://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
- % (video_id, el_type))
+ % (video_id, el_type))
request = urllib2.Request(video_info_url)
try:
video_info_webpage = urllib2.urlopen(request).read()
except UnavailableVideoError:
self._downloader.trouble(u'\nERROR: unable to download video')
+
class GoogleIE(InfoExtractor):
"""Information extractor for video.google.com."""
else: # we need something to pass to process_info
video_thumbnail = ''
-
try:
# Process video information
self._downloader.process_info({
self._downloader.trouble(u'ERROR: unable to extract video description')
return
video_description = mobj.group(1).decode('utf-8')
- if not video_description: video_description = 'No description available.'
+ if not video_description:
+ video_description = 'No description available.'
# Extract video height and width
mobj = re.search(r'<meta name="video_height" content="([0-9]+)" />', webpage)
yv_lg = 'R0xx6idZnW2zlrKP8xxAIR' # not sure what this represents
yv_bitrate = '700' # according to Wikipedia this is hard-coded
request = urllib2.Request('http://cosmos.bcst.yahoo.com/up/yep/process/getPlaylistFOP.php?node_id=' + video_id +
- '&tech=flash&mode=playlist&lg=' + yv_lg + '&bitrate=' + yv_bitrate + '&vidH=' + yv_video_height +
- '&vidW=' + yv_video_width + '&swf=as3&rd=video.yahoo.com&tk=null&adsupported=v1,v2,&eventid=1301797')
+ '&tech=flash&mode=playlist&lg=' + yv_lg + '&bitrate=' + yv_bitrate + '&vidH=' + yv_video_height +
+ '&vidW=' + yv_video_width + '&swf=as3&rd=video.yahoo.com&tk=null&adsupported=v1,v2,&eventid=1301797')
try:
self.report_download_webpage(video_id)
webpage = urllib2.urlopen(request).read()
return
video_url = urllib.unquote(mobj.group(1))
- video_id = os.path.basename(video_url)
+ video_id = os.path.basename(video_url)
# here's a fun little line of code for you:
video_extension = os.path.splitext(video_id)[1][1:]
- video_id = os.path.splitext(video_id)[0]
+ video_id = os.path.splitext(video_id)[0]
# it's tempting to parse this further, but you would
# have to take into account all the variations like
prefix, query = query.split(':')
prefix = prefix[8:]
- query = query.encode('utf-8')
+ query = query.encode('utf-8')
if prefix == '':
self._download_n_results(query, 1)
return
self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query))
return
elif n > self._max_youtube_results:
- self._downloader.to_stderr(u'WARNING: ytsearch returns max %i results (you requested %i)' % (self._max_youtube_results, n))
+ self._downloader.to_stderr(u'WARNING: ytsearch returns max %i results (you requested %i)' % (self._max_youtube_results, n))
n = self._max_youtube_results
self._download_n_results(query, n)
return
pagenum = pagenum + 1
+
class GoogleSearchIE(InfoExtractor):
"""Information Extractor for Google Video search queries."""
_VALID_QUERY = r'gvsearch(\d+|all)?:[\s\S]+'
prefix, query = query.split(':')
prefix = prefix[8:]
- query = query.encode('utf-8')
+ query = query.encode('utf-8')
if prefix == '':
self._download_n_results(query, 1)
return
self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query))
return
elif n > self._max_google_results:
- self._downloader.to_stderr(u'WARNING: gvsearch returns max %i results (you requested %i)' % (self._max_google_results, n))
+ self._downloader.to_stderr(u'WARNING: gvsearch returns max %i results (you requested %i)' % (self._max_google_results, n))
n = self._max_google_results
self._download_n_results(query, n)
return
pagenum = pagenum + 1
+
class YahooSearchIE(InfoExtractor):
"""Information Extractor for Yahoo! Video search queries."""
_VALID_QUERY = r'yvsearch(\d+|all)?:[\s\S]+'
prefix, query = query.split(':')
prefix = prefix[8:]
- query = query.encode('utf-8')
+ query = query.encode('utf-8')
if prefix == '':
self._download_n_results(query, 1)
return
self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query))
return
elif n > self._max_yahoo_results:
- self._downloader.to_stderr(u'WARNING: yvsearch returns max %i results (you requested %i)' % (self._max_yahoo_results, n))
+ self._downloader.to_stderr(u'WARNING: yvsearch returns max %i results (you requested %i)' % (self._max_yahoo_results, n))
n = self._max_yahoo_results
self._download_n_results(query, n)
return
pagenum = pagenum + 1
+
class YoutubePlaylistIE(InfoExtractor):
"""Information Extractor for YouTube playlists."""
self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id)
return
+
class YoutubeUserIE(InfoExtractor):
"""Information Extractor for YouTube users."""
def report_download_page(self, username, start_index):
"""Report attempt to download user page."""
self._downloader.to_screen(u'[youtube] user %s: Downloading video ids from %d to %d' %
- (username, start_index, start_index + self._GDATA_PAGE_SIZE))
+ (username, start_index, start_index + self._GDATA_PAGE_SIZE))
def _real_initialize(self):
self._youtube_ie.initialize()
video_ids = video_ids[playliststart:playlistend]
self._downloader.to_screen("[youtube] user %s: Collected %d video ids (downloading %d of them)" %
- (username, all_ids_count, len(video_ids)))
+ (username, all_ids_count, len(video_ids)))
for video_id in video_ids:
self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % video_id)
except UnavailableVideoError, err:
self._downloader.trouble(u'ERROR: unable to download file')
+
class FacebookIE(InfoExtractor):
"""Information Extractor for Facebook"""
self._downloader.trouble(u'\nERROR: unable to download video')
+class MyVideoIE(InfoExtractor):
+ """Information Extractor for myvideo.de."""
+
+ _VALID_URL = r'(?:http://)?(?:www\.)?myvideo\.de/watch/([0-9]+)/([^?/]+).*'
+
+ def __init__(self, downloader=None):
+ InfoExtractor.__init__(self, downloader)
+
+ @staticmethod
+ def suitable(url):
+ return (re.match(MyVideoIE._VALID_URL, url) is not None)
+
+ def report_download_webpage(self, video_id):
+ """Report webpage download."""
+ self._downloader.to_screen(u'[myvideo] %s: Downloading webpage' % video_id)
+
+ def report_extraction(self, video_id):
+ """Report information extraction."""
+ self._downloader.to_screen(u'[myvideo] %s: Extracting information' % video_id)
+
+ def _real_initialize(self):
+ return
+
+ def _real_extract(self,url):
+ mobj = re.match(self._VALID_URL, url)
+ if mobj is None:
+ self._download.trouble(u'ERROR: invalid URL: %s' % url)
+ return
+
+ video_id = mobj.group(1)
+ simple_title = mobj.group(2).decode('utf-8')
+ # should actually not be necessary
+ simple_title = sanitize_title(simple_title)
+ simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', simple_title)
+
+ # Get video webpage
+ request = urllib2.Request('http://www.myvideo.de/watch/%s' % video_id)
+ try:
+ self.report_download_webpage(video_id)
+ webpage = urllib2.urlopen(request).read()
+ except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+ self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err))
+ return
+
+ self.report_extraction(video_id)
+ mobj = re.search(r'<link rel=\'image_src\' href=\'(http://is[0-9].myvideo\.de/de/movie[0-9]+/[a-f0-9]+)/thumbs/[^.]+\.jpg\' />',
+ webpage)
+ if mobj is None:
+ self._downloader.trouble(u'ERROR: unable to extract media URL')
+ return
+ video_url = mobj.group(1) + ('/%s.flv' % video_id)
+
+ mobj = re.search('<title>([^<]+)</title>', webpage)
+ if mobj is None:
+ self._downloader.trouble(u'ERROR: unable to extract title')
+ return
+
+ video_title = mobj.group(1)
+ video_title = sanitize_title(video_title)
+
+ try:
+ print(video_url)
+ self._downloader.process_info({
+ 'id': video_id,
+ 'url': video_url,
+ 'uploader': u'NA',
+ 'upload_date': u'NA',
+ 'title': video_title,
+ 'stitle': simple_title,
+ 'ext': u'flv',
+ 'format': u'NA',
+ 'player_url': None,
+ })
+ except UnavailableVideoError:
+ self._downloader.trouble(u'\nERROR: Unable to download video')
+
+class ComedyCentralIE(InfoExtractor):
+ """Information extractor for blip.tv"""
+
+ _VALID_URL = r'^(?:https?://)?(www\.)?(thedailyshow|colbertnation)\.com/full-episodes/(.*)$'
+
+ @staticmethod
+ def suitable(url):
+ return (re.match(ComedyCentralIE._VALID_URL, url) is not None)
+
+ def report_extraction(self, episode_id):
+ self._downloader.to_screen(u'[comedycentral] %s: Extracting information' % episode_id)
+
+ def report_config_download(self, episode_id):
+ self._downloader.to_screen(u'[comedycentral] %s: Downloading configuration' % episode_id)
+
+ def report_player_url(self, episode_id):
+ self._downloader.to_screen(u'[comedycentral] %s: Determining player URL' % episode_id)
+
+ def _simplify_title(self, title):
+ res = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', title)
+ res = res.strip(ur'_')
+ return res
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ if mobj is None:
+ self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
+ return
+ epTitle = mobj.group(3)
+
+ req = urllib2.Request(url)
+ self.report_extraction(epTitle)
+ try:
+ html = urllib2.urlopen(req).read()
+ except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+ self._downloader.trouble(u'ERROR: unable to download webpage: %s' % unicode(err))
+ return
+
+ mMovieParams = re.findall('<param name="movie" value="(http://media.mtvnservices.com/(.*?:episode:([^:]*):)(.*?))"/>', html)
+ if len(mMovieParams) == 0:
+ self._downloader.trouble(u'ERROR: unable to find Flash URL in webpage ' + url)
+ return
+ show_id = mMovieParams[0][2]
+ ACT_COUNT = { # TODO: Detect this dynamically
+ 'thedailyshow.com': 4,
+ 'colbertnation.com': 3,
+ }.get(show_id, 4)
+ OFFSET = {
+ 'thedailyshow.com': 1,
+ 'colbertnation.com': 1,
+ }.get(show_id, 1)
+
+ first_player_url = mMovieParams[0][0]
+ startMediaNum = int(mMovieParams[0][3]) + OFFSET
+ movieId = mMovieParams[0][1]
+
+ playerReq = urllib2.Request(first_player_url)
+ self.report_player_url(epTitle)
+ try:
+ playerResponse = urllib2.urlopen(playerReq)
+ except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+ self._downloader.trouble(u'ERROR: unable to download player: %s' % unicode(err))
+ return
+ player_url = playerResponse.geturl()
+
+ for actNum in range(ACT_COUNT):
+ mediaNum = startMediaNum + actNum
+ mediaId = movieId + str(mediaNum)
+ configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
+ urllib.urlencode({'uri': mediaId}))
+ configReq = urllib2.Request(configUrl)
+ self.report_config_download(epTitle)
+ try:
+ configXml = urllib2.urlopen(configReq).read()
+ except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+ self._downloader.trouble(u'ERROR: unable to download webpage: %s' % unicode(err))
+ return
+
+ cdoc = xml.etree.ElementTree.fromstring(configXml)
+ turls = []
+ for rendition in cdoc.findall('.//rendition'):
+ finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
+ turls.append(finfo)
+
+ if len(turls) == 0:
+ self._downloader.trouble(u'\nERROR: unable to download ' + str(mediaNum) + ': No videos found')
+ continue
+
+ # For now, just pick the highest bitrate
+ format,video_url = turls[-1]
+
+ self._downloader.increment_downloads()
+
+ effTitle = show_id.replace('.com', '') + '-' + epTitle
+ info = {
+ 'id': str(mediaNum),
+ 'url': video_url,
+ 'uploader': show_id,
+ 'upload_date': 'NA',
+ 'title': effTitle,
+ 'stitle': self._simplify_title(effTitle),
+ 'ext': 'mp4',
+ 'format': format,
+ 'thumbnail': None,
+ 'description': 'TODO: Not yet supported',
+ 'player_url': player_url
+ }
+
+ try:
+ self._downloader.process_info(info)
+ except UnavailableVideoError, err:
+ self._downloader.trouble(u'\nERROR: unable to download ' + str(mediaNum))
+ continue
+
+
class PostProcessor(object):
"""Post Processor class.
"""
return information # by default, do nothing
+
class FFmpegExtractAudioPP(PostProcessor):
def __init__(self, downloader=None, preferredcodec=None):
# General configuration
cookie_processor = urllib2.HTTPCookieProcessor(jar)
- urllib2.install_opener(urllib2.build_opener(urllib2.ProxyHandler(), cookie_processor, YoutubeDLHandler()))
+ opener = urllib2.build_opener(urllib2.ProxyHandler(), cookie_processor, YoutubeDLHandler())
+ urllib2.install_opener(opener)
socket.setdefaulttimeout(300) # 5 minutes should be enough (famous last words)
# Batch file verification
facebook_ie = FacebookIE()
bliptv_ie = BlipTVIE()
vimeo_ie = VimeoIE()
+ myvideo_ie = MyVideoIE()
+ comedycentral_ie = ComedyCentralIE()
+
generic_ie = GenericIE()
# File downloader
fd.add_info_extractor(facebook_ie)
fd.add_info_extractor(bliptv_ie)
fd.add_info_extractor(vimeo_ie)
+ fd.add_info_extractor(myvideo_ie)
+ fd.add_info_extractor(comedycentral_ie)
# This must come last since it's the
# fallback if none of the others work