updatetime: Use the Last-modified header to set output file timestamps.
writedescription: Write the video description to a .description file
writeinfojson: Write the video description to a .info.json file
+ writesubtitles: Write the video subtitles to a .srt file
"""
params = None
""" Report that the description file is being written """
self.to_screen(u'[info] Writing video description to: ' + descfn)
+ def report_writesubtitles(self, srtfn):
+ """ Report that the subtitles file is being written """
+ self.to_screen(u'[info] Writing video subtitles to: ' + srtfn)
+
def report_writeinfojson(self, infofn):
""" Report that the metadata file has been written """
self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn)
raise MaxDownloadsReached()
filename = self.prepare_filename(info_dict)
-
+
# Forced printings
if self.params.get('forcetitle', False):
print info_dict['title'].encode(preferredencoding(), 'xmlcharrefreplace')
except (OSError, IOError):
self.trouble(u'ERROR: Cannot write description file ' + descfn)
return
+
+ if self.params.get('writesubtitles', False) and 'subtitles' in info_dict and info_dict['subtitles']:
+ # subtitles download errors are already managed as troubles in relevant IE
+ # that way it will silently go on when used with unsupporting IE
+ try:
+ srtfn = filename.rsplit('.', 1)[0] + u'.srt'
+ self.report_writesubtitles(srtfn)
+ srtfile = open(_encodeFilename(srtfn), 'wb')
+ try:
+ srtfile.write(info_dict['subtitles'].encode('utf-8'))
+ finally:
+ srtfile.close()
+ except (OSError, IOError):
+ self.trouble(u'ERROR: Cannot write subtitles file ' + descfn)
+ return
if self.params.get('writeinfojson', False):
infofn = filename + u'.info.json'
except (ContentTooShortError, ), err:
self.trouble(u'ERROR: content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
return
-
+
if success:
try:
self.post_process(filename, info_dict)
'43': '360x640',
'44': '480x854',
'45': '720x1280',
- }
+ }
IE_NAME = u'youtube'
def report_lang(self):
"""Report attempt to download video info webpage."""
self._downloader.to_screen(u'[youtube] %s: Downloading video info webpage' % video_id)
+ def report_video_subtitles_download(self, video_id):
+ """Report attempt to download video info webpage."""
+ self._downloader.to_screen(u'[youtube] %s: Downloading video subtitles' % video_id)
+
def report_information_extraction(self, video_id):
"""Report attempt to extract video information."""
self._downloader.to_screen(u'[youtube] %s: Extracting video information' % video_id)
"""Indicate the download will use the RTMP protocol."""
self._downloader.to_screen(u'[youtube] RTMP download detected')
+ def _closed_captions_xml_to_srt(self, xml_string):
+ srt = ''
+ texts = re.findall(r'<text start="([\d\.]+)"( dur="([\d\.]+)")?>([^<]+)</text>', xml_string, re.MULTILINE)
+ # TODO parse xml instead of regex
+ for n, (start, dur_tag, dur, caption) in enumerate(texts):
+ if not dur: dur = '4'
+ start = float(start)
+ end = start + float(dur)
+ start = "%02i:%02i:%02i,%03i" %(start/(60*60), start/60%60, start%60, start%1*1000)
+ end = "%02i:%02i:%02i,%03i" %(end/(60*60), end/60%60, end%60, end%1*1000)
+ caption = re.sub(ur'(?u)&(.+?);', htmlentity_transform, caption)
+ caption = re.sub(ur'(?u)&(.+?);', htmlentity_transform, caption) # double cycle, inentional
+ srt += str(n) + '\n'
+ srt += start + ' --> ' + end + '\n'
+ srt += caption + '\n\n'
+ return srt
+
def _print_formats(self, formats):
print 'Available formats:'
for x in formats:
vwebpage_doc = lxml.etree.parse(StringIO.StringIO(video_webpage), html_parser)
video_description = u''.join(vwebpage_doc.xpath('id("eow-description")//text()'))
# TODO use another parser
+
+ # closed captions
+ video_subtitles = None
+ if self._downloader.params.get('writesubtitles', False):
+ self.report_video_subtitles_download(video_id)
+ request = urllib2.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
+ try:
+ srt_list = urllib2.urlopen(request).read()
+ except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+ self._downloader.trouble(u'WARNING: unable to download video subtitles: %s' % str(err))
+ else:
+ srt_lang_list = re.findall(r'lang_code="([\w\-]+)"', srt_list)
+ if srt_lang_list:
+ if 'en' in srt_lang_list: srt_lang = 'en'
+ else: srt_lang = srt_lang_list[0] # TODO choose better and provide an override
+ request = urllib2.Request('http://video.google.com/timedtext?hl=en&lang=%s&v=%s' % (srt_lang, video_id))
+ try:
+ srt_xml = urllib2.urlopen(request).read()
+ except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+ self._downloader.trouble(u'WARNING: unable to download video subtitles: %s' % str(err))
+ else:
+ video_subtitles = self._closed_captions_xml_to_srt(srt_xml.decode('utf-8'))
+ else:
+ self._downloader.trouble(u'WARNING: video has no subtitles')
# token
video_token = urllib.unquote_plus(video_info['token'][0])
'thumbnail': video_thumbnail.decode('utf-8'),
'description': video_description,
'player_url': player_url,
+ 'subtitles': video_subtitles
})
except UnavailableVideoError, err:
self._downloader.trouble(u'\nERROR: unable to download video')
video_id = mobj.group(1)
# Retrieve video webpage to extract further information
- request = urllib2.Request("http://vimeo.com/moogaloop/load/clip:%s" % video_id, None, std_headers)
+ request = urllib2.Request(url, None, std_headers)
try:
self.report_download_webpage(video_id)
webpage = urllib2.urlopen(request).read()
# and latter we extract those that are Vimeo specific.
self.report_extraction(video_id)
- # Extract title
- mobj = re.search(r'<caption>(.*?)</caption>', webpage)
- if mobj is None:
- self._downloader.trouble(u'ERROR: unable to extract video title')
+ # Extract the config JSON
+ config = webpage.split(' = {config:')[1].split(',assets:')[0]
+ try:
+ config = json.loads(config)
+ except:
+ self._downloader.trouble(u'ERROR: unable to extract info section')
return
- video_title = mobj.group(1).decode('utf-8')
+
+ # Extract title
+ video_title = config["video"]["title"]
simple_title = _simplify_title(video_title)
# Extract uploader
- mobj = re.search(r'<uploader_url>http://vimeo.com/(.*?)</uploader_url>', webpage)
- if mobj is None:
- self._downloader.trouble(u'ERROR: unable to extract video uploader')
- return
- video_uploader = mobj.group(1).decode('utf-8')
+ video_uploader = config["video"]["owner"]["name"]
# Extract video thumbnail
- mobj = re.search(r'<thumbnail>(.*?)</thumbnail>', webpage)
- if mobj is None:
- self._downloader.trouble(u'ERROR: unable to extract video thumbnail')
- return
- video_thumbnail = mobj.group(1).decode('utf-8')
-
- # # Extract video description
- # mobj = re.search(r'<meta property="og:description" content="(.*)" />', webpage)
- # if mobj is None:
- # 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.'
- video_description = 'Foo.'
-
- # Vimeo specific: extract request signature
- mobj = re.search(r'<request_signature>(.*?)</request_signature>', webpage)
- if mobj is None:
- self._downloader.trouble(u'ERROR: unable to extract request signature')
- return
- sig = mobj.group(1).decode('utf-8')
-
- # Vimeo specific: extract video quality information
- mobj = re.search(r'<isHD>(\d+)</isHD>', webpage)
- if mobj is None:
- self._downloader.trouble(u'ERROR: unable to extract video quality information')
- return
- quality = mobj.group(1).decode('utf-8')
+ video_thumbnail = config["video"]["thumbnail"]
- if int(quality) == 1:
- quality = 'hd'
+ # Extract video description
+ try:
+ lxml.etree
+ except NameError:
+ video_description = u'No description available.'
+ mobj = re.search(r'<meta name="description" content="(.*?)" />', webpage, re.MULTILINE)
+ if mobj is not None:
+ video_description = mobj.group(1)
else:
- quality = 'sd'
+ html_parser = lxml.etree.HTMLParser()
+ vwebpage_doc = lxml.etree.parse(StringIO.StringIO(webpage), html_parser)
+ video_description = u''.join(vwebpage_doc.xpath('id("description")//text()')).strip()
+ # TODO use another parser
- # Vimeo specific: Extract request signature expiration
- mobj = re.search(r'<request_signature_expires>(.*?)</request_signature_expires>', webpage)
- if mobj is None:
- self._downloader.trouble(u'ERROR: unable to extract request signature expiration')
+ # Extract upload date
+ video_upload_date = u'NA'
+ mobj = re.search(r'<span id="clip-date" style="display:none">[^:]*: (.*?)( \([^\(]*\))?</span>', webpage)
+ if mobj is not None:
+ video_upload_date = mobj.group(1)
+
+ # Vimeo specific: extract request signature and timestamp
+ sig = config['request']['signature']
+ timestamp = config['request']['timestamp']
+
+ # Vimeo specific: extract video codec and quality information
+ # TODO bind to format param
+ codecs = [('h264', 'mp4'), ('vp8', 'flv'), ('vp6', 'flv')]
+ for codec in codecs:
+ if codec[0] in config["video"]["files"]:
+ video_codec = codec[0]
+ video_extension = codec[1]
+ if 'hd' in config["video"]["files"][codec[0]]: quality = 'hd'
+ else: quality = 'sd'
+ break
+ else:
+ self._downloader.trouble(u'ERROR: no known codec found')
return
- sig_exp = mobj.group(1).decode('utf-8')
- video_url = "http://vimeo.com/moogaloop/play/clip:%s/%s/%s/?q=%s" % (video_id, sig, sig_exp, quality)
+ video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
+ %(video_id, sig, timestamp, quality, video_codec.upper())
try:
# Process video information
self._downloader.process_info({
- 'id': video_id.decode('utf-8'),
+ 'id': video_id,
'url': video_url,
'uploader': video_uploader,
- 'upload_date': u'NA',
+ 'upload_date': video_upload_date,
'title': video_title,
'stitle': simple_title,
- 'ext': u'mp4',
- 'thumbnail': video_thumbnail.decode('utf-8'),
- 'description': video_description,
+ 'ext': video_extension,
'thumbnail': video_thumbnail,
'description': video_description,
'player_url': None,
class YoutubeSearchIE(InfoExtractor):
"""Information Extractor for YouTube search queries."""
_VALID_URL = r'ytsearch(\d+|all)?:[\s\S]+'
- _TEMPLATE_URL = 'http://www.youtube.com/results?search_query=%s&page=%s&gl=US&hl=en'
- _VIDEO_INDICATOR = r'href="/watch\?v=.+?"'
- _MORE_PAGES_INDICATOR = r'(?m)>\s*Next\s*</a>'
+ _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
_youtube_ie = None
_max_youtube_results = 1000
IE_NAME = u'youtube:search'
"""Downloads a specified number of results for a query"""
video_ids = []
- already_seen = set()
- pagenum = 1
+ pagenum = 0
+ limit = n
- while True:
- self.report_download_page(query, pagenum)
- result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum)
+ while (50 * pagenum) < limit:
+ self.report_download_page(query, pagenum+1)
+ result_url = self._API_URL % (urllib.quote_plus(query), (50*pagenum)+1)
request = urllib2.Request(result_url)
try:
- page = urllib2.urlopen(request).read()
+ data = urllib2.urlopen(request).read()
except (urllib2.URLError, httplib.HTTPException, socket.error), err:
- self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err))
+ self._downloader.trouble(u'ERROR: unable to download API page: %s' % str(err))
return
+ api_response = json.loads(data)['data']
- # Extract video identifiers
- for mobj in re.finditer(self._VIDEO_INDICATOR, page):
- video_id = page[mobj.span()[0]:mobj.span()[1]].split('=')[2][:-1]
- if video_id not in already_seen:
- video_ids.append(video_id)
- already_seen.add(video_id)
- if len(video_ids) == n:
- # Specified n videos reached
- for id in video_ids:
- self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id)
- return
+ new_ids = list(video['id'] for video in api_response['items'])
+ video_ids += new_ids
- if re.search(self._MORE_PAGES_INDICATOR, page) is None:
- for id in video_ids:
- self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id)
- return
+ limit = min(n, api_response['totalItems'])
+ pagenum += 1
- pagenum = pagenum + 1
+ if len(video_ids) > n:
+ video_ids = video_ids[:n]
+ for id in video_ids:
+ self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id)
+ return
class GoogleSearchIE(InfoExtractor):
"""Information Extractor for Google Video search queries."""
_VALID_URL = r'gvsearch(\d+|all)?:[\s\S]+'
_TEMPLATE_URL = 'http://video.google.com/videosearch?q=%s+site:video.google.com&start=%s&hl=en'
- _VIDEO_INDICATOR = r'videoplay\?docid=([^\&>]+)\&'
- _MORE_PAGES_INDICATOR = r'<span>Next</span>'
+ _VIDEO_INDICATOR = r'<a href="http://video\.google\.com/videoplay\?docid=([^"\&]+)'
+ _MORE_PAGES_INDICATOR = r'class="pn" id="pnnext"'
_google_ie = None
_max_google_results = 1000
IE_NAME = u'video.google:search'
"""Downloads a specified number of results for a query"""
video_ids = []
- already_seen = set()
- pagenum = 1
+ pagenum = 0
while True:
self.report_download_page(query, pagenum)
- result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum)
+ result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum*10)
request = urllib2.Request(result_url)
try:
page = urllib2.urlopen(request).read()
# Extract video identifiers
for mobj in re.finditer(self._VIDEO_INDICATOR, page):
video_id = mobj.group(1)
- if video_id not in already_seen:
+ if video_id not in video_ids:
video_ids.append(video_id)
- already_seen.add(video_id)
if len(video_ids) == n:
# Specified n videos reached
for id in video_ids:
# Extract video identifiers
ids_in_page = []
- video_indicator = self._VIDEO_INDICATOR_TEMPLATE % playlist_id
- for mobj in re.finditer(video_indicator, page):
+ for mobj in re.finditer(self._VIDEO_INDICATOR_TEMPLATE % playlist_id, page):
if mobj.group(1) not in ids_in_page:
ids_in_page.append(mobj.group(1))
video_ids.extend(ids_in_page)
playliststart = self._downloader.params.get('playliststart', 1) - 1
playlistend = self._downloader.params.get('playlistend', -1)
-
if playlistend == -1:
video_ids = video_ids[playliststart:]
else:
data = json_data['Post']
else:
data = json_data
-
+
upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
video_url = data['media']['url']
umobj = re.match(self._URL_EXT, video_url)
if umobj is None:
raise ValueError('Can not determine filename extension')
ext = umobj.group(1)
-
+
info = {
'id': data['item_id'],
'url': video_url,
def __init__(self, downloader=None):
InfoExtractor.__init__(self, downloader)
-
+
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, 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)
mobj = re.search('track-description-value"><p>(.*?)</p>', webpage)
if mobj:
description = mobj.group(1)
-
+
# upload date
upload_date = None
mobj = re.search("pretty-date'>on ([\w]+ [\d]+, [\d]+ \d+:\d+)</abbr></h2>", webpage)
url_list = jsonData[fmt][bitrate]
except TypeError: # we have no bitrate info.
url_list = jsonData[fmt]
-
+
return url_list
def check_urls(self, url_list):
info = {
'id': _simplify_title(course + '_' + video),
}
-
+
self.report_extraction(info['id'])
baseUrl = 'http://openclassroom.stanford.edu/MainFolder/courses/' + course + '/videos/'
xmlUrl = baseUrl + video + '.xml'
self._downloader.trouble(u'ERROR: unable to extract performer')
return
performer = _unescapeHTML(mobj.group(1).decode('iso-8859-1'))
- video_title = performer + ' - ' + song_name
+ video_title = performer + ' - ' + song_name
mobj = re.search(r'<meta name="mtvn_uri" content="([^"]+)"/>', webpage)
if mobj is None:
try:
urlh = urllib.urlopen(UPDATE_URL)
newcontent = urlh.read()
-
+
vmatch = re.search("__version__ = '([^']+)'", newcontent)
if vmatch is not None and vmatch.group(1) == __version__:
downloader.to_screen(u'youtube-dl is up-to-date (' + __version__ + ')')
filesystem.add_option('--write-info-json',
action='store_true', dest='writeinfojson',
help='write video metadata to a .info.json file', default=False)
+ filesystem.add_option('--write-srt',
+ action='store_true', dest='writesubtitles',
+ help='write video subtitles to a .srt file', default=False)
postproc.add_option('--extract-audio', action='store_true', dest='extractaudio', default=False,
'updatetime': opts.updatetime,
'writedescription': opts.writedescription,
'writeinfojson': opts.writeinfojson,
+ 'writesubtitles': opts.writesubtitles,
'matchtitle': opts.matchtitle,
'rejecttitle': opts.rejecttitle,
'max_downloads': opts.max_downloads,
parser.error(u'you must provide at least one URL')
else:
sys.exit()
-
+
try:
retcode = fd.download(all_urls)
except MaxDownloadsReached: