X-Git-Url: https://git.jankratochvil.net/?a=blobdiff_plain;f=youtube-dl;h=5c81973cd4ea33c1c470fa44d4940dce2f7ef90c;hb=115d243428b788b98fd8f1654e58a2cf27833c61;hp=b22f1cac274fa165563f1f7e32b5a5ee474644f6;hpb=38ffbc022232762ad2b936d070b4fbaf3f9acb09;p=youtube-dl.git

diff --git a/youtube-dl b/youtube-dl
index b22f1ca..5c81973 100755
--- a/youtube-dl
+++ b/youtube-dl
@@ -18,12 +18,14 @@ __authors__  = (
 	)
 
 __license__ = 'Public Domain'
-__version__ = '2012.01.08'
+__version__ = '2012.02.27'
 
 UPDATE_URL = 'https://raw.github.com/rg3/youtube-dl/master/youtube-dl'
 
+
 import cookielib
 import datetime
+import getpass
 import gzip
 import htmlentitydefs
 import HTMLParser
@@ -31,9 +33,11 @@ import httplib
 import locale
 import math
 import netrc
+import optparse
 import os
 import os.path
 import re
+import shlex
 import socket
 import string
 import subprocess
@@ -305,7 +309,14 @@ def _encodeFilename(s):
 	"""
 
 	assert type(s) == type(u'')
-	return s.encode(sys.getfilesystemencoding(), 'ignore')
+
+	if sys.platform == 'win32' and sys.getwindowsversion().major >= 5:
+		# Pass u'' directly to use Unicode APIs on Windows 2000 and up
+		# (Detecting Windows NT 4 is tricky because 'major >= 4' would
+		# match Windows 9x series as well. Besides, NT 4 is obsolete.)
+		return s
+	else:
+		return s.encode(sys.getfilesystemencoding(), 'ignore')
 
 class DownloadError(Exception):
 	"""Download Error exception.
@@ -479,6 +490,7 @@ class FileDownloader(object):
 	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
@@ -670,6 +682,10 @@ class FileDownloader(object):
 		""" 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)
@@ -797,6 +813,21 @@ class FileDownloader(object):
 			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'
@@ -889,7 +920,15 @@ class FileDownloader(object):
 		# 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]
-		retval = subprocess.call(basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)])
+		args = basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)]
+		if self.params.get('verbose', False):
+			try:
+				import pipes
+				shell_quote = lambda args: ' '.join(map(pipes.quote, args))
+			except ImportError:
+				shell_quote = repr
+			self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(args))
+		retval = subprocess.call(args)
 		while retval == 2 or retval == 1:
 			prevsize = os.path.getsize(_encodeFilename(tmpfilename))
 			self.to_screen(u'\r[rtmpdump] %s bytes' % prevsize, skip_eol=True)
@@ -1187,6 +1226,10 @@ class YoutubeIE(InfoExtractor):
 		"""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)
@@ -1199,6 +1242,23 @@ class YoutubeIE(InfoExtractor):
 		"""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:
@@ -1362,15 +1422,38 @@ class YoutubeIE(InfoExtractor):
 			lxml.etree
 		except NameError:
 			video_description = u'No description available.'
-			if self._downloader.params.get('forcedescription', False) or self._downloader.params.get('writedescription', False):
-				mobj = re.search(r'<meta name="description" content="(.*)"(?:\s*/)?>', video_webpage)
-				if mobj is not None:
-					video_description = mobj.group(1).decode('utf-8')
+			mobj = re.search(r'<meta name="description" content="(.*?)">', video_webpage)
+			if mobj is not None:
+				video_description = mobj.group(1).decode('utf-8')
 		else:
 			html_parser = lxml.etree.HTMLParser(encoding='utf-8')
 			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])
@@ -1443,6 +1526,7 @@ class YoutubeIE(InfoExtractor):
 					'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')
@@ -2040,7 +2124,7 @@ class VimeoIE(InfoExtractor):
 		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()
@@ -2053,77 +2137,75 @@ class VimeoIE(InfoExtractor):
 		# 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')
+		video_thumbnail = config["video"]["thumbnail"]
 
-		# # 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')
-
-		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,
@@ -2322,8 +2404,8 @@ 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'
@@ -2374,12 +2456,11 @@ class GoogleSearchIE(InfoExtractor):
 		"""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()
@@ -2390,9 +2471,8 @@ class GoogleSearchIE(InfoExtractor):
 			# 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:
@@ -2652,7 +2732,7 @@ class YoutubeUserIE(InfoExtractor):
 		else:
 			video_ids = video_ids[playliststart:playlistend]
 
-		self._downloader.to_screen("[youtube] user %s: Collected %d video ids (downloading %d of them)" %
+		self._downloader.to_screen(u"[youtube] user %s: Collected %d video ids (downloading %d of them)" %
 				(username, all_ids_count, len(video_ids)))
 
 		for video_id in video_ids:
@@ -4170,7 +4250,7 @@ def updateSelf(downloader, filename):
 	if not os.access(filename, os.W_OK):
 		sys.exit('ERROR: no write permissions on %s' % filename)
 
-	downloader.to_screen('Updating to latest version...')
+	downloader.to_screen(u'Updating to latest version...')
 
 	try:
 		try:
@@ -4179,7 +4259,7 @@ def updateSelf(downloader, filename):
 			
 			vmatch = re.search("__version__ = '([^']+)'", newcontent)
 			if vmatch is not None and vmatch.group(1) == __version__:
-				downloader.to_screen('youtube-dl is up-to-date (' + __version__ + ')')
+				downloader.to_screen(u'youtube-dl is up-to-date (' + __version__ + ')')
 				return
 		finally:
 			urlh.close()
@@ -4195,14 +4275,9 @@ def updateSelf(downloader, filename):
 	except (IOError, OSError), err:
 		sys.exit('ERROR: unable to overwrite current version')
 
-	downloader.to_screen('Updated youtube-dl. Restart youtube-dl to use the new version.')
+	downloader.to_screen(u'Updated youtube-dl. Restart youtube-dl to use the new version.')
 
 def parseOpts():
-	# Deferred imports
-	import getpass
-	import optparse
-	import shlex
-
 	def _readOptions(filename_bytes):
 		try:
 			optionf = open(filename_bytes)
@@ -4344,6 +4419,8 @@ def parseOpts():
 	verbosity.add_option('--console-title',
 			action='store_true', dest='consoletitle',
 			help='display progress in console titlebar', default=False)
+	verbosity.add_option('-v', '--verbose',
+			action='store_true', dest='verbose', help='print various debugging information', default=False)
 
 
 	filesystem.add_option('-t', '--title',
@@ -4360,7 +4437,7 @@ def parseOpts():
 	filesystem.add_option('-w', '--no-overwrites',
 			action='store_true', dest='nooverwrites', help='do not overwrite files', default=False)
 	filesystem.add_option('-c', '--continue',
-			action='store_true', dest='continue_dl', help='resume partially downloaded files', default=False)
+			action='store_true', dest='continue_dl', help='resume partially downloaded files', default=True)
 	filesystem.add_option('--no-continue',
 			action='store_false', dest='continue_dl',
 			help='do not resume partially downloaded files (restart from beginning)')
@@ -4377,6 +4454,9 @@ def parseOpts():
 	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,
@@ -4480,10 +4560,14 @@ def _real_main():
 
 	# General configuration
 	cookie_processor = urllib2.HTTPCookieProcessor(jar)
-	opener = urllib2.build_opener(urllib2.ProxyHandler(), cookie_processor, YoutubeDLHandler())
+	proxy_handler = urllib2.ProxyHandler()
+	opener = urllib2.build_opener(proxy_handler, cookie_processor, YoutubeDLHandler())
 	urllib2.install_opener(opener)
 	socket.setdefaulttimeout(300) # 5 minutes should be enough (famous last words)
 
+	if opts.verbose:
+		print(u'[debug] Proxy map: ' + str(proxy_handler.proxies))
+
 	extractors = gen_extractors()
 
 	if opts.list_extractors:
@@ -4573,10 +4657,12 @@ def _real_main():
 		'updatetime': opts.updatetime,
 		'writedescription': opts.writedescription,
 		'writeinfojson': opts.writeinfojson,
+		'writesubtitles': opts.writesubtitles,
 		'matchtitle': opts.matchtitle,
 		'rejecttitle': opts.rejecttitle,
 		'max_downloads': opts.max_downloads,
 		'prefer_free_formats': opts.prefer_free_formats,
+		'verbose': opts.verbose,
 		})
 	for extractor in extractors:
 		fd.add_info_extractor(extractor)