X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2F__init__.py;h=5a595901ccc1e8d36a9b24c00966104e05515ebf;hb=6af22cf0efe393825534e19f82e3282a53625d19;hp=789724041be58dc3494057bc8e132efe0e090aa4;hpb=9c228928b6a636ff4ed294e170e3de36d9bb45c5;p=youtube-dl.git diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index 789724041..5a595901c 100755 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -15,6 +15,7 @@ __authors__ = ( 'Kevin Ngo', 'Ori Avtalion', 'shizeeg', + 'Filippo Valsorda', ) __license__ = 'Public Domain' @@ -490,6 +491,8 @@ 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 + subtitleslang: Language of the subtitles to download """ params = None @@ -681,6 +684,10 @@ 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) @@ -808,6 +815,21 @@ 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' @@ -1206,6 +1228,10 @@ 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) @@ -1218,6 +1244,23 @@ 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'([^<]+)', 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: @@ -1389,6 +1432,37 @@ class YoutubeIE(InfoExtractor): 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 self._downloader.params.get('subtitleslang', False): + srt_lang = self._downloader.params.get('subtitleslang') + elif 'en' in srt_lang_list: + srt_lang = 'en' + else: + srt_lang = srt_lang_list[0] + if not srt_lang in srt_lang_list: + self._downloader.trouble(u'WARNING: no closed captions found in the specified language') + else: + 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 closed captions') # token video_token = urllib.unquote_plus(video_info['token'][0]) @@ -1461,6 +1535,7 @@ 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') @@ -2058,7 +2133,7 @@ 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() @@ -2071,77 +2146,75 @@ class VimeoIE(InfoExtractor): # and latter we extract those that are Vimeo specific. self.report_extraction(video_id) - # Extract title - mobj = re.search(r'(.*?)', 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'http://vimeo.com/(.*?)', 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'(.*?)', 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'', 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'(.*?)', webpage) - if mobj is None: - self._downloader.trouble(u'ERROR: unable to extract request signature') - return - sig = mobj.group(1).decode('utf-8') + video_thumbnail = config["video"]["thumbnail"] - # Vimeo specific: extract video quality information - mobj = re.search(r'(\d+)', 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'', 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'(.*?)', 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'', 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, @@ -2168,7 +2241,67 @@ class GenericIE(InfoExtractor): """Report information extraction.""" self._downloader.to_screen(u'[generic] %s: Extracting information' % video_id) + def report_following_redirect(self, new_url): + """Report information extraction.""" + self._downloader.to_screen(u'[redirect] Following redirect to %s' % new_url) + + def _test_redirect(self, url): + """Check if it is a redirect, like url shorteners, in case restart chain.""" + class HeadRequest(urllib2.Request): + def get_method(self): + return "HEAD" + + class HEADRedirectHandler(urllib2.HTTPRedirectHandler): + """ + Subclass the HTTPRedirectHandler to make it use our + HeadRequest also on the redirected URL + """ + def redirect_request(self, req, fp, code, msg, headers, newurl): + if code in (301, 302, 303, 307): + newurl = newurl.replace(' ', '%20') + newheaders = dict((k,v) for k,v in req.headers.items() + if k.lower() not in ("content-length", "content-type")) + return HeadRequest(newurl, + headers=newheaders, + origin_req_host=req.get_origin_req_host(), + unverifiable=True) + else: + raise urllib2.HTTPError(req.get_full_url(), code, msg, headers, fp) + + class HTTPMethodFallback(urllib2.BaseHandler): + """ + Fallback to GET if HEAD is not allowed (405 HTTP error) + """ + def http_error_405(self, req, fp, code, msg, headers): + fp.read() + fp.close() + + newheaders = dict((k,v) for k,v in req.headers.items() + if k.lower() not in ("content-length", "content-type")) + return self.parent.open(urllib2.Request(req.get_full_url(), + headers=newheaders, + origin_req_host=req.get_origin_req_host(), + unverifiable=True)) + + # Build our opener + opener = urllib2.OpenerDirector() + for handler in [urllib2.HTTPHandler, urllib2.HTTPDefaultErrorHandler, + HTTPMethodFallback, HEADRedirectHandler, + urllib2.HTTPErrorProcessor, urllib2.HTTPSHandler]: + opener.add_handler(handler()) + + response = opener.open(HeadRequest(url)) + new_url = response.geturl() + + if url == new_url: return False + + self.report_following_redirect(new_url) + self._downloader.download([new_url]) + return True + def _real_extract(self, url): + if self._test_redirect(url): return + # At this point we have a new video self._downloader.increment_downloads() @@ -2250,9 +2383,7 @@ class GenericIE(InfoExtractor): 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*' + _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' @@ -2303,45 +2434,39 @@ class YoutubeSearchIE(InfoExtractor): """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'Next' + _VIDEO_INDICATOR = r'\s*Next\s*' _youtube_ie = None IE_NAME = u'youtube:playlist' @@ -2571,7 +2694,7 @@ class YoutubePlaylistIE(InfoExtractor): # Extract video identifiers ids_in_page = [] - for mobj in re.finditer(self._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) @@ -2582,7 +2705,10 @@ class YoutubePlaylistIE(InfoExtractor): playliststart = self._downloader.params.get('playliststart', 1) - 1 playlistend = self._downloader.params.get('playlistend', -1) - video_ids = video_ids[playliststart:playlistend] + if playlistend == -1: + video_ids = video_ids[playliststart:] + else: + video_ids = video_ids[playliststart:playlistend] for id in video_ids: self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id) @@ -4328,6 +4454,12 @@ def parseOpts(): action='store', dest='format_limit', metavar='FORMAT', help='highest quality format to download') video_format.add_option('-F', '--list-formats', action='store_true', dest='listformats', help='list all available formats (currently youtube only)') + video_format.add_option('--write-srt', + action='store_true', dest='writesubtitles', + help='write video closed captions to a .srt file (currently youtube only)', default=False) + video_format.add_option('--srt-lang', + action='store', dest='subtitleslang', metavar='LANG', + help='language of the closed captions to download (optional) use IETF language tags like \'en\'') verbosity.add_option('-q', '--quiet', @@ -4592,6 +4724,8 @@ def _real_main(): 'updatetime': opts.updatetime, 'writedescription': opts.writedescription, 'writeinfojson': opts.writeinfojson, + 'writesubtitles': opts.writesubtitles, + 'subtitleslang': opts.subtitleslang, 'matchtitle': opts.matchtitle, 'rejecttitle': opts.rejecttitle, 'max_downloads': opts.max_downloads,