"""Report attempt to download video info webpage."""
self._downloader.to_screen(u'[youtube] %s: Downloading video subtitles' % video_id)
+ def report_video_subtitles_request(self, video_id, lang):
+ """Report attempt to download video info webpage."""
+ self._downloader.to_screen(u'[youtube] %s: Downloading video subtitles for lang: %s' % (video_id,lang))
+
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 = unescapeHTML(caption)
- caption = unescapeHTML(caption) # double cycle, intentional
- srt += str(n+1) + '\n'
- srt += start + ' --> ' + end + '\n'
- srt += caption + '\n\n'
- return srt
-
- def _extract_subtitles(self, video_id):
- self.report_video_subtitles_download(video_id)
+ def _get_available_subtitles(self, video_id):
request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
try:
srt_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
srt_lang_list = dict((l[1], l[0]) for l in srt_lang_list)
if not srt_lang_list:
return (u'WARNING: video has no closed captions', None)
- 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 = list(srt_lang_list.keys())[0]
- if not srt_lang in srt_lang_list:
- return (u'WARNING: no closed captions found in the specified language', None)
+ return srt_lang_list
+
+ def _request_subtitle(self, str_lang, str_name, video_id, format = 'srt'):
+ self.report_video_subtitles_request(video_id, str_lang)
params = compat_urllib_parse.urlencode({
- 'lang': srt_lang,
- 'name': srt_lang_list[srt_lang].encode('utf-8'),
+ 'lang': str_lang,
+ 'name': str_name,
'v': video_id,
+ 'fmt': format,
})
url = 'http://www.youtube.com/api/timedtext?' + params
try:
- srt_xml = compat_urllib_request.urlopen(url).read().decode('utf-8')
+ srt = compat_urllib_request.urlopen(url).read().decode('utf-8')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
return (u'WARNING: unable to download video subtitles: %s' % compat_str(err), None)
- if not srt_xml:
+ if not srt:
return (u'WARNING: Did not fetch video subtitles', None)
- return (None, self._closed_captions_xml_to_srt(srt_xml))
+ return (None, str_lang, srt)
+
+ def _extract_subtitle(self, video_id):
+ self.report_video_subtitles_download(video_id)
+ srt_lang_list = self._get_available_subtitles(video_id)
+
+ 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 = list(srt_lang_list.keys())[0]
+ if not srt_lang in srt_lang_list:
+ return (u'WARNING: no closed captions found in the specified language "%s"' % srt_lang, None)
+
+ sub = self._request_subtitle(srt_lang, srt_lang_list[srt_lang].encode('utf-8'), video_id)
+ return [sub]
+
+ def _extract_all_subtitles(self, video_id):
+ self.report_video_subtitles_download(video_id)
+ srt_lang_list = self._get_available_subtitles(video_id)
+ subs = []
+ for srt_lang in srt_lang_list:
+ sub = self._request_subtitle(srt_lang, srt_lang_list[srt_lang].encode('utf-8'), video_id)
+ subs.append(sub)
+ return subs
def _print_formats(self, formats):
print('Available formats:')
# closed captions
video_subtitles = None
+
if self._downloader.params.get('writesubtitles', False):
- (srt_error, video_subtitles) = self._extract_subtitles(video_id)
- if srt_error:
- self._downloader.trouble(srt_error)
+ video_subtitles = self._extract_subtitle(video_id)
+ if video_subtitles:
+ (srt_error, srt_lang, srt) = video_subtitles[0]
+ if srt_error:
+ self._downloader.trouble(srt_error)
+
+ if self._downloader.params.get('allsubtitles', False):
+ video_subtitles = self._extract_all_subtitles(video_id)
+ for video_subtitle in video_subtitles:
+ (srt_error, srt_lang, srt) = video_subtitle
+ if srt_error:
+ self._downloader.trouble(srt_error)
if 'length_seconds' not in video_info:
self._downloader.trouble(u'WARNING: unable to extract video duration')
opener = compat_urllib_request.OpenerDirector()
for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler,
HTTPMethodFallback, HEADRedirectHandler,
- compat_urllib_error.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]:
+ compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]:
opener.add_handler(handler())
response = opener.open(HeadRequest(url))
if mobj is None:
# Broaden the search a little bit
mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
+ if mobj is None:
+ # Broaden the search a little bit: JWPlayer JS loader
+ mobj = re.search(r'[^A-Za-z0-9]?file:\s*["\'](http[^\'"&]*)', webpage)
if mobj is None:
self._downloader.trouble(u'ERROR: Invalid URL: %s' % url)
return
params_raw = compat_urllib_parse.unquote(data['params'])
params = json.loads(params_raw)
video_url = params['hd_src']
+ if not video_url:
+ video_url = params['sd_src']
+ if not video_url:
+ raise ExtractorError(u'Cannot find video URL')
video_duration = int(params['video_duration'])
m = re.search('<h2 class="uiHeaderTitle">([^<]+)</h2>', webpage)
webpage = self._download_webpage(webpage_url, video_id)
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\' />',
+ 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')
webpage = self._download_webpage(req, video_id)
# Get the video title
- result = re.search(r'videoTitleArea">(?P<title>.*)</h1>', webpage)
+ result = re.search(r'<h1.*?>(?P<title>.*)</h1>', webpage)
if result is None:
- raise ExtractorError(u'ERROR: unable to extract video title')
+ raise ExtractorError(u'Unable to extract video title')
video_title = result.group('title').strip()
# Get the video date
- result = re.search(r'Date:</b>(?P<date>.*)</li>', webpage)
+ result = re.search(r'Date:</label>(?P<date>.*) </li>', webpage)
if result is None:
self._downloader.to_stderr(u'WARNING: unable to extract video date')
upload_date = None
upload_date = result.group('date').strip()
# Get the video uploader
- result = re.search(r'Submitted:</b>(?P<uploader>.*)</li>', webpage)
+ result = re.search(r'Submitted:</label>(?P<uploader>.*)</li>', webpage)
if result is None:
- self._downloader.to_stderr(u'ERROR: unable to extract uploader')
+ self._downloader.to_stderr(u'WARNING: unable to extract uploader')
video_uploader = None
else:
video_uploader = result.group('uploader').strip()
projects / youtube-dl.git / blobdiff