Merge branch 'master' into extract_info_rewrite

This commit is contained in:
Jaime Marquínez Ferrándiz 2013-03-28 13:02:04 +01:00
commit cfa90f4adc
6 changed files with 258 additions and 76 deletions

View file

@ -29,6 +29,7 @@
"simulate": false,
"skip_download": false,
"subtitleslang": null,
"subtitlesformat": "srt",
"test": true,
"updatetime": true,
"usenetrc": false,
@ -36,5 +37,8 @@
"verbose": true,
"writedescription": false,
"writeinfojson": true,
"writesubtitles": false
}
"writesubtitles": false,
"onlysubtitles": false,
"allsubtitles": false,
"listssubtitles": false
}

View file

@ -38,20 +38,63 @@ def download(self, x):
md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
class TestYoutubeSubtitles(unittest.TestCase):
def setUp(self):
DL = FakeDownloader()
DL.params['allsubtitles'] = False
DL.params['writesubtitles'] = False
DL.params['subtitlesformat'] = 'srt'
DL.params['listsubtitles'] = False
def test_youtube_no_subtitles(self):
DL = FakeDownloader()
DL.params['writesubtitles'] = False
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
subtitles = info_dict[0]['subtitles']
self.assertEqual(subtitles, None)
def test_youtube_subtitles(self):
DL = FakeDownloader()
DL.params['writesubtitles'] = True
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
self.assertEqual(md5(info_dict[0]['subtitles']), 'c3228550d59116f3c29fba370b55d033')
sub = info_dict[0]['subtitles'][0]
self.assertEqual(md5(sub[2]), '4cd9278a35ba2305f47354ee13472260')
def test_youtube_subtitles_it(self):
DL = FakeDownloader()
DL.params['writesubtitles'] = True
DL.params['subtitleslang'] = 'it'
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
self.assertEqual(md5(info_dict[0]['subtitles']), '132a88a0daf8e1520f393eb58f1f646a')
sub = info_dict[0]['subtitles'][0]
self.assertEqual(md5(sub[2]), '164a51f16f260476a05b50fe4c2f161d')
def test_youtube_onlysubtitles(self):
DL = FakeDownloader()
DL.params['writesubtitles'] = True
DL.params['onlysubtitles'] = True
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles'][0]
self.assertEqual(md5(sub[2]), '4cd9278a35ba2305f47354ee13472260')
def test_youtube_allsubtitles(self):
DL = FakeDownloader()
DL.params['allsubtitles'] = True
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
subtitles = info_dict[0]['subtitles']
self.assertEqual(len(subtitles), 12)
def test_youtube_subtitles_format(self):
DL = FakeDownloader()
DL.params['writesubtitles'] = True
DL.params['subtitlesformat'] = 'sbv'
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles'][0]
self.assertEqual(md5(sub[2]), '13aeaa0c245a8bed9a451cb643e3ad8b')
def test_youtube_list_subtitles(self):
DL = FakeDownloader()
DL.params['listsubtitles'] = True
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
self.assertEqual(info_dict, None)
if __name__ == '__main__':
unittest.main()

View file

@ -293,5 +293,20 @@
"info_dict": {
"title": "Absolute Mehrheit vom 17.02.2013 - Die Highlights, Teil 2"
}
},
{
"name": "Generic",
"url": "http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html",
"file": "13601338388002.mp4",
"md5": "85b90ccc9d73b4acd9138d3af4c27f89"
},
{
"name": "Spiegel",
"url": "http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html",
"file": "1259285.mp4",
"md5": "2c2754212136f35fb4b19767d242f66e",
"info_dict": {
"title": "Vulkanausbruch in Ecuador: Der \"Feuerschlund\" ist wieder aktiv"
}
}
]

View file

@ -78,7 +78,11 @@ 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
writesubtitles: Write the video subtitles to a file
onlysubtitles: Downloads only the subtitles of the video
allsubtitles: Downloads all the subtitles of the video
listsubtitles: Lists all available subtitles for the video
subtitlesformat: Subtitle format [sbv/srt] (default=srt)
subtitleslang: Language of the subtitles to download
test: Download only first bytes to test the downloader.
keepvideo: Keep the video file after post-processing
@ -301,9 +305,9 @@ def report_writedescription(self, descfn):
""" Report that the description file is being written """
self.to_screen(u'[info] Writing video description to: ' + descfn)
def report_writesubtitles(self, srtfn):
def report_writesubtitles(self, sub_filename):
""" Report that the subtitles file is being written """
self.to_screen(u'[info] Writing video subtitles to: ' + srtfn)
self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename)
def report_writeinfojson(self, infofn):
""" Report that the metadata file has been written """
@ -372,8 +376,11 @@ def prepare_filename(self, info_dict):
filename = self.params['outtmpl'] % template_dict
return filename
except (ValueError, KeyError) as err:
self.trouble(u'ERROR: invalid system charset or erroneous output template')
except KeyError as err:
self.trouble(u'ERROR: Erroneous output template')
return None
except ValueError as err:
self.trouble(u'ERROR: Insufficient system charset ' + repr(preferredencoding()))
return None
def _match_entry(self, info_dict):
@ -519,14 +526,35 @@ def process_info(self, info_dict):
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
subtitle = info_dict['subtitles'][0]
(sub_error, sub_lang, sub) = subtitle
sub_format = self.params.get('subtitlesformat')
try:
srtfn = filename.rsplit('.', 1)[0] + u'.srt'
self.report_writesubtitles(srtfn)
with io.open(encodeFilename(srtfn), 'w', encoding='utf-8') as srtfile:
srtfile.write(info_dict['subtitles'])
sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
self.report_writesubtitles(sub_filename)
with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
subfile.write(sub)
except (OSError, IOError):
self.trouble(u'ERROR: Cannot write subtitles file ' + descfn)
return
if self.params.get('onlysubtitles', False):
return
if self.params.get('allsubtitles', False) and 'subtitles' in info_dict and info_dict['subtitles']:
subtitles = info_dict['subtitles']
sub_format = self.params.get('subtitlesformat')
for subtitle in subtitles:
(sub_error, sub_lang, sub) = subtitle
try:
sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
self.report_writesubtitles(sub_filename)
with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
subfile.write(sub)
except (OSError, IOError):
self.trouble(u'ERROR: Cannot write subtitles file ' + descfn)
return
if self.params.get('onlysubtitles', False):
return
if self.params.get('writeinfojson', False):
infofn = filename + u'.info.json'

View file

@ -48,7 +48,7 @@ class InfoExtractor(object):
uploader_id: Nickname or id of the video uploader.
location: Physical location of the video.
player_url: SWF Player URL (used for rtmpdump).
subtitles: The .srt file contents.
subtitles: The subtitle file contents.
urlhandle: [internal] The urlHandle to be used to download the file,
like returned by urllib.request.urlopen
@ -126,8 +126,14 @@ def _request_webpage(self, url_or_request, video_id, note=None, errnote=None):
def _download_webpage(self, url_or_request, video_id, note=None, errnote=None):
""" Returns the data of the page as a string """
urlh = self._request_webpage(url_or_request, video_id, note, errnote)
content_type = urlh.headers.get('Content-Type', '')
m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
if m:
encoding = m.group(1)
else:
encoding = 'utf-8'
webpage_bytes = urlh.read()
return webpage_bytes.decode('utf-8', 'replace')
return webpage_bytes.decode(encoding, 'replace')
#Methods for following #608
#They set the correct value of the '_type' key
@ -236,7 +242,16 @@ def report_video_info_webpage_download(self, 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)
self._downloader.to_screen(u'[youtube] %s: Checking available subtitles' % video_id)
def report_video_subtitles_request(self, video_id, sub_lang, format):
"""Report attempt to download video info webpage."""
self._downloader.to_screen(u'[youtube] %s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format))
def report_video_subtitles_available(self, video_id, sub_lang_list):
"""Report available subtitles."""
sub_lang = ",".join(list(sub_lang_list.keys()))
self._downloader.to_screen(u'[youtube] %s: Available subtitles for video: %s' % (video_id, sub_lang))
def report_information_extraction(self, video_id):
"""Report attempt to extract video information."""
@ -250,55 +265,63 @@ def report_rtmp_download(self):
"""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):
def _get_available_subtitles(self, video_id):
self.report_video_subtitles_download(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')
sub_list = compat_urllib_request.urlopen(request).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)
srt_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', srt_list)
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)
sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list)
if not sub_lang_list:
return (u'WARNING: video doesn\'t have subtitles', None)
return sub_lang_list
def _list_available_subtitles(self, video_id):
sub_lang_list = self._get_available_subtitles(video_id)
self.report_video_subtitles_available(video_id, sub_lang_list)
def _request_subtitle(self, sub_lang, sub_name, video_id, format):
self.report_video_subtitles_request(video_id, sub_lang, format)
params = compat_urllib_parse.urlencode({
'lang': srt_lang,
'name': srt_lang_list[srt_lang].encode('utf-8'),
'lang': sub_lang,
'name': sub_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')
sub = 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 sub:
return (u'WARNING: Did not fetch video subtitles', None)
return (None, self._closed_captions_xml_to_srt(srt_xml))
return (None, sub_lang, sub)
def _extract_subtitle(self, video_id):
sub_lang_list = self._get_available_subtitles(video_id)
sub_format = self._downloader.params.get('subtitlesformat')
if self._downloader.params.get('subtitleslang', False):
sub_lang = self._downloader.params.get('subtitleslang')
elif 'en' in sub_lang_list:
sub_lang = 'en'
else:
sub_lang = list(sub_lang_list.keys())[0]
if not sub_lang in sub_lang_list:
return (u'WARNING: no closed captions found in the specified language "%s"' % sub_lang, None)
subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
return [subtitle]
def _extract_all_subtitles(self, video_id):
sub_lang_list = self._get_available_subtitles(video_id)
sub_format = self._downloader.params.get('subtitlesformat')
subtitles = []
for sub_lang in sub_lang_list:
subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
subtitles.append(subtitle)
return subtitles
def _print_formats(self, formats):
print('Available formats:')
@ -519,12 +542,26 @@ def _real_extract(self, url):
else:
video_description = ''
# closed captions
# subtitles
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:
(sub_error, sub_lang, sub) = video_subtitles[0]
if sub_error:
self._downloader.trouble(sub_error)
if self._downloader.params.get('allsubtitles', False):
video_subtitles = self._extract_all_subtitles(video_id)
for video_subtitle in video_subtitles:
(sub_error, sub_lang, sub) = video_subtitle
if sub_error:
self._downloader.trouble(sub_error)
if self._downloader.params.get('listsubtitles', False):
sub_lang_list = self._list_available_subtitles(video_id)
return
if 'length_seconds' not in video_info:
self._downloader.trouble(u'WARNING: unable to extract video duration')
@ -1299,7 +1336,8 @@ def __init__(self, downloader=None):
def report_download_webpage(self, video_id):
"""Report webpage download."""
self._downloader.to_screen(u'WARNING: Falling back on generic information extractor.')
if not self._downloader.params.get('test', False):
self._downloader.to_screen(u'WARNING: Falling back on generic information extractor.')
self._downloader.to_screen(u'[generic] %s: Downloading webpage' % video_id)
def report_extraction(self, video_id):
@ -1369,13 +1407,8 @@ def _real_extract(self, url):
if new_url: return [self.url_result(new_url)]
video_id = url.split('/')[-1]
request = compat_urllib_request.Request(url)
try:
self.report_download_webpage(video_id)
webpage = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % compat_str(err))
return
webpage = self._download_webpage(url, video_id)
except ValueError as err:
# since this is the last-resort InfoExtractor, if
# this error is thrown, it'll be thrown here
@ -2576,7 +2609,7 @@ def _real_extract(self, url):
'uploader': showName,
'upload_date': None,
'title': showName,
'ext': 'flv',
'ext': 'mp4',
'thumbnail': imgUrl,
'description': description,
'player_url': playerUrl,
@ -3972,11 +4005,11 @@ def _real_extract(self, url):
webpage = self._download_webpage(url, video_id)
m = re.search(r'<meta property="og:title" content="(?P<title>.+)"', webpage)
title = unescapeHTML(m.group('title'))
m = re.search(r'<div class="bio-names-and-report">[\s\n]+<h4>(?P<uploader>\w+)</h4>', webpage)
uploader = unescapeHTML(m.group('uploader'))
m = re.search(r'<div class="user-name-and-bio">[\S\s]+?<h2>(?P<uploader>.+?)</h2>', webpage)
uploader = clean_html(m.group('uploader'))
info = {
'id':video_id,
'url':video_url,
'id': video_id,
'url': video_url,
'ext': 'mp4',
'title': title,
'thumbnail': thumbnail,
@ -4113,6 +4146,40 @@ def _real_extract(self, url):
}
return [info]
class SpiegelIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?$'
def _real_extract(self, url):
m = re.match(self._VALID_URL, url)
video_id = m.group('videoID')
webpage = self._download_webpage(url, video_id)
m = re.search(r'<div class="spVideoTitle">(.*?)</div>', webpage)
if not m:
raise ExtractorError(u'Cannot find title')
video_title = unescapeHTML(m.group(1))
xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml'
xml_code = self._download_webpage(xml_url, video_id,
note=u'Downloading XML', errnote=u'Failed to download XML')
idoc = xml.etree.ElementTree.fromstring(xml_code)
last_type = idoc[-1]
filename = last_type.findall('./filename')[0].text
duration = float(last_type.findall('./duration')[0].text)
video_url = 'http://video2.spiegel.de/flash/' + filename
video_ext = filename.rpartition('.')[2]
info = {
'id': video_id,
'url': video_url,
'ext': video_ext,
'title': video_title,
'duration': duration,
}
return [info]
def gen_extractors():
""" Return a list of an instance of every supported extractor.
The order does matter; the first extractor matched is the one handling the URL.
@ -4161,6 +4228,7 @@ def gen_extractors():
KeekIE(),
TEDIE(),
MySpassIE(),
SpiegelIE(),
GenericIE()
]

View file

@ -173,12 +173,24 @@ def _find_term_columns():
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',
video_format.add_option('--write-sub', '--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',
help='write subtitle file (currently youtube only)', default=False)
video_format.add_option('--only-sub',
action='store_true', dest='onlysubtitles',
help='downloads only the subtitles (no video)', default=False)
video_format.add_option('--all-subs',
action='store_true', dest='allsubtitles',
help='downloads all the available subtitles of the video (currently youtube only)', default=False)
video_format.add_option('--list-subs',
action='store_true', dest='listsubtitles',
help='lists all available subtitles for the video (currently youtube only)', default=False)
video_format.add_option('--sub-format',
action='store', dest='subtitlesformat', metavar='LANG',
help='subtitle format [srt/sbv] (default=srt) (currently youtube only)', default='srt')
video_format.add_option('--sub-lang', '--srt-lang',
action='store', dest='subtitleslang', metavar='LANG',
help='language of the closed captions to download (optional) use IETF language tags like \'en\'')
help='language of the subtitles to download (optional) use IETF language tags like \'en\'')
verbosity.add_option('-q', '--quiet',
action='store_true', dest='quiet', help='activates quiet mode', default=False)
@ -274,12 +286,20 @@ def _find_term_columns():
xdg_config_home = os.environ.get('XDG_CONFIG_HOME')
if xdg_config_home:
userConf = os.path.join(xdg_config_home, 'youtube-dl.conf')
userConfFile = os.path.join(xdg_config_home, 'youtube-dl.conf')
else:
userConf = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl.conf')
argv = _readOptions('/etc/youtube-dl.conf') + _readOptions(userConf) + sys.argv[1:]
userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl.conf')
systemConf = _readOptions('/etc/youtube-dl.conf')
userConf = _readOptions(userConfFile)
commandLineConf = sys.argv[1:]
argv = systemConf + userConf + commandLineConf
opts, args = parser.parse_args(argv)
if opts.verbose:
print(u'[debug] System config: ' + repr(systemConf))
print(u'[debug] User config: ' + repr(userConf))
print(u'[debug] Command-line args: ' + repr(commandLineConf))
return parser, opts, args
def _real_main():
@ -450,6 +470,10 @@ def _real_main():
'writedescription': opts.writedescription,
'writeinfojson': opts.writeinfojson,
'writesubtitles': opts.writesubtitles,
'onlysubtitles': opts.onlysubtitles,
'allsubtitles': opts.allsubtitles,
'listsubtitles': opts.listsubtitles,
'subtitlesformat': opts.subtitlesformat,
'subtitleslang': opts.subtitleslang,
'matchtitle': decodeOption(opts.matchtitle),
'rejecttitle': decodeOption(opts.rejecttitle),