X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2F__init__.py;h=752d762ebc05fe45b7bd8a475d06ac19114a81b0;hb=c6f45d431482503e86f72a6f4f63d1090a312b28;hp=a1871ca1c2a17920de98ca0e13c4f188fa396c84;hpb=4a34b7252e9909619fe2ba3ba01c2f9471681a81;p=youtube-dl.git
diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py
index a1871ca1c..752d762eb 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'
@@ -66,11 +67,6 @@ try:
except ImportError:
from cgi import parse_qs
-try:
- import lxml.etree
-except ImportError:
- pass # Handled below
-
try:
import xml.etree.ElementTree
except ImportError: # Python<2.5: Not officially supported, but let it slip
@@ -197,6 +193,69 @@ except ImportError: # Python <2.6, use trivialjson (https://github.com/phihag/tr
raise ValueError('Extra data at end of input (index ' + str(i) + ' of ' + repr(s) + ': ' + repr(s[i:]) + ')')
return res
+
+class IDParser(HTMLParser.HTMLParser):
+ """Modified HTMLParser that isolates a tag with the specified id"""
+ def __init__(self, id):
+ self.id = id
+ self.result = None
+ self.started = False
+ self.depth = {}
+ self.html = None
+ self.watch_startpos = False
+ HTMLParser.HTMLParser.__init__(self)
+
+ def loads(self, html):
+ self.html = html
+ self.feed(html)
+ self.close()
+
+ def handle_starttag(self, tag, attrs):
+ attrs = dict(attrs)
+ if self.started:
+ self.find_startpos(None)
+ if 'id' in attrs and attrs['id'] == self.id:
+ self.result = [tag]
+ self.started = True
+ self.watch_startpos = True
+ if self.started:
+ if not tag in self.depth: self.depth[tag] = 0
+ self.depth[tag] += 1
+
+ def handle_endtag(self, tag):
+ if self.started:
+ if tag in self.depth: self.depth[tag] -= 1
+ if self.depth[self.result[0]] == 0:
+ self.started = False
+ self.result.append(self.getpos())
+
+ def find_startpos(self, x):
+ """Needed to put the start position of the result (self.result[1])
+ after the opening tag with the requested id"""
+ if self.watch_startpos:
+ self.watch_startpos = False
+ self.result.append(self.getpos())
+ handle_entityref = handle_charref = handle_data = handle_comment = \
+ handle_decl = handle_pi = unknown_decl = find_startpos
+
+ def get_result(self):
+ if self.result == None: return None
+ if len(self.result) != 3: return None
+ lines = self.html.split('\n')
+ lines = lines[self.result[1][0]-1:self.result[2][0]]
+ lines[0] = lines[0][self.result[1][1]:]
+ if len(lines) == 1:
+ lines[-1] = lines[-1][:self.result[2][1]-self.result[1][1]]
+ lines[-1] = lines[-1][:self.result[2][1]]
+ return '\n'.join(lines).strip()
+
+def get_element_by_id(id, html):
+ """Return the content of the tag with the specified id in the passed HTML document"""
+ parser = IDParser(id)
+ parser.loads(html)
+ return parser.get_result()
+
+
def preferredencoding():
"""Get preferred encoding.
@@ -241,6 +300,18 @@ def htmlentity_transform(matchobj):
return (u'&%s;' % entity)
+def clean_html(html):
+ """Clean an HTML snippet into a readable string"""
+ # Newline vs
+ html = html.replace('\n', ' ')
+ html = re.sub('\s*<\s*br\s*/?\s*>\s*', '\n', html)
+ # Strip html tags
+ html = re.sub('<.*?>', '', html)
+ # Replace html entities
+ html = re.sub(ur'(?u)&(.+?);', htmlentity_transform, html)
+ return html
+
+
def sanitize_title(utitle):
"""Sanitizes a video title so it could be used as part of a filename."""
utitle = re.sub(ur'(?u)&(.+?);', htmlentity_transform, utitle)
@@ -490,6 +561,8 @@ 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 +754,10 @@ 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 +885,21 @@ 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 +1298,10 @@ 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 +1314,23 @@ 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:
@@ -1377,18 +1490,40 @@ class YoutubeIE(InfoExtractor):
pass
# description
- try:
- lxml.etree
- except NameError:
- video_description = u'No description available.'
- mobj = re.search(r'', 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
+ video_description = get_element_by_id("eow-description", video_webpage)
+ if video_description: video_description = clean_html(video_description.decode('utf8'))
+ else: video_description = ''
+
+ # 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 +1596,7 @@ 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')
@@ -2090,18 +2226,9 @@ class VimeoIE(InfoExtractor):
video_thumbnail = config["video"]["thumbnail"]
# 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:
- 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
+ video_description = get_element_by_id("description", webpage)
+ if video_description: video_description = clean_html(video_description.decode('utf8'))
+ else: video_description = ''
# Extract upload date
video_upload_date = u'NA'
@@ -2507,7 +2634,7 @@ class YoutubePlaylistIE(InfoExtractor):
_VALID_URL = r'(?:https?://)?(?:\w+\.)?youtube\.com/(?:(?:course|view_play_list|my_playlists|artist|playlist)\?.*?(p|a|list)=|user/.*?/user/|p/|user/.*?#[pg]/c/)(?:PL)?([0-9A-Za-z-_]+)(?:/.*?/([0-9A-Za-z_-]+))?.*'
_TEMPLATE_URL = 'http://www.youtube.com/%s?%s=%s&page=%s&gl=US&hl=en'
- _VIDEO_INDICATOR = r'/watch\?v=(.+?)&'
+ _VIDEO_INDICATOR_TEMPLATE = r'/watch\?v=(.+?)&list=PL%s&'
_MORE_PAGES_INDICATOR = r'(?m)>\s*Next\s*'
_youtube_ie = None
IE_NAME = u'youtube:playlist'
@@ -2559,7 +2686,7 @@ 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)
@@ -2570,7 +2697,10 @@ 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)
@@ -3265,8 +3395,6 @@ class EscapistIE(InfoExtractor):
self._downloader.to_screen(u'[escapist] %s: Downloading configuration' % showName)
def _real_extract(self, url):
- htmlParser = HTMLParser.HTMLParser()
-
mobj = re.match(self._VALID_URL, url)
if mobj is None:
self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
@@ -3282,11 +3410,11 @@ class EscapistIE(InfoExtractor):
return
descMatch = re.search('