release 2015.02.26.1

This fixes #5066.
Reproducible with
LC_ALL=C youtube-dl "http://www.prosieben.de/tv/germanys-next-topmodel/video/playlist/ganze-folge-episode-2-das-casting-in-muenchen"

AUTHORS

@@ -112,3 +112,4 @@ Frans de Jonge
 Robin de Rooij
 Ryan Schmidt
 Leslie P. Polzer
+Duncan Keall

Makefile

@@ -2,6 +2,7 @@ all: youtube-dl README.md CONTRIBUTING.md README.txt youtube-dl.1 youtube-dl.bas
 
 clean:
 	rm -rf youtube-dl.1.temp.md youtube-dl.1 youtube-dl.bash-completion README.txt MANIFEST build/ dist/ .coverage cover/ youtube-dl.tar.gz youtube-dl.zsh youtube-dl.fish *.dump *.part *.info.json *.mp4 *.flv *.mp3 *.avi CONTRIBUTING.md.tmp youtube-dl youtube-dl.exe
+	find -name "*.pyc" -delete
 
 PREFIX ?= /usr/local
 BINDIR ?= $(PREFIX)/bin

README.md

@@ -139,6 +139,8 @@ which means you can modify it, redistribute it or use it however you like.
                                      dislike_count <? 50 & description" .
     --no-playlist                    If the URL refers to a video and a
                                      playlist, download only the video.
+    --yes-playlist                   If the URL refers to a video and a
+                                     playlist, download the playlist.
     --age-limit YEARS                download only videos suitable for the given
                                      age
     --download-archive FILE          Download only videos not listed in the

docs/supportedsites.md

@@ -17,6 +17,7 @@
  - **AdultSwim**
  - **Aftenposten**
  - **Aftonbladet**
+ - **AirMozilla**
  - **AlJazeera**
  - **Allocine**
  - **AlphaPorno**
@@ -220,6 +221,9 @@
  - **Ku6**
  - **la7.tv**
  - **Laola1Tv**
+ - **Letv**
+ - **LetvPlaylist**
+ - **LetvTv**
  - **lifenews**: LIFE | NEWS
  - **LiveLeak**
  - **livestream**

test/test_subtitles.py

@@ -25,6 +25,7 @@ from youtube_dl.extractor import (
     RaiIE,
     VikiIE,
     ThePlatformIE,
+    RTVEALaCartaIE,
 )
 
 
@@ -305,5 +306,19 @@ class TestThePlatformSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['en']), '97e7670cbae3c4d26ae8bcc7fdd78d4b')
 
 
+class TestRtveSubtitles(BaseTestSubtitles):
+    url = 'http://www.rtve.es/alacarta/videos/los-misterios-de-laura/misterios-laura-capitulo-32-misterio-del-numero-17-2-parte/2428621/'
+    IE = RTVEALaCartaIE
+
+    def test_allsubtitles(self):
+        print('Skipping, only available from Spain')
+        return
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['es']))
+        self.assertEqual(md5(subtitles['es']), '69e70cae2d40574fb7316f31d6eb7fca')
+
+
 if __name__ == '__main__':
     unittest.main()

test/test_utils.py

@@ -85,6 +85,8 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(
             sanitize_filename('New World record at 0:12:34'),
             'New World record at 0_12_34')
+        self.assertEqual(sanitize_filename('--gasdgf'), '_-gasdgf')
+        self.assertEqual(sanitize_filename('--gasdgf', is_id=True), '--gasdgf')
 
         forbidden = '"\0\\/'
         for fc in forbidden:
@@ -244,6 +246,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_duration('2.5 hours'), 9000)
         self.assertEqual(parse_duration('02:03:04'), 7384)
         self.assertEqual(parse_duration('01:02:03:04'), 93784)
+        self.assertEqual(parse_duration('1 hour 3 minutes'), 3780)
 
     def test_fix_xml_ampersands(self):
         self.assertEqual(

youtube_dl/downloader/f4m.py

@@ -11,6 +11,7 @@ from .common import FileDownloader
 from .http import HttpFD
 from ..compat import (
     compat_urlparse,
+    compat_urllib_error,
 )
 from ..utils import (
     struct_pack,
@@ -121,7 +122,8 @@ class FlvReader(io.BytesIO):
 
         self.read_unsigned_int()  # BootstrapinfoVersion
         # Profile,Live,Update,Reserved
-        self.read(1)
+        flags = self.read_unsigned_char()
+        live = flags & 0x20 != 0
         # time scale
         self.read_unsigned_int()
         # CurrentMediaTime
@@ -160,6 +162,7 @@ class FlvReader(io.BytesIO):
         return {
             'segments': segments,
             'fragments': fragments,
+            'live': live,
         }
 
     def read_bootstrap_info(self):
@@ -182,6 +185,10 @@ def build_fragments_list(boot_info):
     for segment, fragments_count in segment_run_table['segment_run']:
         for _ in range(fragments_count):
             res.append((segment, next(fragments_counter)))
+
+    if boot_info['live']:
+        res = res[-2:]
+
     return res
 
 
@@ -246,6 +253,38 @@ class F4mFD(FileDownloader):
             self.report_error('Unsupported DRM')
         return media
 
+    def _get_bootstrap_from_url(self, bootstrap_url):
+        bootstrap = self.ydl.urlopen(bootstrap_url).read()
+        return read_bootstrap_info(bootstrap)
+
+    def _update_live_fragments(self, bootstrap_url, latest_fragment):
+        fragments_list = []
+        retries = 30
+        while (not fragments_list) and (retries > 0):
+            boot_info = self._get_bootstrap_from_url(bootstrap_url)
+            fragments_list = build_fragments_list(boot_info)
+            fragments_list = [f for f in fragments_list if f[1] > latest_fragment]
+            if not fragments_list:
+                # Retry after a while
+                time.sleep(5.0)
+                retries -= 1
+
+        if not fragments_list:
+            self.report_error('Failed to update fragments')
+
+        return fragments_list
+
+    def _parse_bootstrap_node(self, node, base_url):
+        if node.text is None:
+            bootstrap_url = compat_urlparse.urljoin(
+                base_url, node.attrib['url'])
+            boot_info = self._get_bootstrap_from_url(bootstrap_url)
+        else:
+            bootstrap_url = None
+            bootstrap = base64.b64decode(node.text)
+            boot_info = read_bootstrap_info(bootstrap)
+        return (boot_info, bootstrap_url)
+
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
         requested_bitrate = info_dict.get('tbr')
@@ -265,18 +304,13 @@ class F4mFD(FileDownloader):
 
         base_url = compat_urlparse.urljoin(man_url, media.attrib['url'])
         bootstrap_node = doc.find(_add_ns('bootstrapInfo'))
-        if bootstrap_node.text is None:
-            bootstrap_url = compat_urlparse.urljoin(
-                base_url, bootstrap_node.attrib['url'])
-            bootstrap = self.ydl.urlopen(bootstrap_url).read()
-        else:
-            bootstrap = base64.b64decode(bootstrap_node.text)
+        boot_info, bootstrap_url = self._parse_bootstrap_node(bootstrap_node, base_url)
+        live = boot_info['live']
         metadata_node = media.find(_add_ns('metadata'))
         if metadata_node is not None:
             metadata = base64.b64decode(metadata_node.text)
         else:
             metadata = None
-        boot_info = read_bootstrap_info(bootstrap)
 
         fragments_list = build_fragments_list(boot_info)
         if self.params.get('test', False):
@@ -301,7 +335,8 @@ class F4mFD(FileDownloader):
         (dest_stream, tmpfilename) = sanitize_open(tmpfilename, 'wb')
 
         write_flv_header(dest_stream)
-        write_metadata_tag(dest_stream, metadata)
+        if not live:
+            write_metadata_tag(dest_stream, metadata)
 
         # This dict stores the download progress, it's updated by the progress
         # hook
@@ -348,24 +383,45 @@ class F4mFD(FileDownloader):
         http_dl.add_progress_hook(frag_progress_hook)
 
         frags_filenames = []
-        for (seg_i, frag_i) in fragments_list:
+        while fragments_list:
+            seg_i, frag_i = fragments_list.pop(0)
             name = 'Seg%d-Frag%d' % (seg_i, frag_i)
             url = base_url + name
             if akamai_pv:
                 url += '?' + akamai_pv.strip(';')
             frag_filename = '%s-%s' % (tmpfilename, name)
-            success = http_dl.download(frag_filename, {'url': url})
-            if not success:
-                return False
-            with open(frag_filename, 'rb') as down:
-                down_data = down.read()
-                reader = FlvReader(down_data)
-                while True:
-                    _, box_type, box_data = reader.read_box_info()
-                    if box_type == b'mdat':
-                        dest_stream.write(box_data)
-                        break
-            frags_filenames.append(frag_filename)
+            try:
+                success = http_dl.download(frag_filename, {'url': url})
+                if not success:
+                    return False
+                with open(frag_filename, 'rb') as down:
+                    down_data = down.read()
+                    reader = FlvReader(down_data)
+                    while True:
+                        _, box_type, box_data = reader.read_box_info()
+                        if box_type == b'mdat':
+                            dest_stream.write(box_data)
+                            break
+                if live:
+                    os.remove(frag_filename)
+                else:
+                    frags_filenames.append(frag_filename)
+            except (compat_urllib_error.HTTPError, ) as err:
+                if live and (err.code == 404 or err.code == 410):
+                    # We didn't keep up with the live window. Continue
+                    # with the next available fragment.
+                    msg = 'Fragment %d unavailable' % frag_i
+                    self.report_warning(msg)
+                    fragments_list = []
+                else:
+                    raise
+
+            if not fragments_list and live and bootstrap_url:
+                fragments_list = self._update_live_fragments(bootstrap_url, frag_i)
+                total_frags += len(fragments_list)
+                if fragments_list and (fragments_list[0][1] > frag_i + 1):
+                    msg = 'Missed %d fragments' % (fragments_list[0][1] - (frag_i + 1))
+                    self.report_warning(msg)
 
         dest_stream.close()
 

youtube_dl/downloader/rtmp.py

@@ -119,7 +119,9 @@ class RtmpFD(FileDownloader):
         # Download using rtmpdump. rtmpdump returns exit code 2 when
         # the connection was interrumpted and resuming appears to be
         # possible. This is part of rtmpdump's normal usage, AFAIK.
-        basic_args = ['rtmpdump', '--verbose', '-r', url, '-o', tmpfilename]
+        basic_args = [
+            'rtmpdump', '--verbose', '-r', url,
+            '-o', encodeFilename(tmpfilename, True)]
         if player_url is not None:
             basic_args += ['--swfVfy', player_url]
         if page_url is not None:

youtube_dl/extractor/__init__.py

@@ -8,6 +8,7 @@ from .adobetv import AdobeTVIE
 from .adultswim import AdultSwimIE
 from .aftenposten import AftenpostenIE
 from .aftonbladet import AftonbladetIE
+from .airmozilla import AirMozillaIE
 from .aljazeera import AlJazeeraIE
 from .alphaporno import AlphaPornoIE
 from .anitube import AnitubeIE
@@ -237,6 +238,11 @@ from .krasview import KrasViewIE
 from .ku6 import Ku6IE
 from .la7 import LA7IE
 from .laola1tv import Laola1TvIE
+from .letv import (
+    LetvIE,
+    LetvTvIE,
+    LetvPlaylistIE
+)
 from .lifenews import LifeNewsIE
 from .liveleak import LiveLeakIE
 from .livestream import (

youtube_dl/extractor/airmozilla.py

@@ -0,0 +1,74 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    parse_duration,
+    parse_iso8601,
+)
+
+
+class AirMozillaIE(InfoExtractor):
+    _VALID_URL = r'https?://air\.mozilla\.org/(?P<id>[0-9a-z-]+)/?'
+    _TEST = {
+        'url': 'https://air.mozilla.org/privacy-lab-a-meetup-for-privacy-minded-people-in-san-francisco/',
+        'md5': '2e3e7486ba5d180e829d453875b9b8bf',
+        'info_dict': {
+            'id': '6x4q2w',
+            'ext': 'mp4',
+            'title': 'Privacy Lab - a meetup for privacy minded people in San Francisco',
+            'thumbnail': 're:https://\w+\.cloudfront\.net/6x4q2w/poster\.jpg\?t=\d+',
+            'description': 'Brings together privacy professionals and others interested in privacy at for-profits, non-profits, and NGOs in an effort to contribute to the state of the ecosystem...',
+            'timestamp': 1422487800,
+            'upload_date': '20150128',
+            'location': 'SFO Commons',
+            'duration': 3780,
+            'view_count': int,
+            'categories': ['Main'],
+        }
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        video_id = self._html_search_regex(r'//vid.ly/(.*?)/embed', webpage, 'id')
+
+        embed_script = self._download_webpage('https://vid.ly/{0}/embed'.format(video_id), video_id)
+        jwconfig = self._search_regex(r'\svar jwconfig = (\{.*?\});\s', embed_script, 'metadata')
+        metadata = self._parse_json(jwconfig, video_id)
+
+        formats = [{
+            'url': source['file'],
+            'ext': source['type'],
+            'format_id': self._search_regex(r'&format=(.*)$', source['file'], 'video format'),
+            'format': source['label'],
+            'height': int(source['label'].rstrip('p')),
+        } for source in metadata['playlist'][0]['sources']]
+        self._sort_formats(formats)
+
+        view_count = int_or_none(self._html_search_regex(
+            r'Views since archived: ([0-9]+)',
+            webpage, 'view count', fatal=False))
+        timestamp = parse_iso8601(self._html_search_regex(
+            r'<time datetime="(.*?)"', webpage, 'timestamp', fatal=False))
+        duration = parse_duration(self._search_regex(
+            r'Duration:\s*(\d+\s*hours?\s*\d+\s*minutes?)',
+            webpage, 'duration', fatal=False))
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'formats': formats,
+            'url': self._og_search_url(webpage),
+            'display_id': display_id,
+            'thumbnail': metadata['playlist'][0].get('image'),
+            'description': self._og_search_description(webpage),
+            'timestamp': timestamp,
+            'location': self._html_search_regex(r'Location: (.*)', webpage, 'location', default=None),
+            'duration': duration,
+            'view_count': view_count,
+            'categories': re.findall(r'<a href=".*?" class="channel">(.*?)</a>', webpage),
+        }

youtube_dl/extractor/bloomberg.py

@@ -6,7 +6,7 @@ from .common import InfoExtractor
 
 
 class BloombergIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.bloomberg\.com/video/(?P<name>.+?)\.html'
+    _VALID_URL = r'https?://www\.bloomberg\.com/video/(?P<id>.+?)\.html'
 
     _TEST = {
         'url': 'http://www.bloomberg.com/video/shah-s-presentation-on-foreign-exchange-strategies-qurhIVlJSB6hzkVi229d8g.html',
@@ -20,9 +20,9 @@ class BloombergIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        name = mobj.group('name')
+        name = self._match_id(url)
         webpage = self._download_webpage(url, name)
+
         f4m_url = self._search_regex(
             r'<source src="(https?://[^"]+\.f4m.*?)"', webpage,
             'f4m url')

youtube_dl/extractor/common.py

@@ -921,39 +921,57 @@ class InfoExtractor(object):
 
         formats = []
         rtmp_count = 0
-        for video in smil.findall('./body/switch/video'):
-            src = video.get('src')
-            if not src:
-                continue
-            bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
-            width = int_or_none(video.get('width'))
-            height = int_or_none(video.get('height'))
-            proto = video.get('proto')
-            if not proto:
-                if base:
-                    if base.startswith('rtmp'):
-                        proto = 'rtmp'
-                    elif base.startswith('http'):
-                        proto = 'http'
-            ext = video.get('ext')
-            if proto == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(src, video_id, ext))
-            elif proto == 'rtmp':
-                rtmp_count += 1
-                streamer = video.get('streamer') or base
-                formats.append({
-                    'url': streamer,
-                    'play_path': src,
-                    'ext': 'flv',
-                    'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate),
-                    'tbr': bitrate,
-                    'width': width,
-                    'height': height,
-                })
+        if smil.findall('./body/seq/video'):
+            video = smil.findall('./body/seq/video')[0]
+            fmts, rtmp_count = self._parse_smil_video(video, video_id, base, rtmp_count)
+            formats.extend(fmts)
+        else:
+            for video in smil.findall('./body/switch/video'):
+                fmts, rtmp_count = self._parse_smil_video(video, video_id, base, rtmp_count)
+                formats.extend(fmts)
+
         self._sort_formats(formats)
 
         return formats
 
+    def _parse_smil_video(self, video, video_id, base, rtmp_count):
+        src = video.get('src')
+        if not src:
+            return ([], rtmp_count)
+        bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
+        width = int_or_none(video.get('width'))
+        height = int_or_none(video.get('height'))
+        proto = video.get('proto')
+        if not proto:
+            if base:
+                if base.startswith('rtmp'):
+                    proto = 'rtmp'
+                elif base.startswith('http'):
+                    proto = 'http'
+        ext = video.get('ext')
+        if proto == 'm3u8':
+            return (self._extract_m3u8_formats(src, video_id, ext), rtmp_count)
+        elif proto == 'rtmp':
+            rtmp_count += 1
+            streamer = video.get('streamer') or base
+            return ([{
+                'url': streamer,
+                'play_path': src,
+                'ext': 'flv',
+                'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate),
+                'tbr': bitrate,
+                'width': width,
+                'height': height,
+            }], rtmp_count)
+        elif proto.startswith('http'):
+            return ([{
+                'url': base + src,
+                'ext': ext or 'flv',
+                'tbr': bitrate,
+                'width': width,
+                'height': height,
+            }], rtmp_count)
+
     def _live_title(self, name):
         """ Generate the title for a live video """
         now = datetime.datetime.now()

youtube_dl/extractor/eporner.py

@@ -35,10 +35,7 @@ class EpornerIE(InfoExtractor):
         title = self._html_search_regex(
             r'<title>(.*?) - EPORNER', webpage, 'title')
 
-        redirect_code = self._html_search_regex(
-            r'<script type="text/javascript" src="/config5/%s/([a-f\d]+)/">' % video_id,
-            webpage, 'redirect_code')
-        redirect_url = 'http://www.eporner.com/config5/%s/%s' % (video_id, redirect_code)
+        redirect_url = 'http://www.eporner.com/config5/%s' % video_id
         player_code = self._download_webpage(
             redirect_url, display_id, note='Downloading player config')
 
@@ -69,5 +66,5 @@ class EpornerIE(InfoExtractor):
             'duration': duration,
             'view_count': view_count,
             'formats': formats,
-            'age_limit': self._rta_search(webpage),
+            'age_limit': 18,
         }

youtube_dl/extractor/escapist.py

@@ -42,9 +42,17 @@ class EscapistIE(InfoExtractor):
         title = raw_title.partition(' : ')[2]
 
         config_url = compat_urllib_parse.unquote(self._html_search_regex(
-            r'<param\s+name="flashvars"\s+value="config=([^"&]+)', webpage, 'config URL'))
+            r'''(?x)
+            (?:
+                <param\s+name="flashvars".*?\s+value="config=|
+                flashvars=&quot;config=
+            )
+            (https?://[^"&]+)
+            ''',
+            webpage, 'config URL'))
 
         formats = []
+        ad_formats = []
 
         def _add_format(name, cfgurl, quality):
             config = self._download_json(
@@ -54,14 +62,19 @@ class EscapistIE(InfoExtractor):
                 transform_source=js_to_json)
 
             playlist = config['playlist']
-            video_url = next(
-                p['url'] for p in playlist
-                if p.get('eventCategory') == 'Video')
-            formats.append({
-                'url': video_url,
-                'format_id': name,
-                'quality': quality,
-            })
+            for p in playlist:
+                if p.get('eventCategory') == 'Video':
+                    ar = formats
+                elif p.get('eventCategory') == 'Video Postroll':
+                    ar = ad_formats
+                else:
+                    continue
+
+                ar.append({
+                    'url': p['url'],
+                    'format_id': name,
+                    'quality': quality,
+                })
 
         _add_format('normal', config_url, quality=0)
         hq_url = (config_url +
@@ -70,10 +83,9 @@ class EscapistIE(InfoExtractor):
             _add_format('hq', hq_url, quality=1)
         except ExtractorError:
             pass  # That's fine, we'll just use normal quality
-
         self._sort_formats(formats)
 
-        return {
+        res = {
             'id': video_id,
             'formats': formats,
             'uploader': uploader,
@@ -82,3 +94,19 @@ class EscapistIE(InfoExtractor):
             'thumbnail': self._og_search_thumbnail(webpage),
             'description': description,
         }
+
+        if self._downloader.params.get('include_ads') and ad_formats:
+            self._sort_formats(ad_formats)
+            ad_res = {
+                'id': '%s-ad' % video_id,
+                'title': '%s (Postroll)' % title,
+                'formats': ad_formats,
+            }
+            return {
+                '_type': 'playlist',
+                'entries': [res, ad_res],
+                'title': title,
+                'id': video_id,
+            }
+
+        return res

youtube_dl/extractor/facebook.py

@@ -126,11 +126,17 @@ class FacebookIE(InfoExtractor):
         params_raw = compat_urllib_parse.unquote(data['params'])
         params = json.loads(params_raw)
         video_data = params['video_data'][0]
-        video_url = video_data.get('hd_src')
-        if not video_url:
-            video_url = video_data['sd_src']
-        if not video_url:
-            raise ExtractorError('Cannot find video URL')
+
+        formats = []
+        for quality in ['sd', 'hd']:
+            src = video_data.get('%s_src' % quality)
+            if src is not None:
+                formats.append({
+                    'format_id': quality,
+                    'url': src,
+                })
+        if not formats:
+            raise ExtractorError('Cannot find video formats')
 
         video_title = self._html_search_regex(
             r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title',
@@ -146,7 +152,7 @@ class FacebookIE(InfoExtractor):
         return {
             'id': video_id,
             'title': video_title,
-            'url': video_url,
+            'formats': formats,
             'duration': int_or_none(video_data.get('video_duration')),
             'thumbnail': video_data.get('thumbnail_src'),
         }

youtube_dl/extractor/generic.py

@@ -1208,7 +1208,9 @@ class GenericIE(InfoExtractor):
             return entries[0]
         else:
             for num, e in enumerate(entries, start=1):
-                e['title'] = '%s (%d)' % (e['title'], num)
+                # 'url' results don't have a title
+                if e.get('title') is not None:
+                    e['title'] = '%s (%d)' % (e['title'], num)
             return {
                 '_type': 'playlist',
                 'entries': entries,

youtube_dl/extractor/laola1tv.py

@@ -1,23 +1,26 @@
+# -*- coding: utf-8 -*-
 from __future__ import unicode_literals
 
 import random
 import re
 
 from .common import InfoExtractor
-from ..utils import ExtractorError
+from ..utils import (
+    ExtractorError,
+    xpath_text,
+)
 
 
 class Laola1TvIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?laola1\.tv/(?P<lang>[a-z]+)-(?P<portal>[a-z]+)/.*?/(?P<id>[0-9]+)\.html'
     _TEST = {
-        'url': 'http://www.laola1.tv/de-de/live/bwf-bitburger-open-grand-prix-gold-court-1/250019.html',
+        'url': 'http://www.laola1.tv/de-de/video/straubing-tigers-koelner-haie/227883.html',
         'info_dict': {
-            'id': '250019',
+            'id': '227883',
             'ext': 'mp4',
-            'title': 'Bitburger Open Grand Prix Gold - Court 1',
-            'categories': ['Badminton'],
-            'uploader': 'BWF - Badminton World Federation',
-            'is_live': True,
+            'title': 'Straubing Tigers - Kölner Haie',
+            'categories': ['Eishockey'],
+            'is_live': False,
         },
         'params': {
             'skip_download': True,
@@ -43,15 +46,26 @@ class Laola1TvIE(InfoExtractor):
             r'flashvars\.([_a-zA-Z0-9]+)\s*=\s*"([^"]*)";', iframe)
         flashvars = dict((m[0], m[1]) for m in flashvars_m)
 
+        partner_id = self._search_regex(
+            r'partnerid\s*:\s*"([^"]+)"', iframe, 'partner id')
+
         xml_url = ('http://www.laola1.tv/server/hd_video.php?' +
-                   'play=%s&partner=1&portal=%s&v5ident=&lang=%s' % (
-                       video_id, portal, lang))
+                   'play=%s&partner=%s&portal=%s&v5ident=&lang=%s' % (
+                       video_id, partner_id, portal, lang))
         hd_doc = self._download_xml(xml_url, video_id)
 
-        title = hd_doc.find('.//video/title').text
-        flash_url = hd_doc.find('.//video/url').text
-        categories = hd_doc.find('.//video/meta_sports').text.split(',')
-        uploader = hd_doc.find('.//video/meta_organistation').text
+        title = xpath_text(hd_doc, './/video/title', fatal=True)
+        flash_url = xpath_text(hd_doc, './/video/url', fatal=True)
+        uploader = xpath_text(hd_doc, './/video/meta_organistation')
+
+        is_live = xpath_text(hd_doc, './/video/islive') == 'true'
+        if is_live:
+            raise ExtractorError(
+                'Live streams are not supported by the f4m downloader.')
+
+        categories = xpath_text(hd_doc, './/video/meta_sports')
+        if categories:
+            categories = categories.split(',')
 
         ident = random.randint(10000000, 99999999)
         token_url = '%s&ident=%s&klub=0&unikey=0&timestamp=%s&auth=%s' % (
@@ -60,15 +74,16 @@ class Laola1TvIE(InfoExtractor):
         token_doc = self._download_xml(
             token_url, video_id, note='Downloading token')
         token_attrib = token_doc.find('.//token').attrib
-        if token_attrib.get('auth') == 'blocked':
-            raise ExtractorError('Token error: ' % token_attrib.get('comment'))
+        if token_attrib.get('auth') in ('blocked', 'restricted'):
+            raise ExtractorError(
+                'Token error: %s' % token_attrib.get('comment'), expected=True)
 
         video_url = '%s?hdnea=%s&hdcore=3.2.0' % (
             token_attrib['url'], token_attrib['auth'])
 
         return {
             'id': video_id,
-            'is_live': True,
+            'is_live': is_live,
             'title': title,
             'url': video_url,
             'uploader': uploader,

youtube_dl/extractor/letv.py

@@ -0,0 +1,190 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import datetime
+import re
+import time
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urlparse,
+    compat_urllib_parse,
+)
+from ..utils import (
+    determine_ext,
+    ExtractorError,
+    parse_iso8601,
+)
+
+
+class LetvIE(InfoExtractor):
+    _VALID_URL = r'http://www\.letv\.com/ptv/vplay/(?P<id>\d+).html'
+
+    _TESTS = [{
+        'url': 'http://www.letv.com/ptv/vplay/22005890.html',
+        'md5': 'cab23bd68d5a8db9be31c9a222c1e8df',
+        'info_dict': {
+            'id': '22005890',
+            'ext': 'mp4',
+            'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
+            'timestamp': 1424747397,
+            'upload_date': '20150224',
+            'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
+        }
+    }, {
+        'url': 'http://www.letv.com/ptv/vplay/1415246.html',
+        'info_dict': {
+            'id': '1415246',
+            'ext': 'mp4',
+            'title': '美人天下01',
+            'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda',
+        },
+        'expected_warnings': [
+            'publish time'
+        ]
+    }]
+    # http://www.letv.com/ptv/vplay/1118082.html
+    # This video is available only in Mainland China
+
+    @staticmethod
+    def urshift(val, n):
+        return val >> n if val >= 0 else (val + 0x100000000) >> n
+
+    # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
+    def ror(self, param1, param2):
+        _loc3_ = 0
+        while _loc3_ < param2:
+            param1 = self.urshift(param1, 1) + ((param1 & 1) << 31)
+            _loc3_ += 1
+        return param1
+
+    def calc_time_key(self, param1):
+        _loc2_ = 773625421
+        _loc3_ = self.ror(param1, _loc2_ % 13)
+        _loc3_ = _loc3_ ^ _loc2_
+        _loc3_ = self.ror(_loc3_, _loc2_ % 17)
+        return _loc3_
+
+    def _real_extract(self, url):
+        media_id = self._match_id(url)
+        page = self._download_webpage(url, media_id)
+        params = {
+            'id': media_id,
+            'platid': 1,
+            'splatid': 101,
+            'format': 1,
+            'tkey': self.calc_time_key(int(time.time())),
+            'domain': 'www.letv.com'
+        }
+        play_json = self._download_json(
+            'http://api.letv.com/mms/out/video/playJson?' + compat_urllib_parse.urlencode(params),
+            media_id, 'playJson data')
+
+        # Check for errors
+        playstatus = play_json['playstatus']
+        if playstatus['status'] == 0:
+            flag = playstatus['flag']
+            if flag == 1:
+                msg = 'Country %s auth error' % playstatus['country']
+            else:
+                msg = 'Generic error. flag = %d' % flag
+            raise ExtractorError(msg, expected=True)
+
+        playurl = play_json['playurl']
+
+        formats = ['350', '1000', '1300', '720p', '1080p']
+        dispatch = playurl['dispatch']
+
+        urls = []
+        for format_id in formats:
+            if format_id in dispatch:
+                media_url = playurl['domain'][0] + dispatch[format_id][0]
+
+                # Mimic what flvxz.com do
+                url_parts = list(compat_urlparse.urlparse(media_url))
+                qs = dict(compat_urlparse.parse_qs(url_parts[4]))
+                qs.update({
+                    'platid': '14',
+                    'splatid': '1401',
+                    'tss': 'no',
+                    'retry': 1
+                })
+                url_parts[4] = compat_urllib_parse.urlencode(qs)
+                media_url = compat_urlparse.urlunparse(url_parts)
+
+                url_info_dict = {
+                    'url': media_url,
+                    'ext': determine_ext(dispatch[format_id][1])
+                }
+
+                if format_id[-1:] == 'p':
+                    url_info_dict['height'] = format_id[:-1]
+
+                urls.append(url_info_dict)
+
+        publish_time = parse_iso8601(self._html_search_regex(
+            r'发布时间&nbsp;([^<>]+) ', page, 'publish time', fatal=False),
+            delimiter=' ', timezone=datetime.timedelta(hours=8))
+        description = self._html_search_meta('description', page, fatal=False)
+
+        return {
+            'id': media_id,
+            'formats': urls,
+            'title': playurl['title'],
+            'thumbnail': playurl['pic'],
+            'description': description,
+            'timestamp': publish_time,
+        }
+
+
+class LetvTvIE(InfoExtractor):
+    _VALID_URL = r'http://www.letv.com/tv/(?P<id>\d+).html'
+    _TESTS = [{
+        'url': 'http://www.letv.com/tv/46177.html',
+        'info_dict': {
+            'id': '46177',
+            'title': '美人天下',
+            'description': 'md5:395666ff41b44080396e59570dbac01c'
+        },
+        'playlist_count': 35
+    }]
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+        page = self._download_webpage(url, playlist_id)
+
+        media_urls = list(set(re.findall(
+            r'http://www.letv.com/ptv/vplay/\d+.html', page)))
+        entries = [self.url_result(media_url, ie='Letv')
+                   for media_url in media_urls]
+
+        title = self._html_search_meta('keywords', page,
+                                       fatal=False).split('，')[0]
+        description = self._html_search_meta('description', page, fatal=False)
+
+        return self.playlist_result(entries, playlist_id, playlist_title=title,
+                                    playlist_description=description)
+
+
+class LetvPlaylistIE(LetvTvIE):
+    _VALID_URL = r'http://tv.letv.com/[a-z]+/(?P<id>[a-z]+)/index.s?html'
+    _TESTS = [{
+        'url': 'http://tv.letv.com/izt/wuzetian/index.html',
+        'info_dict': {
+            'id': 'wuzetian',
+            'title': '武媚娘传奇',
+            'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
+        },
+        # This playlist contains some extra videos other than the drama itself
+        'playlist_mincount': 96
+    }, {
+        'url': 'http://tv.letv.com/pzt/lswjzzjc/index.shtml',
+        'info_dict': {
+            'id': 'lswjzzjc',
+            # The title should be "劲舞青春", but I can't find a simple way to
+            # determine the playlist title
+            'title': '乐视午间自制剧场',
+            'description': 'md5:b1eef244f45589a7b5b1af9ff25a4489'
+        },
+        'playlist_mincount': 7
+    }]

youtube_dl/extractor/mitele.py

@@ -18,7 +18,7 @@ class MiTeleIE(InfoExtractor):
     IE_NAME = 'mitele.es'
     _VALID_URL = r'http://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
         'md5': '6a75fe9d0d3275bead0cb683c616fddb',
         'info_dict': {
@@ -29,7 +29,7 @@ class MiTeleIE(InfoExtractor):
             'display_id': 'programa-144',
             'duration': 2913,
         },
-    }
+    }]
 
     def _real_extract(self, url):
         episode = self._match_id(url)

youtube_dl/extractor/nrk.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
     float_or_none,
@@ -158,7 +159,9 @@ class NRKTVIE(InfoExtractor):
     def _get_subtitles(self, subtitlesurl, video_id, baseurl):
         url = "%s%s" % (baseurl, subtitlesurl)
         self._debug_print('%s: Subtitle url: %s' % (video_id, url))
-        captions = self._download_xml(url, video_id, 'Downloading subtitles')
+        captions = self._download_xml(
+            url, video_id, 'Downloading subtitles',
+            transform_source=lambda s: s.replace(r'<br />', '\r\n'))
         lang = captions.get('lang', 'no')
         ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/ns/ttml}'))
         srt = ''
@@ -167,8 +170,7 @@ class NRKTVIE(InfoExtractor):
             duration = parse_duration(p.get('dur'))
             starttime = self._seconds2str(begin)
             endtime = self._seconds2str(begin + duration)
-            text = '\n'.join(p.itertext())
-            srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), starttime, endtime, text)
+            srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (compat_str(pos), starttime, endtime, p.text)
         return {lang: [
             {'ext': 'ttml', 'url': url},
             {'ext': 'srt', 'data': srt},

youtube_dl/extractor/rtlnow.py

@@ -146,7 +146,7 @@ class RTLnowIE(InfoExtractor):
                 mobj = re.search(r'.*/(?P<hoster>[^/]+)/videos/(?P<play_path>.+)\.f4m', filename.text)
                 if mobj:
                     fmt = {
-                        'url': 'rtmpe://fmspay-fra2.rtl.de/' + mobj.group('hoster'),
+                        'url': 'rtmpe://fms.rtl.de/' + mobj.group('hoster'),
                         'play_path': 'mp4:' + mobj.group('play_path'),
                         'page_url': url,
                         'player_url': video_page_url + 'includes/vodplayer.swf',

youtube_dl/extractor/rtve.py

@@ -102,14 +102,27 @@ class RTVEALaCartaIE(InfoExtractor):
             video_url = compat_urlparse.urljoin(
                 'http://mvod1.akcdn.rtve.es/', video_path)
 
+        subtitles = None
+        if info.get('sbtFile') is not None:
+            subtitles = self.extract_subtitles(video_id, info['sbtFile'])
+
         return {
             'id': video_id,
             'title': info['title'],
             'url': video_url,
             'thumbnail': info.get('image'),
             'page_url': url,
+            'subtitles': subtitles,
         }
 
+    def _get_subtitles(self, video_id, sub_file):
+        subs = self._download_json(
+            sub_file + '.json', video_id,
+            'Downloading subtitles info')['page']['items']
+        return dict(
+            (s['lang'], [{'ext': 'vtt', 'url': s['src']}])
+            for s in subs)
+
 
 class RTVELiveIE(InfoExtractor):
     IE_NAME = 'rtve.es:live'

youtube_dl/extractor/telecinco.py

@@ -6,9 +6,9 @@ from .mitele import MiTeleIE
 
 class TelecincoIE(MiTeleIE):
     IE_NAME = 'telecinco.es'
-    _VALID_URL = r'https?://www\.telecinco\.es/[^/]+/[^/]+/[^/]+/(?P<id>.*?)\.html'
+    _VALID_URL = r'https?://www\.telecinco\.es/[^/]+/[^/]+/(?:[^/]+/)?(?P<id>.*?)\.html'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html',
         'info_dict': {
             'id': 'MDSVID20141015_0058',
@@ -16,4 +16,7 @@ class TelecincoIE(MiTeleIE):
             'title': 'Con Martín Berasategui, hacer un bacalao al ...',
             'duration': 662,
         },
-    }
+    }, {
+        'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html',
+        'only_matching': True,
+    }]

youtube_dl/extractor/wdr.py

@@ -28,6 +28,7 @@ class WDRIE(InfoExtractor):
                 'title': 'Servicezeit',
                 'description': 'md5:c8f43e5e815eeb54d0b96df2fba906cb',
                 'upload_date': '20140310',
+                'is_live': False
             },
             'params': {
                 'skip_download': True,
@@ -41,6 +42,7 @@ class WDRIE(InfoExtractor):
                 'title': 'Marga Spiegel ist tot',
                 'description': 'md5:2309992a6716c347891c045be50992e4',
                 'upload_date': '20140311',
+                'is_live': False
             },
             'params': {
                 'skip_download': True,
@@ -55,6 +57,7 @@ class WDRIE(InfoExtractor):
                 'title': 'Erlebte Geschichten: Marga Spiegel (29.11.2009)',
                 'description': 'md5:2309992a6716c347891c045be50992e4',
                 'upload_date': '20091129',
+                'is_live': False
             },
         },
         {
@@ -66,6 +69,7 @@ class WDRIE(InfoExtractor):
                 'title': 'Flavia Coelho: Amar é Amar',
                 'description': 'md5:7b29e97e10dfb6e265238b32fa35b23a',
                 'upload_date': '20140717',
+                'is_live': False
             },
         },
         {
@@ -74,6 +78,20 @@ class WDRIE(InfoExtractor):
             'info_dict': {
                 'id': 'mediathek/video/sendungen/quarks_und_co/filterseite-quarks-und-co100',
             }
+        },
+        {
+            'url': 'http://www1.wdr.de/mediathek/video/livestream/index.html',
+            'info_dict': {
+                'id': 'mdb-103364',
+                'title': 're:^WDR Fernsehen [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+                'description': 'md5:ae2ff888510623bf8d4b115f95a9b7c9',
+                'ext': 'flv',
+                'upload_date': '20150212',
+                'is_live': True
+            },
+            'params': {
+                'skip_download': True,
+            },
         }
     ]
 
@@ -119,6 +137,10 @@ class WDRIE(InfoExtractor):
         video_url = flashvars['dslSrc'][0]
         title = flashvars['trackerClipTitle'][0]
         thumbnail = flashvars['startPicture'][0] if 'startPicture' in flashvars else None
+        is_live = flashvars.get('isLive', ['0'])[0] == '1'
+
+        if is_live:
+            title = self._live_title(title)
 
         if 'trackerClipAirTime' in flashvars:
             upload_date = flashvars['trackerClipAirTime'][0]
@@ -131,6 +153,13 @@ class WDRIE(InfoExtractor):
         if video_url.endswith('.f4m'):
             video_url += '?hdcore=3.2.0&plugin=aasp-3.2.0.77.18'
             ext = 'flv'
+        elif video_url.endswith('.smil'):
+            fmt = self._extract_smil_formats(video_url, page_id)[0]
+            video_url = fmt['url']
+            sep = '&' if '?' in video_url else '?'
+            video_url += sep
+            video_url += 'hdcore=3.3.0&plugin=aasp-3.3.0.99.43'
+            ext = fmt['ext']
         else:
             ext = determine_ext(video_url)
 
@@ -144,6 +173,7 @@ class WDRIE(InfoExtractor):
             'description': description,
             'thumbnail': thumbnail,
             'upload_date': upload_date,
+            'is_live': is_live
         }
 
 

youtube_dl/extractor/youtube.py

@@ -1153,13 +1153,13 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
                         |  p/
                         )
                         (
-                            (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
+                            (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
                             # Top tracks, they can also include dots
                             |(?:MC)[\w\.]*
                         )
                         .*
                      |
-                        ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
+                        ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
                      )"""
     _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
     _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)'
@@ -1244,7 +1244,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
             for vid_id in ids]
 
     def _extract_mix(self, playlist_id):
-        # The mixes are generated from a a single video
+        # The mixes are generated from a single video
         # the id of the playlist is just 'RD' + video_id
         url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
         webpage = self._download_webpage(
@@ -1280,7 +1280,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
             else:
                 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
 
-        if playlist_id.startswith('RD'):
+        if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
             # Mixes require a custom extraction process
             return self._extract_mix(playlist_id)
 

youtube_dl/options.py

@@ -272,6 +272,10 @@ def parseOpts(overrideArguments=None):
         '--no-playlist',
         action='store_true', dest='noplaylist', default=False,
         help='If the URL refers to a video and a playlist, download only the video.')
+    selection.add_option(
+        '--yes-playlist',
+        action='store_false', dest='noplaylist', default=False,
+        help='If the URL refers to a video and a playlist, download the playlist.')
     selection.add_option(
         '--age-limit',
         metavar='YEARS', dest='age_limit', default=None, type=int,

youtube_dl/utils.py

@@ -54,7 +54,7 @@ from .compat import (
 compiled_regex_type = type(re.compile(''))
 
 std_headers = {
-    'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0 (Chrome)',
+    'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/20.0 (Chrome)',
     'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
     'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
     'Accept-Encoding': 'gzip, deflate',
@@ -304,6 +304,8 @@ def sanitize_filename(s, restricted=False, is_id=False):
         # Common case of "Foreign band name - English song title"
         if restricted and result.startswith('-_'):
             result = result[2:]
+        if result.startswith('-'):
+            result = '_' + result[len('-'):]
         if not result:
             result = '_'
     return result
@@ -1288,6 +1290,7 @@ def parse_duration(s):
             (?P<only_mins>[0-9.]+)\s*(?:mins?|minutes?)\s*|
             (?P<only_hours>[0-9.]+)\s*(?:hours?)|
 
+            \s*(?P<hours_reversed>[0-9]+)\s*(?:[:h]|hours?)\s*(?P<mins_reversed>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*|
             (?:
                 (?:
                     (?:(?P<days>[0-9]+)\s*(?:[:d]|days?)\s*)?
@@ -1306,10 +1309,14 @@ def parse_duration(s):
         return float_or_none(m.group('only_hours'), invscale=60 * 60)
     if m.group('secs'):
         res += int(m.group('secs'))
+    if m.group('mins_reversed'):
+        res += int(m.group('mins_reversed')) * 60
     if m.group('mins'):
         res += int(m.group('mins')) * 60
     if m.group('hours'):
         res += int(m.group('hours')) * 60 * 60
+    if m.group('hours_reversed'):
+        res += int(m.group('hours_reversed')) * 60 * 60
     if m.group('days'):
         res += int(m.group('days')) * 24 * 60 * 60
     if m.group('ms'):

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2015.02.23.1'
+__version__ = '2015.02.26.1'