release 2013.04.03

Fix crash when subtitles are not found

README.md

@@ -18,7 +18,7 @@ which means you can modify it, redistribute it or use it however you like.
     --version                print program version and exit
     -U, --update             update this program to latest version
     -i, --ignore-errors      continue on download errors
-    -r, --rate-limit LIMIT   download rate limit (e.g. 50k or 44.6m)
+    -r, --rate-limit LIMIT   maximum download rate (e.g. 50k or 44.6m)
     -R, --retries RETRIES    number of retries (default is 10)
     --buffer-size SIZE       size of download buffer (e.g. 1024 or 16k) (default
                              is 1024)
@@ -97,10 +97,16 @@ which means you can modify it, redistribute it or use it however you like.
                              requested
     --max-quality FORMAT     highest quality format to download
     -F, --list-formats       list all available formats (currently youtube only)
-    --write-srt              write video closed captions to a .srt file
+    --write-sub              write subtitle file (currently youtube only)
+    --only-sub               downloads only the subtitles (no video)
+    --all-subs               downloads all the available subtitles of the video
                              (currently youtube only)
-    --srt-lang LANG          language of the closed captions to download
+    --list-subs              lists all available subtitles for the video
-                             (optional) use IETF language tags like 'en'
+                             (currently youtube only)
+    --sub-format LANG        subtitle format [srt/sbv] (default=srt) (currently
+                             youtube only)
+    --sub-lang LANG          language of the subtitles to download (optional)
+                             use IETF language tags like 'en'
 
 ## Authentication Options:
     -u, --username USERNAME  account username

devscripts/gh-pages/update-feed.py

@@ -0,0 +1,57 @@
+#!/usr/bin/env python3
+
+import datetime
+
+import textwrap
+
+import json
+
+atom_template=textwrap.dedent("""\
+								<?xml version='1.0' encoding='utf-8'?>
+								<atom:feed xmlns:atom="http://www.w3.org/2005/Atom">
+									<atom:title>youtube-dl releases</atom:title>
+									<atom:id>youtube-dl-updates-feed</atom:id>
+									<atom:updated>@TIMESTAMP@</atom:updated>
+									@ENTRIES@
+								</atom:feed>""")
+
+entry_template=textwrap.dedent("""
+								<atom:entry>
+									<atom:id>youtube-dl-@VERSION@</atom:id>
+									<atom:title>New version @VERSION@</atom:title>
+									<atom:link href="http://rg3.github.com/youtube-dl" />
+									<atom:content type="xhtml">
+										<div xmlns="http://www.w3.org/1999/xhtml">
+											Downloads available at <a href="http://youtube-dl.org/downloads/@VERSION@/">http://youtube-dl.org/downloads/@VERSION@/</a>
+										</div>
+									</atom:content>
+									<atom:author>
+										<atom:name>The youtube-dl maintainers</atom:name>
+									</atom:author>
+									<atom:updated>@TIMESTAMP@</atom:updated>
+								</atom:entry>
+								""")
+
+now = datetime.datetime.now()
+now_iso = now.isoformat()
+
+atom_template = atom_template.replace('@TIMESTAMP@',now_iso)
+
+entries=[]
+
+versions_info = json.load(open('update/versions.json'))
+versions = list(versions_info['versions'].keys())
+versions.sort()
+
+for v in versions:
+	entry = entry_template.replace('@TIMESTAMP@',v.replace('.','-'))
+	entry = entry.replace('@VERSION@',v)
+	entries.append(entry)
+
+entries_str = textwrap.indent(''.join(entries), '\t')
+atom_template = atom_template.replace('@ENTRIES@', entries_str)
+
+with open('update/releases.atom','w',encoding='utf-8') as atom_file:
+	atom_file.write(atom_template)
+
+

devscripts/release.sh

@@ -22,7 +22,7 @@ if [ ! -f "updates_key.pem" ]; then echo 'ERROR: updates_key.pem missing'; exit
 
 /bin/echo -e "\n### First of all, testing..."
 make cleanall
-nosetests --with-coverage --cover-package=youtube_dl --cover-html test || exit 1
+nosetests --with-coverage --cover-package=youtube_dl --cover-html test --stop || exit 1
 
 /bin/echo -e "\n### Changing version in version.py..."
 sed -i "s/__version__ = '.*'/__version__ = '$version'/" youtube_dl/version.py
@@ -69,6 +69,7 @@ ROOT=$(pwd)
     ORIGIN_URL=$(git config --get remote.origin.url)
     cd build/gh-pages
     "$ROOT/devscripts/gh-pages/add-version.py" $version
+    "$ROOT/devscripts/gh-pages/update-feed.py"
     "$ROOT/devscripts/gh-pages/sign-versions.py" < "$ROOT/updates_key.pem"
     "$ROOT/devscripts/gh-pages/generate-download.py"
     "$ROOT/devscripts/gh-pages/update-copyright.py"

test/parameters.json

@@ -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
 }

test/test_all_urls.py

@@ -11,12 +11,18 @@ from youtube_dl.InfoExtractors import YoutubeIE, YoutubePlaylistIE
 
 class TestAllURLsMatching(unittest.TestCase):
     def test_youtube_playlist_matching(self):
-        self.assertTrue(YoutubePlaylistIE().suitable(u'ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8'))
+        self.assertTrue(YoutubePlaylistIE.suitable(u'ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8'))
-        self.assertTrue(YoutubePlaylistIE().suitable(u'PL63F0C78739B09958'))
+        self.assertTrue(YoutubePlaylistIE.suitable(u'UUBABnxM4Ar9ten8Mdjj1j0Q')) #585
-        self.assertFalse(YoutubePlaylistIE().suitable(u'PLtS2H6bU1M'))
+        self.assertTrue(YoutubePlaylistIE.suitable(u'PL63F0C78739B09958'))
+        self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q'))
+        self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8'))
+        self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC'))
+        self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012')) #668
+        self.assertFalse(YoutubePlaylistIE.suitable(u'PLtS2H6bU1M'))
 
     def test_youtube_matching(self):
-        self.assertTrue(YoutubeIE().suitable(u'PLtS2H6bU1M'))
+        self.assertTrue(YoutubeIE.suitable(u'PLtS2H6bU1M'))
+        self.assertFalse(YoutubeIE.suitable(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012')) #668
 
     def test_youtube_extract(self):
         self.assertEqual(YoutubeIE()._extract_id('http://www.youtube.com/watch?&v=BaW_jenozKc'), 'BaW_jenozKc')

test/test_download.py

@@ -20,6 +20,8 @@ from youtube_dl.utils import *
 DEF_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'tests.json')
 PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
 
+RETRIES = 3
+
 # General configuration (from __init__, not very elegant...)
 jar = compat_cookiejar.CookieJar()
 cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
@@ -79,9 +81,8 @@ def generator(test_case):
         params.update(test_case.get('params', {}))
 
         fd = FileDownloader(params)
-        fd.add_info_extractor(ie())
+        for ie in youtube_dl.InfoExtractors.gen_extractors():
-        for ien in test_case.get('add_ie', []):
+            fd.add_info_extractor(ie)
-            fd.add_info_extractor(getattr(youtube_dl.InfoExtractors, ien + 'IE')())
         finished_hook_called = set()
         def _hook(status):
             if status['status'] == 'finished':
@@ -93,8 +94,20 @@ def generator(test_case):
             _try_rm(tc['file'])
             _try_rm(tc['file'] + '.part')
             _try_rm(tc['file'] + '.info.json')
+        try:
+            for retry in range(1, RETRIES + 1):
                 try:
                     fd.download([test_case['url']])
+                except (DownloadError, ExtractorError) as err:
+                    if retry == RETRIES: raise
+
+                    # Check if the exception is not a network related one
+                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
+                        raise
+
+                    print('Retrying: {0} failed tries\n\n##########\n\n'.format(retry))
+                else:
+                    break
 
             for tc in test_cases:
                 if not test_case.get('params', {}).get('skip_download', False):

test/test_youtube_lists.py

@@ -8,7 +8,7 @@ import json
 import os
 sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from youtube_dl.InfoExtractors import YoutubeUserIE,YoutubePlaylistIE
+from youtube_dl.InfoExtractors import YoutubeUserIE, YoutubePlaylistIE, YoutubeIE
 from youtube_dl.utils import *
 
 PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
@@ -35,39 +35,51 @@ class FakeDownloader(object):
 
 class TestYoutubeLists(unittest.TestCase):
     def test_youtube_playlist(self):
-        DL = FakeDownloader()
+        dl = FakeDownloader()
-        IE = YoutubePlaylistIE(DL)
+        ie = YoutubePlaylistIE(dl)
-        IE.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
+        ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
-        self.assertEqual(DL.result, [
+        ytie_results = [YoutubeIE()._extract_id(r[0]) for r in dl.result]
-            ['http://www.youtube.com/watch?v=bV9L5Ht9LgY'],
+        self.assertEqual(ytie_results, [ 'bV9L5Ht9LgY', 'FXxLjLQi3Fg', 'tU3Bgo5qJZE'])
-            ['http://www.youtube.com/watch?v=FXxLjLQi3Fg'],
+
-            ['http://www.youtube.com/watch?v=tU3Bgo5qJZE']
+    def test_issue_673(self):
-        ])
+        dl = FakeDownloader()
+        ie = YoutubePlaylistIE(dl)
+        ie.extract('PLBB231211A4F62143')
+        self.assertTrue(len(dl.result) > 40)
 
     def test_youtube_playlist_long(self):
-        DL = FakeDownloader()
+        dl = FakeDownloader()
-        IE = YoutubePlaylistIE(DL)
+        ie = YoutubePlaylistIE(dl)
-        IE.extract('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')
+        ie.extract('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')
-        self.assertTrue(len(DL.result) >= 799)
+        self.assertTrue(len(dl.result) >= 799)
+
+    def test_youtube_playlist_with_deleted(self):
+        #651
+        dl = FakeDownloader()
+        ie = YoutubePlaylistIE(dl)
+        ie.extract('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
+        ytie_results = [YoutubeIE()._extract_id(r[0]) for r in dl.result]
+        self.assertFalse('pElCt5oNDuI' in ytie_results)
+        self.assertFalse('KdPEApIVdWM' in ytie_results)
 
     def test_youtube_course(self):
-        DL = FakeDownloader()
+        dl = FakeDownloader()
-        IE = YoutubePlaylistIE(DL)
+        ie = YoutubePlaylistIE(dl)
         # TODO find a > 100 (paginating?) videos course
-        IE.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
+        ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
-        self.assertEqual(DL.result[0], ['http://www.youtube.com/watch?v=j9WZyLZCBzs'])
+        self.assertEqual(YoutubeIE()._extract_id(dl.result[0][0]), 'j9WZyLZCBzs')
-        self.assertEqual(len(DL.result), 25)
+        self.assertEqual(len(dl.result), 25)
-        self.assertEqual(DL.result[-1], ['http://www.youtube.com/watch?v=rYefUsYuEp0'])
+        self.assertEqual(YoutubeIE()._extract_id(dl.result[-1][0]), 'rYefUsYuEp0')
 
     def test_youtube_channel(self):
         # I give up, please find a channel that does paginate and test this like test_youtube_playlist_long
         pass # TODO
 
     def test_youtube_user(self):
-        DL = FakeDownloader()
+        dl = FakeDownloader()
-        IE = YoutubeUserIE(DL)
+        ie = YoutubeUserIE(dl)
-        IE.extract('https://www.youtube.com/user/TheLinuxFoundation')
+        ie.extract('https://www.youtube.com/user/TheLinuxFoundation')
-        self.assertTrue(len(DL.result) >= 320)
+        self.assertTrue(len(dl.result) >= 320)
 
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_subtitles.py

@@ -38,20 +38,63 @@ class FakeDownloader(object):
 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()

test/tests.json

@@ -128,18 +128,6 @@
     "file": "0732f586d7.mp4",
     "md5": "f647e9e90064b53b6e046e75d0241fbd"
   },
-  {
-    "name": "TweetReel",
-    "url": "http://tweetreel.com/?77smq",
-    "file": "77smq.mov",
-    "md5": "56b4d9ca9de467920f3f99a6d91255d6",
-    "info_dict": {
-        "uploader": "itszero",
-        "uploader_id": "itszero",
-        "upload_date": "20091225",
-        "description": "Installing Gentoo Linux on Powerbook G4, it turns out the sleep indicator becomes HDD activity indicator :D"
-    }
-  },
   {
     "name": "Steam",
     "url": "http://store.steampowered.com/video/105600/",
@@ -293,7 +281,8 @@
     "file": "102.mp4",
     "md5": "7bc087e71d16f18f9b8ab9fa62a8a031",
     "info_dict": {
-        "title": "Dan Dennett: The illusion of consciousness"
+        "title": "Dan Dennett: The illusion of consciousness",
+        "thumbnail": "http://images.ted.com/images/ted/488_389x292.jpg"
     }
   },
   {
@@ -304,5 +293,40 @@
     "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"
+    }
+  },
+  {
+    "name": "LiveLeak",
+    "md5":  "0813c2430bea7a46bf13acf3406992f4",
+    "url":  "http://www.liveleak.com/view?i=757_1364311680",
+    "file":  "757_1364311680.mp4",
+    "info_dict": {
+        "title": "Most unlucky car accident",
+        "description": "extremely bad day for this guy..!",
+        "uploader": "ljfriel2"
+    }
+  },
+  {
+    "name": "WorldStarHipHop",
+    "url": "http://www.worldstarhiphop.com/videos/video.php?v=wshh6a7q1ny0G34ZwuIO",
+    "file": "wshh6a7q1ny0G34ZwuIO.mp4",
+    "md5": "9d04de741161603bf7071bbf4e883186",
+    "info_dict": {
+        "title": "Video: KO Of The Week: MMA Fighter Gets Knocked Out By Swift Head Kick! "
+    }
   }
 ]

youtube_dl/FileDownloader.py

@@ -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
@@ -104,7 +108,7 @@ class FileDownloader(object):
         self.params = params
 
         if '%(stitle)s' in self.params['outtmpl']:
-            self.to_stderr(u'WARNING: %(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.')
+            self.report_warning(u'%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.')
 
     @staticmethod
     def format_bytes(bytes):
@@ -227,13 +231,47 @@ class FileDownloader(object):
             self.to_stderr(message)
         if self.params.get('verbose'):
             if tb is None:
+                if sys.exc_info()[0]:  # if .trouble has been called from an except block
+                    tb = u''
+                    if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
+                        tb += u''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
+                    tb += compat_str(traceback.format_exc())
+                else:
                     tb_data = traceback.format_list(traceback.extract_stack())
                     tb = u''.join(tb_data)
             self.to_stderr(tb)
         if not self.params.get('ignoreerrors', False):
-            raise DownloadError(message)
+            if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
+                exc_info = sys.exc_info()[1].exc_info
+            else:
+                exc_info = sys.exc_info()
+            raise DownloadError(message, exc_info)
         self._download_retcode = 1
 
+    def report_warning(self, message):
+        '''
+        Print the message to stderr, it will be prefixed with 'WARNING:'
+        If stderr is a tty file the 'WARNING:' will be colored
+        '''
+        if sys.stderr.isatty():
+            _msg_header=u'\033[0;33mWARNING:\033[0m'
+        else:
+            _msg_header=u'WARNING:'
+        warning_message=u'%s %s' % (_msg_header,message)
+        self.to_stderr(warning_message)
+
+    def report_error(self, message, tb=None):
+        '''
+        Do the same as trouble, but prefixes the message with 'ERROR:', colored
+        in red if stderr is a tty file.
+        '''
+        if sys.stderr.isatty():
+            _msg_header = u'\033[0;31mERROR:\033[0m'
+        else:
+            _msg_header = u'ERROR:'
+        error_message = u'%s %s' % (_msg_header, message)
+        self.trouble(error_message, tb)
+
     def slow_down(self, start_time, byte_counter):
         """Sleep if the download speed is over the rate limit."""
         rate_limit = self.params.get('ratelimit', None)
@@ -265,7 +303,7 @@ class FileDownloader(object):
                 return
             os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
         except (IOError, OSError) as err:
-            self.trouble(u'ERROR: unable to rename file')
+            self.report_error(u'unable to rename file')
 
     def try_utime(self, filename, last_modified_hdr):
         """Try to set the last-modified time of the given file."""
@@ -289,9 +327,9 @@ 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):
+    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 """
@@ -360,8 +398,11 @@ class FileDownloader(object):
 
             filename = self.params['outtmpl'] % template_dict
             return filename
-        except (ValueError, KeyError) as err:
+        except KeyError as err:
-            self.trouble(u'ERROR: invalid system charset or erroneous output template')
+            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):
@@ -370,12 +411,10 @@ class FileDownloader(object):
         title = info_dict['title']
         matchtitle = self.params.get('matchtitle', False)
         if matchtitle:
-            matchtitle = matchtitle.decode('utf8')
             if not re.search(matchtitle, title, re.IGNORECASE):
                 return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"'
         rejecttitle = self.params.get('rejecttitle', False)
         if rejecttitle:
-            rejecttitle = rejecttitle.decode('utf8')
             if re.search(rejecttitle, title, re.IGNORECASE):
                 return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
         return None
@@ -427,7 +466,7 @@ class FileDownloader(object):
             if dn != '' and not os.path.exists(dn): # dn is already encoded
                 os.makedirs(dn)
         except (OSError, IOError) as err:
-            self.trouble(u'ERROR: unable to create directory ' + compat_str(err))
+            self.report_error(u'unable to create directory ' + compat_str(err))
             return
 
         if self.params.get('writedescription', False):
@@ -437,20 +476,47 @@ class FileDownloader(object):
                 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
                     descfile.write(info_dict['description'])
             except (OSError, IOError):
-                self.trouble(u'ERROR: Cannot write description file ' + descfn)
+                self.report_error(u'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
+            subtitle = info_dict['subtitles'][0]
+            (sub_error, sub_lang, sub) = subtitle
+            sub_format = self.params.get('subtitlesformat')
+            if sub_error:
+                self.report_warning("Some error while getting the subtitles")
+            else:
                 try:
-                srtfn = filename.rsplit('.', 1)[0] + u'.srt'
+                    sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
-                self.report_writesubtitles(srtfn)
+                    self.report_writesubtitles(sub_filename)
-                with io.open(encodeFilename(srtfn), 'w', encoding='utf-8') as srtfile:
+                    with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
-                    srtfile.write(info_dict['subtitles'])
+                        subfile.write(sub)
+                except (OSError, IOError):
+                    self.report_error(u'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
+                if sub_error:
+                    self.report_warning("Some error while getting the subtitles")
+                else:
+                    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'
@@ -459,7 +525,7 @@ class FileDownloader(object):
                 json_info_dict = dict((k, v) for k,v in info_dict.items() if not k in ['urlhandle'])
                 write_json_file(json_info_dict, encodeFilename(infofn))
             except (OSError, IOError):
-                self.trouble(u'ERROR: Cannot write metadata to JSON file ' + infofn)
+                self.report_error(u'Cannot write metadata to JSON file ' + infofn)
                 return
 
         if not self.params.get('skip_download', False):
@@ -471,17 +537,17 @@ class FileDownloader(object):
                 except (OSError, IOError) as err:
                     raise UnavailableVideoError()
                 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-                    self.trouble(u'ERROR: unable to download video data: %s' % str(err))
+                    self.report_error(u'unable to download video data: %s' % str(err))
                     return
                 except (ContentTooShortError, ) as err:
-                    self.trouble(u'ERROR: content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
+                    self.report_error(u'content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
                     return
 
             if success:
                 try:
                     self.post_process(filename, info_dict)
                 except (PostProcessingError) as err:
-                    self.trouble(u'ERROR: postprocessing: %s' % str(err))
+                    self.report_error(u'postprocessing: %s' % str(err))
                     return
 
     def download(self, url_list):
@@ -498,7 +564,7 @@ class FileDownloader(object):
 
                 # Warn if the _WORKING attribute is False
                 if not ie.working():
-                    self.to_stderr(u'WARNING: the program functionality for this site has been marked as broken, '
+                    self.report_warning(u'the program functionality for this site has been marked as broken, '
                                         u'and will probably not work. If you want to go on, use the -i option.')
 
                 # Suitable InfoExtractor found
@@ -510,9 +576,12 @@ class FileDownloader(object):
                 except ExtractorError as de: # An error we somewhat expected
                     self.trouble(u'ERROR: ' + compat_str(de), de.format_traceback())
                     break
+                except MaxDownloadsReached:
+                    self.to_screen(u'[info] Maximum number of downloaded files reached.')
+                    raise
                 except Exception as e:
                     if self.params.get('ignoreerrors', False):
-                        self.trouble(u'ERROR: ' + compat_str(e), tb=compat_str(traceback.format_exc()))
+                        self.report_error(u'' + compat_str(e), tb=compat_str(traceback.format_exc()))
                         break
                     else:
                         raise
@@ -526,13 +595,14 @@ class FileDownloader(object):
                         self.increment_downloads()
                         self.process_info(video)
                     except UnavailableVideoError:
-                        self.trouble(u'\nERROR: unable to download video')
+                        self.to_stderr(u"\n")
+                        self.report_error(u'unable to download video')
 
                 # Suitable InfoExtractor had been found; go to next URL
                 break
 
             if not suitable_found:
-                self.trouble(u'ERROR: no suitable InfoExtractor: %s' % url)
+                self.report_error(u'no suitable InfoExtractor: %s' % url)
 
         return self._download_retcode
 
@@ -554,10 +624,10 @@ class FileDownloader(object):
                 self.to_stderr(u'ERROR: ' + e.msg)
         if keep_video is False and not self.params.get('keepvideo', False):
             try:
-                self.to_stderr(u'Deleting original file %s (pass -k to keep)' % filename)
+                self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename)
                 os.remove(encodeFilename(filename))
             except (IOError, OSError):
-                self.to_stderr(u'WARNING: Unable to remove downloaded video file')
+                self.report_warning(u'Unable to remove downloaded video file')
 
     def _download_with_rtmpdump(self, filename, url, player_url, page_url):
         self.report_destination(filename)
@@ -565,9 +635,9 @@ class FileDownloader(object):
 
         # Check for rtmpdump first
         try:
-            subprocess.call(['rtmpdump', '-h'], stdout=(file(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
+            subprocess.call(['rtmpdump', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
         except (OSError, IOError):
-            self.trouble(u'ERROR: RTMP download detected but "rtmpdump" could not be run')
+            self.report_error(u'RTMP download detected but "rtmpdump" could not be run')
             return False
 
         # Download using rtmpdump. rtmpdump returns exit code 2 when
@@ -612,7 +682,8 @@ class FileDownloader(object):
             })
             return True
         else:
-            self.trouble(u'\nERROR: rtmpdump exited with code %d' % retval)
+            self.to_stderr(u"\n")
+            self.report_error(u'rtmpdump exited with code %d' % retval)
             return False
 
     def _do_download(self, filename, info_dict):
@@ -712,7 +783,7 @@ class FileDownloader(object):
                 self.report_retry(count, retries)
 
         if count > retries:
-            self.trouble(u'ERROR: giving up after %s retries' % retries)
+            self.report_error(u'giving up after %s retries' % retries)
             return False
 
         data_len = data.info().get('Content-length', None)
@@ -748,12 +819,13 @@ class FileDownloader(object):
                     filename = self.undo_temp_name(tmpfilename)
                     self.report_destination(filename)
                 except (OSError, IOError) as err:
-                    self.trouble(u'ERROR: unable to open for writing: %s' % str(err))
+                    self.report_error(u'unable to open for writing: %s' % str(err))
                     return False
             try:
                 stream.write(data_block)
             except (IOError, OSError) as err:
-                self.trouble(u'\nERROR: unable to write data: %s' % str(err))
+                self.to_stderr(u"\n")
+                self.report_error(u'unable to write data: %s' % str(err))
                 return False
             if not self.params.get('noresizebuffer', False):
                 block_size = self.best_block_size(after - before, len(data_block))
@@ -779,7 +851,8 @@ class FileDownloader(object):
             self.slow_down(start, byte_counter - resume_len)
 
         if stream is None:
-            self.trouble(u'\nERROR: Did not get any data blocks')
+            self.to_stderr(u"\n")
+            self.report_error(u'Did not get any data blocks')
             return False
         stream.close()
         self.report_finish()

youtube_dl/__init__.py

@@ -126,7 +126,7 @@ def parseOpts():
     general.add_option('-i', '--ignore-errors',
             action='store_true', dest='ignoreerrors', help='continue on download errors', default=False)
     general.add_option('-r', '--rate-limit',
-            dest='ratelimit', metavar='LIMIT', help='download rate limit (e.g. 50k or 44.6m)')
+            dest='ratelimit', metavar='LIMIT', help='maximum download rate (e.g. 50k or 44.6m)')
     general.add_option('-R', '--retries',
             dest='retries', metavar='RETRIES', help='number of retries (default is %default)', default=10)
     general.add_option('--buffer-size',
@@ -173,12 +173,24 @@ 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',
+    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)
+            help='write subtitle file (currently youtube only)', default=False)
-    video_format.add_option('--srt-lang',
+    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 parseOpts():
 
     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')
+        userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl.conf')
-    argv = _readOptions('/etc/youtube-dl.conf') + _readOptions(userConf) + sys.argv[1:]
+    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():
@@ -412,6 +432,7 @@ def _real_main():
             or (opts.useid and u'%(id)s.%(ext)s')
             or (opts.autonumber and u'%(autonumber)s-%(id)s.%(ext)s')
             or u'%(id)s.%(ext)s')
+
     # File downloader
     fd = FileDownloader({
         'usenetrc': opts.usenetrc,
@@ -449,9 +470,13 @@ 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': opts.matchtitle,
+        'matchtitle': decodeOption(opts.matchtitle),
-        'rejecttitle': opts.rejecttitle,
+        'rejecttitle': decodeOption(opts.rejecttitle),
         'max_downloads': opts.max_downloads,
         'prefer_free_formats': opts.prefer_free_formats,
         'verbose': opts.verbose,

youtube_dl/update.py

@@ -77,10 +77,8 @@ def update_self(to_screen, verbose, filename):
 
     to_screen(u'Updating to version ' + versions_info['latest'] + '...')
     version = versions_info['versions'][versions_info['latest']]
-    if version.get('notes'):
+
-        to_screen(u'PLEASE NOTE:')
+    print_notes(versions_info['versions'])
-        for note in version['notes']:
-            to_screen(note)
 
     if not os.access(filename, os.W_OK):
         to_screen(u'ERROR: no write permissions on %s' % filename)
@@ -158,3 +156,13 @@ del "%s"
             return
 
     to_screen(u'Updated youtube-dl. Restart youtube-dl to use the new version.')
+
+def print_notes(versions, fromVersion=__version__):
+    notes = []
+    for v,vdata in sorted(versions.items()):
+        if v > fromVersion:
+            notes.extend(vdata.get('notes', []))
+    if notes:
+        to_screen(u'PLEASE NOTE:')
+        for note in notes:
+            to_screen(note)

youtube_dl/utils.py

@@ -311,7 +311,7 @@ def clean_html(html):
     html = re.sub('<.*?>', '', html)
     # Replace html entities
     html = unescapeHTML(html)
-    return html
+    return html.strip()
 
 
 def sanitize_open(filename, open_mode):
@@ -329,7 +329,7 @@ def sanitize_open(filename, open_mode):
             if sys.platform == 'win32':
                 import msvcrt
                 msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
-            return (sys.stdout, filename)
+            return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename)
         stream = open(encodeFilename(filename), open_mode)
         return (stream, filename)
     except (IOError, OSError) as err:
@@ -420,6 +420,14 @@ def encodeFilename(s):
             encoding = 'utf-8'
         return s.encode(encoding, 'ignore')
 
+def decodeOption(optval):
+    if optval is None:
+        return optval
+    if isinstance(optval, bytes):
+        optval = optval.decode(preferredencoding())
+
+    assert isinstance(optval, compat_str)
+    return optval
 
 class ExtractorError(Exception):
     """Error during info extraction."""
@@ -427,6 +435,7 @@ class ExtractorError(Exception):
         """ tb, if given, is the original traceback (so that it can be printed out). """
         super(ExtractorError, self).__init__(msg)
         self.traceback = tb
+        self.exc_info = sys.exc_info()  # preserve original exception
 
     def format_traceback(self):
         if self.traceback is None:
@@ -441,7 +450,10 @@ class DownloadError(Exception):
     configured to continue on errors. They will contain the appropriate
     error message.
     """
-    pass
+    def __init__(self, msg, exc_info=None):
+        """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """
+        super(DownloadError, self).__init__(msg)
+        self.exc_info = exc_info
 
 
 class SameFileError(Exception):

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2013.02.19'
+__version__ = '2013.04.03'