From: Rogério Brito Date: Wed, 6 Aug 2014 18:35:56 +0000 (-0300) Subject: Imported Upstream version 2014.08.05 X-Git-Url: https://git.rapsys.eu/youtubedl/commitdiff_plain/a4f82408d9a3d921d9c2af7e4d757f44737dc7ea?ds=inline Imported Upstream version 2014.08.05 --- diff --git a/README.md b/README.md index bc5e0f7..a42dfb8 100644 --- a/README.md +++ b/README.md @@ -12,7 +12,7 @@ To install it right away for all UNIX users (Linux, OS X, etc.), type: If you do not have curl, you can alternatively use a recent wget: - sudo wget https://yt-dl.org/downloads/2014.05.13/youtube-dl -O /usr/local/bin/youtube-dl + sudo wget https://yt-dl.org/downloads/latest/youtube-dl -O /usr/local/bin/youtube-dl sudo chmod a+x /usr/local/bin/youtube-dl Windows users can [download a .exe file](https://yt-dl.org/latest/youtube-dl.exe) and place it in their home directory or any other location on their [PATH](http://en.wikipedia.org/wiki/PATH_%28variable%29). @@ -38,12 +38,6 @@ which means you can modify it, redistribute it or use it however you like. playlist or the command line) if an error occurs --dump-user-agent display the current browser identification - --user-agent UA specify a custom user agent - --referer REF specify a custom referer, use if the video - access is restricted to one domain - --add-header FIELD:VALUE specify a custom HTTP header and its value, - separated by a colon ':'. You can use this - option multiple times --list-extractors List all supported extractors and the URLs they would handle --extractor-descriptions Output descriptions of all supported @@ -51,35 +45,22 @@ which means you can modify it, redistribute it or use it however you like. --proxy URL Use the specified HTTP/HTTPS proxy. Pass in an empty string (--proxy "") for direct connection - --no-check-certificate Suppress HTTPS certificate validation. - --prefer-insecure Use an unencrypted connection to retrieve - information about the video. (Currently - supported only for YouTube) - --cache-dir DIR Location in the filesystem where youtube-dl - can store some downloaded information - permanently. By default $XDG_CACHE_HOME - /youtube-dl or ~/.cache/youtube-dl . At the - moment, only YouTube player files (for - videos with obfuscated signatures) are - cached, but that may change. - --no-cache-dir Disable filesystem caching --socket-timeout None Time to wait before giving up, in seconds - --bidi-workaround Work around terminals that lack - bidirectional text support. Requires bidiv - or fribidi executable in PATH --default-search PREFIX Use this prefix for unqualified URLs. For example "gvsearch2:" downloads two videos from google videos for youtube-dl "large apple". Use the value "auto" to let - youtube-dl guess. The default value "error" - just throws an error. + youtube-dl guess ("auto_warning" to emit a + warning when guessing). "error" just throws + an error. The default value "fixup_error" + repairs broken URLs, but emits an error if + this is not possible instead of searching. --ignore-config Do not read configuration files. When given in the global configuration file /etc /youtube-dl.conf: do not read the user configuration in ~/.config/youtube-dl.conf (%APPDATA%/youtube-dl/config.txt on Windows) - --encoding ENCODING Force the specified encoding (experimental) ## Video Selection: --playlist-start NUMBER playlist video to start at (default is 1) @@ -125,9 +106,9 @@ which means you can modify it, redistribute it or use it however you like. of SIZE. ## Filesystem Options: - -t, --title use title in file name (default) + -a, --batch-file FILE file containing URLs to download ('-' for + stdin) --id use only video ID in file name - -l, --literal [deprecated] alias of --title -A, --auto-number number downloaded files starting from 00000 -o, --output TEMPLATE output filename template. Use %(title)s to get the title, %(uploader)s for the @@ -160,18 +141,15 @@ which means you can modify it, redistribute it or use it however you like. --restrict-filenames Restrict filenames to only ASCII characters, and avoid "&" and spaces in filenames - -a, --batch-file FILE file containing URLs to download ('-' for - stdin) - --load-info FILE json file containing the video information - (created with the "--write-json" option) + -t, --title [deprecated] use title in file name + (default) + -l, --literal [deprecated] alias of --title -w, --no-overwrites do not overwrite files -c, --continue force resume of partially downloaded files. By default, youtube-dl will resume downloads if possible. --no-continue do not resume partially downloaded files (restart from beginning) - --cookies FILE file to read cookies from and dump cookie - jar in --no-part do not use .part files --no-mtime do not use the Last-modified header to set the file modification time @@ -181,6 +159,19 @@ which means you can modify it, redistribute it or use it however you like. --write-annotations write video annotations to a .annotation file --write-thumbnail write thumbnail image to disk + --load-info FILE json file containing the video information + (created with the "--write-json" option) + --cookies FILE file to read cookies from and dump cookie + jar in + --cache-dir DIR Location in the filesystem where youtube-dl + can store some downloaded information + permanently. By default $XDG_CACHE_HOME + /youtube-dl or ~/.cache/youtube-dl . At the + moment, only YouTube player files (for + videos with obfuscated signatures) are + cached, but that may change. + --no-cache-dir Disable filesystem caching + --rm-cache-dir Delete all filesystem cache files ## Verbosity / Simulation Options: -q, --quiet activates quiet mode @@ -210,6 +201,22 @@ which means you can modify it, redistribute it or use it however you like. problems --print-traffic Display sent and read HTTP traffic +## Workarounds: + --encoding ENCODING Force the specified encoding (experimental) + --no-check-certificate Suppress HTTPS certificate validation. + --prefer-insecure Use an unencrypted connection to retrieve + information about the video. (Currently + supported only for YouTube) + --user-agent UA specify a custom user agent + --referer REF specify a custom referer, use if the video + access is restricted to one domain + --add-header FIELD:VALUE specify a custom HTTP header and its value, + separated by a colon ':'. You can use this + option multiple times + --bidi-workaround Work around terminals that lack + bidirectional text support. Requires bidiv + or fribidi executable in PATH + ## Video Format Options: -f, --format FORMAT video format code, specify the order of preference using slashes: "-f 22/17/18". diff --git a/README.txt b/README.txt index 5555b2a..4f41be2 100644 --- a/README.txt +++ b/README.txt @@ -15,7 +15,7 @@ To install it right away for all UNIX users (Linux, OS X, etc.), type: If you do not have curl, you can alternatively use a recent wget: - sudo wget https://yt-dl.org/downloads/2014.05.13/youtube-dl -O /usr/local/bin/youtube-dl + sudo wget https://yt-dl.org/downloads/latest/youtube-dl -O /usr/local/bin/youtube-dl sudo chmod a+x /usr/local/bin/youtube-dl Windows users can download a .exe file and place it in their home @@ -50,12 +50,6 @@ OPTIONS playlist or the command line) if an error occurs --dump-user-agent display the current browser identification - --user-agent UA specify a custom user agent - --referer REF specify a custom referer, use if the video - access is restricted to one domain - --add-header FIELD:VALUE specify a custom HTTP header and its value, - separated by a colon ':'. You can use this - option multiple times --list-extractors List all supported extractors and the URLs they would handle --extractor-descriptions Output descriptions of all supported @@ -63,35 +57,22 @@ OPTIONS --proxy URL Use the specified HTTP/HTTPS proxy. Pass in an empty string (--proxy "") for direct connection - --no-check-certificate Suppress HTTPS certificate validation. - --prefer-insecure Use an unencrypted connection to retrieve - information about the video. (Currently - supported only for YouTube) - --cache-dir DIR Location in the filesystem where youtube-dl - can store some downloaded information - permanently. By default $XDG_CACHE_HOME - /youtube-dl or ~/.cache/youtube-dl . At the - moment, only YouTube player files (for - videos with obfuscated signatures) are - cached, but that may change. - --no-cache-dir Disable filesystem caching --socket-timeout None Time to wait before giving up, in seconds - --bidi-workaround Work around terminals that lack - bidirectional text support. Requires bidiv - or fribidi executable in PATH --default-search PREFIX Use this prefix for unqualified URLs. For example "gvsearch2:" downloads two videos from google videos for youtube-dl "large apple". Use the value "auto" to let - youtube-dl guess. The default value "error" - just throws an error. + youtube-dl guess ("auto_warning" to emit a + warning when guessing). "error" just throws + an error. The default value "fixup_error" + repairs broken URLs, but emits an error if + this is not possible instead of searching. --ignore-config Do not read configuration files. When given in the global configuration file /etc /youtube-dl.conf: do not read the user configuration in ~/.config/youtube-dl.conf (%APPDATA%/youtube-dl/config.txt on Windows) - --encoding ENCODING Force the specified encoding (experimental) Video Selection: ---------------- @@ -143,9 +124,9 @@ Download Options: Filesystem Options: ------------------- - -t, --title use title in file name (default) + -a, --batch-file FILE file containing URLs to download ('-' for + stdin) --id use only video ID in file name - -l, --literal [deprecated] alias of --title -A, --auto-number number downloaded files starting from 00000 -o, --output TEMPLATE output filename template. Use %(title)s to get the title, %(uploader)s for the @@ -178,18 +159,15 @@ Filesystem Options: --restrict-filenames Restrict filenames to only ASCII characters, and avoid "&" and spaces in filenames - -a, --batch-file FILE file containing URLs to download ('-' for - stdin) - --load-info FILE json file containing the video information - (created with the "--write-json" option) + -t, --title [deprecated] use title in file name + (default) + -l, --literal [deprecated] alias of --title -w, --no-overwrites do not overwrite files -c, --continue force resume of partially downloaded files. By default, youtube-dl will resume downloads if possible. --no-continue do not resume partially downloaded files (restart from beginning) - --cookies FILE file to read cookies from and dump cookie - jar in --no-part do not use .part files --no-mtime do not use the Last-modified header to set the file modification time @@ -199,6 +177,19 @@ Filesystem Options: --write-annotations write video annotations to a .annotation file --write-thumbnail write thumbnail image to disk + --load-info FILE json file containing the video information + (created with the "--write-json" option) + --cookies FILE file to read cookies from and dump cookie + jar in + --cache-dir DIR Location in the filesystem where youtube-dl + can store some downloaded information + permanently. By default $XDG_CACHE_HOME + /youtube-dl or ~/.cache/youtube-dl . At the + moment, only YouTube player files (for + videos with obfuscated signatures) are + cached, but that may change. + --no-cache-dir Disable filesystem caching + --rm-cache-dir Delete all filesystem cache files Verbosity / Simulation Options: ------------------------------- @@ -230,6 +221,24 @@ Verbosity / Simulation Options: problems --print-traffic Display sent and read HTTP traffic +Workarounds: +------------ + + --encoding ENCODING Force the specified encoding (experimental) + --no-check-certificate Suppress HTTPS certificate validation. + --prefer-insecure Use an unencrypted connection to retrieve + information about the video. (Currently + supported only for YouTube) + --user-agent UA specify a custom user agent + --referer REF specify a custom referer, use if the video + access is restricted to one domain + --add-header FIELD:VALUE specify a custom HTTP header and its value, + separated by a colon ':'. You can use this + option multiple times + --bidi-workaround Work around terminals that lack + bidirectional text support. Requires bidiv + or fribidi executable in PATH + Video Format Options: --------------------- diff --git a/test/helper.py b/test/helper.py index 230d2bd..b7299fb 100644 --- a/test/helper.py +++ b/test/helper.py @@ -137,8 +137,8 @@ def expect_info_dict(self, expected_dict, got_dict): def assertRegexpMatches(self, text, regexp, msg=None): - if hasattr(self, 'assertRegexpMatches'): - return self.assertRegexpMatches(text, regexp, msg) + if hasattr(self, 'assertRegexp'): + return self.assertRegexp(text, regexp, msg) else: m = re.match(regexp, text) if not m: @@ -148,3 +148,10 @@ def assertRegexpMatches(self, text, regexp, msg=None): else: msg = note + ', ' + msg self.assertTrue(m, msg) + + +def assertGreaterEqual(self, got, expected, msg=None): + if not (got >= expected): + if msg is None: + msg = '%r not greater than or equal to %r' % (got, expected) + self.assertTrue(got >= expected, msg) diff --git a/test/swftests/.gitignore b/test/swftests/.gitignore new file mode 100644 index 0000000..da97ff7 --- /dev/null +++ b/test/swftests/.gitignore @@ -0,0 +1 @@ +*.swf diff --git a/test/swftests/ArrayAccess.as b/test/swftests/ArrayAccess.as new file mode 100644 index 0000000..e22caa3 --- /dev/null +++ b/test/swftests/ArrayAccess.as @@ -0,0 +1,19 @@ +// input: [["a", "b", "c", "d"]] +// output: ["c", "b", "a", "d"] + +package { +public class ArrayAccess { + public static function main(ar:Array):Array { + var aa:ArrayAccess = new ArrayAccess(); + return aa.f(ar, 2); + } + + private function f(ar:Array, num:Number):Array{ + var x:String = ar[0]; + var y:String = ar[num % ar.length]; + ar[0] = y; + ar[num] = x; + return ar; + } +} +} diff --git a/test/swftests/ArrayAccess.swf b/test/swftests/ArrayAccess.swf new file mode 100644 index 0000000..3f5b32e Binary files /dev/null and b/test/swftests/ArrayAccess.swf differ diff --git a/test/swftests/ClassCall.as b/test/swftests/ClassCall.as new file mode 100644 index 0000000..aef58da --- /dev/null +++ b/test/swftests/ClassCall.as @@ -0,0 +1,17 @@ +// input: [] +// output: 121 + +package { +public class ClassCall { + public static function main():int{ + var f:OtherClass = new OtherClass(); + return f.func(100,20); + } +} +} + +class OtherClass { + public function func(x: int, y: int):int { + return x+y+1; + } +} diff --git a/test/swftests/ClassCall.swf b/test/swftests/ClassCall.swf new file mode 100644 index 0000000..2f1b31b Binary files /dev/null and b/test/swftests/ClassCall.swf differ diff --git a/test/swftests/ClassConstruction.as b/test/swftests/ClassConstruction.as new file mode 100644 index 0000000..436479f --- /dev/null +++ b/test/swftests/ClassConstruction.as @@ -0,0 +1,15 @@ +// input: [] +// output: 0 + +package { +public class ClassConstruction { + public static function main():int{ + var f:Foo = new Foo(); + return 0; + } +} +} + +class Foo { + +} diff --git a/test/swftests/ClassConstruction.swf b/test/swftests/ClassConstruction.swf new file mode 100644 index 0000000..51e2ad8 Binary files /dev/null and b/test/swftests/ClassConstruction.swf differ diff --git a/test/swftests/LocalVars.as b/test/swftests/LocalVars.as new file mode 100644 index 0000000..b2911a9 --- /dev/null +++ b/test/swftests/LocalVars.as @@ -0,0 +1,13 @@ +// input: [1, 2] +// output: 3 + +package { +public class LocalVars { + public static function main(a:int, b:int):int{ + var c:int = a + b + b; + var d:int = c - b; + var e:int = d; + return e; + } +} +} diff --git a/test/swftests/LocalVars.swf b/test/swftests/LocalVars.swf new file mode 100644 index 0000000..3e9344f Binary files /dev/null and b/test/swftests/LocalVars.swf differ diff --git a/test/swftests/PrivateCall.as b/test/swftests/PrivateCall.as new file mode 100644 index 0000000..f1c110a --- /dev/null +++ b/test/swftests/PrivateCall.as @@ -0,0 +1,21 @@ +// input: [] +// output: 9 + +package { +public class PrivateCall { + public static function main():int{ + var f:OtherClass = new OtherClass(); + return f.func(); + } +} +} + +class OtherClass { + private function pf():int { + return 9; + } + + public function func():int { + return this.pf(); + } +} diff --git a/test/swftests/PrivateCall.swf b/test/swftests/PrivateCall.swf new file mode 100644 index 0000000..55d9eac Binary files /dev/null and b/test/swftests/PrivateCall.swf differ diff --git a/test/swftests/StaticAssignment.as b/test/swftests/StaticAssignment.as new file mode 100644 index 0000000..b061c21 --- /dev/null +++ b/test/swftests/StaticAssignment.as @@ -0,0 +1,13 @@ +// input: [1] +// output: 1 + +package { +public class StaticAssignment { + public static var v:int; + + public static function main(a:int):int{ + v = a; + return v; + } +} +} diff --git a/test/swftests/StaticAssignment.swf b/test/swftests/StaticAssignment.swf new file mode 100644 index 0000000..b28450a Binary files /dev/null and b/test/swftests/StaticAssignment.swf differ diff --git a/test/swftests/StaticRetrieval.as b/test/swftests/StaticRetrieval.as new file mode 100644 index 0000000..c8352d8 --- /dev/null +++ b/test/swftests/StaticRetrieval.as @@ -0,0 +1,16 @@ +// input: [] +// output: 1 + +package { +public class StaticRetrieval { + public static var v:int; + + public static function main():int{ + if (v) { + return 0; + } else { + return 1; + } + } +} +} diff --git a/test/swftests/StaticRetrieval.swf b/test/swftests/StaticRetrieval.swf new file mode 100644 index 0000000..bb0f3de Binary files /dev/null and b/test/swftests/StaticRetrieval.swf differ diff --git a/test/test_all_urls.py b/test/test_all_urls.py index 2bc81f0..0ff47cf 100644 --- a/test/test_all_urls.py +++ b/test/test_all_urls.py @@ -15,7 +15,6 @@ from youtube_dl.extractor import ( FacebookIE, gen_extractors, JustinTVIE, - PBSIE, YoutubeIE, ) diff --git a/test/test_download.py b/test/test_download.py index f171c10..d654058 100644 --- a/test/test_download.py +++ b/test/test_download.py @@ -10,7 +10,6 @@ from test.helper import ( get_params, gettestcases, expect_info_dict, - md5, try_rm, report_warning, ) @@ -24,7 +23,6 @@ import socket import youtube_dl.YoutubeDL from youtube_dl.utils import ( compat_http_client, - compat_str, compat_urllib_error, compat_HTTPError, DownloadError, diff --git a/test/test_playlists.py b/test/test_playlists.py index 1a38a66..4f18834 100644 --- a/test/test_playlists.py +++ b/test/test_playlists.py @@ -11,6 +11,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) from test.helper import ( assertRegexpMatches, + assertGreaterEqual, expect_info_dict, FakeYDL, ) @@ -71,8 +72,8 @@ class TestPlaylists(unittest.TestCase): ie = DailymotionUserIE(dl) result = ie.extract('https://www.dailymotion.com/user/nqtv') self.assertIsPlaylist(result) + assertGreaterEqual(self, len(result['entries']), 100) self.assertEqual(result['title'], 'Rémi Gaillard') - self.assertTrue(len(result['entries']) >= 100) def test_vimeo_channel(self): dl = FakeYDL() @@ -111,7 +112,7 @@ class TestPlaylists(unittest.TestCase): ie = VineUserIE(dl) result = ie.extract('https://vine.co/Visa') self.assertIsPlaylist(result) - self.assertTrue(len(result['entries']) >= 47) + assertGreaterEqual(self, len(result['entries']), 47) def test_ustream_channel(self): dl = FakeYDL() @@ -119,7 +120,7 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('http://www.ustream.tv/channel/channeljapan') self.assertIsPlaylist(result) self.assertEqual(result['id'], '10874166') - self.assertTrue(len(result['entries']) >= 54) + assertGreaterEqual(self, len(result['entries']), 54) def test_soundcloud_set(self): dl = FakeYDL() @@ -127,7 +128,7 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep') self.assertIsPlaylist(result) self.assertEqual(result['title'], 'The Royal Concept EP') - self.assertTrue(len(result['entries']) >= 6) + assertGreaterEqual(self, len(result['entries']), 6) def test_soundcloud_user(self): dl = FakeYDL() @@ -135,7 +136,7 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('https://soundcloud.com/the-concept-band') self.assertIsPlaylist(result) self.assertEqual(result['id'], '9615865') - self.assertTrue(len(result['entries']) >= 12) + assertGreaterEqual(self, len(result['entries']), 12) def test_soundcloud_likes(self): dl = FakeYDL() @@ -143,7 +144,7 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('https://soundcloud.com/the-concept-band/likes') self.assertIsPlaylist(result) self.assertEqual(result['id'], '9615865') - self.assertTrue(len(result['entries']) >= 1) + assertGreaterEqual(self, len(result['entries']), 1) def test_soundcloud_playlist(self): dl = FakeYDL() @@ -153,7 +154,7 @@ class TestPlaylists(unittest.TestCase): self.assertEqual(result['id'], '4110309') self.assertEqual(result['title'], 'TILT Brass - Bowery Poetry Club, August \'03 [Non-Site SCR 02]') assertRegexpMatches( - self, result['description'], r'TILT Brass - Bowery Poetry Club') + self, result['description'], r'.*?TILT Brass - Bowery Poetry Club') self.assertEqual(len(result['entries']), 6) def test_livestream_event(self): @@ -162,7 +163,7 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('http://new.livestream.com/tedx/cityenglish') self.assertIsPlaylist(result) self.assertEqual(result['title'], 'TEDCity2.0 (English)') - self.assertTrue(len(result['entries']) >= 4) + assertGreaterEqual(self, len(result['entries']), 4) def test_livestreamoriginal_folder(self): dl = FakeYDL() @@ -170,7 +171,7 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('https://www.livestream.com/newplay/folder?dirId=a07bf706-d0e4-4e75-a747-b021d84f2fd3') self.assertIsPlaylist(result) self.assertEqual(result['id'], 'a07bf706-d0e4-4e75-a747-b021d84f2fd3') - self.assertTrue(len(result['entries']) >= 28) + assertGreaterEqual(self, len(result['entries']), 28) def test_nhl_videocenter(self): dl = FakeYDL() @@ -187,15 +188,15 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('http://bambuser.com/channel/pixelversity') self.assertIsPlaylist(result) self.assertEqual(result['title'], 'pixelversity') - self.assertTrue(len(result['entries']) >= 60) + assertGreaterEqual(self, len(result['entries']), 60) def test_bandcamp_album(self): dl = FakeYDL() ie = BandcampAlbumIE(dl) - result = ie.extract('http://mpallante.bandcamp.com/album/nightmare-night-ep') + result = ie.extract('http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave') self.assertIsPlaylist(result) - self.assertEqual(result['title'], 'Nightmare Night EP') - self.assertTrue(len(result['entries']) >= 4) + self.assertEqual(result['title'], 'Hierophany of the Open Grave') + assertGreaterEqual(self, len(result['entries']), 9) def test_smotri_community(self): dl = FakeYDL() @@ -204,7 +205,7 @@ class TestPlaylists(unittest.TestCase): self.assertIsPlaylist(result) self.assertEqual(result['id'], 'kommuna') self.assertEqual(result['title'], 'КПРФ') - self.assertTrue(len(result['entries']) >= 4) + assertGreaterEqual(self, len(result['entries']), 4) def test_smotri_user(self): dl = FakeYDL() @@ -213,7 +214,7 @@ class TestPlaylists(unittest.TestCase): self.assertIsPlaylist(result) self.assertEqual(result['id'], 'inspector') self.assertEqual(result['title'], 'Inspector') - self.assertTrue(len(result['entries']) >= 9) + assertGreaterEqual(self, len(result['entries']), 9) def test_AcademicEarthCourse(self): dl = FakeYDL() @@ -232,7 +233,7 @@ class TestPlaylists(unittest.TestCase): self.assertIsPlaylist(result) self.assertEqual(result['id'], 'dvoe_iz_lartsa') self.assertEqual(result['title'], 'Двое из ларца (2006 - 2008)') - self.assertTrue(len(result['entries']) >= 24) + assertGreaterEqual(self, len(result['entries']), 24) def test_ivi_compilation_season(self): dl = FakeYDL() @@ -241,7 +242,7 @@ class TestPlaylists(unittest.TestCase): self.assertIsPlaylist(result) self.assertEqual(result['id'], 'dvoe_iz_lartsa/season1') self.assertEqual(result['title'], 'Двое из ларца (2006 - 2008) 1 сезон') - self.assertTrue(len(result['entries']) >= 12) + assertGreaterEqual(self, len(result['entries']), 12) def test_imdb_list(self): dl = FakeYDL() @@ -260,7 +261,7 @@ class TestPlaylists(unittest.TestCase): self.assertEqual(result['id'], 'cryptography') self.assertEqual(result['title'], 'Journey into cryptography') self.assertEqual(result['description'], 'How have humans protected their secret messages through history? What has changed today?') - self.assertTrue(len(result['entries']) >= 3) + assertGreaterEqual(self, len(result['entries']), 3) def test_EveryonesMixtape(self): dl = FakeYDL() @@ -277,7 +278,7 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('http://rutube.ru/tags/video/1800/') self.assertIsPlaylist(result) self.assertEqual(result['id'], '1800') - self.assertTrue(len(result['entries']) >= 68) + assertGreaterEqual(self, len(result['entries']), 68) def test_rutube_person(self): dl = FakeYDL() @@ -285,7 +286,7 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('http://rutube.ru/video/person/313878/') self.assertIsPlaylist(result) self.assertEqual(result['id'], '313878') - self.assertTrue(len(result['entries']) >= 37) + assertGreaterEqual(self, len(result['entries']), 37) def test_multiple_brightcove_videos(self): # https://github.com/rg3/youtube-dl/issues/2283 @@ -322,7 +323,7 @@ class TestPlaylists(unittest.TestCase): self.assertIsPlaylist(result) self.assertEqual(result['id'], '10') self.assertEqual(result['title'], 'Who are the hackers?') - self.assertTrue(len(result['entries']) >= 6) + assertGreaterEqual(self, len(result['entries']), 6) def test_toypics_user(self): dl = FakeYDL() @@ -330,7 +331,7 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('http://videos.toypics.net/Mikey') self.assertIsPlaylist(result) self.assertEqual(result['id'], 'Mikey') - self.assertTrue(len(result['entries']) >= 17) + assertGreaterEqual(self, len(result['entries']), 17) def test_xtube_user(self): dl = FakeYDL() @@ -338,7 +339,7 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('http://www.xtube.com/community/profile.php?user=greenshowers') self.assertIsPlaylist(result) self.assertEqual(result['id'], 'greenshowers') - self.assertTrue(len(result['entries']) >= 155) + assertGreaterEqual(self, len(result['entries']), 155) def test_InstagramUser(self): dl = FakeYDL() @@ -346,7 +347,7 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('http://instagram.com/porsche') self.assertIsPlaylist(result) self.assertEqual(result['id'], 'porsche') - self.assertTrue(len(result['entries']) >= 2) + assertGreaterEqual(self, len(result['entries']), 2) test_video = next( e for e in result['entries'] if e['id'] == '614605558512799803_462752227') @@ -385,7 +386,7 @@ class TestPlaylists(unittest.TestCase): self.assertEqual(result['id'], '152147') self.assertEqual( result['title'], 'Brace Yourself - Today\'s Weirdest News') - self.assertTrue(len(result['entries']) >= 10) + assertGreaterEqual(self, len(result['entries']), 10) def test_TeacherTubeUser(self): dl = FakeYDL() @@ -393,7 +394,7 @@ class TestPlaylists(unittest.TestCase): result = ie.extract('http://www.teachertube.com/user/profile/rbhagwati2') self.assertIsPlaylist(result) self.assertEqual(result['id'], 'rbhagwati2') - self.assertTrue(len(result['entries']) >= 179) + assertGreaterEqual(self, len(result['entries']), 179) if __name__ == '__main__': unittest.main() diff --git a/test/test_swfinterp.py b/test/test_swfinterp.py new file mode 100644 index 0000000..b42cd74 --- /dev/null +++ b/test/test_swfinterp.py @@ -0,0 +1,77 @@ +#!/usr/bin/env python + +# Allow direct execution +import os +import sys +import unittest +sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) + + +import errno +import io +import json +import re +import subprocess + +from youtube_dl.swfinterp import SWFInterpreter + + +TEST_DIR = os.path.join( + os.path.dirname(os.path.abspath(__file__)), 'swftests') + + +class TestSWFInterpreter(unittest.TestCase): + pass + + +def _make_testfunc(testfile): + m = re.match(r'^(.*)\.(as)$', testfile) + if not m: + return + test_id = m.group(1) + + def test_func(self): + as_file = os.path.join(TEST_DIR, testfile) + swf_file = os.path.join(TEST_DIR, test_id + '.swf') + if ((not os.path.exists(swf_file)) + or os.path.getmtime(swf_file) < os.path.getmtime(as_file)): + # Recompile + try: + subprocess.check_call(['mxmlc', '-output', swf_file, as_file]) + except OSError as ose: + if ose.errno == errno.ENOENT: + print('mxmlc not found! Skipping test.') + return + raise + + with open(swf_file, 'rb') as swf_f: + swf_content = swf_f.read() + swfi = SWFInterpreter(swf_content) + + with io.open(as_file, 'r', encoding='utf-8') as as_f: + as_content = as_f.read() + + def _find_spec(key): + m = re.search( + r'(?m)^//\s*%s:\s*(.*?)\n' % re.escape(key), as_content) + if not m: + raise ValueError('Cannot find %s in %s' % (key, testfile)) + return json.loads(m.group(1)) + + input_args = _find_spec('input') + output = _find_spec('output') + + swf_class = swfi.extract_class(test_id) + func = swfi.extract_function(swf_class, 'main') + res = func(input_args) + self.assertEqual(res, output) + + test_func.__name__ = str('test_swf_' + test_id) + setattr(TestSWFInterpreter, test_func.__name__, test_func) + + +for testfile in os.listdir(TEST_DIR): + _make_testfunc(testfile) + +if __name__ == '__main__': + unittest.main() diff --git a/test/test_youtube_signature.py b/test/test_youtube_signature.py index d955339..604e76a 100644 --- a/test/test_youtube_signature.py +++ b/test/test_youtube_signature.py @@ -1,5 +1,7 @@ #!/usr/bin/env python +from __future__ import unicode_literals + # Allow direct execution import os import sys @@ -16,35 +18,65 @@ from youtube_dl.utils import compat_str, compat_urlretrieve _TESTS = [ ( - u'https://s.ytimg.com/yts/jsbin/html5player-vflHOr_nV.js', - u'js', + 'https://s.ytimg.com/yts/jsbin/html5player-vflHOr_nV.js', + 'js', 86, - u'>=<;:/.-[+*)(\'&%$#"!ZYX0VUTSRQPONMLKJIHGFEDCBA\\yxwvutsrqponmlkjihgfedcba987654321', + '>=<;:/.-[+*)(\'&%$#"!ZYX0VUTSRQPONMLKJIHGFEDCBA\\yxwvutsrqponmlkjihgfedcba987654321', ), ( - u'https://s.ytimg.com/yts/jsbin/html5player-vfldJ8xgI.js', - u'js', + 'https://s.ytimg.com/yts/jsbin/html5player-vfldJ8xgI.js', + 'js', 85, - u'3456789a0cdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRS[UVWXYZ!"#$%&\'()*+,-./:;<=>?@', + '3456789a0cdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRS[UVWXYZ!"#$%&\'()*+,-./:;<=>?@', ), ( - u'https://s.ytimg.com/yts/jsbin/html5player-vfle-mVwz.js', - u'js', + 'https://s.ytimg.com/yts/jsbin/html5player-vfle-mVwz.js', + 'js', 90, - u']\\[@?>=<;:/.-,+*)(\'&%$#"hZYXWVUTSRQPONMLKJIHGFEDCBAzyxwvutsrqponmlkjiagfedcb39876', + ']\\[@?>=<;:/.-,+*)(\'&%$#"hZYXWVUTSRQPONMLKJIHGFEDCBAzyxwvutsrqponmlkjiagfedcb39876', + ), + ( + 'https://s.ytimg.com/yts/jsbin/html5player-en_US-vfl0Cbn9e.js', + 'js', + 84, + 'O1I3456789abcde0ghijklmnopqrstuvwxyzABCDEFGHfJKLMN2PQRSTUVW@YZ!"#$%&\'()*+,-./:;<=', + ), + ( + 'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflXGBaUN.js', + 'js', + '2ACFC7A61CA478CD21425E5A57EBD73DDC78E22A.2094302436B2D377D14A3BBA23022D023B8BC25AA', + 'A52CB8B320D22032ABB3A41D773D2B6342034902.A22E87CDD37DBE75A5E52412DC874AC16A7CFCA2', ), ( - u'https://s.ytimg.com/yts/jsbin/html5player-en_US-vfl0Cbn9e.js', - u'js', + 'http://s.ytimg.com/yts/swfbin/player-vfl5vIhK2/watch_as3.swf', + 'swf', + 86, + 'O1I3456789abcde0ghijklmnopqrstuvwxyzABCDEFGHfJKLMN2PQRSTUVWXY\\!"#$%&\'()*+,-./:;<=>?' + ), + ( + 'http://s.ytimg.com/yts/swfbin/player-vflmDyk47/watch_as3.swf', + 'swf', + 'F375F75BF2AFDAAF2666E43868D46816F83F13E81C46.3725A8218E446A0DECD33F79DC282994D6AA92C92C9', + '9C29AA6D499282CD97F33DCED0A644E8128A5273.64C18E31F38361864D86834E6662FAADFA2FB57F' + ), + ( + 'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflBb0OQx.js', + 'js', 84, - u'O1I3456789abcde0ghijklmnopqrstuvwxyzABCDEFGHfJKLMN2PQRSTUVW@YZ!"#$%&\'()*+,-./:;<=', + '123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQ0STUVWXYZ!"#$%&\'()*+,@./:;<=>' ), ( - u'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflXGBaUN.js', - u'js', - u'2ACFC7A61CA478CD21425E5A57EBD73DDC78E22A.2094302436B2D377D14A3BBA23022D023B8BC25AA', - u'A52CB8B320D22032ABB3A41D773D2B6342034902.A22E87CDD37DBE75A5E52412DC874AC16A7CFCA2', + 'https://s.ytimg.com/yts/jsbin/html5player-en_US-vfl9FYC6l.js', + 'js', + 83, + '123456789abcdefghijklmnopqr0tuvwxyzABCDETGHIJKLMNOPQRS>UVWXYZ!"#$%&\'()*+,-./:;<=F' ), + ( + 'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflCGk6yw/html5player.js', + 'js', + '4646B5181C6C3020DF1D9C7FCFEA.AD80ABF70C39BD369CCCAE780AFBB98FA6B6CB42766249D9488C288', + '82C8849D94266724DC6B6AF89BBFA087EACCD963.B93C07FBA084ACAEFCF7C9D1FD0203C6C1815B6B' + ) ] @@ -57,12 +89,12 @@ class TestSignature(unittest.TestCase): def make_tfunc(url, stype, sig_input, expected_sig): - basename = url.rpartition('/')[2] - m = re.match(r'.*-([a-zA-Z0-9_-]+)\.[a-z]+$', basename) - assert m, '%r should follow URL format' % basename + m = re.match(r'.*-([a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.[a-z]+$', url) + assert m, '%r should follow URL format' % url test_id = m.group(1) def test_func(self): + basename = 'player-%s.%s' % (test_id, stype) fn = os.path.join(self.TESTDATA_DIR, basename) if not os.path.exists(fn): diff --git a/youtube-dl b/youtube-dl index 73304fe..bb583ea 100755 Binary files a/youtube-dl and b/youtube-dl differ diff --git a/youtube-dl.1 b/youtube-dl.1 index 26833e8..59edc71 100644 --- a/youtube-dl.1 +++ b/youtube-dl.1 @@ -29,12 +29,6 @@ redistribute it or use it however you like. \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ playlist\ or\ the\ command\ line)\ if\ an\ error \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ occurs \-\-dump\-user\-agent\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ display\ the\ current\ browser\ identification -\-\-user\-agent\ UA\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ specify\ a\ custom\ user\ agent -\-\-referer\ REF\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ specify\ a\ custom\ referer,\ use\ if\ the\ video -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ access\ is\ restricted\ to\ one\ domain -\-\-add\-header\ FIELD:VALUE\ \ \ \ \ \ \ \ \ specify\ a\ custom\ HTTP\ header\ and\ its\ value, -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ separated\ by\ a\ colon\ \[aq]:\[aq].\ You\ can\ use\ this -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ option\ multiple\ times \-\-list\-extractors\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ List\ all\ supported\ extractors\ and\ the\ URLs \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ they\ would\ handle \-\-extractor\-descriptions\ \ \ \ \ \ \ \ \ Output\ descriptions\ of\ all\ supported @@ -42,35 +36,22 @@ redistribute it or use it however you like. \-\-proxy\ URL\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Use\ the\ specified\ HTTP/HTTPS\ proxy.\ Pass\ in \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ an\ empty\ string\ (\-\-proxy\ "")\ for\ direct \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ connection -\-\-no\-check\-certificate\ \ \ \ \ \ \ \ \ \ \ Suppress\ HTTPS\ certificate\ validation. -\-\-prefer\-insecure\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Use\ an\ unencrypted\ connection\ to\ retrieve -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ information\ about\ the\ video.\ (Currently -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ supported\ only\ for\ YouTube) -\-\-cache\-dir\ DIR\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Location\ in\ the\ filesystem\ where\ youtube\-dl -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ can\ store\ some\ downloaded\ information -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ permanently.\ By\ default\ $XDG_CACHE_HOME -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ /youtube\-dl\ or\ ~/.cache/youtube\-dl\ .\ At\ the -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ moment,\ only\ YouTube\ player\ files\ (for -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ videos\ with\ obfuscated\ signatures)\ are -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ cached,\ but\ that\ may\ change. -\-\-no\-cache\-dir\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Disable\ filesystem\ caching \-\-socket\-timeout\ None\ \ \ \ \ \ \ \ \ \ \ \ Time\ to\ wait\ before\ giving\ up,\ in\ seconds -\-\-bidi\-workaround\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Work\ around\ terminals\ that\ lack -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ bidirectional\ text\ support.\ Requires\ bidiv -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ or\ fribidi\ executable\ in\ PATH \-\-default\-search\ PREFIX\ \ \ \ \ \ \ \ \ \ Use\ this\ prefix\ for\ unqualified\ URLs.\ For \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ example\ "gvsearch2:"\ downloads\ two\ videos \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ from\ google\ videos\ for\ \ youtube\-dl\ "large \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ apple".\ Use\ the\ value\ "auto"\ to\ let -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ youtube\-dl\ guess.\ The\ default\ value\ "error" -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ just\ throws\ an\ error. +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ youtube\-dl\ guess\ ("auto_warning"\ to\ emit\ a +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ warning\ when\ guessing).\ "error"\ just\ throws +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ an\ error.\ The\ default\ value\ "fixup_error" +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ repairs\ broken\ URLs,\ but\ emits\ an\ error\ if +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ this\ is\ not\ possible\ instead\ of\ searching. \-\-ignore\-config\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Do\ not\ read\ configuration\ files.\ When\ given \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ in\ the\ global\ configuration\ file\ /etc \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ /youtube\-dl.conf:\ do\ not\ read\ the\ user \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ configuration\ in\ ~/.config/youtube\-dl.conf \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ (%APPDATA%/youtube\-dl/config.txt\ on \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Windows) -\-\-encoding\ ENCODING\ \ \ \ \ \ \ \ \ \ \ \ \ \ Force\ the\ specified\ encoding\ (experimental) \f[] .fi .SS Video Selection: @@ -128,9 +109,9 @@ redistribute it or use it however you like. .IP .nf \f[C] -\-t,\ \-\-title\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ use\ title\ in\ file\ name\ (default) +\-a,\ \-\-batch\-file\ FILE\ \ \ \ \ \ \ \ \ \ \ \ file\ containing\ URLs\ to\ download\ (\[aq]\-\[aq]\ for +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ stdin) \-\-id\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ use\ only\ video\ ID\ in\ file\ name -\-l,\ \-\-literal\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ [deprecated]\ alias\ of\ \-\-title \-A,\ \-\-auto\-number\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ number\ downloaded\ files\ starting\ from\ 00000 \-o,\ \-\-output\ TEMPLATE\ \ \ \ \ \ \ \ \ \ \ \ output\ filename\ template.\ Use\ %(title)s\ to \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ get\ the\ title,\ %(uploader)s\ for\ the @@ -163,18 +144,15 @@ redistribute it or use it however you like. \-\-restrict\-filenames\ \ \ \ \ \ \ \ \ \ \ \ \ Restrict\ filenames\ to\ only\ ASCII \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ characters,\ and\ avoid\ "&"\ and\ spaces\ in \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ filenames -\-a,\ \-\-batch\-file\ FILE\ \ \ \ \ \ \ \ \ \ \ \ file\ containing\ URLs\ to\ download\ (\[aq]\-\[aq]\ for -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ stdin) -\-\-load\-info\ FILE\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ json\ file\ containing\ the\ video\ information -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ (created\ with\ the\ "\-\-write\-json"\ option) +\-t,\ \-\-title\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ [deprecated]\ use\ title\ in\ file\ name +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ (default) +\-l,\ \-\-literal\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ [deprecated]\ alias\ of\ \-\-title \-w,\ \-\-no\-overwrites\ \ \ \ \ \ \ \ \ \ \ \ \ \ do\ not\ overwrite\ files \-c,\ \-\-continue\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ force\ resume\ of\ partially\ downloaded\ files. \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ By\ default,\ youtube\-dl\ will\ resume \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ downloads\ if\ possible. \-\-no\-continue\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ do\ not\ resume\ partially\ downloaded\ files \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ (restart\ from\ beginning) -\-\-cookies\ FILE\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ file\ to\ read\ cookies\ from\ and\ dump\ cookie -\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ jar\ in \-\-no\-part\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ do\ not\ use\ .part\ files \-\-no\-mtime\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ do\ not\ use\ the\ Last\-modified\ header\ to\ set \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ the\ file\ modification\ time @@ -184,6 +162,19 @@ redistribute it or use it however you like. \-\-write\-annotations\ \ \ \ \ \ \ \ \ \ \ \ \ \ write\ video\ annotations\ to\ a\ .annotation \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ file \-\-write\-thumbnail\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ write\ thumbnail\ image\ to\ disk +\-\-load\-info\ FILE\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ json\ file\ containing\ the\ video\ information +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ (created\ with\ the\ "\-\-write\-json"\ option) +\-\-cookies\ FILE\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ file\ to\ read\ cookies\ from\ and\ dump\ cookie +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ jar\ in +\-\-cache\-dir\ DIR\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Location\ in\ the\ filesystem\ where\ youtube\-dl +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ can\ store\ some\ downloaded\ information +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ permanently.\ By\ default\ $XDG_CACHE_HOME +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ /youtube\-dl\ or\ ~/.cache/youtube\-dl\ .\ At\ the +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ moment,\ only\ YouTube\ player\ files\ (for +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ videos\ with\ obfuscated\ signatures)\ are +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ cached,\ but\ that\ may\ change. +\-\-no\-cache\-dir\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Disable\ filesystem\ caching +\-\-rm\-cache\-dir\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Delete\ all\ filesystem\ cache\ files \f[] .fi .SS Verbosity / Simulation Options: @@ -218,6 +209,26 @@ redistribute it or use it however you like. \-\-print\-traffic\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Display\ sent\ and\ read\ HTTP\ traffic \f[] .fi +.SS Workarounds: +.IP +.nf +\f[C] +\-\-encoding\ ENCODING\ \ \ \ \ \ \ \ \ \ \ \ \ \ Force\ the\ specified\ encoding\ (experimental) +\-\-no\-check\-certificate\ \ \ \ \ \ \ \ \ \ \ Suppress\ HTTPS\ certificate\ validation. +\-\-prefer\-insecure\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Use\ an\ unencrypted\ connection\ to\ retrieve +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ information\ about\ the\ video.\ (Currently +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ supported\ only\ for\ YouTube) +\-\-user\-agent\ UA\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ specify\ a\ custom\ user\ agent +\-\-referer\ REF\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ specify\ a\ custom\ referer,\ use\ if\ the\ video +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ access\ is\ restricted\ to\ one\ domain +\-\-add\-header\ FIELD:VALUE\ \ \ \ \ \ \ \ \ specify\ a\ custom\ HTTP\ header\ and\ its\ value, +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ separated\ by\ a\ colon\ \[aq]:\[aq].\ You\ can\ use\ this +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ option\ multiple\ times +\-\-bidi\-workaround\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Work\ around\ terminals\ that\ lack +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ bidirectional\ text\ support.\ Requires\ bidiv +\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ or\ fribidi\ executable\ in\ PATH +\f[] +.fi .SS Video Format Options: .IP .nf diff --git a/youtube-dl.bash-completion b/youtube-dl.bash-completion index 498e841..4077a16 100644 --- a/youtube-dl.bash-completion +++ b/youtube-dl.bash-completion @@ -4,7 +4,7 @@ __youtube_dl() COMPREPLY=() cur="${COMP_WORDS[COMP_CWORD]}" prev="${COMP_WORDS[COMP_CWORD-1]}" - opts="--help --version --update --ignore-errors --abort-on-error --dump-user-agent --user-agent --referer --add-header --list-extractors --extractor-descriptions --proxy --no-check-certificate --prefer-insecure --cache-dir --no-cache-dir --socket-timeout --bidi-workaround --default-search --ignore-config --encoding --playlist-start --playlist-end --match-title --reject-title --max-downloads --min-filesize --max-filesize --date --datebefore --dateafter --min-views --max-views --no-playlist --age-limit --download-archive --include-ads --youtube-include-dash-manifest --rate-limit --retries --buffer-size --no-resize-buffer --test --title --id --literal --auto-number --output --autonumber-size --restrict-filenames --batch-file --load-info --no-overwrites --continue --no-continue --cookies --no-part --no-mtime --write-description --write-info-json --write-annotations --write-thumbnail --quiet --no-warnings --simulate --skip-download --get-url --get-title --get-id --get-thumbnail --get-description --get-duration --get-filename --get-format --dump-json --newline --no-progress --console-title --verbose --dump-intermediate-pages --write-pages --youtube-print-sig-code --print-traffic --format --all-formats --prefer-free-formats --max-quality --list-formats --write-sub --write-auto-sub --all-subs --list-subs --sub-format --sub-lang --username --password --netrc --video-password --extract-audio --audio-format --audio-quality --recode-video --keep-video --no-post-overwrites --embed-subs --embed-thumbnail --add-metadata --xattrs --prefer-avconv --prefer-ffmpeg" + opts="--help --version --update --ignore-errors --abort-on-error --dump-user-agent --list-extractors --extractor-descriptions --proxy --socket-timeout --default-search --ignore-config --playlist-start --playlist-end --match-title --reject-title --max-downloads --min-filesize --max-filesize --date --datebefore --dateafter --min-views --max-views --no-playlist --age-limit --download-archive --include-ads --youtube-include-dash-manifest --rate-limit --retries --buffer-size --no-resize-buffer --test --batch-file --id --auto-number --output --autonumber-size --restrict-filenames --title --literal --no-overwrites --continue --no-continue --no-part --no-mtime --write-description --write-info-json --write-annotations --write-thumbnail --load-info --cookies --cache-dir --no-cache-dir --rm-cache-dir --quiet --no-warnings --simulate --skip-download --get-url --get-title --get-id --get-thumbnail --get-description --get-duration --get-filename --get-format --dump-json --newline --no-progress --console-title --verbose --dump-intermediate-pages --write-pages --youtube-print-sig-code --print-traffic --encoding --no-check-certificate --prefer-insecure --user-agent --referer --add-header --bidi-workaround --format --all-formats --prefer-free-formats --max-quality --list-formats --write-sub --write-auto-sub --all-subs --list-subs --sub-format --sub-lang --username --password --netrc --video-password --extract-audio --audio-format --audio-quality --recode-video --keep-video --no-post-overwrites --embed-subs --embed-thumbnail --add-metadata --xattrs --prefer-avconv --prefer-ffmpeg" keywords=":ytfavorites :ytrecommended :ytsubscriptions :ytwatchlater :ythistory" fileopts="-a|--batch-file|--download-archive|--cookies|--load-info" diropts="--cache-dir" diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py deleted file mode 100644 index 5c8e676..0000000 --- a/youtube_dl/FileDownloader.py +++ /dev/null @@ -1,12 +0,0 @@ -# Legacy file for backwards compatibility, use youtube_dl.downloader instead! -from .downloader import FileDownloader as RealFileDownloader -from .downloader import get_suitable_downloader - - -# This class reproduces the old behaviour of FileDownloader -class FileDownloader(RealFileDownloader): - def _do_download(self, filename, info_dict): - real_fd = get_suitable_downloader(info_dict)(self.ydl, self.params) - for ph in self._progress_hooks: - real_fd.add_progress_hook(ph) - return real_fd.download(filename, info_dict) diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 3dff723..14a1d06 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -275,7 +275,7 @@ class YoutubeDL(object): return message assert hasattr(self, '_output_process') - assert type(message) == type('') + assert isinstance(message, compat_str) line_count = message.count('\n') + 1 self._output_process.stdin.write((message + '\n').encode('utf-8')) self._output_process.stdin.flush() @@ -303,7 +303,7 @@ class YoutubeDL(object): def to_stderr(self, message): """Print message to stderr.""" - assert type(message) == type('') + assert isinstance(message, compat_str) if self.params.get('logger'): self.params['logger'].error(message) else: @@ -849,7 +849,7 @@ class YoutubeDL(object): # Keep for backwards compatibility info_dict['stitle'] = info_dict['title'] - if not 'format' in info_dict: + if 'format' not in info_dict: info_dict['format'] = info_dict['ext'] reason = self._match_entry(info_dict) @@ -999,7 +999,7 @@ class YoutubeDL(object): if info_dict.get('requested_formats') is not None: downloaded = [] success = True - merger = FFmpegMergerPP(self) + merger = FFmpegMergerPP(self, not self.params.get('keepvideo')) if not merger._get_executable(): postprocessors = [] self.report_warning('You have requested multiple ' @@ -1197,6 +1197,10 @@ class YoutubeDL(object): if res: res += ', ' res += format_bytes(fdict['filesize']) + elif fdict.get('filesize_approx') is not None: + if res: + res += ', ' + res += '~' + format_bytes(fdict['filesize_approx']) return res def list_formats(self, info_dict): @@ -1230,14 +1234,18 @@ class YoutubeDL(object): if not self.params.get('verbose'): return - write_string( + if type('') is not compat_str: + # Python 2.6 on SLES11 SP1 (https://github.com/rg3/youtube-dl/issues/3326) + self.report_warning( + 'Your Python is broken! Update to a newer and supported version') + + encoding_str = ( '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % ( locale.getpreferredencoding(), sys.getfilesystemencoding(), sys.stdout.encoding, - self.get_encoding()), - encoding=None - ) + self.get_encoding())) + write_string(encoding_str, encoding=None) self._write_string('[debug] youtube-dl version ' + __version__ + '\n') try: diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index 5e16a54..f9c16e4 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -64,18 +64,20 @@ __authors__ = ( 'Adam Malcontenti-Wilson', 'Tobias Bell', 'Naglis Jonaitis', + 'Charles Chen', + 'Hassaan Ali', + 'Dobrosław Å»ybort', ) __license__ = 'Public Domain' import codecs import io -import locale import optparse import os import random -import re import shlex +import shutil import sys @@ -97,7 +99,7 @@ from .utils import ( write_string, ) from .update import update_self -from .FileDownloader import ( +from .downloader import ( FileDownloader, ) from .extractor import gen_extractors @@ -222,6 +224,7 @@ def parseOpts(overrideArguments=None): downloader = optparse.OptionGroup(parser, 'Download Options') postproc = optparse.OptionGroup(parser, 'Post-processing Options') filesystem = optparse.OptionGroup(parser, 'Filesystem Options') + workarounds = optparse.OptionGroup(parser, 'Workarounds') verbosity = optparse.OptionGroup(parser, 'Verbosity / Simulation Options') general.add_option('-h', '--help', @@ -238,14 +241,6 @@ def parseOpts(overrideArguments=None): general.add_option('--dump-user-agent', action='store_true', dest='dump_user_agent', help='display the current browser identification', default=False) - general.add_option('--user-agent', - dest='user_agent', help='specify a custom user agent', metavar='UA') - general.add_option('--referer', - dest='referer', help='specify a custom referer, use if the video access is restricted to one domain', - metavar='REF', default=None) - general.add_option('--add-header', - dest='headers', help='specify a custom HTTP header and its value, separated by a colon \':\'. You can use this option multiple times', action="append", - metavar='FIELD:VALUE') general.add_option('--list-extractors', action='store_true', dest='list_extractors', help='List all supported extractors and the URLs they would handle', default=False) @@ -255,33 +250,17 @@ def parseOpts(overrideArguments=None): general.add_option( '--proxy', dest='proxy', default=None, metavar='URL', help='Use the specified HTTP/HTTPS proxy. Pass in an empty string (--proxy "") for direct connection') - general.add_option('--no-check-certificate', action='store_true', dest='no_check_certificate', default=False, help='Suppress HTTPS certificate validation.') - general.add_option( - '--prefer-insecure', '--prefer-unsecure', action='store_true', dest='prefer_insecure', - help='Use an unencrypted connection to retrieve information about the video. (Currently supported only for YouTube)') - general.add_option( - '--cache-dir', dest='cachedir', default=get_cachedir(), metavar='DIR', - help='Location in the filesystem where youtube-dl can store some downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl . At the moment, only YouTube player files (for videos with obfuscated signatures) are cached, but that may change.') - general.add_option( - '--no-cache-dir', action='store_const', const=None, dest='cachedir', - help='Disable filesystem caching') general.add_option( '--socket-timeout', dest='socket_timeout', type=float, default=None, help=u'Time to wait before giving up, in seconds') - general.add_option( - '--bidi-workaround', dest='bidi_workaround', action='store_true', - help=u'Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH') general.add_option( '--default-search', dest='default_search', metavar='PREFIX', - help='Use this prefix for unqualified URLs. For example "gvsearch2:" downloads two videos from google videos for youtube-dl "large apple". Use the value "auto" to let youtube-dl guess. The default value "error" just throws an error.') + help='Use this prefix for unqualified URLs. For example "gvsearch2:" downloads two videos from google videos for youtube-dl "large apple". Use the value "auto" to let youtube-dl guess ("auto_warning" to emit a warning when guessing). "error" just throws an error. The default value "fixup_error" repairs broken URLs, but emits an error if this is not possible instead of searching.') general.add_option( '--ignore-config', action='store_true', help='Do not read configuration files. When given in the global configuration file /etc/youtube-dl.conf: do not read the user configuration in ~/.config/youtube-dl.conf (%APPDATA%/youtube-dl/config.txt on Windows)') - general.add_option( - '--encoding', dest='encoding', metavar='ENCODING', - help='Force the specified encoding (experimental)') selection.add_option( '--playlist-start', @@ -382,6 +361,33 @@ def parseOpts(overrideArguments=None): help='do not automatically adjust the buffer size. By default, the buffer size is automatically resized from an initial value of SIZE.', default=False) downloader.add_option('--test', action='store_true', dest='test', default=False, help=optparse.SUPPRESS_HELP) + workarounds.add_option( + '--encoding', dest='encoding', metavar='ENCODING', + help='Force the specified encoding (experimental)') + workarounds.add_option( + '--no-check-certificate', action='store_true', + dest='no_check_certificate', default=False, + help='Suppress HTTPS certificate validation.') + workarounds.add_option( + '--prefer-insecure', '--prefer-unsecure', action='store_true', dest='prefer_insecure', + help='Use an unencrypted connection to retrieve information about the video. (Currently supported only for YouTube)') + workarounds.add_option( + '--user-agent', metavar='UA', + dest='user_agent', help='specify a custom user agent') + workarounds.add_option( + '--referer', metavar='REF', + dest='referer', default=None, + help='specify a custom referer, use if the video access is restricted to one domain', + ) + workarounds.add_option( + '--add-header', metavar='FIELD:VALUE', + dest='headers', action='append', + help='specify a custom HTTP header and its value, separated by a colon \':\'. You can use this option multiple times', + ) + workarounds.add_option( + '--bidi-workaround', dest='bidi_workaround', action='store_true', + help=u'Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH') + verbosity.add_option('-q', '--quiet', action='store_true', dest='quiet', help='activates quiet mode', default=False) verbosity.add_option( @@ -439,12 +445,10 @@ def parseOpts(overrideArguments=None): help='Display sent and read HTTP traffic') - filesystem.add_option('-t', '--title', - action='store_true', dest='usetitle', help='use title in file name (default)', default=False) + filesystem.add_option('-a', '--batch-file', + dest='batchfile', metavar='FILE', help='file containing URLs to download (\'-\' for stdin)') filesystem.add_option('--id', action='store_true', dest='useid', help='use only video ID in file name', default=False) - filesystem.add_option('-l', '--literal', - action='store_true', dest='usetitle', help='[deprecated] alias of --title', default=False) filesystem.add_option('-A', '--auto-number', action='store_true', dest='autonumber', help='number downloaded files starting from 00000', default=False) @@ -470,11 +474,10 @@ def parseOpts(overrideArguments=None): filesystem.add_option('--restrict-filenames', action='store_true', dest='restrictfilenames', help='Restrict filenames to only ASCII characters, and avoid "&" and spaces in filenames', default=False) - filesystem.add_option('-a', '--batch-file', - dest='batchfile', metavar='FILE', help='file containing URLs to download (\'-\' for stdin)') - filesystem.add_option('--load-info', - dest='load_info_filename', metavar='FILE', - help='json file containing the video information (created with the "--write-json" option)') + filesystem.add_option('-t', '--title', + action='store_true', dest='usetitle', help='[deprecated] use title in file name (default)', default=False) + filesystem.add_option('-l', '--literal', + action='store_true', dest='usetitle', help='[deprecated] alias of --title', default=False) filesystem.add_option('-w', '--no-overwrites', action='store_true', dest='nooverwrites', help='do not overwrite files', default=False) filesystem.add_option('-c', '--continue', @@ -482,8 +485,6 @@ def parseOpts(overrideArguments=None): filesystem.add_option('--no-continue', action='store_false', dest='continue_dl', help='do not resume partially downloaded files (restart from beginning)') - filesystem.add_option('--cookies', - dest='cookiefile', metavar='FILE', help='file to read cookies from and dump cookie jar in') filesystem.add_option('--no-part', action='store_true', dest='nopart', help='do not use .part files', default=False) filesystem.add_option('--no-mtime', @@ -501,6 +502,20 @@ def parseOpts(overrideArguments=None): filesystem.add_option('--write-thumbnail', action='store_true', dest='writethumbnail', help='write thumbnail image to disk', default=False) + filesystem.add_option('--load-info', + dest='load_info_filename', metavar='FILE', + help='json file containing the video information (created with the "--write-json" option)') + filesystem.add_option('--cookies', + dest='cookiefile', metavar='FILE', help='file to read cookies from and dump cookie jar in') + filesystem.add_option( + '--cache-dir', dest='cachedir', default=get_cachedir(), metavar='DIR', + help='Location in the filesystem where youtube-dl can store some downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl . At the moment, only YouTube player files (for videos with obfuscated signatures) are cached, but that may change.') + filesystem.add_option( + '--no-cache-dir', action='store_const', const=None, dest='cachedir', + help='Disable filesystem caching') + filesystem.add_option( + '--rm-cache-dir', action='store_true', dest='rm_cachedir', + help='Delete all filesystem cache files') postproc.add_option('-x', '--extract-audio', action='store_true', dest='extractaudio', default=False, @@ -534,6 +549,7 @@ def parseOpts(overrideArguments=None): parser.add_option_group(downloader) parser.add_option_group(filesystem) parser.add_option_group(verbosity) + parser.add_option_group(workarounds) parser.add_option_group(video_format) parser.add_option_group(subtitles) parser.add_option_group(authentication) @@ -633,7 +649,7 @@ def _real_main(argv=None): if desc is False: continue if hasattr(ie, 'SEARCH_KEY'): - _SEARCHES = (u'cute kittens', u'slithering pythons', u'falling cat', u'angry poodle', u'purple fish', u'running tortoise') + _SEARCHES = (u'cute kittens', u'slithering pythons', u'falling cat', u'angry poodle', u'purple fish', u'running tortoise', u'sleeping bunny') _COUNTS = (u'', u'5', u'10', u'all') desc += u' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES)) compat_print(desc) @@ -694,7 +710,7 @@ def _real_main(argv=None): date = DateRange.day(opts.date) else: date = DateRange(opts.dateafter, opts.datebefore) - if opts.default_search not in ('auto', 'auto_warning', None) and ':' not in opts.default_search: + if opts.default_search not in ('auto', 'auto_warning', 'error', 'fixup_error', None) and ':' not in opts.default_search: parser.error(u'--default-search invalid; did you forget a colon (:) at the end?') # Do not download videos when there are audio-only formats @@ -833,9 +849,26 @@ def _real_main(argv=None): if opts.update_self: update_self(ydl.to_screen, opts.verbose) + # Remove cache dir + if opts.rm_cachedir: + if opts.cachedir is None: + ydl.to_screen(u'No cache dir specified (Did you combine --no-cache-dir and --rm-cache-dir?)') + else: + if ('.cache' not in opts.cachedir) or ('youtube-dl' not in opts.cachedir): + ydl.to_screen(u'Not removing directory %s - this does not look like a cache dir') + retcode = 141 + else: + ydl.to_screen( + u'Removing cache dir %s .' % opts.cachedir, + skip_eol=True) + if os.path.exists(opts.cachedir): + ydl.to_screen(u'.', skip_eol=True) + shutil.rmtree(opts.cachedir) + ydl.to_screen(u'.') + # Maybe do nothing if (len(all_urls) < 1) and (opts.load_info_filename is None): - if not opts.update_self: + if not (opts.update_self or opts.rm_cachedir): parser.error(u'you must provide at least one URL') else: sys.exit() diff --git a/youtube_dl/downloader/f4m.py b/youtube_dl/downloader/f4m.py index e6be6ae..71353f6 100644 --- a/youtube_dl/downloader/f4m.py +++ b/youtube_dl/downloader/f4m.py @@ -220,6 +220,7 @@ class F4mFD(FileDownloader): def real_download(self, filename, info_dict): man_url = info_dict['url'] + requested_bitrate = info_dict.get('tbr') self.to_screen('[download] Downloading f4m manifest') manifest = self.ydl.urlopen(man_url).read() self.report_destination(filename) @@ -233,8 +234,14 @@ class F4mFD(FileDownloader): doc = etree.fromstring(manifest) formats = [(int(f.attrib.get('bitrate', -1)), f) for f in doc.findall(_add_ns('media'))] - formats = sorted(formats, key=lambda f: f[0]) - rate, media = formats[-1] + if requested_bitrate is None: + # get the best format + formats = sorted(formats, key=lambda f: f[0]) + rate, media = formats[-1] + else: + rate, media = list(filter( + lambda f: int(f[0]) == requested_bitrate, formats))[0] + base_url = compat_urlparse.urljoin(man_url, media.attrib['url']) bootstrap = base64.b64decode(doc.find(_add_ns('bootstrapInfo')).text) metadata = base64.b64decode(media.find(_add_ns('metadata')).text) diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index e49ac3e..66c8737 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -1,5 +1,7 @@ +from .abc import ABCIE from .academicearth import AcademicEarthCourseIE from .addanime import AddAnimeIE +from .adultswim import AdultSwimIE from .aftonbladet import AftonbladetIE from .anitube import AnitubeIE from .aol import AolIE @@ -52,6 +54,7 @@ from .cnn import ( from .collegehumor import CollegeHumorIE from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE from .condenast import CondeNastIE +from .cracked import CrackedIE from .criterion import CriterionIE from .crunchyroll import CrunchyrollIE from .cspan import CSpanIE @@ -62,6 +65,7 @@ from .dailymotion import ( DailymotionUserIE, ) from .daum import DaumIE +from .dfb import DFBIE from .dotsub import DotsubIE from .dreisat import DreiSatIE from .drtv import DRTVIE @@ -108,9 +112,11 @@ from .funnyordie import FunnyOrDieIE from .gamekings import GamekingsIE from .gameone import GameOneIE from .gamespot import GameSpotIE +from .gamestar import GameStarIE from .gametrailers import GametrailersIE from .gdcvault import GDCVaultIE from .generic import GenericIE +from .godtube import GodTubeIE from .googleplus import GooglePlusIE from .googlesearch import GoogleSearchIE from .gorillavid import GorillaVidIE @@ -137,6 +143,7 @@ from .ivi import ( IviIE, IviCompilationIE ) +from .izlesene import IzleseneIE from .jadorecettepub import JadoreCettePubIE from .jeuxvideo import JeuxVideoIE from .jukebox import JukeboxIE @@ -148,6 +155,7 @@ from .khanacademy import KhanAcademyIE from .kickstarter import KickStarterIE from .keek import KeekIE from .kontrtube import KontrTubeIE +from .krasview import KrasViewIE from .ku6 import Ku6IE from .la7 import LA7IE from .lifenews import LifeNewsIE @@ -170,6 +178,7 @@ from .metacafe import MetacafeIE from .metacritic import MetacriticIE from .mit import TechTVMITIE, MITIE, OCWMITIE from .mixcloud import MixcloudIE +from .mlb import MLBIE from .mpora import MporaIE from .mofosex import MofosexIE from .mooshare import MooshareIE @@ -249,10 +258,12 @@ from .rutube import ( RutubePersonIE, ) from .rutv import RUTVIE +from .sapo import SapoIE from .savefrom import SaveFromIE from .scivee import SciVeeIE from .screencast import ScreencastIE from .servingsys import ServingSysIE +from .shared import SharedIE from .sina import SinaIE from .slideshare import SlideshareIE from .slutload import SlutloadIE @@ -262,6 +273,8 @@ from .smotri import ( SmotriUserIE, SmotriBroadcastIE, ) +from .snotr import SnotrIE +from .sockshare import SockshareIE from .sohu import SohuIE from .soundcloud import ( SoundcloudIE, @@ -313,6 +326,8 @@ from .tumblr import TumblrIE from .tutv import TutvIE from .tvigle import TvigleIE from .tvp import TvpIE +from .tvplay import TVPlayIE +from .ubu import UbuIE from .udemy import ( UdemyIE, UdemyCourseIE @@ -334,6 +349,7 @@ from .videofyme import VideofyMeIE from .videopremium import VideoPremiumIE from .videott import VideoTtIE from .videoweed import VideoWeedIE +from .vidme import VidmeIE from .vimeo import ( VimeoIE, VimeoChannelIE, @@ -396,6 +412,7 @@ from .youtube import ( YoutubeUserIE, YoutubeWatchLaterIE, ) + from .zdf import ZDFIE diff --git a/youtube_dl/extractor/abc.py b/youtube_dl/extractor/abc.py new file mode 100644 index 0000000..7d89f44 --- /dev/null +++ b/youtube_dl/extractor/abc.py @@ -0,0 +1,48 @@ +from __future__ import unicode_literals + +import re +import json + +from .common import InfoExtractor + + +class ABCIE(InfoExtractor): + IE_NAME = 'abc.net.au' + _VALID_URL = r'http://www\.abc\.net\.au/news/[^/]+/[^/]+/(?P\d+)' + + _TEST = { + 'url': 'http://www.abc.net.au/news/2014-07-25/bringing-asylum-seekers-to-australia-would-give/5624716', + 'md5': 'dad6f8ad011a70d9ddf887ce6d5d0742', + 'info_dict': { + 'id': '5624716', + 'ext': 'mp4', + 'title': 'Bringing asylum seekers to Australia would give them right to asylum claims: professor', + 'description': 'md5:ba36fa5e27e5c9251fd929d339aea4af', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + webpage = self._download_webpage(url, video_id) + + urls_info_json = self._search_regex( + r'inlineVideoData\.push\((.*?)\);', webpage, 'video urls', + flags=re.DOTALL) + urls_info = json.loads(urls_info_json.replace('\'', '"')) + formats = [{ + 'url': url_info['url'], + 'width': int(url_info['width']), + 'height': int(url_info['height']), + 'tbr': int(url_info['bitrate']), + 'filesize': int(url_info['filesize']), + } for url_info in urls_info] + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': self._og_search_title(webpage), + 'formats': formats, + 'description': self._og_search_description(webpage), + 'thumbnail': self._og_search_thumbnail(webpage), + } diff --git a/youtube_dl/extractor/adultswim.py b/youtube_dl/extractor/adultswim.py new file mode 100644 index 0000000..a00bfcb --- /dev/null +++ b/youtube_dl/extractor/adultswim.py @@ -0,0 +1,139 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor + +class AdultSwimIE(InfoExtractor): + _VALID_URL = r'https?://video\.adultswim\.com/(?P.+?)(?:\.html)?(?:\?.*)?(?:#.*)?$' + _TEST = { + 'url': 'http://video.adultswim.com/rick-and-morty/close-rick-counters-of-the-rick-kind.html?x=y#title', + 'playlist': [ + { + 'md5': '4da359ec73b58df4575cd01a610ba5dc', + 'info_dict': { + 'id': '8a250ba1450996e901453d7f02ca02f5', + 'ext': 'flv', + 'title': 'Rick and Morty Close Rick-Counters of the Rick Kind part 1', + 'description': 'Rick has a run in with some old associates, resulting in a fallout with Morty. You got any chips, broh?', + 'uploader': 'Rick and Morty', + 'thumbnail': 'http://i.cdn.turner.com/asfix/repository/8a250ba13f865824013fc9db8b6b0400/thumbnail_267549017116827057.jpg' + } + }, + { + 'md5': 'ffbdf55af9331c509d95350bd0cc1819', + 'info_dict': { + 'id': '8a250ba1450996e901453d7f4bd102f6', + 'ext': 'flv', + 'title': 'Rick and Morty Close Rick-Counters of the Rick Kind part 2', + 'description': 'Rick has a run in with some old associates, resulting in a fallout with Morty. You got any chips, broh?', + 'uploader': 'Rick and Morty', + 'thumbnail': 'http://i.cdn.turner.com/asfix/repository/8a250ba13f865824013fc9db8b6b0400/thumbnail_267549017116827057.jpg' + } + }, + { + 'md5': 'b92409635540304280b4b6c36bd14a0a', + 'info_dict': { + 'id': '8a250ba1450996e901453d7fa73c02f7', + 'ext': 'flv', + 'title': 'Rick and Morty Close Rick-Counters of the Rick Kind part 3', + 'description': 'Rick has a run in with some old associates, resulting in a fallout with Morty. You got any chips, broh?', + 'uploader': 'Rick and Morty', + 'thumbnail': 'http://i.cdn.turner.com/asfix/repository/8a250ba13f865824013fc9db8b6b0400/thumbnail_267549017116827057.jpg' + } + }, + { + 'md5': 'e8818891d60e47b29cd89d7b0278156d', + 'info_dict': { + 'id': '8a250ba1450996e901453d7fc8ba02f8', + 'ext': 'flv', + 'title': 'Rick and Morty Close Rick-Counters of the Rick Kind part 4', + 'description': 'Rick has a run in with some old associates, resulting in a fallout with Morty. You got any chips, broh?', + 'uploader': 'Rick and Morty', + 'thumbnail': 'http://i.cdn.turner.com/asfix/repository/8a250ba13f865824013fc9db8b6b0400/thumbnail_267549017116827057.jpg' + } + } + ] + } + + _video_extensions = { + '3500': 'flv', + '640': 'mp4', + '150': 'mp4', + 'ipad': 'm3u8', + 'iphone': 'm3u8' + } + _video_dimensions = { + '3500': (1280, 720), + '640': (480, 270), + '150': (320, 180) + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_path = mobj.group('path') + + webpage = self._download_webpage(url, video_path) + episode_id = self._html_search_regex(r'', webpage, 'episode_id') + title = self._og_search_title(webpage) + + index_url = 'http://asfix.adultswim.com/asfix-svc/episodeSearch/getEpisodesByIDs?networkName=AS&ids=%s' % episode_id + idoc = self._download_xml(index_url, title, 'Downloading episode index', 'Unable to download episode index') + + episode_el = idoc.find('.//episode') + show_title = episode_el.attrib.get('collectionTitle') + episode_title = episode_el.attrib.get('title') + thumbnail = episode_el.attrib.get('thumbnailUrl') + description = episode_el.find('./description').text.strip() + + entries = [] + segment_els = episode_el.findall('./segments/segment') + + for part_num, segment_el in enumerate(segment_els): + segment_id = segment_el.attrib.get('id') + segment_title = '%s %s part %d' % (show_title, episode_title, part_num + 1) + thumbnail = segment_el.attrib.get('thumbnailUrl') + duration = segment_el.attrib.get('duration') + + segment_url = 'http://asfix.adultswim.com/asfix-svc/episodeservices/getCvpPlaylist?networkName=AS&id=%s' % segment_id + idoc = self._download_xml(segment_url, segment_title, 'Downloading segment information', 'Unable to download segment information') + + formats = [] + file_els = idoc.findall('.//files/file') + + for file_el in file_els: + bitrate = file_el.attrib.get('bitrate') + type = file_el.attrib.get('type') + width, height = self._video_dimensions.get(bitrate, (None, None)) + formats.append({ + 'format_id': '%s-%s' % (bitrate, type), + 'url': file_el.text, + 'ext': self._video_extensions.get(bitrate, 'mp4'), + # The bitrate may not be a number (for example: 'iphone') + 'tbr': int(bitrate) if bitrate.isdigit() else None, + 'height': height, + 'width': width + }) + + self._sort_formats(formats) + + entries.append({ + 'id': segment_id, + 'title': segment_title, + 'formats': formats, + 'uploader': show_title, + 'thumbnail': thumbnail, + 'duration': duration, + 'description': description + }) + + return { + '_type': 'playlist', + 'id': episode_id, + 'display_id': video_path, + 'entries': entries, + 'title': '%s %s' % (show_title, episode_title), + 'description': description, + 'thumbnail': thumbnail + } diff --git a/youtube_dl/extractor/allocine.py b/youtube_dl/extractor/allocine.py index 34f0cd4..7bd7978 100644 --- a/youtube_dl/extractor/allocine.py +++ b/youtube_dl/extractor/allocine.py @@ -32,7 +32,7 @@ class AllocineIE(InfoExtractor): 'id': '19540403', 'ext': 'mp4', 'title': 'Planes 2 Bande-annonce VF', - 'description': 'md5:c4b1f7bd682a91de6491ada267ec0f4d', + 'description': 'md5:eeaffe7c2d634525e21159b93acf3b1e', 'thumbnail': 're:http://.*\.jpg', }, }, { @@ -42,7 +42,7 @@ class AllocineIE(InfoExtractor): 'id': '19544709', 'ext': 'mp4', 'title': 'Dragons 2 - Bande annonce finale VF', - 'description': 'md5:e74a4dc750894bac300ece46c7036490', + 'description': 'md5:71742e3a74b0d692c7fce0dd2017a4ac', 'thumbnail': 're:http://.*\.jpg', }, }] diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py index b36a4d4..957bdef 100644 --- a/youtube_dl/extractor/ard.py +++ b/youtube_dl/extractor/ard.py @@ -7,23 +7,34 @@ from .common import InfoExtractor from ..utils import ( determine_ext, ExtractorError, + qualities, + compat_urllib_parse_urlparse, + compat_urllib_parse, ) class ARDIE(InfoExtractor): - _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P[^/\?]+)(?:\?.*)?' + _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?' - _TEST = { - 'url': 'http://www.ardmediathek.de/das-erste/guenther-jauch/edward-snowden-im-interview-held-oder-verraeter?documentId=19288786', - 'file': '19288786.mp4', - 'md5': '515bf47ce209fb3f5a61b7aad364634c', + _TESTS = [{ + 'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht', + 'file': '22429276.mp4', + 'md5': '469751912f1de0816a9fc9df8336476c', 'info_dict': { - 'title': 'Edward Snowden im Interview - Held oder Verräter?', - 'description': 'Edward Snowden hat alles aufs Spiel gesetzt, um die weltweite \xdcberwachung durch die Geheimdienste zu enttarnen. Nun stellt sich der ehemalige NSA-Mitarbeiter erstmals weltweit in einem TV-Interview den Fragen eines NDR-Journalisten. Die Sendung vom Sonntagabend.', - 'thumbnail': 'http://www.ardmediathek.de/ard/servlet/contentblob/19/28/87/90/19288790/bild/2250037', + 'title': 'Vertrauen ist gut, Spionieren ist besser - Geht so deutsch-amerikanische Freundschaft?', + 'description': 'Das Erste Mediathek [ARD]: Vertrauen ist gut, Spionieren ist besser - Geht so deutsch-amerikanische Freundschaft?, Anne Will, Über die Spionage-Affäre diskutieren Clemens Binninger, Katrin Göring-Eckardt, Georg Mascolo, Andrew B. Denison und Constanze Kurz.. Das Video zur Sendung Anne Will am Mittwoch, 16.07.2014', }, 'skip': 'Blocked outside of Germany', - } + }, { + 'url': 'http://www.ardmediathek.de/tv/Tatort/Das-Wunder-von-Wolbeck-Video-tgl-ab-20/Das-Erste/Video?documentId=22490580&bcastId=602916', + 'info_dict': { + 'id': '22490580', + 'ext': 'mp4', + 'title': 'Das Wunder von Wolbeck (Video tgl. ab 20 Uhr)', + 'description': 'Auf einem restaurierten Hof bei Wolbeck wird der Heilpraktiker Raffael Lembeck eines morgens von seiner Frau Stella tot aufgefunden. Das Opfer war offensichtlich in seiner Praxis zu Fall gekommen und ist dann verblutet, erklärt Prof. Boerne am Tatort.', + }, + 'skip': 'Blocked outside of Germany', + }] def _real_extract(self, url): # determine video id from url @@ -35,6 +46,9 @@ class ARDIE(InfoExtractor): else: video_id = m.group('video_id') + urlp = compat_urllib_parse_urlparse(url) + url = urlp._replace(path=compat_urllib_parse.quote(urlp.path.encode('utf-8'))).geturl() + webpage = self._download_webpage(url, video_id) title = self._html_search_regex( @@ -43,40 +57,64 @@ class ARDIE(InfoExtractor): r'

(.*?)

'], webpage, 'title') description = self._html_search_meta( - 'dcterms.abstract', webpage, 'description') - thumbnail = self._og_search_thumbnail(webpage) - - - media_info = self._download_json( - 'http://www.ardmediathek.de/play/media/%s' % video_id, video_id) - # The second element of the _mediaArray contains the standard http urls - streams = media_info['_mediaArray'][1]['_mediaStreamArray'] - if not streams: - if '"fsk"' in webpage: - raise ExtractorError('This video is only available after 20:00') - - formats = [] - - for s in streams: - if type(s['_stream']) == list: - for index, url in enumerate(s['_stream'][::-1]): - quality = s['_quality'] + index - formats.append({ - 'quality': quality, - 'url': url, - 'format_id': '%s-%s' % (determine_ext(url), quality) + 'dcterms.abstract', webpage, 'description', default=None) + if description is None: + description = self._html_search_meta( + 'description', webpage, 'meta description') + + # Thumbnail is sometimes not present. + # It is in the mobile version, but that seems to use a different URL + # structure altogether. + thumbnail = self._og_search_thumbnail(webpage, default=None) + + media_streams = re.findall(r'''(?x) + mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s* + "([^"]+)"''', webpage) + + if media_streams: + QUALITIES = qualities(['lo', 'hi', 'hq']) + formats = [] + for furl in set(media_streams): + if furl.endswith('.f4m'): + fid = 'f4m' + else: + fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl) + fid = fid_m.group(1) if fid_m else None + formats.append({ + 'quality': QUALITIES(fid), + 'format_id': fid, + 'url': furl, + }) + else: # request JSON file + media_info = self._download_json( + 'http://www.ardmediathek.de/play/media/%s' % video_id, video_id) + # The second element of the _mediaArray contains the standard http urls + streams = media_info['_mediaArray'][1]['_mediaStreamArray'] + if not streams: + if '"fsk"' in webpage: + raise ExtractorError('This video is only available after 20:00') + + formats = [] + for s in streams: + if type(s['_stream']) == list: + for index, url in enumerate(s['_stream'][::-1]): + quality = s['_quality'] + index + formats.append({ + 'quality': quality, + 'url': url, + 'format_id': '%s-%s' % (determine_ext(url), quality) }) - continue + continue - format = { - 'quality': s['_quality'], - 'url': s['_stream'], - } + format = { + 'quality': s['_quality'], + 'url': s['_stream'], + } - format['format_id'] = '%s-%s' % ( - determine_ext(format['url']), format['quality']) + format['format_id'] = '%s-%s' % ( + determine_ext(format['url']), format['quality']) - formats.append(format) + formats.append(format) self._sort_formats(formats) diff --git a/youtube_dl/extractor/blinkx.py b/youtube_dl/extractor/blinkx.py index 7d558e2..3e461e7 100644 --- a/youtube_dl/extractor/blinkx.py +++ b/youtube_dl/extractor/blinkx.py @@ -52,7 +52,7 @@ class BlinkxIE(InfoExtractor): 'height': int(m['h']), }) elif m['type'] == 'original': - duration = m['d'] + duration = float(m['d']) elif m['type'] == 'youtube': yt_id = m['link'] self.to_screen('Youtube video detected: %s' % yt_id) diff --git a/youtube_dl/extractor/bloomberg.py b/youtube_dl/extractor/bloomberg.py index 25fb79e..c51a97c 100644 --- a/youtube_dl/extractor/bloomberg.py +++ b/youtube_dl/extractor/bloomberg.py @@ -10,7 +10,7 @@ class BloombergIE(InfoExtractor): _TEST = { 'url': 'http://www.bloomberg.com/video/shah-s-presentation-on-foreign-exchange-strategies-qurhIVlJSB6hzkVi229d8g.html', - 'md5': '7bf08858ff7c203c870e8a6190e221e5', + # The md5 checksum changes 'info_dict': { 'id': 'qurhIVlJSB6hzkVi229d8g', 'ext': 'flv', @@ -31,8 +31,7 @@ class BloombergIE(InfoExtractor): return { 'id': name.split('-')[-1], 'title': title, - 'url': f4m_url, - 'ext': 'flv', + 'formats': self._extract_f4m_formats(f4m_url, name), 'description': self._og_search_description(webpage), 'thumbnail': self._og_search_thumbnail(webpage), } diff --git a/youtube_dl/extractor/br.py b/youtube_dl/extractor/br.py index 9933607..86f0c28 100644 --- a/youtube_dl/extractor/br.py +++ b/youtube_dl/extractor/br.py @@ -7,12 +7,13 @@ from .common import InfoExtractor from ..utils import ( ExtractorError, int_or_none, + parse_duration, ) class BRIE(InfoExtractor): IE_DESC = 'Bayerischer Rundfunk Mediathek' - _VALID_URL = r'https?://(?:www\.)?br\.de/(?:[a-z0-9\-]+/)+(?P[a-z0-9\-]+)\.html' + _VALID_URL = r'https?://(?:www\.)?br\.de/(?:[a-z0-9\-_]+/)+(?P[a-z0-9\-_]+)\.html' _BASE_URL = 'http://www.br.de' _TESTS = [ @@ -22,8 +23,9 @@ class BRIE(InfoExtractor): 'info_dict': { 'id': '25e279aa-1ffd-40fd-9955-5325bd48a53a', 'ext': 'mp4', - 'title': 'Am 1. und 2. August in Oberammergau', - 'description': 'md5:dfd224e5aa6819bc1fcbb7826a932021', + 'title': 'Wenn das Traditions-Theater wackelt', + 'description': 'Heimatsound-Festival 2014: Wenn das Traditions-Theater wackelt', + 'duration': 34, } }, { @@ -34,6 +36,7 @@ class BRIE(InfoExtractor): 'ext': 'mp4', 'title': 'Über den Pass', 'description': 'Die Eroberung der Alpen: Über den Pass', + 'duration': 2588, } }, { @@ -44,6 +47,7 @@ class BRIE(InfoExtractor): 'ext': 'aac', 'title': '"Keine neuen Schulden im nächsten Jahr"', 'description': 'Haushaltsentwurf: "Keine neuen Schulden im nächsten Jahr"', + 'duration': 64, } }, { @@ -54,6 +58,7 @@ class BRIE(InfoExtractor): 'ext': 'mp4', 'title': 'Umweltbewusster Häuslebauer', 'description': 'Uwe Erdelt: Umweltbewusster Häuslebauer', + 'duration': 116, } }, { @@ -64,6 +69,7 @@ class BRIE(InfoExtractor): 'ext': 'mp4', 'title': 'Folge 1 - Metaphysik', 'description': 'Kant für Anfänger: Folge 1 - Metaphysik', + 'duration': 893, 'uploader': 'Eva Maria Steimle', 'upload_date': '20140117', } @@ -84,6 +90,7 @@ class BRIE(InfoExtractor): media = { 'id': xml_media.get('externalId'), 'title': xml_media.find('title').text, + 'duration': parse_duration(xml_media.find('duration').text), 'formats': self._extract_formats(xml_media.find('assets')), 'thumbnails': self._extract_thumbnails(xml_media.find('teaserImage/variants')), 'description': ' '.join(xml_media.find('shareTitle').text.splitlines()), diff --git a/youtube_dl/extractor/cbs.py b/youtube_dl/extractor/cbs.py index ac03158..822f9a7 100644 --- a/youtube_dl/extractor/cbs.py +++ b/youtube_dl/extractor/cbs.py @@ -1,24 +1,42 @@ +from __future__ import unicode_literals + import re from .common import InfoExtractor class CBSIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?cbs\.com/shows/[^/]+/video/(?P[^/]+)/.*' + _VALID_URL = r'https?://(?:www\.)?cbs\.com/shows/[^/]+/(?:video|artist)/(?P[^/]+)/.*' - _TEST = { - u'url': u'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/', - u'file': u'4JUVEwq3wUT7.flv', - u'info_dict': { - u'title': u'Connect Chat feat. Garth Brooks', - u'description': u'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!', - u'duration': 1495, + _TESTS = [{ + 'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/', + 'info_dict': { + 'id': '4JUVEwq3wUT7', + 'ext': 'flv', + 'title': 'Connect Chat feat. Garth Brooks', + 'description': 'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!', + 'duration': 1495, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + '_skip': 'Blocked outside the US', + }, { + 'url': 'http://www.cbs.com/shows/liveonletterman/artist/221752/st-vincent/', + 'info_dict': { + 'id': 'P9gjWjelt6iP', + 'ext': 'flv', + 'title': 'Live on Letterman - St. Vincent', + 'description': 'Live On Letterman: St. Vincent in concert from New York\'s Ed Sullivan Theater on Tuesday, July 16, 2014.', + 'duration': 3221, }, - u'params': { + 'params': { # rtmp download - u'skip_download': True, + 'skip_download': True, }, - } + '_skip': 'Blocked outside the US', + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -26,5 +44,5 @@ class CBSIE(InfoExtractor): webpage = self._download_webpage(url, video_id) real_id = self._search_regex( r"video\.settings\.pid\s*=\s*'([^']+)';", - webpage, u'real video ID') + webpage, 'real video ID') return self.url_result(u'theplatform:%s' % real_id) diff --git a/youtube_dl/extractor/chilloutzone.py b/youtube_dl/extractor/chilloutzone.py index 02d5ba5..a62395d 100644 --- a/youtube_dl/extractor/chilloutzone.py +++ b/youtube_dl/extractor/chilloutzone.py @@ -42,7 +42,7 @@ class ChilloutzoneIE(InfoExtractor): 'id': '85523671', 'ext': 'mp4', 'title': 'The Sunday Times - Icons', - 'description': 'md5:3e1c0dc6047498d6728dcdaad0891762', + 'description': 'md5:a5f7ff82e2f7a9ed77473fe666954e84', 'uploader': 'Us', 'uploader_id': 'usfilms', 'upload_date': '20140131' diff --git a/youtube_dl/extractor/cnet.py b/youtube_dl/extractor/cnet.py index a94f425..710d500 100644 --- a/youtube_dl/extractor/cnet.py +++ b/youtube_dl/extractor/cnet.py @@ -43,7 +43,11 @@ class CNETIE(InfoExtractor): raise ExtractorError('Cannot find video data') video_id = vdata['id'] - title = vdata['headline'] + title = vdata.get('headline') + if title is None: + title = vdata.get('title') + if title is None: + raise ExtractorError('Cannot find title!') description = vdata.get('dek') thumbnail = vdata.get('image', {}).get('path') author = vdata.get('author') diff --git a/youtube_dl/extractor/comedycentral.py b/youtube_dl/extractor/comedycentral.py index 8af0aba..c81ce5a 100644 --- a/youtube_dl/extractor/comedycentral.py +++ b/youtube_dl/extractor/comedycentral.py @@ -14,13 +14,13 @@ from ..utils import ( class ComedyCentralIE(MTVServicesInfoExtractor): - _VALID_URL = r'''(?x)https?://(?:www\.)?(comedycentral|cc)\.com/ - (video-clips|episodes|cc-studios|video-collections) + _VALID_URL = r'''(?x)https?://(?:www\.)?cc\.com/ + (video-clips|episodes|cc-studios|video-collections|full-episodes) /(?P.*)''' _FEED_URL = 'http://comedycentral.com/feeds/mrss/' _TEST = { - 'url': 'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother', + 'url': 'http://www.cc.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother', 'md5': 'c4f48e9eda1b16dd10add0744344b6d8', 'info_dict': { 'id': 'cef0cbb3-e776-4bc9-b62e-8016deccb354', diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index e686573..342bfb8 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -18,6 +18,7 @@ from ..utils import ( clean_html, compiled_regex_type, ExtractorError, + int_or_none, RegexNotFoundError, sanitize_filename, unescapeHTML, @@ -69,6 +70,7 @@ class InfoExtractor(object): * vcodec Name of the video codec in use * container Name of the container format * filesize The number of bytes, if known in advance + * filesize_approx An estimate for the number of bytes * player_url SWF Player URL (used for rtmpdump). * protocol The protocol that will be used for the actual download, lower-case. @@ -300,8 +302,12 @@ class InfoExtractor(object): def _download_json(self, url_or_request, video_id, note=u'Downloading JSON metadata', errnote=u'Unable to download JSON metadata', - transform_source=None): - json_string = self._download_webpage(url_or_request, video_id, note, errnote) + transform_source=None, + fatal=True): + json_string = self._download_webpage( + url_or_request, video_id, note, errnote, fatal=fatal) + if (not fatal) and json_string is False: + return None if transform_source: json_string = transform_source(json_string) try: @@ -368,7 +374,8 @@ class InfoExtractor(object): else: for p in pattern: mobj = re.search(p, string, flags) - if mobj: break + if mobj: + break if os.name != 'nt' and sys.stderr.isatty(): _name = u'\033[0;34m%s\033[0m' % name @@ -468,7 +475,7 @@ class InfoExtractor(object): display_name = name return self._html_search_regex( r'''(?ix)<meta - (?=[^>]+(?:itemprop|name|property)=["\']%s["\']) + (?=[^>]+(?:itemprop|name|property)=["\']?%s["\']?) [^>]+content=["\']([^"\']+)["\']''' % re.escape(name), html, display_name, fatal=fatal, **kwargs) @@ -555,6 +562,7 @@ class InfoExtractor(object): f.get('abr') if f.get('abr') is not None else -1, audio_ext_preference, f.get('filesize') if f.get('filesize') is not None else -1, + f.get('filesize_approx') if f.get('filesize_approx') is not None else -1, f.get('format_id'), ) formats.sort(key=_formats_key) @@ -583,6 +591,24 @@ class InfoExtractor(object): self.to_screen(msg) time.sleep(timeout) + def _extract_f4m_formats(self, manifest_url, video_id): + manifest = self._download_xml( + manifest_url, video_id, 'Downloading f4m manifest', + 'Unable to download f4m manifest') + + formats = [] + for media_el in manifest.findall('{http://ns.adobe.com/f4m/1.0}media'): + formats.append({ + 'url': manifest_url, + 'ext': 'flv', + 'tbr': int_or_none(media_el.attrib.get('bitrate')), + 'width': int_or_none(media_el.attrib.get('width')), + 'height': int_or_none(media_el.attrib.get('height')), + }) + self._sort_formats(formats) + + return formats + class SearchInfoExtractor(InfoExtractor): """ diff --git a/youtube_dl/extractor/cracked.py b/youtube_dl/extractor/cracked.py new file mode 100644 index 0000000..74b880f --- /dev/null +++ b/youtube_dl/extractor/cracked.py @@ -0,0 +1,65 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + parse_iso8601, + str_to_int, +) + + +class CrackedIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?cracked\.com/video_(?P<id>\d+)_[\da-z-]+\.html' + _TEST = { + 'url': 'http://www.cracked.com/video_19006_4-plot-holes-you-didnt-notice-in-your-favorite-movies.html', + 'md5': '4b29a5eeec292cd5eca6388c7558db9e', + 'info_dict': { + 'id': '19006', + 'ext': 'mp4', + 'title': '4 Plot Holes You Didn\'t Notice in Your Favorite Movies', + 'description': 'md5:3b909e752661db86007d10e5ec2df769', + 'timestamp': 1405659600, + 'upload_date': '20140718', + } + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + + video_url = self._html_search_regex( + [r'var\s+CK_vidSrc\s*=\s*"([^"]+)"', r'<video\s+src="([^"]+)"'], webpage, 'video URL') + + title = self._og_search_title(webpage) + description = self._og_search_description(webpage) + + timestamp = self._html_search_regex(r'<time datetime="([^"]+)"', webpage, 'upload date', fatal=False) + if timestamp: + timestamp = parse_iso8601(timestamp[:-6]) + + view_count = str_to_int(self._html_search_regex( + r'<span class="views" id="viewCounts">([\d,\.]+) Views</span>', webpage, 'view count', fatal=False)) + comment_count = str_to_int(self._html_search_regex( + r'<span id="commentCounts">([\d,\.]+)</span>', webpage, 'comment count', fatal=False)) + + m = re.search(r'_(?P<width>\d+)X(?P<height>\d+)\.mp4$', video_url) + if m: + width = int(m.group('width')) + height = int(m.group('height')) + else: + width = height = None + + return { + 'id': video_id, + 'url':video_url, + 'title': title, + 'description': description, + 'timestamp': timestamp, + 'view_count': view_count, + 'comment_count': comment_count, + 'height': height, + 'width': width, + } \ No newline at end of file diff --git a/youtube_dl/extractor/dfb.py b/youtube_dl/extractor/dfb.py new file mode 100644 index 0000000..cb8e068 --- /dev/null +++ b/youtube_dl/extractor/dfb.py @@ -0,0 +1,44 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor + + +class DFBIE(InfoExtractor): + IE_NAME = 'tv.dfb.de' + _VALID_URL = r'https?://tv\.dfb\.de/video/[^/]+/(?P<id>\d+)' + + _TEST = { + 'url': 'http://tv.dfb.de/video/highlights-des-empfangs-in-berlin/9070/', + # The md5 is different each time + 'info_dict': { + 'id': '9070', + 'ext': 'flv', + 'title': 'Highlights des Empfangs in Berlin', + 'upload_date': '20140716', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + player_info = self._download_xml( + 'http://tv.dfb.de/server/hd_video.php?play=%s' % video_id, + video_id) + video_info = player_info.find('video') + + f4m_info = self._download_xml(video_info.find('url').text, video_id) + token_el = f4m_info.find('token') + manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth'] + '&hdcore=3.2.0' + + return { + 'id': video_id, + 'title': video_info.find('title').text, + 'url': manifest_url, + 'ext': 'flv', + 'thumbnail': self._og_search_thumbnail(webpage), + 'upload_date': ''.join(video_info.find('time_date').text.split('.')[::-1]), + } diff --git a/youtube_dl/extractor/dropbox.py b/youtube_dl/extractor/dropbox.py index 41208c9..9f569aa 100644 --- a/youtube_dl/extractor/dropbox.py +++ b/youtube_dl/extractor/dropbox.py @@ -5,24 +5,26 @@ import os.path import re from .common import InfoExtractor +from ..utils import compat_urllib_parse_unquote class DropboxIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?dropbox[.]com/s/(?P<id>[a-zA-Z0-9]{15})/(?P<title>[^?#]*)' _TEST = { - 'url': 'https://www.dropbox.com/s/0qr9sai2veej4f8/THE_DOCTOR_GAMES.mp4', - 'md5': '8ae17c51172fb7f93bdd6a214cc8c896', + 'url': 'https://www.dropbox.com/s/nelirfsxnmcfbfh/youtube-dl%20test%20video%20%27%C3%A4%22BaW_jenozKc.mp4', + 'md5': '8a3d905427a6951ccb9eb292f154530b', 'info_dict': { - 'id': '0qr9sai2veej4f8', + 'id': 'nelirfsxnmcfbfh', 'ext': 'mp4', - 'title': 'THE_DOCTOR_GAMES' + 'title': 'youtube-dl test video \'ä"BaW_jenozKc' } } def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') - title = os.path.splitext(mobj.group('title'))[0] + fn = compat_urllib_parse_unquote(mobj.group('title')) + title = os.path.splitext(fn)[0] video_url = url + '?dl=1' return { diff --git a/youtube_dl/extractor/firedrive.py b/youtube_dl/extractor/firedrive.py index d26145d..6d73c8a 100644 --- a/youtube_dl/extractor/firedrive.py +++ b/youtube_dl/extractor/firedrive.py @@ -8,7 +8,6 @@ from ..utils import ( ExtractorError, compat_urllib_parse, compat_urllib_request, - determine_ext, ) diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index f3e0f38..1b0e8e5 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -19,17 +19,35 @@ class FranceTVBaseInfoExtractor(InfoExtractor): + video_id, video_id, 'Downloading XML config') manifest_url = info.find('videos/video/url').text - video_url = manifest_url.replace('manifest.f4m', 'index_2_av.m3u8') - video_url = video_url.replace('/z/', '/i/') + manifest_url = manifest_url.replace('/z/', '/i/') + + if manifest_url.startswith('rtmp'): + formats = [{'url': manifest_url, 'ext': 'flv'}] + else: + formats = [] + available_formats = self._search_regex(r'/[^,]*,(.*?),k\.mp4', manifest_url, 'available formats') + for index, format_descr in enumerate(available_formats.split(',')): + format_info = { + 'url': manifest_url.replace('manifest.f4m', 'index_%d_av.m3u8' % index), + 'ext': 'mp4', + } + m_resolution = re.search(r'(?P<width>\d+)x(?P<height>\d+)', format_descr) + if m_resolution is not None: + format_info.update({ + 'width': int(m_resolution.group('width')), + 'height': int(m_resolution.group('height')), + }) + formats.append(format_info) + thumbnail_path = info.find('image').text - return {'id': video_id, - 'ext': 'flv' if video_url.startswith('rtmp') else 'mp4', - 'url': video_url, - 'title': info.find('titre').text, - 'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', thumbnail_path), - 'description': info.find('synopsis').text, - } + return { + 'id': video_id, + 'title': info.find('titre').text, + 'formats': formats, + 'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', thumbnail_path), + 'description': info.find('synopsis').text, + } class PluzzIE(FranceTVBaseInfoExtractor): @@ -48,7 +66,7 @@ class PluzzIE(FranceTVBaseInfoExtractor): class FranceTvInfoIE(FranceTVBaseInfoExtractor): IE_NAME = 'francetvinfo.fr' - _VALID_URL = r'https?://www\.francetvinfo\.fr/.*/(?P<title>.+)\.html' + _VALID_URL = r'https?://(?:www|mobile)\.francetvinfo\.fr/.*/(?P<title>.+)\.html' _TESTS = [{ 'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html', @@ -211,7 +229,7 @@ class GenerationQuoiIE(InfoExtractor): class CultureboxIE(FranceTVBaseInfoExtractor): IE_NAME = 'culturebox.francetvinfo.fr' - _VALID_URL = r'https?://culturebox\.francetvinfo\.fr/(?P<name>.*?)(\?|$)' + _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?P<name>.*?)(\?|$)' _TEST = { 'url': 'http://culturebox.francetvinfo.fr/einstein-on-the-beach-au-theatre-du-chatelet-146813', diff --git a/youtube_dl/extractor/funnyordie.py b/youtube_dl/extractor/funnyordie.py index 6e6b666..721e5fc 100644 --- a/youtube_dl/extractor/funnyordie.py +++ b/youtube_dl/extractor/funnyordie.py @@ -26,7 +26,7 @@ class FunnyOrDieIE(InfoExtractor): 'id': 'e402820827', 'ext': 'mp4', 'title': 'Please Use This Song (Jon Lajoie)', - 'description': 'md5:2ed27d364f5a805a6dba199faaf6681d', + 'description': 'Please use this to sell something. www.jonlajoie.com', 'thumbnail': 're:^http:.*\.jpg$', }, }] diff --git a/youtube_dl/extractor/gamestar.py b/youtube_dl/extractor/gamestar.py new file mode 100644 index 0000000..50f8fc7 --- /dev/null +++ b/youtube_dl/extractor/gamestar.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, + str_to_int, + unified_strdate, +) + + +class GameStarIE(InfoExtractor): + _VALID_URL = r'http://www\.gamestar\.de/videos/.*,(?P<id>[0-9]+)\.html' + _TEST = { + 'url': 'http://www.gamestar.de/videos/trailer,3/hobbit-3-die-schlacht-der-fuenf-heere,76110.html', + 'md5': '96974ecbb7fd8d0d20fca5a00810cea7', + 'info_dict': { + 'id': '76110', + 'ext': 'mp4', + 'title': 'Hobbit 3: Die Schlacht der Fünf Heere - Teaser-Trailer zum dritten Teil', + 'description': 'Der Teaser-Trailer zu Hobbit 3: Die Schlacht der Fünf Heere zeigt einige Szenen aus dem dritten Teil der Saga und kündigt den vollständigen Trailer an.', + 'thumbnail': 'http://images.gamestar.de/images/idgwpgsgp/bdb/2494525/600x.jpg', + 'upload_date': '20140728', + 'duration': 17 + } + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + + og_title = self._og_search_title(webpage) + title = og_title.replace(' - Video bei GameStar.de', '').strip() + + url = 'http://gamestar.de/_misc/videos/portal/getVideoUrl.cfm?premium=0&videoId=' + video_id + + description = self._og_search_description(webpage).strip() + + thumbnail = self._proto_relative_url( + self._og_search_thumbnail(webpage), scheme='http:') + + upload_date = unified_strdate(self._html_search_regex( + r'<span style="float:left;font-size:11px;">Datum: ([0-9]+\.[0-9]+\.[0-9]+)  ', + webpage, 'upload_date', fatal=False)) + + duration = parse_duration(self._html_search_regex( + r'  Länge: ([0-9]+:[0-9]+)</span>', webpage, 'duration', + fatal=False)) + + view_count = str_to_int(self._html_search_regex( + r'  Zuschauer: ([0-9\.]+)  ', webpage, + 'view_count', fatal=False)) + + comment_count = int_or_none(self._html_search_regex( + r'>Kommentieren \(([0-9]+)\)</a>', webpage, 'comment_count', + fatal=False)) + + return { + 'id': video_id, + 'title': title, + 'url': url, + 'ext': 'mp4', + 'thumbnail': thumbnail, + 'description': description, + 'upload_date': upload_date, + 'duration': duration, + 'view_count': view_count, + 'comment_count': comment_count + } diff --git a/youtube_dl/extractor/gdcvault.py b/youtube_dl/extractor/gdcvault.py index 89d5994..de14ae1 100644 --- a/youtube_dl/extractor/gdcvault.py +++ b/youtube_dl/extractor/gdcvault.py @@ -8,6 +8,7 @@ from ..utils import ( compat_urllib_request, ) + class GDCVaultIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)/(?P<name>(\w|-)+)' _TESTS = [ @@ -31,6 +32,15 @@ class GDCVaultIE(InfoExtractor): 'skip_download': True, # Requires rtmpdump } }, + { + 'url': 'http://www.gdcvault.com/play/1015301/Thexder-Meets-Windows-95-or', + 'md5': 'a5eb77996ef82118afbbe8e48731b98e', + 'info_dict': { + 'id': '1015301', + 'ext': 'flv', + 'title': 'Thexder Meets Windows 95, or Writing Great Games in the Windows 95 Environment', + } + } ] def _parse_mp4(self, xml_description): @@ -103,18 +113,40 @@ class GDCVaultIE(InfoExtractor): webpage_url = 'http://www.gdcvault.com/play/' + video_id start_page = self._download_webpage(webpage_url, video_id) - xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root', None, False) + direct_url = self._search_regex( + r's1\.addVariable\("file",\s*encodeURIComponent\("(/[^"]+)"\)\);', + start_page, 'url', default=None) + if direct_url: + video_url = 'http://www.gdcvault.com/' + direct_url + title = self._html_search_regex( + r'<td><strong>Session Name</strong></td>\s*<td>(.*?)</td>', + start_page, 'title') + + return { + 'id': video_id, + 'url': video_url, + 'ext': 'flv', + 'title': title, + } + xml_root = self._html_search_regex( + r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', + start_page, 'xml root', default=None) if xml_root is None: # Probably need to authenticate - start_page = self._login(webpage_url, video_id) - if start_page is None: + login_res = self._login(webpage_url, video_id) + if login_res is None: self.report_warning('Could not login.') else: + start_page = login_res # Grab the url from the authenticated page - xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root') + xml_root = self._html_search_regex( + r'<iframe src="(.*?)player.html.*?".*?</iframe>', + start_page, 'xml root') - xml_name = self._html_search_regex(r'<iframe src=".*?\?xml=(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename', None, False) + xml_name = self._html_search_regex( + r'<iframe src=".*?\?xml=(.+?\.xml).*?".*?</iframe>', + start_page, 'xml filename', default=None) if xml_name is None: # Fallback to the older format xml_name = self._html_search_regex(r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename') diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index f97b598..bcb0765 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -383,13 +383,13 @@ class GenericIE(InfoExtractor): if not parsed_url.scheme: default_search = self._downloader.params.get('default_search') if default_search is None: - default_search = 'error' + default_search = 'fixup_error' - if default_search in ('auto', 'auto_warning'): + if default_search in ('auto', 'auto_warning', 'fixup_error'): if '/' in url: self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http') return self.url_result('http://' + url) - else: + elif default_search != 'fixup_error': if default_search == 'auto_warning': if re.match(r'^(?:url|URL)$', url): raise ExtractorError( @@ -399,10 +399,11 @@ class GenericIE(InfoExtractor): self._downloader.report_warning( 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url) return self.url_result('ytsearch:' + url) - elif default_search == 'error': + + if default_search in ('error', 'fixup_error'): raise ExtractorError( ('%r is not a valid URL. ' - 'Set --default-search "ytseach" (or run youtube-dl "ytsearch:%s" ) to search YouTube' + 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube' ) % (url, url), expected=True) else: assert ':' in default_search diff --git a/youtube_dl/extractor/godtube.py b/youtube_dl/extractor/godtube.py new file mode 100644 index 0000000..73bd6d8 --- /dev/null +++ b/youtube_dl/extractor/godtube.py @@ -0,0 +1,58 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + parse_duration, + parse_iso8601, +) + + +class GodTubeIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?godtube\.com/watch/\?v=(?P<id>[\da-zA-Z]+)' + _TESTS = [ + { + 'url': 'https://www.godtube.com/watch/?v=0C0CNNNU', + 'md5': '77108c1e4ab58f48031101a1a2119789', + 'info_dict': { + 'id': '0C0CNNNU', + 'ext': 'mp4', + 'title': 'Woman at the well.', + 'duration': 159, + 'timestamp': 1205712000, + 'uploader': 'beverlybmusic', + 'upload_date': '20080317', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + }, + ] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + config = self._download_xml( + 'http://www.godtube.com/resource/mediaplayer/%s.xml' % video_id.lower(), + video_id, 'Downloading player config XML') + + video_url = config.find('.//file').text + uploader = config.find('.//author').text + timestamp = parse_iso8601(config.find('.//date').text) + duration = parse_duration(config.find('.//duration').text) + thumbnail = config.find('.//image').text + + media = self._download_xml( + 'http://www.godtube.com/media/xml/?v=%s' % video_id, video_id, 'Downloading media XML') + + title = media.find('.//title').text + + return { + 'id': video_id, + 'url': video_url, + 'title': title, + 'thumbnail': thumbnail, + 'timestamp': timestamp, + 'uploader': uploader, + 'duration': duration, + } diff --git a/youtube_dl/extractor/izlesene.py b/youtube_dl/extractor/izlesene.py new file mode 100644 index 0000000..79e8430 --- /dev/null +++ b/youtube_dl/extractor/izlesene.py @@ -0,0 +1,97 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + get_element_by_id, + parse_iso8601, + determine_ext, + int_or_none, + str_to_int, +) + + +class IzleseneIE(InfoExtractor): + _VALID_URL = r'https?://(?:(?:www|m)\.)?izlesene\.com/(?:video|embedplayer)/(?:[^/]+/)?(?P<id>[0-9]+)' + _STREAM_URL = 'http://panel.izlesene.com/api/streamurl/{id:}/{format:}' + _TEST = { + 'url': 'http://www.izlesene.com/video/sevincten-cildirtan-dogum-gunu-hediyesi/7599694', + 'md5': '4384f9f0ea65086734b881085ee05ac2', + 'info_dict': { + 'id': '7599694', + 'ext': 'mp4', + 'title': 'Sevinçten Çıldırtan Doğum Günü Hediyesi', + 'description': 'Annesi oğluna doğum günü hediyesi olarak minecraft cd si alıyor, ve çocuk hunharca seviniyor', + 'thumbnail': 're:^http://.*\.jpg', + 'uploader_id': 'pelikzzle', + 'timestamp': 1404298698, + 'upload_date': '20140702', + 'duration': 95.395, + 'age_limit': 0, + } + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + url = 'http://www.izlesene.com/video/%s' % video_id + + webpage = self._download_webpage(url, video_id) + + title = self._og_search_title(webpage) + description = self._og_search_description(webpage) + thumbnail = self._og_search_thumbnail(webpage) + + uploader = self._html_search_regex( + r"adduserUsername\s*=\s*'([^']+)';", webpage, 'uploader', fatal=False, default='') + timestamp = parse_iso8601(self._html_search_meta( + 'uploadDate', webpage, 'upload date', fatal=False)) + + duration = int_or_none(self._html_search_regex( + r'"videoduration"\s*:\s*"([^"]+)"', webpage, 'duration', fatal=False)) + if duration: + duration /= 1000.0 + + view_count = str_to_int(get_element_by_id('videoViewCount', webpage)) + comment_count = self._html_search_regex( + r'comment_count\s*=\s*\'([^\']+)\';', webpage, 'uploader', fatal=False) + + family_friendly = self._html_search_meta( + 'isFamilyFriendly', webpage, 'age limit', fatal=False) + + content_url = self._html_search_meta( + 'contentURL', webpage, 'content URL', fatal=False) + ext = determine_ext(content_url, 'mp4') + + # Might be empty for some videos. + qualities = self._html_search_regex( + r'"quality"\s*:\s*"([^"]+)"', webpage, 'qualities', fatal=False, default='') + + formats = [] + for quality in qualities.split('|'): + json = self._download_json( + self._STREAM_URL.format(id=video_id, format=quality), video_id, + note='Getting video URL for "%s" quality' % quality, + errnote='Failed to get video URL for "%s" quality' % quality + ) + formats.append({ + 'url': json.get('streamurl'), + 'ext': ext, + 'format_id': '%sp' % quality if quality else 'sd', + }) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'uploader_id': uploader, + 'timestamp': timestamp, + 'duration': duration, + 'view_count': int_or_none(view_count), + 'comment_count': int_or_none(comment_count), + 'age_limit': 18 if family_friendly == 'False' else 0, + 'formats': formats, + } diff --git a/youtube_dl/extractor/justintv.py b/youtube_dl/extractor/justintv.py index 7083db1..27017e8 100644 --- a/youtube_dl/extractor/justintv.py +++ b/youtube_dl/extractor/justintv.py @@ -1,5 +1,6 @@ from __future__ import unicode_literals +import itertools import json import os import re @@ -43,10 +44,11 @@ class JustinTVIE(InfoExtractor): } # Return count of items, list of *valid* items - def _parse_page(self, url, video_id): - info_json = self._download_webpage(url, video_id, - 'Downloading video info JSON', - 'unable to download video info JSON') + def _parse_page(self, url, video_id, counter): + info_json = self._download_webpage( + url, video_id, + 'Downloading video info JSON on page %d' % counter, + 'Unable to download video info JSON %d' % counter) response = json.loads(info_json) if type(response) != list: @@ -138,11 +140,10 @@ class JustinTVIE(InfoExtractor): entries = [] offset = 0 limit = self._JUSTIN_PAGE_LIMIT - while True: - if paged: - self.report_download_page(video_id, offset) + for counter in itertools.count(1): page_url = api + ('?offset=%d&limit=%d' % (offset, limit)) - page_count, page_info = self._parse_page(page_url, video_id) + page_count, page_info = self._parse_page( + page_url, video_id, counter) entries.extend(page_info) if not paged or page_count != limit: break diff --git a/youtube_dl/extractor/kickstarter.py b/youtube_dl/extractor/kickstarter.py index 961dd1a..56a7638 100644 --- a/youtube_dl/extractor/kickstarter.py +++ b/youtube_dl/extractor/kickstarter.py @@ -8,7 +8,7 @@ from .common import InfoExtractor class KickStarterIE(InfoExtractor): _VALID_URL = r'https?://www\.kickstarter\.com/projects/(?P<id>[^/]*)/.*' - _TEST = { + _TESTS = [{ 'url': 'https://www.kickstarter.com/projects/1404461844/intersection-the-story-of-josh-grant?ref=home_location', 'md5': 'c81addca81327ffa66c642b5d8b08cab', 'info_dict': { @@ -18,22 +18,45 @@ class KickStarterIE(InfoExtractor): 'description': 'A unique motocross documentary that examines the ' 'life and mind of one of sports most elite athletes: Josh Grant.', }, - } + }, { + 'note': 'Embedded video (not using the native kickstarter video service)', + 'url': 'https://www.kickstarter.com/projects/597507018/pebble-e-paper-watch-for-iphone-and-android/posts/659178', + 'playlist': [ + { + 'info_dict': { + 'id': '78704821', + 'ext': 'mp4', + 'uploader_id': 'pebble', + 'uploader': 'Pebble Technology', + 'title': 'Pebble iOS Notifications', + } + } + ], + }] def _real_extract(self, url): m = re.match(self._VALID_URL, url) video_id = m.group('id') webpage = self._download_webpage(url, video_id) - video_url = self._search_regex(r'data-video-url="(.*?)"', - webpage, 'video URL') - video_title = self._html_search_regex(r'<title>(.*?)', - webpage, 'title').rpartition('— Kickstarter')[0].strip() + title = self._html_search_regex( + r'\s*(.*?)(?:\s*— Kickstarter)?\s*', + webpage, 'title') + video_url = self._search_regex( + r'data-video-url="(.*?)"', + webpage, 'video URL', default=None) + if video_url is None: # No native kickstarter, look for embedded videos + return { + '_type': 'url_transparent', + 'ie_key': 'Generic', + 'url': url, + 'title': title, + } return { 'id': video_id, 'url': video_url, - 'title': video_title, + 'title': title, 'description': self._og_search_description(webpage), 'thumbnail': self._og_search_thumbnail(webpage), } diff --git a/youtube_dl/extractor/krasview.py b/youtube_dl/extractor/krasview.py new file mode 100644 index 0000000..6f3d234 --- /dev/null +++ b/youtube_dl/extractor/krasview.py @@ -0,0 +1,59 @@ +# encoding: utf-8 +from __future__ import unicode_literals + +import json +import re + +from .common import InfoExtractor +from ..utils import ( + int_or_none, + unescapeHTML, +) + + +class KrasViewIE(InfoExtractor): + IE_DESC = 'Красвью' + _VALID_URL = r'https?://krasview\.ru/video/(?P\d+)' + + _TEST = { + 'url': 'http://krasview.ru/video/512228', + 'md5': '3b91003cf85fc5db277870c8ebd98eae', + 'info_dict': { + 'id': '512228', + 'ext': 'mp4', + 'title': 'Снег, лёд, заносы', + 'description': 'Снято в городе Нягань, в Ханты-Мансийском автономном округе.', + 'duration': 27, + 'thumbnail': 're:^https?://.*\.jpg', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + + flashvars = json.loads(self._search_regex( + r'flashvars\s*:\s*({.+?})\s*}\);', webpage, 'flashvars')) + + video_url = flashvars['url'] + title = unescapeHTML(flashvars['title']) + description = unescapeHTML(flashvars.get('subtitle') or self._og_search_description(webpage, default=None)) + thumbnail = flashvars['image'] + duration = int(flashvars['duration']) + filesize = int(flashvars['size']) + width = int_or_none(self._og_search_property('video:width', webpage, 'video width')) + height = int_or_none(self._og_search_property('video:height', webpage, 'video height')) + + return { + 'id': video_id, + 'url': video_url, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'filesize': filesize, + 'width': width, + 'height': height, + } diff --git a/youtube_dl/extractor/livestream.py b/youtube_dl/extractor/livestream.py index 2c100d4..c0c2d9b 100644 --- a/youtube_dl/extractor/livestream.py +++ b/youtube_dl/extractor/livestream.py @@ -5,11 +5,14 @@ import json from .common import InfoExtractor from ..utils import ( + compat_str, compat_urllib_parse_urlparse, compat_urlparse, - xpath_with_ns, - compat_str, + ExtractorError, + find_xpath_attr, + int_or_none, orderedSet, + xpath_with_ns, ) @@ -24,18 +27,82 @@ class LivestreamIE(InfoExtractor): 'ext': 'mp4', 'title': 'Live from Webster Hall NYC', 'upload_date': '20121012', + 'like_count': int, + 'view_count': int, + 'thumbnail': 're:^http://.*\.jpg$' } } + def _parse_smil(self, video_id, smil_url): + formats = [] + _SWITCH_XPATH = ( + './/{http://www.w3.org/2001/SMIL20/Language}body/' + '{http://www.w3.org/2001/SMIL20/Language}switch') + smil_doc = self._download_xml( + smil_url, video_id, + note='Downloading SMIL information', + errnote='Unable to download SMIL information', + fatal=False) + if smil_doc is False: # Download failed + return formats + title_node = find_xpath_attr( + smil_doc, './/{http://www.w3.org/2001/SMIL20/Language}meta', + 'name', 'title') + if title_node is None: + self.report_warning('Cannot find SMIL id') + switch_node = smil_doc.find(_SWITCH_XPATH) + else: + title_id = title_node.attrib['content'] + switch_node = find_xpath_attr( + smil_doc, _SWITCH_XPATH, 'id', title_id) + if switch_node is None: + raise ExtractorError('Cannot find switch node') + video_nodes = switch_node.findall( + '{http://www.w3.org/2001/SMIL20/Language}video') + + for vn in video_nodes: + tbr = int_or_none(vn.attrib.get('system-bitrate')) + furl = ( + 'http://livestream-f.akamaihd.net/%s?v=3.0.3&fp=WIN%%2014,0,0,145' % + (vn.attrib['src'])) + if 'clipBegin' in vn.attrib: + furl += '&ssek=' + vn.attrib['clipBegin'] + formats.append({ + 'url': furl, + 'format_id': 'smil_%d' % tbr, + 'ext': 'flv', + 'tbr': tbr, + 'preference': -1000, + }) + return formats + def _extract_video_info(self, video_data): - video_url = video_data.get('progressive_url_hd') or video_data.get('progressive_url') + video_id = compat_str(video_data['id']) + + FORMAT_KEYS = ( + ('sd', 'progressive_url'), + ('hd', 'progressive_url_hd'), + ) + formats = [{ + 'format_id': format_id, + 'url': video_data[key], + 'quality': i + 1, + } for i, (format_id, key) in enumerate(FORMAT_KEYS) + if video_data.get(key)] + + smil_url = video_data.get('smil_url') + if smil_url: + formats.extend(self._parse_smil(video_id, smil_url)) + self._sort_formats(formats) + return { - 'id': compat_str(video_data['id']), - 'url': video_url, - 'ext': 'mp4', + 'id': video_id, + 'formats': formats, 'title': video_data['caption'], - 'thumbnail': video_data['thumbnail_url'], + 'thumbnail': video_data.get('thumbnail_url'), 'upload_date': video_data['updated_at'].replace('-', '')[:8], + 'like_count': video_data.get('likes', {}).get('total'), + 'view_count': video_data.get('views'), } def _real_extract(self, url): @@ -50,7 +117,8 @@ class LivestreamIE(InfoExtractor): r'window.config = ({.*?});', webpage, 'window config') info = json.loads(config_json)['event'] videos = [self._extract_video_info(video_data['data']) - for video_data in info['feed']['data'] if video_data['type'] == 'video'] + for video_data in info['feed']['data'] + if video_data['type'] == 'video'] return self.playlist_result(videos, info['id'], info['full_name']) else: og_video = self._og_search_video_url(webpage, 'player url') diff --git a/youtube_dl/extractor/mlb.py b/youtube_dl/extractor/mlb.py new file mode 100644 index 0000000..37c72bc --- /dev/null +++ b/youtube_dl/extractor/mlb.py @@ -0,0 +1,116 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + parse_duration, + parse_iso8601, + find_xpath_attr, +) + + +class MLBIE(InfoExtractor): + _VALID_URL = r'https?://m\.mlb\.com/(?:.*?/)?video/(?:topic/[\da-z_-]+/)?v(?Pn?\d+)' + _TESTS = [ + { + 'url': 'http://m.mlb.com/sea/video/topic/51231442/v34698933/nymsea-ackley-robs-a-home-run-with-an-amazing-catch/?c_id=sea', + 'md5': 'ff56a598c2cf411a9a38a69709e97079', + 'info_dict': { + 'id': '34698933', + 'ext': 'mp4', + 'title': "Ackley's spectacular catch", + 'description': 'md5:7f5a981eb4f3cbc8daf2aeffa2215bf0', + 'duration': 66, + 'timestamp': 1405980600, + 'upload_date': '20140721', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + }, + { + 'url': 'http://m.mlb.com/video/topic/81536970/v34496663/mianym-stanton-practices-for-the-home-run-derby', + 'md5': 'd9c022c10d21f849f49c05ae12a8a7e9', + 'info_dict': { + 'id': '34496663', + 'ext': 'mp4', + 'title': 'Stanton prepares for Derby', + 'description': 'md5:d00ce1e5fd9c9069e9c13ab4faedfa57', + 'duration': 46, + 'timestamp': 1405105800, + 'upload_date': '20140711', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + }, + { + 'url': 'http://m.mlb.com/video/topic/vtp_hrd_sponsor/v34578115/hrd-cespedes-wins-2014-gillette-home-run-derby', + 'md5': '0e6e73d509321e142409b695eadd541f', + 'info_dict': { + 'id': '34578115', + 'ext': 'mp4', + 'title': 'Cespedes repeats as Derby champ', + 'description': 'md5:08df253ce265d4cf6fb09f581fafad07', + 'duration': 488, + 'timestamp': 1405399936, + 'upload_date': '20140715', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + }, + { + 'url': 'http://m.mlb.com/video/v34577915/bautista-on-derby-captaining-duties-his-performance', + 'md5': 'b8fd237347b844365d74ea61d4245967', + 'info_dict': { + 'id': '34577915', + 'ext': 'mp4', + 'title': 'Bautista on Home Run Derby', + 'description': 'md5:b80b34031143d0986dddc64a8839f0fb', + 'duration': 52, + 'timestamp': 1405390722, + 'upload_date': '20140715', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + }, + ] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + detail = self._download_xml( + 'http://m.mlb.com/gen/multimedia/detail/%s/%s/%s/%s.xml' + % (video_id[-3], video_id[-2], video_id[-1], video_id), video_id) + + title = detail.find('./headline').text + description = detail.find('./big-blurb').text + duration = parse_duration(detail.find('./duration').text) + timestamp = parse_iso8601(detail.attrib['date'][:-5]) + + thumbnail = find_xpath_attr( + detail, './thumbnailScenarios/thumbnailScenario', 'type', '45').text + + formats = [] + for media_url in detail.findall('./url'): + playback_scenario = media_url.attrib['playback_scenario'] + fmt = { + 'url': media_url.text, + 'format_id': playback_scenario, + } + m = re.search(r'(?P\d+)K_(?P\d+)X(?P\d+)', playback_scenario) + if m: + fmt.update({ + 'vbr': int(m.group('vbr')) * 1000, + 'width': int(m.group('width')), + 'height': int(m.group('height')), + }) + formats.append(fmt) + + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + 'thumbnail': thumbnail, + } diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py index aa34665..d2e4acb 100644 --- a/youtube_dl/extractor/nbc.py +++ b/youtube_dl/extractor/nbc.py @@ -4,7 +4,11 @@ import re import json from .common import InfoExtractor -from ..utils import find_xpath_attr, compat_str +from ..utils import ( + compat_str, + ExtractorError, + find_xpath_attr, +) class NBCIE(InfoExtractor): @@ -85,11 +89,25 @@ class NBCNewsIE(InfoExtractor): flags=re.MULTILINE) bootstrap = json.loads(bootstrap_json) info = bootstrap['results'][0]['video'] - playlist_url = info['fallbackPlaylistUrl'] + '?form=MPXNBCNewsAPI' mpxid = info['mpxId'] - all_videos = self._download_json(playlist_url, title)['videos'] - # The response contains additional videos - info = next(v for v in all_videos if v['mpxId'] == mpxid) + + base_urls = [ + info['fallbackPlaylistUrl'], + info['associatedPlaylistUrl'], + ] + + for base_url in base_urls: + playlist_url = base_url + '?form=MPXNBCNewsAPI' + all_videos = self._download_json(playlist_url, title)['videos'] + + try: + info = next(v for v in all_videos if v['mpxId'] == mpxid) + break + except StopIteration: + continue + + if info is None: + raise ExtractorError('Could not find video in playlists') return { '_type': 'url', diff --git a/youtube_dl/extractor/npo.py b/youtube_dl/extractor/npo.py index fbcbe1f..12e85a7 100644 --- a/youtube_dl/extractor/npo.py +++ b/youtube_dl/extractor/npo.py @@ -32,7 +32,7 @@ class NPOIE(InfoExtractor): 'http://e.omroep.nl/metadata/aflevering/%s' % video_id, video_id, # We have to remove the javascript callback - transform_source=lambda j: re.sub(r'parseMetadata\((.*?)\);\n//epc', r'\1', j) + transform_source=lambda j: re.sub(r'parseMetadata\((.*?)\);\n//.*$', r'\1', j) ) token_page = self._download_webpage( 'http://ida.omroep.nl/npoplayer/i.js', diff --git a/youtube_dl/extractor/pbs.py b/youtube_dl/extractor/pbs.py index 64cded7..ec95d07 100644 --- a/youtube_dl/extractor/pbs.py +++ b/youtube_dl/extractor/pbs.py @@ -32,13 +32,21 @@ class PBSIE(InfoExtractor): }, } - def _real_extract(self, url): + def _extract_ids(self, url): mobj = re.match(self._VALID_URL, url) presumptive_id = mobj.group('presumptive_id') display_id = presumptive_id if presumptive_id: webpage = self._download_webpage(url, display_id) + + # frontline video embed + media_id = self._search_regex( + r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'", + webpage, 'frontline video ID', fatal=False, default=None) + if media_id: + return media_id, presumptive_id + url = self._search_regex( r'', webpage, 'player URL') @@ -57,6 +65,11 @@ class PBSIE(InfoExtractor): video_id = mobj.group('id') display_id = video_id + return video_id, display_id + + def _real_extract(self, url): + video_id, display_id = self._extract_ids(url) + info_url = 'http://video.pbs.org/videoInfo/%s?format=json' % video_id info = self._download_json(info_url, display_id) diff --git a/youtube_dl/extractor/redtube.py b/youtube_dl/extractor/redtube.py index 4295cf9..d1e12dd 100644 --- a/youtube_dl/extractor/redtube.py +++ b/youtube_dl/extractor/redtube.py @@ -35,9 +35,7 @@ class RedTubeIE(InfoExtractor): r'

(.+?)

', webpage, u'title') - video_thumbnail = self._html_search_regex( - r'playerInnerHTML.+?[\da-zA-Z]{20})' + + _TESTS = [ + { + 'url': 'http://videos.sapo.pt/UBz95kOtiWYUMTA5Ghfi', + 'md5': '79ee523f6ecb9233ac25075dee0eda83', + 'note': 'SD video', + 'info_dict': { + 'id': 'UBz95kOtiWYUMTA5Ghfi', + 'ext': 'mp4', + 'title': 'Benfica - Marcas na Hitória', + 'description': 'md5:c9082000a128c3fd57bf0299e1367f22', + 'duration': 264, + 'uploader': 'tiago_1988', + 'upload_date': '20080229', + 'categories': ['benfica', 'cabral', 'desporto', 'futebol', 'geovanni', 'hooijdonk', 'joao', 'karel', 'lisboa', 'miccoli'], + }, + }, + { + 'url': 'http://videos.sapo.pt/IyusNAZ791ZdoCY5H5IF', + 'md5': '90a2f283cfb49193fe06e861613a72aa', + 'note': 'HD video', + 'info_dict': { + 'id': 'IyusNAZ791ZdoCY5H5IF', + 'ext': 'mp4', + 'title': 'Codebits VII - Report', + 'description': 'md5:6448d6fd81ce86feac05321f354dbdc8', + 'duration': 144, + 'uploader': 'codebits', + 'upload_date': '20140427', + 'categories': ['codebits', 'codebits2014'], + }, + }, + { + 'url': 'http://v2.videos.sapo.pt/yLqjzPtbTimsn2wWBKHz', + 'md5': 'e5aa7cc0bdc6db9b33df1a48e49a15ac', + 'note': 'v2 video', + 'info_dict': { + 'id': 'yLqjzPtbTimsn2wWBKHz', + 'ext': 'mp4', + 'title': 'Hipnose Condicionativa 4', + 'description': 'md5:ef0481abf8fb4ae6f525088a6dadbc40', + 'duration': 692, + 'uploader': 'sapozen', + 'upload_date': '20090609', + 'categories': ['condicionativa', 'heloisa', 'hipnose', 'miranda', 'sapo', 'zen'], + }, + }, + ] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + item = self._download_xml( + 'http://rd3.videos.sapo.pt/%s/rss2' % video_id, video_id).find('./channel/item') + + title = item.find('./title').text + description = item.find('./{http://videos.sapo.pt/mrss/}synopse').text + thumbnail = item.find('./{http://search.yahoo.com/mrss/}content').get('url') + duration = parse_duration(item.find('./{http://videos.sapo.pt/mrss/}time').text) + uploader = item.find('./{http://videos.sapo.pt/mrss/}author').text + upload_date = unified_strdate(item.find('./pubDate').text) + view_count = int(item.find('./{http://videos.sapo.pt/mrss/}views').text) + comment_count = int(item.find('./{http://videos.sapo.pt/mrss/}comment_count').text) + tags = item.find('./{http://videos.sapo.pt/mrss/}tags').text + categories = tags.split() if tags else [] + age_limit = 18 if item.find('./{http://videos.sapo.pt/mrss/}m18').text == 'true' else 0 + + video_url = item.find('./{http://videos.sapo.pt/mrss/}videoFile').text + video_size = item.find('./{http://videos.sapo.pt/mrss/}videoSize').text.split('x') + + formats = [{ + 'url': video_url, + 'ext': 'mp4', + 'format_id': 'sd', + 'width': int(video_size[0]), + 'height': int(video_size[1]), + }] + + if item.find('./{http://videos.sapo.pt/mrss/}HD').text == 'true': + formats.append({ + 'url': re.sub(r'/mov/1$', '/mov/39', video_url), + 'ext': 'mp4', + 'format_id': 'hd', + 'width': 1280, + 'height': 720, + }) + + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'uploader': uploader, + 'upload_date': upload_date, + 'view_count': view_count, + 'comment_count': comment_count, + 'categories': categories, + 'age_limit': age_limit, + 'formats': formats, + } diff --git a/youtube_dl/extractor/savefrom.py b/youtube_dl/extractor/savefrom.py index 198a08c..5b7367b 100644 --- a/youtube_dl/extractor/savefrom.py +++ b/youtube_dl/extractor/savefrom.py @@ -20,7 +20,7 @@ class SaveFromIE(InfoExtractor): 'upload_date': '20120816', 'uploader': 'Howcast', 'uploader_id': 'Howcast', - 'description': 'md5:4f0aac94361a12e1ce57d74f85265175', + 'description': 're:(?s).* Hi, my name is Rene Dreifuss\. And I\'m here to show you some MMA.*', }, 'params': { 'skip_download': True diff --git a/youtube_dl/extractor/shared.py b/youtube_dl/extractor/shared.py new file mode 100644 index 0000000..8607482 --- /dev/null +++ b/youtube_dl/extractor/shared.py @@ -0,0 +1,57 @@ +from __future__ import unicode_literals + +import re +import base64 + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + compat_urllib_request, + compat_urllib_parse, + int_or_none, +) + + +class SharedIE(InfoExtractor): + _VALID_URL = r'http://shared\.sx/(?P[\da-z]{10})' + + _TEST = { + 'url': 'http://shared.sx/0060718775', + 'md5': '53e1c58fc3e777ae1dfe9e57ba2f9c72', + 'info_dict': { + 'id': '0060718775', + 'ext': 'mp4', + 'title': 'Big Buck Bunny Trailer', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + page = self._download_webpage(url, video_id) + + if re.search(r'>File does not exist<', page) is not None: + raise ExtractorError('Video %s does not exist' % video_id, expected=True) + + download_form = dict(re.findall(r'\d+)/([\w]+)' + _TESTS = [{ + 'url': 'http://www.snotr.com/video/13708/Drone_flying_through_fireworks', + 'info_dict': { + 'id': '13708', + 'ext': 'flv', + 'title': 'Drone flying through fireworks!', + 'duration': 247, + 'filesize_approx': 98566144, + 'description': 'A drone flying through Fourth of July Fireworks', + } + }, { + 'url': 'http://www.snotr.com/video/530/David_Letteman_-_George_W_Bush_Top_10', + 'info_dict': { + 'id': '530', + 'ext': 'flv', + 'title': 'David Letteman - George W. Bush Top 10', + 'duration': 126, + 'filesize_approx': 8912896, + 'description': 'The top 10 George W. Bush moments, brought to you by David Letterman!', + } + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + title = self._og_search_title(webpage) + + description = self._og_search_description(webpage) + video_url = "http://cdn.videos.snotr.com/%s.flv" % video_id + + view_count = str_to_int(self._html_search_regex( + r'

\nViews:\n([\d,\.]+)

', + webpage, 'view count', fatal=False)) + + duration = parse_duration(self._html_search_regex( + r'

\nLength:\n\s*([0-9:]+).*?

', + webpage, 'duration', fatal=False)) + + filesize_approx = float_or_none(self._html_search_regex( + r'

\nFilesize:\n\s*([0-9.]+)\s*megabyte

', + webpage, 'filesize', fatal=False), invscale=1024 * 1024) + + return { + 'id': video_id, + 'description': description, + 'title': title, + 'url': video_url, + 'view_count': view_count, + 'duration': duration, + 'filesize_approx': filesize_approx, + } diff --git a/youtube_dl/extractor/sockshare.py b/youtube_dl/extractor/sockshare.py new file mode 100644 index 0000000..dc9f805 --- /dev/null +++ b/youtube_dl/extractor/sockshare.py @@ -0,0 +1,80 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from ..utils import ( + ExtractorError, + compat_urllib_parse, + compat_urllib_request, + determine_ext, +) +import re + +from .common import InfoExtractor + + +class SockshareIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?sockshare\.com/file/(?P[0-9A-Za-z]+)' + _FILE_DELETED_REGEX = r'This file doesn\'t exist, or has been removed\.' + _TEST = { + 'url': 'http://www.sockshare.com/file/437BE28B89D799D7', + 'md5': '9d0bf1cfb6dbeaa8d562f6c97506c5bd', + 'info_dict': { + 'id': '437BE28B89D799D7', + 'title': 'big_buck_bunny_720p_surround.avi', + 'ext': 'avi', + 'thumbnail': 're:^http://.*\.jpg$', + } + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + url = 'http://sockshare.com/file/%s' % video_id + webpage = self._download_webpage(url, video_id) + + if re.search(self._FILE_DELETED_REGEX, webpage) is not None: + raise ExtractorError('Video %s does not exist' % video_id, + expected=True) + + confirm_hash = self._html_search_regex(r'''(?x)(.+)', webpage, 'title') + thumbnail = self._html_search_regex( + r'[a-zA-Z0-9_-]+)/(?P[^#?]*)\.html' _TEST = { - u'url': u'http://streamcloud.eu/skp9j99s4bpz/youtube-dl_test_video_____________-BaW_jenozKc.mp4.html', - u'file': u'skp9j99s4bpz.mp4', - u'md5': u'6bea4c7fa5daaacc2a946b7146286686', - u'info_dict': { - u'title': u'youtube-dl test video \'/\\ ä ↭', - u'duration': 9, + 'url': 'http://streamcloud.eu/skp9j99s4bpz/youtube-dl_test_video_____________-BaW_jenozKc.mp4.html', + 'md5': '6bea4c7fa5daaacc2a946b7146286686', + 'info_dict': { + 'id': 'skp9j99s4bpz', + 'ext': 'mp4', + 'title': 'youtube-dl test video \'/\\ ä ↭', }, - u'skip': u'Only available from the EU' + 'skip': 'Only available from the EU' } def _real_extract(self, url): @@ -46,21 +48,17 @@ class StreamcloudIE(InfoExtractor): req = compat_urllib_request.Request(url, post, headers) webpage = self._download_webpage( - req, video_id, note=u'Downloading video page ...') + req, video_id, note='Downloading video page ...') title = self._html_search_regex( - r']*>([^<]+)<', webpage, u'title') + r']*>([^<]+)<', webpage, 'title') video_url = self._search_regex( - r'file:\s*"([^"]+)"', webpage, u'video URL') - duration_str = self._search_regex( - r'duration:\s*"?([0-9]+)"?', webpage, u'duration', fatal=False) - duration = None if duration_str is None else int(duration_str) + r'file:\s*"([^"]+)"', webpage, 'video URL') thumbnail = self._search_regex( - r'image:\s*"([^"]+)"', webpage, u'thumbnail URL', fatal=False) + r'image:\s*"([^"]+)"', webpage, 'thumbnail URL', fatal=False) return { 'id': video_id, 'title': title, 'url': video_url, - 'duration': duration, 'thumbnail': thumbnail, } diff --git a/youtube_dl/extractor/swrmediathek.py b/youtube_dl/extractor/swrmediathek.py index 6c688c5..5d9d703 100644 --- a/youtube_dl/extractor/swrmediathek.py +++ b/youtube_dl/extractor/swrmediathek.py @@ -8,7 +8,7 @@ from ..utils import parse_duration class SWRMediathekIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?swrmediathek\.de/player\.htm\?show=(?P[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})' + _VALID_URL = r'https?://(?:www\.)?swrmediathek\.de/(?:content/)?player\.htm\?show=(?P[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})' _TESTS = [{ 'url': 'http://swrmediathek.de/player.htm?show=849790d0-dab8-11e3-a953-0026b975f2e6', @@ -52,6 +52,20 @@ class SWRMediathekIE(InfoExtractor): 'uploader': 'SWR 2', 'uploader_id': '284670', } + }, { + 'url': 'http://swrmediathek.de/content/player.htm?show=52dc7e00-15c5-11e4-84bc-0026b975f2e6', + 'md5': '881531487d0633080a8cc88d31ef896f', + 'info_dict': { + 'id': '52dc7e00-15c5-11e4-84bc-0026b975f2e6', + 'ext': 'mp4', + 'title': 'Familienspaß am Bodensee', + 'description': 'md5:0b591225a32cfde7be1629ed49fe4315', + 'thumbnail': 're:http://.*\.jpg', + 'duration': 1784, + 'upload_date': '20140727', + 'uploader': 'SWR Fernsehen BW', + 'uploader_id': '281130', + } }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/tagesschau.py b/youtube_dl/extractor/tagesschau.py index 25b9864..b870474 100644 --- a/youtube_dl/extractor/tagesschau.py +++ b/youtube_dl/extractor/tagesschau.py @@ -19,16 +19,6 @@ class TagesschauIE(InfoExtractor): 'description': 'md5:69da3c61275b426426d711bde96463ab', 'thumbnail': 're:^http:.*\.jpg$', }, - }, { - 'url': 'http://www.tagesschau.de/multimedia/video/video-5964.html', - 'md5': '66652566900963a3f962333579eeffcf', - 'info_dict': { - 'id': '5964', - 'ext': 'mp4', - 'title': 'Nahost-Konflikt: Israel bombadiert Ziele im Gazastreifen und Westjordanland', - 'description': 'md5:07bfc78c48eec3145ed4805299a1900a', - 'thumbnail': 're:http://.*\.jpg', - }, }] _FORMATS = { diff --git a/youtube_dl/extractor/teachertube.py b/youtube_dl/extractor/teachertube.py index 2c2113b..46d727d 100644 --- a/youtube_dl/extractor/teachertube.py +++ b/youtube_dl/extractor/teachertube.py @@ -62,7 +62,7 @@ class TeacherTubeIE(InfoExtractor): webpage = self._download_webpage(url, video_id) - title = self._html_search_meta('title', webpage, 'title') + title = self._html_search_meta('title', webpage, 'title', fatal=True) TITLE_SUFFIX = ' - TeacherTube' if title.endswith(TITLE_SUFFIX): title = title[:-len(TITLE_SUFFIX)].strip() @@ -101,7 +101,11 @@ class TeacherTubeUserIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?teachertube\.com/(user/profile|collection)/(?P[0-9a-zA-Z]+)/?' - _MEDIA_RE = r'(?s)"sidebar_thumb_time">[0-9:]+.+?' + _MEDIA_RE = r'''(?sx) + class="?sidebar_thumb_time"?>[0-9:]+ + \s* + \d+)' + _TESTS = [ + { + 'url': 'http://www.tvplay.lv/parraides/vinas-melo-labak/418113?autostart=true', + 'info_dict': { + 'id': '418113', + 'ext': 'flv', + 'title': 'Kādi ir īri? - Viņas melo labāk', + 'description': 'Baiba apsmej īrus, kādi tie ir un ko viņi dara.', + 'duration': 25, + 'timestamp': 1406097056, + 'upload_date': '20140723', + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }, + ] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + video = self._download_json( + 'http://playapi.mtgx.tv/v1/videos/%s' % video_id, video_id, 'Downloading video JSON') + + if video['is_geo_blocked']: + raise ExtractorError( + 'This content is not available in your country due to copyright reasons', expected=True) + + streams = self._download_json( + 'http://playapi.mtgx.tv/v1/videos/stream/%s' % video_id, video_id, 'Downloading streams JSON') + + quality = qualities(['hls', 'medium', 'high']) + formats = [] + for format_id, video_url in streams['streams'].items(): + if not video_url: + continue + fmt = { + 'format_id': format_id, + 'preference': quality(format_id), + } + if video_url.startswith('rtmp'): + m = re.search(r'^(?Prtmp://[^/]+/(?P[^/]+))/(?P.+)$', video_url) + if not m: + continue + fmt.update({ + 'ext': 'flv', + 'url': m.group('url'), + 'app': m.group('app'), + 'play_path': m.group('playpath'), + }) + else: + fmt.update({ + 'url': video_url, + }) + formats.append(fmt) + + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': video['title'], + 'description': video['description'], + 'duration': video['duration'], + 'timestamp': parse_iso8601(video['created_at']), + 'view_count': video['views']['total'], + 'age_limit': video.get('age_limit', 0), + 'formats': formats, + } diff --git a/youtube_dl/extractor/ubu.py b/youtube_dl/extractor/ubu.py new file mode 100644 index 0000000..0182d67 --- /dev/null +++ b/youtube_dl/extractor/ubu.py @@ -0,0 +1,56 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import int_or_none + + +class UbuIE(InfoExtractor): + _VALID_URL = r'http://(?:www\.)?ubu\.com/film/(?P[\da-z_-]+)\.html' + _TEST = { + 'url': 'http://ubu.com/film/her_noise.html', + 'md5': '8edd46ee8aa6b265fb5ed6cf05c36bc9', + 'info_dict': { + 'id': 'her_noise', + 'ext': 'mp4', + 'title': 'Her Noise - The Making Of (2007)', + 'duration': 3600, + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + + title = self._html_search_regex( + r'.+?Film & Video: ([^<]+)', webpage, 'title') + + duration = int_or_none(self._html_search_regex( + r'Duration: (\d+) minutes', webpage, 'duration', fatal=False, default=None)) + if duration: + duration *= 60 + + formats = [] + + FORMAT_REGEXES = [ + ['sq', r"'flashvars'\s*,\s*'file=([^']+)'"], + ['hq', r'href="(http://ubumexico\.centro\.org\.mx/video/[^"]+)"'] + ] + + for format_id, format_regex in FORMAT_REGEXES: + m = re.search(format_regex, webpage) + if m: + formats.append({ + 'url': m.group(1), + 'format_id': format_id, + }) + + return { + 'id': video_id, + 'title': title, + 'duration': duration, + 'formats': formats, + } diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py index eada13c..d2ffd1b 100644 --- a/youtube_dl/extractor/vevo.py +++ b/youtube_dl/extractor/vevo.py @@ -177,6 +177,7 @@ class VevoIE(InfoExtractor): self._downloader.report_warning( 'Cannot download SMIL information, falling back to JSON ..') + self._sort_formats(formats) timestamp_ms = int(self._search_regex( r'/Date\((\d+)\)/', video_info['launchDate'], 'launch date')) diff --git a/youtube_dl/extractor/vidme.py b/youtube_dl/extractor/vidme.py new file mode 100644 index 0000000..5c89824 --- /dev/null +++ b/youtube_dl/extractor/vidme.py @@ -0,0 +1,68 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + int_or_none, + float_or_none, + str_to_int, +) + + +class VidmeIE(InfoExtractor): + _VALID_URL = r'https?://vid\.me/(?:e/)?(?P[\da-zA-Z]+)' + _TEST = { + 'url': 'https://vid.me/QNB', + 'md5': 'f42d05e7149aeaec5c037b17e5d3dc82', + 'info_dict': { + 'id': 'QNB', + 'ext': 'mp4', + 'title': 'Fishing for piranha - the easy way', + 'description': 'source: https://www.facebook.com/photo.php?v=312276045600871', + 'duration': 119.92, + 'timestamp': 1406313244, + 'upload_date': '20140725', + 'thumbnail': 're:^https?://.*\.jpg', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + + video_url = self._html_search_regex(r'\s*([\d,\.]+)\s*plays?', webpage, 'view count', fatal=False)) + like_count = str_to_int(self._html_search_regex( + r'class="score js-video-vote-score"[^>]+data-score="([\d,\.\s]+)">', + webpage, 'like count', fatal=False)) + comment_count = str_to_int(self._html_search_regex( + r'class="js-comment-count"[^>]+data-count="([\d,\.\s]+)">', + webpage, 'comment count', fatal=False)) + + return { + 'id': video_id, + 'url': video_url, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'timestamp': timestamp, + 'width': width, + 'height': height, + 'duration': duration, + 'view_count': view_count, + 'like_count': like_count, + 'comment_count': comment_count, + } diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 2558555..11c7d7e 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -98,7 +98,7 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): 'info_dict': { 'id': '54469442', 'ext': 'mp4', - 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software', + 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012', 'uploader': 'The BLN & Business of Software', 'uploader_id': 'theblnbusinessofsoftware', 'duration': 3610, @@ -121,6 +121,21 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): 'videopassword': 'youtube-dl', }, }, + { + 'url': 'http://vimeo.com/channels/keypeele/75629013', + 'md5': '2f86a05afe9d7abc0b9126d229bbe15d', + 'note': 'Video is freely available via original URL ' + 'and protected with password when accessed via http://vimeo.com/75629013', + 'info_dict': { + 'id': '75629013', + 'ext': 'mp4', + 'title': 'Key & Peele: Terrorist Interrogation', + 'description': 'md5:8678b246399b070816b12313e8b4eb5c', + 'uploader_id': 'atencio', + 'uploader': 'Peter Atencio', + 'duration': 187, + }, + }, { 'url': 'http://vimeo.com/76979871', 'md5': '3363dd6ffebe3784d56f4132317fd446', @@ -196,8 +211,6 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): video_id = mobj.group('id') if mobj.group('pro') or mobj.group('player'): url = 'http://player.vimeo.com/video/' + video_id - else: - url = 'https://vimeo.com/' + video_id # Retrieve video webpage to extract further information request = compat_urllib_request.Request(url, None, headers) @@ -263,7 +276,7 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): if video_thumbnail is None: video_thumbs = config["video"].get("thumbs") if video_thumbs and isinstance(video_thumbs, dict): - _, video_thumbnail = sorted((int(width), t_url) for (width, t_url) in video_thumbs.items())[-1] + _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1] # Extract video description video_description = None diff --git a/youtube_dl/extractor/vodlocker.py b/youtube_dl/extractor/vodlocker.py index 68c5936..6d3b787 100644 --- a/youtube_dl/extractor/vodlocker.py +++ b/youtube_dl/extractor/vodlocker.py @@ -10,7 +10,7 @@ from ..utils import ( class VodlockerIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?vodlocker.com/(?P[0-9a-zA-Z]+)(?:\..*?)?' + _VALID_URL = r'https?://(?:www\.)?vodlocker\.com/(?P[0-9a-zA-Z]+)(?:\..*?)?' _TESTS = [{ 'url': 'http://vodlocker.com/e8wvyzz4sl42', diff --git a/youtube_dl/extractor/vube.py b/youtube_dl/extractor/vube.py index 7b77865..f1b9e9a 100644 --- a/youtube_dl/extractor/vube.py +++ b/youtube_dl/extractor/vube.py @@ -1,5 +1,6 @@ from __future__ import unicode_literals +import json import re from .common import InfoExtractor @@ -20,12 +21,14 @@ class VubeIE(InfoExtractor): 'ext': 'mp4', 'title': 'Chiara Grispo - Price Tag by Jessie J', 'description': 'md5:8ea652a1f36818352428cb5134933313', - 'thumbnail': 'http://frame.thestaticvube.com/snap/228x128/102e7e63057-5ebc-4f5c-4065-6ce4ebde131f.jpg', + 'thumbnail': 're:^http://frame\.thestaticvube\.com/snap/[0-9x]+/102e7e63057-5ebc-4f5c-4065-6ce4ebde131f\.jpg$', 'uploader': 'Chiara.Grispo', - 'uploader_id': '1u3hX0znhP', 'timestamp': 1388743358, 'upload_date': '20140103', - 'duration': 170.56 + 'duration': 170.56, + 'like_count': int, + 'dislike_count': int, + 'comment_count': int, } }, { @@ -36,12 +39,30 @@ class VubeIE(InfoExtractor): 'ext': 'mp4', 'title': 'My 7 year old Sister and I singing "Alive" by Krewella', 'description': 'md5:40bcacb97796339f1690642c21d56f4a', - 'thumbnail': 'http://frame.thestaticvube.com/snap/228x128/102265d5a9f-0f17-4f6b-5753-adf08484ee1e.jpg', + 'thumbnail': 're:^http://frame\.thestaticvube\.com/snap/[0-9x]+/102265d5a9f-0f17-4f6b-5753-adf08484ee1e\.jpg$', 'uploader': 'Seraina', - 'uploader_id': 'XU9VE2BQ2q', 'timestamp': 1396492438, 'upload_date': '20140403', - 'duration': 240.107 + 'duration': 240.107, + 'like_count': int, + 'dislike_count': int, + 'comment_count': int, + } + }, { + 'url': 'http://vube.com/vote/Siren+Gene/0nmsMY5vEq?n=2&t=s', + 'md5': '0584fc13b50f887127d9d1007589d27f', + 'info_dict': { + 'id': '0nmsMY5vEq', + 'ext': 'mp4', + 'title': 'Frozen - Let It Go Cover by Siren Gene', + 'description': 'My rendition of "Let It Go" originally sung by Idina Menzel.', + 'uploader': 'Siren Gene', + 'uploader_id': 'Siren', + 'thumbnail': 're:^http://frame\.thestaticvube\.com/snap/[0-9x]+/10283ab622a-86c9-4681-51f2-30d1f65774af\.jpg$', + 'duration': 221.788, + 'like_count': int, + 'dislike_count': int, + 'comment_count': int, } } ] @@ -50,8 +71,16 @@ class VubeIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') - video = self._download_json( - 'http://vube.com/api/v2/video/%s' % video_id, video_id, 'Downloading video JSON') + webpage = self._download_webpage(url, video_id) + data_json = self._search_regex( + r'(?s)window\["(?:tapiVideoData|vubeOriginalVideoData)"\]\s*=\s*(\{.*?\n});\n', + webpage, 'video data' + ) + data = json.loads(data_json) + video = ( + data.get('video') or + data) + assert isinstance(video, dict) public_id = video['public_id'] @@ -69,21 +98,31 @@ class VubeIE(InfoExtractor): title = video['title'] description = video.get('description') - thumbnail = video['thumbnail_src'] - if thumbnail.startswith('//'): - thumbnail = 'http:' + thumbnail - uploader = video['user_alias'] - uploader_id = video['user_url_id'] - timestamp = int(video['upload_time']) + thumbnail = self._proto_relative_url( + video.get('thumbnail') or video.get('thumbnail_src'), + scheme='http:') + uploader = data.get('user', {}).get('channel', {}).get('name') or video.get('user_alias') + uploader_id = data.get('user', {}).get('name') + timestamp = int_or_none(video.get('upload_time')) duration = video['duration'] view_count = video.get('raw_view_count') - like_count = video.get('total_likes') - dislike_count= video.get('total_hates') - - comment = self._download_json( - 'http://vube.com/api/video/%s/comment' % video_id, video_id, 'Downloading video comment JSON') + like_count = video.get('rlikes') + if like_count is None: + like_count = video.get('total_likes') + dislike_count = video.get('rhates') + if dislike_count is None: + dislike_count = video.get('total_hates') - comment_count = int_or_none(comment.get('total')) + comments = video.get('comments') + comment_count = None + if comments is None: + comment_data = self._download_json( + 'http://vube.com/api/video/%s/comment' % video_id, + video_id, 'Downloading video comment JSON', fatal=False) + if comment_data is not None: + comment_count = int_or_none(comment_data.get('total')) + else: + comment_count = len(comments) return { 'id': video_id, diff --git a/youtube_dl/extractor/wdr.py b/youtube_dl/extractor/wdr.py index f741ba5..54d37da 100644 --- a/youtube_dl/extractor/wdr.py +++ b/youtube_dl/extractor/wdr.py @@ -55,14 +55,14 @@ class WDRIE(InfoExtractor): }, }, { - 'url': 'http://www.funkhauseuropa.de/av/audiosuepersongsoulbossanova100-audioplayer.html', - 'md5': '24e83813e832badb0a8d7d1ef9ef0691', + 'url': 'http://www.funkhauseuropa.de/av/audioflaviacoelhoamaramar100-audioplayer.html', + 'md5': '99a1443ff29af19f6c52cf6f4dc1f4aa', 'info_dict': { - 'id': 'mdb-463528', + 'id': 'mdb-478135', 'ext': 'mp3', - 'title': 'Süpersong: Soul Bossa Nova', + 'title': 'Flavia Coelho: Amar é Amar', 'description': 'md5:7b29e97e10dfb6e265238b32fa35b23a', - 'upload_date': '20140630', + 'upload_date': '20140717', }, }, ] @@ -81,7 +81,7 @@ class WDRIE(InfoExtractor): ] return self.playlist_result(entries, page_id) - flashvars = compat_urlparse.parse_qs( + flashvars = compat_parse_qs( self._html_search_regex(r'[a-zA-Z0-9_-]+)\.(?P[a-z]+)$', - player_url) + def _signature_cache_id(self, example_sig): + """ Return a string representation of a signature """ + return u'.'.join(compat_str(len(part)) for part in example_sig.split('.')) + + def _extract_signature_function(self, video_id, player_url, example_sig): + id_m = re.match( + r'.*-(?P[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P[a-z]+)$', + player_url) + if not id_m: + raise ExtractorError('Cannot identify player %r' % player_url) player_type = id_m.group('ext') player_id = id_m.group('id') # Read from filesystem cache - func_id = '%s_%s_%d' % (player_type, player_id, slen) + func_id = '%s_%s_%s' % ( + player_type, player_id, self._signature_cache_id(example_sig)) assert os.path.basename(func_id) == func_id cache_dir = get_cachedir(self._downloader.params) @@ -387,7 +393,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): if cache_enabled: try: - test_string = u''.join(map(compat_chr, range(slen))) + test_string = u''.join(map(compat_chr, range(len(example_sig)))) cache_res = res(test_string) cache_spec = [ord(c) for c in cache_res] try: @@ -403,7 +409,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): return res - def _print_sig_code(self, func, slen): + def _print_sig_code(self, func, example_sig): def gen_sig_code(idxs): def _genslice(start, end, step): starts = u'' if start == 0 else str(start) @@ -432,11 +438,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): else: yield _genslice(start, i, step) - test_string = u''.join(map(compat_chr, range(slen))) + test_string = u''.join(map(compat_chr, range(len(example_sig)))) cache_res = func(test_string) cache_spec = [ord(c) for c in cache_res] expr_code = u' + '.join(gen_sig_code(cache_spec)) - code = u'if len(s) == %d:\n return %s\n' % (slen, expr_code) + signature_id_tuple = '(%s)' % ( + ', '.join(compat_str(len(p)) for p in example_sig.split('.'))) + code = (u'if tuple(len(p) for p in s.split(\'.\')) == %s:\n' + u' return %s\n') % (signature_id_tuple, expr_code) self.to_screen(u'Extracted signature function:\n' + code) def _parse_sig_js(self, jscode): @@ -449,417 +458,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): return lambda s: initial_function([s]) def _parse_sig_swf(self, file_contents): - if file_contents[1:3] != b'WS': - raise ExtractorError( - u'Not an SWF file; header is %r' % file_contents[:3]) - if file_contents[:1] == b'C': - content = zlib.decompress(file_contents[8:]) - else: - raise NotImplementedError(u'Unsupported compression format %r' % - file_contents[:1]) - - def extract_tags(content): - pos = 0 - while pos < len(content): - header16 = struct.unpack('> 6 - tag_len = header16 & 0x3f - if tag_len == 0x3f: - tag_len = struct.unpack('> 4 - methods = {} - if kind in [0x00, 0x06]: # Slot or Const - u30() # Slot id - u30() # type_name_idx - vindex = u30() - if vindex != 0: - read_byte() # vkind - elif kind in [0x01, 0x02, 0x03]: # Method / Getter / Setter - u30() # disp_id - method_idx = u30() - methods[multinames[trait_name_idx]] = method_idx - elif kind == 0x04: # Class - u30() # slot_id - u30() # classi - elif kind == 0x05: # Function - u30() # slot_id - function_idx = u30() - methods[function_idx] = multinames[trait_name_idx] - else: - raise ExtractorError(u'Unsupported trait kind %d' % kind) - - if attrs & 0x4 != 0: # Metadata present - metadata_count = u30() - for _c3 in range(metadata_count): - u30() # metadata index - - return methods - - # Classes + swfi = SWFInterpreter(file_contents) TARGET_CLASSNAME = u'SignatureDecipher' - searched_idx = multinames.index(TARGET_CLASSNAME) - searched_class_id = None - class_count = u30() - for class_id in range(class_count): - name_idx = u30() - if name_idx == searched_idx: - # We found the class we're looking for! - searched_class_id = class_id - u30() # super_name idx - flags = read_byte() - if flags & 0x08 != 0: # Protected namespace is present - u30() # protected_ns_idx - intrf_count = u30() - for _c2 in range(intrf_count): - u30() - u30() # iinit - trait_count = u30() - for _c2 in range(trait_count): - parse_traits_info() - - if searched_class_id is None: - raise ExtractorError(u'Target class %r not found' % - TARGET_CLASSNAME) - - method_names = {} - method_idxs = {} - for class_id in range(class_count): - u30() # cinit - trait_count = u30() - for _c2 in range(trait_count): - trait_methods = parse_traits_info() - if class_id == searched_class_id: - method_names.update(trait_methods.items()) - method_idxs.update(dict( - (idx, name) - for name, idx in trait_methods.items())) - - # Scripts - script_count = u30() - for _c in range(script_count): - u30() # init - trait_count = u30() - for _c2 in range(trait_count): - parse_traits_info() - - # Method bodies - method_body_count = u30() - Method = collections.namedtuple('Method', ['code', 'local_count']) - methods = {} - for _c in range(method_body_count): - method_idx = u30() - u30() # max_stack - local_count = u30() - u30() # init_scope_depth - u30() # max_scope_depth - code_length = u30() - code = read_bytes(code_length) - if method_idx in method_idxs: - m = Method(code, local_count) - methods[method_idxs[method_idx]] = m - exception_count = u30() - for _c2 in range(exception_count): - u30() # from - u30() # to - u30() # target - u30() # exc_type - u30() # var_name - trait_count = u30() - for _c2 in range(trait_count): - parse_traits_info() - - assert p + code_reader.tell() == len(code_tag) - assert len(methods) == len(method_idxs) - - method_pyfunctions = {} - - def extract_function(func_name): - if func_name in method_pyfunctions: - return method_pyfunctions[func_name] - if func_name not in methods: - raise ExtractorError(u'Cannot find function %r' % func_name) - m = methods[func_name] - - def resfunc(args): - registers = ['(this)'] + list(args) + [None] * m.local_count - stack = [] - coder = io.BytesIO(m.code) - while True: - opcode = struct.unpack('!B', coder.read(1))[0] - if opcode == 36: # pushbyte - v = struct.unpack('!B', coder.read(1))[0] - stack.append(v) - elif opcode == 44: # pushstring - idx = u30(coder) - stack.append(constant_strings[idx]) - elif opcode == 48: # pushscope - # We don't implement the scope register, so we'll just - # ignore the popped value - stack.pop() - elif opcode == 70: # callproperty - index = u30(coder) - mname = multinames[index] - arg_count = u30(coder) - args = list(reversed( - [stack.pop() for _ in range(arg_count)])) - obj = stack.pop() - if mname == u'split': - assert len(args) == 1 - assert isinstance(args[0], compat_str) - assert isinstance(obj, compat_str) - if args[0] == u'': - res = list(obj) - else: - res = obj.split(args[0]) - stack.append(res) - elif mname == u'slice': - assert len(args) == 1 - assert isinstance(args[0], int) - assert isinstance(obj, list) - res = obj[args[0]:] - stack.append(res) - elif mname == u'join': - assert len(args) == 1 - assert isinstance(args[0], compat_str) - assert isinstance(obj, list) - res = args[0].join(obj) - stack.append(res) - elif mname in method_pyfunctions: - stack.append(method_pyfunctions[mname](args)) - else: - raise NotImplementedError( - u'Unsupported property %r on %r' - % (mname, obj)) - elif opcode == 72: # returnvalue - res = stack.pop() - return res - elif opcode == 79: # callpropvoid - index = u30(coder) - mname = multinames[index] - arg_count = u30(coder) - args = list(reversed( - [stack.pop() for _ in range(arg_count)])) - obj = stack.pop() - if mname == u'reverse': - assert isinstance(obj, list) - obj.reverse() - else: - raise NotImplementedError( - u'Unsupported (void) property %r on %r' - % (mname, obj)) - elif opcode == 93: # findpropstrict - index = u30(coder) - mname = multinames[index] - res = extract_function(mname) - stack.append(res) - elif opcode == 97: # setproperty - index = u30(coder) - value = stack.pop() - idx = stack.pop() - obj = stack.pop() - assert isinstance(obj, list) - assert isinstance(idx, int) - obj[idx] = value - elif opcode == 98: # getlocal - index = u30(coder) - stack.append(registers[index]) - elif opcode == 99: # setlocal - index = u30(coder) - value = stack.pop() - registers[index] = value - elif opcode == 102: # getproperty - index = u30(coder) - pname = multinames[index] - if pname == u'length': - obj = stack.pop() - assert isinstance(obj, list) - stack.append(len(obj)) - else: # Assume attribute access - idx = stack.pop() - assert isinstance(idx, int) - obj = stack.pop() - assert isinstance(obj, list) - stack.append(obj[idx]) - elif opcode == 128: # coerce - u30(coder) - elif opcode == 133: # coerce_s - assert isinstance(stack[-1], (type(None), compat_str)) - elif opcode == 164: # modulo - value2 = stack.pop() - value1 = stack.pop() - res = value1 % value2 - stack.append(res) - elif opcode == 208: # getlocal_0 - stack.append(registers[0]) - elif opcode == 209: # getlocal_1 - stack.append(registers[1]) - elif opcode == 210: # getlocal_2 - stack.append(registers[2]) - elif opcode == 211: # getlocal_3 - stack.append(registers[3]) - elif opcode == 214: # setlocal_2 - registers[2] = stack.pop() - elif opcode == 215: # setlocal_3 - registers[3] = stack.pop() - else: - raise NotImplementedError( - u'Unsupported opcode %d' % opcode) - - method_pyfunctions[func_name] = resfunc - return resfunc - - initial_function = extract_function(u'decipher') + searched_class = swfi.extract_class(TARGET_CLASSNAME) + initial_function = swfi.extract_function(searched_class, u'decipher') return lambda s: initial_function([s]) def _decrypt_signature(self, s, video_id, player_url, age_gate=False): @@ -871,20 +473,20 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): if player_url.startswith(u'//'): player_url = u'https:' + player_url try: - player_id = (player_url, len(s)) + player_id = (player_url, self._signature_cache_id(s)) if player_id not in self._player_cache: func = self._extract_signature_function( - video_id, player_url, len(s) + video_id, player_url, s ) self._player_cache[player_id] = func func = self._player_cache[player_id] if self._downloader.params.get('youtube_print_sig_code'): - self._print_sig_code(func, len(s)) + self._print_sig_code(func, s) return func(s) except Exception as e: tb = traceback.format_exc() raise ExtractorError( - u'Automatic signature extraction failed: ' + tb, cause=e) + u'Signature extraction failed: ' + tb, cause=e) def _get_available_subtitles(self, video_id, webpage): try: @@ -1014,14 +616,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): age_gate = True # We simulate the access to the video from www.youtube.com/v/{video_id} # this can be viewed without login into Youtube - data = compat_urllib_parse.urlencode({'video_id': video_id, - 'el': 'player_embedded', - 'gl': 'US', - 'hl': 'en', - 'eurl': 'https://youtube.googleapis.com/v/' + video_id, - 'asv': 3, - 'sts':'1588', - }) + data = compat_urllib_parse.urlencode({ + 'video_id': video_id, + 'eurl': 'https://youtube.googleapis.com/v/' + video_id, + 'sts': self._search_regex( + r'"sts"\s*:\s*(\d+)', video_webpage, 'sts'), + }) video_info_url = proto + '://www.youtube.com/get_video_info?' + data video_info_webpage = self._download_webpage(video_info_url, video_id, note=False, @@ -1214,43 +814,54 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): url_map = {} for url_data_str in encoded_url_map.split(','): url_data = compat_parse_qs(url_data_str) - if 'itag' in url_data and 'url' in url_data: - url = url_data['url'][0] - if 'sig' in url_data: - url += '&signature=' + url_data['sig'][0] - elif 's' in url_data: - encrypted_sig = url_data['s'][0] - if self._downloader.params.get('verbose'): - if age_gate: - if player_url is None: - player_version = 'unknown' - else: - player_version = self._search_regex( - r'-(.+)\.swf$', player_url, - u'flash player', fatal=False) + if 'itag' not in url_data or 'url' not in url_data: + continue + format_id = url_data['itag'][0] + url = url_data['url'][0] + + if 'sig' in url_data: + url += '&signature=' + url_data['sig'][0] + elif 's' in url_data: + encrypted_sig = url_data['s'][0] + + if not age_gate: + jsplayer_url_json = self._search_regex( + r'"assets":.+?"js":\s*("[^"]+")', + video_webpage, u'JS player URL') + player_url = json.loads(jsplayer_url_json) + if player_url is None: + player_url_json = self._search_regex( + r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")', + video_webpage, u'age gate player URL') + player_url = json.loads(player_url_json) + + if self._downloader.params.get('verbose'): + if player_url is None: + player_version = 'unknown' + player_desc = 'unknown' + else: + if player_url.endswith('swf'): + player_version = self._search_regex( + r'-(.+?)(?:/watch_as3)?\.swf$', player_url, + u'flash player', fatal=False) player_desc = 'flash player %s' % player_version else: player_version = self._search_regex( - r'html5player-(.+?)\.js', video_webpage, + r'html5player-([^/]+?)(?:/html5player)?\.js', + player_url, 'html5 player', fatal=False) player_desc = u'html5 player %s' % player_version - parts_sizes = u'.'.join(compat_str(len(part)) for part in encrypted_sig.split('.')) - self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % - (len(encrypted_sig), parts_sizes, url_data['itag'][0], player_desc)) - - if not age_gate: - jsplayer_url_json = self._search_regex( - r'"assets":.+?"js":\s*("[^"]+")', - video_webpage, u'JS player URL') - player_url = json.loads(jsplayer_url_json) - - signature = self._decrypt_signature( - encrypted_sig, video_id, player_url, age_gate) - url += '&signature=' + signature - if 'ratebypass' not in url: - url += '&ratebypass=yes' - url_map[url_data['itag'][0]] = url + parts_sizes = self._signature_cache_id(encrypted_sig) + self.to_screen(u'{%s} signature length %s, %s' % + (format_id, parts_sizes, player_desc)) + + signature = self._decrypt_signature( + encrypted_sig, video_id, player_url, age_gate) + url += '&signature=' + signature + if 'ratebypass' not in url: + url += '&ratebypass=yes' + url_map[format_id] = url formats = _map_to_format_list(url_map) elif video_info.get('hlsvp'): manifest_url = video_info['hlsvp'][0] diff --git a/youtube_dl/jsinterp.py b/youtube_dl/jsinterp.py index ae5bca2..c40cd37 100644 --- a/youtube_dl/jsinterp.py +++ b/youtube_dl/jsinterp.py @@ -1,5 +1,6 @@ from __future__ import unicode_literals +import json import re from .utils import ( @@ -40,8 +41,9 @@ class JSInterpreter(object): assign = lambda v: v expr = stmt[len('return '):] else: - raise ExtractorError( - 'Cannot determine left side of statement in %r' % stmt) + # Try interpreting it as an expression + expr = stmt + assign = lambda v: v v = self.interpret_expression(expr, local_vars, allow_recursion) return assign(v) @@ -53,35 +55,63 @@ class JSInterpreter(object): if expr.isalpha(): return local_vars[expr] - m = re.match(r'^(?P[a-z]+)\.(?P.*)$', expr) + try: + return json.loads(expr) + except ValueError: + pass + + m = re.match( + r'^(?P[a-zA-Z0-9_]+)\.(?P[^(]+)(?:\(+(?P[^()]*)\))?$', + expr) if m: + variable = m.group('var') member = m.group('member') - variable = m.group('in') + arg_str = m.group('args') - if variable not in local_vars: + if variable in local_vars: + obj = local_vars[variable] + else: if variable not in self._objects: self._objects[variable] = self.extract_object(variable) obj = self._objects[variable] - key, args = member.split('(', 1) - args = args.strip(')') - argvals = [int(v) if v.isdigit() else local_vars[v] - for v in args.split(',')] - return obj[key](argvals) - - val = local_vars[variable] - if member == 'split("")': - return list(val) - if member == 'join("")': - return ''.join(val) - if member == 'length': - return len(val) - if member == 'reverse()': - return val[::-1] - slice_m = re.match(r'slice\((?P.*)\)', member) - if slice_m: - idx = self.interpret_expression( - slice_m.group('idx'), local_vars, allow_recursion - 1) - return val[idx:] + + if arg_str is None: + # Member access + if member == 'length': + return len(obj) + return obj[member] + + assert expr.endswith(')') + # Function call + if arg_str == '': + argvals = tuple() + else: + argvals = tuple([ + self.interpret_expression(v, local_vars, allow_recursion) + for v in arg_str.split(',')]) + + if member == 'split': + assert argvals == ('',) + return list(obj) + if member == 'join': + assert len(argvals) == 1 + return argvals[0].join(obj) + if member == 'reverse': + assert len(argvals) == 0 + obj.reverse() + return obj + if member == 'slice': + assert len(argvals) == 1 + return obj[argvals[0]:] + if member == 'splice': + assert isinstance(obj, list) + index, howMany = argvals + res = [] + for i in range(index, min(index + howMany, len(obj))): + res.append(obj.pop(index)) + return res + + return obj[member](argvals) m = re.match( r'^(?P[a-z]+)\[(?P.+)\]$', expr) @@ -103,10 +133,11 @@ class JSInterpreter(object): r'^(?P[a-zA-Z$]+)\((?P[a-z0-9,]+)\)$', expr) if m: fname = m.group('func') + argvals = tuple([ + int(v) if v.isdigit() else local_vars[v] + for v in m.group('args').split(',')]) if fname not in self._functions: self._functions[fname] = self.extract_function(fname) - argvals = [int(v) if v.isdigit() else local_vars[v] - for v in m.group('args').split(',')] return self._functions[fname](argvals) raise ExtractorError('Unsupported JS expression %r' % expr) @@ -114,13 +145,13 @@ class JSInterpreter(object): obj = {} obj_m = re.search( (r'(?:var\s+)?%s\s*=\s*\{' % re.escape(objname)) + - r'\s*(?P([a-zA-Z$]+\s*:\s*function\(.*?\)\s*\{.*?\})*)' + + r'\s*(?P([a-zA-Z$0-9]+\s*:\s*function\(.*?\)\s*\{.*?\})*)' + r'\}\s*;', self.code) fields = obj_m.group('fields') # Currently, it only supports function definitions fields_m = re.finditer( - r'(?P[a-zA-Z$]+)\s*:\s*function' + r'(?P[a-zA-Z$0-9]+)\s*:\s*function' r'\((?P[a-z,]+)\){(?P[^}]+)}', fields) for f in fields_m: diff --git a/youtube_dl/postprocessor/ffmpeg.py b/youtube_dl/postprocessor/ffmpeg.py index 45328ed..8c5f7c4 100644 --- a/youtube_dl/postprocessor/ffmpeg.py +++ b/youtube_dl/postprocessor/ffmpeg.py @@ -18,14 +18,15 @@ from ..utils import ( ) - class FFmpegPostProcessorError(PostProcessingError): pass + class FFmpegPostProcessor(PostProcessor): - def __init__(self,downloader=None): + def __init__(self, downloader=None, deletetempfiles=False): PostProcessor.__init__(self, downloader) self._exes = self.detect_executables() + self._deletetempfiles = deletetempfiles @staticmethod def detect_executables(): @@ -60,6 +61,9 @@ class FFmpegPostProcessor(PostProcessor): stderr = stderr.decode('utf-8', 'replace') msg = stderr.strip().split('\n')[-1] raise FFmpegPostProcessorError(msg) + if self._deletetempfiles: + for ipath in input_paths: + os.remove(ipath) def run_ffmpeg(self, path, out_path, opts): self.run_ffmpeg_multiple_files([path], out_path, opts) diff --git a/youtube_dl/swfinterp.py b/youtube_dl/swfinterp.py new file mode 100644 index 0000000..b63c65b --- /dev/null +++ b/youtube_dl/swfinterp.py @@ -0,0 +1,609 @@ +from __future__ import unicode_literals + +import collections +import io +import zlib + +from .utils import ( + compat_str, + ExtractorError, + struct_unpack, +) + + +def _extract_tags(file_contents): + if file_contents[1:3] != b'WS': + raise ExtractorError( + 'Not an SWF file; header is %r' % file_contents[:3]) + if file_contents[:1] == b'C': + content = zlib.decompress(file_contents[8:]) + else: + raise NotImplementedError( + 'Unsupported compression format %r' % + file_contents[:1]) + + # Determine number of bits in framesize rectangle + framesize_nbits = struct_unpack('!B', content[:1])[0] >> 3 + framesize_len = (5 + 4 * framesize_nbits + 7) // 8 + + pos = framesize_len + 2 + 2 + while pos < len(content): + header16 = struct_unpack('> 6 + tag_len = header16 & 0x3f + if tag_len == 0x3f: + tag_len = struct_unpack('= 0x80) else b'\x00' + return struct_unpack('= 0 + resb = reader.read(count) + assert len(resb) == count + return resb + + +def _read_byte(reader): + resb = _read_bytes(1, reader=reader) + res = struct_unpack('> 4 + methods = {} + if kind in [0x00, 0x06]: # Slot or Const + u30() # Slot id + u30() # type_name_idx + vindex = u30() + if vindex != 0: + read_byte() # vkind + elif kind in [0x01, 0x02, 0x03]: # Method / Getter / Setter + u30() # disp_id + method_idx = u30() + methods[self.multinames[trait_name_idx]] = method_idx + elif kind == 0x04: # Class + u30() # slot_id + u30() # classi + elif kind == 0x05: # Function + u30() # slot_id + function_idx = u30() + methods[function_idx] = self.multinames[trait_name_idx] + else: + raise ExtractorError('Unsupported trait kind %d' % kind) + + if attrs & 0x4 != 0: # Metadata present + metadata_count = u30() + for _c3 in range(metadata_count): + u30() # metadata index + + return methods + + # Classes + class_count = u30() + classes = [] + for class_id in range(class_count): + name_idx = u30() + + cname = self.multinames[name_idx] + avm_class = _AVMClass(name_idx, cname) + classes.append(avm_class) + + u30() # super_name idx + flags = read_byte() + if flags & 0x08 != 0: # Protected namespace is present + u30() # protected_ns_idx + intrf_count = u30() + for _c2 in range(intrf_count): + u30() + u30() # iinit + trait_count = u30() + for _c2 in range(trait_count): + trait_methods = parse_traits_info() + avm_class.register_methods(trait_methods) + + assert len(classes) == class_count + self._classes_by_name = dict((c.name, c) for c in classes) + + for avm_class in classes: + u30() # cinit + trait_count = u30() + for _c2 in range(trait_count): + trait_methods = parse_traits_info() + avm_class.register_methods(trait_methods) + + # Scripts + script_count = u30() + for _c in range(script_count): + u30() # init + trait_count = u30() + for _c2 in range(trait_count): + parse_traits_info() + + # Method bodies + method_body_count = u30() + Method = collections.namedtuple('Method', ['code', 'local_count']) + for _c in range(method_body_count): + method_idx = u30() + u30() # max_stack + local_count = u30() + u30() # init_scope_depth + u30() # max_scope_depth + code_length = u30() + code = read_bytes(code_length) + for avm_class in classes: + if method_idx in avm_class.method_idxs: + m = Method(code, local_count) + avm_class.methods[avm_class.method_idxs[method_idx]] = m + exception_count = u30() + for _c2 in range(exception_count): + u30() # from + u30() # to + u30() # target + u30() # exc_type + u30() # var_name + trait_count = u30() + for _c2 in range(trait_count): + parse_traits_info() + + assert p + code_reader.tell() == len(code_tag) + + def extract_class(self, class_name): + try: + return self._classes_by_name[class_name] + except KeyError: + raise ExtractorError('Class %r not found' % class_name) + + def extract_function(self, avm_class, func_name): + if func_name in avm_class.method_pyfunctions: + return avm_class.method_pyfunctions[func_name] + if func_name in self._classes_by_name: + return self._classes_by_name[func_name].make_object() + if func_name not in avm_class.methods: + raise ExtractorError('Cannot find function %s.%s' % ( + avm_class.name, func_name)) + m = avm_class.methods[func_name] + + def resfunc(args): + # Helper functions + coder = io.BytesIO(m.code) + s24 = lambda: _s24(coder) + u30 = lambda: _u30(coder) + + registers = [avm_class.variables] + list(args) + [None] * m.local_count + stack = [] + scopes = collections.deque([ + self._classes_by_name, avm_class.variables]) + while True: + opcode = _read_byte(coder) + if opcode == 17: # iftrue + offset = s24() + value = stack.pop() + if value: + coder.seek(coder.tell() + offset) + elif opcode == 18: # iffalse + offset = s24() + value = stack.pop() + if not value: + coder.seek(coder.tell() + offset) + elif opcode == 36: # pushbyte + v = _read_byte(coder) + stack.append(v) + elif opcode == 42: # dup + value = stack[-1] + stack.append(value) + elif opcode == 44: # pushstring + idx = u30() + stack.append(self.constant_strings[idx]) + elif opcode == 48: # pushscope + new_scope = stack.pop() + scopes.append(new_scope) + elif opcode == 66: # construct + arg_count = u30() + args = list(reversed( + [stack.pop() for _ in range(arg_count)])) + obj = stack.pop() + res = obj.avm_class.make_object() + stack.append(res) + elif opcode == 70: # callproperty + index = u30() + mname = self.multinames[index] + arg_count = u30() + args = list(reversed( + [stack.pop() for _ in range(arg_count)])) + obj = stack.pop() + + if isinstance(obj, _AVMClass_Object): + func = self.extract_function(obj.avm_class, mname) + res = func(args) + stack.append(res) + continue + elif isinstance(obj, _ScopeDict): + if mname in obj.avm_class.method_names: + func = self.extract_function(obj.avm_class, mname) + res = func(args) + else: + res = obj[mname] + stack.append(res) + continue + elif isinstance(obj, compat_str): + if mname == 'split': + assert len(args) == 1 + assert isinstance(args[0], compat_str) + if args[0] == '': + res = list(obj) + else: + res = obj.split(args[0]) + stack.append(res) + continue + elif isinstance(obj, list): + if mname == 'slice': + assert len(args) == 1 + assert isinstance(args[0], int) + res = obj[args[0]:] + stack.append(res) + continue + elif mname == 'join': + assert len(args) == 1 + assert isinstance(args[0], compat_str) + res = args[0].join(obj) + stack.append(res) + continue + raise NotImplementedError( + 'Unsupported property %r on %r' + % (mname, obj)) + elif opcode == 72: # returnvalue + res = stack.pop() + return res + elif opcode == 74: # constructproperty + index = u30() + arg_count = u30() + args = list(reversed( + [stack.pop() for _ in range(arg_count)])) + obj = stack.pop() + + mname = self.multinames[index] + assert isinstance(obj, _AVMClass) + + # We do not actually call the constructor for now; + # we just pretend it does nothing + stack.append(obj.make_object()) + elif opcode == 79: # callpropvoid + index = u30() + mname = self.multinames[index] + arg_count = u30() + args = list(reversed( + [stack.pop() for _ in range(arg_count)])) + obj = stack.pop() + if mname == 'reverse': + assert isinstance(obj, list) + obj.reverse() + else: + raise NotImplementedError( + 'Unsupported (void) property %r on %r' + % (mname, obj)) + elif opcode == 86: # newarray + arg_count = u30() + arr = [] + for i in range(arg_count): + arr.append(stack.pop()) + arr = arr[::-1] + stack.append(arr) + elif opcode == 93: # findpropstrict + index = u30() + mname = self.multinames[index] + for s in reversed(scopes): + if mname in s: + res = s + break + else: + res = scopes[0] + stack.append(res[mname]) + elif opcode == 94: # findproperty + index = u30() + mname = self.multinames[index] + for s in reversed(scopes): + if mname in s: + res = s + break + else: + res = avm_class.variables + stack.append(res) + elif opcode == 96: # getlex + index = u30() + mname = self.multinames[index] + for s in reversed(scopes): + if mname in s: + scope = s + break + else: + scope = avm_class.variables + # I cannot find where static variables are initialized + # so let's just return None + res = scope.get(mname) + stack.append(res) + elif opcode == 97: # setproperty + index = u30() + value = stack.pop() + idx = self.multinames[index] + if isinstance(idx, _Multiname): + idx = stack.pop() + obj = stack.pop() + obj[idx] = value + elif opcode == 98: # getlocal + index = u30() + stack.append(registers[index]) + elif opcode == 99: # setlocal + index = u30() + value = stack.pop() + registers[index] = value + elif opcode == 102: # getproperty + index = u30() + pname = self.multinames[index] + if pname == 'length': + obj = stack.pop() + assert isinstance(obj, list) + stack.append(len(obj)) + else: # Assume attribute access + idx = stack.pop() + assert isinstance(idx, int) + obj = stack.pop() + assert isinstance(obj, list) + stack.append(obj[idx]) + elif opcode == 115: # convert_ + value = stack.pop() + intvalue = int(value) + stack.append(intvalue) + elif opcode == 128: # coerce + u30() + elif opcode == 133: # coerce_s + assert isinstance(stack[-1], (type(None), compat_str)) + elif opcode == 160: # add + value2 = stack.pop() + value1 = stack.pop() + res = value1 + value2 + stack.append(res) + elif opcode == 161: # subtract + value2 = stack.pop() + value1 = stack.pop() + res = value1 - value2 + stack.append(res) + elif opcode == 164: # modulo + value2 = stack.pop() + value1 = stack.pop() + res = value1 % value2 + stack.append(res) + elif opcode == 175: # greaterequals + value2 = stack.pop() + value1 = stack.pop() + result = value1 >= value2 + stack.append(result) + elif opcode == 208: # getlocal_0 + stack.append(registers[0]) + elif opcode == 209: # getlocal_1 + stack.append(registers[1]) + elif opcode == 210: # getlocal_2 + stack.append(registers[2]) + elif opcode == 211: # getlocal_3 + stack.append(registers[3]) + elif opcode == 212: # setlocal_0 + registers[0] = stack.pop() + elif opcode == 213: # setlocal_1 + registers[1] = stack.pop() + elif opcode == 214: # setlocal_2 + registers[2] = stack.pop() + elif opcode == 215: # setlocal_3 + registers[3] = stack.pop() + else: + raise NotImplementedError( + 'Unsupported opcode %d' % opcode) + + avm_class.method_pyfunctions[func_name] = resfunc + return resfunc + diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 64a9618..e40b367 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -91,11 +91,9 @@ except ImportError: compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w') try: - from urllib.parse import parse_qs as compat_parse_qs -except ImportError: # Python 2 - # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib. - # Python 2's version is apparently totally broken - def _unquote(string, encoding='utf-8', errors='replace'): + from urllib.parse import unquote as compat_urllib_parse_unquote +except ImportError: + def compat_urllib_parse_unquote(string, encoding='utf-8', errors='replace'): if string == '': return string res = string.split('%') @@ -130,6 +128,13 @@ except ImportError: # Python 2 string += pct_sequence.decode(encoding, errors) return string + +try: + from urllib.parse import parse_qs as compat_parse_qs +except ImportError: # Python 2 + # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib. + # Python 2's version is apparently totally broken + def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False, encoding='utf-8', errors='replace'): qs, _coerce_result = qs, unicode @@ -149,10 +154,12 @@ except ImportError: # Python 2 continue if len(nv[1]) or keep_blank_values: name = nv[0].replace('+', ' ') - name = _unquote(name, encoding=encoding, errors=errors) + name = compat_urllib_parse_unquote( + name, encoding=encoding, errors=errors) name = _coerce_result(name) value = nv[1].replace('+', ' ') - value = _unquote(value, encoding=encoding, errors=errors) + value = compat_urllib_parse_unquote( + value, encoding=encoding, errors=errors) value = _coerce_result(value) r.append((name, value)) return r @@ -235,8 +242,8 @@ else: if sys.version_info >= (2,7): def find_xpath_attr(node, xpath, key, val): """ Find the xpath xpath[@key=val] """ - assert re.match(r'^[a-zA-Z]+$', key) - assert re.match(r'^[a-zA-Z0-9@\s:._]*$', val) + assert re.match(r'^[a-zA-Z-]+$', key) + assert re.match(r'^[a-zA-Z0-9@\s:._-]*$', val) expr = xpath + u"[@%s='%s']" % (key, val) return node.find(expr) else: @@ -845,6 +852,8 @@ def unified_strdate(date_str): return upload_date def determine_ext(url, default_ext=u'unknown_video'): + if url is None: + return default_ext guess = url.partition(u'?')[0].rpartition(u'.')[2] if re.match(r'^[A-Za-z0-9]+$', guess): return guess @@ -1193,11 +1202,6 @@ def format_bytes(bytes): return u'%.2f%s' % (converted, suffix) -def str_to_int(int_str): - int_str = re.sub(r'[,\.]', u'', int_str) - return int(int_str) - - def get_term_width(): columns = os.environ.get('COLUMNS', None) if columns: @@ -1265,15 +1269,22 @@ class HEADRequest(compat_urllib_request.Request): return "HEAD" -def int_or_none(v, scale=1, default=None, get_attr=None): +def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1): if get_attr: if v is not None: v = getattr(v, get_attr, None) - return default if v is None else (int(v) // scale) + return default if v is None else (int(v) * invscale // scale) + + +def str_to_int(int_str): + if int_str is None: + return None + int_str = re.sub(r'[,\.]', u'', int_str) + return int(int_str) -def float_or_none(v, scale=1, default=None): - return default if v is None else (float(v) / scale) +def float_or_none(v, scale=1, invscale=1, default=None): + return default if v is None else (float(v) * invscale / scale) def parse_duration(s): diff --git a/youtube_dl/version.py b/youtube_dl/version.py index 4d606c3..08b5339 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,2 +1,2 @@ -__version__ = '2014.07.15' +__version__ = '2014.08.05'