release 2015.02.02.1

'' might actually be passed in, so check for None.

.travis.yml

@@ -4,6 +4,9 @@ python:
   - "2.7"
   - "3.3"
   - "3.4"
+before_install:
+  - sudo apt-get update -qq
+  - sudo apt-get install -yqq rtmpdump
 script: nosetests test --verbose
 notifications:
   email:

AUTHORS

@@ -104,3 +104,7 @@ Ondřej Caletka
 Dinesh S
 Johan K. Jensen
 Yen Chi Hsuan
+Enam Mijbah Noor
+David Luhmer
+Shaya Goldberg
+Paul Hartmann

README.md

@@ -93,6 +93,14 @@ which means you can modify it, redistribute it or use it however you like.
 ## Video Selection:
     --playlist-start NUMBER          playlist video to start at (default is 1)
     --playlist-end NUMBER            playlist video to end at (default is last)
+    --playlist-items ITEM_SPEC       playlist video items to download. Specify
+                                     indices of the videos in the playlist
+                                     seperated by commas like: "--playlist-items
+                                     1,2,5,8" if you want to download videos
+                                     indexed 1, 2, 5, 8 in the playlist. You can
+                                     specify range: "--playlist-items
+                                     1-3,7,10-13", it will download the videos
+                                     at index 1, 2, 3, 7, 10, 11, 12 and 13.
     --match-title REGEX              download only matching titles (regex or
                                      caseless sub-string)
     --reject-title REGEX             skip download for matching titles (regex or
@@ -124,7 +132,8 @@ which means you can modify it, redistribute it or use it however you like.
 ## Download Options:
     -r, --rate-limit LIMIT           maximum download rate in bytes per second
                                      (e.g. 50K or 4.2M)
-    -R, --retries RETRIES            number of retries (default is 10)
+    -R, --retries RETRIES            number of retries (default is 10), or
+                                     "infinite".
     --buffer-size SIZE               size of download buffer (e.g. 1024 or 16K)
                                      (default is 1024)
     --no-resize-buffer               do not automatically adjust the buffer
@@ -132,6 +141,11 @@ which means you can modify it, redistribute it or use it however you like.
                                      automatically resized from an initial value
                                      of SIZE.
     --playlist-reverse               Download playlist videos in reverse order
+    --xattr-set-filesize             (experimental) set file xattribute
+                                     ytdl.filesize with expected filesize
+    --external-downloader COMMAND    (experimental) Use the specified external
+                                     downloader. Currently supports
+                                     aria2c,curl,wget
 
 ## Filesystem Options:
     -a, --batch-file FILE            file containing URLs to download ('-' for
@@ -191,7 +205,6 @@ which means you can modify it, redistribute it or use it however you like.
     --write-info-json                write video metadata to a .info.json file
     --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
@@ -206,6 +219,12 @@ which means you can modify it, redistribute it or use it however you like.
     --no-cache-dir                   Disable filesystem caching
     --rm-cache-dir                   Delete all filesystem cache files
 
+## Thumbnail images:
+    --write-thumbnail                write thumbnail image to disk
+    --write-all-thumbnails           write all thumbnail image formats to disk
+    --list-thumbnails                Simulate and list all available thumbnail
+                                     formats
+
 ## Verbosity / Simulation Options:
     -q, --quiet                      activates quiet mode
     --no-warnings                    Ignore warnings
@@ -259,6 +278,8 @@ which means you can modify it, redistribute it or use it however you like.
     --bidi-workaround                Work around terminals that lack
                                      bidirectional text support. Requires bidiv
                                      or fribidi executable in PATH
+    --sleep-interval SECONDS         Number of seconds to sleep before each
+                                     download.
 
 ## Video Format Options:
     -f, --format FORMAT              video format code, specify the order of
@@ -271,9 +292,9 @@ which means you can modify it, redistribute it or use it however you like.
                                      video results by putting a condition in
                                      brackets, as in -f "best[height=720]" (or
                                      -f "[filesize>10M]").  This works for
-                                     filesize, height, width, tbr, abr, and vbr
-                                     and the comparisons <, <=, >, >=, =, != .
-                                     Formats for which the value is not known
+                                     filesize, height, width, tbr, abr, vbr, and
+                                     fps and the comparisons <, <=, >, >=, =, !=
+                                     . Formats for which the value is not known
                                      are excluded unless you put a question mark
                                      (?) after the operator. You can combine
                                      format filters, so  -f "[height <=?
@@ -347,11 +368,11 @@ which means you can modify it, redistribute it or use it however you like.
     --add-metadata                   write metadata to the video file
     --xattrs                         write metadata to the video file's xattrs
                                      (using dublin core and xdg standards)
-    --fixup POLICY                   (experimental) Automatically correct known
-                                     faults of the file. One of never (do
-                                     nothing), warn (only emit a warning),
-                                     detect_or_warn(check whether we can do
-                                     anything about it, warn otherwise
+    --fixup POLICY                   Automatically correct known faults of the
+                                     file. One of never (do nothing), warn (only
+                                     emit a warning), detect_or_warn(the
+                                     default; fix file if we can, warn
+                                     otherwise)
     --prefer-avconv                  Prefer avconv over ffmpeg for running the
                                      postprocessors (default)
     --prefer-ffmpeg                  Prefer ffmpeg over avconv for running the
@@ -504,9 +525,16 @@ From then on, after restarting your shell, you will be able to access both youtu
 
 Use the `-o` to specify an [output template](#output-template), for example `-o "/home/user/videos/%(title)s-%(id)s.%(ext)s"`. If you want this for all of your downloads, put the option into your [configuration file](#configuration).
 
+### How do I download a video starting with a `-` ?
+
+Either prepend `http://www.youtube.com/watch?v=` or separate the ID from the options with `--`:
+
+    youtube-dl -- -wNyEUrxzFU
+    youtube-dl "http://www.youtube.com/watch?v=-wNyEUrxzFU"
+
 ### How can I detect whether a given URL is supported by youtube-dl?
 
-For one, have a look at the [list of supported sites](docs/supportedsites). Note that it can sometimes happen that the site changes its URL scheme (say, from http://example.com/v/1234567 to http://example.com/v/1234567 ) and youtube-dl reports an URL of a service in that list as unsupported. In that case, simply report a bug.
+For one, have a look at the [list of supported sites](docs/supportedsites.md). Note that it can sometimes happen that the site changes its URL scheme (say, from http://example.com/v/1234567 to http://example.com/v/1234567 ) and youtube-dl reports an URL of a service in that list as unsupported. In that case, simply report a bug.
 
 It is *not* possible to detect whether a URL is supported or not. That's because youtube-dl contains a generic extractor which matches **all** URLs. You may be tempted to disable, exclude, or remove the generic extractor, but the generic extractor not only allows users to extract videos from lots of websites that embed a video from another service, but may also be used to extract video from a service that it's hosting itself. Therefore, we neither recommend nor support disabling, excluding, or removing the generic extractor.
 
@@ -584,7 +612,7 @@ If you want to add support for a new site, you can follow this quick list (assum
 5. Add an import in [`youtube_dl/extractor/__init__.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/__init__.py).
 6. Run `python test/test_download.py TestDownload.test_YourExtractor`. This *should fail* at first, but you can continually re-run it until you're done. If you decide to add more than one test, then rename ``_TEST`` to ``_TESTS`` and make it into a list of dictionaries. The tests will be then be named `TestDownload.test_YourExtractor`, `TestDownload.test_YourExtractor_1`, `TestDownload.test_YourExtractor_2`, etc.
 7. Have a look at [`youtube_dl/common/extractor/common.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py) for possible helper methods and a [detailed description of what your extractor should return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L38). Add tests and code for as many as you want.
-8. If you can, check the code with [pyflakes](https://pypi.python.org/pypi/pyflakes) (a good idea) and [pep8](https://pypi.python.org/pypi/pep8) (optional, ignore E501).
+8. If you can, check the code with [flake8](https://pypi.python.org/pypi/flake8).
 9. When the tests pass, [add](http://git-scm.com/docs/git-add) the new files and [commit](http://git-scm.com/docs/git-commit) them and [push](http://git-scm.com/docs/git-push) the result, like this:
 
         $ git add youtube_dl/extractor/__init__.py

setup.cfg

@@ -2,5 +2,5 @@
 universal = True
 
 [flake8]
-exclude = youtube_dl/extractor/__init__.py,devscripts/buildserver.py,setup.py,build
+exclude = youtube_dl/extractor/__init__.py,devscripts/buildserver.py,setup.py,build,.git
 ignore = E501

test/helper.py

@@ -140,7 +140,7 @@ def expect_info_dict(self, got_dict, expected_dict):
     # Are checkable fields missing from the test case definition?
     test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))
                           for key, value in got_dict.items()
-                          if value and key in ('title', 'description', 'uploader', 'upload_date', 'timestamp', 'uploader_id', 'location'))
+                          if value and key in ('id', 'title', 'description', 'uploader', 'upload_date', 'timestamp', 'uploader_id', 'location'))
     missing_keys = set(test_info_dict.keys()) - set(expected_dict.keys())
     if missing_keys:
         def _repr(v):
@@ -148,9 +148,15 @@ def expect_info_dict(self, got_dict, expected_dict):
                 return "'%s'" % v.replace('\\', '\\\\').replace("'", "\\'").replace('\n', '\\n')
             else:
                 return repr(v)
-        info_dict_str = ''.join(
-            '    %s: %s,\n' % (_repr(k), _repr(v))
-            for k, v in test_info_dict.items())
+        info_dict_str = ''
+        if len(missing_keys) != len(expected_dict):
+            info_dict_str += ''.join(
+                '    %s: %s,\n' % (_repr(k), _repr(v))
+                for k, v in test_info_dict.items() if k not in missing_keys)
+            info_dict_str += '\n'
+        info_dict_str += ''.join(
+            '    %s: %s,\n' % (_repr(k), _repr(test_info_dict[k]))
+            for k in missing_keys)
         write_string(
             '\n\'info_dict\': {\n' + info_dict_str + '}\n', out=sys.stderr)
         self.assertFalse(

test/test_download.py

@@ -89,7 +89,7 @@ def generator(test_case):
 
         for tc in test_cases:
             info_dict = tc.get('info_dict', {})
-            if not tc.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
+            if not (info_dict.get('id') and info_dict.get('ext')):
                 raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?')
 
         if 'skip' in test_case:
@@ -116,7 +116,7 @@ def generator(test_case):
         expect_warnings(ydl, test_case.get('expected_warnings', []))
 
         def get_tc_filename(tc):
-            return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
+            return ydl.prepare_filename(tc.get('info_dict', {}))
 
         res_dict = None
 

test/test_http.py

@@ -0,0 +1,72 @@
+#!/usr/bin/env python
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl import YoutubeDL
+from youtube_dl.compat import compat_http_server
+import ssl
+import threading
+
+TEST_DIR = os.path.dirname(os.path.abspath(__file__))
+
+
+class HTTPTestRequestHandler(compat_http_server.BaseHTTPRequestHandler):
+    def log_message(self, format, *args):
+        pass
+
+    def do_GET(self):
+        if self.path == '/video.html':
+            self.send_response(200)
+            self.send_header('Content-Type', 'text/html; charset=utf-8')
+            self.end_headers()
+            self.wfile.write(b'<html><video src="/vid.mp4" /></html>')
+        elif self.path == '/vid.mp4':
+            self.send_response(200)
+            self.send_header('Content-Type', 'video/mp4')
+            self.end_headers()
+            self.wfile.write(b'\x00\x00\x00\x00\x20\x66\x74[video]')
+        else:
+            assert False
+
+
+class FakeLogger(object):
+    def debug(self, msg):
+        pass
+
+    def warning(self, msg):
+        pass
+
+    def error(self, msg):
+        pass
+
+
+class TestHTTP(unittest.TestCase):
+    def setUp(self):
+        certfn = os.path.join(TEST_DIR, 'testcert.pem')
+        self.httpd = compat_http_server.HTTPServer(
+            ('localhost', 0), HTTPTestRequestHandler)
+        self.httpd.socket = ssl.wrap_socket(
+            self.httpd.socket, certfile=certfn, server_side=True)
+        self.port = self.httpd.socket.getsockname()[1]
+        self.server_thread = threading.Thread(target=self.httpd.serve_forever)
+        self.server_thread.daemon = True
+        self.server_thread.start()
+
+    def test_nocheckcertificate(self):
+        if sys.version_info >= (2, 7, 9):  # No certificate checking anyways
+            ydl = YoutubeDL({'logger': FakeLogger()})
+            self.assertRaises(
+                Exception,
+                ydl.extract_info, 'https://localhost:%d/video.html' % self.port)
+
+        ydl = YoutubeDL({'logger': FakeLogger(), 'nocheckcertificate': True})
+        r = ydl.extract_info('https://localhost:%d/video.html' % self.port)
+        self.assertEqual(r['url'], 'https://localhost:%d/vid.mp4' % self.port)
+
+if __name__ == '__main__':
+    unittest.main()

test/test_jsinterp.py

@@ -0,0 +1,95 @@
+#!/usr/bin/env python
+
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl.jsinterp import JSInterpreter
+
+
+class TestJSInterpreter(unittest.TestCase):
+    def test_basic(self):
+        jsi = JSInterpreter('function x(){;}')
+        self.assertEqual(jsi.call_function('x'), None)
+
+        jsi = JSInterpreter('function x3(){return 42;}')
+        self.assertEqual(jsi.call_function('x3'), 42)
+
+    def test_calc(self):
+        jsi = JSInterpreter('function x4(a){return 2*a+1;}')
+        self.assertEqual(jsi.call_function('x4', 3), 7)
+
+    def test_empty_return(self):
+        jsi = JSInterpreter('function f(){return; y()}')
+        self.assertEqual(jsi.call_function('f'), None)
+
+    def test_morespace(self):
+        jsi = JSInterpreter('function x (a) { return 2 * a + 1 ; }')
+        self.assertEqual(jsi.call_function('x', 3), 7)
+
+        jsi = JSInterpreter('function f () { x =  2  ; return x; }')
+        self.assertEqual(jsi.call_function('f'), 2)
+
+    def test_strange_chars(self):
+        jsi = JSInterpreter('function $_xY1 ($_axY1) { var $_axY2 = $_axY1 + 1; return $_axY2; }')
+        self.assertEqual(jsi.call_function('$_xY1', 20), 21)
+
+    def test_operators(self):
+        jsi = JSInterpreter('function f(){return 1 << 5;}')
+        self.assertEqual(jsi.call_function('f'), 32)
+
+        jsi = JSInterpreter('function f(){return 19 & 21;}')
+        self.assertEqual(jsi.call_function('f'), 17)
+
+        jsi = JSInterpreter('function f(){return 11 >> 2;}')
+        self.assertEqual(jsi.call_function('f'), 2)
+
+    def test_array_access(self):
+        jsi = JSInterpreter('function f(){var x = [1,2,3]; x[0] = 4; x[0] = 5; x[2] = 7; return x;}')
+        self.assertEqual(jsi.call_function('f'), [5, 2, 7])
+
+    def test_parens(self):
+        jsi = JSInterpreter('function f(){return (1) + (2) * ((( (( (((((3)))))) )) ));}')
+        self.assertEqual(jsi.call_function('f'), 7)
+
+        jsi = JSInterpreter('function f(){return (1 + 2) * 3;}')
+        self.assertEqual(jsi.call_function('f'), 9)
+
+    def test_assignments(self):
+        jsi = JSInterpreter('function f(){var x = 20; x = 30 + 1; return x;}')
+        self.assertEqual(jsi.call_function('f'), 31)
+
+        jsi = JSInterpreter('function f(){var x = 20; x += 30 + 1; return x;}')
+        self.assertEqual(jsi.call_function('f'), 51)
+
+        jsi = JSInterpreter('function f(){var x = 20; x -= 30 + 1; return x;}')
+        self.assertEqual(jsi.call_function('f'), -11)
+
+    def test_comments(self):
+        jsi = JSInterpreter('''
+        function x() {
+            var x = /* 1 + */ 2;
+            var y = /* 30
+            * 40 */ 50;
+            return x + y;
+        }
+        ''')
+        self.assertEqual(jsi.call_function('x'), 52)
+
+    def test_precedence(self):
+        jsi = JSInterpreter('''
+        function x() {
+            var a = [10, 20, 30, 40, 50];
+            var b = 6;
+            a[0]=a[b%a.length];
+            return a;
+        }''')
+        self.assertEqual(jsi.call_function('x'), [20, 20, 30, 40, 50])
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_utils.py

@@ -28,6 +28,7 @@ from youtube_dl.utils import (
     fix_xml_ampersands,
     InAdvancePagedList,
     intlist_to_bytes,
+    is_html,
     js_to_json,
     limit_length,
     OnDemandPagedList,
@@ -51,6 +52,7 @@ from youtube_dl.utils import (
     urlencode_postdata,
     version_tuple,
     xpath_with_ns,
+    render_table,
 )
 
 
@@ -417,5 +419,31 @@ ffmpeg version 2.4.4 Copyright (c) 2000-2014 the FFmpeg ...'''), '2.4.4')
         self.assertTrue(age_restricted(18, 14))
         self.assertFalse(age_restricted(18, 18))
 
+    def test_is_html(self):
+        self.assertFalse(is_html(b'\x49\x44\x43<html'))
+        self.assertTrue(is_html(b'<!DOCTYPE foo>\xaaa'))
+        self.assertTrue(is_html(  # UTF-8 with BOM
+            b'\xef\xbb\xbf<!DOCTYPE foo>\xaaa'))
+        self.assertTrue(is_html(  # UTF-16-LE
+            b'\xff\xfe<\x00h\x00t\x00m\x00l\x00>\x00\xe4\x00'
+        ))
+        self.assertTrue(is_html(  # UTF-16-BE
+            b'\xfe\xff\x00<\x00h\x00t\x00m\x00l\x00>\x00\xe4'
+        ))
+        self.assertTrue(is_html(  # UTF-32-BE
+            b'\x00\x00\xFE\xFF\x00\x00\x00<\x00\x00\x00h\x00\x00\x00t\x00\x00\x00m\x00\x00\x00l\x00\x00\x00>\x00\x00\x00\xe4'))
+        self.assertTrue(is_html(  # UTF-32-LE
+            b'\xFF\xFE\x00\x00<\x00\x00\x00h\x00\x00\x00t\x00\x00\x00m\x00\x00\x00l\x00\x00\x00>\x00\x00\x00\xe4\x00\x00\x00'))
+
+    def test_render_table(self):
+        self.assertEqual(
+            render_table(
+                ['a', 'bcd'],
+                [[123, 4], [9999, 51]]),
+            'a    bcd\n'
+            '123  4\n'
+            '9999 51')
+
+
 if __name__ == '__main__':
     unittest.main()

test/testcert.pem

@@ -0,0 +1,52 @@
+-----BEGIN PRIVATE KEY-----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+-----END PRIVATE KEY-----
+-----BEGIN CERTIFICATE-----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+-----END CERTIFICATE-----

youtube_dl/YoutubeDL.py

@@ -25,6 +25,7 @@ if os.name == 'nt':
     import ctypes
 
 from .compat import (
+    compat_basestring,
     compat_cookiejar,
     compat_expanduser,
     compat_http_client,
@@ -54,8 +55,10 @@ from .utils import (
     PostProcessingError,
     platform_name,
     preferredencoding,
+    render_table,
     SameFileError,
     sanitize_filename,
+    std_headers,
     subtitles_filename,
     takewhile_inclusive,
     UnavailableVideoError,
@@ -73,6 +76,7 @@ from .extractor import get_info_extractor, gen_extractors
 from .downloader import get_suitable_downloader
 from .downloader.rtmp import rtmpdump_version
 from .postprocessor import (
+    FFmpegFixupM4aPP,
     FFmpegFixupStretchedPP,
     FFmpegMergerPP,
     FFmpegPostProcessor,
@@ -134,6 +138,7 @@ class YoutubeDL(object):
     nooverwrites:      Prevent overwriting files.
     playliststart:     Playlist item to start at.
     playlistend:       Playlist item to end at.
+    playlist_items:    Specific indices of playlist to download.
     playlistreverse:   Download playlist items in reverse order.
     matchtitle:        Download only matching titles.
     rejecttitle:       Reject downloads for matching titles.
@@ -143,6 +148,7 @@ class YoutubeDL(object):
     writeinfojson:     Write the video description to a .info.json file
     writeannotations:  Write the video annotations to a .annotations.xml file
     writethumbnail:    Write the thumbnail image to a file
+    write_all_thumbnails:  Write all thumbnail formats to files
     writesubtitles:    Write the video subtitles to a file
     writeautomaticsub: Write the automatic subtitles to a file
     allsubtitles:      Downloads all the subtitles of the video
@@ -193,11 +199,12 @@ class YoutubeDL(object):
                        postprocessor.
     progress_hooks:    A list of functions that get called on download
                        progress, with a dictionary with the entries
-                       * filename: The final filename
-                       * status: One of "downloading" and "finished"
-
-                       The dict may also have some of the following entries:
+                       * status: One of "downloading" and "finished".
+                                 Check this first and ignore unknown values.
 
+                       If status is one of "downloading" or "finished", the
+                       following properties may also be present:
+                       * filename: The final filename (always present)
                        * downloaded_bytes: Bytes on disk
                        * total_bytes: Size of the whole file, None if unknown
                        * tmpfilename: The filename we're currently writing to
@@ -213,16 +220,21 @@ class YoutubeDL(object):
                        - "never": do nothing
                        - "warn": only emit a warning
                        - "detect_or_warn": check whether we can do anything
-                                           about it, warn otherwise
+                                           about it, warn otherwise (default)
     source_address:    (Experimental) Client-side IP address to bind to.
     call_home:         Boolean, true iff we are allowed to contact the
                        youtube-dl servers for debugging.
+    sleep_interval:    Number of seconds to sleep before each download.
+    external_downloader:  Executable of the external downloader to call.
+    listformats:       Print an overview of available video formats and exit.
+    list_thumbnails:   Print a table of all thumbnails and exit.
 
 
     The following parameters are not used by YoutubeDL itself, they are used by
     the FileDownloader:
     nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
-    noresizebuffer, retries, continuedl, noprogress, consoletitle
+    noresizebuffer, retries, continuedl, noprogress, consoletitle,
+    xattr_set_filesize.
 
     The following options are used by the post processors:
     prefer_ffmpeg:     If True, use ffmpeg instead of avconv if both are available,
@@ -532,6 +544,11 @@ class YoutubeDL(object):
             outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
             tmpl = compat_expanduser(outtmpl)
             filename = tmpl % template_dict
+            # Temporary fix for #4787
+            # 'Treat' all problem characters by passing filename through preferredencoding
+            # to workaround encoding issues with subprocess on python2 @ Windows
+            if sys.version_info < (3, 0) and sys.platform == 'win32':
+                filename = encodeFilename(filename, True).decode(preferredencoding())
             return filename
         except ValueError as err:
             self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
@@ -695,24 +712,51 @@ class YoutubeDL(object):
             if playlistend == -1:
                 playlistend = None
 
+            playlistitems_str = self.params.get('playlist_items', None)
+            playlistitems = None
+            if playlistitems_str is not None:
+                def iter_playlistitems(format):
+                    for string_segment in format.split(','):
+                        if '-' in string_segment:
+                            start, end = string_segment.split('-')
+                            for item in range(int(start), int(end) + 1):
+                                yield int(item)
+                        else:
+                            yield int(string_segment)
+                playlistitems = iter_playlistitems(playlistitems_str)
+
             ie_entries = ie_result['entries']
             if isinstance(ie_entries, list):
                 n_all_entries = len(ie_entries)
-                entries = ie_entries[playliststart:playlistend]
+                if playlistitems:
+                    entries = [ie_entries[i - 1] for i in playlistitems]
+                else:
+                    entries = ie_entries[playliststart:playlistend]
                 n_entries = len(entries)
                 self.to_screen(
                     "[%s] playlist %s: Collected %d video ids (downloading %d of them)" %
                     (ie_result['extractor'], playlist, n_all_entries, n_entries))
             elif isinstance(ie_entries, PagedList):
-                entries = ie_entries.getslice(
-                    playliststart, playlistend)
+                if playlistitems:
+                    entries = []
+                    for item in playlistitems:
+                        entries.extend(ie_entries.getslice(
+                            item - 1, item
+                        ))
+                else:
+                    entries = ie_entries.getslice(
+                        playliststart, playlistend)
                 n_entries = len(entries)
                 self.to_screen(
                     "[%s] playlist %s: Downloading %d videos" %
                     (ie_result['extractor'], playlist, n_entries))
             else:  # iterable
-                entries = list(itertools.islice(
-                    ie_entries, playliststart, playlistend))
+                if playlistitems:
+                    entry_list = list(ie_entries)
+                    entries = [entry_list[i - 1] for i in playlistitems]
+                else:
+                    entries = list(itertools.islice(
+                        ie_entries, playliststart, playlistend))
                 n_entries = len(entries)
                 self.to_screen(
                     "[%s] playlist %s: Downloading %d videos" %
@@ -782,7 +826,7 @@ class YoutubeDL(object):
             '!=': operator.ne,
         }
         operator_rex = re.compile(r'''(?x)\s*\[
-            (?P<key>width|height|tbr|abr|vbr|filesize)
+            (?P<key>width|height|tbr|abr|vbr|filesize|fps)
             \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
             (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
             \]$
@@ -862,6 +906,42 @@ class YoutubeDL(object):
                 return matches[-1]
         return None
 
+    def _calc_headers(self, info_dict):
+        res = std_headers.copy()
+
+        add_headers = info_dict.get('http_headers')
+        if add_headers:
+            res.update(add_headers)
+
+        cookies = self._calc_cookies(info_dict)
+        if cookies:
+            res['Cookie'] = cookies
+
+        return res
+
+    def _calc_cookies(self, info_dict):
+        class _PseudoRequest(object):
+            def __init__(self, url):
+                self.url = url
+                self.headers = {}
+                self.unverifiable = False
+
+            def add_unredirected_header(self, k, v):
+                self.headers[k] = v
+
+            def get_full_url(self):
+                return self.url
+
+            def is_unverifiable(self):
+                return self.unverifiable
+
+            def has_header(self, h):
+                return h in self.headers
+
+        pr = _PseudoRequest(info_dict['url'])
+        self.cookiejar.add_cookie_header(pr)
+        return pr.headers.get('Cookie')
+
     def process_video_result(self, info_dict, download=True):
         assert info_dict.get('_type', 'video') == 'video'
 
@@ -876,9 +956,14 @@ class YoutubeDL(object):
             info_dict['playlist_index'] = None
 
         thumbnails = info_dict.get('thumbnails')
+        if thumbnails is None:
+            thumbnail = info_dict.get('thumbnail')
+            if thumbnail:
+                info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
         if thumbnails:
             thumbnails.sort(key=lambda t: (
-                t.get('width'), t.get('height'), t.get('url')))
+                t.get('preference'), t.get('width'), t.get('height'),
+                t.get('id'), t.get('url')))
             for t in thumbnails:
                 if 'width' in t and 'height' in t:
                     t['resolution'] = '%dx%d' % (t['width'], t['height'])
@@ -930,6 +1015,11 @@ class YoutubeDL(object):
             # Automatically determine file extension if missing
             if 'ext' not in format:
                 format['ext'] = determine_ext(format['url']).lower()
+            # Add HTTP headers, so that external programs can use them from the
+            # json output
+            full_format_info = info_dict.copy()
+            full_format_info.update(format)
+            format['http_headers'] = self._calc_headers(full_format_info)
 
         format_limit = self.params.get('format_limit', None)
         if format_limit:
@@ -945,9 +1035,12 @@ class YoutubeDL(object):
             # element in the 'formats' field in info_dict is info_dict itself,
             # wich can't be exported to json
             info_dict['formats'] = formats
-        if self.params.get('listformats', None):
+        if self.params.get('listformats'):
             self.list_formats(info_dict)
             return
+        if self.params.get('list_thumbnails'):
+            self.list_thumbnails(info_dict)
+            return
 
         req_format = self.params.get('format')
         if req_format is None:
@@ -982,7 +1075,8 @@ class YoutubeDL(object):
                             selected_format = {
                                 'requested_formats': formats_info,
                                 'format': rf,
-                                'ext': formats_info[0]['ext'],
+                                'format_id': '%s+%s' % (formats_info[0].get('format_id'),
+                                                        formats_info[1].get('format_id')),
                                 'width': formats_info[0].get('width'),
                                 'height': formats_info[0].get('height'),
                                 'resolution': formats_info[0].get('resolution'),
@@ -1043,7 +1137,7 @@ class YoutubeDL(object):
 
         self._num_downloads += 1
 
-        filename = self.prepare_filename(info_dict)
+        info_dict['_filename'] = filename = self.prepare_filename(info_dict)
 
         # Forced printings
         if self.params.get('forcetitle', False):
@@ -1068,10 +1162,7 @@ class YoutubeDL(object):
         if self.params.get('forceformat', False):
             self.to_stdout(info_dict['format'])
         if self.params.get('forcejson', False):
-            info_dict['_filename'] = filename
             self.to_stdout(json.dumps(info_dict))
-        if self.params.get('dump_single_json', False):
-            info_dict['_filename'] = filename
 
         # Do nothing else if in simulate mode
         if self.params.get('simulate', False):
@@ -1154,35 +1245,18 @@ class YoutubeDL(object):
                     self.report_error('Cannot write metadata to JSON file ' + infofn)
                     return
 
-        if self.params.get('writethumbnail', False):
-            if info_dict.get('thumbnail') is not None:
-                thumb_format = determine_ext(info_dict['thumbnail'], 'jpg')
-                thumb_filename = os.path.splitext(filename)[0] + '.' + thumb_format
-                if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
-                    self.to_screen('[%s] %s: Thumbnail is already present' %
-                                   (info_dict['extractor'], info_dict['id']))
-                else:
-                    self.to_screen('[%s] %s: Downloading thumbnail ...' %
-                                   (info_dict['extractor'], info_dict['id']))
-                    try:
-                        uf = self.urlopen(info_dict['thumbnail'])
-                        with open(thumb_filename, 'wb') as thumbf:
-                            shutil.copyfileobj(uf, thumbf)
-                        self.to_screen('[%s] %s: Writing thumbnail to: %s' %
-                                       (info_dict['extractor'], info_dict['id'], thumb_filename))
-                    except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-                        self.report_warning('Unable to download thumbnail "%s": %s' %
-                                            (info_dict['thumbnail'], compat_str(err)))
+        self._write_thumbnails(info_dict, filename)
 
         if not self.params.get('skip_download', False):
             try:
                 def dl(name, info):
-                    fd = get_suitable_downloader(info)(self, self.params)
+                    fd = get_suitable_downloader(info, self.params)(self, self.params)
                     for ph in self._progress_hooks:
                         fd.add_progress_hook(ph)
                     if self.params.get('verbose'):
                         self.to_stdout('[debug] Invoking downloader on %r' % info.get('url'))
                     return fd.download(name, info)
+
                 if info_dict.get('requested_formats') is not None:
                     downloaded = []
                     success = True
@@ -1218,11 +1292,12 @@ class YoutubeDL(object):
 
             if success:
                 # Fixup content
+                fixup_policy = self.params.get('fixup')
+                if fixup_policy is None:
+                    fixup_policy = 'detect_or_warn'
+
                 stretched_ratio = info_dict.get('stretched_ratio')
                 if stretched_ratio is not None and stretched_ratio != 1:
-                    fixup_policy = self.params.get('fixup')
-                    if fixup_policy is None:
-                        fixup_policy = 'detect_or_warn'
                     if fixup_policy == 'warn':
                         self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
                             info_dict['id'], stretched_ratio))
@@ -1236,7 +1311,23 @@ class YoutubeDL(object):
                                 '%s: Non-uniform pixel ratio (%s). Install ffmpeg or avconv to fix this automatically.' % (
                                     info_dict['id'], stretched_ratio))
                     else:
-                        assert fixup_policy == 'ignore'
+                        assert fixup_policy in ('ignore', 'never')
+
+                if info_dict.get('requested_formats') is None and info_dict.get('container') == 'm4a_dash':
+                    if fixup_policy == 'warn':
+                        self.report_warning('%s: writing DASH m4a. Only some players support this container.' % (
+                            info_dict['id']))
+                    elif fixup_policy == 'detect_or_warn':
+                        fixup_pp = FFmpegFixupM4aPP(self)
+                        if fixup_pp.available:
+                            info_dict.setdefault('__postprocessors', [])
+                            info_dict['__postprocessors'].append(fixup_pp)
+                        else:
+                            self.report_warning(
+                                '%s: writing DASH m4a. Only some players support this container. Install ffmpeg or avconv to fix this automatically.' % (
+                                    info_dict['id']))
+                    else:
+                        assert fixup_policy in ('ignore', 'never')
 
                 try:
                     self.post_process(filename, info_dict)
@@ -1438,8 +1529,26 @@ class YoutubeDL(object):
         header_line = line({
             'format_id': 'format code', 'ext': 'extension',
             'resolution': 'resolution', 'format_note': 'note'}, idlen=idlen)
-        self.to_screen('[info] Available formats for %s:\n%s\n%s' %
-                       (info_dict['id'], header_line, '\n'.join(formats_s)))
+        self.to_screen(
+            '[info] Available formats for %s:\n%s\n%s' %
+            (info_dict['id'], header_line, '\n'.join(formats_s)))
+
+    def list_thumbnails(self, info_dict):
+        thumbnails = info_dict.get('thumbnails')
+        if not thumbnails:
+            tn_url = info_dict.get('thumbnail')
+            if tn_url:
+                thumbnails = [{'id': '0', 'url': tn_url}]
+            else:
+                self.to_screen(
+                    '[info] No thumbnails present for %s' % info_dict['id'])
+                return
+
+        self.to_screen(
+            '[info] Thumbnails for %s:' % info_dict['id'])
+        self.to_screen(render_table(
+            ['ID', 'width', 'height', 'URL'],
+            [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
 
     def urlopen(self, req):
         """ Start an HTTP download """
@@ -1450,7 +1559,7 @@ class YoutubeDL(object):
         # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
         # To work around aforementioned issue we will replace request's original URL with
         # percent-encoded one
-        req_is_string = isinstance(req, basestring if sys.version_info < (3, 0) else compat_str)
+        req_is_string = isinstance(req, compat_basestring)
         url = req if req_is_string else req.get_full_url()
         url_escaped = escape_url(url)
 
@@ -1585,3 +1694,39 @@ class YoutubeDL(object):
         if encoding is None:
             encoding = preferredencoding()
         return encoding
+
+    def _write_thumbnails(self, info_dict, filename):
+        if self.params.get('writethumbnail', False):
+            thumbnails = info_dict.get('thumbnails')
+            if thumbnails:
+                thumbnails = [thumbnails[-1]]
+        elif self.params.get('write_all_thumbnails', False):
+            thumbnails = info_dict.get('thumbnails')
+        else:
+            return
+
+        if not thumbnails:
+            # No thumbnails present, so return immediately
+            return
+
+        for t in thumbnails:
+            thumb_ext = determine_ext(t['url'], 'jpg')
+            suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
+            thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
+            thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
+
+            if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
+                self.to_screen('[%s] %s: Thumbnail %sis already present' %
+                               (info_dict['extractor'], info_dict['id'], thumb_display_id))
+            else:
+                self.to_screen('[%s] %s: Downloading thumbnail %s...' %
+                               (info_dict['extractor'], info_dict['id'], thumb_display_id))
+                try:
+                    uf = self.urlopen(t['url'])
+                    with open(thumb_filename, 'wb') as thumbf:
+                        shutil.copyfileobj(uf, thumbf)
+                    self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
+                                   (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
+                except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
+                    self.report_warning('Unable to download thumbnail "%s": %s' %
+                                        (t['url'], compat_str(err)))

youtube_dl/__init__.py

@@ -143,10 +143,13 @@ def _real_main(argv=None):
             parser.error('invalid max_filesize specified')
         opts.max_filesize = numeric_limit
     if opts.retries is not None:
-        try:
-            opts.retries = int(opts.retries)
-        except (TypeError, ValueError):
-            parser.error('invalid retry count specified')
+        if opts.retries in ('inf', 'infinite'):
+            opts_retries = float('inf')
+        else:
+            try:
+                opts_retries = int(opts.retries)
+            except (TypeError, ValueError):
+                parser.error('invalid retry count specified')
     if opts.buffersize is not None:
         numeric_buffersize = FileDownloader.parse_bytes(opts.buffersize)
         if numeric_buffersize is None:
@@ -238,6 +241,12 @@ def _real_main(argv=None):
             'verboseOutput': opts.verbose,
             'exec_cmd': opts.exec_cmd,
         })
+    if opts.xattr_set_filesize:
+        try:
+            import xattr
+            xattr  # Confuse flake8
+        except ImportError:
+            parser.error('setting filesize xattr requested but python-xattr is not available')
 
     ydl_opts = {
         'usenetrc': opts.usenetrc,
@@ -268,7 +277,7 @@ def _real_main(argv=None):
         'ignoreerrors': opts.ignoreerrors,
         'ratelimit': opts.ratelimit,
         'nooverwrites': opts.nooverwrites,
-        'retries': opts.retries,
+        'retries': opts_retries,
         'buffersize': opts.buffersize,
         'noresizebuffer': opts.noresizebuffer,
         'continuedl': opts.continue_dl,
@@ -286,6 +295,7 @@ def _real_main(argv=None):
         'writeannotations': opts.writeannotations,
         'writeinfojson': opts.writeinfojson,
         'writethumbnail': opts.writethumbnail,
+        'write_all_thumbnails': opts.write_all_thumbnails,
         'writesubtitles': opts.writesubtitles,
         'writeautomaticsub': opts.writeautomaticsub,
         'allsubtitles': opts.allsubtitles,
@@ -329,6 +339,11 @@ def _real_main(argv=None):
         'fixup': opts.fixup,
         'source_address': opts.source_address,
         'call_home': opts.call_home,
+        'sleep_interval': opts.sleep_interval,
+        'external_downloader': opts.external_downloader,
+        'list_thumbnails': opts.list_thumbnails,
+        'playlist_items': opts.playlist_items,
+        'xattr_set_filesize': opts.xattr_set_filesize,
     }
 
     with YoutubeDL(ydl_opts) as ydl:
@@ -346,7 +361,9 @@ def _real_main(argv=None):
                 sys.exit()
 
             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
-            parser.error('you must provide at least one URL')
+            parser.error(
+                'You must provide at least one URL.\n'
+                'Type youtube-dl --help to see a list of all options.')
 
         try:
             if opts.load_info_filename is not None:

youtube_dl/compat.py

@@ -71,6 +71,11 @@ try:
 except ImportError:
     compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w')
 
+try:
+    import http.server as compat_http_server
+except ImportError:
+    import BaseHTTPServer as compat_http_server
+
 try:
     from urllib.parse import unquote as compat_urllib_parse_unquote
 except ImportError:
@@ -109,6 +114,26 @@ except ImportError:
             string += pct_sequence.decode(encoding, errors)
         return string
 
+try:
+    compat_str = unicode  # Python 2
+except NameError:
+    compat_str = str
+
+try:
+    compat_basestring = basestring  # Python 2
+except NameError:
+    compat_basestring = str
+
+try:
+    compat_chr = unichr  # Python 2
+except NameError:
+    compat_chr = chr
+
+try:
+    from xml.etree.ElementTree import ParseError as compat_xml_parse_error
+except ImportError:  # Python 2.6
+    from xml.parsers.expat import ExpatError as compat_xml_parse_error
+
 
 try:
     from urllib.parse import parse_qs as compat_parse_qs
@@ -118,7 +143,7 @@ except ImportError:  # Python 2
 
     def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False,
                    encoding='utf-8', errors='replace'):
-        qs, _coerce_result = qs, unicode
+        qs, _coerce_result = qs, compat_str
         pairs = [s2 for s1 in qs.split('&') for s2 in s1.split(';')]
         r = []
         for name_value in pairs:
@@ -157,21 +182,6 @@ except ImportError:  # Python 2
                 parsed_result[name] = [value]
         return parsed_result
 
-try:
-    compat_str = unicode  # Python 2
-except NameError:
-    compat_str = str
-
-try:
-    compat_chr = unichr  # Python 2
-except NameError:
-    compat_chr = chr
-
-try:
-    from xml.etree.ElementTree import ParseError as compat_xml_parse_error
-except ImportError:  # Python 2.6
-    from xml.parsers.expat import ExpatError as compat_xml_parse_error
-
 try:
     from shlex import quote as shlex_quote
 except ImportError:  # Python < 3.3
@@ -357,6 +367,7 @@ def workaround_optparse_bug9161():
 
 __all__ = [
     'compat_HTTPError',
+    'compat_basestring',
     'compat_chr',
     'compat_cookiejar',
     'compat_expanduser',
@@ -365,6 +376,7 @@ __all__ = [
     'compat_html_entities',
     'compat_html_parser',
     'compat_http_client',
+    'compat_http_server',
     'compat_kwargs',
     'compat_ord',
     'compat_parse_qs',

youtube_dl/downloader/__init__.py

@@ -1,35 +1,41 @@
 from __future__ import unicode_literals
 
 from .common import FileDownloader
+from .external import get_external_downloader
+from .f4m import F4mFD
 from .hls import HlsFD
 from .hls import NativeHlsFD
 from .http import HttpFD
 from .mplayer import MplayerFD
 from .rtmp import RtmpFD
-from .f4m import F4mFD
 
 from ..utils import (
-    determine_ext,
+    determine_protocol,
 )
 
+PROTOCOL_MAP = {
+    'rtmp': RtmpFD,
+    'm3u8_native': NativeHlsFD,
+    'm3u8': HlsFD,
+    'mms': MplayerFD,
+    'rtsp': MplayerFD,
+    'f4m': F4mFD,
+}
 
-def get_suitable_downloader(info_dict):
+
+def get_suitable_downloader(info_dict, params={}):
     """Get the downloader class that can handle the info dict."""
-    url = info_dict['url']
-    protocol = info_dict.get('protocol')
+    protocol = determine_protocol(info_dict)
+    info_dict['protocol'] = protocol
+
+    external_downloader = params.get('external_downloader')
+    if external_downloader is not None:
+        ed = get_external_downloader(external_downloader)
+        if ed.supports(info_dict):
+            return ed
+
+    return PROTOCOL_MAP.get(protocol, HttpFD)
 
-    if url.startswith('rtmp'):
-        return RtmpFD
-    if protocol == 'm3u8_native':
-        return NativeHlsFD
-    if (protocol == 'm3u8') or (protocol is None and determine_ext(url) == 'm3u8'):
-        return HlsFD
-    if url.startswith('mms') or url.startswith('rtsp'):
-        return MplayerFD
-    if determine_ext(url) == 'f4m':
-        return F4mFD
-    else:
-        return HttpFD
 
 __all__ = [
     'get_suitable_downloader',

youtube_dl/downloader/common.py

@@ -25,21 +25,23 @@ class FileDownloader(object):
 
     Available options:
 
-    verbose:           Print additional info to stdout.
-    quiet:             Do not print messages to stdout.
-    ratelimit:         Download speed limit, in bytes/sec.
-    retries:           Number of times to retry for HTTP error 5xx
-    buffersize:        Size of download buffer in bytes.
-    noresizebuffer:    Do not automatically resize the download buffer.
-    continuedl:        Try to continue downloads if possible.
-    noprogress:        Do not print the progress bar.
-    logtostderr:       Log messages to stderr instead of stdout.
-    consoletitle:      Display progress in console window's titlebar.
-    nopart:            Do not use temporary .part files.
-    updatetime:        Use the Last-modified header to set output file timestamps.
-    test:              Download only first bytes to test the downloader.
-    min_filesize:      Skip files smaller than this size
-    max_filesize:      Skip files larger than this size
+    verbose:            Print additional info to stdout.
+    quiet:              Do not print messages to stdout.
+    ratelimit:          Download speed limit, in bytes/sec.
+    retries:            Number of times to retry for HTTP error 5xx
+    buffersize:         Size of download buffer in bytes.
+    noresizebuffer:     Do not automatically resize the download buffer.
+    continuedl:         Try to continue downloads if possible.
+    noprogress:         Do not print the progress bar.
+    logtostderr:        Log messages to stderr instead of stdout.
+    consoletitle:       Display progress in console window's titlebar.
+    nopart:             Do not use temporary .part files.
+    updatetime:         Use the Last-modified header to set output file timestamps.
+    test:               Download only first bytes to test the downloader.
+    min_filesize:       Skip files smaller than this size
+    max_filesize:       Skip files larger than this size
+    xattr_set_filesize: Set ytdl.filesize user xattribute with expected size.
+                        (experimenatal)
 
     Subclasses of this one must re-define the real_download method.
     """
@@ -284,6 +286,7 @@ class FileDownloader(object):
         """Download to a filename using the info from info_dict
         Return True on success and False otherwise
         """
+
         nooverwrites_and_exists = (
             self.params.get('nooverwrites', False)
             and os.path.exists(encodeFilename(filename))
@@ -305,6 +308,11 @@ class FileDownloader(object):
             })
             return True
 
+        sleep_interval = self.params.get('sleep_interval')
+        if sleep_interval:
+            self.to_screen('[download] Sleeping %s seconds...' % sleep_interval)
+            time.sleep(sleep_interval)
+
         return self.real_download(filename, info_dict)
 
     def real_download(self, filename, info_dict):
@@ -319,3 +327,24 @@ class FileDownloader(object):
         # See YoutubeDl.py (search for progress_hooks) for a description of
         # this interface
         self._progress_hooks.append(ph)
+
+    def _debug_cmd(self, args, subprocess_encoding, exe=None):
+        if not self.params.get('verbose', False):
+            return
+
+        if exe is None:
+            exe = os.path.basename(args[0])
+
+        if subprocess_encoding:
+            str_args = [
+                a.decode(subprocess_encoding) if isinstance(a, bytes) else a
+                for a in args]
+        else:
+            str_args = args
+        try:
+            import pipes
+            shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
+        except ImportError:
+            shell_quote = repr
+        self.to_screen('[debug] %s command line: %s' % (
+            exe, shell_quote(str_args)))

youtube_dl/downloader/external.py

@@ -0,0 +1,126 @@
+from __future__ import unicode_literals
+
+import os.path
+import subprocess
+import sys
+
+from .common import FileDownloader
+from ..utils import (
+    encodeFilename,
+)
+
+
+class ExternalFD(FileDownloader):
+    def real_download(self, filename, info_dict):
+        self.report_destination(filename)
+        tmpfilename = self.temp_name(filename)
+
+        retval = self._call_downloader(tmpfilename, info_dict)
+        if retval == 0:
+            fsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen('\r[%s] Downloaded %s bytes' % (self.get_basename(), fsize))
+            self.try_rename(tmpfilename, filename)
+            self._hook_progress({
+                'downloaded_bytes': fsize,
+                'total_bytes': fsize,
+                'filename': filename,
+                'status': 'finished',
+            })
+            return True
+        else:
+            self.to_stderr('\n')
+            self.report_error('%s exited with code %d' % (
+                self.get_basename(), retval))
+            return False
+
+    @classmethod
+    def get_basename(cls):
+        return cls.__name__[:-2].lower()
+
+    @property
+    def exe(self):
+        return self.params.get('external_downloader')
+
+    @classmethod
+    def supports(cls, info_dict):
+        return info_dict['protocol'] in ('http', 'https', 'ftp', 'ftps')
+
+    def _source_address(self, command_option):
+        source_address = self.params.get('source_address')
+        if source_address is None:
+            return []
+        return [command_option, source_address]
+
+    def _call_downloader(self, tmpfilename, info_dict):
+        """ Either overwrite this or implement _make_cmd """
+        cmd = self._make_cmd(tmpfilename, info_dict)
+
+        if sys.platform == 'win32' and sys.version_info < (3, 0):
+            # Windows subprocess module does not actually support Unicode
+            # on Python 2.x
+            # See http://stackoverflow.com/a/9951851/35070
+            subprocess_encoding = sys.getfilesystemencoding()
+            cmd = [a.encode(subprocess_encoding, 'ignore') for a in cmd]
+        else:
+            subprocess_encoding = None
+        self._debug_cmd(cmd, subprocess_encoding)
+
+        p = subprocess.Popen(
+            cmd, stderr=subprocess.PIPE)
+        _, stderr = p.communicate()
+        if p.returncode != 0:
+            self.to_stderr(stderr)
+        return p.returncode
+
+
+class CurlFD(ExternalFD):
+    def _make_cmd(self, tmpfilename, info_dict):
+        cmd = [self.exe, '-o', tmpfilename]
+        for key, val in info_dict['http_headers'].items():
+            cmd += ['--header', '%s: %s' % (key, val)]
+        cmd += self._source_address('--interface')
+        cmd += ['--', info_dict['url']]
+        return cmd
+
+
+class WgetFD(ExternalFD):
+    def _make_cmd(self, tmpfilename, info_dict):
+        cmd = [self.exe, '-O', tmpfilename, '-nv', '--no-cookies']
+        for key, val in info_dict['http_headers'].items():
+            cmd += ['--header', '%s: %s' % (key, val)]
+        cmd += self._source_address('--bind-address')
+        cmd += ['--', info_dict['url']]
+        return cmd
+
+
+class Aria2cFD(ExternalFD):
+    def _make_cmd(self, tmpfilename, info_dict):
+        cmd = [
+            self.exe, '-c',
+            '--min-split-size', '1M', '--max-connection-per-server', '4']
+        dn = os.path.dirname(tmpfilename)
+        if dn:
+            cmd += ['--dir', dn]
+        cmd += ['--out', os.path.basename(tmpfilename)]
+        for key, val in info_dict['http_headers'].items():
+            cmd += ['--header', '%s: %s' % (key, val)]
+        cmd += self._source_address('--interface')
+        cmd += ['--', info_dict['url']]
+        return cmd
+
+_BY_NAME = dict(
+    (klass.get_basename(), klass)
+    for name, klass in globals().items()
+    if name.endswith('FD') and name != 'ExternalFD'
+)
+
+
+def list_external_downloaders():
+    return sorted(_BY_NAME.keys())
+
+
+def get_external_downloader(external_downloader):
+    """ Given the name of the executable, see whether we support the given
+        downloader . """
+    bn = os.path.basename(external_downloader)
+    return _BY_NAME[bn]

youtube_dl/downloader/f4m.py

@@ -177,13 +177,12 @@ def build_fragments_list(boot_info):
     """ Return a list of (segment, fragment) for each fragment in the video """
     res = []
     segment_run_table = boot_info['segments'][0]
-    # I've only found videos with one segment
-    segment_run_entry = segment_run_table['segment_run'][0]
-    n_frags = segment_run_entry[1]
     fragment_run_entry_table = boot_info['fragments'][0]['fragments']
     first_frag_number = fragment_run_entry_table[0]['first']
-    for (i, frag_number) in zip(range(1, n_frags + 1), itertools.count(first_frag_number)):
-        res.append((1, frag_number))
+    fragments_counter = itertools.count(first_frag_number)
+    for segment, fragments_count in segment_run_table['segment_run']:
+        for _ in range(fragments_count):
+            res.append((segment, next(fragments_counter)))
     return res
 
 
@@ -231,6 +230,23 @@ class F4mFD(FileDownloader):
     A downloader for f4m manifests or AdobeHDS.
     """
 
+    def _get_unencrypted_media(self, doc):
+        media = doc.findall(_add_ns('media'))
+        if not media:
+            self.report_error('No media found')
+        for e in (doc.findall(_add_ns('drmAdditionalHeader')) +
+                  doc.findall(_add_ns('drmAdditionalHeaderSet'))):
+            # If id attribute is missing it's valid for all media nodes
+            # without drmAdditionalHeaderId or drmAdditionalHeaderSetId attribute
+            if 'id' not in e.attrib:
+                self.report_error('Missing ID in f4m DRM')
+        media = list(filter(lambda e: 'drmAdditionalHeaderId' not in e.attrib and
+                                      'drmAdditionalHeaderSetId' not in e.attrib,
+                            media))
+        if not media:
+            self.report_error('Unsupported DRM')
+        return media
+
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
         requested_bitrate = info_dict.get('tbr')
@@ -249,7 +265,8 @@ class F4mFD(FileDownloader):
         )
 
         doc = etree.fromstring(manifest)
-        formats = [(int(f.attrib.get('bitrate', -1)), f) for f in doc.findall(_add_ns('media'))]
+        formats = [(int(f.attrib.get('bitrate', -1)), f)
+                   for f in self._get_unencrypted_media(doc)]
         if requested_bitrate is None:
             # get the best format
             formats = sorted(formats, key=lambda f: f[0])

youtube_dl/downloader/hls.py

@@ -11,6 +11,7 @@ from ..compat import (
     compat_urllib_request,
 )
 from ..utils import (
+    encodeArgument,
     encodeFilename,
 )
 
@@ -21,23 +22,22 @@ class HlsFD(FileDownloader):
         self.report_destination(filename)
         tmpfilename = self.temp_name(filename)
 
-        args = [
-            '-y', '-i', url, '-f', 'mp4', '-c', 'copy',
-            '-bsf:a', 'aac_adtstoasc',
-            encodeFilename(tmpfilename, for_subprocess=True)]
-
         ffpp = FFmpegPostProcessor(downloader=self)
         program = ffpp._executable
         if program is None:
             self.report_error('m3u8 download detected but ffmpeg or avconv could not be found. Please install one.')
             return False
         ffpp.check_version()
-        cmd = [program] + args
 
-        retval = subprocess.call(cmd)
+        args = [
+            encodeArgument(opt)
+            for opt in (program, '-y', '-i', url, '-f', 'mp4', '-c', 'copy', '-bsf:a', 'aac_adtstoasc')]
+        args.append(encodeFilename(tmpfilename, True))
+
+        retval = subprocess.call(args)
         if retval == 0:
             fsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen('\r[%s] %s bytes' % (cmd[0], fsize))
+            self.to_screen('\r[%s] %s bytes' % (args[0], fsize))
             self.try_rename(tmpfilename, filename)
             self._hook_progress({
                 'downloaded_bytes': fsize,

youtube_dl/downloader/http.py

@@ -3,6 +3,9 @@ from __future__ import unicode_literals
 import os
 import time
 
+from socket import error as SocketError
+import errno
+
 from .common import FileDownloader
 from ..compat import (
     compat_urllib_request,
@@ -24,10 +27,6 @@ class HttpFD(FileDownloader):
 
         # Do not include the Accept-Encoding header
         headers = {'Youtubedl-no-compression': 'True'}
-        if 'user_agent' in info_dict:
-            headers['Youtubedl-user-agent'] = info_dict['user_agent']
-        if 'http_referer' in info_dict:
-            headers['Referer'] = info_dict['http_referer']
         add_headers = info_dict.get('http_headers')
         if add_headers:
             headers.update(add_headers)
@@ -103,6 +102,11 @@ class HttpFD(FileDownloader):
                             resume_len = 0
                             open_mode = 'wb'
                             break
+            except SocketError as e:
+                if e.errno != errno.ECONNRESET:
+                    # Connection reset is no problem, just retry
+                    raise
+
             # Retry
             count += 1
             if count <= retries:
@@ -161,6 +165,14 @@ class HttpFD(FileDownloader):
                 except (OSError, IOError) as err:
                     self.report_error('unable to open for writing: %s' % str(err))
                     return False
+
+                if self.params.get('xattr_set_filesize', False) and data_len is not None:
+                    try:
+                        import xattr
+                        xattr.setxattr(tmpfilename, 'user.ytdl.filesize', str(data_len))
+                    except(OSError, IOError, ImportError) as err:
+                        self.report_error('unable to set filesize xattr: %s' % str(err))
+
             try:
                 stream.write(data_block)
             except (IOError, OSError) as err:

youtube_dl/downloader/rtmp.py

@@ -104,6 +104,9 @@ class RtmpFD(FileDownloader):
         live = info_dict.get('rtmp_live', False)
         conn = info_dict.get('rtmp_conn', None)
         protocol = info_dict.get('rtmp_protocol', None)
+        real_time = info_dict.get('rtmp_real_time', False)
+        no_resume = info_dict.get('no_resume', False)
+        continue_dl = info_dict.get('continuedl', False)
 
         self.report_destination(filename)
         tmpfilename = self.temp_name(filename)
@@ -141,7 +144,14 @@ class RtmpFD(FileDownloader):
             basic_args += ['--conn', conn]
         if protocol is not None:
             basic_args += ['--protocol', protocol]
-        args = basic_args + [[], ['--resume', '--skip', '1']][not live and self.params.get('continuedl', False)]
+        if real_time:
+            basic_args += ['--realtime']
+
+        args = basic_args
+        if not no_resume and continue_dl and not live:
+            args += ['--resume']
+        if not live and continue_dl:
+            args += ['--skip', '1']
 
         if sys.platform == 'win32' and sys.version_info < (3, 0):
             # Windows subprocess module does not actually support Unicode
@@ -152,19 +162,7 @@ class RtmpFD(FileDownloader):
         else:
             subprocess_encoding = None
 
-        if self.params.get('verbose', False):
-            if subprocess_encoding:
-                str_args = [
-                    a.decode(subprocess_encoding) if isinstance(a, bytes) else a
-                    for a in args]
-            else:
-                str_args = args
-            try:
-                import pipes
-                shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
-            except ImportError:
-                shell_quote = repr
-            self.to_screen('[debug] rtmpdump command line: ' + shell_quote(str_args))
+        self._debug_cmd(args, subprocess_encoding, exe='rtmpdump')
 
         RD_SUCCESS = 0
         RD_FAILED = 1

youtube_dl/extractor/__init__.py

@@ -29,7 +29,6 @@ from .arte import (
 from .atresplayer import AtresPlayerIE
 from .atttechchannel import ATTTechChannelIE
 from .audiomack import AudiomackIE, AudiomackAlbumIE
-from .auengine import AUEngineIE
 from .azubu import AzubuIE
 from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE
@@ -83,6 +82,7 @@ from .crunchyroll import (
     CrunchyrollShowPlaylistIE
 )
 from .cspan import CSpanIE
+from .ctsnews import CtsNewsIE
 from .dailymotion import (
     DailymotionIE,
     DailymotionPlaylistIE,
@@ -90,6 +90,7 @@ from .dailymotion import (
 )
 from .daum import DaumIE
 from .dbtv import DBTVIE
+from .dctp import DctpTvIE
 from .deezer import DeezerPlaylistIE
 from .dfb import DFBIE
 from .dotsub import DotsubIE
@@ -181,6 +182,7 @@ from .heise import HeiseIE
 from .hellporno import HellPornoIE
 from .helsinki import HelsinkiIE
 from .hentaistigma import HentaiStigmaIE
+from .historicfilms import HistoricFilmsIE
 from .hitbox import HitboxIE, HitboxLiveIE
 from .hornbunny import HornBunnyIE
 from .hostingbulk import HostingBulkIE
@@ -285,9 +287,19 @@ from .netzkino import NetzkinoIE
 from .nerdcubed import NerdCubedFeedIE
 from .newgrounds import NewgroundsIE
 from .newstube import NewstubeIE
+from .nextmedia import (
+    NextMediaIE,
+    NextMediaActionNewsIE,
+    AppleDailyRealtimeNewsIE,
+    AppleDailyAnimationNewsIE
+)
 from .nfb import NFBIE
 from .nfl import NFLIE
-from .nhl import NHLIE, NHLVideocenterIE
+from .nhl import (
+    NHLIE,
+    NHLNewsIE,
+    NHLVideocenterIE,
+)
 from .niconico import NiconicoIE, NiconicoPlaylistIE
 from .ninegag import NineGagIE
 from .noco import NocoIE
@@ -350,6 +362,7 @@ from .rtbf import RTBFIE
 from .rte import RteIE
 from .rtlnl import RtlXlIE
 from .rtlnow import RTLnowIE
+from .rtl2 import RTL2IE
 from .rtp import RTPIE
 from .rts import RTSIE
 from .rtve import RTVEALaCartaIE, RTVELiveIE
@@ -467,6 +480,7 @@ from .twitch import (
     TwitchVodIE,
     TwitchProfileIE,
     TwitchPastBroadcastsIE,
+    TwitchBookmarksIE,
     TwitchStreamIE,
 )
 from .ubu import UbuIE
@@ -545,6 +559,7 @@ from .xminus import XMinusIE
 from .xnxx import XNXXIE
 from .xvideos import XVideosIE
 from .xtube import XTubeUserIE, XTubeIE
+from .xuite import XuiteIE
 from .xxxymovies import XXXYMoviesIE
 from .yahoo import (
     YahooIE,

youtube_dl/extractor/appletrailers.py

@@ -122,14 +122,15 @@ class AppleTrailersIE(InfoExtractor):
             playlist.append({
                 '_type': 'video',
                 'id': video_id,
-                'title': title,
                 'formats': formats,
                 'title': title,
                 'duration': duration,
                 'thumbnail': thumbnail,
                 'upload_date': upload_date,
                 'uploader_id': uploader_id,
-                'user_agent': 'QuickTime compatible (youtube-dl)',
+                'http_headers': {
+                    'User-Agent': 'QuickTime compatible (youtube-dl)',
+                },
             })
 
         return {

youtube_dl/extractor/ard.py

@@ -23,13 +23,7 @@ class ARDMediathekIE(InfoExtractor):
 
     _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': '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',
+        'only_matching': True,
     }, {
         'url': 'http://www.ardmediathek.de/tv/Tatort/Das-Wunder-von-Wolbeck-Video-tgl-ab-20/Das-Erste/Video?documentId=22490580&bcastId=602916',
         'info_dict': {

youtube_dl/extractor/atresplayer.py

@@ -3,7 +3,7 @@ from __future__ import unicode_literals
 import time
 import hmac
 
-from .common import InfoExtractor
+from .subtitles import SubtitlesInfoExtractor
 from ..compat import (
     compat_str,
     compat_urllib_parse,
@@ -17,7 +17,7 @@ from ..utils import (
 )
 
 
-class AtresPlayerIE(InfoExtractor):
+class AtresPlayerIE(SubtitlesInfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?atresplayer\.com/television/[^/]+/[^/]+/[^/]+/(?P<id>.+?)_\d+\.html'
     _TESTS = [
         {
@@ -95,7 +95,7 @@ class AtresPlayerIE(InfoExtractor):
         for fmt in ['windows', 'android_tablet']:
             request = compat_urllib_request.Request(
                 self._URL_VIDEO_TEMPLATE.format(fmt, episode_id, timestamp_shifted, token))
-            request.add_header('Youtubedl-user-agent', self._USER_AGENT)
+            request.add_header('User-Agent', self._USER_AGENT)
 
             fmt_json = self._download_json(
                 request, video_id, 'Downloading %s video JSON' % fmt)
@@ -105,13 +105,22 @@ class AtresPlayerIE(InfoExtractor):
                 raise ExtractorError(
                     '%s returned error: %s' % (self.IE_NAME, result), expected=True)
 
-            for _, video_url in fmt_json['resultObject'].items():
+            for format_id, video_url in fmt_json['resultObject'].items():
+                if format_id == 'token' or not video_url.startswith('http'):
+                    continue
                 if video_url.endswith('/Manifest'):
-                    formats.extend(self._extract_f4m_formats(video_url[:-9] + '/manifest.f4m', video_id))
+                    if 'geodeswowsmpra3player' in video_url:
+                        f4m_path = video_url.split('smil:', 1)[-1].split('free_', 1)[0]
+                        f4m_url = 'http://drg.antena3.com/{0}hds/es/sd.f4m'.format(f4m_path)
+                        # this videos are protected by DRM, the f4m downloader doesn't support them
+                        continue
+                    else:
+                        f4m_url = video_url[:-9] + '/manifest.f4m'
+                    formats.extend(self._extract_f4m_formats(f4m_url, video_id))
                 else:
                     formats.append({
                         'url': video_url,
-                        'format_id': 'android',
+                        'format_id': 'android-%s' % format_id,
                         'preference': 1,
                     })
         self._sort_formats(formats)
@@ -134,6 +143,15 @@ class AtresPlayerIE(InfoExtractor):
         description = xpath_text(art, './description', 'description')
         thumbnail = xpath_text(episode, './media/asset/files/background', 'thumbnail')
 
+        subtitles = {}
+        subtitle = xpath_text(episode, './media/asset/files/subtitle', 'subtitle')
+        if subtitle:
+            subtitles['es'] = subtitle
+
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(video_id, subtitles)
+            return
+
         return {
             'id': video_id,
             'title': title,
@@ -141,4 +159,5 @@ class AtresPlayerIE(InfoExtractor):
             'thumbnail': thumbnail,
             'duration': duration,
             'formats': formats,
+            'subtitles': self.extract_subtitles(video_id, subtitles),
         }

youtube_dl/extractor/audiomack.py

@@ -88,16 +88,21 @@ class AudiomackAlbumIE(InfoExtractor):
         # Album playlist ripped from fakeshoredrive with no metadata
         {
             'url': 'http://www.audiomack.com/album/fakeshoredrive/ppp-pistol-p-project',
+            'info_dict': {
+                'title': 'PPP (Pistol P Project)',
+                'id': '837572',
+            },
             'playlist': [{
                 'info_dict': {
-                    'title': '9.-heaven-or-hell-chimaca-ft-zuse-prod-by-dj-fu',
-                    'id': '9.-heaven-or-hell-chimaca-ft-zuse-prod-by-dj-fu',
+                    'title': 'PPP (Pistol P Project) - 9. Heaven or Hell (CHIMACA) ft Zuse (prod by DJ FU)',
+                    'id': '837577',
                     'ext': 'mp3',
+                    'uploader': 'Lil Herb a.k.a. G Herbo',
                 }
             }],
             'params': {
-                'playliststart': 8,
-                'playlistend': 8,
+                'playliststart': 9,
+                'playlistend': 9,
             }
         }
     ]

youtube_dl/extractor/auengine.py

@@ -1,50 +0,0 @@
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..compat import compat_urllib_parse
-from ..utils import (
-    determine_ext,
-    ExtractorError,
-    remove_end,
-)
-
-
-class AUEngineIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?auengine\.com/embed\.php\?.*?file=(?P<id>[^&]+).*?'
-
-    _TEST = {
-        'url': 'http://auengine.com/embed.php?file=lfvlytY6&w=650&h=370',
-        'md5': '48972bdbcf1a3a2f5533e62425b41d4f',
-        'info_dict': {
-            'id': 'lfvlytY6',
-            'ext': 'mp4',
-            'title': '[Commie]The Legend of the Legendary Heroes - 03 - Replication Eye (Alpha Stigma)[F9410F5A]'
-        }
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, video_id)
-        title = self._html_search_regex(
-            r'<title>\s*(?P<title>.+?)\s*</title>', webpage, 'title')
-        video_urls = re.findall(r'http://\w+.auengine.com/vod/.*[^\W]', webpage)
-        video_url = compat_urllib_parse.unquote(video_urls[0])
-        thumbnails = re.findall(r'http://\w+.auengine.com/thumb/.*[^\W]', webpage)
-        thumbnail = compat_urllib_parse.unquote(thumbnails[0])
-
-        if not video_url:
-            raise ExtractorError('Could not find video URL')
-
-        ext = '.' + determine_ext(video_url)
-        title = remove_end(title, ext)
-
-        return {
-            'id': video_id,
-            'url': video_url,
-            'title': title,
-            'thumbnail': thumbnail,
-            'http_referer': 'http://www.auengine.com/flowplayer/flowplayer.commercial-3.2.14.swf',
-        }

youtube_dl/extractor/bbccouk.py

@@ -10,7 +10,7 @@ from ..compat import compat_HTTPError
 class BBCCoUkIE(SubtitlesInfoExtractor):
     IE_NAME = 'bbc.co.uk'
     IE_DESC = 'BBC iPlayer'
-    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})'
 
     _TESTS = [
         {
@@ -118,6 +118,9 @@ class BBCCoUkIE(SubtitlesInfoExtractor):
         }, {
             'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
             'only_matching': True,
+        }, {
+            'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
+            'only_matching': True,
         }
     ]
 

youtube_dl/extractor/bliptv.py

@@ -199,7 +199,7 @@ class BlipTVIE(SubtitlesInfoExtractor):
         # For some weird reason, blip.tv serves a video instead of subtitles
         # when we request with a common UA
         req = compat_urllib_request.Request(url)
-        req.add_header('Youtubedl-user-agent', 'youtube-dl')
+        req.add_header('User-Agent', 'youtube-dl')
         return self._download_webpage(req, None, note=False)
 
 

youtube_dl/extractor/cinchcast.py

@@ -28,12 +28,10 @@ class CinchcastIE(InfoExtractor):
             item, './{http://developer.longtailvideo.com/trac/}date')
         upload_date = unified_strdate(date_str, day_first=False)
         # duration is present but wrong
-        formats = []
-        formats.append({
+        formats = [{
             'format_id': 'main',
-            'url': item.find(
-                './{http://search.yahoo.com/mrss/}content').attrib['url'],
-        })
+            'url': item.find('./{http://search.yahoo.com/mrss/}content').attrib['url'],
+        }]
         backup_url = xpath_text(
             item, './{http://developer.longtailvideo.com/trac/}backupContent')
         if backup_url:

youtube_dl/extractor/cliphunter.py

@@ -1,9 +1,7 @@
 from __future__ import unicode_literals
 
-import json
-import re
-
 from .common import InfoExtractor
+from ..utils import determine_ext
 
 
 _translation_table = {
@@ -27,10 +25,10 @@ class CliphunterIE(InfoExtractor):
     '''
     _TEST = {
         'url': 'http://www.cliphunter.com/w/1012420/Fun_Jynx_Maze_solo',
-        'md5': 'a2ba71eebf523859fe527a61018f723e',
+        'md5': 'b7c9bbd4eb3a226ab91093714dcaa480',
         'info_dict': {
             'id': '1012420',
-            'ext': 'mp4',
+            'ext': 'flv',
             'title': 'Fun Jynx Maze solo',
             'thumbnail': 're:^https?://.*\.jpg$',
             'age_limit': 18,
@@ -44,39 +42,31 @@ class CliphunterIE(InfoExtractor):
         video_title = self._search_regex(
             r'mediaTitle = "([^"]+)"', webpage, 'title')
 
-        pl_fiji = self._search_regex(
-            r'pl_fiji = \'([^\']+)\'', webpage, 'video data')
-        pl_c_qual = self._search_regex(
-            r'pl_c_qual = "(.)"', webpage, 'video quality')
-        video_url = _decode(pl_fiji)
-        formats = [{
-            'url': video_url,
-            'format_id': 'default-%s' % pl_c_qual,
-        }]
+        fmts = {}
+        for fmt in ('mp4', 'flv'):
+            fmt_list = self._parse_json(self._search_regex(
+                r'var %sjson\s*=\s*(\[.*?\]);' % fmt, webpage, '%s formats' % fmt), video_id)
+            for f in fmt_list:
+                fmts[f['fname']] = _decode(f['sUrl'])
 
-        qualities_json = self._search_regex(
-            r'var pl_qualities\s*=\s*(.*?);\n', webpage, 'quality info')
-        qualities_data = json.loads(qualities_json)
+        qualities = self._parse_json(self._search_regex(
+            r'var player_btns\s*=\s*(.*?);\n', webpage, 'quality info'), video_id)
 
-        for i, t in enumerate(
-                re.findall(r"pl_fiji_([a-z0-9]+)\s*=\s*'([^']+')", webpage)):
-            quality_id, crypted_url = t
-            video_url = _decode(crypted_url)
+        formats = []
+        for fname, url in fmts.items():
             f = {
-                'format_id': quality_id,
-                'url': video_url,
-                'quality': i,
+                'url': url,
             }
-            if quality_id in qualities_data:
-                qd = qualities_data[quality_id]
-                m = re.match(
-                    r'''(?x)<b>(?P<width>[0-9]+)x(?P<height>[0-9]+)<\\/b>
-                        \s*\(\s*(?P<tbr>[0-9]+)\s*kb\\/s''', qd)
-                if m:
-                    f['width'] = int(m.group('width'))
-                    f['height'] = int(m.group('height'))
-                    f['tbr'] = int(m.group('tbr'))
+            if fname in qualities:
+                qual = qualities[fname]
+                f.update({
+                    'format_id': '%s_%sp' % (determine_ext(url), qual['h']),
+                    'width': qual['w'],
+                    'height': qual['h'],
+                    'tbr': qual['br'],
+                })
             formats.append(f)
+
         self._sort_formats(formats)
 
         thumbnail = self._search_regex(

youtube_dl/extractor/comedycentral.py

@@ -49,7 +49,9 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor):
                               |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*))
                           )|
                           (?P<interview>
-                              extended-interviews/(?P<interID>[0-9a-z]+)/(?:playlist_tds_extended_)?(?P<interview_title>.*?)(/.*?)?)))
+                              extended-interviews/(?P<interID>[0-9a-z]+)/
+                              (?:playlist_tds_extended_)?(?P<interview_title>[^/?#]*?)
+                              (?:/[^/?#]?|[?#]|$))))
                      '''
     _TESTS = [{
         'url': 'http://thedailyshow.cc.com/watch/thu-december-13-2012/kristen-stewart',
@@ -62,6 +64,38 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor):
             'uploader': 'thedailyshow',
             'title': 'thedailyshow kristen-stewart part 1',
         }
+    }, {
+        'url': 'http://thedailyshow.cc.com/extended-interviews/b6364d/sarah-chayes-extended-interview',
+        'info_dict': {
+            'id': 'sarah-chayes-extended-interview',
+            'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
+            'title': 'thedailyshow Sarah Chayes Extended Interview',
+        },
+        'playlist': [
+            {
+                'info_dict': {
+                    'id': '0baad492-cbec-4ec1-9e50-ad91c291127f',
+                    'ext': 'mp4',
+                    'upload_date': '20150129',
+                    'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
+                    'uploader': 'thedailyshow',
+                    'title': 'thedailyshow sarah-chayes-extended-interview part 1',
+                },
+            },
+            {
+                'info_dict': {
+                    'id': '1e4fb91b-8ce7-4277-bd7c-98c9f1bbd283',
+                    'ext': 'mp4',
+                    'upload_date': '20150129',
+                    'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
+                    'uploader': 'thedailyshow',
+                    'title': 'thedailyshow sarah-chayes-extended-interview part 2',
+                },
+            },
+        ],
+        'params': {
+            'skip_download': True,
+        },
     }, {
         'url': 'http://thedailyshow.cc.com/extended-interviews/xm3fnq/andrew-napolitano-extended-interview',
         'only_matching': True,
@@ -230,6 +264,7 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor):
 
         return {
             '_type': 'playlist',
+            'id': epTitle,
             'entries': entries,
             'title': show_name + ' ' + title,
             'description': description,

youtube_dl/extractor/common.py

@@ -14,6 +14,7 @@ import xml.etree.ElementTree
 
 from ..compat import (
     compat_cookiejar,
+    compat_HTTPError,
     compat_http_client,
     compat_urllib_error,
     compat_urllib_parse_urlparse,
@@ -26,6 +27,7 @@ from ..utils import (
     compiled_regex_type,
     ExtractorError,
     float_or_none,
+    HEADRequest,
     int_or_none,
     RegexNotFoundError,
     sanitize_filename,
@@ -87,7 +89,8 @@ class InfoExtractor(object):
                     * player_url SWF Player URL (used for rtmpdump).
                     * protocol   The protocol that will be used for the actual
                                  download, lower-case.
-                                 "http", "https", "rtsp", "rtmp", "m3u8" or so.
+                                 "http", "https", "rtsp", "rtmp", "rtmpe",
+                                 "m3u8", or "m3u8_native".
                     * preference Order number of this format. If this field is
                                  present and not None, the formats get sorted
                                  by this field, regardless of all other values.
@@ -108,15 +111,17 @@ class InfoExtractor(object):
                                   (quality takes higher priority)
                                  -1 for default (order by other properties),
                                  -2 or smaller for less than default.
-                    * http_referer  HTTP Referer header value to set.
                     * http_method  HTTP method to use for the download.
                     * http_headers  A dictionary of additional HTTP headers
                                  to add to the request.
                     * http_post_data  Additional data to send with a POST
                                  request.
                     * stretched_ratio  If given and not 1, indicates that the
-                                       video's pixels are not square.
-                                       width : height ratio as float.
+                                 video's pixels are not square.
+                                 width : height ratio as float.
+                    * no_resume  The server does not support resuming the
+                                 (HTTP or RTMP) download. Boolean.
+
     url:            Final video URL.
     ext:            Video filename extension.
     format:         The video format, defaults to ext (used for --get-format)
@@ -130,7 +135,9 @@ class InfoExtractor(object):
                     something like "4234987", title "Dancing naked mole rats",
                     and display_id "dancing-naked-mole-rats"
     thumbnails:     A list of dictionaries, with the following entries:
+                        * "id" (optional, string) - Thumbnail format ID
                         * "url"
+                        * "preference" (optional, int) - quality of the image
                         * "width" (optional, int)
                         * "height" (optional, int)
                         * "resolution" (optional, string "{width}x{height"},
@@ -712,6 +719,27 @@ class InfoExtractor(object):
             )
         formats.sort(key=_formats_key)
 
+    def _check_formats(self, formats, video_id):
+        if formats:
+            formats[:] = filter(
+                lambda f: self._is_valid_url(
+                    f['url'], video_id,
+                    item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'),
+                formats)
+
+    def _is_valid_url(self, url, video_id, item='video'):
+        try:
+            self._request_webpage(
+                HEADRequest(url), video_id,
+                'Checking %s URL' % item)
+            return True
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError):
+                self.report_warning(
+                    '%s URL is invalid, skipping' % item, video_id)
+                return False
+            raise
+
     def http_scheme(self):
         """ Either "http:" or "https:", depending on the user's preferences """
         return (

youtube_dl/extractor/ctsnews.py

@@ -0,0 +1,93 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import parse_iso8601, ExtractorError
+
+
+class CtsNewsIE(InfoExtractor):
+    # https connection failed (Connection reset)
+    _VALID_URL = r'http://news\.cts\.com\.tw/[a-z]+/[a-z]+/\d+/(?P<id>\d+)\.html'
+    _TESTS = [{
+        'url': 'http://news.cts.com.tw/cts/international/201501/201501291578109.html',
+        'md5': 'a9875cb790252b08431186d741beaabe',
+        'info_dict': {
+            'id': '201501291578109',
+            'ext': 'mp4',
+            'title': '以色列.真主黨交火 3人死亡',
+            'description': 'md5:95e9b295c898b7ff294f09d450178d7d',
+            'timestamp': 1422528540,
+            'upload_date': '20150129',
+        }
+    }, {
+        # News count not appear on page but still available in database
+        'url': 'http://news.cts.com.tw/cts/international/201309/201309031304098.html',
+        'md5': '3aee7e0df7cdff94e43581f54c22619e',
+        'info_dict': {
+            'id': '201309031304098',
+            'ext': 'mp4',
+            'title': '韓國31歲童顏男 貌如十多歲小孩',
+            'description': 'md5:f183feeba3752b683827aab71adad584',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'timestamp': 1378205880,
+            'upload_date': '20130903',
+        }
+    }, {
+        # With Youtube embedded video
+        'url': 'http://news.cts.com.tw/cts/money/201501/201501291578003.html',
+        'md5': '1d842c771dc94c8c3bca5af2cc1db9c5',
+        'add_ie': ['Youtube'],
+        'info_dict': {
+            'id': 'OVbfO7d0_hQ',
+            'ext': 'mp4',
+            'title': 'iPhone6熱銷 蘋果財報亮眼',
+            'description': 'md5:f395d4f485487bb0f992ed2c4b07aa7d',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'upload_date': '20150128',
+            'uploader_id': 'TBSCTS',
+            'uploader': '中華電視公司',
+        }
+    }]
+
+    def _real_extract(self, url):
+        news_id = self._match_id(url)
+        page = self._download_webpage(url, news_id)
+
+        if self._search_regex(r'(CTSPlayer2)', page, 'CTSPlayer2 identifier', default=None):
+            feed_url = self._html_search_regex(
+                r'(http://news\.cts\.com\.tw/action/mp4feed\.php\?news_id=\d+)',
+                page, 'feed url')
+            video_url = self._download_webpage(
+                feed_url, news_id, note='Fetching feed')
+        else:
+            self.to_screen('Not CTSPlayer video, trying Youtube...')
+            youtube_url = self._search_regex(
+                r'src="(//www\.youtube\.com/embed/[^"]+)"', page, 'youtube url',
+                default=None)
+            if not youtube_url:
+                raise ExtractorError('The news includes no videos!', expected=True)
+
+            return {
+                '_type': 'url',
+                'url': youtube_url,
+                'ie_key': 'Youtube',
+            }
+
+        description = self._html_search_meta('description', page)
+        title = self._html_search_meta('title', page)
+        thumbnail = self._html_search_meta('image', page)
+
+        datetime_str = self._html_search_regex(
+            r'(\d{4}/\d{2}/\d{2} \d{2}:\d{2})', page, 'date and time')
+        # Transform into ISO 8601 format with timezone info
+        datetime_str = datetime_str.replace('/', '-') + ':00+0800'
+        timestamp = parse_iso8601(datetime_str, delimiter=' ')
+
+        return {
+            'id': news_id,
+            'url': video_url,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
+        }

youtube_dl/extractor/dctp.py

@@ -0,0 +1,57 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+
+
+class DctpTvIE(InfoExtractor):
+    _VALID_URL = r'http://www.dctp.tv/(#/)?filme/(?P<id>.+?)/$'
+    _TEST = {
+        'url': 'http://www.dctp.tv/filme/videoinstallation-fuer-eine-kaufhausfassade/',
+        'info_dict': {
+            'id': '1324',
+            'display_id': 'videoinstallation-fuer-eine-kaufhausfassade',
+            'ext': 'flv',
+            'title': 'Videoinstallation für eine Kaufhausfassade'
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        base_url = 'http://dctp-ivms2-restapi.s3.amazonaws.com/'
+        version_json = self._download_json(
+            base_url + 'version.json',
+            video_id, note='Determining file version')
+        version = version_json['version_name']
+        info_json = self._download_json(
+            '{0}{1}/restapi/slugs/{2}.json'.format(base_url, version, video_id),
+            video_id, note='Fetching object ID')
+        object_id = compat_str(info_json['object_id'])
+        meta_json = self._download_json(
+            '{0}{1}/restapi/media/{2}.json'.format(base_url, version, object_id),
+            video_id, note='Downloading metadata')
+        uuid = meta_json['uuid']
+        title = meta_json['title']
+        wide = meta_json['is_wide']
+        if wide:
+            ratio = '16x9'
+        else:
+            ratio = '4x3'
+        play_path = 'mp4:{0}_dctp_0500_{1}.m4v'.format(uuid, ratio)
+
+        servers_json = self._download_json(
+            'http://www.dctp.tv/streaming_servers/',
+            video_id, note='Downloading server list')
+        url = servers_json[0]['endpoint']
+
+        return {
+            'id': object_id,
+            'title': title,
+            'format': 'rtmp',
+            'url': url,
+            'play_path': play_path,
+            'rtmp_real_time': True,
+            'ext': 'flv',
+            'display_id': video_id
+        }

youtube_dl/extractor/defense.py

@@ -1,40 +1,38 @@
 from __future__ import unicode_literals
 
-import re
-import json
-
 from .common import InfoExtractor
 
 
 class DefenseGouvFrIE(InfoExtractor):
     IE_NAME = 'defense.gouv.fr'
-    _VALID_URL = (r'http://.*?\.defense\.gouv\.fr/layout/set/'
-                  r'ligthboxvideo/base-de-medias/webtv/(.*)')
+    _VALID_URL = r'http://.*?\.defense\.gouv\.fr/layout/set/ligthboxvideo/base-de-medias/webtv/(?P<id>[^/?#]*)'
 
     _TEST = {
         'url': 'http://www.defense.gouv.fr/layout/set/ligthboxvideo/base-de-medias/webtv/attaque-chimique-syrienne-du-21-aout-2013-1',
-        'file': '11213.mp4',
         'md5': '75bba6124da7e63d2d60b5244ec9430c',
-        "info_dict": {
-            "title": "attaque-chimique-syrienne-du-21-aout-2013-1"
+        'info_dict': {
+            'id': '11213',
+            'ext': 'mp4',
+            'title': 'attaque-chimique-syrienne-du-21-aout-2013-1'
         }
     }
 
     def _real_extract(self, url):
-        title = re.match(self._VALID_URL, url).group(1)
+        title = self._match_id(url)
         webpage = self._download_webpage(url, title)
+
         video_id = self._search_regex(
             r"flashvars.pvg_id=\"(\d+)\";",
             webpage, 'ID')
 
         json_url = ('http://static.videos.gouv.fr/brightcovehub/export/json/'
                     + video_id)
-        info = self._download_webpage(json_url, title,
-                                      'Downloading JSON config')
-        video_url = json.loads(info)['renditions'][0]['url']
+        info = self._download_json(json_url, title, 'Downloading JSON config')
+        video_url = info['renditions'][0]['url']
 
-        return {'id': video_id,
-                'ext': 'mp4',
-                'url': video_url,
-                'title': title,
-                }
+        return {
+            'id': video_id,
+            'ext': 'mp4',
+            'url': video_url,
+            'title': title,
+        }

youtube_dl/extractor/drtv.py

@@ -6,7 +6,7 @@ from ..utils import parse_iso8601
 
 
 class DRTVIE(SubtitlesInfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dr\.dk/tv/se/(?:[^/]+/)+(?P<id>[\da-z-]+)(?:[/#?]|$)'
+    _VALID_URL = r'https?://(?:www\.)?dr\.dk/tv/se/(?:[^/]+/)*(?P<id>[\da-z-]+)(?:[/#?]|$)'
 
     _TEST = {
         'url': 'http://www.dr.dk/tv/se/partiets-mand/partiets-mand-7-8',
@@ -25,9 +25,15 @@ class DRTVIE(SubtitlesInfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        programcard = self._download_json(
-            'http://www.dr.dk/mu/programcard/expanded/%s' % video_id, video_id, 'Downloading video JSON')
+        webpage = self._download_webpage(url, video_id)
 
+        video_id = self._search_regex(
+            r'data-(?:material-identifier|episode-slug)="([^"]+)"',
+            webpage, 'video id')
+
+        programcard = self._download_json(
+            'http://www.dr.dk/mu/programcard/expanded/%s' % video_id,
+            video_id, 'Downloading video JSON')
         data = programcard['Data'][0]
 
         title = data['Title']
@@ -48,14 +54,20 @@ class DRTVIE(SubtitlesInfoExtractor):
             elif asset['Kind'] == 'VideoResource':
                 duration = asset['DurationInMilliseconds'] / 1000.0
                 restricted_to_denmark = asset['RestrictedToDenmark']
+                spoken_subtitles = asset['Target'] == 'SpokenSubtitles'
                 for link in asset['Links']:
                     target = link['Target']
                     uri = link['Uri']
+                    format_id = target
+                    preference = -1 if target == 'HDS' else -2
+                    if spoken_subtitles:
+                        preference -= 2
+                        format_id += '-spoken-subtitles'
                     formats.append({
                         'url': uri + '?hdcore=3.3.0&plugin=aasp-3.3.0.99.43' if target == 'HDS' else uri,
-                        'format_id': target,
+                        'format_id': format_id,
                         'ext': link['FileFormat'],
-                        'preference': -1 if target == 'HDS' else -2,
+                        'preference': preference,
                     })
                 subtitles_list = asset.get('SubtitlesList')
                 if isinstance(subtitles_list, list):

youtube_dl/extractor/fc2.py

@@ -5,6 +5,7 @@ import hashlib
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_urllib_parse,
     compat_urllib_request,
     compat_urlparse,
 )
@@ -16,7 +17,8 @@ from ..utils import (
 class FC2IE(InfoExtractor):
     _VALID_URL = r'^http://video\.fc2\.com/(?:[^/]+/)?content/(?P<id>[^/]+)'
     IE_NAME = 'fc2'
-    _TEST = {
+    _NETRC_MACHINE = 'fc2'
+    _TESTS = [{
         'url': 'http://video.fc2.com/en/content/20121103kUan1KHs',
         'md5': 'a6ebe8ebe0396518689d963774a54eb7',
         'info_dict': {
@@ -24,12 +26,57 @@ class FC2IE(InfoExtractor):
             'ext': 'flv',
             'title': 'Boxing again with Puff',
         },
-    }
+    }, {
+        'url': 'http://video.fc2.com/en/content/20150125cEva0hDn/',
+        'info_dict': {
+            'id': '20150125cEva0hDn',
+            'ext': 'mp4',
+        },
+        'params': {
+            'username': 'ytdl@yt-dl.org',
+            'password': '(snip)',
+            'skip': 'requires actual password'
+        }
+    }]
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None or password is None:
+            return False
+
+        # Log in
+        login_form_strs = {
+            'email': username,
+            'password': password,
+            'done': 'video',
+            'Submit': ' Login ',
+        }
+
+        # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
+        # chokes on unicode
+        login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
+        login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
+        request = compat_urllib_request.Request(
+            'https://secure.id.fc2.com/index.php?mode=login&switch_language=en', login_data)
+
+        login_results = self._download_webpage(request, None, note='Logging in', errnote='Unable to log in')
+        if 'mode=redirect&login=done' not in login_results:
+            self.report_warning('unable to log in: bad username or password')
+            return False
+
+        # this is also needed
+        login_redir = compat_urllib_request.Request('http://id.fc2.com/?mode=redirect&login=done')
+        self._download_webpage(
+            login_redir, None, note='Login redirect', errnote='Login redirect failed')
+
+        return True
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+        self._login()
         webpage = self._download_webpage(url, video_id)
         self._downloader.cookiejar.clear_session_cookies()  # must clear
+        self._login()
 
         title = self._og_search_title(webpage)
         thumbnail = self._og_search_thumbnail(webpage)
@@ -46,7 +93,12 @@ class FC2IE(InfoExtractor):
         info = compat_urlparse.parse_qs(info_webpage)
 
         if 'err_code' in info:
-            raise ExtractorError('Error code: %s' % info['err_code'][0])
+            # most of the time we can still download wideo even if err_code is 403 or 602
+            self.report_warning(
+                'Error code was: %s... but still trying' % info['err_code'][0])
+
+        if 'filepath' not in info:
+            raise ExtractorError('Cannot download file. Are you logged in?')
 
         video_url = info['filepath'][0] + '?mid=' + info['mid'][0]
         title_info = info.get('title')

youtube_dl/extractor/folketinget.py

@@ -16,6 +16,7 @@ class FolketingetIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?ft\.dk/webtv/video/[^?#]*?\.(?P<id>[0-9]+)\.aspx'
     _TEST = {
         'url': 'http://www.ft.dk/webtv/video/20141/eru/td.1165642.aspx?as=1#player',
+        'md5': '6269e8626fa1a891bf5369b386ae996a',
         'info_dict': {
             'id': '1165642',
             'ext': 'mp4',
@@ -29,9 +30,6 @@ class FolketingetIE(InfoExtractor):
             'upload_date': '20141120',
             'duration': 3960,
         },
-        'params': {
-            'skip_download': 'rtmpdump required',
-        }
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/francetv.py

@@ -230,12 +230,13 @@ class FranceTVIE(FranceTVBaseInfoExtractor):
 
 class GenerationQuoiIE(InfoExtractor):
     IE_NAME = 'france2.fr:generation-quoi'
-    _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<name>.*)(\?|$)'
+    _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<id>[^/?#]+)'
 
     _TEST = {
         'url': 'http://generation-quoi.france2.fr/portrait/garde-a-vous',
-        'file': 'k7FJX8VBcvvLmX4wA5Q.mp4',
         'info_dict': {
+            'id': 'k7FJX8VBcvvLmX4wA5Q',
+            'ext': 'mp4',
             'title': 'Génération Quoi - Garde à Vous',
             'uploader': 'Génération Quoi',
         },
@@ -243,14 +244,12 @@ class GenerationQuoiIE(InfoExtractor):
             # It uses Dailymotion
             'skip_download': True,
         },
-        'skip': 'Only available from France',
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        name = mobj.group('name')
-        info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % name)
-        info_json = self._download_webpage(info_url, name)
+        display_id = self._match_id(url)
+        info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % display_id)
+        info_json = self._download_webpage(info_url, display_id)
         info = json.loads(info_json)
         return self.url_result('http://www.dailymotion.com/video/%s' % info['id'],
                                ie='Dailymotion')

youtube_dl/extractor/gamestar.py

@@ -1,8 +1,6 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
@@ -29,9 +27,7 @@ class GameStarIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
         og_title = self._og_search_title(webpage)

youtube_dl/extractor/generic.py

@@ -17,6 +17,7 @@ from ..utils import (
     ExtractorError,
     float_or_none,
     HEADRequest,
+    is_html,
     orderedSet,
     parse_xml,
     smuggle_url,
@@ -361,7 +362,7 @@ class GenericIE(InfoExtractor):
             'info_dict': {
                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
                 'title': 'Zero Punctuation',
-                'description': 're:'
+                'description': 're:.*groundbreaking video review series.*'
             },
             'playlist_mincount': 11,
         },
@@ -488,6 +489,29 @@ class GenericIE(InfoExtractor):
                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
             }
         },
+        # Cinerama player
+        {
+            'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
+            'info_dict': {
+                'id': '730m_DandD_1901_512k',
+                'ext': 'mp4',
+                'uploader': 'www.abc.net.au',
+                'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
+            }
+        },
+        # embedded viddler video
+        {
+            'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597',
+            'info_dict': {
+                'id': '4d03aad9',
+                'ext': 'mp4',
+                'uploader': 'deadspin',
+                'title': 'WALL-TO-GORTAT',
+                'timestamp': 1422285291,
+                'upload_date': '20150126',
+            },
+            'add_ie': ['Viddler'],
+        }
     ]
 
     def report_following_redirect(self, new_url):
@@ -647,7 +671,7 @@ class GenericIE(InfoExtractor):
         # Maybe it's a direct link to a video?
         # Be careful not to download the whole thing!
         first_bytes = full_response.read(512)
-        if not re.match(r'^\s*<', first_bytes.decode('utf-8', 'replace')):
+        if not is_html(first_bytes):
             self._downloader.report_warning(
                 'URL could be a direct video link, returning it as such.')
             upload_date = unified_strdate(
@@ -849,9 +873,16 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group('url'))
 
+        # Look for embedded Viddler player
+        mobj = re.search(
+            r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1',
+            webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'))
+
         # Look for Ooyala videos
-        mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
-                re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
+        mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
+                re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
         if mobj is not None:
             return OoyalaIE._build_url_result(mobj.group('ec'))
 
@@ -1042,9 +1073,13 @@ class GenericIE(InfoExtractor):
             found = filter_video(re.findall(r'''(?xs)
                 flowplayer\("[^"]+",\s*
                     \{[^}]+?\}\s*,
-                    \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
+                    \s*\{[^}]+? ["']?clip["']?\s*:\s*\{\s*
                         ["']?url["']?\s*:\s*["']([^"']+)["']
             ''', webpage))
+        if not found:
+            # Cinerama player
+            found = re.findall(
+                r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
         if not found:
             # Try to find twitter cards info
             found = filter_video(re.findall(

youtube_dl/extractor/globo.py

@@ -70,6 +70,19 @@ class GloboIE(InfoExtractor):
                 'like_count': int,
             }
         },
+        {
+            'url': 'http://globotv.globo.com/canal-brasil/sangue-latino/t/todos-os-videos/v/ator-e-diretor-argentino-ricado-darin-fala-sobre-utopias-e-suas-perdas/3928201/',
+            'md5': 'c1defca721ce25b2354e927d3e4b3dec',
+            'info_dict': {
+                'id': '3928201',
+                'ext': 'mp4',
+                'title': 'Ator e diretor argentino, Ricado Darín fala sobre utopias e suas perdas',
+                'duration': 1472.906,
+                'uploader': 'Canal Brasil',
+                'uploader_id': 705,
+                'like_count': int,
+            }
+        },
     ]
 
     class MD5():
@@ -381,11 +394,16 @@ class GloboIE(InfoExtractor):
             signed_md5 = self.MD5.b64_md5(received_md5 + compat_str(sign_time) + padding)
             signed_hash = hash_code + compat_str(received_time) + received_random + compat_str(sign_time) + padding + signed_md5
 
-            formats.append({
-                'url': '%s?h=%s&k=%s' % (resource['url'], signed_hash, 'flash'),
-                'format_id': resource_id,
-                'height': resource['height']
-            })
+            resource_url = resource['url']
+            signed_url = '%s?h=%s&k=%s' % (resource_url, signed_hash, 'flash')
+            if resource_id.endswith('m3u8') or resource_url.endswith('.m3u8'):
+                formats.extend(self._extract_m3u8_formats(signed_url, resource_id, 'mp4'))
+            else:
+                formats.append({
+                    'url': signed_url,
+                    'format_id': resource_id,
+                    'height': resource.get('height'),
+                })
 
         self._sort_formats(formats)
 

youtube_dl/extractor/grooveshark.py

@@ -83,7 +83,7 @@ class GroovesharkIE(InfoExtractor):
         return compat_urlparse.urlunparse((uri.scheme, uri.netloc, obj['attrs']['data'], None, None, None))
 
     def _transform_bootstrap(self, js):
-        return re.split('(?m)^\s*try\s*{', js)[0] \
+        return re.split('(?m)^\s*try\s*\{', js)[0] \
                  .split(' = ', 1)[1].strip().rstrip(';')
 
     def _transform_meta(self, js):

youtube_dl/extractor/historicfilms.py

@@ -0,0 +1,46 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import parse_duration
+
+
+class HistoricFilmsIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?historicfilms\.com/(?:tapes/|play)(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.historicfilms.com/tapes/4728',
+        'md5': 'd4a437aec45d8d796a38a215db064e9a',
+        'info_dict': {
+            'id': '4728',
+            'ext': 'mov',
+            'title': 'Historic Films: GP-7',
+            'description': 'md5:1a86a0f3ac54024e419aba97210d959a',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 2096,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        tape_id = self._search_regex(
+            r'class="tapeId">([^<]+)<', webpage, 'tape id')
+
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage)
+        thumbnail = self._html_search_meta(
+            'thumbnailUrl', webpage, 'thumbnails') or self._og_search_thumbnail(webpage)
+        duration = parse_duration(self._html_search_meta(
+            'duration', webpage, 'duration'))
+
+        video_url = 'http://www.historicfilms.com/video/%s_%s_web.mov' % (tape_id, video_id)
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+        }

youtube_dl/extractor/ivi.py

@@ -16,7 +16,7 @@ from ..utils import (
 class IviIE(InfoExtractor):
     IE_DESC = 'ivi.ru'
     IE_NAME = 'ivi'
-    _VALID_URL = r'https?://(?:www\.)?ivi\.ru/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<videoid>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?ivi\.ru/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<id>\d+)'
 
     _TESTS = [
         # Single movie
@@ -63,29 +63,34 @@ class IviIE(InfoExtractor):
         return int(m.group('commentcount')) if m is not None else 0
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('videoid')
+        video_id = self._match_id(url)
 
         api_url = 'http://api.digitalaccess.ru/api/json/'
 
-        data = {'method': 'da.content.get',
-                'params': [video_id, {'site': 's183',
-                                      'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
-                                      'contentid': video_id
-                                      }
-                           ]
+        data = {
+            'method': 'da.content.get',
+            'params': [
+                video_id, {
+                    'site': 's183',
+                    'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
+                    'contentid': video_id
                 }
+            ]
+        }
 
         request = compat_urllib_request.Request(api_url, json.dumps(data))
 
-        video_json_page = self._download_webpage(request, video_id, 'Downloading video JSON')
+        video_json_page = self._download_webpage(
+            request, video_id, 'Downloading video JSON')
         video_json = json.loads(video_json_page)
 
         if 'error' in video_json:
             error = video_json['error']
             if error['origin'] == 'NoRedisValidData':
                 raise ExtractorError('Video %s does not exist' % video_id, expected=True)
-            raise ExtractorError('Unable to download video %s: %s' % (video_id, error['message']), expected=True)
+            raise ExtractorError(
+                'Unable to download video %s: %s' % (video_id, error['message']),
+                expected=True)
 
         result = video_json['result']
 

youtube_dl/extractor/kankan.py

@@ -13,17 +13,17 @@ class KankanIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://yinyue.kankan.com/vod/48/48863.shtml',
-        'file': '48863.flv',
         'md5': '29aca1e47ae68fc28804aca89f29507e',
         'info_dict': {
+            'id': '48863',
+            'ext': 'flv',
             'title': 'Ready To Go',
         },
         'skip': 'Only available from China',
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
         title = self._search_regex(r'(?:G_TITLE=|G_MOVIE_TITLE = )[\'"](.+?)[\'"]', webpage, 'video title')

youtube_dl/extractor/keezmovies.py

@@ -7,10 +7,6 @@ from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse_urlparse,
     compat_urllib_request,
-    compat_urllib_parse,
-)
-from ..aes import (
-    aes_decrypt_text
 )
 
 
@@ -18,9 +14,10 @@ class KeezMoviesIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?keezmovies\.com/video/.+?(?P<id>[0-9]+)(?:[/?&]|$)'
     _TEST = {
         'url': 'http://www.keezmovies.com/video/petite-asian-lady-mai-playing-in-bathtub-1214711',
-        'file': '1214711.mp4',
         'md5': '6e297b7e789329923fcf83abb67c9289',
         'info_dict': {
+            'id': '1214711',
+            'ext': 'mp4',
             'title': 'Petite Asian Lady Mai Playing In Bathtub',
             'age_limit': 18,
         }
@@ -39,11 +36,10 @@ class KeezMoviesIE(InfoExtractor):
             embedded_url = mobj.group(1)
             return self.url_result(embedded_url)
 
-        video_title = self._html_search_regex(r'<h1 [^>]*>([^<]+)', webpage, 'title')
-        video_url = compat_urllib_parse.unquote(self._html_search_regex(r'video_url=(.+?)&amp;', webpage, 'video_url'))
-        if 'encrypted=true' in webpage:
-            password = self._html_search_regex(r'video_title=(.+?)&amp;', webpage, 'password')
-            video_url = aes_decrypt_text(video_url, password, 32).decode('utf-8')
+        video_title = self._html_search_regex(
+            r'<h1 [^>]*>([^<]+)', webpage, 'title')
+        video_url = self._html_search_regex(
+            r'(?s)html5VideoPlayer = .*?src="([^"]+)"', webpage, 'video URL')
         path = compat_urllib_parse_urlparse(video_url).path
         extension = os.path.splitext(path)[1][1:]
         format = path.split('/')[4].split('_')[:2]

youtube_dl/extractor/krasview.py

@@ -2,18 +2,17 @@
 from __future__ import unicode_literals
 
 import json
-import re
 
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
-    unescapeHTML,
+    js_to_json,
 )
 
 
 class KrasViewIE(InfoExtractor):
     IE_DESC = 'Красвью'
-    _VALID_URL = r'https?://krasview\.ru/video/(?P<id>\d+)'
+    _VALID_URL = r'https?://krasview\.ru/(?:video|embed)/(?P<id>\d+)'
 
     _TEST = {
         'url': 'http://krasview.ru/video/512228',
@@ -29,20 +28,18 @@ class KrasViewIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
 
-        flashvars = json.loads(self._search_regex(
-            r'flashvars\s*:\s*({.+?})\s*}\);', webpage, 'flashvars'))
+        flashvars = json.loads(js_to_json(self._search_regex(
+            r'video_Init\(({.+?})', 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'])
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage, default=None)
+        thumbnail = flashvars.get('image') or self._og_search_thumbnail(webpage)
+        duration = int_or_none(flashvars.get('duration'))
         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'))
 
@@ -53,7 +50,6 @@ class KrasViewIE(InfoExtractor):
             'description': description,
             'thumbnail': thumbnail,
             'duration': duration,
-            'filesize': filesize,
             'width': width,
             'height': height,
         }

youtube_dl/extractor/la7.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     parse_duration,
@@ -20,9 +18,10 @@ class LA7IE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.la7.tv/richplayer/?assetid=50355319',
-        'file': '50355319.mp4',
         'md5': 'ec7d1f0224d20ba293ab56cf2259651f',
         'info_dict': {
+            'id': '50355319',
+            'ext': 'mp4',
             'title': 'IL DIVO',
             'description': 'Un film di Paolo Sorrentino con Toni Servillo, Anna Bonaiuto, Giulio Bosetti  e Flavio Bucci',
             'duration': 6254,
@@ -31,9 +30,7 @@ class LA7IE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
+        video_id = self._match_id(url)
         xml_url = 'http://www.la7.tv/repliche/content/index.php?contentId=%s' % video_id
         doc = self._download_xml(xml_url, video_id)
 

youtube_dl/extractor/liveleak.py

@@ -8,20 +8,20 @@ from ..utils import int_or_none
 
 
 class LiveLeakIE(InfoExtractor):
-    _VALID_URL = r'^(?:http://)?(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<video_id>[\w_]+)(?:.*)'
+    _VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<id>[\w_]+)(?:.*)'
     _TESTS = [{
         'url': 'http://www.liveleak.com/view?i=757_1364311680',
-        'md5': '0813c2430bea7a46bf13acf3406992f4',
+        'md5': '50f79e05ba149149c1b4ea961223d5b3',
         'info_dict': {
             'id': '757_1364311680',
-            'ext': 'mp4',
+            'ext': 'flv',
             'description': 'extremely bad day for this guy..!',
             'uploader': 'ljfriel2',
             'title': 'Most unlucky car accident'
         }
     }, {
         'url': 'http://www.liveleak.com/view?i=f93_1390833151',
-        'md5': 'd3f1367d14cc3c15bf24fbfbe04b9abf',
+        'md5': 'b13a29626183c9d33944e6a04f41aafc',
         'info_dict': {
             'id': 'f93_1390833151',
             'ext': 'mp4',
@@ -43,8 +43,7 @@ class LiveLeakIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('video_id')
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
         video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
@@ -81,9 +80,19 @@ class LiveLeakIE(InfoExtractor):
         sources = json.loads(sources_json)
 
         formats = [{
+            'format_id': '%s' % i,
             'format_note': s.get('label'),
             'url': s['file'],
-        } for s in sources]
+        } for i, s in enumerate(sources)]
+        for i, s in enumerate(sources):
+            orig_url = s['file'].replace('.h264_base.mp4', '')
+            if s['file'] != orig_url:
+                formats.append({
+                    'format_id': 'original-%s' % i,
+                    'format_note': s.get('label'),
+                    'url': orig_url,
+                    'preference': 1,
+                })
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/lnkgo.py

@@ -6,13 +6,12 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
-    js_to_json,
     unified_strdate,
 )
 
 
 class LnkGoIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?lnkgo\.alfa\.lt/visi\-video/(?P<show>[^/]+)/ziurek\-(?P<display_id>[A-Za-z0-9\-]+)'
+    _VALID_URL = r'https?://(?:www\.)?lnkgo\.alfa\.lt/visi-video/(?P<show>[^/]+)/ziurek-(?P<id>[A-Za-z0-9-]+)'
     _TESTS = [{
         'url': 'http://lnkgo.alfa.lt/visi-video/yra-kaip-yra/ziurek-yra-kaip-yra-162',
         'info_dict': {
@@ -51,8 +50,7 @@ class LnkGoIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('display_id')
+        display_id = self._match_id(url)
 
         webpage = self._download_webpage(
             url, display_id, 'Downloading player webpage')
@@ -61,6 +59,8 @@ class LnkGoIE(InfoExtractor):
             r'data-ep="([^"]+)"', webpage, 'video ID')
         title = self._og_search_title(webpage)
         description = self._og_search_description(webpage)
+        upload_date = unified_strdate(self._search_regex(
+            r'class="[^"]*meta-item[^"]*air-time[^"]*">.*?<strong>([^<]+)</strong>', webpage, 'upload date', fatal=False))
 
         thumbnail_w = int_or_none(
             self._og_search_property('image:width', webpage, 'thumbnail width', fatal=False))
@@ -75,39 +75,28 @@ class LnkGoIE(InfoExtractor):
                 'height': thumbnail_h,
             })
 
-        upload_date = unified_strdate(self._search_regex(
-            r'class="meta-item\sair-time">.*?<strong>([^<]+)</strong>', webpage, 'upload date', fatal=False))
-        duration = int_or_none(self._search_regex(
-            r'VideoDuration = "([^"]+)"', webpage, 'duration', fatal=False))
+        config = self._parse_json(self._search_regex(
+            r'episodePlayer\((\{.*?\}),\s*\{', webpage, 'sources'), video_id)
 
-        pg_rating = self._search_regex(
-            r'pgrating="([^"]+)"', webpage, 'PG rating', fatal=False, default='')
-        age_limit = self._AGE_LIMITS.get(pg_rating.upper(), 0)
+        if config.get('pGeo'):
+            self.report_warning(
+                'This content might not be available in your country due to copyright reasons')
 
-        sources_js = self._search_regex(
-            r'(?s)sources:\s(\[.*?\]),', webpage, 'sources')
-        sources = self._parse_json(
-            sources_js, video_id, transform_source=js_to_json)
+        formats = [{
+            'format_id': 'hls',
+            'ext': 'mp4',
+            'url': config['EpisodeVideoLink_HLS'],
+        }]
 
-        formats = []
-        for source in sources:
-            if source.get('provider') == 'rtmp':
-                m = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<play_path>.+)$', source['file'])
-                if not m:
-                    continue
-                formats.append({
-                    'format_id': 'rtmp',
-                    'ext': 'flv',
-                    'url': m.group('url'),
-                    'play_path': m.group('play_path'),
-                    'page_url': url,
-                })
-            elif source.get('file').endswith('.m3u8'):
-                formats.append({
-                    'format_id': 'hls',
-                    'ext': source.get('type', 'mp4'),
-                    'url': source['file'],
-                })
+        m = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<play_path>.+)$', config['EpisodeVideoLink'])
+        if m:
+            formats.append({
+                'format_id': 'rtmp',
+                'ext': 'flv',
+                'url': m.group('url'),
+                'play_path': m.group('play_path'),
+                'page_url': url,
+            })
 
         self._sort_formats(formats)
 
@@ -117,8 +106,8 @@ class LnkGoIE(InfoExtractor):
             'title': title,
             'formats': formats,
             'thumbnails': [thumbnail],
-            'duration': duration,
+            'duration': int_or_none(config.get('VideoTime')),
             'description': description,
-            'age_limit': age_limit,
+            'age_limit': self._AGE_LIMITS.get(config.get('PGRating'), 0),
             'upload_date': upload_date,
         }

youtube_dl/extractor/lynda.py

@@ -85,6 +85,7 @@ class LyndaIE(SubtitlesInfoExtractor):
                 } for format_id, video_url in prioritized_streams['0'].items()
             ])
 
+        self._check_formats(formats, video_id)
         self._sort_formats(formats)
 
         if self._downloader.params.get('listsubtitles', False):

youtube_dl/extractor/macgamestore.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import ExtractorError
 
@@ -13,21 +11,22 @@ class MacGameStoreIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.macgamestore.com/mediaviewer.php?trailer=2450',
-        'file': '2450.m4v',
         'md5': '8649b8ea684b6666b4c5be736ecddc61',
         'info_dict': {
+            'id': '2450',
+            'ext': 'm4v',
             'title': 'Crow',
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(
+            url, video_id, 'Downloading trailer page')
 
-        webpage = self._download_webpage(url, video_id, 'Downloading trailer page')
-
-        if re.search(r'>Missing Media<', webpage) is not None:
-            raise ExtractorError('Trailer %s does not exist' % video_id, expected=True)
+        if '>Missing Media<' in webpage:
+            raise ExtractorError(
+                'Trailer %s does not exist' % video_id, expected=True)
 
         video_title = self._html_search_regex(
             r'<title>MacGameStore: (.*?) Trailer</title>', webpage, 'title')

youtube_dl/extractor/mixcloud.py

@@ -9,7 +9,7 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     HEADRequest,
-    int_or_none,
+    str_to_int,
     parse_iso8601,
 )
 
@@ -85,15 +85,17 @@ class MixcloudIE(InfoExtractor):
         uploader_id = self._search_regex(
             r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
         description = self._og_search_description(webpage)
-        like_count = int_or_none(self._search_regex(
-            r'<meta itemprop="interactionCount" content="UserLikes:([0-9]+)"',
+        like_count = str_to_int(self._search_regex(
+            [r'<meta itemprop="interactionCount" content="UserLikes:([0-9]+)"',
+             r'/favorites/?">([0-9]+)<'],
             webpage, 'like count', fatal=False))
-        view_count = int_or_none(self._search_regex(
-            r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
+        view_count = str_to_int(self._search_regex(
+            [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
+             r'/listeners/?">([0-9,.]+)</a>'],
             webpage, 'play count', fatal=False))
         timestamp = parse_iso8601(self._search_regex(
             r'<time itemprop="dateCreated" datetime="([^"]+)">',
-            webpage, 'upload date'))
+            webpage, 'upload date', default=None))
 
         return {
             'id': track_id,

youtube_dl/extractor/mpora.py

@@ -1,21 +1,19 @@
 from __future__ import unicode_literals
 
-import json
-import re
-
 from .common import InfoExtractor
 from ..utils import int_or_none
 
 
 class MporaIE(InfoExtractor):
-    _VALID_URL = r'^https?://(www\.)?mpora\.(?:com|de)/videos/(?P<id>[^?#/]+)'
+    _VALID_URL = r'https?://(www\.)?mpora\.(?:com|de)/videos/(?P<id>[^?#/]+)'
     IE_NAME = 'MPORA'
 
     _TEST = {
         'url': 'http://mpora.de/videos/AAdo8okx4wiz/embed?locale=de',
-        'file': 'AAdo8okx4wiz.mp4',
         'md5': 'a7a228473eedd3be741397cf452932eb',
         'info_dict': {
+            'id': 'AAdo8okx4wiz',
+            'ext': 'mp4',
             'title': 'Katy Curd -  Winter in the Forest',
             'duration': 416,
             'uploader': 'Peter Newman Media',
@@ -23,14 +21,12 @@ class MporaIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        m = re.match(self._VALID_URL, url)
-        video_id = m.group('id')
-
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
+
         data_json = self._search_regex(
             r"new FM\.Player\('[^']+',\s*(\{.*?)\).player;", webpage, 'json')
-
-        data = json.loads(data_json)
+        data = self._parse_json(data_json, video_id)
 
         uploader = data['info_overlay'].get('username')
         duration = data['video']['duration'] // 1000

youtube_dl/extractor/mtv.py

@@ -2,10 +2,11 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
+from .subtitles import SubtitlesInfoExtractor
 from ..compat import (
     compat_urllib_parse,
     compat_urllib_request,
+    compat_str,
 )
 from ..utils import (
     ExtractorError,
@@ -22,7 +23,7 @@ def _media_xml_tag(tag):
     return '{http://search.yahoo.com/mrss/}%s' % tag
 
 
-class MTVServicesInfoExtractor(InfoExtractor):
+class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
     _MOBILE_TEMPLATE = None
 
     @staticmethod
@@ -53,7 +54,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         webpage_url = self._MOBILE_TEMPLATE % mtvn_id
         req = compat_urllib_request.Request(webpage_url)
         # Otherwise we get a webpage that would execute some javascript
-        req.add_header('Youtubedl-user-agent', 'curl/7')
+        req.add_header('User-Agent', 'curl/7')
         webpage = self._download_webpage(req, mtvn_id,
                                          'Downloading mobile page')
         metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url'))
@@ -78,17 +79,42 @@ class MTVServicesInfoExtractor(InfoExtractor):
             try:
                 _, _, ext = rendition.attrib['type'].partition('/')
                 rtmp_video_url = rendition.find('./src').text
-                formats.append({'ext': ext,
-                                'url': self._transform_rtmp_url(rtmp_video_url),
-                                'format_id': rendition.get('bitrate'),
-                                'width': int(rendition.get('width')),
-                                'height': int(rendition.get('height')),
-                                })
+                if rtmp_video_url.endswith('siteunavail.png'):
+                    continue
+                formats.append({
+                    'ext': ext,
+                    'url': self._transform_rtmp_url(rtmp_video_url),
+                    'format_id': rendition.get('bitrate'),
+                    'width': int(rendition.get('width')),
+                    'height': int(rendition.get('height')),
+                })
             except (KeyError, TypeError):
                 raise ExtractorError('Invalid rendition field.')
         self._sort_formats(formats)
         return formats
 
+    def _extract_subtitles(self, mdoc, mtvn_id):
+        subtitles = {}
+        FORMATS = {
+            'scc': 'cea-608',
+            'eia-608': 'cea-608',
+            'xml': 'ttml',
+        }
+        subtitles_format = FORMATS.get(
+            self._downloader.params.get('subtitlesformat'), 'ttml')
+        for transcript in mdoc.findall('.//transcript'):
+            if transcript.get('kind') != 'captions':
+                continue
+            lang = transcript.get('srclang')
+            for typographic in transcript.findall('./typographic'):
+                captions_format = typographic.get('format')
+                if captions_format == subtitles_format:
+                    subtitles[lang] = compat_str(typographic.get('src'))
+                    break
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(mtvn_id, subtitles)
+        return self.extract_subtitles(mtvn_id, subtitles)
+
     def _get_video_info(self, itemdoc):
         uri = itemdoc.find('guid').text
         video_id = self._id_from_uri(uri)
@@ -135,6 +161,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         return {
             'title': title,
             'formats': self._extract_video_formats(mediagen_doc, mtvn_id),
+            'subtitles': self._extract_subtitles(mediagen_doc, mtvn_id),
             'id': video_id,
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
             'description': description,
@@ -167,7 +194,11 @@ class MTVServicesInfoExtractor(InfoExtractor):
             mgid = self._search_regex(
                 [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
                 webpage, 'mgid')
-        return self._get_videos_info(mgid)
+
+        videos_info = self._get_videos_info(mgid)
+        if self._downloader.params.get('listsubtitles', False):
+            return
+        return videos_info
 
 
 class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
@@ -212,25 +243,14 @@ class MTVIE(MTVServicesInfoExtractor):
     _TESTS = [
         {
             'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
-            'file': '853555.mp4',
             'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
             'info_dict': {
+                'id': '853555',
+                'ext': 'mp4',
                 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
                 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
             },
         },
-        {
-            'add_ie': ['Vevo'],
-            'url': 'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
-            'file': 'USCJY1331283.mp4',
-            'md5': '73b4e7fcadd88929292fe52c3ced8caf',
-            'info_dict': {
-                'title': 'Everything Has Changed',
-                'upload_date': '20130606',
-                'uploader': 'Taylor Swift',
-            },
-            'skip': 'VEVO is only available in some countries',
-        },
     ]
 
     def _get_thumbnail_url(self, uri, itemdoc):
@@ -244,8 +264,8 @@ class MTVIE(MTVServicesInfoExtractor):
             webpage = self._download_webpage(url, video_id)
 
             # Some videos come from Vevo.com
-            m_vevo = re.search(r'isVevoVideo = true;.*?vevoVideoId = "(.*?)";',
-                               webpage, re.DOTALL)
+            m_vevo = re.search(
+                r'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?)";', webpage)
             if m_vevo:
                 vevo_id = m_vevo.group(1)
                 self.to_screen('Vevo video detected: %s' % vevo_id)

youtube_dl/extractor/nerdcubed.py

@@ -11,6 +11,7 @@ class NerdCubedFeedIE(InfoExtractor):
     _TEST = {
         'url': 'http://www.nerdcubed.co.uk/feed.json',
         'info_dict': {
+            'id': 'nerdcubed-feed',
             'title': 'nerdcubed.co.uk feed',
         },
         'playlist_mincount': 1300,

youtube_dl/extractor/nextmedia.py

@@ -0,0 +1,163 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import parse_iso8601
+
+
+class NextMediaIE(InfoExtractor):
+    _VALID_URL = r'http://hk.apple.nextmedia.com/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://hk.apple.nextmedia.com/realtime/news/20141108/53109199',
+        'md5': 'dff9fad7009311c421176d1ac90bfe4f',
+        'info_dict': {
+            'id': '53109199',
+            'ext': 'mp4',
+            'title': '【佔領金鐘】50外國領事議員撐場 讚學生勇敢香港有希望',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'description': 'md5:28222b9912b6665a21011b034c70fcc7',
+            'timestamp': 1415456273,
+            'upload_date': '20141108',
+        }
+    }]
+
+    _URL_PATTERN = r'\{ url: \'(.+)\' \}'
+
+    def _real_extract(self, url):
+        news_id = self._match_id(url)
+        page = self._download_webpage(url, news_id)
+        return self._extract_from_nextmedia_page(news_id, url, page)
+
+    def _extract_from_nextmedia_page(self, news_id, url, page):
+        title = self._fetch_title(page)
+        video_url = self._search_regex(self._URL_PATTERN, page, 'video url')
+
+        attrs = {
+            'id': news_id,
+            'title': title,
+            'url': video_url,  # ext can be inferred from url
+            'thumbnail': self._fetch_thumbnail(page),
+            'description': self._fetch_description(page),
+        }
+
+        timestamp = self._fetch_timestamp(page)
+        if timestamp:
+            attrs['timestamp'] = timestamp
+        else:
+            attrs['upload_date'] = self._fetch_upload_date(url)
+
+        return attrs
+
+    def _fetch_title(self, page):
+        return self._og_search_title(page)
+
+    def _fetch_thumbnail(self, page):
+        return self._og_search_thumbnail(page)
+
+    def _fetch_timestamp(self, page):
+        dateCreated = self._search_regex('"dateCreated":"([^"]+)"', page, 'created time')
+        return parse_iso8601(dateCreated)
+
+    def _fetch_upload_date(self, url):
+        return self._search_regex(self._VALID_URL, url, 'upload date', group='date')
+
+    def _fetch_description(self, page):
+        return self._og_search_property('description', page)
+
+
+class NextMediaActionNewsIE(NextMediaIE):
+    _VALID_URL = r'http://hk.dv.nextmedia.com/actionnews/[^/]+/(?P<date>\d+)/(?P<id>\d+)/\d+'
+    _TESTS = [{
+        'url': 'http://hk.dv.nextmedia.com/actionnews/hit/20150121/19009428/20061460',
+        'md5': '05fce8ffeed7a5e00665d4b7cf0f9201',
+        'info_dict': {
+            'id': '19009428',
+            'ext': 'mp4',
+            'title': '【壹週刊】細10年男友偷食　50歲邵美琪再失戀',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'description': 'md5:cd802fad1f40fd9ea178c1e2af02d659',
+            'timestamp': 1421791200,
+            'upload_date': '20150120',
+        }
+    }]
+
+    def _real_extract(self, url):
+        news_id = self._match_id(url)
+        actionnews_page = self._download_webpage(url, news_id)
+        article_url = self._og_search_url(actionnews_page)
+        article_page = self._download_webpage(article_url, news_id)
+        return self._extract_from_nextmedia_page(news_id, url, article_page)
+
+
+class AppleDailyRealtimeNewsIE(NextMediaIE):
+    _VALID_URL = r'http://(www|ent).appledaily.com.tw/(realtimenews|enews)/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)(/.*)?'
+    _TESTS = [{
+        'url': 'http://ent.appledaily.com.tw/enews/article/entertainment/20150128/36354694',
+        'md5': 'a843ab23d150977cc55ef94f1e2c1e4d',
+        'info_dict': {
+            'id': '36354694',
+            'ext': 'mp4',
+            'title': '周亭羽走過摩鐵陰霾2男陪吃 九把刀孤寒看醫生',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'description': 'md5:b23787119933404ce515c6356a8c355c',
+            'upload_date': '20150128',
+        }
+    }, {
+        'url': 'http://www.appledaily.com.tw/realtimenews/article/strange/20150128/550549/%E4%B8%8D%E6%BB%BF%E8%A2%AB%E8%B8%A9%E8%85%B3%E3%80%80%E5%B1%B1%E6%9D%B1%E5%85%A9%E5%A4%A7%E5%AA%BD%E4%B8%80%E8%B7%AF%E6%89%93%E4%B8%8B%E8%BB%8A',
+        'md5': '86b4e9132d158279c7883822d94ccc49',
+        'info_dict': {
+            'id': '550549',
+            'ext': 'mp4',
+            'title': '不滿被踩腳　山東兩大媽一路打下車',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'description': 'md5:2648aaf6fc4f401f6de35a91d111aa1d',
+            'upload_date': '20150128',
+        }
+    }]
+
+    _URL_PATTERN = r'\{url: \'(.+)\'\}'
+
+    def _fetch_title(self, page):
+        return self._html_search_regex(r'<h1 id="h1">([^<>]+)</h1>', page, 'news title')
+
+    def _fetch_thumbnail(self, page):
+        return self._html_search_regex(r"setInitialImage\(\'([^']+)'\)", page, 'video thumbnail', fatal=False)
+
+    def _fetch_timestamp(self, page):
+        return None
+
+
+class AppleDailyAnimationNewsIE(AppleDailyRealtimeNewsIE):
+    _VALID_URL = 'http://www.appledaily.com.tw/animation/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)(/.*)?'
+    _TESTS = [{
+        'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003671',
+        'md5': '03df296d95dedc2d5886debbb80cb43f',
+        'info_dict': {
+            'id': '5003671',
+            'ext': 'mp4',
+            'title': '20正妹熱舞　《刀龍傳說Online》火辣上市',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'description': 'md5:23c0aac567dc08c9c16a3161a2c2e3cd',
+            'upload_date': '20150128',
+        }
+    }, {
+        # No thumbnail
+        'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003673/',
+        'md5': 'b06182cd386ea7bc6115ec7ff0f72aeb',
+        'info_dict': {
+            'id': '5003673',
+            'ext': 'mp4',
+            'title': '半夜尿尿　好像會看到___',
+            'description': 'md5:61d2da7fe117fede148706cdb85ac066',
+            'upload_date': '20150128',
+        },
+        'expected_warnings': [
+            'video thumbnail',
+        ]
+    }]
+
+    def _fetch_title(self, page):
+        return self._html_search_meta('description', page, 'news title')
+
+    def _fetch_description(self, page):
+        return self._html_search_meta('description', page, 'news description')

youtube_dl/extractor/nhl.py

@@ -20,6 +20,12 @@ class NHLBaseInfoExtractor(InfoExtractor):
     def _fix_json(json_string):
         return json_string.replace('\\\'', '\'')
 
+    def _real_extract_video(self, video_id):
+        json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
+        data = self._download_json(
+            json_url, video_id, transform_source=self._fix_json)
+        return self._extract_video(data[0])
+
     def _extract_video(self, info):
         video_id = info['id']
         self.report_extraction(video_id)
@@ -54,7 +60,7 @@ class NHLBaseInfoExtractor(InfoExtractor):
 
 class NHLIE(NHLBaseInfoExtractor):
     IE_NAME = 'nhl.com'
-    _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/console(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)'
+    _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/(?:console)?(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)'
 
     _TESTS = [{
         'url': 'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614',
@@ -92,15 +98,41 @@ class NHLIE(NHLBaseInfoExtractor):
     }, {
         'url': 'http://video.flames.nhl.com/videocenter/console?id=630616',
         'only_matching': True,
+    }, {
+        'url': 'http://video.nhl.com/videocenter/?id=736722',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
-        data = self._download_json(
-            json_url, video_id, transform_source=self._fix_json)
-        return self._extract_video(data[0])
+        video_id = self._match_id(url)
+        return self._real_extract_video(video_id)
+
+
+class NHLNewsIE(NHLBaseInfoExtractor):
+    IE_NAME = 'nhl.com:news'
+    IE_DESC = 'NHL news'
+    _VALID_URL = r'https?://(?:www\.)?nhl\.com/ice/news\.html?(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)'
+
+    _TEST = {
+        'url': 'http://www.nhl.com/ice/news.htm?id=750727',
+        'md5': '4b3d1262e177687a3009937bd9ec0be8',
+        'info_dict': {
+            'id': '736722',
+            'ext': 'mp4',
+            'title': 'Cal Clutterbuck has been fined $2,000',
+            'description': 'md5:45fe547d30edab88b23e0dd0ab1ed9e6',
+            'duration': 37,
+            'upload_date': '20150128',
+        },
+    }
+
+    def _real_extract(self, url):
+        news_id = self._match_id(url)
+        webpage = self._download_webpage(url, news_id)
+        video_id = self._search_regex(
+            [r'pVid(\d+)', r"nlid\s*:\s*'(\d+)'"],
+            webpage, 'video id')
+        return self._real_extract_video(video_id)
 
 
 class NHLVideocenterIE(NHLBaseInfoExtractor):

youtube_dl/extractor/ringtv.py

@@ -6,12 +6,13 @@ from .common import InfoExtractor
 
 
 class RingTVIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?ringtv\.craveonline\.com/(?P<type>news|videos/video)/(?P<id>[^/?#]+)'
+    _VALID_URL = r'http://(?:www\.)?ringtv\.craveonline\.com/(?P<type>news|videos/video)/(?P<id>[^/?#]+)'
     _TEST = {
         "url": "http://ringtv.craveonline.com/news/310833-luis-collazo-says-victor-ortiz-better-not-quit-on-jan-30",
-        "file": "857645.mp4",
         "md5": "d25945f5df41cdca2d2587165ac28720",
         "info_dict": {
+            'id': '857645',
+            'ext': 'mp4',
             "title": 'Video: Luis Collazo says Victor Ortiz "better not quit on Jan. 30" - Ring TV',
             "description": 'Luis Collazo is excited about his Jan. 30 showdown with fellow former welterweight titleholder Victor Ortiz at Barclays Center in his hometown of Brooklyn. The SuperBowl week fight headlines a Golden Boy Live! card on Fox Sports 1.',
         }

youtube_dl/extractor/rottentomatoes.py

@@ -10,8 +10,9 @@ class RottenTomatoesIE(VideoDetectiveIE):
 
     _TEST = {
         'url': 'http://www.rottentomatoes.com/m/toy_story_3/trailers/11028566/',
-        'file': '613340.mp4',
         'info_dict': {
+            'id': '613340',
+            'ext': 'mp4',
             'title': 'TOY STORY 3',
             'description': 'From the creators of the beloved TOY STORY films, comes a story that will reunite the gang in a whole new way.',
         },

youtube_dl/extractor/rtl2.py

@@ -0,0 +1,72 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class RTL2IE(InfoExtractor):
+    _VALID_URL = r'http?://(?:www\.)?rtl2\.de/[^?#]*?/(?P<id>[^?#/]*?)(?:$|/(?:$|[?#]))'
+    _TESTS = [{
+        'url': 'http://www.rtl2.de/sendung/grip-das-motormagazin/folge/folge-203-0',
+        'md5': 'bfcc179030535b08dc2b36b469b5adc7',
+        'info_dict': {
+            'id': 'folge-203-0',
+            'ext': 'f4v',
+            'title': 'GRIP sucht den Sommerkönig',
+            'description': 'Matthias, Det und Helge treten gegeneinander an.'
+        },
+    }, {
+        'url': 'http://www.rtl2.de/sendung/koeln-50667/video/5512-anna/21040-anna-erwischt-alex/',
+        'md5': 'ffcd517d2805b57ce11a58a2980c2b02',
+        'info_dict': {
+            'id': '21040-anna-erwischt-alex',
+            'ext': 'mp4',
+            'title': 'Anna erwischt Alex!',
+            'description': 'Anna ist Alex\' Tochter bei Köln 50667.'
+        },
+    }]
+
+    def _real_extract(self, url):
+        # Some rtl2 urls have no slash at the end, so append it.
+        if not url.endswith('/'):
+            url += '/'
+
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        vico_id = self._html_search_regex(
+            r'vico_id\s*:\s*([0-9]+)', webpage, 'vico_id')
+        vivi_id = self._html_search_regex(
+            r'vivi_id\s*:\s*([0-9]+)', webpage, 'vivi_id')
+        info_url = 'http://www.rtl2.de/video/php/get_video.php?vico_id=' + vico_id + '&vivi_id=' + vivi_id
+        webpage = self._download_webpage(info_url, '')
+
+        info = self._download_json(info_url, video_id)
+        video_info = info['video']
+        title = video_info['titel']
+        description = video_info.get('beschreibung')
+        thumbnail = video_info.get('image')
+
+        download_url = video_info['streamurl']
+        download_url = download_url.replace('\\', '')
+        stream_url = 'mp4:' + self._html_search_regex(r'ondemand/(.*)', download_url, 'stream URL')
+        rtmp_conn = ["S:connect", "O:1", "NS:pageUrl:" + url, "NB:fpad:0", "NN:videoFunction:1", "O:0"]
+
+        formats = [{
+            'url': download_url,
+            'play_path': stream_url,
+            'player_url': 'http://www.rtl2.de/flashplayer/vipo_player.swf',
+            'page_url': url,
+            'flash_version': 'LNX 11,2,202,429',
+            'rtmp_conn': rtmp_conn,
+            'no_resume': True,
+        }]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'description': description,
+            'formats': formats,
+        }

youtube_dl/extractor/rtve.py

@@ -57,7 +57,7 @@ def _decrypt_url(png):
 class RTVEALaCartaIE(InfoExtractor):
     IE_NAME = 'rtve.es:alacarta'
     IE_DESC = 'RTVE a la carta'
-    _VALID_URL = r'http://www\.rtve\.es/alacarta/videos/[^/]+/[^/]+/(?P<id>\d+)'
+    _VALID_URL = r'http://www\.rtve\.es/(m/)?alacarta/videos/[^/]+/[^/]+/(?P<id>\d+)'
 
     _TESTS = [{
         'url': 'http://www.rtve.es/alacarta/videos/balonmano/o-swiss-cup-masculina-final-espana-suecia/2491869/',
@@ -74,7 +74,11 @@ class RTVEALaCartaIE(InfoExtractor):
             'id': '1694255',
             'ext': 'flv',
             'title': 'TODO',
-        }
+        },
+        'skip': 'The f4m manifest can\'t be used yet',
+    }, {
+        'url': 'http://www.rtve.es/m/alacarta/videos/cuentame-como-paso/cuentame-como-paso-t16-ultimo-minuto-nuestra-vida-capitulo-276/2969138/?media=tve',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -86,6 +90,18 @@ class RTVEALaCartaIE(InfoExtractor):
         png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/default/videos/%s.png' % video_id
         png = self._download_webpage(png_url, video_id, 'Downloading url information')
         video_url = _decrypt_url(png)
+        if not video_url.endswith('.f4m'):
+            auth_url = video_url.replace(
+                'resources/', 'auth/resources/'
+            ).replace('.net.rtve', '.multimedia.cdn.rtve')
+            video_path = self._download_webpage(
+                auth_url, video_id, 'Getting video url')
+            # Use mvod.akcdn instead of flash.akamaihd.multimedia.cdn to get
+            # the right Content-Length header and the mp4 format
+            video_url = (
+                'http://mvod.akcdn.rtve.es/{0}&v=2.6.8'
+                '&fp=MAC%2016,0,0,296&r=MRUGG&g=OEOJWFXNFGCP'.format(video_path)
+            )
 
         return {
             'id': video_id,

youtube_dl/extractor/rutv.py

@@ -162,10 +162,8 @@ class RUTVIE(InfoExtractor):
                         'vbr': int(quality),
                     }
                 elif transport == 'm3u8':
-                    fmt = {
-                        'url': url,
-                        'ext': 'mp4',
-                    }
+                    formats.extend(self._extract_m3u8_formats(url, video_id, 'mp4'))
+                    continue
                 else:
                     fmt = {
                         'url': url

youtube_dl/extractor/servingsys.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
@@ -13,10 +11,15 @@ class ServingSysIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://bs.serving-sys.com/BurstingPipe/adServer.bs?cn=is&c=23&pl=VAST&pli=5349193&PluID=0&pos=7135&ord=[timestamp]&cim=1?',
+        'info_dict': {
+            'id': '5349193',
+            'title': 'AdAPPter_Hyundai_demo',
+        },
         'playlist': [{
-            'file': '29955898.flv',
             'md5': 'baed851342df6846eb8677a60a011a0f',
             'info_dict': {
+                'id': '29955898',
+                'ext': 'flv',
                 'title': 'AdAPPter_Hyundai_demo (1)',
                 'duration': 74,
                 'tbr': 1378,
@@ -24,9 +27,10 @@ class ServingSysIE(InfoExtractor):
                 'height': 400,
             },
         }, {
-            'file': '29907998.flv',
             'md5': '979b4da2655c4bc2d81aeb915a8c5014',
             'info_dict': {
+                'id': '29907998',
+                'ext': 'flv',
                 'title': 'AdAPPter_Hyundai_demo (2)',
                 'duration': 34,
                 'width': 854,
@@ -37,14 +41,13 @@ class ServingSysIE(InfoExtractor):
         'params': {
             'playlistend': 2,
         },
-        'skip': 'Blocked in the US [sic]',
+        '_skip': 'Blocked in the US [sic]',
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        pl_id = mobj.group('id')
-
+        pl_id = self._match_id(url)
         vast_doc = self._download_xml(url, pl_id)
+
         title = vast_doc.find('.//AdTitle').text
         media = vast_doc.find('.//MediaFile').text
         info_url = self._search_regex(r'&adData=([^&]+)&', media, 'info URL')

youtube_dl/extractor/sina.py

@@ -11,7 +11,7 @@ from ..compat import (
 
 
 class SinaIE(InfoExtractor):
-    _VALID_URL = r'''https?://(.*?\.)?video\.sina\.com\.cn/
+    _VALID_URL = r'''(?x)https?://(.*?\.)?video\.sina\.com\.cn/
                         (
                             (.+?/(((?P<pseudo_id>\d+).html)|(.*?(\#|(vid=)|b/)(?P<id>\d+?)($|&|\-))))
                             |
@@ -23,9 +23,10 @@ class SinaIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'http://video.sina.com.cn/news/vlist/zt/chczlj2013/?opsubject_id=top12#110028898',
-            'file': '110028898.flv',
             'md5': 'd65dd22ddcf44e38ce2bf58a10c3e71f',
             'info_dict': {
+                'id': '110028898',
+                'ext': 'flv',
                 'title': '《中国新闻》 朝鲜要求巴拿马立即释放被扣船员',
             }
         },
@@ -39,10 +40,6 @@ class SinaIE(InfoExtractor):
         },
     ]
 
-    @classmethod
-    def suitable(cls, url):
-        return re.match(cls._VALID_URL, url, flags=re.VERBOSE) is not None
-
     def _extract_video(self, video_id):
         data = compat_urllib_parse.urlencode({'vid': video_id})
         url_doc = self._download_xml('http://v.iask.com/v_play.php?%s' % data,
@@ -59,7 +56,7 @@ class SinaIE(InfoExtractor):
                 }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
+        mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         if mobj.group('token') is not None:
             # The video id is in the redirected url

youtube_dl/extractor/smotri.py

@@ -102,12 +102,13 @@ class SmotriIE(InfoExtractor):
                 'uploader_id': 'mopeder',
                 'duration': 71,
                 'thumbnail': 'http://frame9.loadup.ru/d7/32/2888853.2.3.jpg',
+                'upload_date': '20150114',
             },
         },
         # swf player
         {
             'url': 'http://pics.smotri.com/scrubber_custom8.swf?file=v9188090500',
-            'md5': '4d47034979d9390d14acdf59c4935bc2',
+            'md5': '31099eeb4bc906712c5f40092045108d',
             'info_dict': {
                 'id': 'v9188090500',
                 'ext': 'mp4',
@@ -138,9 +139,6 @@ class SmotriIE(InfoExtractor):
     def _search_meta(self, name, html, display_name=None):
         if display_name is None:
             display_name = name
-        return self._html_search_regex(
-            r'<meta itemprop="%s" content="([^"]+)" />' % re.escape(name),
-            html, display_name, fatal=False)
         return self._html_search_meta(name, html, display_name)
 
     def _real_extract(self, url):

youtube_dl/extractor/soundcloud.py

@@ -246,6 +246,7 @@ class SoundcloudSetIE(SoundcloudIE):
     _TESTS = [{
         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep',
         'info_dict': {
+            'id': '2284613',
             'title': 'The Royal Concept EP',
         },
         'playlist_mincount': 6,
@@ -279,7 +280,7 @@ class SoundcloudSetIE(SoundcloudIE):
         return {
             '_type': 'playlist',
             'entries': [self._extract_info_dict(track, secret_token=token) for track in info['tracks']],
-            'id': info['id'],
+            'id': '%s' % info['id'],
             'title': info['title'],
         }
 

youtube_dl/extractor/spiegel.py

@@ -4,14 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urlparse,
-    compat_HTTPError,
-)
-from ..utils import (
-    HEADRequest,
-    ExtractorError,
-)
+from ..compat import compat_urlparse
 from .spiegeltv import SpiegeltvIE
 
 
@@ -72,16 +65,6 @@ class SpiegelIE(InfoExtractor):
             if n.tag.startswith('type') and n.tag != 'type6':
                 format_id = n.tag.rpartition('type')[2]
                 video_url = base_url + n.find('./filename').text
-                # Test video URLs beforehand as some of them are invalid
-                try:
-                    self._request_webpage(
-                        HEADRequest(video_url), video_id,
-                        'Checking %s video URL' % format_id)
-                except ExtractorError as e:
-                    if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
-                        self.report_warning(
-                            '%s video URL is invalid, skipping' % format_id, video_id)
-                        continue
                 formats.append({
                     'format_id': format_id,
                     'url': video_url,
@@ -94,6 +77,7 @@ class SpiegelIE(InfoExtractor):
                 })
         duration = float(idoc[0].findall('./duration')[0].text)
 
+        self._check_formats(formats, video_id)
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/spike.py

@@ -1,14 +1,12 @@
 from __future__ import unicode_literals
 
-import re
-
 from .mtv import MTVServicesInfoExtractor
 
 
 class SpikeIE(MTVServicesInfoExtractor):
     _VALID_URL = r'''(?x)https?://
-        (www\.spike\.com/(video-clips|episodes)/.+|
-         m\.spike\.com/videos/video.rbml\?id=(?P<mobile_id>[^&]+))
+        (?:www\.spike\.com/(?:video-clips|(?:full-)?episodes)/.+|
+         m\.spike\.com/videos/video\.rbml\?id=(?P<id>[^&]+))
         '''
     _TEST = {
         'url': 'http://www.spike.com/video-clips/lhtu8m/auction-hunters-can-allen-ride-a-hundred-year-old-motorcycle',
@@ -25,8 +23,7 @@ class SpikeIE(MTVServicesInfoExtractor):
     _MOBILE_TEMPLATE = 'http://m.spike.com/videos/video.rbml?id=%s'
 
     def _real_extract(self, url):
-        mobj = re.search(self._VALID_URL, url)
-        mobile_id = mobj.group('mobile_id')
-        if mobile_id is not None:
+        mobile_id = self._match_id(url)
+        if mobile_id:
             url = 'http://www.spike.com/video-clips/%s' % mobile_id
         return super(SpikeIE, self)._real_extract(url)

youtube_dl/extractor/srmediathek.py

@@ -8,7 +8,7 @@ from ..utils import js_to_json
 
 
 class SRMediathekIE(InfoExtractor):
-    IE_DESC = 'Süddeutscher Rundfunk'
+    IE_DESC = 'Saarländischer Rundfunk'
     _VALID_URL = r'https?://sr-mediathek\.sr-online\.de/index\.php\?.*?&id=(?P<id>[0-9]+)'
 
     _TEST = {

youtube_dl/extractor/teamcoco.py

@@ -10,17 +10,19 @@ class TeamcocoIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
-            'file': '80187.mp4',
             'md5': '3f7746aa0dc86de18df7539903d399ea',
             'info_dict': {
+                'id': '80187',
+                'ext': 'mp4',
                 'title': 'Conan Becomes A Mary Kay Beauty Consultant',
                 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.'
             }
         }, {
             'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
-            'file': '19705.mp4',
             'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
             'info_dict': {
+                'id': '19705',
+                'ext': 'mp4',
                 "description": "Louis C.K. got starstruck by George W. Bush, so what? Part one.",
                 "title": "Louis C.K. Interview Pt. 1 11/3/11"
             }
@@ -36,7 +38,7 @@ class TeamcocoIE(InfoExtractor):
         video_id = mobj.group("video_id")
         if not video_id:
             video_id = self._html_search_regex(
-                r'data-node-id="(\d+?)"',
+                r'<div\s+class="player".*?data-id="(\d+?)"',
                 webpage, 'video id')
 
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id

youtube_dl/extractor/teletask.py

@@ -11,6 +11,7 @@ class TeleTaskIE(InfoExtractor):
     _TEST = {
         'url': 'http://www.tele-task.de/archive/video/html5/26168/',
         'info_dict': {
+            'id': '26168',
             'title': 'Duplicate Detection',
         },
         'playlist': [{
@@ -34,7 +35,6 @@ class TeleTaskIE(InfoExtractor):
 
     def _real_extract(self, url):
         lecture_id = self._match_id(url)
-
         webpage = self._download_webpage(url, lecture_id)
 
         title = self._html_search_regex(

youtube_dl/extractor/testtube.py

@@ -1,7 +1,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    int_or_none,
+    qualities,
+)
 
 
 class TestTubeIE(InfoExtractor):
@@ -46,13 +49,22 @@ class TestTubeIE(InfoExtractor):
         self._sort_formats(formats)
 
         duration = int_or_none(info.get('duration'))
+        images = info.get('images')
+        thumbnails = None
+        preference = qualities(['mini', 'small', 'medium', 'large'])
+        if images:
+            thumbnails = [{
+                'id': thumbnail_id,
+                'url': img_url,
+                'preference': preference(thumbnail_id)
+            } for thumbnail_id, img_url in images.items()]
 
         return {
             'id': video_id,
             'display_id': display_id,
             'title': info['title'],
             'description': info.get('summary'),
-            'thumbnail': info.get('images', {}).get('large'),
+            'thumbnails': thumbnails,
             'uploader': info.get('show', {}).get('name'),
             'uploader_id': info.get('show', {}).get('slug'),
             'duration': duration,

youtube_dl/extractor/toutv.py

@@ -16,8 +16,9 @@ class TouTvIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.tou.tv/30-vies/S04E41',
-        'file': '30-vies_S04E41.mp4',
         'info_dict': {
+            'id': '30-vies_S04E41',
+            'ext': 'mp4',
             'title': '30 vies Saison 4 / Épisode 41',
             'description': 'md5:da363002db82ccbe4dafeb9cab039b09',
             'age_limit': 8,

youtube_dl/extractor/twitch.py

@@ -220,12 +220,18 @@ class TwitchPlaylistBaseIE(TwitchBaseIE):
             response = self._download_json(
                 self._PLAYLIST_URL % (channel_id, offset, limit),
                 channel_id, 'Downloading %s videos JSON page %d' % (self._PLAYLIST_TYPE, counter))
-            videos = response['videos']
-            if not videos:
+            page_entries = self._extract_playlist_page(response)
+            if not page_entries:
                 break
-            entries.extend([self.url_result(video['url']) for video in videos])
+            entries.extend(page_entries)
             offset += limit
-        return self.playlist_result(entries, channel_id, channel_name)
+        return self.playlist_result(
+            [self.url_result(entry) for entry in set(entries)],
+            channel_id, channel_name)
+
+    def _extract_playlist_page(self, response):
+        videos = response.get('videos')
+        return [video['url'] for video in videos] if videos else []
 
     def _real_extract(self, url):
         return self._extract_playlist(self._match_id(url))
@@ -262,6 +268,31 @@ class TwitchPastBroadcastsIE(TwitchPlaylistBaseIE):
     }
 
 
+class TwitchBookmarksIE(TwitchPlaylistBaseIE):
+    IE_NAME = 'twitch:bookmarks'
+    _VALID_URL = r'%s/(?P<id>[^/]+)/profile/bookmarks/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
+    _PLAYLIST_URL = '%s/api/bookmark/?user=%%s&offset=%%d&limit=%%d' % TwitchBaseIE._API_BASE
+    _PLAYLIST_TYPE = 'bookmarks'
+
+    _TEST = {
+        'url': 'http://www.twitch.tv/ognos/profile/bookmarks',
+        'info_dict': {
+            'id': 'ognos',
+            'title': 'Ognos',
+        },
+        'playlist_mincount': 3,
+    }
+
+    def _extract_playlist_page(self, response):
+        entries = []
+        for bookmark in response.get('bookmarks', []):
+            video = bookmark.get('video')
+            if not video:
+                continue
+            entries.append(video['url'])
+        return entries
+
+
 class TwitchStreamIE(TwitchBaseIE):
     IE_NAME = 'twitch:stream'
     _VALID_URL = r'%s/(?P<id>[^/]+)/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE

youtube_dl/extractor/ubu.py

@@ -3,50 +3,51 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    int_or_none,
+    qualities,
+)
 
 
 class UbuIE(InfoExtractor):
     _VALID_URL = r'http://(?:www\.)?ubu\.com/film/(?P<id>[\da-z_-]+)\.html'
     _TEST = {
         'url': 'http://ubu.com/film/her_noise.html',
-        'md5': '8edd46ee8aa6b265fb5ed6cf05c36bc9',
+        'md5': '138d5652618bf0f03878978db9bef1ee',
         'info_dict': {
             'id': 'her_noise',
-            'ext': 'mp4',
+            'ext': 'm4v',
             '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')
-
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
         title = self._html_search_regex(
             r'<title>.+?Film &amp; Video: ([^<]+)</title>', webpage, 'title')
 
         duration = int_or_none(self._html_search_regex(
-            r'Duration: (\d+) minutes', webpage, 'duration', fatal=False, default=None))
-        if duration:
-            duration *= 60
+            r'Duration: (\d+) minutes', webpage, 'duration', fatal=False),
+            invscale=60)
 
         formats = []
-
         FORMAT_REGEXES = [
-            ['sq', r"'flashvars'\s*,\s*'file=([^']+)'"],
-            ['hq', r'href="(http://ubumexico\.centro\.org\.mx/video/[^"]+)"']
+            ('sq', r"'flashvars'\s*,\s*'file=([^']+)'"),
+            ('hq', r'href="(http://ubumexico\.centro\.org\.mx/video/[^"]+)"'),
         ]
-
+        preference = qualities([fid for fid, _ in FORMAT_REGEXES])
         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,
+                    'preference': preference(format_id),
                 })
+        self._sort_formats(formats)
 
         return {
             'id': video_id,

youtube_dl/extractor/viddler.py

@@ -5,27 +5,58 @@ from ..utils import (
     float_or_none,
     int_or_none,
 )
+from ..compat import (
+    compat_urllib_request
+)
 
 
 class ViddlerIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?viddler\.com/(?:v|embed|player)/(?P<id>[a-z0-9]+)'
-    _TEST = {
-        "url": "http://www.viddler.com/v/43903784",
+    _TESTS = [{
+        'url': 'http://www.viddler.com/v/43903784',
         'md5': 'ae43ad7cb59431ce043f0ff7fa13cbf4',
         'info_dict': {
             'id': '43903784',
             'ext': 'mp4',
-            "title": "Video Made Easy",
-            'description': 'You don\'t need to be a professional to make high-quality video content. Viddler provides some quick and easy tips on how to produce great video content with limited resources. ',
-            "uploader": "viddler",
+            'title': 'Video Made Easy',
+            'description': 'md5:6a697ebd844ff3093bd2e82c37b409cd',
+            'uploader': 'viddler',
             'timestamp': 1335371429,
             'upload_date': '20120425',
-            "duration": 100.89,
+            'duration': 100.89,
             'thumbnail': 're:^https?://.*\.jpg$',
             'view_count': int,
+            'comment_count': int,
             'categories': ['video content', 'high quality video', 'video made easy', 'how to produce video with limited resources', 'viddler'],
         }
-    }
+    }, {
+        'url': 'http://www.viddler.com/v/4d03aad9/',
+        'md5': 'faa71fbf70c0bee7ab93076fd007f4b0',
+        'info_dict': {
+            'id': '4d03aad9',
+            'ext': 'mp4',
+            'title': 'WALL-TO-GORTAT',
+            'upload_date': '20150126',
+            'uploader': 'deadspin',
+            'timestamp': 1422285291,
+            'view_count': int,
+            'comment_count': int,
+        }
+    }, {
+        'url': 'http://www.viddler.com/player/221ebbbd/0/',
+        'md5': '0defa2bd0ea613d14a6e9bd1db6be326',
+        'info_dict': {
+            'id': '221ebbbd',
+            'ext': 'mp4',
+            'title': 'LETeens-Grammar-snack-third-conditional',
+            'description': ' ',
+            'upload_date': '20140929',
+            'uploader': 'BCLETeens',
+            'timestamp': 1411997190,
+            'view_count': int,
+            'comment_count': int,
+        }
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -33,14 +64,17 @@ class ViddlerIE(InfoExtractor):
         json_url = (
             'http://api.viddler.com/api/v2/viddler.videos.getPlaybackDetails.json?video_id=%s&key=v0vhrt7bg2xq1vyxhkct' %
             video_id)
-        data = self._download_json(json_url, video_id)['video']
+        headers = {'Referer': 'http://static.cdn-ec.viddler.com/js/arpeggio/v2/embed.html'}
+        request = compat_urllib_request.Request(json_url, None, headers)
+        data = self._download_json(request, video_id)['video']
 
         formats = []
         for filed in data['files']:
             if filed.get('status', 'ready') != 'ready':
                 continue
+            format_id = filed.get('profile_id') or filed['profile_name']
             f = {
-                'format_id': filed['profile_id'],
+                'format_id': format_id,
                 'format_note': filed['profile_name'],
                 'url': self._proto_relative_url(filed['url']),
                 'width': int_or_none(filed.get('width')),
@@ -53,16 +87,15 @@ class ViddlerIE(InfoExtractor):
 
             if filed.get('cdn_url'):
                 f = f.copy()
-                f['url'] = self._proto_relative_url(filed['cdn_url'])
-                f['format_id'] = filed['profile_id'] + '-cdn'
+                f['url'] = self._proto_relative_url(filed['cdn_url'], 'http:')
+                f['format_id'] = format_id + '-cdn'
                 f['source_preference'] = 1
                 formats.append(f)
 
             if filed.get('html5_video_source'):
                 f = f.copy()
-                f['url'] = self._proto_relative_url(
-                    filed['html5_video_source'])
-                f['format_id'] = filed['profile_id'] + '-html5'
+                f['url'] = self._proto_relative_url(filed['html5_video_source'])
+                f['format_id'] = format_id + '-html5'
                 f['source_preference'] = 0
                 formats.append(f)
         self._sort_formats(formats)
@@ -71,7 +104,6 @@ class ViddlerIE(InfoExtractor):
             t.get('text') for t in data.get('tags', []) if 'text' in t]
 
         return {
-            '_type': 'video',
             'id': video_id,
             'title': data['title'],
             'formats': formats,
@@ -81,5 +113,6 @@ class ViddlerIE(InfoExtractor):
             'uploader': data.get('author'),
             'duration': float_or_none(data.get('length')),
             'view_count': int_or_none(data.get('view_count')),
+            'comment_count': int_or_none(data.get('comment_count')),
             'categories': categories,
         }

youtube_dl/extractor/videomega.py

@@ -62,5 +62,7 @@ class VideoMegaIE(InfoExtractor):
             'title': title,
             'formats': formats,
             'thumbnail': thumbnail,
-            'http_referer': iframe_url,
+            'http_headers': {
+                'Referer': iframe_url,
+            },
         }

youtube_dl/extractor/videott.py

@@ -13,9 +13,9 @@ from ..utils import (
 class VideoTtIE(InfoExtractor):
     ID_NAME = 'video.tt'
     IE_DESC = 'video.tt - Your True Tube'
-    _VALID_URL = r'http://(?:www\.)?video\.tt/(?:video/|watch_video\.php\?v=)(?P<id>[\da-zA-Z]{9})'
+    _VALID_URL = r'http://(?:www\.)?video\.tt/(?:(?:video|embed)/|watch_video\.php\?v=)(?P<id>[\da-zA-Z]{9})'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.video.tt/watch_video.php?v=amd5YujV8',
         'md5': 'b13aa9e2f267effb5d1094443dff65ba',
         'info_dict': {
@@ -26,7 +26,10 @@ class VideoTtIE(InfoExtractor):
             'upload_date': '20130827',
             'uploader': 'joseph313',
         }
-    }
+    }, {
+        'url': 'http://video.tt/embed/amd5YujV8',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/vimeo.py

@@ -501,9 +501,10 @@ class VimeoReviewIE(InfoExtractor):
     _VALID_URL = r'https?://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
     _TESTS = [{
         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
-        'file': '75524534.mp4',
         'md5': 'c507a72f780cacc12b2248bb4006d253',
         'info_dict': {
+            'id': '75524534',
+            'ext': 'mp4',
             'title': "DICK HARDWICK 'Comedian'",
             'uploader': 'Richard Hardwick',
         }

youtube_dl/extractor/washingtonpost.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -11,9 +12,10 @@ from ..utils import (
 
 class WashingtonPostIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?washingtonpost\.com/.*?/(?P<id>[^/]+)/(?:$|[?#])'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.washingtonpost.com/sf/national/2014/03/22/sinkhole-of-bureaucracy/',
         'info_dict': {
+            'id': 'sinkhole-of-bureaucracy',
             'title': 'Sinkhole of bureaucracy',
         },
         'playlist': [{
@@ -40,15 +42,38 @@ class WashingtonPostIE(InfoExtractor):
                 'upload_date': '20140322',
                 'uploader': 'The Washington Post',
             },
+        }],
+    }, {
+        'url': 'http://www.washingtonpost.com/blogs/wonkblog/wp/2014/12/31/one-airline-figured-out-how-to-make-sure-its-airplanes-never-disappear/',
+        'info_dict': {
+            'id': 'one-airline-figured-out-how-to-make-sure-its-airplanes-never-disappear',
+            'title': 'One airline figured out how to make sure its airplanes never disappear',
+        },
+        'playlist': [{
+            'md5': 'a7c1b5634ba5e57a6a82cdffa5b1e0d0',
+            'info_dict': {
+                'id': '0e4bb54c-9065-11e4-a66f-0ca5037a597d',
+                'ext': 'mp4',
+                'description': 'Washington Post transportation reporter Ashley Halsey III explains why a plane\'s black box needs to be recovered from a crash site instead of having its information streamed in real time throughout the flight.',
+                'upload_date': '20141230',
+                'uploader': 'The Washington Post',
+                'timestamp': 1419974765,
+                'title': 'Why black boxes don’t transmit data in real time',
+            }
         }]
-    }
+    }]
 
     def _real_extract(self, url):
         page_id = self._match_id(url)
         webpage = self._download_webpage(url, page_id)
 
         title = self._og_search_title(webpage)
-        uuids = re.findall(r'data-video-uuid="([^"]+)"', webpage)
+
+        uuids = re.findall(r'''(?x)
+            (?:
+                <div\s+class="posttv-video-embed[^>]*?data-uuid=|
+                data-video-uuid=
+            )"([^"]+)"''', webpage)
         entries = []
         for i, uuid in enumerate(uuids, start=1):
             vinfo_all = self._download_json(
@@ -75,10 +100,11 @@ class WashingtonPostIE(InfoExtractor):
                 'filesize': s.get('fileSize'),
                 'url': s.get('url'),
                 'ext': 'mp4',
+                'preference': -100 if s.get('type') == 'smil' else None,
                 'protocol': {
                     'MP4': 'http',
                     'F4F': 'f4m',
-                }.get(s.get('type'))
+                }.get(s.get('type')),
             } for s in vinfo.get('streams', [])]
             source_media_url = vinfo.get('sourceMediaURL')
             if source_media_url:

youtube_dl/extractor/wdr.py

@@ -71,6 +71,9 @@ class WDRIE(InfoExtractor):
         {
             'url': 'http://www1.wdr.de/mediathek/video/sendungen/quarks_und_co/filterseite-quarks-und-co100.html',
             'playlist_mincount': 146,
+            'info_dict': {
+                'id': 'mediathek/video/sendungen/quarks_und_co/filterseite-quarks-und-co100',
+            }
         }
     ]
 
@@ -169,7 +172,9 @@ class WDRMobileIE(InfoExtractor):
             'title': mobj.group('title'),
             'age_limit': int(mobj.group('age_limit')),
             'url': url,
-            'user_agent': 'mobile',
+            'http_headers': {
+                'User-Agent': 'mobile',
+            },
         }
 
 

youtube_dl/extractor/xuite.py

@@ -0,0 +1,142 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+
+import base64
+
+from .common import InfoExtractor
+from ..compat import compat_urllib_parse_unquote
+from ..utils import (
+    ExtractorError,
+    parse_iso8601,
+    parse_duration,
+)
+
+
+class XuiteIE(InfoExtractor):
+    _REGEX_BASE64 = r'(?:[A-Za-z0-9+/]{4})*(?:[A-Za-z0-9+/]{2}==|[A-Za-z0-9+/]{3}=)?'
+    _VALID_URL = r'https?://vlog\.xuite\.net/(?:play|embed)/(?P<id>%s)' % _REGEX_BASE64
+    _TESTS = [{
+        # Audio
+        'url': 'http://vlog.xuite.net/play/RGkzc1ZULTM4NjA5MTQuZmx2',
+        'md5': '63a42c705772aa53fd4c1a0027f86adf',
+        'info_dict': {
+            'id': '3860914',
+            'ext': 'mp3',
+            'title': '孤單南半球-歐德陽',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 247.246,
+            'timestamp': 1314932940,
+            'upload_date': '20110902',
+            'uploader': '阿能',
+            'uploader_id': '15973816',
+            'categories': ['個人短片'],
+        },
+    }, {
+        # Video with only one format
+        'url': 'http://vlog.xuite.net/play/TkRZNjhULTM0NDE2MjkuZmx2',
+        'md5': 'c45737fc8ac5dc8ac2f92ecbcecf505e',
+        'info_dict': {
+            'id': '3441629',
+            'ext': 'mp4',
+            'title': '孫燕姿 - 眼淚成詩',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 217.399,
+            'timestamp': 1299383640,
+            'upload_date': '20110306',
+            'uploader': 'Valen',
+            'uploader_id': '10400126',
+            'categories': ['影視娛樂'],
+        },
+    }, {
+        # Video with two formats
+        'url': 'http://vlog.xuite.net/play/bWo1N1pLLTIxMzAxMTcwLmZsdg==',
+        'md5': '1166e0f461efe55b62e26a2d2a68e6de',
+        'info_dict': {
+            'id': '21301170',
+            'ext': 'mp4',
+            'title': '暗殺教室 02',
+            'description': '字幕:【極影字幕社】',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 1384.907,
+            'timestamp': 1421481240,
+            'upload_date': '20150117',
+            'uploader': '我只是想認真點',
+            'uploader_id': '242127761',
+            'categories': ['電玩動漫'],
+        },
+    }, {
+        'url': 'http://vlog.xuite.net/play/S1dDUjdyLTMyOTc3NjcuZmx2/%E5%AD%AB%E7%87%95%E5%A7%BF-%E7%9C%BC%E6%B7%9A%E6%88%90%E8%A9%A9',
+        'only_matching': True,
+    }]
+
+    def _extract_flv_config(self, media_id):
+        base64_media_id = base64.b64encode(media_id.encode('utf-8')).decode('utf-8')
+        flv_config = self._download_xml(
+            'http://vlog.xuite.net/flash/player?media=%s' % base64_media_id,
+            'flv config')
+        prop_dict = {}
+        for prop in flv_config.findall('./property'):
+            prop_id = base64.b64decode(prop.attrib['id']).decode('utf-8')
+            # CDATA may be empty in flv config
+            if not prop.text:
+                continue
+            encoded_content = base64.b64decode(prop.text).decode('utf-8')
+            prop_dict[prop_id] = compat_urllib_parse_unquote(encoded_content)
+        return prop_dict
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        error_msg = self._search_regex(
+            r'<div id="error-message-content">([^<]+)',
+            webpage, 'error message', default=None)
+        if error_msg:
+            raise ExtractorError(
+                '%s returned error: %s' % (self.IE_NAME, error_msg),
+                expected=True)
+
+        video_id = self._html_search_regex(
+            r'data-mediaid="(\d+)"', webpage, 'media id')
+        flv_config = self._extract_flv_config(video_id)
+
+        FORMATS = {
+            'audio': 'mp3',
+            'video': 'mp4',
+        }
+
+        formats = []
+        for format_tag in ('src', 'hq_src'):
+            video_url = flv_config.get(format_tag)
+            if not video_url:
+                continue
+            format_id = self._search_regex(
+                r'\bq=(.+?)\b', video_url, 'format id', default=format_tag)
+            formats.append({
+                'url': video_url,
+                'ext': FORMATS.get(flv_config['type'], 'mp4'),
+                'format_id': format_id,
+                'height': int(format_id) if format_id.isnumeric() else None,
+            })
+        self._sort_formats(formats)
+
+        timestamp = flv_config.get('publish_datetime')
+        if timestamp:
+            timestamp = parse_iso8601(timestamp + ' +0800', ' ')
+
+        category = flv_config.get('category')
+        categories = [category] if category else []
+
+        return {
+            'id': video_id,
+            'title': flv_config['title'],
+            'description': flv_config.get('description'),
+            'thumbnail': flv_config.get('thumb'),
+            'timestamp': timestamp,
+            'uploader': flv_config.get('author_name'),
+            'uploader_id': flv_config.get('author_id'),
+            'duration': parse_duration(flv_config.get('duration')),
+            'categories': categories,
+            'formats': formats,
+        }

youtube_dl/extractor/youtube.py

@@ -264,9 +264,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
 
         # Dash mp4 audio
-        '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50},
-        '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50},
-        '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50},
+        '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
+        '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
+        '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
 
         # Dash webm
         '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
@@ -809,6 +809,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             player_url = None
 
         # Get video info
+        embed_webpage = None
         if re.search(r'player-age-gate-content">', video_webpage) is not None:
             age_gate = True
             # We simulate the access to the video from www.youtube.com/v/{video_id}
@@ -1016,10 +1017,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                     url += '&signature=' + url_data['sig'][0]
                 elif 's' in url_data:
                     encrypted_sig = url_data['s'][0]
+                    ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
 
                     jsplayer_url_json = self._search_regex(
-                        r'"assets":.+?"js":\s*("[^"]+")',
-                        embed_webpage if age_gate else video_webpage, 'JS player URL')
+                        ASSETS_RE,
+                        embed_webpage if age_gate else video_webpage,
+                        'JS player URL (1)', default=None)
+                    if not jsplayer_url_json and not age_gate:
+                        # We need the embed website after all
+                        if embed_webpage is None:
+                            embed_url = proto + '://www.youtube.com/embed/%s' % video_id
+                            embed_webpage = self._download_webpage(
+                                embed_url, video_id, 'Downloading embed webpage')
+                        jsplayer_url_json = self._search_regex(
+                            ASSETS_RE, embed_webpage, 'JS player URL')
+
                     player_url = json.loads(jsplayer_url_json)
                     if player_url is None:
                         player_url_json = self._search_regex(
@@ -1148,6 +1160,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
     }, {
         'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
         'info_dict': {
+            'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
             'title': 'YDL_Empty_List',
         },
         'playlist_count': 0,
@@ -1156,6 +1169,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
         'info_dict': {
             'title': '29C3: Not my department',
+            'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
         },
         'playlist_count': 95,
     }, {
@@ -1163,6 +1177,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'url': 'PLBB231211A4F62143',
         'info_dict': {
             'title': '[OLD]Team Fortress 2 (Class-based LP)',
+            'id': 'PLBB231211A4F62143',
         },
         'playlist_mincount': 26,
     }, {
@@ -1170,12 +1185,14 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
         'info_dict': {
             'title': 'Uploads from Cauchemar',
+            'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
         },
         'playlist_mincount': 799,
     }, {
         'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
         'info_dict': {
             'title': 'YDL_safe_search',
+            'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
         },
         'playlist_count': 2,
     }, {
@@ -1184,6 +1201,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'playlist_count': 4,
         'info_dict': {
             'title': 'JODA15',
+            'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
         }
     }, {
         'note': 'Embedded SWF player',
@@ -1191,12 +1209,14 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'playlist_count': 4,
         'info_dict': {
             'title': 'JODA7',
+            'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
         }
     }, {
         'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
         'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
         'info_dict': {
-                'title': 'Uploads from Interstellar Movie',
+            'title': 'Uploads from Interstellar Movie',
+            'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
         },
         'playlist_mincout': 21,
     }]
@@ -1302,6 +1322,9 @@ class YoutubeChannelIE(InfoExtractor):
         'note': 'paginated channel',
         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
         'playlist_mincount': 91,
+        'info_dict': {
+            'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+        }
     }]
 
     def extract_videos_from_page(self, page):
@@ -1682,11 +1705,18 @@ class YoutubeTruncatedURLIE(InfoExtractor):
     IE_NAME = 'youtube:truncated_url'
     IE_DESC = False  # Do not list
     _VALID_URL = r'''(?x)
-        (?:https?://)?[^/]+/watch\?(?:
+        (?:https?://)?
+        (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
+        (?:watch\?(?:
             feature=[a-z_]+|
-            annotation_id=annotation_[^&]+
-        )?$|
-        (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
+            annotation_id=annotation_[^&]+|
+            x-yt-cl=[0-9]+|
+            hl=[^&]*|
+        )?
+        |
+            attribution_link\?a=[^&]+
+        )
+        $
     '''
 
     _TESTS = [{
@@ -1695,6 +1725,15 @@ class YoutubeTruncatedURLIE(InfoExtractor):
     }, {
         'url': 'http://www.youtube.com/watch?',
         'only_matching': True,
+    }, {
+        'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.youtube.com/watch?feature=foo',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.youtube.com/watch?hl=en-GB',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -1710,7 +1749,7 @@ class YoutubeTruncatedURLIE(InfoExtractor):
 class YoutubeTruncatedIDIE(InfoExtractor):
     IE_NAME = 'youtube:truncated_id'
     IE_DESC = False  # Do not list
-    _VALID_URL = r'https?://(?:www\.)youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
+    _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
 
     _TESTS = [{
         'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',

youtube_dl/jsinterp.py

@@ -1,59 +1,122 @@
 from __future__ import unicode_literals
 
 import json
+import operator
 import re
 
 from .utils import (
     ExtractorError,
 )
 
+_OPERATORS = [
+    ('|', operator.or_),
+    ('^', operator.xor),
+    ('&', operator.and_),
+    ('>>', operator.rshift),
+    ('<<', operator.lshift),
+    ('-', operator.sub),
+    ('+', operator.add),
+    ('%', operator.mod),
+    ('/', operator.truediv),
+    ('*', operator.mul),
+]
+_ASSIGN_OPERATORS = [(op + '=', opfunc) for op, opfunc in _OPERATORS]
+_ASSIGN_OPERATORS.append(('=', lambda cur, right: right))
+
+_NAME_RE = r'[a-zA-Z_$][a-zA-Z_$0-9]*'
+
 
 class JSInterpreter(object):
-    def __init__(self, code):
-        self.code = code
+    def __init__(self, code, objects=None):
+        if objects is None:
+            objects = {}
+        self.code = self._remove_comments(code)
         self._functions = {}
-        self._objects = {}
+        self._objects = objects
 
-    def interpret_statement(self, stmt, local_vars, allow_recursion=20):
+    def _remove_comments(self, code):
+        return re.sub(r'(?s)/\*.*?\*/', '', code)
+
+    def interpret_statement(self, stmt, local_vars, allow_recursion=100):
         if allow_recursion < 0:
             raise ExtractorError('Recursion limit reached')
 
-        if stmt.startswith('var '):
-            stmt = stmt[len('var '):]
-        ass_m = re.match(r'^(?P<out>[a-z]+)(?:\[(?P<index>[^\]]+)\])?' +
-                         r'=(?P<expr>.*)$', stmt)
-        if ass_m:
-            if ass_m.groupdict().get('index'):
-                def assign(val):
-                    lvar = local_vars[ass_m.group('out')]
-                    idx = self.interpret_expression(
-                        ass_m.group('index'), local_vars, allow_recursion)
-                    assert isinstance(idx, int)
-                    lvar[idx] = val
-                    return val
-                expr = ass_m.group('expr')
-            else:
-                def assign(val):
-                    local_vars[ass_m.group('out')] = val
-                    return val
-                expr = ass_m.group('expr')
-        elif stmt.startswith('return '):
-            assign = lambda v: v
-            expr = stmt[len('return '):]
+        should_abort = False
+        stmt = stmt.lstrip()
+        stmt_m = re.match(r'var\s', stmt)
+        if stmt_m:
+            expr = stmt[len(stmt_m.group(0)):]
         else:
-            # Try interpreting it as an expression
-            expr = stmt
-            assign = lambda v: v
+            return_m = re.match(r'return(?:\s+|$)', stmt)
+            if return_m:
+                expr = stmt[len(return_m.group(0)):]
+                should_abort = True
+            else:
+                # Try interpreting it as an expression
+                expr = stmt
 
         v = self.interpret_expression(expr, local_vars, allow_recursion)
-        return assign(v)
+        return v, should_abort
 
     def interpret_expression(self, expr, local_vars, allow_recursion):
+        expr = expr.strip()
+
+        if expr == '':  # Empty expression
+            return None
+
+        if expr.startswith('('):
+            parens_count = 0
+            for m in re.finditer(r'[()]', expr):
+                if m.group(0) == '(':
+                    parens_count += 1
+                else:
+                    parens_count -= 1
+                    if parens_count == 0:
+                        sub_expr = expr[1:m.start()]
+                        sub_result = self.interpret_expression(
+                            sub_expr, local_vars, allow_recursion)
+                        remaining_expr = expr[m.end():].strip()
+                        if not remaining_expr:
+                            return sub_result
+                        else:
+                            expr = json.dumps(sub_result) + remaining_expr
+                        break
+            else:
+                raise ExtractorError('Premature end of parens in %r' % expr)
+
+        for op, opfunc in _ASSIGN_OPERATORS:
+            m = re.match(r'''(?x)
+                (?P<out>%s)(?:\[(?P<index>[^\]]+?)\])?
+                \s*%s
+                (?P<expr>.*)$''' % (_NAME_RE, re.escape(op)), expr)
+            if not m:
+                continue
+            right_val = self.interpret_expression(
+                m.group('expr'), local_vars, allow_recursion - 1)
+
+            if m.groupdict().get('index'):
+                lvar = local_vars[m.group('out')]
+                idx = self.interpret_expression(
+                    m.group('index'), local_vars, allow_recursion)
+                assert isinstance(idx, int)
+                cur = lvar[idx]
+                val = opfunc(cur, right_val)
+                lvar[idx] = val
+                return val
+            else:
+                cur = local_vars.get(m.group('out'))
+                val = opfunc(cur, right_val)
+                local_vars[m.group('out')] = val
+                return val
+
         if expr.isdigit():
             return int(expr)
 
-        if expr.isalpha():
-            return local_vars[expr]
+        var_m = re.match(
+            r'(?!if|return|true|false)(?P<name>%s)$' % _NAME_RE,
+            expr)
+        if var_m:
+            return local_vars[var_m.group('name')]
 
         try:
             return json.loads(expr)
@@ -61,7 +124,7 @@ class JSInterpreter(object):
             pass
 
         m = re.match(
-            r'^(?P<var>[$a-zA-Z0-9_]+)\.(?P<member>[^(]+)(?:\(+(?P<args>[^()]*)\))?$',
+            r'(?P<var>%s)\.(?P<member>[^(]+)(?:\(+(?P<args>[^()]*)\))?$' % _NAME_RE,
             expr)
         if m:
             variable = m.group('var')
@@ -114,23 +177,31 @@ class JSInterpreter(object):
             return obj[member](argvals)
 
         m = re.match(
-            r'^(?P<in>[a-z]+)\[(?P<idx>.+)\]$', expr)
+            r'(?P<in>%s)\[(?P<idx>.+)\]$' % _NAME_RE, expr)
         if m:
             val = local_vars[m.group('in')]
             idx = self.interpret_expression(
                 m.group('idx'), local_vars, allow_recursion - 1)
             return val[idx]
 
-        m = re.match(r'^(?P<a>.+?)(?P<op>[%])(?P<b>.+?)$', expr)
-        if m:
-            a = self.interpret_expression(
-                m.group('a'), local_vars, allow_recursion)
-            b = self.interpret_expression(
-                m.group('b'), local_vars, allow_recursion)
-            return a % b
+        for op, opfunc in _OPERATORS:
+            m = re.match(r'(?P<x>.+?)%s(?P<y>.+)' % re.escape(op), expr)
+            if not m:
+                continue
+            x, abort = self.interpret_statement(
+                m.group('x'), local_vars, allow_recursion - 1)
+            if abort:
+                raise ExtractorError(
+                    'Premature left-side return of %s in %r' % (op, expr))
+            y, abort = self.interpret_statement(
+                m.group('y'), local_vars, allow_recursion - 1)
+            if abort:
+                raise ExtractorError(
+                    'Premature right-side return of %s in %r' % (op, expr))
+            return opfunc(x, y)
 
         m = re.match(
-            r'^(?P<func>[a-zA-Z$]+)\((?P<args>[a-z0-9,]+)\)$', expr)
+            r'^(?P<func>%s)\((?P<args>[a-zA-Z0-9_$,]+)\)$' % _NAME_RE, expr)
         if m:
             fname = m.group('func')
             argvals = tuple([
@@ -139,6 +210,7 @@ class JSInterpreter(object):
             if fname not in self._functions:
                 self._functions[fname] = self.extract_function(fname)
             return self._functions[fname](argvals)
+
         raise ExtractorError('Unsupported JS expression %r' % expr)
 
     def extract_object(self, objname):
@@ -162,9 +234,11 @@ class JSInterpreter(object):
 
     def extract_function(self, funcname):
         func_m = re.search(
-            (r'(?:function %s|[{;]%s\s*=\s*function)' % (
-                re.escape(funcname), re.escape(funcname))) +
-            r'\((?P<args>[a-z,]+)\){(?P<code>[^}]+)}',
+            r'''(?x)
+                (?:function\s+%s|[{;]%s\s*=\s*function)\s*
+                \((?P<args>[^)]*)\)\s*
+                \{(?P<code>[^}]+)\}''' % (
+                re.escape(funcname), re.escape(funcname)),
             self.code)
         if func_m is None:
             raise ExtractorError('Could not find JS function %r' % funcname)
@@ -172,10 +246,16 @@ class JSInterpreter(object):
 
         return self.build_function(argnames, func_m.group('code'))
 
+    def call_function(self, funcname, *args):
+        f = self.extract_function(funcname)
+        return f(args)
+
     def build_function(self, argnames, code):
         def resf(args):
             local_vars = dict(zip(argnames, args))
             for stmt in code.split(';'):
-                res = self.interpret_statement(stmt, local_vars)
+                res, abort = self.interpret_statement(stmt, local_vars)
+                if abort:
+                    break
             return res
         return resf

youtube_dl/options.py

@@ -5,6 +5,7 @@ import optparse
 import shlex
 import sys
 
+from .downloader.external import list_external_downloaders
 from .compat import (
     compat_expanduser,
     compat_getenv,
@@ -199,6 +200,10 @@ def parseOpts(overrideArguments=None):
         '--playlist-end',
         dest='playlistend', metavar='NUMBER', default=None, type=int,
         help='playlist video to end at (default is last)')
+    selection.add_option(
+        '--playlist-items',
+        dest='playlist_items', metavar='ITEM_SPEC', default=None,
+        help='playlist video items to download. Specify indices of the videos in the playlist seperated by commas like: "--playlist-items 1,2,5,8" if you want to download videos indexed 1, 2, 5, 8 in the playlist. You can specify range: "--playlist-items 1-3,7,10-13", it will download the videos at index 1, 2, 3, 7, 10, 11, 12 and 13.')
     selection.add_option(
         '--match-title',
         dest='matchtitle', metavar='REGEX',
@@ -292,7 +297,7 @@ def parseOpts(overrideArguments=None):
             ' You can filter the video results by putting a condition in'
             ' brackets, as in -f "best[height=720]"'
             ' (or -f "[filesize>10M]"). '
-            ' This works for filesize, height, width, tbr, abr, and vbr'
+            ' This works for filesize, height, width, tbr, abr, vbr, and fps'
             ' and the comparisons <, <=, >, >=, =, != .'
             ' Formats for which the value is not known are excluded unless you'
             ' put a question mark (?) after the operator.'
@@ -372,7 +377,7 @@ def parseOpts(overrideArguments=None):
     downloader.add_option(
         '-R', '--retries',
         dest='retries', metavar='RETRIES', default=10,
-        help='number of retries (default is %default)')
+        help='number of retries (default is %default), or "infinite".')
     downloader.add_option(
         '--buffer-size',
         dest='buffersize', metavar='SIZE', default='1024',
@@ -389,6 +394,15 @@ def parseOpts(overrideArguments=None):
         '--playlist-reverse',
         action='store_true',
         help='Download playlist videos in reverse order')
+    downloader.add_option(
+        '--xattr-set-filesize',
+        dest='xattr_set_filesize', action='store_true',
+        help='(experimental) set file xattribute ytdl.filesize with expected filesize')
+    downloader.add_option(
+        '--external-downloader',
+        dest='external_downloader', metavar='COMMAND',
+        help='(experimental) Use the specified external downloader. '
+             'Currently supports %s' % ','.join(list_external_downloaders()))
 
     workarounds = optparse.OptionGroup(parser, 'Workarounds')
     workarounds.add_option(
@@ -421,6 +435,10 @@ def parseOpts(overrideArguments=None):
         '--bidi-workaround',
         dest='bidi_workaround', action='store_true',
         help='Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH')
+    workarounds.add_option(
+        '--sleep-interval', metavar='SECONDS',
+        dest='sleep_interval', type=float,
+        help='Number of seconds to sleep before each download.')
 
     verbosity = optparse.OptionGroup(parser, 'Verbosity / Simulation Options')
     verbosity.add_option(
@@ -604,10 +622,6 @@ def parseOpts(overrideArguments=None):
         '--write-annotations',
         action='store_true', dest='writeannotations', default=False,
         help='write video annotations to a .annotation file')
-    filesystem.add_option(
-        '--write-thumbnail',
-        action='store_true', dest='writethumbnail', default=False,
-        help='write thumbnail image to disk')
     filesystem.add_option(
         '--load-info',
         dest='load_info_filename', metavar='FILE',
@@ -627,6 +641,20 @@ def parseOpts(overrideArguments=None):
         action='store_true', dest='rm_cachedir',
         help='Delete all filesystem cache files')
 
+    thumbnail = optparse.OptionGroup(parser, 'Thumbnail images')
+    thumbnail.add_option(
+        '--write-thumbnail',
+        action='store_true', dest='writethumbnail', default=False,
+        help='write thumbnail image to disk')
+    thumbnail.add_option(
+        '--write-all-thumbnails',
+        action='store_true', dest='write_all_thumbnails', default=False,
+        help='write all thumbnail image formats to disk')
+    thumbnail.add_option(
+        '--list-thumbnails',
+        action='store_true', dest='list_thumbnails', default=False,
+        help='Simulate and list all available thumbnail formats')
+
     postproc = optparse.OptionGroup(parser, 'Post-processing Options')
     postproc.add_option(
         '-x', '--extract-audio',
@@ -670,10 +698,9 @@ def parseOpts(overrideArguments=None):
     postproc.add_option(
         '--fixup',
         metavar='POLICY', dest='fixup', default='detect_or_warn',
-        help='(experimental) Automatically correct known faults of the file. '
+        help='Automatically correct known faults of the file. '
              'One of never (do nothing), warn (only emit a warning), '
-             'detect_or_warn(check whether we can do anything about it, warn '
-             'otherwise')
+             'detect_or_warn(the default; fix file if we can, warn otherwise)')
     postproc.add_option(
         '--prefer-avconv',
         action='store_false', dest='prefer_ffmpeg',
@@ -692,6 +719,7 @@ def parseOpts(overrideArguments=None):
     parser.add_option_group(selection)
     parser.add_option_group(downloader)
     parser.add_option_group(filesystem)
+    parser.add_option_group(thumbnail)
     parser.add_option_group(verbosity)
     parser.add_option_group(workarounds)
     parser.add_option_group(video_format)

youtube_dl/postprocessor/__init__.py

@@ -7,6 +7,7 @@ from .ffmpeg import (
     FFmpegEmbedSubtitlePP,
     FFmpegExtractAudioPP,
     FFmpegFixupStretchedPP,
+    FFmpegFixupM4aPP,
     FFmpegMergerPP,
     FFmpegMetadataPP,
     FFmpegVideoConvertorPP,
@@ -25,6 +26,7 @@ __all__ = [
     'FFmpegAudioFixPP',
     'FFmpegEmbedSubtitlePP',
     'FFmpegExtractAudioPP',
+    'FFmpegFixupM4aPP',
     'FFmpegFixupStretchedPP',
     'FFmpegMergerPP',
     'FFmpegMetadataPP',

youtube_dl/postprocessor/ffmpeg.py

@@ -509,6 +509,10 @@ class FFmpegMetadataPP(FFmpegPostProcessor):
             metadata['artist'] = info['uploader']
         elif info.get('uploader_id') is not None:
             metadata['artist'] = info['uploader_id']
+        if info.get('description') is not None:
+            metadata['description'] = info['description']
+        if info.get('webpage_url') is not None:
+            metadata['comment'] = info['webpage_url']
 
         if not metadata:
             self._downloader.to_screen('[ffmpeg] There isn\'t any metadata to add')
@@ -560,7 +564,7 @@ class FFmpegFixupStretchedPP(FFmpegPostProcessor):
     def run(self, info):
         stretched_ratio = info.get('stretched_ratio')
         if stretched_ratio is None or stretched_ratio == 1:
-            return
+            return True, info
 
         filename = info['filepath']
         temp_filename = prepend_extension(filename, 'temp')
@@ -573,3 +577,21 @@ class FFmpegFixupStretchedPP(FFmpegPostProcessor):
         os.rename(encodeFilename(temp_filename), encodeFilename(filename))
 
         return True, info
+
+
+class FFmpegFixupM4aPP(FFmpegPostProcessor):
+    def run(self, info):
+        if info.get('container') != 'm4a_dash':
+            return True, info
+
+        filename = info['filepath']
+        temp_filename = prepend_extension(filename, 'temp')
+
+        options = ['-c', 'copy', '-f', 'mp4']
+        self._downloader.to_screen('[ffmpeg] Correcting container in "%s"' % filename)
+        self.run_ffmpeg(filename, temp_filename, options)
+
+        os.remove(encodeFilename(filename))
+        os.rename(encodeFilename(temp_filename), encodeFilename(filename))
+
+        return True, info

youtube_dl/utils.py

@@ -32,6 +32,7 @@ import xml.etree.ElementTree
 import zlib
 
 from .compat import (
+    compat_basestring,
     compat_chr,
     compat_getenv,
     compat_html_entities,
@@ -140,7 +141,7 @@ else:
     def find_xpath_attr(node, xpath, key, val):
         # Here comes the crazy part: In 2.6, if the xpath is a unicode,
         # .//node does not match if a node is a direct child of . !
-        if isinstance(xpath, unicode):
+        if isinstance(xpath, compat_str):
             xpath = xpath.encode('ascii')
 
         for f in node.findall(xpath):
@@ -411,25 +412,9 @@ def make_HTTPS_handler(params, **kwargs):
             pass
 
     if sys.version_info < (3, 2):
-        import httplib
-
-        class HTTPSConnectionV3(httplib.HTTPSConnection):
-            def __init__(self, *args, **kwargs):
-                httplib.HTTPSConnection.__init__(self, *args, **kwargs)
-
-            def connect(self):
-                sock = socket.create_connection((self.host, self.port), self.timeout)
-                if getattr(self, '_tunnel_host', False):
-                    self.sock = sock
-                    self._tunnel()
-                try:
-                    self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_TLSv1)
-                except ssl.SSLError:
-                    self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv23)
-
-        return YoutubeDLHTTPSHandler(params, https_conn_class=HTTPSConnectionV3, **kwargs)
+        return YoutubeDLHTTPSHandler(params, **kwargs)
     else:  # Python < 3.4
-        context = ssl.SSLContext(ssl.PROTOCOL_SSLv23)
+        context = ssl.SSLContext(ssl.PROTOCOL_TLSv1)
         context.verify_mode = (ssl.CERT_NONE
                                if opts_no_check_certificate
                                else ssl.CERT_REQUIRED)
@@ -560,7 +545,9 @@ def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs):
                 sock = compat_socket_create_connection(
                     (self.host, self.port), self.timeout, sa)
                 if is_https:
-                    self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file)
+                    self.sock = ssl.wrap_socket(
+                        sock, self.key_file, self.cert_file,
+                        ssl_version=ssl.PROTOCOL_TLSv1)
                 else:
                     self.sock = sock
             hc.connect = functools.partial(_hc_connect, hc)
@@ -612,17 +599,14 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
 
     def http_request(self, req):
         for h, v in std_headers.items():
-            if h not in req.headers:
+            # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275
+            # The dict keys are capitalized because of this bug by urllib
+            if h.capitalize() not in req.headers:
                 req.add_header(h, v)
         if 'Youtubedl-no-compression' in req.headers:
             if 'Accept-encoding' in req.headers:
                 del req.headers['Accept-encoding']
             del req.headers['Youtubedl-no-compression']
-        if 'Youtubedl-user-agent' in req.headers:
-            if 'User-agent' in req.headers:
-                del req.headers['User-agent']
-            req.headers['User-agent'] = req.headers['Youtubedl-user-agent']
-            del req.headers['Youtubedl-user-agent']
 
         if sys.version_info < (2, 7) and '#' in req.get_full_url():
             # Python 2.6 is brain-dead when it comes to fragments
@@ -671,9 +655,14 @@ class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler):
         self._params = params
 
     def https_open(self, req):
+        kwargs = {}
+        if hasattr(self, '_context'):  # python > 2.6
+            kwargs['context'] = self._context
+        if hasattr(self, '_check_hostname'):  # python 3.x
+            kwargs['check_hostname'] = self._check_hostname
         return self.do_open(functools.partial(
             _create_http_connection, self, self._https_conn_class, True),
-            req)
+            req, **kwargs)
 
 
 def parse_iso8601(date_str, delimiter='T'):
@@ -875,6 +864,9 @@ def _windows_write_string(s, out):
     except AttributeError:
         # If the output stream doesn't have a fileno, it's virtual
         return False
+    except io.UnsupportedOperation:
+        # Some strange Windows pseudo files?
+        return False
     if fileno not in WIN_OUTPUT_IDS:
         return False
 
@@ -1271,7 +1263,7 @@ def float_or_none(v, scale=1, invscale=1, default=None):
 
 
 def parse_duration(s):
-    if not isinstance(s, basestring if sys.version_info < (3, 0) else compat_str):
+    if not isinstance(s, compat_basestring):
         return None
 
     s = s.strip()
@@ -1435,7 +1427,7 @@ def uppercase_escape(s):
 
 def escape_rfc3986(s):
     """Escape non-ASCII characters as suggested by RFC 3986"""
-    if sys.version_info < (3, 0) and isinstance(s, unicode):
+    if sys.version_info < (3, 0) and isinstance(s, compat_str):
         s = s.encode('utf-8')
     return compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]")
 
@@ -1631,3 +1623,53 @@ def age_restricted(content_limit, age_limit):
     if content_limit is None:
         return False  # Content available for everyone
     return age_limit < content_limit
+
+
+def is_html(first_bytes):
+    """ Detect whether a file contains HTML by examining its first bytes. """
+
+    BOMS = [
+        (b'\xef\xbb\xbf', 'utf-8'),
+        (b'\x00\x00\xfe\xff', 'utf-32-be'),
+        (b'\xff\xfe\x00\x00', 'utf-32-le'),
+        (b'\xff\xfe', 'utf-16-le'),
+        (b'\xfe\xff', 'utf-16-be'),
+    ]
+    for bom, enc in BOMS:
+        if first_bytes.startswith(bom):
+            s = first_bytes[len(bom):].decode(enc, 'replace')
+            break
+    else:
+        s = first_bytes.decode('utf-8', 'replace')
+
+    return re.match(r'^\s*<', s)
+
+
+def determine_protocol(info_dict):
+    protocol = info_dict.get('protocol')
+    if protocol is not None:
+        return protocol
+
+    url = info_dict['url']
+    if url.startswith('rtmp'):
+        return 'rtmp'
+    elif url.startswith('mms'):
+        return 'mms'
+    elif url.startswith('rtsp'):
+        return 'rtsp'
+
+    ext = determine_ext(url)
+    if ext == 'm3u8':
+        return 'm3u8'
+    elif ext == 'f4m':
+        return 'f4m'
+
+    return compat_urllib_parse_urlparse(url).scheme
+
+
+def render_table(header_row, data):
+    """ Render a list of rows, each as a list of values """
+    table = [header_row] + data
+    max_lens = [max(len(compat_str(v)) for v in col) for col in zip(*table)]
+    format_str = ' '.join('%-' + compat_str(ml + 1) + 's' for ml in max_lens[:-1]) + '%s'
+    return '\n'.join(format_str % tuple(row) for row in table)

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2015.01.23.1'
+__version__ = '2015.02.02.1'