diff --git a/AUTHORS b/AUTHORS index cc552bcb2..cdb56de3b 100644 --- a/AUTHORS +++ b/AUTHORS @@ -144,3 +144,6 @@ Lee Jenkins Anssi Hannula Lukáš Lalinský Qijiang Fan +Rémy Léone +Marco Ferragina +reiv diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index 32c2fd84c..09ce98ca2 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -1,6 +1,6 @@ **Please include the full output of youtube-dl when run with `-v`**. -The output (including the first lines) contain important debugging information. Issues without the full output are often not reproducible and therefore do not get solved in short order, if ever. +The output (including the first lines) contains important debugging information. Issues without the full output are often not reproducible and therefore do not get solved in short order, if ever. Please re-read your issue once again to avoid a couple of common mistakes (you can and should use this as a checklist): @@ -114,12 +114,13 @@ If you want to add support for a new site, you can follow this quick list (assum webpage = self._download_webpage(url, video_id) # TODO more code goes here, for example ... - title = self._html_search_regex(r'

(.*?)

', webpage, 'title') + title = self._html_search_regex(r'

(.+?)

', webpage, 'title') return { 'id': video_id, 'title': title, 'description': self._og_search_description(webpage), + 'uploader': self._search_regex(r']+id="uploader"[^>]*>([^<]+)<', webpage, 'uploader', fatal=False), # TODO more properties (see youtube_dl/extractor/common.py) } ``` diff --git a/Makefile b/Makefile index fdb1abb60..f826c1685 100644 --- a/Makefile +++ b/Makefile @@ -61,34 +61,34 @@ youtube-dl: youtube_dl/*.py youtube_dl/*/*.py chmod a+x youtube-dl README.md: youtube_dl/*.py youtube_dl/*/*.py - COLUMNS=80 python youtube_dl/__main__.py --help | python devscripts/make_readme.py + COLUMNS=80 $(PYTHON) youtube_dl/__main__.py --help | $(PYTHON) devscripts/make_readme.py CONTRIBUTING.md: README.md - python devscripts/make_contributing.py README.md CONTRIBUTING.md + $(PYTHON) devscripts/make_contributing.py README.md CONTRIBUTING.md supportedsites: - python devscripts/make_supportedsites.py docs/supportedsites.md + $(PYTHON) devscripts/make_supportedsites.py docs/supportedsites.md README.txt: README.md pandoc -f markdown -t plain README.md -o README.txt youtube-dl.1: README.md - python devscripts/prepare_manpage.py >youtube-dl.1.temp.md + $(PYTHON) devscripts/prepare_manpage.py >youtube-dl.1.temp.md pandoc -s -f markdown -t man youtube-dl.1.temp.md -o youtube-dl.1 rm -f youtube-dl.1.temp.md youtube-dl.bash-completion: youtube_dl/*.py youtube_dl/*/*.py devscripts/bash-completion.in - python devscripts/bash-completion.py + $(PYTHON) devscripts/bash-completion.py bash-completion: youtube-dl.bash-completion youtube-dl.zsh: youtube_dl/*.py youtube_dl/*/*.py devscripts/zsh-completion.in - python devscripts/zsh-completion.py + $(PYTHON) devscripts/zsh-completion.py zsh-completion: youtube-dl.zsh youtube-dl.fish: youtube_dl/*.py youtube_dl/*/*.py devscripts/fish-completion.in - python devscripts/fish-completion.py + $(PYTHON) devscripts/fish-completion.py fish-completion: youtube-dl.fish diff --git a/README.md b/README.md index cf4aebf3d..df419abe8 100644 --- a/README.md +++ b/README.md @@ -319,7 +319,8 @@ which means you can modify it, redistribute it or use it however you like. --all-formats Download all available video formats --prefer-free-formats Prefer free video formats unless a specific one is requested - -F, --list-formats List all available formats + -F, --list-formats List all available formats of specified + videos --youtube-skip-dash-manifest Do not download the DASH manifests and related data on YouTube videos --merge-output-format FORMAT If a merge is required (e.g. @@ -329,8 +330,8 @@ which means you can modify it, redistribute it or use it however you like. ## Subtitle Options: --write-sub Write subtitle file - --write-auto-sub Write automatic subtitle file (YouTube - only) + --write-auto-sub Write automatically generated subtitle file + (YouTube only) --all-subs Download all the available subtitles of the video --list-subs List all available subtitles for the video @@ -534,6 +535,12 @@ Most people asking this question are not aware that youtube-dl now defaults to d Apparently YouTube requires you to pass a CAPTCHA test if you download too much. We're [considering to provide a way to let you solve the CAPTCHA](https://github.com/rg3/youtube-dl/issues/154), but at the moment, your best course of action is pointing a webbrowser to the youtube URL, solving the CAPTCHA, and restart youtube-dl. +### Do I need any other programs? + +youtube-dl works fine on its own on most sites. However, if you want to convert video/audio, you'll need [avconv](https://libav.org/) or [ffmpeg](https://www.ffmpeg.org/). On some sites - most notably YouTube - videos can be retrieved in a higher quality format without sound. youtube-dl will detect whether avconv/ffmpeg is present and automatically pick the best option. + +Videos or video formats streamed via RTMP protocol can only be downloaded when [rtmpdump](https://rtmpdump.mplayerhq.hu/) is installed. Downloading MMS and RTSP videos requires either [mplayer](http://mplayerhq.hu/) or [mpv](https://mpv.io/) to be installed. + ### I have downloaded a video but how can I play it? Once the video is fully downloaded, use any video player, such as [vlc](http://www.videolan.org) or [mplayer](http://www.mplayerhq.hu/). @@ -710,12 +717,13 @@ If you want to add support for a new site, you can follow this quick list (assum webpage = self._download_webpage(url, video_id) # TODO more code goes here, for example ... - title = self._html_search_regex(r'

(.*?)

', webpage, 'title') + title = self._html_search_regex(r'

(.+?)

', webpage, 'title') return { 'id': video_id, 'title': title, 'description': self._og_search_description(webpage), + 'uploader': self._search_regex(r']+id="uploader"[^>]*>([^<]+)<', webpage, 'uploader', fatal=False), # TODO more properties (see youtube_dl/extractor/common.py) } ``` @@ -794,7 +802,7 @@ Bugs and suggestions should be reported at: + + ''' self.assertEqual(ie._og_search_title(html), 'Foo') self.assertEqual(ie._og_search_description(html), 'Some video\'s description ') self.assertEqual(ie._og_search_thumbnail(html), 'http://domain.com/pic.jpg?key1=val1&key2=val2') self.assertEqual(ie._og_search_video_url(html, default=None), None) self.assertEqual(ie._og_search_property('foobar', html), 'Foo') + self.assertEqual(ie._og_search_property('test1', html), 'foo > < bar') + self.assertEqual(ie._og_search_property('test2', html), 'foo >//< bar') def test_html_search_meta(self): ie = self.ie diff --git a/test/test_compat.py b/test/test_compat.py index 4ee0dc99d..b6bfad05e 100644 --- a/test/test_compat.py +++ b/test/test_compat.py @@ -13,8 +13,10 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) from youtube_dl.utils import get_filesystem_encoding from youtube_dl.compat import ( compat_getenv, + compat_etree_fromstring, compat_expanduser, compat_shlex_split, + compat_str, compat_urllib_parse_unquote, compat_urllib_parse_unquote_plus, ) @@ -71,5 +73,20 @@ class TestCompat(unittest.TestCase): def test_compat_shlex_split(self): self.assertEqual(compat_shlex_split('-option "one two"'), ['-option', 'one two']) + def test_compat_etree_fromstring(self): + xml = ''' + + foo + 中文 + spam + + ''' + doc = compat_etree_fromstring(xml.encode('utf-8')) + self.assertTrue(isinstance(doc.attrib['foo'], compat_str)) + self.assertTrue(isinstance(doc.attrib['spam'], compat_str)) + self.assertTrue(isinstance(doc.find('normal').text, compat_str)) + self.assertTrue(isinstance(doc.find('chinese').text, compat_str)) + self.assertTrue(isinstance(doc.find('foo/bar').text, compat_str)) + if __name__ == '__main__': unittest.main() diff --git a/test/test_download.py b/test/test_download.py index 284418834..a3f1c0644 100644 --- a/test/test_download.py +++ b/test/test_download.py @@ -102,7 +102,7 @@ def generator(test_case): params = get_params(test_case.get('params', {})) if is_playlist and 'playlist' not in test_case: - params.setdefault('extract_flat', True) + params.setdefault('extract_flat', 'in_playlist') params.setdefault('skip_download', True) ydl = YoutubeDL(params, auto_init=False) diff --git a/test/test_jsinterp.py b/test/test_jsinterp.py index fc73e5dc2..63c350b8f 100644 --- a/test/test_jsinterp.py +++ b/test/test_jsinterp.py @@ -19,6 +19,9 @@ class TestJSInterpreter(unittest.TestCase): jsi = JSInterpreter('function x3(){return 42;}') self.assertEqual(jsi.call_function('x3'), 42) + jsi = JSInterpreter('var x5 = function(){return 42;}') + self.assertEqual(jsi.call_function('x5'), 42) + def test_calc(self): jsi = JSInterpreter('function x4(a){return 2*a+1;}') self.assertEqual(jsi.call_function('x4', 3), 7) diff --git a/test/test_subtitles.py b/test/test_subtitles.py index 0343967d9..75f0ea75f 100644 --- a/test/test_subtitles.py +++ b/test/test_subtitles.py @@ -28,6 +28,7 @@ from youtube_dl.extractor import ( ThePlatformFeedIE, RTVEALaCartaIE, FunnyOrDieIE, + DemocracynowIE, ) @@ -346,5 +347,25 @@ class TestFunnyOrDieSubtitles(BaseTestSubtitles): self.assertEqual(md5(subtitles['en']), 'c5593c193eacd353596c11c2d4f9ecc4') +class TestDemocracynowSubtitles(BaseTestSubtitles): + url = 'http://www.democracynow.org/shows/2015/7/3' + IE = DemocracynowIE + + def test_allsubtitles(self): + self.DL.params['writesubtitles'] = True + self.DL.params['allsubtitles'] = True + subtitles = self.getSubtitles() + self.assertEqual(set(subtitles.keys()), set(['en'])) + self.assertEqual(md5(subtitles['en']), 'acaca989e24a9e45a6719c9b3d60815c') + + def test_subtitles_in_page(self): + self.url = 'http://www.democracynow.org/2015/7/3/this_flag_comes_down_today_bree' + self.DL.params['writesubtitles'] = True + self.DL.params['allsubtitles'] = True + subtitles = self.getSubtitles() + self.assertEqual(set(subtitles.keys()), set(['en'])) + self.assertEqual(md5(subtitles['en']), 'acaca989e24a9e45a6719c9b3d60815c') + + if __name__ == '__main__': unittest.main() diff --git a/test/test_utils.py b/test/test_utils.py index a5f164c49..501355c74 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -21,6 +21,7 @@ from youtube_dl.utils import ( clean_html, DateRange, detect_exe_version, + determine_ext, encodeFilename, escape_rfc3986, escape_url, @@ -68,6 +69,9 @@ from youtube_dl.utils import ( cli_valueless_option, cli_bool_option, ) +from youtube_dl.compat import ( + compat_etree_fromstring, +) class TestUtil(unittest.TestCase): @@ -207,8 +211,8 @@ class TestUtil(unittest.TestCase): self.assertEqual(unescapeHTML('%20;'), '%20;') self.assertEqual(unescapeHTML('/'), '/') self.assertEqual(unescapeHTML('/'), '/') - self.assertEqual( - unescapeHTML('é'), 'é') + self.assertEqual(unescapeHTML('é'), 'é') + self.assertEqual(unescapeHTML('�'), '�') def test_daterange(self): _20century = DateRange("19000101", "20000101") @@ -233,6 +237,14 @@ class TestUtil(unittest.TestCase): unified_strdate('2/2/2015 6:47:40 PM', day_first=False), '20150202') self.assertEqual(unified_strdate('25-09-2014'), '20140925') + self.assertEqual(unified_strdate('UNKNOWN DATE FORMAT'), None) + + def test_determine_ext(self): + self.assertEqual(determine_ext('http://example.com/foo/bar.mp4/?download'), 'mp4') + self.assertEqual(determine_ext('http://example.com/foo/bar/?download', None), None) + self.assertEqual(determine_ext('http://example.com/foo/bar.nonext/?download', None), None) + self.assertEqual(determine_ext('http://example.com/foo/bar/mp4?download', None), None) + self.assertEqual(determine_ext('http://example.com/foo/bar.m3u8//?download'), 'm3u8') def test_find_xpath_attr(self): testxml = ''' @@ -242,7 +254,7 @@ class TestUtil(unittest.TestCase): ''' - doc = xml.etree.ElementTree.fromstring(testxml) + doc = compat_etree_fromstring(testxml) self.assertEqual(find_xpath_attr(doc, './/fourohfour', 'n'), None) self.assertEqual(find_xpath_attr(doc, './/fourohfour', 'n', 'v'), None) @@ -263,7 +275,7 @@ class TestUtil(unittest.TestCase): http://server.com/download.mp3 ''' - doc = xml.etree.ElementTree.fromstring(testxml) + doc = compat_etree_fromstring(testxml) find = lambda p: doc.find(xpath_with_ns(p, {'media': 'http://example.com/'})) self.assertTrue(find('media:song') is not None) self.assertEqual(find('media:song/media:author').text, 'The Author') @@ -275,9 +287,16 @@ class TestUtil(unittest.TestCase): p = xml.etree.ElementTree.SubElement(div, 'p') p.text = 'Foo' self.assertEqual(xpath_element(doc, 'div/p'), p) + self.assertEqual(xpath_element(doc, ['div/p']), p) + self.assertEqual(xpath_element(doc, ['div/bar', 'div/p']), p) self.assertEqual(xpath_element(doc, 'div/bar', default='default'), 'default') + self.assertEqual(xpath_element(doc, ['div/bar'], default='default'), 'default') self.assertTrue(xpath_element(doc, 'div/bar') is None) + self.assertTrue(xpath_element(doc, ['div/bar']) is None) + self.assertTrue(xpath_element(doc, ['div/bar'], 'div/baz') is None) self.assertRaises(ExtractorError, xpath_element, doc, 'div/bar', fatal=True) + self.assertRaises(ExtractorError, xpath_element, doc, ['div/bar'], fatal=True) + self.assertRaises(ExtractorError, xpath_element, doc, ['div/bar', 'div/baz'], fatal=True) def test_xpath_text(self): testxml = ''' @@ -285,7 +304,7 @@ class TestUtil(unittest.TestCase):

Foo

''' - doc = xml.etree.ElementTree.fromstring(testxml) + doc = compat_etree_fromstring(testxml) self.assertEqual(xpath_text(doc, 'div/p'), 'Foo') self.assertEqual(xpath_text(doc, 'div/bar', default='default'), 'default') self.assertTrue(xpath_text(doc, 'div/bar') is None) @@ -297,7 +316,7 @@ class TestUtil(unittest.TestCase):

Foo

''' - doc = xml.etree.ElementTree.fromstring(testxml) + doc = compat_etree_fromstring(testxml) self.assertEqual(xpath_attr(doc, 'div/p', 'x'), 'a') self.assertEqual(xpath_attr(doc, 'div/bar', 'x'), None) self.assertEqual(xpath_attr(doc, 'div/p', 'y'), None) @@ -425,6 +444,8 @@ class TestUtil(unittest.TestCase): self.assertEqual(parse_iso8601('2014-03-23T22:04:26+0000'), 1395612266) self.assertEqual(parse_iso8601('2014-03-23T22:04:26Z'), 1395612266) self.assertEqual(parse_iso8601('2014-03-23T22:04:26.1234Z'), 1395612266) + self.assertEqual(parse_iso8601('2015-09-29T08:27:31.727'), 1443515251) + self.assertEqual(parse_iso8601('2015-09-29T08-27-31.727'), None) def test_strip_jsonp(self): stripped = strip_jsonp('cb ([ {"id":"532cb",\n\n\n"x":\n3}\n]\n);') @@ -495,6 +516,9 @@ class TestUtil(unittest.TestCase): "playlist":[{"controls":{"all":null}}] }''') + inp = '''"The CW\\'s \\'Crazy Ex-Girlfriend\\'"''' + self.assertEqual(js_to_json(inp), '''"The CW's 'Crazy Ex-Girlfriend'"''') + inp = '"SAND Number: SAND 2013-7800P\\nPresenter: Tom Russo\\nHabanero Software Training - Xyce Software\\nXyce, Sandia\\u0027s"' json_code = js_to_json(inp) self.assertEqual(json.loads(json_code), json.loads(inp)) diff --git a/test/test_youtube_lists.py b/test/test_youtube_lists.py index c889b6f15..26aadb34f 100644 --- a/test/test_youtube_lists.py +++ b/test/test_youtube_lists.py @@ -57,5 +57,14 @@ class TestYoutubeLists(unittest.TestCase): entries = result['entries'] self.assertEqual(len(entries), 100) + def test_youtube_flat_playlist_titles(self): + dl = FakeYDL() + dl.params['extract_flat'] = True + ie = YoutubePlaylistIE(dl) + result = ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re') + self.assertIsPlaylist(result) + for entry in result['entries']: + self.assertTrue(entry.get('title')) + if __name__ == '__main__': unittest.main() diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index adf70d658..9a8c7da05 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -28,6 +28,7 @@ if os.name == 'nt': import ctypes from .compat import ( + compat_basestring, compat_cookiejar, compat_expanduser, compat_get_terminal_size, @@ -37,6 +38,7 @@ from .compat import ( compat_tokenize_tokenize, compat_urllib_error, compat_urllib_request, + compat_urllib_request_DataHandler, ) from .utils import ( ContentTooShortError, @@ -62,6 +64,7 @@ from .utils import ( SameFileError, sanitize_filename, sanitize_path, + sanitized_Request, std_headers, subtitles_filename, UnavailableVideoError, @@ -155,7 +158,7 @@ class YoutubeDL(object): 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 + writeautomaticsub: Write the automatically generated subtitles to a file allsubtitles: Downloads all the subtitles of the video (requires writesubtitles or writeautomaticsub) listsubtitles: Lists all available subtitles for the video @@ -571,7 +574,7 @@ class YoutubeDL(object): if v is not None) template_dict = collections.defaultdict(lambda: 'NA', template_dict) - outtmpl = sanitize_path(self.params.get('outtmpl', DEFAULT_OUTTMPL)) + outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL) tmpl = compat_expanduser(outtmpl) filename = tmpl % template_dict # Temporary fix for #4787 @@ -579,7 +582,7 @@ class YoutubeDL(object): # 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 + return sanitize_path(filename) except ValueError as err: self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')') return None @@ -832,6 +835,7 @@ class YoutubeDL(object): extra_info=extra) playlist_results.append(entry_result) ie_result['entries'] = playlist_results + self.to_screen('[download] Finished downloading playlist: %s' % playlist) return ie_result elif result_type == 'compat_list': self.report_warning( @@ -936,7 +940,7 @@ class YoutubeDL(object): filter_parts.append(string) def _remove_unused_ops(tokens): - # Remove operators that we don't use and join them with the sourrounding strings + # Remove operators that we don't use and join them with the surrounding strings # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9' ALLOWED_OPS = ('/', '+', ',', '(', ')') last_string, last_start, last_end, last_line = None, None, None, None @@ -1185,7 +1189,7 @@ class YoutubeDL(object): return res def _calc_cookies(self, info_dict): - pr = compat_urllib_request.Request(info_dict['url']) + pr = sanitized_Request(info_dict['url']) self.cookiejar.add_cookie_header(pr) return pr.get_header('Cookie') @@ -1869,6 +1873,8 @@ class YoutubeDL(object): def urlopen(self, req): """ Start an HTTP download """ + if isinstance(req, compat_basestring): + req = sanitized_Request(req) return self._opener.open(req, timeout=self._socket_timeout) def print_debug_header(self): @@ -1967,8 +1973,9 @@ class YoutubeDL(object): debuglevel = 1 if self.params.get('debug_printtraffic') else 0 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel) ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel) + data_handler = compat_urllib_request_DataHandler() opener = compat_urllib_request.build_opener( - proxy_handler, https_handler, cookie_processor, ydlh) + proxy_handler, https_handler, cookie_processor, ydlh, data_handler) # Delete the default user-agent header, which would otherwise apply in # cases where our custom HTTP handler doesn't come into play diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index 5e2ed4d4b..9f131f5db 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -377,7 +377,7 @@ def _real_main(argv=None): with YoutubeDL(ydl_opts) as ydl: # Update version if opts.update_self: - update_self(ydl.to_screen, opts.verbose) + update_self(ydl.to_screen, opts.verbose, ydl._opener) # Remove cache dir if opts.rm_cachedir: diff --git a/youtube_dl/compat.py b/youtube_dl/compat.py index 192e1c515..a3e85264a 100644 --- a/youtube_dl/compat.py +++ b/youtube_dl/compat.py @@ -1,7 +1,10 @@ from __future__ import unicode_literals +import binascii import collections +import email import getpass +import io import optparse import os import re @@ -11,6 +14,7 @@ import socket import subprocess import sys import itertools +import xml.etree.ElementTree try: @@ -38,6 +42,11 @@ try: except ImportError: # Python 2 import urlparse as compat_urlparse +try: + import urllib.response as compat_urllib_response +except ImportError: # Python 2 + import urllib as compat_urllib_response + try: import http.cookiejar as compat_cookiejar except ImportError: # Python 2 @@ -155,6 +164,40 @@ except ImportError: # Python 2 string = string.replace('+', ' ') return compat_urllib_parse_unquote(string, encoding, errors) +try: + from urllib.request import DataHandler as compat_urllib_request_DataHandler +except ImportError: # Python < 3.4 + # Ported from CPython 98774:1733b3bd46db, Lib/urllib/request.py + class compat_urllib_request_DataHandler(compat_urllib_request.BaseHandler): + def data_open(self, req): + # data URLs as specified in RFC 2397. + # + # ignores POSTed data + # + # syntax: + # dataurl := "data:" [ mediatype ] [ ";base64" ] "," data + # mediatype := [ type "/" subtype ] *( ";" parameter ) + # data := *urlchar + # parameter := attribute "=" value + url = req.get_full_url() + + scheme, data = url.split(":", 1) + mediatype, data = data.split(",", 1) + + # even base64 encoded data URLs might be quoted so unquote in any case: + data = compat_urllib_parse_unquote_to_bytes(data) + if mediatype.endswith(";base64"): + data = binascii.a2b_base64(data) + mediatype = mediatype[:-7] + + if not mediatype: + mediatype = "text/plain;charset=US-ASCII" + + headers = email.message_from_string( + "Content-type: %s\nContent-length: %d\n" % (mediatype, len(data))) + + return compat_urllib_response.addinfourl(io.BytesIO(data), headers, url) + try: compat_basestring = basestring # Python 2 except NameError: @@ -170,6 +213,43 @@ try: except ImportError: # Python 2.6 from xml.parsers.expat import ExpatError as compat_xml_parse_error +if sys.version_info[0] >= 3: + compat_etree_fromstring = xml.etree.ElementTree.fromstring +else: + # python 2.x tries to encode unicode strings with ascii (see the + # XMLParser._fixtext method) + etree = xml.etree.ElementTree + + try: + _etree_iter = etree.Element.iter + except AttributeError: # Python <=2.6 + def _etree_iter(root): + for el in root.findall('*'): + yield el + for sub in _etree_iter(el): + yield sub + + # on 2.6 XML doesn't have a parser argument, function copied from CPython + # 2.7 source + def _XML(text, parser=None): + if not parser: + parser = etree.XMLParser(target=etree.TreeBuilder()) + parser.feed(text) + return parser.close() + + def _element_factory(*args, **kwargs): + el = etree.Element(*args, **kwargs) + for k, v in el.items(): + if isinstance(v, bytes): + el.set(k, v.decode('utf-8')) + return el + + def compat_etree_fromstring(text): + doc = _XML(text, parser=etree.XMLParser(target=etree.TreeBuilder(element_factory=_element_factory))) + for el in _etree_iter(doc): + if el.text is not None and isinstance(el.text, bytes): + el.text = el.text.decode('utf-8') + return doc try: from urllib.parse import parse_qs as compat_parse_qs @@ -465,6 +545,7 @@ __all__ = [ 'compat_chr', 'compat_cookiejar', 'compat_cookies', + 'compat_etree_fromstring', 'compat_expanduser', 'compat_get_terminal_size', 'compat_getenv', @@ -489,6 +570,8 @@ __all__ = [ 'compat_urllib_parse_unquote_to_bytes', 'compat_urllib_parse_urlparse', 'compat_urllib_request', + 'compat_urllib_request_DataHandler', + 'compat_urllib_response', 'compat_urlparse', 'compat_urlretrieve', 'compat_xml_parse_error', diff --git a/youtube_dl/downloader/common.py b/youtube_dl/downloader/common.py index 29a4500d3..b8bf8daf8 100644 --- a/youtube_dl/downloader/common.py +++ b/youtube_dl/downloader/common.py @@ -42,7 +42,7 @@ class FileDownloader(object): 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) + (experimental) external_downloader_args: A list of additional command-line arguments for the external downloader. diff --git a/youtube_dl/downloader/dash.py b/youtube_dl/downloader/dash.py index 8b6fa2753..535f2a7fc 100644 --- a/youtube_dl/downloader/dash.py +++ b/youtube_dl/downloader/dash.py @@ -3,7 +3,7 @@ from __future__ import unicode_literals import re from .common import FileDownloader -from ..compat import compat_urllib_request +from ..utils import sanitized_Request class DashSegmentsFD(FileDownloader): @@ -22,7 +22,7 @@ class DashSegmentsFD(FileDownloader): def append_url_to_file(outf, target_url, target_name, remaining_bytes=None): self.to_screen('[DashSegments] %s: Downloading %s' % (info_dict['id'], target_name)) - req = compat_urllib_request.Request(target_url) + req = sanitized_Request(target_url) if remaining_bytes is not None: req.add_header('Range', 'bytes=0-%d' % (remaining_bytes - 1)) diff --git a/youtube_dl/downloader/f4m.py b/youtube_dl/downloader/f4m.py index 174180db5..6170cc155 100644 --- a/youtube_dl/downloader/f4m.py +++ b/youtube_dl/downloader/f4m.py @@ -5,12 +5,13 @@ import io import itertools import os import time -import xml.etree.ElementTree as etree from .fragment import FragmentFD from ..compat import ( + compat_etree_fromstring, compat_urlparse, compat_urllib_error, + compat_urllib_parse_urlparse, ) from ..utils import ( encodeFilename, @@ -285,9 +286,11 @@ class F4mFD(FragmentFD): man_url = info_dict['url'] requested_bitrate = info_dict.get('tbr') self.to_screen('[%s] Downloading f4m manifest' % self.FD_NAME) - manifest = self.ydl.urlopen(man_url).read() + urlh = self.ydl.urlopen(man_url) + man_url = urlh.geturl() + manifest = urlh.read() - doc = etree.fromstring(manifest) + doc = compat_etree_fromstring(manifest) formats = [(int(f.attrib.get('bitrate', -1)), f) for f in self._get_unencrypted_media(doc)] if requested_bitrate is None: @@ -329,20 +332,25 @@ class F4mFD(FragmentFD): if not live: write_metadata_tag(dest_stream, metadata) + base_url_parsed = compat_urllib_parse_urlparse(base_url) + self._start_frag_download(ctx) frags_filenames = [] while fragments_list: seg_i, frag_i = fragments_list.pop(0) name = 'Seg%d-Frag%d' % (seg_i, frag_i) - url = base_url + name + query = [] + if base_url_parsed.query: + query.append(base_url_parsed.query) if akamai_pv: - url += '?' + akamai_pv.strip(';') + query.append(akamai_pv.strip(';')) if info_dict.get('extra_param_to_segment_url'): - url += info_dict.get('extra_param_to_segment_url') + query.append(info_dict['extra_param_to_segment_url']) + url_parsed = base_url_parsed._replace(path=base_url_parsed.path + name, query='&'.join(query)) frag_filename = '%s-%s' % (ctx['tmpfilename'], name) try: - success = ctx['dl'].download(frag_filename, {'url': url}) + success = ctx['dl'].download(frag_filename, {'url': url_parsed.geturl()}) if not success: return False (down, frag_sanitized) = sanitize_open(frag_filename, 'rb') diff --git a/youtube_dl/downloader/hls.py b/youtube_dl/downloader/hls.py index a62d2047b..b5a3e1167 100644 --- a/youtube_dl/downloader/hls.py +++ b/youtube_dl/downloader/hls.py @@ -13,6 +13,7 @@ from ..utils import ( encodeArgument, encodeFilename, sanitize_open, + handle_youtubedl_headers, ) @@ -30,12 +31,13 @@ class HlsFD(FileDownloader): args = [ffpp.executable, '-y'] - if info_dict['http_headers']: + if info_dict['http_headers'] and re.match(r'^https?://', url): # Trailing \r\n after each HTTP header is important to prevent warning from ffmpeg/avconv: # [http @ 00000000003d2fa0] No trailing CRLF found in HTTP header. + headers = handle_youtubedl_headers(info_dict['http_headers']) args += [ '-headers', - ''.join('%s: %s\r\n' % (key, val) for key, val in info_dict['http_headers'].items())] + ''.join('%s: %s\r\n' % (key, val) for key, val in headers.items())] args += ['-i', url, '-f', 'mp4', '-c', 'copy', '-bsf:a', 'aac_adtstoasc'] diff --git a/youtube_dl/downloader/http.py b/youtube_dl/downloader/http.py index a29f5cf31..56840e026 100644 --- a/youtube_dl/downloader/http.py +++ b/youtube_dl/downloader/http.py @@ -7,14 +7,12 @@ import time import re from .common import FileDownloader -from ..compat import ( - compat_urllib_request, - compat_urllib_error, -) +from ..compat import compat_urllib_error from ..utils import ( ContentTooShortError, encodeFilename, sanitize_open, + sanitized_Request, ) @@ -29,8 +27,8 @@ class HttpFD(FileDownloader): add_headers = info_dict.get('http_headers') if add_headers: headers.update(add_headers) - basic_request = compat_urllib_request.Request(url, None, headers) - request = compat_urllib_request.Request(url, None, headers) + basic_request = sanitized_Request(url, None, headers) + request = sanitized_Request(url, None, headers) is_test = self.params.get('test', False) diff --git a/youtube_dl/downloader/rtmp.py b/youtube_dl/downloader/rtmp.py index f1d219ba9..14d56db47 100644 --- a/youtube_dl/downloader/rtmp.py +++ b/youtube_dl/downloader/rtmp.py @@ -117,7 +117,7 @@ class RtmpFD(FileDownloader): return False # Download using rtmpdump. rtmpdump returns exit code 2 when - # the connection was interrumpted and resuming appears to be + # the connection was interrupted and resuming appears to be # possible. This is part of rtmpdump's normal usage, AFAIK. basic_args = [ 'rtmpdump', '--verbose', '-r', url, diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index f9c40e6cd..5b842204f 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -38,6 +38,7 @@ from .arte import ( ) from .atresplayer import AtresPlayerIE from .atttechchannel import ATTTechChannelIE +from .audimedia import AudiMediaIE from .audiomack import AudiomackIE, AudiomackAlbumIE from .azubu import AzubuIE from .baidu import BaiduVideoIE @@ -45,6 +46,7 @@ from .bambuser import BambuserIE, BambuserChannelIE from .bandcamp import BandcampIE, BandcampAlbumIE from .bbc import ( BBCCoUkIE, + BBCCoUkArticleIE, BBCIE, ) from .beeg import BeegIE @@ -58,7 +60,10 @@ from .bloomberg import BloombergIE from .bpb import BpbIE from .br import BRIE from .breakcom import BreakIE -from .brightcove import BrightcoveIE +from .brightcove import ( + BrightcoveLegacyIE, + BrightcoveNewIE, +) from .buzzfeed import BuzzFeedIE from .byutv import BYUtvIE from .c56 import C56IE @@ -88,6 +93,7 @@ from .cliphunter import CliphunterIE from .clipsyndicate import ClipsyndicateIE from .cloudy import CloudyIE from .clubic import ClubicIE +from .clyp import ClypIE from .cmt import CMTIE from .cnet import CNETIE from .cnn import ( @@ -121,10 +127,12 @@ from .dbtv import DBTVIE from .dcn import DCNIE from .dctp import DctpTvIE from .deezer import DeezerPlaylistIE +from .democracynow import DemocracynowIE from .dfb import DFBIE from .dhm import DHMIE from .dotsub import DotsubIE from .douyutv import DouyuTVIE +from .dplay import DPlayIE from .dramafever import ( DramaFeverIE, DramaFeverSeriesIE, @@ -208,13 +216,15 @@ from .gfycat import GfycatIE from .giantbomb import GiantBombIE from .giga import GigaIE from .glide import GlideIE -from .globo import GloboIE +from .globo import ( + GloboIE, + GloboArticleIE, +) from .godtube import GodTubeIE from .goldenmoustache import GoldenMoustacheIE from .golem import GolemIE from .googleplus import GooglePlusIE from .googlesearch import GoogleSearchIE -from .gorillavid import GorillaVidIE from .goshgay import GoshgayIE from .groupon import GrouponIE from .hark import HarkIE @@ -413,7 +423,10 @@ from .nowness import ( NownessPlaylistIE, NownessSeriesIE, ) -from .nowtv import NowTVIE +from .nowtv import ( + NowTVIE, + NowTVListIE, +) from .nowvideo import NowVideoIE from .npo import ( NPOIE, @@ -451,10 +464,7 @@ from .orf import ( from .parliamentliveuk import ParliamentLiveUKIE from .patreon import PatreonIE from .pbs import PBSIE -from .periscope import ( - PeriscopeIE, - QuickscopeIE, -) +from .periscope import PeriscopeIE from .philharmoniedeparis import PhilharmonieDeParisIE from .phoenix import PhoenixIE from .photobucket import PhotobucketIE @@ -546,6 +556,10 @@ from .shahid import ShahidIE from .shared import SharedIE from .sharesix import ShareSixIE from .sina import SinaIE +from .skynewsarabia import ( + SkyNewsArabiaIE, + SkyNewsArabiaArticleIE, +) from .slideshare import SlideshareIE from .slutload import SlutloadIE from .smotri import ( @@ -568,7 +582,8 @@ from .soundcloud import ( SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE, - SoundcloudPlaylistIE + SoundcloudPlaylistIE, + SoundcloudSearchIE ) from .soundgasm import ( SoundgasmIE, @@ -587,6 +602,7 @@ from .spankwire import SpankwireIE from .spiegel import SpiegelIE, SpiegelArticleIE from .spiegeltv import SpiegeltvIE from .spike import SpikeIE +from .stitcher import StitcherIE from .sport5 import Sport5IE from .sportbox import ( SportBoxIE, @@ -691,7 +707,7 @@ from .twitch import ( TwitchBookmarksIE, TwitchStreamIE, ) -from .twitter import TwitterCardIE +from .twitter import TwitterCardIE, TwitterIE from .ubu import UbuIE from .udemy import ( UdemyIE, @@ -718,7 +734,6 @@ from .vh1 import VH1IE from .vice import ViceIE from .viddler import ViddlerIE from .videodetective import VideoDetectiveIE -from .videolecturesnet import VideoLecturesNetIE from .videofyme import VideofyMeIE from .videomega import VideoMegaIE from .videopremium import VideoPremiumIE @@ -728,6 +743,7 @@ from .vidme import VidmeIE from .vidzi import VidziIE from .vier import VierIE, VierVideosIE from .viewster import ViewsterIE +from .viidea import ViideaIE from .vimeo import ( VimeoIE, VimeoAlbumIE, @@ -780,6 +796,7 @@ from .wrzuta import WrzutaIE from .wsj import WSJIE from .xbef import XBefIE from .xboxclips import XboxClipsIE +from .xfileshare import XFileShareIE from .xhamster import ( XHamsterIE, XHamsterEmbedIE, @@ -823,6 +840,7 @@ from .youtube import ( YoutubeTruncatedIDIE, YoutubeTruncatedURLIE, YoutubeUserIE, + YoutubeUserPlaylistsIE, YoutubeWatchLaterIE, ) from .zapiks import ZapiksIE diff --git a/youtube_dl/extractor/abc.py b/youtube_dl/extractor/abc.py index f9a389f67..c0e5d1abf 100644 --- a/youtube_dl/extractor/abc.py +++ b/youtube_dl/extractor/abc.py @@ -12,7 +12,7 @@ from ..utils import ( class ABCIE(InfoExtractor): IE_NAME = 'abc.net.au' - _VALID_URL = r'http://www\.abc\.net\.au/news/[^/]+/[^/]+/(?P\d+)' + _VALID_URL = r'http://www\.abc\.net\.au/news/(?:[^/]+/){1,2}(?P\d+)' _TESTS = [{ 'url': 'http://www.abc.net.au/news/2014-11-05/australia-to-staff-ebola-treatment-centre-in-sierra-leone/5868334', @@ -36,6 +36,18 @@ class ABCIE(InfoExtractor): 'title': 'Marriage Equality: Warren Entsch introduces same sex marriage bill', }, 'add_ie': ['Youtube'], + }, { + 'url': 'http://www.abc.net.au/news/2015-10-23/nab-lifts-interest-rates-following-westpac-and-cba/6880080', + 'md5': 'b96eee7c9edf4fc5a358a0252881cc1f', + 'info_dict': { + 'id': '6880080', + 'ext': 'mp3', + 'title': 'NAB lifts interest rates, following Westpac and CBA', + 'description': 'md5:f13d8edc81e462fce4a0437c7dc04728', + }, + }, { + 'url': 'http://www.abc.net.au/news/2015-10-19/6866214', + 'only_matching': True, }] def _real_extract(self, url): @@ -43,7 +55,7 @@ class ABCIE(InfoExtractor): webpage = self._download_webpage(url, video_id) mobj = re.search( - r'inline(?PVideo|YouTube)Data\.push\((?P[^)]+)\);', + r'inline(?PVideo|Audio|YouTube)Data\.push\((?P[^)]+)\);', webpage) if mobj is None: raise ExtractorError('Unable to extract video urls') @@ -60,11 +72,13 @@ class ABCIE(InfoExtractor): formats = [{ 'url': url_info['url'], + 'vcodec': url_info.get('codec') if mobj.group('type') == 'Video' else 'none', 'width': int_or_none(url_info.get('width')), 'height': int_or_none(url_info.get('height')), 'tbr': int_or_none(url_info.get('bitrate')), 'filesize': int_or_none(url_info.get('filesize')), } for url_info in urls_info] + self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/adultswim.py b/youtube_dl/extractor/adultswim.py index 130afe791..3ae618e71 100644 --- a/youtube_dl/extractor/adultswim.py +++ b/youtube_dl/extractor/adultswim.py @@ -183,7 +183,7 @@ class AdultSwimIE(InfoExtractor): media_url = file_el.text if determine_ext(media_url) == 'm3u8': formats.extend(self._extract_m3u8_formats( - media_url, segment_title, 'mp4', 'm3u8_native', preference=0, m3u8_id='hls')) + media_url, segment_title, 'mp4', preference=0, m3u8_id='hls')) else: formats.append({ 'format_id': '%s_%s' % (bitrate, ftype), diff --git a/youtube_dl/extractor/aljazeera.py b/youtube_dl/extractor/aljazeera.py index 184a14a4f..5b2c0dc9a 100644 --- a/youtube_dl/extractor/aljazeera.py +++ b/youtube_dl/extractor/aljazeera.py @@ -15,7 +15,7 @@ class AlJazeeraIE(InfoExtractor): 'description': 'As a birth attendant advocating for family planning, Remy is on the frontline of Tondo\'s battle with overcrowding.', 'uploader': 'Al Jazeera English', }, - 'add_ie': ['Brightcove'], + 'add_ie': ['BrightcoveLegacy'], 'skip': 'Not accessible from Travis CI server', } @@ -32,5 +32,5 @@ class AlJazeeraIE(InfoExtractor): 'playerKey=AQ~~%2CAAAAmtVJIFk~%2CTVGOQ5ZTwJbeMWnq5d_H4MOM57xfzApc' '&%40videoPlayer={0}'.format(brightcove_id) ), - 'ie_key': 'Brightcove', + 'ie_key': 'BrightcoveLegacy', } diff --git a/youtube_dl/extractor/anitube.py b/youtube_dl/extractor/anitube.py index 31f0d417c..23f942ae2 100644 --- a/youtube_dl/extractor/anitube.py +++ b/youtube_dl/extractor/anitube.py @@ -26,8 +26,8 @@ class AnitubeIE(InfoExtractor): video_id = mobj.group('id') webpage = self._download_webpage(url, video_id) - key = self._html_search_regex( - r'http://www\.anitube\.se/embed/([A-Za-z0-9_-]*)', webpage, 'key') + key = self._search_regex( + r'src=["\']https?://[^/]+/embed/([A-Za-z0-9_-]+)', webpage, 'key') config_xml = self._download_xml( 'http://www.anitube.se/nuevo/econfig.php?key=%s' % key, key) diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py index 6f465789b..73be6d204 100644 --- a/youtube_dl/extractor/ard.py +++ b/youtube_dl/extractor/ard.py @@ -14,8 +14,8 @@ from ..utils import ( parse_duration, unified_strdate, xpath_text, - parse_xml, ) +from ..compat import compat_etree_fromstring class ARDMediathekIE(InfoExtractor): @@ -161,7 +161,7 @@ class ARDMediathekIE(InfoExtractor): raise ExtractorError('This program is only suitable for those aged 12 and older. Video %s is therefore only available between 20 pm and 6 am.' % video_id, expected=True) if re.search(r'[\?&]rss($|[=&])', url): - doc = parse_xml(webpage) + doc = compat_etree_fromstring(webpage.encode('utf-8')) if doc.tag == 'rss': return GenericIE()._extract_rss(url, video_id, doc) diff --git a/youtube_dl/extractor/atresplayer.py b/youtube_dl/extractor/atresplayer.py index 29f8795d3..50e47ba0a 100644 --- a/youtube_dl/extractor/atresplayer.py +++ b/youtube_dl/extractor/atresplayer.py @@ -7,11 +7,11 @@ from .common import InfoExtractor from ..compat import ( compat_str, compat_urllib_parse, - compat_urllib_request, ) from ..utils import ( int_or_none, float_or_none, + sanitized_Request, xpath_text, ExtractorError, ) @@ -63,7 +63,7 @@ class AtresPlayerIE(InfoExtractor): 'j_password': password, } - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) request.add_header('Content-Type', 'application/x-www-form-urlencoded') response = self._download_webpage( @@ -94,7 +94,7 @@ class AtresPlayerIE(InfoExtractor): formats = [] for fmt in ['windows', 'android_tablet']: - request = compat_urllib_request.Request( + request = sanitized_Request( self._URL_VIDEO_TEMPLATE.format(fmt, episode_id, timestamp_shifted, token)) request.add_header('User-Agent', self._USER_AGENT) diff --git a/youtube_dl/extractor/audimedia.py b/youtube_dl/extractor/audimedia.py new file mode 100644 index 000000000..cad503510 --- /dev/null +++ b/youtube_dl/extractor/audimedia.py @@ -0,0 +1,80 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + int_or_none, + parse_iso8601, + sanitized_Request, +) + + +class AudiMediaIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?audimedia\.tv/(?:en|de)/vid/(?P[^/?#]+)' + _TEST = { + 'url': 'https://audimedia.tv/en/vid/60-seconds-of-audi-sport-104-2015-wec-bahrain-rookie-test', + 'md5': '79a8b71c46d49042609795ab59779b66', + 'info_dict': { + 'id': '1564', + 'ext': 'mp4', + 'title': '60 Seconds of Audi Sport 104/2015 - WEC Bahrain, Rookie Test', + 'description': 'md5:60e5d30a78ced725f7b8d34370762941', + 'upload_date': '20151124', + 'timestamp': 1448354940, + 'duration': 74022, + 'view_count': int, + } + } + # extracted from https://audimedia.tv/assets/embed/embedded-player.js (dataSourceAuthToken) + _AUTH_TOKEN = 'e25b42847dba18c6c8816d5d8ce94c326e06823ebf0859ed164b3ba169be97f2' + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + + raw_payload = self._search_regex(r']+class="amtv-embed"[^>]+id="([^"]+)"', webpage, 'raw payload'); + _, stage_mode, video_id, lang = raw_payload.split('-') + + # TODO: handle s and e stage_mode (live streams and ended live streams) + if stage_mode not in ('s', 'e'): + request = sanitized_Request( + 'https://audimedia.tv/api/video/v1/videos/%s?embed[]=video_versions&embed[]=thumbnail_image&where[content_language_iso]=%s' % (video_id, lang), + headers={'X-Auth-Token': self._AUTH_TOKEN}) + json_data = self._download_json(request, video_id)['results'] + formats = [] + + stream_url_hls = json_data.get('stream_url_hls') + if stream_url_hls: + m3u8_formats = self._extract_m3u8_formats(stream_url_hls, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + + stream_url_hds = json_data.get('stream_url_hds') + if stream_url_hds: + f4m_formats = self._extract_f4m_formats(json_data.get('stream_url_hds') + '?hdcore=3.4.0', video_id, -1, f4m_id='hds', fatal=False) + if f4m_formats: + formats.extend(f4m_formats) + + for video_version in json_data.get('video_versions'): + video_version_url = video_version.get('download_url') or video_version.get('stream_url') + if not video_version_url: + continue + formats.append({ + 'url': video_version_url, + 'width': int_or_none(video_version.get('width')), + 'height': int_or_none(video_version.get('height')), + 'abr': int_or_none(video_version.get('audio_bitrate')), + 'vbr': int_or_none(video_version.get('video_bitrate')), + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': json_data['title'], + 'description': json_data.get('subtitle'), + 'thumbnail': json_data.get('thumbnail_image', {}).get('file'), + 'timestamp': parse_iso8601(json_data.get('publication_date')), + 'duration': int_or_none(json_data.get('duration')), + 'view_count': int_or_none(json_data.get('view_count')), + 'formats': formats, + } diff --git a/youtube_dl/extractor/bambuser.py b/youtube_dl/extractor/bambuser.py index 8dff1d6e3..da986e063 100644 --- a/youtube_dl/extractor/bambuser.py +++ b/youtube_dl/extractor/bambuser.py @@ -6,13 +6,13 @@ import itertools from .common import InfoExtractor from ..compat import ( compat_urllib_parse, - compat_urllib_request, compat_str, ) from ..utils import ( ExtractorError, int_or_none, float_or_none, + sanitized_Request, ) @@ -57,7 +57,7 @@ class BambuserIE(InfoExtractor): 'pass': password, } - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) request.add_header('Referer', self._LOGIN_URL) response = self._download_webpage( @@ -126,7 +126,7 @@ class BambuserChannelIE(InfoExtractor): '&sort=created&access_mode=0%2C1%2C2&limit={count}' '&method=broadcast&format=json&vid_older_than={last}' ).format(user=user, count=self._STEP, last=last_id) - req = compat_urllib_request.Request(req_url) + req = sanitized_Request(req_url) # Without setting this header, we wouldn't get any result req.add_header('Referer', 'http://bambuser.com/channel/%s' % user) data = self._download_json( diff --git a/youtube_dl/extractor/bandcamp.py b/youtube_dl/extractor/bandcamp.py index a27f3e748..c1ef8051d 100644 --- a/youtube_dl/extractor/bandcamp.py +++ b/youtube_dl/extractor/bandcamp.py @@ -10,6 +10,8 @@ from ..compat import ( ) from ..utils import ( ExtractorError, + float_or_none, + int_or_none, ) @@ -52,11 +54,11 @@ class BandcampIE(InfoExtractor): ext, abr_str = format_id.split('-', 1) formats.append({ 'format_id': format_id, - 'url': format_url, + 'url': self._proto_relative_url(format_url, 'http:'), 'ext': ext, 'vcodec': 'none', 'acodec': ext, - 'abr': int(abr_str), + 'abr': int_or_none(abr_str), }) self._sort_formats(formats) @@ -65,7 +67,7 @@ class BandcampIE(InfoExtractor): 'id': compat_str(data['id']), 'title': data['title'], 'formats': formats, - 'duration': float(data['duration']), + 'duration': float_or_none(data.get('duration')), } else: raise ExtractorError('No free songs found') diff --git a/youtube_dl/extractor/bbc.py b/youtube_dl/extractor/bbc.py index 68995f81e..7fb80aa38 100644 --- a/youtube_dl/extractor/bbc.py +++ b/youtube_dl/extractor/bbc.py @@ -2,7 +2,6 @@ from __future__ import unicode_literals import re -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( @@ -14,18 +13,22 @@ from ..utils import ( remove_end, unescapeHTML, ) -from ..compat import compat_HTTPError +from ..compat import ( + compat_etree_fromstring, + compat_HTTPError, +) class BBCCoUkIE(InfoExtractor): IE_NAME = 'bbc.co.uk' IE_DESC = 'BBC iPlayer' - _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P[\da-z]{8})' + _ID_REGEX = r'[pb][\da-z]{7}' + _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:programmes/(?!articles/)|iplayer(?:/[^/]+)?/(?:episode/|playlist/))|music/clips[/#])(?P%s)' % _ID_REGEX _MEDIASELECTOR_URLS = [ # Provides HQ HLS streams with even better quality that pc mediaset but fails # with geolocation in some cases when it's even not geo restricted at all (e.g. - # http://www.bbc.co.uk/programmes/b06bp7lf) + # http://www.bbc.co.uk/programmes/b06bp7lf). Also may fail with selectionunavailable. 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s', 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s', ] @@ -332,7 +335,7 @@ class BBCCoUkIE(InfoExtractor): return self._download_media_selector_url( mediaselector_url % programme_id, programme_id) except BBCCoUkIE.MediaSelectionError as e: - if e.id in ('notukerror', 'geolocation'): + if e.id in ('notukerror', 'geolocation', 'selectionunavailable'): last_exception = e continue self._raise_extractor_error(e) @@ -343,8 +346,8 @@ class BBCCoUkIE(InfoExtractor): media_selection = self._download_xml( url, programme_id, 'Downloading media selection XML') except ExtractorError as ee: - if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403: - media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8')) + if isinstance(ee.cause, compat_HTTPError) and ee.cause.code in (403, 404): + media_selection = compat_etree_fromstring(ee.cause.read().decode('utf-8')) else: raise return self._process_media_selector(media_selection, programme_id) @@ -421,7 +424,7 @@ class BBCCoUkIE(InfoExtractor): continue title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS) - description = description_el.text if description_el else None + description = description_el.text if description_el is not None else None def get_programme_id(item): def get_from_attributes(item): @@ -463,7 +466,7 @@ class BBCCoUkIE(InfoExtractor): if not programme_id: programme_id = self._search_regex( - r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None) + r'"vpid"\s*:\s*"(%s)"' % self._ID_REGEX, webpage, 'vpid', fatal=False, default=None) if programme_id: formats, subtitles = self._download_media_selector(programme_id) @@ -625,6 +628,7 @@ class BBCIE(BBCCoUkIE): 'id': 'p02xycnp', 'ext': 'mp4', 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?', + 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.', 'duration': 140, }, 'params': { @@ -651,7 +655,7 @@ class BBCIE(BBCCoUkIE): @classmethod def suitable(cls, url): - return False if BBCCoUkIE.suitable(url) else super(BBCIE, cls).suitable(url) + return False if BBCCoUkIE.suitable(url) or BBCCoUkArticleIE.suitable(url) else super(BBCIE, cls).suitable(url) def _extract_from_media_meta(self, media_meta, video_id): # Direct links to media in media metadata (e.g. @@ -777,8 +781,9 @@ class BBCIE(BBCCoUkIE): # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret) programme_id = self._search_regex( - [r'data-video-player-vpid="([\da-z]{8})"', - r']+name="externalIdentifier"[^>]+value="([\da-z]{8})"'], + [r'data-video-player-vpid="(%s)"' % self._ID_REGEX, + r']+name="externalIdentifier"[^>]+value="(%s)"' % self._ID_REGEX, + r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX], webpage, 'vpid', default=None) if programme_id: @@ -813,7 +818,7 @@ class BBCIE(BBCCoUkIE): # Multiple video article (e.g. # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460) - EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+[\da-z]{8}(?:\b[^"]+)?' + EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX entries = [] for match in extract_all(r'new\s+SMP\(({.+?})\)'): embed_url = match.get('playerSettings', {}).get('externalEmbedUrl') @@ -902,3 +907,33 @@ class BBCIE(BBCCoUkIE): }) return self.playlist_result(entries, playlist_id, playlist_title, playlist_description) + + +class BBCCoUkArticleIE(InfoExtractor): + _VALID_URL = 'http://www.bbc.co.uk/programmes/articles/(?P[a-zA-Z0-9]+)' + IE_NAME = 'bbc.co.uk:article' + IE_DESC = 'BBC articles' + + _TEST = { + 'url': 'http://www.bbc.co.uk/programmes/articles/3jNQLTMrPlYGTBn0WV6M2MS/not-your-typical-role-model-ada-lovelace-the-19th-century-programmer', + 'info_dict': { + 'id': '3jNQLTMrPlYGTBn0WV6M2MS', + 'title': 'Calculating Ada: The Countess of Computing - Not your typical role model: Ada Lovelace the 19th century programmer - BBC Four', + 'description': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming.', + }, + 'playlist_count': 4, + 'add_ie': ['BBCCoUk'], + } + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + title = self._og_search_title(webpage) + description = self._og_search_description(webpage).strip() + + entries = [self.url_result(programme_url) for programme_url in re.findall( + r']+typeof="Clip"[^>]+resource="([^"]+)"', webpage)] + + return self.playlist_result(entries, playlist_id, title, description) diff --git a/youtube_dl/extractor/beeg.py b/youtube_dl/extractor/beeg.py index e6c928699..e63c2ac00 100644 --- a/youtube_dl/extractor/beeg.py +++ b/youtube_dl/extractor/beeg.py @@ -1,6 +1,11 @@ from __future__ import unicode_literals from .common import InfoExtractor +from ..compat import ( + compat_chr, + compat_ord, + compat_urllib_parse_unquote, +) from ..utils import ( int_or_none, parse_iso8601, @@ -29,16 +34,35 @@ class BeegIE(InfoExtractor): video_id = self._match_id(url) video = self._download_json( - 'http://beeg.com/api/v1/video/%s' % video_id, video_id) + 'http://beeg.com/api/v3/video/%s' % video_id, video_id) + + def decrypt_key(key): + # Reverse engineered from http://static.beeg.com/cpl/1067.js + a = '8RPUUCS35ZWp3ADnKcSmpH71ZusrROo' + e = compat_urllib_parse_unquote(key) + return ''.join([ + compat_chr(compat_ord(e[n]) - compat_ord(a[n % len(a)]) % 25) + for n in range(len(e))]) + + def decrypt_url(encrypted_url): + encrypted_url = self._proto_relative_url( + encrypted_url.replace('{DATA_MARKERS}', ''), 'http:') + key = self._search_regex( + r'/key=(.*?)%2Cend=', encrypted_url, 'key', default=None) + if not key: + return encrypted_url + return encrypted_url.replace(key, decrypt_key(key)) formats = [] for format_id, video_url in video.items(): + if not video_url: + continue height = self._search_regex( r'^(\d+)[pP]$', format_id, 'height', default=None) if not height: continue formats.append({ - 'url': self._proto_relative_url(video_url.replace('{DATA_MARKERS}', ''), 'http:'), + 'url': decrypt_url(video_url), 'format_id': format_id, 'height': int(height), }) diff --git a/youtube_dl/extractor/bilibili.py b/youtube_dl/extractor/bilibili.py index ecc17ebeb..1c3644587 100644 --- a/youtube_dl/extractor/bilibili.py +++ b/youtube_dl/extractor/bilibili.py @@ -2,141 +2,116 @@ from __future__ import unicode_literals import re -import itertools import json -import xml.etree.ElementTree as ET from .common import InfoExtractor +from ..compat import ( + compat_etree_fromstring, +) from ..utils import ( int_or_none, - unified_strdate, + unescapeHTML, ExtractorError, ) class BiliBiliIE(InfoExtractor): - _VALID_URL = r'http://www\.bilibili\.(?:tv|com)/video/av(?P[0-9]+)/' + _VALID_URL = r'http://www\.bilibili\.(?:tv|com)/video/av(?P\d+)(?:/index_(?P\d+).html)?' _TESTS = [{ 'url': 'http://www.bilibili.tv/video/av1074402/', 'md5': '2c301e4dab317596e837c3e7633e7d86', 'info_dict': { - 'id': '1074402_part1', + 'id': '1554319', 'ext': 'flv', 'title': '【金坷垃】金泡沫', - 'duration': 308, + 'duration': 308313, 'upload_date': '20140420', 'thumbnail': 're:^https?://.+\.jpg', + 'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923', + 'timestamp': 1397983878, + 'uploader': '菊子桑', }, }, { 'url': 'http://www.bilibili.com/video/av1041170/', 'info_dict': { 'id': '1041170', 'title': '【BD1080P】刀语【诸神&异域】', + 'description': '这是个神奇的故事~每个人不留弹幕不给走哦~切利哦!~', + 'uploader': '枫叶逝去', + 'timestamp': 1396501299, }, 'playlist_count': 9, }] def _real_extract(self, url): - video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + page_num = mobj.group('page_num') or '1' - if '(此视频不存在或被删除)' in webpage: - raise ExtractorError( - 'The video does not exist or was deleted', expected=True) + view_data = self._download_json( + 'http://api.bilibili.com/view?type=json&appkey=8e9fc618fbd41e28&id=%s&page=%s' % (video_id, page_num), + video_id) + if 'error' in view_data: + raise ExtractorError('%s said: %s' % (self.IE_NAME, view_data['error']), expected=True) - if '>你没有权限浏览! 由于版权相关问题 我们不对您所在的地区提供服务<' in webpage: - raise ExtractorError( - 'The video is not available in your region due to copyright reasons', - expected=True) + cid = view_data['cid'] + title = unescapeHTML(view_data['title']) - video_code = self._search_regex( - r'(?s)
(.*?)
', webpage, 'video code') - - title = self._html_search_meta( - 'media:title', video_code, 'title', fatal=True) - duration_str = self._html_search_meta( - 'duration', video_code, 'duration') - if duration_str is None: - duration = None - else: - duration_mobj = re.match( - r'^T(?:(?P[0-9]+)H)?(?P[0-9]+)M(?P[0-9]+)S$', - duration_str) - duration = ( - int_or_none(duration_mobj.group('hours'), default=0) * 3600 + - int(duration_mobj.group('minutes')) * 60 + - int(duration_mobj.group('seconds'))) - upload_date = unified_strdate(self._html_search_meta( - 'uploadDate', video_code, fatal=False)) - thumbnail = self._html_search_meta( - 'thumbnailUrl', video_code, 'thumbnail', fatal=False) - - cid = self._search_regex(r'cid=(\d+)', webpage, 'cid') - - entries = [] - - lq_page = self._download_webpage( - 'http://interface.bilibili.com/v_cdn_play?appkey=1&cid=%s' % cid, - video_id, - note='Downloading LQ video info' + page = self._download_webpage( + 'http://interface.bilibili.com/v_cdn_play?appkey=8e9fc618fbd41e28&cid=%s' % cid, + cid, + 'Downloading page %s/%s' % (page_num, view_data['pages']) ) try: - err_info = json.loads(lq_page) + err_info = json.loads(page) raise ExtractorError( 'BiliBili said: ' + err_info['error_text'], expected=True) except ValueError: pass - lq_doc = ET.fromstring(lq_page) - lq_durls = lq_doc.findall('./durl') + doc = compat_etree_fromstring(page) - hq_doc = self._download_xml( - 'http://interface.bilibili.com/playurl?appkey=1&cid=%s' % cid, - video_id, - note='Downloading HQ video info', - fatal=False, - ) - if hq_doc is not False: - hq_durls = hq_doc.findall('./durl') - assert len(lq_durls) == len(hq_durls) - else: - hq_durls = itertools.repeat(None) + entries = [] - i = 1 - for lq_durl, hq_durl in zip(lq_durls, hq_durls): + for durl in doc.findall('./durl'): + size = durl.find('./filesize|./size') formats = [{ - 'format_id': 'lq', - 'quality': 1, - 'url': lq_durl.find('./url').text, - 'filesize': int_or_none( - lq_durl.find('./size'), get_attr='text'), + 'url': durl.find('./url').text, + 'filesize': int_or_none(size.text) if size else None, + 'ext': 'flv', }] - if hq_durl is not None: - formats.append({ - 'format_id': 'hq', - 'quality': 2, - 'ext': 'flv', - 'url': hq_durl.find('./url').text, - 'filesize': int_or_none( - hq_durl.find('./size'), get_attr='text'), - }) - self._sort_formats(formats) + backup_urls = durl.find('./backup_url') + if backup_urls is not None: + for backup_url in backup_urls.findall('./url'): + formats.append({'url': backup_url.text}) + formats.reverse() entries.append({ - 'id': '%s_part%d' % (video_id, i), + 'id': '%s_part%s' % (cid, durl.find('./order').text), 'title': title, + 'duration': int_or_none(durl.find('./length').text) // 1000, 'formats': formats, - 'duration': duration, - 'upload_date': upload_date, - 'thumbnail': thumbnail, }) - i += 1 - - return { - '_type': 'multi_video', - 'entries': entries, - 'id': video_id, - 'title': title + info = { + 'id': str(cid), + 'title': title, + 'description': view_data.get('description'), + 'thumbnail': view_data.get('pic'), + 'uploader': view_data.get('author'), + 'timestamp': int_or_none(view_data.get('created')), + 'view_count': view_data.get('play'), + 'duration': int_or_none(doc.find('./timelength').text), } + + if len(entries) == 1: + entries[0].update(info) + return entries[0] + else: + info.update({ + '_type': 'multi_video', + 'id': video_id, + 'entries': entries, + }) + return info diff --git a/youtube_dl/extractor/bloomberg.py b/youtube_dl/extractor/bloomberg.py index 0dca29b71..ebeef8f2a 100644 --- a/youtube_dl/extractor/bloomberg.py +++ b/youtube_dl/extractor/bloomberg.py @@ -6,9 +6,9 @@ from .common import InfoExtractor class BloombergIE(InfoExtractor): - _VALID_URL = r'https?://www\.bloomberg\.com/news/videos/[^/]+/(?P[^/?#]+)' + _VALID_URL = r'https?://(?:www\.)?bloomberg\.com/(?:[^/]+/)*(?P[^/?#]+)' - _TEST = { + _TESTS = [{ 'url': 'http://www.bloomberg.com/news/videos/b/aaeae121-5949-481e-a1ce-4562db6f5df2', # The md5 checksum changes 'info_dict': { @@ -17,22 +17,39 @@ class BloombergIE(InfoExtractor): 'title': 'Shah\'s Presentation on Foreign-Exchange Strategies', 'description': 'md5:a8ba0302912d03d246979735c17d2761', }, - } + }, { + 'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets', + 'only_matching': True, + }, { + 'url': 'http://www.bloomberg.com/politics/videos/2015-11-25/karl-rove-on-jeb-bush-s-struggles-stopping-trump', + 'only_matching': True, + }] def _real_extract(self, url): name = self._match_id(url) webpage = self._download_webpage(url, name) - video_id = self._search_regex(r'"bmmrId":"(.+?)"', webpage, 'id') + video_id = self._search_regex( + r'["\']bmmrId["\']\s*:\s*(["\'])(?P.+?)\1', + webpage, 'id', group='url') title = re.sub(': Video$', '', self._og_search_title(webpage)) embed_info = self._download_json( 'http://www.bloomberg.com/api/embed?id=%s' % video_id, video_id) formats = [] for stream in embed_info['streams']: - if stream["muxing_format"] == "TS": - formats.extend(self._extract_m3u8_formats(stream['url'], video_id)) + stream_url = stream.get('url') + if not stream_url: + continue + if stream['muxing_format'] == 'TS': + m3u8_formats = self._extract_m3u8_formats( + stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) else: - formats.extend(self._extract_f4m_formats(stream['url'], video_id)) + f4m_formats = self._extract_f4m_formats( + stream_url, video_id, f4m_id='hds', fatal=False) + if f4m_formats: + formats.extend(f4m_formats) self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py index 4721c2293..f5ebae1e6 100644 --- a/youtube_dl/extractor/brightcove.py +++ b/youtube_dl/extractor/brightcove.py @@ -3,15 +3,14 @@ from __future__ import unicode_literals import re import json -import xml.etree.ElementTree from .common import InfoExtractor from ..compat import ( + compat_etree_fromstring, compat_parse_qs, compat_str, compat_urllib_parse, compat_urllib_parse_urlparse, - compat_urllib_request, compat_urlparse, compat_xml_parse_error, ) @@ -20,12 +19,18 @@ from ..utils import ( ExtractorError, find_xpath_attr, fix_xml_ampersands, + float_or_none, + js_to_json, + int_or_none, + parse_iso8601, + sanitized_Request, unescapeHTML, unsmuggle_url, ) -class BrightcoveIE(InfoExtractor): +class BrightcoveLegacyIE(InfoExtractor): + IE_NAME = 'brightcove:legacy' _VALID_URL = r'(?:https?://.*brightcove\.com/(services|viewer).*?\?|brightcove:)(?P.*)' _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s' @@ -119,7 +124,7 @@ class BrightcoveIE(InfoExtractor): object_str = fix_xml_ampersands(object_str) try: - object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8')) + object_doc = compat_etree_fromstring(object_str.encode('utf-8')) except compat_xml_parse_error: return @@ -245,7 +250,7 @@ class BrightcoveIE(InfoExtractor): def _get_video_info(self, video_id, query_str, query, referer=None): request_url = self._FEDERATED_URL_TEMPLATE % query_str - req = compat_urllib_request.Request(request_url) + req = sanitized_Request(request_url) linkBase = query.get('linkBaseURL') if linkBase is not None: referer = linkBase[0] @@ -346,3 +351,172 @@ class BrightcoveIE(InfoExtractor): if 'url' not in info and not info.get('formats'): raise ExtractorError('Unable to extract video url for %s' % info['id']) return info + + +class BrightcoveNewIE(InfoExtractor): + IE_NAME = 'brightcove:new' + _VALID_URL = r'https?://players\.brightcove\.net/(?P\d+)/(?P[^/]+)_(?P[^/]+)/index\.html\?.*videoId=(?P\d+)' + _TESTS = [{ + 'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001', + 'md5': 'c8100925723840d4b0d243f7025703be', + 'info_dict': { + 'id': '4463358922001', + 'ext': 'mp4', + 'title': 'Meet the man behind Popcorn Time', + 'description': 'md5:eac376a4fe366edc70279bfb681aea16', + 'duration': 165.768, + 'timestamp': 1441391203, + 'upload_date': '20150904', + 'uploader_id': '929656772001', + 'formats': 'mincount:22', + }, + }, { + # with rtmp streams + 'url': 'http://players.brightcove.net/4036320279001/5d112ed9-283f-485f-a7f9-33f42e8bc042_default/index.html?videoId=4279049078001', + 'info_dict': { + 'id': '4279049078001', + 'ext': 'mp4', + 'title': 'Titansgrave: Chapter 0', + 'description': 'Titansgrave: Chapter 0', + 'duration': 1242.058, + 'timestamp': 1433556729, + 'upload_date': '20150606', + 'uploader_id': '4036320279001', + 'formats': 'mincount:41', + }, + 'params': { + 'skip_download': True, + } + }] + + @staticmethod + def _extract_urls(webpage): + # Reference: + # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe + # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript) + # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/embed-in-page.html + + entries = [] + + # Look for iframe embeds [1] + for _, url in re.findall( + r']+src=(["\'])((?:https?:)//players\.brightcove\.net/\d+/[^/]+/index\.html.+?)\1', webpage): + entries.append(url) + + # Look for embed_in_page embeds [2] + for video_id, account_id, player_id, embed in re.findall( + # According to examples from [3] it's unclear whether video id + # may be optional and what to do when it is + r'''(?sx) + ]+ + data-video-id=["\'](\d+)["\'][^>]*>.*? + .*? + ]+ + src=["\'](?:https?:)?//players\.brightcove\.net/ + (\d+)/([\da-f-]+)_([^/]+)/index\.min\.js + ''', webpage): + entries.append( + 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s' + % (account_id, player_id, embed, video_id)) + + return entries + + def _real_extract(self, url): + account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups() + + webpage = self._download_webpage( + 'http://players.brightcove.net/%s/%s_%s/index.min.js' + % (account_id, player_id, embed), video_id) + + policy_key = None + + catalog = self._search_regex( + r'catalog\(({.+?})\);', webpage, 'catalog', default=None) + if catalog: + catalog = self._parse_json( + js_to_json(catalog), video_id, fatal=False) + if catalog: + policy_key = catalog.get('policyKey') + + if not policy_key: + policy_key = self._search_regex( + r'policyKey\s*:\s*(["\'])(?P.+?)\1', + webpage, 'policy key', group='pk') + + req = sanitized_Request( + 'https://edge.api.brightcove.com/playback/v1/accounts/%s/videos/%s' + % (account_id, video_id), + headers={'Accept': 'application/json;pk=%s' % policy_key}) + json_data = self._download_json(req, video_id) + + title = json_data['name'] + + formats = [] + for source in json_data.get('sources', []): + source_type = source.get('type') + src = source.get('src') + if source_type == 'application/x-mpegURL': + if not src: + continue + m3u8_formats = self._extract_m3u8_formats( + src, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + else: + streaming_src = source.get('streaming_src') + stream_name, app_name = source.get('stream_name'), source.get('app_name') + if not src and not streaming_src and (not stream_name or not app_name): + continue + tbr = float_or_none(source.get('avg_bitrate'), 1000) + height = int_or_none(source.get('height')) + f = { + 'tbr': tbr, + 'width': int_or_none(source.get('width')), + 'height': height, + 'filesize': int_or_none(source.get('size')), + 'container': source.get('container'), + 'vcodec': source.get('codec'), + 'ext': source.get('container').lower(), + } + + def build_format_id(kind): + format_id = kind + if tbr: + format_id += '-%dk' % int(tbr) + if height: + format_id += '-%dp' % height + return format_id + + if src or streaming_src: + f.update({ + 'url': src or streaming_src, + 'format_id': build_format_id('http' if src else 'http-streaming'), + 'preference': 2 if src else 1, + }) + else: + f.update({ + 'url': app_name, + 'play_path': stream_name, + 'format_id': build_format_id('rtmp'), + }) + formats.append(f) + self._sort_formats(formats) + + description = json_data.get('description') + thumbnail = json_data.get('thumbnail') + timestamp = parse_iso8601(json_data.get('published_at')) + duration = float_or_none(json_data.get('duration'), 1000) + tags = json_data.get('tags', []) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'timestamp': timestamp, + 'uploader_id': account_id, + 'formats': formats, + 'tags': tags, + } diff --git a/youtube_dl/extractor/canalc2.py b/youtube_dl/extractor/canalc2.py index c4fefefe4..f6a1ff381 100644 --- a/youtube_dl/extractor/canalc2.py +++ b/youtube_dl/extractor/canalc2.py @@ -4,38 +4,53 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..utils import parse_duration class Canalc2IE(InfoExtractor): IE_NAME = 'canalc2.tv' - _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?.*?idVideo=(?P\d+)' + _VALID_URL = r'https?://(?:www\.)?canalc2\.tv/video/(?P\d+)' _TEST = { - 'url': 'http://www.canalc2.tv/video.asp?idVideo=12163&voir=oui', + 'url': 'http://www.canalc2.tv/video/12163', 'md5': '060158428b650f896c542dfbb3d6487f', 'info_dict': { 'id': '12163', - 'ext': 'mp4', - 'title': 'Terrasses du Numérique' + 'ext': 'flv', + 'title': 'Terrasses du Numérique', + 'duration': 122, + }, + 'params': { + 'skip_download': True, # Requires rtmpdump } } def _real_extract(self, url): - video_id = re.match(self._VALID_URL, url).group('id') - # We need to set the voir field for getting the file name - url = 'http://www.canalc2.tv/video.asp?idVideo=%s&voir=oui' % video_id + video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - file_name = self._search_regex( - r"so\.addVariable\('file','(.*?)'\);", - webpage, 'file name') - video_url = 'http://vod-flash.u-strasbg.fr:8080/' + file_name + video_url = self._search_regex( + r'jwplayer\((["\'])Player\1\)\.setup\({[^}]*file\s*:\s*(["\'])(?P.+?)\2', + webpage, 'video_url', group='file') + formats = [{'url': video_url}] + if video_url.startswith('rtmp://'): + rtmp = re.search(r'^(?Prtmp://[^/]+/(?P.+/))(?Pmp4:.+)$', video_url) + formats[0].update({ + 'url': rtmp.group('url'), + 'ext': 'flv', + 'app': rtmp.group('app'), + 'play_path': rtmp.group('play_path'), + 'page_url': url, + }) title = self._html_search_regex( - r'class="evenement8">(.*?)', webpage, 'title') + r'(?s)class="[^"]*col_description[^"]*">.*?

(.*?)

', webpage, 'title') + duration = parse_duration(self._search_regex( + r'id=["\']video_duree["\'][^>]*>([^<]+)', + webpage, 'duration', fatal=False)) return { 'id': video_id, - 'ext': 'mp4', - 'url': video_url, 'title': title, + 'duration': duration, + 'formats': formats, } diff --git a/youtube_dl/extractor/cbs.py b/youtube_dl/extractor/cbs.py index 75fffb156..40d07ab18 100644 --- a/youtube_dl/extractor/cbs.py +++ b/youtube_dl/extractor/cbs.py @@ -1,6 +1,10 @@ from __future__ import unicode_literals from .common import InfoExtractor +from ..utils import ( + sanitized_Request, + smuggle_url, +) class CBSIE(InfoExtractor): @@ -46,13 +50,19 @@ class CBSIE(InfoExtractor): def _real_extract(self, url): display_id = self._match_id(url) - webpage = self._download_webpage(url, display_id) + request = sanitized_Request(url) + # Android UA is served with higher quality (720p) streams (see + # https://github.com/rg3/youtube-dl/issues/7490) + request.add_header('User-Agent', 'Mozilla/5.0 (Linux; Android 4.4; Nexus 5)') + webpage = self._download_webpage(request, display_id) real_id = self._search_regex( [r"video\.settings\.pid\s*=\s*'([^']+)';", r"cbsplayer\.pid\s*=\s*'([^']+)';"], webpage, 'real video ID') return { '_type': 'url_transparent', 'ie_key': 'ThePlatform', - 'url': 'theplatform:%s' % real_id, + 'url': smuggle_url( + 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true&manifest=m3u' % real_id, + {'force_smil_url': True}), 'display_id': display_id, } diff --git a/youtube_dl/extractor/cbsnews.py b/youtube_dl/extractor/cbsnews.py index 52e61d85b..f9a64a0a2 100644 --- a/youtube_dl/extractor/cbsnews.py +++ b/youtube_dl/extractor/cbsnews.py @@ -67,9 +67,12 @@ class CBSNewsIE(InfoExtractor): 'format_id': format_id, } if uri.startswith('rtmp'): + play_path = re.sub( + r'{slistFilePath}', '', + uri.split('')[-1].split('{break}')[-1]) fmt.update({ 'app': 'ondemand?auth=cbs', - 'play_path': 'mp4:' + uri.split('')[-1], + 'play_path': 'mp4:' + play_path, 'player_url': 'http://www.cbsnews.com/[[IMPORT]]/vidtech.cbsinteractive.com/player/3_3_0/CBSI_PLAYER_HD.swf', 'page_url': 'http://www.cbsnews.com', 'ext': 'flv', diff --git a/youtube_dl/extractor/ceskatelevize.py b/youtube_dl/extractor/ceskatelevize.py index e857e66f4..6f7b2a70d 100644 --- a/youtube_dl/extractor/ceskatelevize.py +++ b/youtube_dl/extractor/ceskatelevize.py @@ -5,7 +5,6 @@ import re from .common import InfoExtractor from ..compat import ( - compat_urllib_request, compat_urllib_parse, compat_urllib_parse_unquote, compat_urllib_parse_urlparse, @@ -13,6 +12,7 @@ from ..compat import ( from ..utils import ( ExtractorError, float_or_none, + sanitized_Request, ) @@ -100,7 +100,7 @@ class CeskaTelevizeIE(InfoExtractor): 'requestSource': 'iVysilani', } - req = compat_urllib_request.Request( + req = sanitized_Request( 'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist', data=compat_urllib_parse.urlencode(data)) @@ -115,7 +115,7 @@ class CeskaTelevizeIE(InfoExtractor): if playlist_url == 'error_region': raise ExtractorError(NOT_AVAILABLE_STRING, expected=True) - req = compat_urllib_request.Request(compat_urllib_parse_unquote(playlist_url)) + req = sanitized_Request(compat_urllib_parse_unquote(playlist_url)) req.add_header('Referer', url) playlist_title = self._og_search_title(webpage) diff --git a/youtube_dl/extractor/clyp.py b/youtube_dl/extractor/clyp.py new file mode 100644 index 000000000..57e643799 --- /dev/null +++ b/youtube_dl/extractor/clyp.py @@ -0,0 +1,57 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + float_or_none, + parse_iso8601, +) + + +class ClypIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?clyp\.it/(?P[a-z0-9]+)' + _TEST = { + 'url': 'https://clyp.it/ojz2wfah', + 'md5': '1d4961036c41247ecfdcc439c0cddcbb', + 'info_dict': { + 'id': 'ojz2wfah', + 'ext': 'mp3', + 'title': 'Krisson80 - bits wip wip', + 'description': '#Krisson80BitsWipWip #chiptune\n#wip', + 'duration': 263.21, + 'timestamp': 1443515251, + 'upload_date': '20150929', + }, + } + + def _real_extract(self, url): + audio_id = self._match_id(url) + + metadata = self._download_json( + 'https://api.clyp.it/%s' % audio_id, audio_id) + + formats = [] + for secure in ('', 'Secure'): + for ext in ('Ogg', 'Mp3'): + format_id = '%s%s' % (secure, ext) + format_url = metadata.get('%sUrl' % format_id) + if format_url: + formats.append({ + 'url': format_url, + 'format_id': format_id, + 'vcodec': 'none', + }) + self._sort_formats(formats) + + title = metadata['Title'] + description = metadata.get('Description') + duration = float_or_none(metadata.get('Duration')) + timestamp = parse_iso8601(metadata.get('DateCreated')) + + return { + 'id': audio_id, + 'title': title, + 'description': description, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + } diff --git a/youtube_dl/extractor/cmt.py b/youtube_dl/extractor/cmt.py index e96c59f71..f1311b14f 100644 --- a/youtube_dl/extractor/cmt.py +++ b/youtube_dl/extractor/cmt.py @@ -4,7 +4,7 @@ from .mtv import MTVIE class CMTIE(MTVIE): IE_NAME = 'cmt.com' - _VALID_URL = r'https?://www\.cmt\.com/videos/.+?/(?P[^/]+)\.jhtml' + _VALID_URL = r'https?://www\.cmt\.com/(?:videos|shows)/(?:[^/]+/)*(?P\d+)' _FEED_URL = 'http://www.cmt.com/sitewide/apps/player/embed/rss/' _TESTS = [{ @@ -16,4 +16,7 @@ class CMTIE(MTVIE): 'title': 'Garth Brooks - "The Call (featuring Trisha Yearwood)"', 'description': 'Blame It All On My Roots', }, + }, { + 'url': 'http://www.cmt.com/shows/party-down-south/party-down-south-ep-407-gone-girl/1738172/playlist/#id=1738172', + 'only_matching': True, }] diff --git a/youtube_dl/extractor/collegerama.py b/youtube_dl/extractor/collegerama.py index fedd48490..40667a0f1 100644 --- a/youtube_dl/extractor/collegerama.py +++ b/youtube_dl/extractor/collegerama.py @@ -3,10 +3,10 @@ from __future__ import unicode_literals import json from .common import InfoExtractor -from ..compat import compat_urllib_request from ..utils import ( float_or_none, int_or_none, + sanitized_Request, ) @@ -52,7 +52,7 @@ class CollegeRamaIE(InfoExtractor): } } - request = compat_urllib_request.Request( + request = sanitized_Request( 'http://collegerama.tudelft.nl/Mediasite/PlayerService/PlayerService.svc/json/GetPlayerOptions', json.dumps(player_options_request)) request.add_header('Content-Type', 'application/json') diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index a0c4af92f..6ab2d68d6 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -10,20 +10,18 @@ import re import socket import sys import time -import xml.etree.ElementTree from ..compat import ( compat_cookiejar, compat_cookies, compat_getpass, - compat_HTTPError, compat_http_client, compat_urllib_error, compat_urllib_parse, compat_urllib_parse_urlparse, - compat_urllib_request, compat_urlparse, compat_str, + compat_etree_fromstring, ) from ..utils import ( NO_DEFAULT, @@ -38,6 +36,7 @@ from ..utils import ( int_or_none, RegexNotFoundError, sanitize_filename, + sanitized_Request, unescapeHTML, unified_strdate, url_basename, @@ -168,10 +167,11 @@ class InfoExtractor(object): "ext" will be calculated from URL if missing automatic_captions: Like 'subtitles', used by the YoutubeIE for automatically generated captions - duration: Length of the video in seconds, as an integer. + duration: Length of the video in seconds, as an integer or float. view_count: How many users have watched the video on the platform. like_count: Number of positive ratings of the video dislike_count: Number of negative ratings of the video + repost_count: Number of reposts of the video average_rating: Average rating give by users, the scale used depends on the webpage comment_count: Number of comments on the video comments: A list of comments, each with one or more of the following @@ -310,11 +310,11 @@ class InfoExtractor(object): @classmethod def ie_key(cls): """A string for getting the InfoExtractor with get_info_extractor""" - return cls.__name__[:-2] + return compat_str(cls.__name__[:-2]) @property def IE_NAME(self): - return type(self).__name__[:-2] + return compat_str(type(self).__name__[:-2]) def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): """ Returns the response handle """ @@ -461,7 +461,7 @@ class InfoExtractor(object): return xml_string if transform_source: xml_string = transform_source(xml_string) - return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8')) + return compat_etree_fromstring(xml_string.encode('utf-8')) def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata', @@ -645,7 +645,7 @@ class InfoExtractor(object): # Helper functions for extracting OpenGraph info @staticmethod def _og_regexes(prop): - content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\'|\s*([^\s"\'=<>`]+?))' + content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))' property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)' % {'prop': re.escape(prop)}) template = r']+?%s[^>]+?%s' @@ -841,7 +841,7 @@ class InfoExtractor(object): self._request_webpage(url, video_id, 'Checking %s URL' % item) return True except ExtractorError as e: - if isinstance(e.cause, compat_HTTPError): + if isinstance(e.cause, compat_urllib_error.URLError): self.to_screen( '%s: %s URL is invalid, skipping' % (video_id, item)) return False @@ -891,6 +891,11 @@ class InfoExtractor(object): if not media_nodes: manifest_version = '2.0' media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media') + base_url = xpath_text( + manifest, ['{http://ns.adobe.com/f4m/1.0}baseURL', '{http://ns.adobe.com/f4m/2.0}baseURL'], + 'base URL', default=None) + if base_url: + base_url = base_url.strip() for i, media_el in enumerate(media_nodes): if manifest_version == '2.0': media_url = media_el.attrib.get('href') or media_el.attrib.get('url') @@ -898,7 +903,7 @@ class InfoExtractor(object): continue manifest_url = ( media_url if media_url.startswith('http://') or media_url.startswith('https://') - else ('/'.join(manifest_url.split('/')[:-1]) + '/' + media_url)) + else ((base_url or '/'.join(manifest_url.split('/')[:-1])) + '/' + media_url)) # If media_url is itself a f4m manifest do the recursive extraction # since bitrates in parent manifest (this one) and media_url manifest # may differ leading to inability to resolve the format by requested @@ -943,13 +948,15 @@ class InfoExtractor(object): if re.match(r'^https?://', u) else compat_urlparse.urljoin(m3u8_url, u)) - m3u8_doc = self._download_webpage( + res = self._download_webpage_handle( m3u8_url, video_id, note=note or 'Downloading m3u8 information', errnote=errnote or 'Failed to download m3u8 information', fatal=fatal) - if m3u8_doc is False: - return m3u8_doc + if res is False: + return res + m3u8_doc, urlh = res + m3u8_url = urlh.geturl() last_info = None last_media = None kv_rex = re.compile( @@ -1278,7 +1285,7 @@ class InfoExtractor(object): def _get_cookies(self, url): """ Return a compat_cookies.SimpleCookie with the cookies for the url """ - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) self._downloader.cookiejar.add_cookie_header(req) return compat_cookies.SimpleCookie(req.get_header('Cookie')) diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py index 95952bc29..00d943f77 100644 --- a/youtube_dl/extractor/crunchyroll.py +++ b/youtube_dl/extractor/crunchyroll.py @@ -5,12 +5,12 @@ import re import json import base64 import zlib -import xml.etree.ElementTree from hashlib import sha1 from math import pow, sqrt, floor from .common import InfoExtractor from ..compat import ( + compat_etree_fromstring, compat_urllib_parse, compat_urllib_parse_unquote, compat_urllib_request, @@ -21,7 +21,9 @@ from ..utils import ( bytes_to_intlist, intlist_to_bytes, int_or_none, + lowercase_escape, remove_end, + sanitized_Request, unified_strdate, urlencode_postdata, xpath_text, @@ -32,9 +34,29 @@ from ..aes import ( class CrunchyrollBaseIE(InfoExtractor): + _NETRC_MACHINE = 'crunchyroll' + + def _login(self): + (username, password) = self._get_login_info() + if username is None: + return + self.report_login() + login_url = 'https://www.crunchyroll.com/?a=formhandler' + data = urlencode_postdata({ + 'formname': 'RpcApiUser_Login', + 'name': username, + 'password': password, + }) + login_request = sanitized_Request(login_url, data) + login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') + self._download_webpage(login_request, None, False, 'Wrong login info') + + def _real_initialize(self): + self._login() + def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None): request = (url_or_request if isinstance(url_or_request, compat_urllib_request.Request) - else compat_urllib_request.Request(url_or_request)) + else sanitized_Request(url_or_request)) # Accept-Language must be set explicitly to accept any language to avoid issues # similar to https://github.com/rg3/youtube-dl/issues/6797. # Along with IP address Crunchyroll uses Accept-Language to guess whether georestriction @@ -46,10 +68,22 @@ class CrunchyrollBaseIE(InfoExtractor): return super(CrunchyrollBaseIE, self)._download_webpage( request, video_id, note, errnote, fatal, tries, timeout, encoding) + @staticmethod + def _add_skip_wall(url): + parsed_url = compat_urlparse.urlparse(url) + qs = compat_urlparse.parse_qs(parsed_url.query) + # Always force skip_wall to bypass maturity wall, namely 18+ confirmation message: + # > This content may be inappropriate for some people. + # > Are you sure you want to continue? + # since it's not disabled by default in crunchyroll account's settings. + # See https://github.com/rg3/youtube-dl/issues/7202. + qs['skip_wall'] = ['1'] + return compat_urlparse.urlunparse( + parsed_url._replace(query=compat_urllib_parse.urlencode(qs, True))) + class CrunchyrollIE(CrunchyrollBaseIE): _VALID_URL = r'https?://(?:(?Pwww|m)\.)?(?Pcrunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P[0-9]+))(?:[/?&]|$)' - _NETRC_MACHINE = 'crunchyroll' _TESTS = [{ 'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513', 'info_dict': { @@ -72,7 +106,7 @@ class CrunchyrollIE(CrunchyrollBaseIE): 'id': '589804', 'ext': 'flv', 'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11', - 'description': 'md5:fe2743efedb49d279552926d0bd0cd9e', + 'description': 'md5:2fbc01f90b87e8e9137296f37b461c12', 'thumbnail': 're:^https?://.*\.jpg$', 'uploader': 'Danny Choo Network', 'upload_date': '20120213', @@ -81,10 +115,13 @@ class CrunchyrollIE(CrunchyrollBaseIE): # rtmp 'skip_download': True, }, - }, { 'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697', 'only_matching': True, + }, { + # geo-restricted (US), 18+ maturity wall, non-premium available + 'url': 'http://www.crunchyroll.com/cosplay-complex-ova/episode-1-the-birth-of-the-cosplay-club-565617', + 'only_matching': True, }] _FORMAT_IDS = { @@ -94,24 +131,6 @@ class CrunchyrollIE(CrunchyrollBaseIE): '1080': ('80', '108'), } - def _login(self): - (username, password) = self._get_login_info() - if username is None: - return - self.report_login() - login_url = 'https://www.crunchyroll.com/?a=formhandler' - data = urlencode_postdata({ - 'formname': 'RpcApiUser_Login', - 'name': username, - 'password': password, - }) - login_request = compat_urllib_request.Request(login_url, data) - login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - self._download_webpage(login_request, None, False, 'Wrong login info') - - def _real_initialize(self): - self._login() - def _decrypt_subtitles(self, data, iv, id): data = bytes_to_intlist(base64.b64decode(data.encode('utf-8'))) iv = bytes_to_intlist(base64.b64decode(iv.encode('utf-8'))) @@ -217,7 +236,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text return output def _extract_subtitles(self, subtitle): - sub_root = xml.etree.ElementTree.fromstring(subtitle) + sub_root = compat_etree_fromstring(subtitle) return [{ 'ext': 'srt', 'data': self._convert_subtitles_to_srt(sub_root), @@ -228,7 +247,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text def _get_subtitles(self, video_id, webpage): subtitles = {} - for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage): + for sub_id, sub_name in re.findall(r'\bssid=([0-9]+)"[^>]+?\btitle="([^"]+)', webpage): sub_page = self._download_webpage( 'http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id=' + sub_id, video_id, note='Downloading subtitles for ' + sub_name) @@ -254,7 +273,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text else: webpage_url = 'http://www.' + mobj.group('url') - webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage') + webpage = self._download_webpage(self._add_skip_wall(webpage_url), video_id, 'Downloading webpage') note_m = self._html_search_regex( r'
(.+?)
', webpage, 'trailer-notice', default='') @@ -270,11 +289,15 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text if 'To view this, please log in to verify you are 18 or older.' in webpage: self.raise_login_required() - video_title = self._html_search_regex(r']*>(.+?)', webpage, 'video_title', flags=re.DOTALL) + video_title = self._html_search_regex( + r'(?s)]*>((?:(?!]+itemprop=["\']title["\'][^>]*>(?:(?!', + webpage, 'video_title') video_title = re.sub(r' {2,}', ' ', video_title) - video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='') - if not video_description: - video_description = None + video_description = self._html_search_regex( + r']*>\s*.+?\[media_id=%s\].+?"description"\s*:\s*"([^"]+)' % video_id, + webpage, 'description', default=None) + if video_description: + video_description = lowercase_escape(video_description.replace(r'\r\n', '\n')) video_upload_date = self._html_search_regex( [r'
Availability for free users:(.+?)
', r'
[^<>]+\s*(.+?\d{4})\s*
'], webpage, 'video_upload_date', fatal=False, flags=re.DOTALL) @@ -285,7 +308,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text 'video_uploader', fatal=False) playerdata_url = compat_urllib_parse_unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url')) - playerdata_req = compat_urllib_request.Request(playerdata_url) + playerdata_req = sanitized_Request(playerdata_url) playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url}) playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded') playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info') @@ -297,7 +320,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage): stream_quality, stream_format = self._FORMAT_IDS[fmt] video_format = fmt + 'p' - streamdata_req = compat_urllib_request.Request( + streamdata_req = sanitized_Request( 'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s' % (stream_id, stream_format, stream_quality), compat_urllib_parse.urlencode({'current_page': url}).encode('utf-8')) @@ -352,7 +375,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE): IE_NAME = "crunchyroll:playlist" - _VALID_URL = r'https?://(?:(?Pwww|m)\.)?(?Pcrunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P[\w\-]+))/?$' + _VALID_URL = r'https?://(?:(?Pwww|m)\.)?(?Pcrunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P[\w\-]+))/?(?:\?|$)' _TESTS = [{ 'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi', @@ -361,12 +384,25 @@ class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE): 'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi' }, 'playlist_count': 13, + }, { + # geo-restricted (US), 18+ maturity wall, non-premium available + 'url': 'http://www.crunchyroll.com/cosplay-complex-ova', + 'info_dict': { + 'id': 'cosplay-complex-ova', + 'title': 'Cosplay Complex OVA' + }, + 'playlist_count': 3, + 'skip': 'Georestricted', + }, { + # geo-restricted (US), 18+ maturity wall, non-premium will be available since 2015.11.14 + 'url': 'http://www.crunchyroll.com/ladies-versus-butlers?skip_wall=1', + 'only_matching': True, }] def _real_extract(self, url): show_id = self._match_id(url) - webpage = self._download_webpage(url, show_id) + webpage = self._download_webpage(self._add_skip_wall(url), show_id) title = self._html_search_regex( r'(?s)]*>\s*(.*?)', webpage, 'title') diff --git a/youtube_dl/extractor/cspan.py b/youtube_dl/extractor/cspan.py index fbefd37d0..7b685d157 100644 --- a/youtube_dl/extractor/cspan.py +++ b/youtube_dl/extractor/cspan.py @@ -9,6 +9,7 @@ from ..utils import ( find_xpath_attr, smuggle_url, determine_ext, + ExtractorError, ) from .senateisvp import SenateISVPIE @@ -18,33 +19,32 @@ class CSpanIE(InfoExtractor): IE_DESC = 'C-SPAN' _TESTS = [{ 'url': 'http://www.c-span.org/video/?313572-1/HolderonV', - 'md5': '8e44ce11f0f725527daccc453f553eb0', + 'md5': '94b29a4f131ff03d23471dd6f60b6a1d', 'info_dict': { 'id': '315139', 'ext': 'mp4', 'title': 'Attorney General Eric Holder on Voting Rights Act Decision', - 'description': 'Attorney General Eric Holder spoke to reporters following the Supreme Court decision in Shelby County v. Holder in which the court ruled that the preclearance provisions of the Voting Rights Act could not be enforced until Congress established new guidelines for review.', + 'description': 'Attorney General Eric Holder speaks to reporters following the Supreme Court decision in [Shelby County v. Holder], in which the court ruled that the preclearance provisions of the Voting Rights Act could not be enforced.', }, 'skip': 'Regularly fails on travis, for unknown reasons', }, { 'url': 'http://www.c-span.org/video/?c4486943/cspan-international-health-care-models', - # For whatever reason, the served video alternates between - # two different ones + 'md5': '8e5fbfabe6ad0f89f3012a7943c1287b', 'info_dict': { - 'id': '340723', + 'id': 'c4486943', 'ext': 'mp4', - 'title': 'International Health Care Models', + 'title': 'CSPAN - International Health Care Models', 'description': 'md5:7a985a2d595dba00af3d9c9f0783c967', } }, { 'url': 'http://www.c-span.org/video/?318608-1/gm-ignition-switch-recall', - 'md5': '446562a736c6bf97118e389433ed88d4', + 'md5': '2ae5051559169baadba13fc35345ae74', 'info_dict': { 'id': '342759', 'ext': 'mp4', 'title': 'General Motors Ignition Switch Recall', 'duration': 14848, - 'description': 'md5:70c7c3b8fa63fa60d42772440596034c' + 'description': 'md5:118081aedd24bf1d3b68b3803344e7f3' }, }, { # Video from senate.gov @@ -57,67 +57,77 @@ class CSpanIE(InfoExtractor): }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - page_id = mobj.group('id') - webpage = self._download_webpage(url, page_id) - video_id = self._search_regex(r'progid=\'?([0-9]+)\'?>', webpage, 'video id') + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + matches = re.search(r'data-(prog|clip)id=\'([0-9]+)\'', webpage) + if matches: + video_type, video_id = matches.groups() + if video_type == 'prog': + video_type = 'program' + else: + senate_isvp_url = SenateISVPIE._search_iframe_url(webpage) + if senate_isvp_url: + title = self._og_search_title(webpage) + surl = smuggle_url(senate_isvp_url, {'force_title': title}) + return self.url_result(surl, 'SenateISVP', video_id, title) - description = self._html_search_regex( - [ - # The full description - r'
(.*?)(.*?)

' - ], - webpage, 'description', flags=re.DOTALL, default=None) + def get_text_attr(d, attr): + return d.get(attr, {}).get('#text') - info_url = 'http://c-spanvideo.org/videoLibrary/assets/player/ajax-player.php?os=android&html5=program&id=' + video_id - data = self._download_json(info_url, video_id) + data = self._download_json( + 'http://www.c-span.org/assets/player/ajax-player.php?os=android&html5=%s&id=%s' % (video_type, video_id), + video_id)['video'] + if data['@status'] != 'Success': + raise ExtractorError('%s said: %s' % (self.IE_NAME, get_text_attr(data, 'error')), expected=True) doc = self._download_xml( - 'http://www.c-span.org/common/services/flashXml.php?programid=' + video_id, + 'http://www.c-span.org/common/services/flashXml.php?%sid=%s' % (video_type, video_id), video_id) + description = self._html_search_meta('description', webpage) + title = find_xpath_attr(doc, './/string', 'name', 'title').text thumbnail = find_xpath_attr(doc, './/string', 'name', 'poster').text - senate_isvp_url = SenateISVPIE._search_iframe_url(webpage) - if senate_isvp_url: - surl = smuggle_url(senate_isvp_url, {'force_title': title}) - return self.url_result(surl, 'SenateISVP', video_id, title) + files = data['files'] + capfile = get_text_attr(data, 'capfile') - files = data['video']['files'] - try: - capfile = data['video']['capfile']['#text'] - except KeyError: - capfile = None - - entries = [{ - 'id': '%s_%d' % (video_id, partnum + 1), - 'title': ( - title if len(files) == 1 else - '%s part %d' % (title, partnum + 1)), - 'url': unescapeHTML(f['path']['#text']), - 'description': description, - 'thumbnail': thumbnail, - 'duration': int_or_none(f.get('length', {}).get('#text')), - 'subtitles': { - 'en': [{ - 'url': capfile, - 'ext': determine_ext(capfile, 'dfxp') - }], - } if capfile else None, - } for partnum, f in enumerate(files)] + entries = [] + for partnum, f in enumerate(files): + formats = [] + for quality in f['qualities']: + formats.append({ + 'format_id': '%s-%sp' % (get_text_attr(quality, 'bitrate'), get_text_attr(quality, 'height')), + 'url': unescapeHTML(get_text_attr(quality, 'file')), + 'height': int_or_none(get_text_attr(quality, 'height')), + 'tbr': int_or_none(get_text_attr(quality, 'bitrate')), + }) + self._sort_formats(formats) + entries.append({ + 'id': '%s_%d' % (video_id, partnum + 1), + 'title': ( + title if len(files) == 1 else + '%s part %d' % (title, partnum + 1)), + 'formats': formats, + 'description': description, + 'thumbnail': thumbnail, + 'duration': int_or_none(get_text_attr(f, 'length')), + 'subtitles': { + 'en': [{ + 'url': capfile, + 'ext': determine_ext(capfile, 'dfxp') + }], + } if capfile else None, + }) if len(entries) == 1: entry = dict(entries[0]) - entry['id'] = video_id + entry['id'] = 'c' + video_id if video_type == 'clip' else video_id return entry else: return { '_type': 'playlist', 'entries': entries, 'title': title, - 'id': video_id, + 'id': 'c' + video_id if video_type == 'clip' else video_id, } diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 80a05cfee..ab7f3aec4 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -7,15 +7,13 @@ import itertools from .common import InfoExtractor -from ..compat import ( - compat_str, - compat_urllib_request, -) +from ..compat import compat_str from ..utils import ( ExtractorError, determine_ext, int_or_none, parse_iso8601, + sanitized_Request, str_to_int, unescapeHTML, ) @@ -25,7 +23,7 @@ class DailymotionBaseInfoExtractor(InfoExtractor): @staticmethod def _build_request(url): """Build a request with the family filter disabled""" - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) request.add_header('Cookie', 'family_filter=off; ff=off') return request @@ -96,6 +94,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'uploader': 'HotWaves1012', 'age_limit': 18, } + }, + # geo-restricted, player v5 + { + 'url': 'http://www.dailymotion.com/video/xhza0o', + 'only_matching': True, } ] @@ -124,6 +127,9 @@ class DailymotionIE(DailymotionBaseInfoExtractor): if player_v5: player = self._parse_json(player_v5, video_id) metadata = player['metadata'] + + self._check_error(metadata) + formats = [] for quality, media_list in metadata['qualities'].items(): for media in media_list: @@ -133,9 +139,17 @@ class DailymotionIE(DailymotionBaseInfoExtractor): type_ = media.get('type') if type_ == 'application/vnd.lumberjack.manifest': continue - if type_ == 'application/x-mpegURL' or determine_ext(media_url) == 'm3u8': - formats.extend(self._extract_m3u8_formats( - media_url, video_id, 'mp4', m3u8_id='hls')) + ext = determine_ext(media_url) + if type_ == 'application/x-mpegURL' or ext == 'm3u8': + m3u8_formats = self._extract_m3u8_formats( + media_url, video_id, 'mp4', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + elif type_ == 'application/f4m' or ext == 'f4m': + f4m_formats = self._extract_f4m_formats( + media_url, video_id, preference=-1, f4m_id='hds', fatal=False) + if f4m_formats: + formats.extend(f4m_formats) else: f = { 'url': media_url, @@ -201,9 +215,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'video info', flags=re.MULTILINE), video_id) - if info.get('error') is not None: - msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title'] - raise ExtractorError(msg, expected=True) + self._check_error(info) formats = [] for (key, format_id) in self._FORMATS: @@ -246,6 +258,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'duration': info['duration'] } + def _check_error(self, info): + if info.get('error') is not None: + raise ExtractorError( + '%s said: %s' % (self.IE_NAME, info['error']['title']), expected=True) + def _get_subtitles(self, video_id, webpage): try: sub_list = self._download_webpage( diff --git a/youtube_dl/extractor/dbtv.py b/youtube_dl/extractor/dbtv.py index 212217625..133cdc50b 100644 --- a/youtube_dl/extractor/dbtv.py +++ b/youtube_dl/extractor/dbtv.py @@ -13,8 +13,8 @@ from ..utils import ( class DBTVIE(InfoExtractor): - _VALID_URL = r'http://dbtv\.no/(?P[0-9]+)#(?P.+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?dbtv\.no/(?:(?:lazyplayer|player)/)?(?P[0-9]+)(?:#(?P.+))?' + _TESTS = [{ 'url': 'http://dbtv.no/3649835190001#Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen', 'md5': 'b89953ed25dacb6edb3ef6c6f430f8bc', 'info_dict': { @@ -30,12 +30,18 @@ class DBTVIE(InfoExtractor): 'view_count': int, 'categories': list, } - } + }, { + 'url': 'http://dbtv.no/3649835190001', + 'only_matching': True, + }, { + 'url': 'http://www.dbtv.no/lazyplayer/4631135248001', + 'only_matching': True, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') - display_id = mobj.group('display_id') + display_id = mobj.group('display_id') or video_id data = self._download_json( 'http://api.dbtv.no/discovery/%s' % video_id, display_id) diff --git a/youtube_dl/extractor/dcn.py b/youtube_dl/extractor/dcn.py index 6f2fea5ff..9737cff14 100644 --- a/youtube_dl/extractor/dcn.py +++ b/youtube_dl/extractor/dcn.py @@ -2,13 +2,11 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( int_or_none, parse_iso8601, + sanitized_Request, ) @@ -36,7 +34,7 @@ class DCNIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - request = compat_urllib_request.Request( + request = sanitized_Request( 'http://admin.mangomolo.com/analytics/index.php/plus/video?id=%s' % video_id, headers={'Origin': 'http://www.dcndigital.ae'}) diff --git a/youtube_dl/extractor/democracynow.py b/youtube_dl/extractor/democracynow.py new file mode 100644 index 000000000..6cd395e11 --- /dev/null +++ b/youtube_dl/extractor/democracynow.py @@ -0,0 +1,88 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re +import os.path + +from .common import InfoExtractor +from ..compat import compat_urlparse +from ..utils import ( + url_basename, + remove_start, +) + + +class DemocracynowIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?democracynow.org/(?P[^\?]*)' + IE_NAME = 'democracynow' + _TESTS = [{ + 'url': 'http://www.democracynow.org/shows/2015/7/3', + 'md5': 'fbb8fe3d7a56a5e12431ce2f9b2fab0d', + 'info_dict': { + 'id': '2015-0703-001', + 'ext': 'mp4', + 'title': 'July 03, 2015 - Democracy Now!', + 'description': 'A daily independent global news hour with Amy Goodman & Juan González "What to the Slave is 4th of July?": James Earl Jones Reads Frederick Douglass\u2019 Historic Speech : "This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag : "We Shall Overcome": Remembering Folk Icon, Activist Pete Seeger in His Own Words & Songs', + }, + }, { + 'url': 'http://www.democracynow.org/2015/7/3/this_flag_comes_down_today_bree', + 'md5': 'fbb8fe3d7a56a5e12431ce2f9b2fab0d', + 'info_dict': { + 'id': '2015-0703-001', + 'ext': 'mp4', + 'title': '"This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag', + 'description': 'md5:4d2bc4f0d29f5553c2210a4bc7761a21', + }, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + description = self._og_search_description(webpage) + + json_data = self._parse_json(self._search_regex( + r']+type="text/json"[^>]*>\s*({[^>]+})', webpage, 'json'), + display_id) + video_id = None + formats = [] + + default_lang = 'en' + + subtitles = {} + + def add_subtitle_item(lang, info_dict): + if lang not in subtitles: + subtitles[lang] = [] + subtitles[lang].append(info_dict) + + # chapter_file are not subtitles + if 'caption_file' in json_data: + add_subtitle_item(default_lang, { + 'url': compat_urlparse.urljoin(url, json_data['caption_file']), + }) + + for subtitle_item in json_data.get('captions', []): + lang = subtitle_item.get('language', '').lower() or default_lang + add_subtitle_item(lang, { + 'url': compat_urlparse.urljoin(url, subtitle_item['url']), + }) + + for key in ('file', 'audio', 'video'): + media_url = json_data.get(key, '') + if not media_url: + continue + media_url = re.sub(r'\?.*', '', compat_urlparse.urljoin(url, media_url)) + video_id = video_id or remove_start(os.path.splitext(url_basename(media_url))[0], 'dn') + formats.append({ + 'url': media_url, + }) + + self._sort_formats(formats) + + return { + 'id': video_id or display_id, + 'title': json_data['title'], + 'description': description, + 'subtitles': subtitles, + 'formats': formats, + } diff --git a/youtube_dl/extractor/dplay.py b/youtube_dl/extractor/dplay.py new file mode 100644 index 000000000..6cda56a7f --- /dev/null +++ b/youtube_dl/extractor/dplay.py @@ -0,0 +1,51 @@ +# encoding: utf-8 +from __future__ import unicode_literals + +import time + +from .common import InfoExtractor +from ..utils import int_or_none + + +class DPlayIE(InfoExtractor): + _VALID_URL = r'http://www\.dplay\.se/[^/]+/(?P[^/?#]+)' + + _TEST = { + 'url': 'http://www.dplay.se/nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet/', + 'info_dict': { + 'id': '3172', + 'ext': 'mp4', + 'display_id': 'season-1-svensken-lar-sig-njuta-av-livet', + 'title': 'Svensken lär sig njuta av livet', + 'duration': 2650, + }, + } + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + video_id = self._search_regex( + r'data-video-id="(\d+)"', webpage, 'video id') + + info = self._download_json( + 'http://www.dplay.se/api/v2/ajax/videos?video_id=' + video_id, + video_id)['data'][0] + + self._set_cookie( + 'secure.dplay.se', 'dsc-geo', + '{"countryCode":"NL","expiry":%d}' % ((time.time() + 20 * 60) * 1000)) + # TODO: consider adding support for 'stream_type=hds', it seems to + # require setting some cookies + manifest_url = self._download_json( + 'https://secure.dplay.se/secure/api/v2/user/authorization/stream/%s?stream_type=hls' % video_id, + video_id, 'Getting manifest url for hls stream')['hls'] + formats = self._extract_m3u8_formats( + manifest_url, video_id, ext='mp4', entry_protocol='m3u8_native') + + return { + 'id': video_id, + 'display_id': display_id, + 'title': info['title'], + 'formats': formats, + 'duration': int_or_none(info.get('video_metadata_length'), scale=1000), + } diff --git a/youtube_dl/extractor/dramafever.py b/youtube_dl/extractor/dramafever.py index 38e6597c8..d836c1a6c 100644 --- a/youtube_dl/extractor/dramafever.py +++ b/youtube_dl/extractor/dramafever.py @@ -7,7 +7,6 @@ from .common import InfoExtractor from ..compat import ( compat_HTTPError, compat_urllib_parse, - compat_urllib_request, compat_urlparse, ) from ..utils import ( @@ -16,6 +15,7 @@ from ..utils import ( determine_ext, int_or_none, parse_iso8601, + sanitized_Request, ) @@ -51,7 +51,7 @@ class DramaFeverBaseIE(InfoExtractor): 'password': password, } - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) response = self._download_webpage( request, None, 'Logging in as %s' % username) diff --git a/youtube_dl/extractor/dumpert.py b/youtube_dl/extractor/dumpert.py index 1f00386fe..e5aadcd25 100644 --- a/youtube_dl/extractor/dumpert.py +++ b/youtube_dl/extractor/dumpert.py @@ -2,14 +2,17 @@ from __future__ import unicode_literals import base64 +import re from .common import InfoExtractor -from ..compat import compat_urllib_request -from ..utils import qualities +from ..utils import ( + qualities, + sanitized_Request, +) class DumpertIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?dumpert\.nl/(?:mediabase|embed)/(?P[0-9]+/[0-9a-zA-Z]+)' + _VALID_URL = r'(?Phttps?)://(?:www\.)?dumpert\.nl/(?:mediabase|embed)/(?P[0-9]+/[0-9a-zA-Z]+)' _TESTS = [{ 'url': 'http://www.dumpert.nl/mediabase/6646981/951bc60f/', 'md5': '1b9318d7d5054e7dcb9dc7654f21d643', @@ -26,10 +29,12 @@ class DumpertIE(InfoExtractor): }] def _real_extract(self, url): - video_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + protocol = mobj.group('protocol') - url = 'https://www.dumpert.nl/mediabase/' + video_id - req = compat_urllib_request.Request(url) + url = '%s://www.dumpert.nl/mediabase/%s' % (protocol, video_id) + req = sanitized_Request(url) req.add_header('Cookie', 'nsfw=1; cpc=10') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/eagleplatform.py b/youtube_dl/extractor/eagleplatform.py index e529b9b96..7bbf617d4 100644 --- a/youtube_dl/extractor/eagleplatform.py +++ b/youtube_dl/extractor/eagleplatform.py @@ -87,7 +87,7 @@ class EaglePlatformIE(InfoExtractor): m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON') formats = self._extract_m3u8_formats( m3u8_url, video_id, - 'mp4', entry_protocol='m3u8_native') + 'mp4', entry_protocol='m3u8_native', m3u8_id='hls') mp4_url = self._get_video_url( # Secure mp4 URL is constructed according to Player.prototype.mp4 from diff --git a/youtube_dl/extractor/eitb.py b/youtube_dl/extractor/eitb.py index 2cba82532..c83845fc2 100644 --- a/youtube_dl/extractor/eitb.py +++ b/youtube_dl/extractor/eitb.py @@ -1,39 +1,92 @@ # encoding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor -from .brightcove import BrightcoveIE -from ..utils import ExtractorError +from ..utils import ( + float_or_none, + int_or_none, + parse_iso8601, + sanitized_Request, +) class EitbIE(InfoExtractor): IE_NAME = 'eitb.tv' - _VALID_URL = r'https?://www\.eitb\.tv/(eu/bideoa|es/video)/[^/]+/(?P\d+)/(?P\d+)' + _VALID_URL = r'https?://(?:www\.)?eitb\.tv/(?:eu/bideoa|es/video)/[^/]+/\d+/(?P\d+)' _TEST = { - 'add_ie': ['Brightcove'], - 'url': 'http://www.eitb.tv/es/video/60-minutos-60-minutos-2013-2014/2677100210001/2743577154001/lasa-y-zabala-30-anos/', + 'url': 'http://www.eitb.tv/es/video/60-minutos-60-minutos-2013-2014/4104995148001/4090227752001/lasa-y-zabala-30-anos/', 'md5': 'edf4436247185adee3ea18ce64c47998', 'info_dict': { - 'id': '2743577154001', + 'id': '4090227752001', 'ext': 'mp4', 'title': '60 minutos (Lasa y Zabala, 30 años)', - # All videos from eitb has this description in the brightcove info - 'description': '.', - 'uploader': 'Euskal Telebista', + 'description': 'Programa de reportajes de actualidad.', + 'duration': 3996.76, + 'timestamp': 1381789200, + 'upload_date': '20131014', + 'tags': list, }, } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - chapter_id = mobj.group('chapter_id') - webpage = self._download_webpage(url, chapter_id) - bc_url = BrightcoveIE._extract_brightcove_url(webpage) - if bc_url is None: - raise ExtractorError('Could not extract the Brightcove url') - # The BrightcoveExperience object doesn't contain the video id, we set - # it manually - bc_url += '&%40videoPlayer={0}'.format(chapter_id) - return self.url_result(bc_url, BrightcoveIE.ie_key()) + video_id = self._match_id(url) + + video = self._download_json( + 'http://mam.eitb.eus/mam/REST/ServiceMultiweb/Video/MULTIWEBTV/%s/' % video_id, + video_id, 'Downloading video JSON') + + media = video['web_media'][0] + + formats = [] + for rendition in media['RENDITIONS']: + video_url = rendition.get('PMD_URL') + if not video_url: + continue + tbr = float_or_none(rendition.get('ENCODING_RATE'), 1000) + format_id = 'http' + if tbr: + format_id += '-%d' % int(tbr) + formats.append({ + 'url': rendition['PMD_URL'], + 'format_id': format_id, + 'width': int_or_none(rendition.get('FRAME_WIDTH')), + 'height': int_or_none(rendition.get('FRAME_HEIGHT')), + 'tbr': tbr, + }) + + hls_url = media.get('HLS_SURL') + if hls_url: + request = sanitized_Request( + 'http://mam.eitb.eus/mam/REST/ServiceMultiweb/DomainRestrictedSecurity/TokenAuth/', + headers={'Referer': url}) + token_data = self._download_json( + request, video_id, 'Downloading auth token', fatal=False) + if token_data: + token = token_data.get('token') + if token: + m3u8_formats = self._extract_m3u8_formats( + '%s?hdnts=%s' % (hls_url, token), video_id, m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + + hds_url = media.get('HDS_SURL') + if hds_url: + f4m_formats = self._extract_f4m_formats( + '%s?hdcore=3.7.0' % hds_url.replace('euskalsvod', 'euskalvod'), + video_id, f4m_id='hds', fatal=False) + if f4m_formats: + formats.extend(f4m_formats) + + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': media.get('NAME_ES') or media.get('name') or media['NAME_EU'], + 'description': media.get('SHORT_DESC_ES') or video.get('desc_group') or media.get('SHORT_DESC_EU'), + 'thumbnail': media.get('STILL_URL') or media.get('THUMBNAIL_URL'), + 'duration': float_or_none(media.get('LENGTH'), 1000), + 'timestamp': parse_iso8601(media.get('BROADCST_DATE'), ' '), + 'tags': media.get('TAGS'), + 'formats': formats, + } diff --git a/youtube_dl/extractor/escapist.py b/youtube_dl/extractor/escapist.py index c85b4c458..a3d7bbbcb 100644 --- a/youtube_dl/extractor/escapist.py +++ b/youtube_dl/extractor/escapist.py @@ -3,13 +3,12 @@ from __future__ import unicode_literals import json from .common import InfoExtractor -from ..compat import compat_urllib_request - from ..utils import ( determine_ext, clean_html, int_or_none, float_or_none, + sanitized_Request, ) @@ -75,7 +74,7 @@ class EscapistIE(InfoExtractor): video_id = ims_video['videoID'] key = ims_video['hash'] - config_req = compat_urllib_request.Request( + config_req = sanitized_Request( 'http://www.escapistmagazine.com/videos/' 'vidconfig.php?videoID=%s&hash=%s' % (video_id, key)) config_req.add_header('Referer', url) diff --git a/youtube_dl/extractor/everyonesmixtape.py b/youtube_dl/extractor/everyonesmixtape.py index d872d828f..493d38af8 100644 --- a/youtube_dl/extractor/everyonesmixtape.py +++ b/youtube_dl/extractor/everyonesmixtape.py @@ -3,11 +3,9 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -42,7 +40,7 @@ class EveryonesMixtapeIE(InfoExtractor): playlist_id = mobj.group('id') pllist_url = 'http://everyonesmixtape.com/mixtape.php?a=getMixes&u=-1&linked=%s&explore=' % playlist_id - pllist_req = compat_urllib_request.Request(pllist_url) + pllist_req = sanitized_Request(pllist_url) pllist_req.add_header('X-Requested-With', 'XMLHttpRequest') playlist_list = self._download_json( @@ -55,7 +53,7 @@ class EveryonesMixtapeIE(InfoExtractor): raise ExtractorError('Playlist id not found') pl_url = 'http://everyonesmixtape.com/mixtape.php?a=getMix&id=%s&userId=null&code=' % playlist_no - pl_req = compat_urllib_request.Request(pl_url) + pl_req = sanitized_Request(pl_url) pl_req.add_header('X-Requested-With', 'XMLHttpRequest') playlist = self._download_json( pl_req, playlist_id, note='Downloading playlist info') diff --git a/youtube_dl/extractor/extremetube.py b/youtube_dl/extractor/extremetube.py index c826a5404..3403581fd 100644 --- a/youtube_dl/extractor/extremetube.py +++ b/youtube_dl/extractor/extremetube.py @@ -3,23 +3,20 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_parse_qs, - compat_urllib_request, -) from ..utils import ( - qualities, + int_or_none, + sanitized_Request, str_to_int, ) class ExtremeTubeIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?(?Pextremetube\.com/.*?video/.+?(?P[0-9]+))(?:[/?&]|$)' + _VALID_URL = r'https?://(?:www\.)?extremetube\.com/(?:[^/]+/)?video/(?P[^/#?&]+)' _TESTS = [{ 'url': 'http://www.extremetube.com/video/music-video-14-british-euro-brit-european-cumshots-swallow-652431', 'md5': '344d0c6d50e2f16b06e49ca011d8ac69', 'info_dict': { - 'id': '652431', + 'id': 'music-video-14-british-euro-brit-european-cumshots-swallow-652431', 'ext': 'mp4', 'title': 'Music Video 14 british euro brit european cumshots swallow', 'uploader': 'unknown', @@ -29,14 +26,18 @@ class ExtremeTubeIE(InfoExtractor): }, { 'url': 'http://www.extremetube.com/gay/video/abcde-1234', 'only_matching': True, + }, { + 'url': 'http://www.extremetube.com/video/latina-slut-fucked-by-fat-black-dick', + 'only_matching': True, + }, { + 'url': 'http://www.extremetube.com/video/652431', + 'only_matching': True, }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - url = 'http://www.' + mobj.group('url') + video_id = self._match_id(url) - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) @@ -49,20 +50,36 @@ class ExtremeTubeIE(InfoExtractor): r'Views:\s*\s*([\d,\.]+)', webpage, 'view count', fatal=False)) - flash_vars = compat_parse_qs(self._search_regex( - r']+?name="flashvars"[^>]+?value="([^"]+)"', webpage, 'flash vars')) + flash_vars = self._parse_json( + self._search_regex( + r'var\s+flashvars\s*=\s*({.+?});', webpage, 'flash vars'), + video_id) formats = [] - quality = qualities(['180p', '240p', '360p', '480p', '720p', '1080p']) - for k, vals in flash_vars.items(): - m = re.match(r'quality_(?P[0-9]+p)$', k) - if m is not None: - formats.append({ - 'format_id': m.group('quality'), - 'quality': quality(m.group('quality')), - 'url': vals[0], + for quality_key, video_url in flash_vars.items(): + height = int_or_none(self._search_regex( + r'quality_(\d+)[pP]$', quality_key, 'height', default=None)) + if not height: + continue + f = { + 'url': video_url, + } + mobj = re.search( + r'/(?P\d{3,4})[pP]_(?P\d+)[kK]_\d+', video_url) + if mobj: + height = int(mobj.group('height')) + bitrate = int(mobj.group('bitrate')) + f.update({ + 'format_id': '%dp-%dk' % (height, bitrate), + 'height': height, + 'tbr': bitrate, }) - + else: + f.update({ + 'format_id': '%dp' % height, + 'height': height, + }) + formats.append(f) self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py index 178a7ca4c..321eec59e 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -10,12 +10,11 @@ from ..compat import ( compat_str, compat_urllib_error, compat_urllib_parse_unquote, - compat_urllib_request, ) from ..utils import ( ExtractorError, - int_or_none, limit_length, + sanitized_Request, urlencode_postdata, get_element_by_id, clean_html, @@ -74,7 +73,7 @@ class FacebookIE(InfoExtractor): if useremail is None: return - login_page_req = compat_urllib_request.Request(self._LOGIN_URL) + login_page_req = sanitized_Request(self._LOGIN_URL) login_page_req.add_header('Cookie', 'locale=en_US') login_page = self._download_webpage(login_page_req, None, note='Downloading login page', @@ -95,7 +94,7 @@ class FacebookIE(InfoExtractor): 'timezone': '-60', 'trynum': '1', } - request = compat_urllib_request.Request(self._LOGIN_URL, urlencode_postdata(login_form)) + request = sanitized_Request(self._LOGIN_URL, urlencode_postdata(login_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') try: login_results = self._download_webpage(request, None, @@ -110,7 +109,7 @@ class FacebookIE(InfoExtractor): r'name="h"\s+(?:\w+="[^"]+"\s+)*?value="([^"]+)"', login_results, 'h'), 'name_action_selected': 'dont_save', } - check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form)) + check_req = sanitized_Request(self._CHECKPOINT_URL, urlencode_postdata(check_form)) check_req.add_header('Content-Type', 'application/x-www-form-urlencoded') check_response = self._download_webpage(check_req, None, note='Confirming login') @@ -142,16 +141,20 @@ class FacebookIE(InfoExtractor): data = dict(json.loads(m.group(1))) params_raw = compat_urllib_parse_unquote(data['params']) params = json.loads(params_raw) - video_data = params['video_data'][0] formats = [] - for quality in ['sd', 'hd']: - src = video_data.get('%s_src' % quality) - if src is not None: - formats.append({ - 'format_id': quality, - 'url': src, - }) + for format_id, f in params['video_data'].items(): + if not f or not isinstance(f, list): + continue + for quality in ('sd', 'hd'): + for src_type in ('src', 'src_no_ratelimit'): + src = f[0].get('%s_%s' % (quality, src_type)) + if src: + formats.append({ + 'format_id': '%s_%s_%s' % (format_id, quality, src_type), + 'url': src, + 'preference': -10 if format_id == 'progressive' else 0, + }) if not formats: raise ExtractorError('Cannot find video formats') @@ -161,7 +164,7 @@ class FacebookIE(InfoExtractor): if not video_title: video_title = self._html_search_regex( r'(?s)(.*?)', - webpage, 'alternative title', fatal=False) + webpage, 'alternative title', default=None) video_title = limit_length(video_title, 80) if not video_title: video_title = 'Facebook video #%s' % video_id @@ -171,7 +174,5 @@ class FacebookIE(InfoExtractor): 'id': video_id, 'title': video_title, 'formats': formats, - 'duration': int_or_none(video_data.get('video_duration')), - 'thumbnail': video_data.get('thumbnail_src'), 'uploader': uploader, } diff --git a/youtube_dl/extractor/fc2.py b/youtube_dl/extractor/fc2.py index a406945e8..92e8c571f 100644 --- a/youtube_dl/extractor/fc2.py +++ b/youtube_dl/extractor/fc2.py @@ -12,6 +12,7 @@ from ..compat import ( from ..utils import ( encode_dict, ExtractorError, + sanitized_Request, ) @@ -57,7 +58,7 @@ class FC2IE(InfoExtractor): } login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('utf-8') - request = compat_urllib_request.Request( + request = sanitized_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') @@ -66,7 +67,7 @@ class FC2IE(InfoExtractor): return False # this is also needed - login_redir = compat_urllib_request.Request('http://id.fc2.com/?mode=redirect&login=done') + login_redir = sanitized_Request('http://id.fc2.com/?mode=redirect&login=done') self._download_webpage( login_redir, None, note='Login redirect', errnote='Login redirect failed') diff --git a/youtube_dl/extractor/flickr.py b/youtube_dl/extractor/flickr.py index 2fe76d661..91cd46e76 100644 --- a/youtube_dl/extractor/flickr.py +++ b/youtube_dl/extractor/flickr.py @@ -3,10 +3,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_urllib_request from ..utils import ( ExtractorError, find_xpath_attr, + sanitized_Request, ) @@ -30,7 +30,7 @@ class FlickrIE(InfoExtractor): video_id = mobj.group('id') video_uploader_id = mobj.group('uploader_id') webpage_url = 'http://www.flickr.com/photos/' + video_uploader_id + '/' + video_id - req = compat_urllib_request.Request(webpage_url) + req = sanitized_Request(webpage_url) req.add_header( 'User-Agent', # it needs a more recent version diff --git a/youtube_dl/extractor/fourtube.py b/youtube_dl/extractor/fourtube.py index fb6d108c0..fc4a5a0fb 100644 --- a/youtube_dl/extractor/fourtube.py +++ b/youtube_dl/extractor/fourtube.py @@ -3,12 +3,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( parse_duration, parse_iso8601, + sanitized_Request, str_to_int, ) @@ -93,7 +91,7 @@ class FourTubeIE(InfoExtractor): b'Content-Type': b'application/x-www-form-urlencoded', b'Origin': b'http://www.4tube.com', } - token_req = compat_urllib_request.Request(token_url, b'{}', headers) + token_req = sanitized_Request(token_url, b'{}', headers) tokens = self._download_json(token_req, video_id) formats = [{ 'url': tokens[format]['token'], diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index 129984a5f..8e60cf60f 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -83,6 +83,14 @@ class FranceTVBaseInfoExtractor(InfoExtractor): if subtitle: title += ' - %s' % subtitle + subtitles = {} + subtitles_list = [{ + 'url': subformat['url'], + 'ext': subformat.get('format'), + } for subformat in info.get('subtitles', []) if subformat.get('url')] + if subtitles_list: + subtitles['fr'] = subtitles_list + return { 'id': video_id, 'title': title, @@ -91,20 +99,27 @@ class FranceTVBaseInfoExtractor(InfoExtractor): 'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']), 'timestamp': int_or_none(info['diffusion']['timestamp']), 'formats': formats, + 'subtitles': subtitles, } class PluzzIE(FranceTVBaseInfoExtractor): IE_NAME = 'pluzz.francetv.fr' - _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html' + _VALID_URL = r'https?://(?:m\.)?pluzz\.francetv\.fr/videos/(?P.+?)\.html' # Can't use tests, videos expire in 7 days def _real_extract(self, url): - title = re.match(self._VALID_URL, url).group(1) - webpage = self._download_webpage(url, title) - video_id = self._search_regex( - r'data-diffusion="(\d+)"', webpage, 'ID') + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + video_id = self._html_search_meta( + 'id_video', webpage, 'video id', default=None) + if not video_id: + video_id = self._search_regex( + r'data-diffusion=["\'](\d+)', webpage, 'video id') + return self._extract_video(video_id, 'Pluzz') @@ -120,6 +135,9 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): 'title': 'Soir 3', 'upload_date': '20130826', 'timestamp': 1377548400, + 'subtitles': { + 'fr': 'mincount:2', + }, }, }, { 'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html', diff --git a/youtube_dl/extractor/funnyordie.py b/youtube_dl/extractor/funnyordie.py index f5f13689c..7f21d7410 100644 --- a/youtube_dl/extractor/funnyordie.py +++ b/youtube_dl/extractor/funnyordie.py @@ -45,11 +45,20 @@ class FunnyOrDieIE(InfoExtractor): links.sort(key=lambda link: 1 if link[1] == 'mp4' else 0) - bitrates = self._html_search_regex(r']+src=(["\'])(?P.+?/master\.m3u8)\1', + webpage, 'm3u8 url', default=None, group='url') formats = [] + + m3u8_formats = self._extract_m3u8_formats( + m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + + bitrates = [int(bitrate) for bitrate in re.findall(r'[,/]v(\d+)[,/]', m3u8_url)] + bitrates.sort() + for bitrate in bitrates: for link in links: formats.append({ diff --git a/youtube_dl/extractor/gametrailers.py b/youtube_dl/extractor/gametrailers.py index a6ab795ae..c3f031d9c 100644 --- a/youtube_dl/extractor/gametrailers.py +++ b/youtube_dl/extractor/gametrailers.py @@ -1,19 +1,62 @@ from __future__ import unicode_literals -from .mtv import MTVServicesInfoExtractor +from .common import InfoExtractor +from ..utils import ( + int_or_none, + parse_age_limit, + url_basename, +) -class GametrailersIE(MTVServicesInfoExtractor): - _VALID_URL = r'http://www\.gametrailers\.com/(?Pvideos|reviews|full-episodes)/(?P.*?)/(?P.*)' +class GametrailersIE(InfoExtractor): + _VALID_URL = r'http://www\.gametrailers\.com/videos/view/[^/]+/(?P<id>.+)' + _TEST = { - 'url': 'http://www.gametrailers.com/videos/zbvr8i/mirror-s-edge-2-e3-2013--debut-trailer', - 'md5': '4c8e67681a0ea7ec241e8c09b3ea8cf7', + 'url': 'http://www.gametrailers.com/videos/view/gametrailers-com/116437-Just-Cause-3-Review', + 'md5': 'f28c4efa0bdfaf9b760f6507955b6a6a', 'info_dict': { - 'id': '70e9a5d7-cf25-4a10-9104-6f3e7342ae0d', + 'id': '2983958', 'ext': 'mp4', - 'title': 'E3 2013: Debut Trailer', - 'description': 'Faith is back! Check out the World Premiere trailer for Mirror\'s Edge 2 straight from the EA Press Conference at E3 2013!', + 'display_id': '116437-Just-Cause-3-Review', + 'title': 'Just Cause 3 - Review', + 'description': 'It\'s a lot of fun to shoot at things and then watch them explode in Just Cause 3, but should there be more to the experience than that?', }, } - _FEED_URL = 'http://www.gametrailers.com/feeds/mrss' + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + title = self._html_search_regex( + r'<title>(.+?)\|', webpage, 'title').strip() + embed_url = self._proto_relative_url( + self._search_regex( + r'src=\'(//embed.gametrailers.com/embed/[^\']+)\'', webpage, + 'embed url'), + scheme='http:') + video_id = url_basename(embed_url) + embed_page = self._download_webpage(embed_url, video_id) + embed_vars_json = self._search_regex( + r'(?s)var embedVars = (\{.*?\})\s*</script>', embed_page, + 'embed vars') + info = self._parse_json(embed_vars_json, video_id) + + formats = [] + for media in info['media']: + if media['mediaPurpose'] == 'play': + formats.append({ + 'url': media['uri'], + 'height': media['height'], + 'width:': media['width'], + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'display_id': display_id, + 'title': title, + 'formats': formats, + 'thumbnail': info.get('thumbUri'), + 'description': self._og_search_description(webpage), + 'duration': int_or_none(info.get('videoLengthInSeconds')), + 'age_limit': parse_age_limit(info.get('audienceRating')), + } diff --git a/youtube_dl/extractor/gdcvault.py b/youtube_dl/extractor/gdcvault.py index a6834db43..3befd3e7b 100644 --- a/youtube_dl/extractor/gdcvault.py +++ b/youtube_dl/extractor/gdcvault.py @@ -3,13 +3,11 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( remove_end, HEADRequest, + sanitized_Request, ) @@ -125,7 +123,7 @@ class GDCVaultIE(InfoExtractor): 'password': password, } - request = compat_urllib_request.Request(login_url, compat_urllib_parse.urlencode(login_form)) + request = sanitized_Request(login_url, compat_urllib_parse.urlencode(login_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') self._download_webpage(request, display_id, 'Logging in') start_page = self._download_webpage(webpage_url, display_id, 'Getting authenticated video page') diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 285c0ff66..4d38b0c9d 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -9,8 +9,8 @@ import sys from .common import InfoExtractor from .youtube import YoutubeIE from ..compat import ( + compat_etree_fromstring, compat_urllib_parse_unquote, - compat_urllib_request, compat_urlparse, compat_xml_parse_error, ) @@ -21,7 +21,7 @@ from ..utils import ( HEADRequest, is_html, orderedSet, - parse_xml, + sanitized_Request, smuggle_url, unescapeHTML, unified_strdate, @@ -30,7 +30,10 @@ from ..utils import ( url_basename, xpath_text, ) -from .brightcove import BrightcoveIE +from .brightcove import ( + BrightcoveLegacyIE, + BrightcoveNewIE, +) from .nbc import NBCSportsVPlayerIE from .ooyala import OoyalaIE from .rutv import RUTVIE @@ -140,6 +143,7 @@ class GenericIE(InfoExtractor): 'ext': 'mp4', 'title': 'Automatics, robotics and biocybernetics', 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482', + 'upload_date': '20130627', 'formats': 'mincount:16', 'subtitles': 'mincount:1', }, @@ -273,7 +277,7 @@ class GenericIE(InfoExtractor): # it also tests brightcove videos that need to set the 'Referer' in the # http requests { - 'add_ie': ['Brightcove'], + 'add_ie': ['BrightcoveLegacy'], 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/', 'info_dict': { 'id': '2765128793001', @@ -297,7 +301,7 @@ class GenericIE(InfoExtractor): 'uploader': 'thestar.com', 'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.', }, - 'add_ie': ['Brightcove'], + 'add_ie': ['BrightcoveLegacy'], }, { 'url': 'http://www.championat.com/video/football/v/87/87499.html', @@ -312,7 +316,7 @@ class GenericIE(InfoExtractor): }, { # https://github.com/rg3/youtube-dl/issues/3541 - 'add_ie': ['Brightcove'], + 'add_ie': ['BrightcoveLegacy'], 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1', 'info_dict': { 'id': '3866516442001', @@ -818,6 +822,19 @@ class GenericIE(InfoExtractor): 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014', }, }, + # Kaltura embed protected with referrer + { + 'url': 'http://www.disney.nl/disney-channel/filmpjes/achter-de-schermen#/videoId/violetta-achter-de-schermen-ruggero', + 'info_dict': { + 'id': '1_g4fbemnq', + 'ext': 'mp4', + 'title': 'Violetta - Achter De Schermen - Ruggero', + 'description': 'Achter de schermen met Ruggero', + 'timestamp': 1435133761, + 'upload_date': '20150624', + 'uploader_id': 'echojecka', + }, + }, # Eagle.Platform embed (generic URL) { 'url': 'http://lenta.ru/news/2015/03/06/navalny/', @@ -1029,6 +1046,31 @@ class GenericIE(InfoExtractor): 'ext': 'mp4', 'title': 'cinemasnob', }, + }, + # BrightcoveInPageEmbed embed + { + 'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/', + 'info_dict': { + 'id': '4238694884001', + 'ext': 'flv', + 'title': 'Tabletop: Dread, Last Thoughts', + 'description': 'Tabletop: Dread, Last Thoughts', + 'duration': 51690, + }, + }, + # JWPlayer with M3U8 + { + 'url': 'http://ren.tv/novosti/2015-09-25/sluchaynyy-prohozhiy-poymal-avtougonshchika-v-murmanske-video', + 'info_dict': { + 'id': 'playlist', + 'ext': 'mp4', + 'title': 'Случайный прохожий поймал автоугонщика в Мурманске. ВИДЕО | РЕН ТВ', + 'uploader': 'ren.tv', + }, + 'params': { + # m3u8 downloads + 'skip_download': True, + } } ] @@ -1172,7 +1214,7 @@ class GenericIE(InfoExtractor): full_response = None if head_response is False: - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) request.add_header('Accept-Encoding', '*') full_response = self._request_webpage(request, video_id) head_response = full_response @@ -1201,7 +1243,7 @@ class GenericIE(InfoExtractor): '%s on generic information extractor.' % ('Forcing' if force else 'Falling back')) if not full_response: - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) # Some webservers may serve compressed content of rather big size (e.g. gzipped flac) # making it impossible to download only chunk of the file (yet we need only 512kB to # test whether it's HTML or not). According to youtube-dl default Accept-Encoding @@ -1236,7 +1278,7 @@ class GenericIE(InfoExtractor): # Is it an RSS feed, a SMIL file or a XSPF playlist? try: - doc = parse_xml(webpage) + doc = compat_etree_fromstring(webpage.encode('utf-8')) if doc.tag == 'rss': return self._extract_rss(url, video_id, doc) elif re.match(r'^(?:{[^}]+})?smil$', doc.tag): @@ -1288,14 +1330,14 @@ class GenericIE(InfoExtractor): return self.playlist_result( urlrs, playlist_id=video_id, playlist_title=video_title) - # Look for BrightCove: - bc_urls = BrightcoveIE._extract_brightcove_urls(webpage) + # Look for Brightcove Legacy Studio embeds + bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage) if bc_urls: self.to_screen('Brightcove video detected.') entries = [{ '_type': 'url', 'url': smuggle_url(bc_url, {'Referer': url}), - 'ie_key': 'Brightcove' + 'ie_key': 'BrightcoveLegacy' } for bc_url in bc_urls] return { @@ -1305,6 +1347,11 @@ class GenericIE(InfoExtractor): 'entries': entries, } + # Look for Brightcove New Studio embeds + bc_urls = BrightcoveNewIE._extract_urls(webpage) + if bc_urls: + return _playlist_from_matches(bc_urls, ie='BrightcoveNew') + # Look for embedded rtl.nl player matches = re.findall( r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"', @@ -1665,10 +1712,12 @@ class GenericIE(InfoExtractor): return self.url_result(mobj.group('url'), 'Zapiks') # Look for Kaltura embeds - mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage) or - re.search(r'(?s)(["\'])(?:https?:)?//cdnapisec\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?\1.*?entry_id\s*:\s*(["\'])(?P<id>[^\2]+?)\2', webpage)) + mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_?[Ii]d'\s*:\s*'(?P<id>[^']+)',", webpage) or + re.search(r'(?s)(?P<q1>["\'])(?:https?:)?//cdnapi(?:sec)?\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?(?P=q1).*?entry_?[Ii]d\s*:\s*(?P<q2>["\'])(?P<id>.+?)(?P=q2)', webpage)) if mobj is not None: - return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura') + return self.url_result(smuggle_url( + 'kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), + {'source_url': url}), 'Kaltura') # Look for Eagle.Platform embeds mobj = re.search( @@ -1713,7 +1762,7 @@ class GenericIE(InfoExtractor): # Look for UDN embeds mobj = re.search( - r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._VALID_URL, webpage) + r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage) if mobj is not None: return self.url_result( compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed') @@ -1833,6 +1882,7 @@ class GenericIE(InfoExtractor): entries = [] for video_url in found: + video_url = video_url.replace('\\/', '/') video_url = compat_urlparse.urljoin(url, video_url) video_id = compat_urllib_parse_unquote(os.path.basename(video_url)) @@ -1844,25 +1894,24 @@ class GenericIE(InfoExtractor): # here's a fun little line of code for you: video_id = os.path.splitext(video_id)[0] + entry_info_dict = { + 'id': video_id, + 'uploader': video_uploader, + 'title': video_title, + 'age_limit': age_limit, + } + ext = determine_ext(video_url) if ext == 'smil': - entries.append({ - 'id': video_id, - 'formats': self._extract_smil_formats(video_url, video_id), - 'uploader': video_uploader, - 'title': video_title, - 'age_limit': age_limit, - }) + entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id) elif ext == 'xspf': return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id) + elif ext == 'm3u8': + entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4') else: - entries.append({ - 'id': video_id, - 'url': video_url, - 'uploader': video_uploader, - 'title': video_title, - 'age_limit': age_limit, - }) + entry_info_dict['url'] = video_url + + entries.append(entry_info_dict) if len(entries) == 1: return entries[0] diff --git a/youtube_dl/extractor/globo.py b/youtube_dl/extractor/globo.py index 33d6432a6..c65ef6bcf 100644 --- a/youtube_dl/extractor/globo.py +++ b/youtube_dl/extractor/globo.py @@ -14,79 +14,58 @@ from ..utils import ( ExtractorError, float_or_none, int_or_none, + str_or_none, ) class GloboIE(InfoExtractor): - _VALID_URL = 'https?://.+?\.globo\.com/(?P<id>.+)' + _VALID_URL = '(?:globo:|https?://.+?\.globo\.com/(?:[^/]+/)*(?:v/(?:[^/]+/)?|videos/))(?P<id>\d{7,})' _API_URL_TEMPLATE = 'http://api.globovideos.com/videos/%s/playlist' _SECURITY_URL_TEMPLATE = 'http://security.video.globo.com/videos/%s/hash?player=flash&version=17.0.0.132&resource_id=%s' - _VIDEOID_REGEXES = [ - r'\bdata-video-id="(\d+)"', - r'\bdata-player-videosids="(\d+)"', - r'<div[^>]+\bid="(\d+)"', - ] - _RESIGN_EXPIRATION = 86400 - _TESTS = [ - { - 'url': 'http://globotv.globo.com/sportv/futebol-nacional/v/os-gols-de-atletico-mg-3-x-2-santos-pela-24a-rodada-do-brasileirao/3654973/', - 'md5': '03ebf41cb7ade43581608b7d9b71fab0', - 'info_dict': { - 'id': '3654973', - 'ext': 'mp4', - 'title': 'Os gols de Atlético-MG 3 x 2 Santos pela 24ª rodada do Brasileirão', - 'duration': 251.585, - 'uploader': 'SporTV', - 'uploader_id': 698, - 'like_count': int, - } + _TESTS = [{ + 'url': 'http://g1.globo.com/carros/autoesporte/videos/t/exclusivos-do-g1/v/mercedes-benz-gla-passa-por-teste-de-colisao-na-europa/3607726/', + 'md5': 'b3ccc801f75cd04a914d51dadb83a78d', + 'info_dict': { + 'id': '3607726', + 'ext': 'mp4', + 'title': 'Mercedes-Benz GLA passa por teste de colisão na Europa', + 'duration': 103.204, + 'uploader': 'Globo.com', + 'uploader_id': '265', }, - { - 'url': 'http://g1.globo.com/carros/autoesporte/videos/t/exclusivos-do-g1/v/mercedes-benz-gla-passa-por-teste-de-colisao-na-europa/3607726/', - 'md5': 'b3ccc801f75cd04a914d51dadb83a78d', - 'info_dict': { - 'id': '3607726', - 'ext': 'mp4', - 'title': 'Mercedes-Benz GLA passa por teste de colisão na Europa', - 'duration': 103.204, - 'uploader': 'Globo.com', - 'uploader_id': 265, - 'like_count': int, - } + }, { + 'url': 'http://globoplay.globo.com/v/4581987/', + 'md5': 'f36a1ecd6a50da1577eee6dd17f67eff', + 'info_dict': { + 'id': '4581987', + 'ext': 'mp4', + 'title': 'Acidentes de trânsito estão entre as maiores causas de queda de energia em SP', + 'duration': 137.973, + 'uploader': 'Rede Globo', + 'uploader_id': '196', }, - { - 'url': 'http://g1.globo.com/jornal-nacional/noticia/2014/09/novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes.html', - 'md5': '307fdeae4390ccfe6ba1aa198cf6e72b', - 'info_dict': { - 'id': '3652183', - 'ext': 'mp4', - 'title': 'Receita Federal explica como vai fiscalizar bagagens de quem retorna ao Brasil de avião', - 'duration': 110.711, - 'uploader': 'Rede Globo', - 'uploader_id': 196, - '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, - } - }, - ] + }, { + 'url': 'http://canalbrasil.globo.com/programas/sangue-latino/videos/3928201.html', + 'only_matching': True, + }, { + 'url': 'http://globosatplay.globo.com/globonews/v/4472924/', + 'only_matching': True, + }, { + 'url': 'http://globotv.globo.com/t/programa/v/clipe-sexo-e-as-negas-adeus/3836166/', + 'only_matching': True, + }, { + '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/', + 'only_matching': True, + }, { + 'url': 'http://canaloff.globo.com/programas/desejar-profundo/videos/4518560.html', + 'only_matching': True, + }] - class MD5(): + class MD5: HEX_FORMAT_LOWERCASE = 0 HEX_FORMAT_UPPERCASE = 1 BASE64_PAD_CHARACTER_DEFAULT_COMPLIANCE = '' @@ -353,9 +332,6 @@ class GloboIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - video_id = self._search_regex(self._VIDEOID_REGEXES, webpage, 'video id') - video = self._download_json( self._API_URL_TEMPLATE % video_id, video_id)['videos'][0] @@ -364,7 +340,7 @@ class GloboIE(InfoExtractor): formats = [] for resource in video['resources']: resource_id = resource.get('_id') - if not resource_id: + if not resource_id or resource_id.endswith('manifest'): continue security = self._download_json( @@ -393,20 +369,23 @@ class GloboIE(InfoExtractor): 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')) + m3u8_formats = self._extract_m3u8_formats( + signed_url, resource_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) else: formats.append({ 'url': signed_url, - 'format_id': resource_id, - 'height': resource.get('height'), + 'format_id': 'http-%s' % resource_id, + 'height': int_or_none(resource.get('height')), }) self._sort_formats(formats) duration = float_or_none(video.get('duration'), 1000) - like_count = int_or_none(video.get('likes')) uploader = video.get('channel') - uploader_id = video.get('channel_id') + uploader_id = str_or_none(video.get('channel_id')) return { 'id': video_id, @@ -414,6 +393,46 @@ class GloboIE(InfoExtractor): 'duration': duration, 'uploader': uploader, 'uploader_id': uploader_id, - 'like_count': like_count, 'formats': formats } + + +class GloboArticleIE(InfoExtractor): + _VALID_URL = 'https?://.+?\.globo\.com/(?:[^/]+/)*(?P<id>[^/]+)\.html' + + _VIDEOID_REGEXES = [ + r'\bdata-video-id=["\'](\d{7,})', + r'\bdata-player-videosids=["\'](\d{7,})', + r'\bvideosIDs\s*:\s*["\'](\d{7,})', + r'\bdata-id=["\'](\d{7,})', + r'<div[^>]+\bid=["\'](\d{7,})', + ] + + _TESTS = [{ + 'url': 'http://g1.globo.com/jornal-nacional/noticia/2014/09/novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes.html', + 'md5': '307fdeae4390ccfe6ba1aa198cf6e72b', + 'info_dict': { + 'id': '3652183', + 'ext': 'mp4', + 'title': 'Receita Federal explica como vai fiscalizar bagagens de quem retorna ao Brasil de avião', + 'duration': 110.711, + 'uploader': 'Rede Globo', + 'uploader_id': '196', + } + }, { + 'url': 'http://gq.globo.com/Prazeres/Poder/noticia/2015/10/all-o-desafio-assista-ao-segundo-capitulo-da-serie.html', + 'only_matching': True, + }, { + 'url': 'http://gshow.globo.com/programas/tv-xuxa/O-Programa/noticia/2014/01/xuxa-e-junno-namoram-muuuito-em-luau-de-zeze-di-camargo-e-luciano.html', + 'only_matching': True, + }] + + @classmethod + def suitable(cls, url): + return False if GloboIE.suitable(url) else super(GloboArticleIE, cls).suitable(url) + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + video_id = self._search_regex(self._VIDEOID_REGEXES, webpage, 'video id') + return self.url_result('globo:%s' % video_id, 'Globo') diff --git a/youtube_dl/extractor/googleplus.py b/youtube_dl/extractor/googleplus.py index fcefe54cd..731bacd67 100644 --- a/youtube_dl/extractor/googleplus.py +++ b/youtube_dl/extractor/googleplus.py @@ -61,7 +61,7 @@ class GooglePlusIE(InfoExtractor): 'width': int(width), 'height': int(height), } for width, height, video_url in re.findall( - r'\d+,(\d+),(\d+),"(https?://redirector\.googlevideo\.com.*?)"', webpage)] + r'\d+,(\d+),(\d+),"(https?://[^.]+\.googleusercontent.com.*?)"', webpage)] self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/hearthisat.py b/youtube_dl/extractor/hearthisat.py index a19b31ac0..7d8698655 100644 --- a/youtube_dl/extractor/hearthisat.py +++ b/youtube_dl/extractor/hearthisat.py @@ -4,12 +4,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( HEADRequest, + sanitized_Request, str_to_int, urlencode_postdata, urlhandle_detect_ext, @@ -47,7 +45,7 @@ class HearThisAtIE(InfoExtractor): r'intTrackId\s*=\s*(\d+)', webpage, 'track ID') payload = urlencode_postdata({'tracks[]': track_id}) - req = compat_urllib_request.Request(self._PLAYLIST_URL, payload) + req = sanitized_Request(self._PLAYLIST_URL, payload) req.add_header('Content-type', 'application/x-www-form-urlencoded') track = self._download_json(req, track_id, 'Downloading playlist')[0] diff --git a/youtube_dl/extractor/hotnewhiphop.py b/youtube_dl/extractor/hotnewhiphop.py index 651784b73..31e219945 100644 --- a/youtube_dl/extractor/hotnewhiphop.py +++ b/youtube_dl/extractor/hotnewhiphop.py @@ -3,13 +3,11 @@ from __future__ import unicode_literals import base64 from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, HEADRequest, + sanitized_Request, ) @@ -41,7 +39,7 @@ class HotNewHipHopIE(InfoExtractor): ('mediaType', 's'), ('mediaId', video_id), ]) - r = compat_urllib_request.Request( + r = sanitized_Request( 'http://www.hotnewhiphop.com/ajax/media/getActions/', data=reqdata) r.add_header('Content-Type', 'application/x-www-form-urlencoded') mkd = self._download_json( diff --git a/youtube_dl/extractor/hypem.py b/youtube_dl/extractor/hypem.py index aa0724a02..cca3dd498 100644 --- a/youtube_dl/extractor/hypem.py +++ b/youtube_dl/extractor/hypem.py @@ -4,12 +4,10 @@ import json import time from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -32,7 +30,7 @@ class HypemIE(InfoExtractor): data = {'ax': 1, 'ts': time.time()} data_encoded = compat_urllib_parse.urlencode(data) complete_url = url + "?" + data_encoded - request = compat_urllib_request.Request(complete_url) + request = sanitized_Request(complete_url) response, urlh = self._download_webpage_handle( request, track_id, 'Downloading webpage with the url') cookie = urlh.headers.get('Set-Cookie', '') @@ -52,7 +50,7 @@ class HypemIE(InfoExtractor): title = track['song'] serve_url = "http://hypem.com/serve/source/%s/%s" % (track_id, key) - request = compat_urllib_request.Request( + request = sanitized_Request( serve_url, '', {'Content-Type': 'application/json'}) request.add_header('cookie', cookie) song_data = self._download_json(request, track_id, 'Downloading metadata') diff --git a/youtube_dl/extractor/imdb.py b/youtube_dl/extractor/imdb.py index 4bb574cf3..02e1e428e 100644 --- a/youtube_dl/extractor/imdb.py +++ b/youtube_dl/extractor/imdb.py @@ -4,8 +4,8 @@ import re import json from .common import InfoExtractor -from ..compat import ( - compat_urlparse, +from ..utils import ( + qualities, ) @@ -30,24 +30,33 @@ class ImdbIE(InfoExtractor): descr = self._html_search_regex( r'(?s)<span itemprop="description">(.*?)</span>', webpage, 'description', fatal=False) - available_formats = re.findall( - r'case \'(?P<f_id>.*?)\' :$\s+url = \'(?P<path>.*?)\'', webpage, - flags=re.MULTILINE) + player_url = 'http://www.imdb.com/video/imdb/vi%s/imdb/single' % video_id + player_page = self._download_webpage( + player_url, video_id, 'Downloading player page') + # the player page contains the info for the default format, we have to + # fetch other pages for the rest of the formats + extra_formats = re.findall(r'href="(?P<url>%s.*?)".*?>(?P<name>.*?)<' % re.escape(player_url), player_page) + format_pages = [ + self._download_webpage( + f_url, video_id, 'Downloading info for %s format' % f_name) + for f_url, f_name in extra_formats] + format_pages.append(player_page) + + quality = qualities(['SD', '480p', '720p']) formats = [] - for f_id, f_path in available_formats: - f_path = f_path.strip() - format_page = self._download_webpage( - compat_urlparse.urljoin(url, f_path), - 'Downloading info for %s format' % f_id) + for format_page in format_pages: json_data = self._search_regex( r'<script[^>]+class="imdb-player-data"[^>]*?>(.*?)</script>', format_page, 'json data', flags=re.DOTALL) info = json.loads(json_data) format_info = info['videoPlayerObject']['video'] + f_id = format_info['ffname'] formats.append({ 'format_id': f_id, 'url': format_info['videoInfoList'][0]['videoUrl'], + 'quality': quality(f_id), }) + self._sort_formats(formats) return { 'id': video_id, diff --git a/youtube_dl/extractor/instagram.py b/youtube_dl/extractor/instagram.py index 3d78f78c4..c158f2064 100644 --- a/youtube_dl/extractor/instagram.py +++ b/youtube_dl/extractor/instagram.py @@ -10,8 +10,8 @@ from ..utils import ( class InstagramIE(InfoExtractor): - _VALID_URL = r'https://instagram\.com/p/(?P<id>[\da-zA-Z]+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?instagram\.com/p/(?P<id>[^/?#&]+)' + _TESTS = [{ 'url': 'https://instagram.com/p/aye83DjauH/?foo=bar#abc', 'md5': '0d2da106a9d2631273e192b372806516', 'info_dict': { @@ -21,7 +21,10 @@ class InstagramIE(InfoExtractor): 'title': 'Video by naomipq', 'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8', } - } + }, { + 'url': 'https://instagram.com/p/-Cmh1cukG2/', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) diff --git a/youtube_dl/extractor/iprima.py b/youtube_dl/extractor/iprima.py index 821c8ec10..36baf3245 100644 --- a/youtube_dl/extractor/iprima.py +++ b/youtube_dl/extractor/iprima.py @@ -6,12 +6,10 @@ from random import random from math import floor from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( ExtractorError, remove_end, + sanitized_Request, ) @@ -61,7 +59,7 @@ class IPrimaIE(InfoExtractor): (floor(random() * 1073741824), floor(random() * 1073741824)) ) - req = compat_urllib_request.Request(player_url) + req = sanitized_Request(player_url) req.add_header('Referer', url) playerpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/iqiyi.py b/youtube_dl/extractor/iqiyi.py index 0e53cb154..2df1da3f0 100644 --- a/youtube_dl/extractor/iqiyi.py +++ b/youtube_dl/extractor/iqiyi.py @@ -205,9 +205,9 @@ class IqiyiIE(InfoExtractor): def get_enc_key(self, swf_url, video_id): # TODO: automatic key extraction - # last update at 2015-10-10 for Zombie::bite - # '7239670519b6ac209a0bee4ef0446a6b24894b8ac2751506e42116212a0d0272e505'[2:66][1::2] - enc_key = '97596c0abee04ab49ba25564161ad225' + # last update at 2015-10-22 for Zombie::bite + # '7223c67061dbea1259d0ceb44f44b6d62288f4f80c972170de5201d2321060270e05'[2:66][0::2] + enc_key = '2c76de15dcb44bd28ff0927d50d31620' return enc_key def _real_extract(self, url): diff --git a/youtube_dl/extractor/ivi.py b/youtube_dl/extractor/ivi.py index e82594444..029878d24 100644 --- a/youtube_dl/extractor/ivi.py +++ b/youtube_dl/extractor/ivi.py @@ -5,11 +5,9 @@ import re import json from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -78,7 +76,7 @@ class IviIE(InfoExtractor): ] } - request = compat_urllib_request.Request(api_url, json.dumps(data)) + request = sanitized_Request(api_url, json.dumps(data)) video_json_page = self._download_webpage( request, video_id, 'Downloading video JSON') diff --git a/youtube_dl/extractor/jeuxvideo.py b/youtube_dl/extractor/jeuxvideo.py index 1df084d87..eef7daa29 100644 --- a/youtube_dl/extractor/jeuxvideo.py +++ b/youtube_dl/extractor/jeuxvideo.py @@ -28,7 +28,7 @@ class JeuxVideoIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) title = mobj.group(1) webpage = self._download_webpage(url, title) - title = self._html_search_meta('name', webpage) + title = self._html_search_meta('name', webpage) or self._og_search_title(webpage) config_url = self._html_search_regex( r'data-src="(/contenu/medias/video.php.*?)"', webpage, 'config URL') diff --git a/youtube_dl/extractor/kaltura.py b/youtube_dl/extractor/kaltura.py index 3dca0e566..583b1a5ad 100644 --- a/youtube_dl/extractor/kaltura.py +++ b/youtube_dl/extractor/kaltura.py @@ -2,12 +2,18 @@ from __future__ import unicode_literals import re +import base64 from .common import InfoExtractor -from ..compat import compat_urllib_parse +from ..compat import ( + compat_urllib_parse, + compat_urlparse, +) from ..utils import ( + clean_html, ExtractorError, int_or_none, + unsmuggle_url, ) @@ -16,7 +22,7 @@ class KalturaIE(InfoExtractor): (?: kaltura:(?P<partner_id_s>\d+):(?P<id_s>[0-9a-z_]+)| https?:// - (:?(?:www|cdnapisec)\.)?kaltura\.com/ + (:?(?:www|cdnapi(?:sec)?)\.)?kaltura\.com/ (?: (?: # flash player @@ -121,31 +127,47 @@ class KalturaIE(InfoExtractor): video_id, actions, note='Downloading video info JSON') def _real_extract(self, url): + url, smuggled_data = unsmuggle_url(url, {}) + mobj = re.match(self._VALID_URL, url) partner_id = mobj.group('partner_id_s') or mobj.group('partner_id') or mobj.group('partner_id_html5') entry_id = mobj.group('id_s') or mobj.group('id') or mobj.group('id_html5') info, source_data = self._get_video_info(entry_id, partner_id) - formats = [{ - 'format_id': '%(fileExt)s-%(bitrate)s' % f, - 'ext': f['fileExt'], - 'tbr': f['bitrate'], - 'fps': f.get('frameRate'), - 'filesize_approx': int_or_none(f.get('size'), invscale=1024), - 'container': f.get('containerFormat'), - 'vcodec': f.get('videoCodecId'), - 'height': f.get('height'), - 'width': f.get('width'), - 'url': '%s/flavorId/%s' % (info['dataUrl'], f['id']), - } for f in source_data['flavorAssets']] + source_url = smuggled_data.get('source_url') + if source_url: + referrer = base64.b64encode( + '://'.join(compat_urlparse.urlparse(source_url)[:2]) + .encode('utf-8')).decode('utf-8') + else: + referrer = None + + formats = [] + for f in source_data['flavorAssets']: + video_url = '%s/flavorId/%s' % (info['dataUrl'], f['id']) + if referrer: + video_url += '?referrer=%s' % referrer + formats.append({ + 'format_id': '%(fileExt)s-%(bitrate)s' % f, + 'ext': f.get('fileExt'), + 'tbr': int_or_none(f['bitrate']), + 'fps': int_or_none(f.get('frameRate')), + 'filesize_approx': int_or_none(f.get('size'), invscale=1024), + 'container': f.get('containerFormat'), + 'vcodec': f.get('videoCodecId'), + 'height': int_or_none(f.get('height')), + 'width': int_or_none(f.get('width')), + 'url': video_url, + }) + self._check_formats(formats, entry_id) self._sort_formats(formats) return { 'id': entry_id, 'title': info['name'], 'formats': formats, - 'description': info.get('description'), + 'description': clean_html(info.get('description')), 'thumbnail': info.get('thumbnailUrl'), 'duration': info.get('duration'), 'timestamp': info.get('createdAt'), diff --git a/youtube_dl/extractor/keezmovies.py b/youtube_dl/extractor/keezmovies.py index 82eddec51..d79261bb5 100644 --- a/youtube_dl/extractor/keezmovies.py +++ b/youtube_dl/extractor/keezmovies.py @@ -4,10 +4,8 @@ import os import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_urlparse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse_urlparse +from ..utils import sanitized_Request class KeezMoviesIE(InfoExtractor): @@ -26,7 +24,7 @@ class KeezMoviesIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/letv.py b/youtube_dl/extractor/letv.py index a28abb0f0..be648000e 100644 --- a/youtube_dl/extractor/letv.py +++ b/youtube_dl/extractor/letv.py @@ -8,14 +8,15 @@ import time from .common import InfoExtractor from ..compat import ( compat_urllib_parse, - compat_urllib_request, - compat_urlparse, + compat_ord, ) from ..utils import ( determine_ext, ExtractorError, parse_iso8601, + sanitized_Request, int_or_none, + encode_data_uri, ) @@ -25,15 +26,16 @@ class LetvIE(InfoExtractor): _TESTS = [{ 'url': 'http://www.letv.com/ptv/vplay/22005890.html', - 'md5': 'cab23bd68d5a8db9be31c9a222c1e8df', + 'md5': 'edadcfe5406976f42f9f266057ee5e40', 'info_dict': { 'id': '22005890', 'ext': 'mp4', 'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家', - 'timestamp': 1424747397, - 'upload_date': '20150224', 'description': 'md5:a9cb175fd753e2962176b7beca21a47c', - } + }, + 'params': { + 'hls_prefer_native': True, + }, }, { 'url': 'http://www.letv.com/ptv/vplay/1415246.html', 'info_dict': { @@ -42,16 +44,22 @@ class LetvIE(InfoExtractor): 'title': '美人天下01', 'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda', }, + 'params': { + 'hls_prefer_native': True, + }, }, { 'note': 'This video is available only in Mainland China, thus a proxy is needed', 'url': 'http://www.letv.com/ptv/vplay/1118082.html', - 'md5': 'f80936fbe20fb2f58648e81386ff7927', + 'md5': '2424c74948a62e5f31988438979c5ad1', 'info_dict': { 'id': '1118082', 'ext': 'mp4', 'title': '与龙共舞 完整版', 'description': 'md5:7506a5eeb1722bb9d4068f85024e3986', }, + 'params': { + 'hls_prefer_native': True, + }, 'skip': 'Only available in China', }] @@ -74,6 +82,27 @@ class LetvIE(InfoExtractor): _loc3_ = self.ror(_loc3_, _loc2_ % 17) return _loc3_ + # see M3U8Encryption class in KLetvPlayer.swf + @staticmethod + def decrypt_m3u8(encrypted_data): + if encrypted_data[:5].decode('utf-8').lower() != 'vc_01': + return encrypted_data + encrypted_data = encrypted_data[5:] + + _loc4_ = bytearray() + while encrypted_data: + b = compat_ord(encrypted_data[0]) + _loc4_.extend([b // 16, b & 0x0f]) + encrypted_data = encrypted_data[1:] + idx = len(_loc4_) - 11 + _loc4_ = _loc4_[idx:] + _loc4_[:idx] + _loc7_ = bytearray() + while _loc4_: + _loc7_.append(_loc4_[0] * 16 + _loc4_[1]) + _loc4_ = _loc4_[2:] + + return bytes(_loc7_) + def _real_extract(self, url): media_id = self._match_id(url) page = self._download_webpage(url, media_id) @@ -85,7 +114,7 @@ class LetvIE(InfoExtractor): 'tkey': self.calc_time_key(int(time.time())), 'domain': 'www.letv.com' } - play_json_req = compat_urllib_request.Request( + play_json_req = sanitized_Request( 'http://api.letv.com/mms/out/video/playJson?' + compat_urllib_parse.urlencode(params) ) cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') @@ -115,23 +144,28 @@ class LetvIE(InfoExtractor): for format_id in formats: if format_id in dispatch: media_url = playurl['domain'][0] + dispatch[format_id][0] - - # Mimic what flvxz.com do - url_parts = list(compat_urlparse.urlparse(media_url)) - qs = dict(compat_urlparse.parse_qs(url_parts[4])) - qs.update({ - 'platid': '14', - 'splatid': '1401', - 'tss': 'no', - 'retry': 1 + media_url += '&' + compat_urllib_parse.urlencode({ + 'm3v': 1, + 'format': 1, + 'expect': 3, + 'rateid': format_id, }) - url_parts[4] = compat_urllib_parse.urlencode(qs) - media_url = compat_urlparse.urlunparse(url_parts) + + nodes_data = self._download_json( + media_url, media_id, + 'Download JSON metadata for format %s' % format_id) + + req = self._request_webpage( + nodes_data['nodelist'][0]['location'], media_id, + note='Downloading m3u8 information for format %s' % format_id) + + m3u8_data = self.decrypt_m3u8(req.read()) url_info_dict = { - 'url': media_url, + 'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'), 'ext': determine_ext(dispatch[format_id][1]), 'format_id': format_id, + 'protocol': 'm3u8', } if format_id[-1:] == 'p': diff --git a/youtube_dl/extractor/lynda.py b/youtube_dl/extractor/lynda.py index 378117270..d4e1ae99d 100644 --- a/youtube_dl/extractor/lynda.py +++ b/youtube_dl/extractor/lynda.py @@ -7,12 +7,12 @@ from .common import InfoExtractor from ..compat import ( compat_str, compat_urllib_parse, - compat_urllib_request, ) from ..utils import ( ExtractorError, clean_html, int_or_none, + sanitized_Request, ) @@ -25,7 +25,7 @@ class LyndaBaseIE(InfoExtractor): self._login() def _login(self): - (username, password) = self._get_login_info() + username, password = self._get_login_info() if username is None: return @@ -35,7 +35,7 @@ class LyndaBaseIE(InfoExtractor): 'remember': 'false', 'stayPut': 'false' } - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) login_page = self._download_webpage( request, None, 'Logging in as %s' % username) @@ -64,7 +64,7 @@ class LyndaBaseIE(InfoExtractor): 'remember': 'false', 'stayPut': 'false', } - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form).encode('utf-8')) login_page = self._download_webpage( request, None, @@ -82,6 +82,15 @@ class LyndaBaseIE(InfoExtractor): expected=True) raise ExtractorError('Unable to log in') + def _logout(self): + username, _ = self._get_login_info() + if username is None: + return + + self._download_webpage( + 'http://www.lynda.com/ajax/logout.aspx', None, + 'Logging out', 'Unable to log out', fatal=False) + class LyndaIE(LyndaBaseIE): IE_NAME = 'lynda' @@ -108,50 +117,47 @@ class LyndaIE(LyndaBaseIE): def _real_extract(self, url): video_id = self._match_id(url) - page = self._download_webpage( + video = self._download_json( 'http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id, video_id, 'Downloading video JSON') - video_json = json.loads(page) - if 'Status' in video_json: + if 'Status' in video: raise ExtractorError( - 'lynda returned error: %s' % video_json['Message'], expected=True) + 'lynda returned error: %s' % video['Message'], expected=True) - if video_json['HasAccess'] is False: + if video.get('HasAccess') is False: self.raise_login_required('Video %s is only available for members' % video_id) - video_id = compat_str(video_json['ID']) - duration = video_json['DurationInSeconds'] - title = video_json['Title'] + video_id = compat_str(video.get('ID') or video_id) + duration = int_or_none(video.get('DurationInSeconds')) + title = video['Title'] formats = [] - fmts = video_json.get('Formats') + fmts = video.get('Formats') if fmts: - formats.extend([ - { - 'url': fmt['Url'], - 'ext': fmt['Extension'], - 'width': fmt['Width'], - 'height': fmt['Height'], - 'filesize': fmt['FileSize'], - 'format_id': str(fmt['Resolution']) - } for fmt in fmts]) + formats.extend([{ + 'url': f['Url'], + 'ext': f.get('Extension'), + 'width': int_or_none(f.get('Width')), + 'height': int_or_none(f.get('Height')), + 'filesize': int_or_none(f.get('FileSize')), + 'format_id': compat_str(f.get('Resolution')) if f.get('Resolution') else None, + } for f in fmts if f.get('Url')]) - prioritized_streams = video_json.get('PrioritizedStreams') + prioritized_streams = video.get('PrioritizedStreams') if prioritized_streams: - formats.extend([ - { + for prioritized_stream_id, prioritized_stream in prioritized_streams.items(): + formats.extend([{ 'url': video_url, 'width': int_or_none(format_id), - 'format_id': format_id, - } for format_id, video_url in prioritized_streams['0'].items() - ]) + 'format_id': '%s-%s' % (prioritized_stream_id, format_id), + } for format_id, video_url in prioritized_stream.items()]) self._check_formats(formats, video_id) self._sort_formats(formats) - subtitles = self.extract_subtitles(video_id, page) + subtitles = self.extract_subtitles(video_id) return { 'id': video_id, @@ -182,7 +188,7 @@ class LyndaIE(LyndaBaseIE): if srt: return srt - def _get_subtitles(self, video_id, webpage): + def _get_subtitles(self, video_id): url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id subs = self._download_json(url, None, False) if subs: @@ -204,12 +210,13 @@ class LyndaCourseIE(LyndaBaseIE): course_path = mobj.group('coursepath') course_id = mobj.group('courseid') - page = self._download_webpage( + course = self._download_json( 'http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id, course_id, 'Downloading course JSON') - course_json = json.loads(page) - if 'Status' in course_json and course_json['Status'] == 'NotFound': + self._logout() + + if course.get('Status') == 'NotFound': raise ExtractorError( 'Course %s does not exist' % course_id, expected=True) @@ -219,12 +226,13 @@ class LyndaCourseIE(LyndaBaseIE): # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided # by single video API anymore - for chapter in course_json['Chapters']: - for video in chapter['Videos']: - if video['HasAccess'] is False: + for chapter in course['Chapters']: + for video in chapter.get('Videos', []): + if video.get('HasAccess') is False: unaccessible_videos += 1 continue - videos.append(video['ID']) + if video.get('ID'): + videos.append(video['ID']) if unaccessible_videos > 0: self._downloader.report_warning( @@ -237,6 +245,6 @@ class LyndaCourseIE(LyndaBaseIE): 'Lynda') for video_id in videos] - course_title = course_json['Title'] + course_title = course.get('Title') return self.playlist_result(entries, course_id, course_title) diff --git a/youtube_dl/extractor/mdr.py b/youtube_dl/extractor/mdr.py index fc7499958..88334889e 100644 --- a/youtube_dl/extractor/mdr.py +++ b/youtube_dl/extractor/mdr.py @@ -1,64 +1,169 @@ +# coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor +from ..compat import compat_urlparse +from ..utils import ( + determine_ext, + int_or_none, + parse_duration, + parse_iso8601, + xpath_text, +) class MDRIE(InfoExtractor): - _VALID_URL = r'^(?P<domain>https?://(?:www\.)?mdr\.de)/(?:.*)/(?P<type>video|audio)(?P<video_id>[^/_]+)(?:_|\.html)' + IE_DESC = 'MDR.DE and KiKA' + _VALID_URL = r'https?://(?:www\.)?(?:mdr|kika)\.de/(?:.*)/[a-z]+(?P<id>\d+)(?:_.+?)?\.html' - # No tests, MDR regularily deletes its videos - _TEST = { + _TESTS = [{ + # MDR regularily deletes its videos 'url': 'http://www.mdr.de/fakt/video189002.html', 'only_matching': True, - } + }, { + # audio + 'url': 'http://www.mdr.de/kultur/audio1312272_zc-15948bad_zs-86171fdd.html', + 'md5': '64c4ee50f0a791deb9479cd7bbe9d2fa', + 'info_dict': { + 'id': '1312272', + 'ext': 'mp3', + 'title': 'Feuilleton vom 30. Oktober 2015', + 'duration': 250, + 'uploader': 'MITTELDEUTSCHER RUNDFUNK', + }, + }, { + 'url': 'http://www.kika.de/baumhaus/videos/video19636.html', + 'md5': '4930515e36b06c111213e80d1e4aad0e', + 'info_dict': { + 'id': '19636', + 'ext': 'mp4', + 'title': 'Baumhaus vom 30. Oktober 2015', + 'duration': 134, + 'uploader': 'KIKA', + }, + }, { + 'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/videos/video8182.html', + 'md5': '5fe9c4dd7d71e3b238f04b8fdd588357', + 'info_dict': { + 'id': '8182', + 'ext': 'mp4', + 'title': 'Beutolomäus und der geheime Weihnachtswunsch', + 'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd', + 'timestamp': 1419047100, + 'upload_date': '20141220', + 'duration': 4628, + 'uploader': 'KIKA', + }, + }, { + 'url': 'http://www.kika.de/baumhaus/sendungen/video19636_zc-fea7f8a0_zs-4bf89c60.html', + 'only_matching': True, + }, { + 'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/einzelsendung2534.html', + 'only_matching': True, + }] def _real_extract(self, url): - m = re.match(self._VALID_URL, url) - video_id = m.group('video_id') - domain = m.group('domain') + video_id = self._match_id(url) - # determine title and media streams from webpage - html = self._download_webpage(url, video_id) + webpage = self._download_webpage(url, video_id) - title = self._html_search_regex(r'<h[12]>(.*?)</h[12]>', html, 'title') - xmlurl = self._search_regex( - r'dataURL:\'(/(?:.+)/(?:video|audio)[0-9]+-avCustom.xml)', html, 'XML URL') + data_url = self._search_regex( + r'dataURL\s*:\s*(["\'])(?P<url>/.+/(?:video|audio)[0-9]+-avCustom\.xml)\1', + webpage, 'data url', group='url') + + doc = self._download_xml( + compat_urlparse.urljoin(url, data_url), video_id) + + title = xpath_text(doc, ['./title', './broadcast/broadcastName'], 'title', fatal=True) - doc = self._download_xml(domain + xmlurl, video_id) formats = [] - for a in doc.findall('./assets/asset'): - url_el = a.find('./progressiveDownloadUrl') - if url_el is None: - continue - abr = int(a.find('bitrateAudio').text) // 1000 - media_type = a.find('mediaType').text - format = { - 'abr': abr, - 'filesize': int(a.find('fileSize').text), - 'url': url_el.text, - } + processed_urls = [] + for asset in doc.findall('./assets/asset'): + for source in ( + 'progressiveDownload', + 'dynamicHttpStreamingRedirector', + 'adaptiveHttpStreamingRedirector'): + url_el = asset.find('./%sUrl' % source) + if url_el is None: + continue + + video_url = url_el.text + if video_url in processed_urls: + continue + + processed_urls.append(video_url) + + vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000) + abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000) + + ext = determine_ext(url_el.text) + if ext == 'm3u8': + url_formats = self._extract_m3u8_formats( + video_url, video_id, 'mp4', entry_protocol='m3u8_native', + preference=0, m3u8_id='HLS', fatal=False) + elif ext == 'f4m': + url_formats = self._extract_f4m_formats( + video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id, + preference=0, f4m_id='HDS', fatal=False) + else: + media_type = xpath_text(asset, './mediaType', 'media type', default='MP4') + vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000) + abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000) + filesize = int_or_none(xpath_text(asset, './fileSize', 'file size')) + + f = { + 'url': video_url, + 'format_id': '%s-%d' % (media_type, vbr or abr), + 'filesize': filesize, + 'abr': abr, + 'preference': 1, + } + + if vbr: + width = int_or_none(xpath_text(asset, './frameWidth', 'width')) + height = int_or_none(xpath_text(asset, './frameHeight', 'height')) + f.update({ + 'vbr': vbr, + 'width': width, + 'height': height, + }) + + url_formats = [f] + + if not url_formats: + continue + + if not vbr: + for f in url_formats: + abr = f.get('tbr') or abr + if 'tbr' in f: + del f['tbr'] + f.update({ + 'abr': abr, + 'vcodec': 'none', + }) + + formats.extend(url_formats) - vbr_el = a.find('bitrateVideo') - if vbr_el is None: - format.update({ - 'vcodec': 'none', - 'format_id': '%s-%d' % (media_type, abr), - }) - else: - vbr = int(vbr_el.text) // 1000 - format.update({ - 'vbr': vbr, - 'width': int(a.find('frameWidth').text), - 'height': int(a.find('frameHeight').text), - 'format_id': '%s-%d' % (media_type, vbr), - }) - formats.append(format) self._sort_formats(formats) + description = xpath_text(doc, './broadcast/broadcastDescription', 'description') + timestamp = parse_iso8601( + xpath_text( + doc, [ + './broadcast/broadcastDate', + './broadcast/broadcastStartDate', + './broadcast/broadcastEndDate'], + 'timestamp', default=None)) + duration = parse_duration(xpath_text(doc, './duration', 'duration')) + uploader = xpath_text(doc, './rights', 'uploader') + return { 'id': video_id, 'title': title, + 'description': description, + 'timestamp': timestamp, + 'duration': duration, + 'uploader': uploader, 'formats': formats, } diff --git a/youtube_dl/extractor/metacafe.py b/youtube_dl/extractor/metacafe.py index 6e2e73a51..3c786a36d 100644 --- a/youtube_dl/extractor/metacafe.py +++ b/youtube_dl/extractor/metacafe.py @@ -7,12 +7,12 @@ from ..compat import ( compat_parse_qs, compat_urllib_parse, compat_urllib_parse_unquote, - compat_urllib_request, ) from ..utils import ( determine_ext, ExtractorError, int_or_none, + sanitized_Request, ) @@ -117,7 +117,7 @@ class MetacafeIE(InfoExtractor): 'filters': '0', 'submit': "Continue - I'm over 18", } - request = compat_urllib_request.Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form)) + request = sanitized_Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') self.report_age_confirmation() self._download_webpage(request, None, False, 'Unable to confirm age') @@ -142,7 +142,7 @@ class MetacafeIE(InfoExtractor): return self.url_result('theplatform:%s' % ext_id, 'ThePlatform') # Retrieve video webpage to extract further information - req = compat_urllib_request.Request('http://www.metacafe.com/watch/%s/' % video_id) + req = sanitized_Request('http://www.metacafe.com/watch/%s/' % video_id) # AnyClip videos require the flashversion cookie so that we get the link # to the mp4 file diff --git a/youtube_dl/extractor/minhateca.py b/youtube_dl/extractor/minhateca.py index 14934b7ec..e46b23a6f 100644 --- a/youtube_dl/extractor/minhateca.py +++ b/youtube_dl/extractor/minhateca.py @@ -2,14 +2,12 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( int_or_none, parse_duration, parse_filesize, + sanitized_Request, ) @@ -39,7 +37,7 @@ class MinhatecaIE(InfoExtractor): ('fileId', video_id), ('__RequestVerificationToken', token), ] - req = compat_urllib_request.Request( + req = sanitized_Request( 'http://minhateca.com.br/action/License/Download', data=compat_urllib_parse.urlencode(token_data)) req.add_header('Content-Type', 'application/x-www-form-urlencoded') diff --git a/youtube_dl/extractor/miomio.py b/youtube_dl/extractor/miomio.py index a784fc5fb..170ebd9eb 100644 --- a/youtube_dl/extractor/miomio.py +++ b/youtube_dl/extractor/miomio.py @@ -8,6 +8,7 @@ from ..utils import ( xpath_text, int_or_none, ExtractorError, + sanitized_Request, ) @@ -51,6 +52,8 @@ class MioMioIE(InfoExtractor): mioplayer_path = self._search_regex( r'src="(/mioplayer/[^"]+)"', webpage, 'ref_path') + http_headers = {'Referer': 'http://www.miomio.tv%s' % mioplayer_path} + xml_config = self._search_regex( r'flashvars="type=(?:sina|video)&(.+?)&', webpage, 'xml config') @@ -60,14 +63,12 @@ class MioMioIE(InfoExtractor): 'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/xml.php?id=%s&r=%s' % (id, random.randint(100, 999)), video_id) - # the following xml contains the actual configuration information on the video file(s) - vid_config = self._download_xml( + vid_config_request = sanitized_Request( 'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/sina.php?{0}'.format(xml_config), - video_id) + headers=http_headers) - http_headers = { - 'Referer': 'http://www.miomio.tv%s' % mioplayer_path, - } + # the following xml contains the actual configuration information on the video file(s) + vid_config = self._download_xml(vid_config_request, video_id) if not int_or_none(xpath_text(vid_config, 'timelength')): raise ExtractorError('Unable to load videos!', expected=True) diff --git a/youtube_dl/extractor/mit.py b/youtube_dl/extractor/mit.py index f088ab9e2..29ca45778 100644 --- a/youtube_dl/extractor/mit.py +++ b/youtube_dl/extractor/mit.py @@ -86,7 +86,7 @@ class MITIE(TechTVMITIE): webpage = self._download_webpage(url, page_title) embed_url = self._search_regex( r'<iframe .*?src="(.+?)"', webpage, 'embed url') - return self.url_result(embed_url, ie='TechTVMIT') + return self.url_result(embed_url) class OCWMITIE(InfoExtractor): diff --git a/youtube_dl/extractor/mitele.py b/youtube_dl/extractor/mitele.py index 54993e2c9..c595f2077 100644 --- a/youtube_dl/extractor/mitele.py +++ b/youtube_dl/extractor/mitele.py @@ -1,7 +1,10 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import compat_urllib_parse +from ..compat import ( + compat_urllib_parse, + compat_urlparse, +) from ..utils import ( encode_dict, get_element_by_attribute, @@ -15,7 +18,7 @@ class MiTeleIE(InfoExtractor): _TESTS = [{ 'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/', - 'md5': 'ace7635b2a0b286aaa37d3ff192d2a8a', + 'md5': '0ff1a13aebb35d9bc14081ff633dd324', 'info_dict': { 'id': '0NF1jJnxS1Wu3pHrmvFyw2', 'display_id': 'programa-144', @@ -34,6 +37,7 @@ class MiTeleIE(InfoExtractor): config_url = self._search_regex( r'data-config\s*=\s*"([^"]+)"', webpage, 'data config url') + config_url = compat_urlparse.urljoin(url, config_url) config = self._download_json( config_url, display_id, 'Downloading config JSON') @@ -56,7 +60,7 @@ class MiTeleIE(InfoExtractor): 'sta': '0', } media = self._download_json( - '%s/?%s' % (gat, compat_urllib_parse.urlencode(encode_dict(token_data)).encode('utf-8')), + '%s/?%s' % (gat, compat_urllib_parse.urlencode(encode_dict(token_data))), display_id, 'Downloading %s JSON' % location['loc']) file_ = media.get('file') if not file_: diff --git a/youtube_dl/extractor/moevideo.py b/youtube_dl/extractor/moevideo.py index 5a66302f6..d930b9634 100644 --- a/youtube_dl/extractor/moevideo.py +++ b/youtube_dl/extractor/moevideo.py @@ -5,13 +5,11 @@ import json import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, int_or_none, + sanitized_Request, ) @@ -80,7 +78,7 @@ class MoeVideoIE(InfoExtractor): ] r_json = json.dumps(r) post = compat_urllib_parse.urlencode({'r': r_json}) - req = compat_urllib_request.Request(self._API_URL, post) + req = sanitized_Request(self._API_URL, post) req.add_header('Content-type', 'application/x-www-form-urlencoded') response = self._download_json(req, video_id) diff --git a/youtube_dl/extractor/mofosex.py b/youtube_dl/extractor/mofosex.py index 9bf99a54a..f8226cbb2 100644 --- a/youtube_dl/extractor/mofosex.py +++ b/youtube_dl/extractor/mofosex.py @@ -7,8 +7,8 @@ from .common import InfoExtractor from ..compat import ( compat_urllib_parse_unquote, compat_urllib_parse_urlparse, - compat_urllib_request, ) +from ..utils import sanitized_Request class MofosexIE(InfoExtractor): @@ -29,7 +29,7 @@ class MofosexIE(InfoExtractor): video_id = mobj.group('id') url = 'http://www.' + mobj.group('url') - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/moniker.py b/youtube_dl/extractor/moniker.py index 69e4bcd1a..f6bf94f2f 100644 --- a/youtube_dl/extractor/moniker.py +++ b/youtube_dl/extractor/moniker.py @@ -5,19 +5,17 @@ import os.path import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, remove_start, + sanitized_Request, ) class MonikerIE(InfoExtractor): IE_DESC = 'allmyvideos.net and vidspot.net' - _VALID_URL = r'https?://(?:www\.)?(?:allmyvideos|vidspot)\.net/(?P<id>[a-zA-Z0-9_-]+)' + _VALID_URL = r'https?://(?:www\.)?(?:allmyvideos|vidspot)\.net/(?:(?:2|v)/v-)?(?P<id>[a-zA-Z0-9_-]+)' _TESTS = [{ 'url': 'http://allmyvideos.net/jih3nce3x6wn', @@ -46,6 +44,18 @@ class MonikerIE(InfoExtractor): }, { 'url': 'https://www.vidspot.net/l2ngsmhs8ci5', 'only_matching': True, + }, { + 'url': 'http://vidspot.net/2/v-ywDf99', + 'md5': '5f8254ce12df30479428b0152fb8e7ba', + 'info_dict': { + 'id': 'ywDf99', + 'ext': 'mp4', + 'title': 'IL FAIT LE MALIN EN PORSHE CAYENNE ( mais pas pour longtemps)', + 'description': 'IL FAIT LE MALIN EN PORSHE CAYENNE.', + }, + }, { + 'url': 'http://allmyvideos.net/v/v-HXZm5t', + 'only_matching': True, }] def _real_extract(self, url): @@ -64,18 +74,30 @@ class MonikerIE(InfoExtractor): raise ExtractorError( '%s returned error: %s' % (self.IE_NAME, error), expected=True) - fields = re.findall(r'type="hidden" name="(.+?)"\s* value="?(.+?)">', orig_webpage) - data = dict(fields) + builtin_url = self._search_regex( + r'<iframe[^>]+src=(["\'])(?P<url>.+?/builtin-.+?)\1', + orig_webpage, 'builtin URL', default=None, group='url') - post = compat_urllib_parse.urlencode(data) - headers = { - b'Content-Type': b'application/x-www-form-urlencoded', - } - req = compat_urllib_request.Request(url, post, headers) - webpage = self._download_webpage( - req, video_id, note='Downloading video page ...') + if builtin_url: + req = sanitized_Request(builtin_url) + req.add_header('Referer', url) + webpage = self._download_webpage(req, video_id, 'Downloading builtin page') + title = self._og_search_title(orig_webpage).strip() + description = self._og_search_description(orig_webpage).strip() + else: + fields = re.findall(r'type="hidden" name="(.+?)"\s* value="?(.+?)">', orig_webpage) + data = dict(fields) - title = os.path.splitext(data['fname'])[0] + post = compat_urllib_parse.urlencode(data) + headers = { + b'Content-Type': b'application/x-www-form-urlencoded', + } + req = sanitized_Request(url, post, headers) + webpage = self._download_webpage( + req, video_id, note='Downloading video page ...') + + title = os.path.splitext(data['fname'])[0] + description = None # Could be several links with different quality links = re.findall(r'"file" : "?(.+?)",', webpage) @@ -89,5 +111,6 @@ class MonikerIE(InfoExtractor): return { 'id': video_id, 'title': title, + 'description': description, 'formats': formats, } diff --git a/youtube_dl/extractor/mooshare.py b/youtube_dl/extractor/mooshare.py index 7603af5e2..7cc7f054f 100644 --- a/youtube_dl/extractor/mooshare.py +++ b/youtube_dl/extractor/mooshare.py @@ -3,12 +3,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urllib_parse, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -59,7 +57,7 @@ class MooshareIE(InfoExtractor): 'hash': hash_key, } - request = compat_urllib_request.Request( + request = sanitized_Request( 'http://mooshare.biz/%s' % video_id, compat_urllib_parse.urlencode(download_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') diff --git a/youtube_dl/extractor/movieclips.py b/youtube_dl/extractor/movieclips.py index 04e17d055..1564cb71f 100644 --- a/youtube_dl/extractor/movieclips.py +++ b/youtube_dl/extractor/movieclips.py @@ -1,80 +1,40 @@ +# coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor -from ..compat import ( - compat_str, -) -from ..utils import ( - ExtractorError, - clean_html, -) +from ..utils import sanitized_Request class MovieClipsIE(InfoExtractor): - _VALID_URL = r'https?://movieclips\.com/(?P<id>[\da-zA-Z]+)(?:-(?P<display_id>[\da-z-]+))?' + _VALID_URL = r'https?://(?:www.)?movieclips\.com/videos/(?P<id>[^/?#]+)' _TEST = { - 'url': 'http://movieclips.com/Wy7ZU-my-week-with-marilyn-movie-do-you-love-me/', + 'url': 'http://www.movieclips.com/videos/warcraft-trailer-1-561180739597?autoPlay=true&playlistId=5', 'info_dict': { - 'id': 'Wy7ZU', - 'display_id': 'my-week-with-marilyn-movie-do-you-love-me', + 'id': 'pKIGmG83AqD9', + 'display_id': 'warcraft-trailer-1-561180739597', 'ext': 'mp4', - 'title': 'My Week with Marilyn - Do You Love Me?', - 'description': 'md5:e86795bd332fe3cff461e7c8dc542acb', + 'title': 'Warcraft Trailer 1', + 'description': 'Watch Trailer 1 from Warcraft (2016). Legendary’s WARCRAFT is a 3D epic adventure of world-colliding conflict based.', 'thumbnail': 're:^https?://.*\.jpg$', }, - 'params': { - # rtmp download - 'skip_download': True, - } + 'add_ie': ['ThePlatform'], } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - display_id = mobj.group('display_id') - show_id = display_id or video_id + display_id = self._match_id(url) - config = self._download_xml( - 'http://config.movieclips.com/player/config/%s' % video_id, - show_id, 'Downloading player config') - - if config.find('./country-region').text == 'false': - raise ExtractorError( - '%s said: %s' % (self.IE_NAME, config.find('./region_alert').text), expected=True) - - properties = config.find('./video/properties') - smil_file = properties.attrib['smil_file'] - - smil = self._download_xml(smil_file, show_id, 'Downloading SMIL') - base_url = smil.find('./head/meta').attrib['base'] - - formats = [] - for video in smil.findall('./body/switch/video'): - vbr = int(video.attrib['system-bitrate']) / 1000 - src = video.attrib['src'] - formats.append({ - 'url': base_url, - 'play_path': src, - 'ext': src.split(':')[0], - 'vbr': vbr, - 'format_id': '%dk' % vbr, - }) - - self._sort_formats(formats) - - title = '%s - %s' % (properties.attrib['clip_movie_title'], properties.attrib['clip_title']) - description = clean_html(compat_str(properties.attrib['clip_description'])) - thumbnail = properties.attrib['image'] - categories = properties.attrib['clip_categories'].split(',') + req = sanitized_Request(url) + # it doesn't work if it thinks the browser it's too old + req.add_header('User-Agent', 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/43.0 (Chrome)') + webpage = self._download_webpage(req, display_id) + theplatform_link = self._html_search_regex(r'src="(http://player.theplatform.com/p/.*?)"', webpage, 'theplatform link') + title = self._html_search_regex(r'<title[^>]*>([^>]+)-\s*\d+\s*|\s*Movieclips.com', webpage, 'title') + description = self._html_search_meta('description', webpage) return { - 'id': video_id, - 'display_id': display_id, + '_type': 'url_transparent', + 'url': theplatform_link, 'title': title, + 'display_id': display_id, 'description': description, - 'thumbnail': thumbnail, - 'categories': categories, - 'formats': formats, } diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index 302c9bf35..d887583e6 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -5,7 +5,6 @@ import re from .common import InfoExtractor from ..compat import ( compat_urllib_parse, - compat_urllib_request, compat_str, ) from ..utils import ( @@ -13,6 +12,7 @@ from ..utils import ( find_xpath_attr, fix_xml_ampersands, HEADRequest, + sanitized_Request, unescapeHTML, url_basename, RegexNotFoundError, @@ -53,7 +53,7 @@ class MTVServicesInfoExtractor(InfoExtractor): def _extract_mobile_video_formats(self, mtvn_id): webpage_url = self._MOBILE_TEMPLATE % mtvn_id - req = compat_urllib_request.Request(webpage_url) + req = sanitized_Request(webpage_url) # Otherwise we get a webpage that would execute some javascript req.add_header('User-Agent', 'curl/7') webpage = self._download_webpage(req, mtvn_id, diff --git a/youtube_dl/extractor/myvideo.py b/youtube_dl/extractor/myvideo.py index c96f472a3..36ab388b2 100644 --- a/youtube_dl/extractor/myvideo.py +++ b/youtube_dl/extractor/myvideo.py @@ -11,10 +11,10 @@ from ..compat import ( compat_ord, compat_urllib_parse, compat_urllib_parse_unquote, - compat_urllib_request, ) from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -83,7 +83,7 @@ class MyVideoIE(InfoExtractor): mobj = re.search(r'data-video-service="/service/data/video/%s/config' % video_id, webpage) if mobj is not None: - request = compat_urllib_request.Request('http://www.myvideo.de/service/data/video/%s/config' % video_id, '') + request = sanitized_Request('http://www.myvideo.de/service/data/video/%s/config' % video_id, '') response = self._download_webpage(request, video_id, 'Downloading video info') info = json.loads(base64.b64decode(response).decode('utf-8')) diff --git a/youtube_dl/extractor/ndr.py b/youtube_dl/extractor/ndr.py index e3cc6fde8..16213eed9 100644 --- a/youtube_dl/extractor/ndr.py +++ b/youtube_dl/extractor/ndr.py @@ -14,7 +14,8 @@ from ..utils import ( class NDRBaseIE(InfoExtractor): def _real_extract(self, url): - display_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + display_id = next(group for group in mobj.groups() if group) webpage = self._download_webpage(url, display_id) return self._extract_embed(webpage, display_id) @@ -22,7 +23,7 @@ class NDRBaseIE(InfoExtractor): class NDRIE(NDRBaseIE): IE_NAME = 'ndr' IE_DESC = 'NDR.de - Norddeutscher Rundfunk' - _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)+(?P[^/?#]+),[\da-z]+\.html' + _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)*(?P[^/?#]+),[\da-z]+\.html' _TESTS = [{ # httpVideo, same content id 'url': 'http://www.ndr.de/fernsehen/Party-Poette-und-Parade,hafengeburtstag988.html', @@ -77,6 +78,9 @@ class NDRIE(NDRBaseIE): 'params': { 'skip_download': True, }, + }, { + 'url': 'https://www.ndr.de/Fettes-Brot-Ferris-MC-und-Thees-Uhlmann-live-on-stage,festivalsommer116.html', + 'only_matching': True, }] def _extract_embed(self, webpage, display_id): @@ -101,7 +105,7 @@ class NDRIE(NDRBaseIE): class NJoyIE(NDRBaseIE): IE_NAME = 'njoy' IE_DESC = 'N-JOY' - _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)+(?P[^/?#]+),[\da-z]+\.html' + _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)*(?:(?P[^/?#]+),)?(?P[\da-z]+)\.html' _TESTS = [{ # httpVideo, same content id 'url': 'http://www.n-joy.de/entertainment/comedy/comedy_contest/Benaissa-beim-NDR-Comedy-Contest,comedycontest2480.html', @@ -136,6 +140,9 @@ class NJoyIE(NDRBaseIE): 'params': { 'skip_download': True, }, + }, { + 'url': 'http://www.n-joy.de/radio/webradio/morningshow209.html', + 'only_matching': True, }] def _extract_embed(self, webpage, display_id): @@ -231,7 +238,7 @@ class NDREmbedBaseIE(InfoExtractor): class NDREmbedIE(NDREmbedBaseIE): IE_NAME = 'ndr:embed' - _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)+(?P[\da-z]+)-(?:player|externalPlayer)\.html' + _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)*(?P[\da-z]+)-(?:player|externalPlayer)\.html' _TESTS = [{ 'url': 'http://www.ndr.de/fernsehen/sendungen/ndr_aktuell/ndraktuell28488-player.html', 'md5': '8b9306142fe65bbdefb5ce24edb6b0a9', @@ -325,7 +332,7 @@ class NDREmbedIE(NDREmbedBaseIE): class NJoyEmbedIE(NDREmbedBaseIE): IE_NAME = 'njoy:embed' - _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)+(?P[\da-z]+)-(?:player|externalPlayer)_[^/]+\.html' + _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)*(?P[\da-z]+)-(?:player|externalPlayer)_[^/]+\.html' _TESTS = [{ # httpVideo 'url': 'http://www.n-joy.de/events/reeperbahnfestival/doku948-player_image-bc168e87-5263-4d6d-bd27-bb643005a6de_theme-n-joy.html', diff --git a/youtube_dl/extractor/neteasemusic.py b/youtube_dl/extractor/neteasemusic.py index a8e0a64ed..15eca825a 100644 --- a/youtube_dl/extractor/neteasemusic.py +++ b/youtube_dl/extractor/neteasemusic.py @@ -8,11 +8,11 @@ import re from .common import InfoExtractor from ..compat import ( - compat_urllib_request, compat_urllib_parse, compat_str, compat_itertools_count, ) +from ..utils import sanitized_Request class NetEaseMusicBaseIE(InfoExtractor): @@ -40,7 +40,7 @@ class NetEaseMusicBaseIE(InfoExtractor): if not details: continue formats.append({ - 'url': 'http://m1.music.126.net/%s/%s.%s' % + 'url': 'http://m5.music.126.net/%s/%s.%s' % (cls._encrypt(details['dfsId']), details['dfsId'], details['extension']), 'ext': details.get('extension'), @@ -56,7 +56,7 @@ class NetEaseMusicBaseIE(InfoExtractor): return int(round(ms / 1000.0)) def query_api(self, endpoint, video_id, note): - req = compat_urllib_request.Request('%s%s' % (self._API_BASE, endpoint)) + req = sanitized_Request('%s%s' % (self._API_BASE, endpoint)) req.add_header('Referer', self._API_BASE) return self._download_json(req, video_id, note) diff --git a/youtube_dl/extractor/nfb.py b/youtube_dl/extractor/nfb.py index ea077254b..5bd15f7a7 100644 --- a/youtube_dl/extractor/nfb.py +++ b/youtube_dl/extractor/nfb.py @@ -1,10 +1,8 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urllib_parse, -) +from ..compat import compat_urllib_parse +from ..utils import sanitized_Request class NFBIE(InfoExtractor): @@ -40,8 +38,9 @@ class NFBIE(InfoExtractor): uploader = self._html_search_regex(r'([^<]+)', page, 'director name', fatal=False) - request = compat_urllib_request.Request('https://www.nfb.ca/film/%s/player_config' % video_id, - compat_urllib_parse.urlencode({'getConfig': 'true'}).encode('ascii')) + request = sanitized_Request( + 'https://www.nfb.ca/film/%s/player_config' % video_id, + compat_urllib_parse.urlencode({'getConfig': 'true'}).encode('ascii')) request.add_header('Content-Type', 'application/x-www-form-urlencoded') request.add_header('X-NFB-Referer', 'http://www.nfb.ca/medias/flash/NFBVideoPlayer.swf') diff --git a/youtube_dl/extractor/niconico.py b/youtube_dl/extractor/niconico.py index bda1cff05..586e52a4a 100644 --- a/youtube_dl/extractor/niconico.py +++ b/youtube_dl/extractor/niconico.py @@ -8,7 +8,6 @@ import datetime from .common import InfoExtractor from ..compat import ( compat_urllib_parse, - compat_urllib_request, compat_urlparse, ) from ..utils import ( @@ -17,6 +16,7 @@ from ..utils import ( int_or_none, parse_duration, parse_iso8601, + sanitized_Request, xpath_text, determine_ext, ) @@ -102,7 +102,7 @@ class NiconicoIE(InfoExtractor): 'password': password, } login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('utf-8') - request = compat_urllib_request.Request( + request = sanitized_Request( 'https://secure.nicovideo.jp/secure/login', login_data) login_results = self._download_webpage( request, None, note='Logging in', errnote='Unable to log in') @@ -145,7 +145,7 @@ class NiconicoIE(InfoExtractor): 'k': thumb_play_key, 'v': video_id }) - flv_info_request = compat_urllib_request.Request( + flv_info_request = sanitized_Request( 'http://ext.nicovideo.jp/thumb_watch', flv_info_data, {'Content-Type': 'application/x-www-form-urlencoded'}) flv_info_webpage = self._download_webpage( diff --git a/youtube_dl/extractor/noco.py b/youtube_dl/extractor/noco.py index a53e27b27..76bd21e6d 100644 --- a/youtube_dl/extractor/noco.py +++ b/youtube_dl/extractor/noco.py @@ -9,7 +9,6 @@ from .common import InfoExtractor from ..compat import ( compat_str, compat_urllib_parse, - compat_urllib_request, ) from ..utils import ( clean_html, @@ -17,6 +16,7 @@ from ..utils import ( int_or_none, float_or_none, parse_iso8601, + sanitized_Request, ) @@ -74,7 +74,7 @@ class NocoIE(InfoExtractor): 'username': username, 'password': password, } - request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form)) + request = sanitized_Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded; charset=UTF-8') login = self._download_json(request, None, 'Logging in as %s' % username) diff --git a/youtube_dl/extractor/nosvideo.py b/youtube_dl/extractor/nosvideo.py index f5ef856db..eab816e49 100644 --- a/youtube_dl/extractor/nosvideo.py +++ b/youtube_dl/extractor/nosvideo.py @@ -4,11 +4,9 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( ExtractorError, + sanitized_Request, urlencode_postdata, xpath_text, xpath_with_ns, @@ -41,7 +39,7 @@ class NosVideoIE(InfoExtractor): 'op': 'download1', 'method_free': 'Continue to Video', } - req = compat_urllib_request.Request(url, urlencode_postdata(fields)) + req = sanitized_Request(url, urlencode_postdata(fields)) req.add_header('Content-type', 'application/x-www-form-urlencoded') webpage = self._download_webpage(req, video_id, 'Downloading download page') diff --git a/youtube_dl/extractor/novamov.py b/youtube_dl/extractor/novamov.py index 04d779890..6163e8855 100644 --- a/youtube_dl/extractor/novamov.py +++ b/youtube_dl/extractor/novamov.py @@ -3,11 +3,13 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( ExtractorError, + NO_DEFAULT, + encode_dict, + sanitized_Request, + urlencode_postdata, ) @@ -38,19 +40,40 @@ class NovaMovIE(InfoExtractor): } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = self._match_id(url) - page = self._download_webpage( - 'http://%s/video/%s' % (self._HOST, video_id), video_id, 'Downloading video page') + url = 'http://%s/video/%s' % (self._HOST, video_id) - if re.search(self._FILE_DELETED_REGEX, page) is not None: + webpage = self._download_webpage( + url, video_id, 'Downloading video page') + + if re.search(self._FILE_DELETED_REGEX, webpage) is not None: raise ExtractorError('Video %s does not exist' % video_id, expected=True) - filekey = self._search_regex(self._FILEKEY_REGEX, page, 'filekey') + def extract_filekey(default=NO_DEFAULT): + return self._search_regex( + self._FILEKEY_REGEX, webpage, 'filekey', default=default) - title = self._html_search_regex(self._TITLE_REGEX, page, 'title', fatal=False) - description = self._html_search_regex(self._DESCRIPTION_REGEX, page, 'description', default='', fatal=False) + filekey = extract_filekey(default=None) + + if not filekey: + fields = self._hidden_inputs(webpage) + post_url = self._search_regex( + r']+action=(["\'])(?P.+?)\1', webpage, + 'post url', default=url, group='url') + if not post_url.startswith('http'): + post_url = compat_urlparse.urljoin(url, post_url) + request = sanitized_Request( + post_url, urlencode_postdata(encode_dict(fields))) + request.add_header('Content-Type', 'application/x-www-form-urlencoded') + request.add_header('Referer', post_url) + webpage = self._download_webpage( + request, video_id, 'Downloading continue to the video page') + + filekey = extract_filekey() + + title = self._html_search_regex(self._TITLE_REGEX, webpage, 'title', fatal=False) + description = self._html_search_regex(self._DESCRIPTION_REGEX, webpage, 'description', default='', fatal=False) api_response = self._download_webpage( 'http://%s/api/player.api.php?key=%s&file=%s' % (self._HOST, filekey, video_id), video_id, diff --git a/youtube_dl/extractor/nowness.py b/youtube_dl/extractor/nowness.py index b97f62fdb..d480fb58c 100644 --- a/youtube_dl/extractor/nowness.py +++ b/youtube_dl/extractor/nowness.py @@ -1,12 +1,12 @@ # encoding: utf-8 from __future__ import unicode_literals -from .brightcove import BrightcoveIE +from .brightcove import BrightcoveLegacyIE from .common import InfoExtractor -from ..utils import ExtractorError -from ..compat import ( - compat_str, - compat_urllib_request, +from ..compat import compat_str +from ..utils import ( + ExtractorError, + sanitized_Request, ) @@ -22,10 +22,10 @@ class NownessBaseIE(InfoExtractor): 'http://www.nowness.com/iframe?id=%s' % video_id, video_id, note='Downloading player JavaScript', errnote='Unable to download player JavaScript') - bc_url = BrightcoveIE._extract_brightcove_url(player_code) + bc_url = BrightcoveLegacyIE._extract_brightcove_url(player_code) if bc_url is None: raise ExtractorError('Could not find player definition') - return self.url_result(bc_url, 'Brightcove') + return self.url_result(bc_url, 'BrightcoveLegacy') elif source == 'vimeo': return self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo') elif source == 'youtube': @@ -37,7 +37,7 @@ class NownessBaseIE(InfoExtractor): def _api_request(self, url, request_path): display_id = self._match_id(url) - request = compat_urllib_request.Request( + request = sanitized_Request( 'http://api.nowness.com/api/' + request_path % display_id, headers={ 'X-Nowness-Language': 'zh-cn' if 'cn.nowness.com' in url else 'en-us', diff --git a/youtube_dl/extractor/nowtv.py b/youtube_dl/extractor/nowtv.py index b0bdffc4e..67e34b294 100644 --- a/youtube_dl/extractor/nowtv.py +++ b/youtube_dl/extractor/nowtv.py @@ -1,6 +1,8 @@ # coding: utf-8 from __future__ import unicode_literals +import re + from .common import InfoExtractor from ..compat import compat_str from ..utils import ( @@ -13,8 +15,63 @@ from ..utils import ( ) -class NowTVIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?nowtv\.(?:de|at|ch)/(?:rtl|rtl2|rtlnitro|superrtl|ntv|vox)/(?P.+?)/(?:player|preview)' +class NowTVBaseIE(InfoExtractor): + _VIDEO_FIELDS = ( + 'id', 'title', 'free', 'geoblocked', 'articleLong', 'articleShort', + 'broadcastStartDate', 'seoUrl', 'duration', 'files', + 'format.defaultImage169Format', 'format.defaultImage169Logo') + + def _extract_video(self, info, display_id=None): + video_id = compat_str(info['id']) + + files = info['files'] + if not files: + if info.get('geoblocked', False): + raise ExtractorError( + 'Video %s is not available from your location due to geo restriction' % video_id, + expected=True) + if not info.get('free', True): + raise ExtractorError( + 'Video %s is not available for free' % video_id, expected=True) + + formats = [] + for item in files['items']: + if determine_ext(item['path']) != 'f4v': + continue + app, play_path = remove_start(item['path'], '/').split('/', 1) + formats.append({ + 'url': 'rtmpe://fms.rtl.de', + 'app': app, + 'play_path': 'mp4:%s' % play_path, + 'ext': 'flv', + 'page_url': 'http://rtlnow.rtl.de', + 'player_url': 'http://cdn.static-fra.de/now/vodplayer.swf', + 'tbr': int_or_none(item.get('bitrate')), + }) + self._sort_formats(formats) + + title = info['title'] + description = info.get('articleLong') or info.get('articleShort') + timestamp = parse_iso8601(info.get('broadcastStartDate'), ' ') + duration = parse_duration(info.get('duration')) + + f = info.get('format', {}) + thumbnail = f.get('defaultImage169Format') or f.get('defaultImage169Logo') + + return { + 'id': video_id, + 'display_id': display_id or info.get('seoUrl'), + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'timestamp': timestamp, + 'duration': duration, + 'formats': formats, + } + + +class NowTVIE(NowTVBaseIE): + _VALID_URL = r'https?://(?:www\.)?nowtv\.(?:de|at|ch)/(?:rtl|rtl2|rtlnitro|superrtl|ntv|vox)/(?P[^/]+)/(?:list/[^/]+/)?(?P[^/]+)/(?:player|preview)' _TESTS = [{ # rtl @@ -23,7 +80,7 @@ class NowTVIE(InfoExtractor): 'id': '203519', 'display_id': 'bauer-sucht-frau/die-neuen-bauern-und-eine-hochzeit', 'ext': 'flv', - 'title': 'Die neuen Bauern und eine Hochzeit', + 'title': 'Inka Bause stellt die neuen Bauern vor', 'description': 'md5:e234e1ed6d63cf06be5c070442612e7e', 'thumbnail': 're:^https?://.*\.jpg$', 'timestamp': 1432580700, @@ -136,58 +193,65 @@ class NowTVIE(InfoExtractor): }] def _real_extract(self, url): - display_id = self._match_id(url) - display_id_split = display_id.split('/') - if len(display_id) > 2: - display_id = '/'.join((display_id_split[0], display_id_split[-1])) + mobj = re.match(self._VALID_URL, url) + display_id = '%s/%s' % (mobj.group('show_id'), mobj.group('id')) info = self._download_json( - 'https://api.nowtv.de/v3/movies/%s?fields=id,title,free,geoblocked,articleLong,articleShort,broadcastStartDate,seoUrl,duration,format,files' % display_id, - display_id) + 'https://api.nowtv.de/v3/movies/%s?fields=%s' + % (display_id, ','.join(self._VIDEO_FIELDS)), display_id) - video_id = compat_str(info['id']) + return self._extract_video(info, display_id) - files = info['files'] - if not files: - if info.get('geoblocked', False): - raise ExtractorError( - 'Video %s is not available from your location due to geo restriction' % video_id, - expected=True) - if not info.get('free', True): - raise ExtractorError( - 'Video %s is not available for free' % video_id, expected=True) - formats = [] - for item in files['items']: - if determine_ext(item['path']) != 'f4v': - continue - app, play_path = remove_start(item['path'], '/').split('/', 1) - formats.append({ - 'url': 'rtmpe://fms.rtl.de', - 'app': app, - 'play_path': 'mp4:%s' % play_path, - 'ext': 'flv', - 'page_url': 'http://rtlnow.rtl.de', - 'player_url': 'http://cdn.static-fra.de/now/vodplayer.swf', - 'tbr': int_or_none(item.get('bitrate')), - }) - self._sort_formats(formats) +class NowTVListIE(NowTVBaseIE): + _VALID_URL = r'https?://(?:www\.)?nowtv\.(?:de|at|ch)/(?:rtl|rtl2|rtlnitro|superrtl|ntv|vox)/(?P[^/]+)/list/(?P[^?/#&]+)$' - title = info['title'] - description = info.get('articleLong') or info.get('articleShort') - timestamp = parse_iso8601(info.get('broadcastStartDate'), ' ') - duration = parse_duration(info.get('duration')) + _SHOW_FIELDS = ('title', ) + _SEASON_FIELDS = ('id', 'headline', 'seoheadline', ) - f = info.get('format', {}) - thumbnail = f.get('defaultImage169Format') or f.get('defaultImage169Logo') + _TESTS = [{ + 'url': 'http://www.nowtv.at/rtl/stern-tv/list/aktuell', + 'info_dict': { + 'id': '17006', + 'title': 'stern TV - Aktuell', + }, + 'playlist_count': 1, + }, { + 'url': 'http://www.nowtv.at/rtl/das-supertalent/list/free-staffel-8', + 'info_dict': { + 'id': '20716', + 'title': 'Das Supertalent - FREE Staffel 8', + }, + 'playlist_count': 14, + }] - return { - 'id': video_id, - 'display_id': display_id, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'timestamp': timestamp, - 'duration': duration, - 'formats': formats, - } + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + show_id = mobj.group('show_id') + season_id = mobj.group('id') + + fields = [] + fields.extend(self._SHOW_FIELDS) + fields.extend('formatTabs.%s' % field for field in self._SEASON_FIELDS) + fields.extend( + 'formatTabs.formatTabPages.container.movies.%s' % field + for field in self._VIDEO_FIELDS) + + list_info = self._download_json( + 'https://api.nowtv.de/v3/formats/seo?fields=%s&name=%s.php' + % (','.join(fields), show_id), + season_id) + + season = next( + season for season in list_info['formatTabs']['items'] + if season.get('seoheadline') == season_id) + + title = '%s - %s' % (list_info['title'], season['headline']) + + entries = [] + for container in season['formatTabPages']['items']: + for info in ((container.get('container') or {}).get('movies') or {}).get('items') or []: + entries.append(self._extract_video(info)) + + return self.playlist_result( + entries, compat_str(season.get('id') or season_id), title) diff --git a/youtube_dl/extractor/nowvideo.py b/youtube_dl/extractor/nowvideo.py index 17baa9679..57ee3d366 100644 --- a/youtube_dl/extractor/nowvideo.py +++ b/youtube_dl/extractor/nowvideo.py @@ -7,9 +7,9 @@ class NowVideoIE(NovaMovIE): IE_NAME = 'nowvideo' IE_DESC = 'NowVideo' - _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'nowvideo\.(?:ch|ec|sx|eu|at|ag|co|li)'} + _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'nowvideo\.(?:to|ch|ec|sx|eu|at|ag|co|li)'} - _HOST = 'www.nowvideo.ch' + _HOST = 'www.nowvideo.to' _FILE_DELETED_REGEX = r'>This file no longer exists on our servers.<' _FILEKEY_REGEX = r'var fkzd="([^"]+)";' diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py index 8ac38a174..6ff13050d 100644 --- a/youtube_dl/extractor/nrk.py +++ b/youtube_dl/extractor/nrk.py @@ -6,6 +6,7 @@ import re from .common import InfoExtractor from ..compat import compat_urlparse from ..utils import ( + determine_ext, ExtractorError, float_or_none, parse_duration, @@ -48,12 +49,22 @@ class NRKIE(InfoExtractor): 'http://v8.psapi.nrk.no/mediaelement/%s' % video_id, video_id, 'Downloading media JSON') - if data['usageRights']['isGeoBlocked']: - raise ExtractorError( - 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge', - expected=True) + media_url = data.get('mediaUrl') - video_url = data['mediaUrl'] + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81' + if not media_url: + if data['usageRights']['isGeoBlocked']: + raise ExtractorError( + 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge', + expected=True) + + if determine_ext(media_url) == 'f4m': + formats = self._extract_f4m_formats( + media_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81', video_id, f4m_id='hds') + else: + formats = [{ + 'url': media_url, + 'ext': 'flv', + }] duration = parse_duration(data.get('duration')) @@ -67,12 +78,11 @@ class NRKIE(InfoExtractor): return { 'id': video_id, - 'url': video_url, - 'ext': 'flv', 'title': data['title'], 'description': data['description'], 'duration': duration, 'thumbnail': thumbnail, + 'formats': formats, } diff --git a/youtube_dl/extractor/nuvid.py b/youtube_dl/extractor/nuvid.py index 57928f2ae..9fa7cefad 100644 --- a/youtube_dl/extractor/nuvid.py +++ b/youtube_dl/extractor/nuvid.py @@ -3,11 +3,9 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( parse_duration, + sanitized_Request, unified_strdate, ) @@ -33,7 +31,7 @@ class NuvidIE(InfoExtractor): formats = [] for dwnld_speed, format_id in [(0, '3gp'), (5, 'mp4')]: - request = compat_urllib_request.Request( + request = sanitized_Request( 'http://m.nuvid.com/play/%s' % video_id) request.add_header('Cookie', 'skip_download_page=1; dwnld_speed=%d; adv_show=1' % dwnld_speed) webpage = self._download_webpage( diff --git a/youtube_dl/extractor/odnoklassniki.py b/youtube_dl/extractor/odnoklassniki.py index ccc88cfb1..184c7a323 100644 --- a/youtube_dl/extractor/odnoklassniki.py +++ b/youtube_dl/extractor/odnoklassniki.py @@ -13,7 +13,7 @@ from ..utils import ( class OdnoklassnikiIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?(?:odnoklassniki|ok)\.ru/(?:video|web-api/video/moviePlayer)/(?P[\d-]+)' + _VALID_URL = r'https?://(?:www\.)?(?:odnoklassniki|ok)\.ru/(?:video(?:embed)?|web-api/video/moviePlayer)/(?P[\d-]+)' _TESTS = [{ # metadata in JSON 'url': 'http://ok.ru/video/20079905452', @@ -66,6 +66,9 @@ class OdnoklassnikiIE(InfoExtractor): }, { 'url': 'http://www.ok.ru/video/20648036891', 'only_matching': True, + }, { + 'url': 'http://www.ok.ru/videoembed/20648036891', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/patreon.py b/youtube_dl/extractor/patreon.py index 6cdc2638b..ec8876c28 100644 --- a/youtube_dl/extractor/patreon.py +++ b/youtube_dl/extractor/patreon.py @@ -2,9 +2,7 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import ( - js_to_json, -) +from ..utils import js_to_json class PatreonIE(InfoExtractor): @@ -65,7 +63,7 @@ class PatreonIE(InfoExtractor): 'password': password, } - request = compat_urllib_request.Request( + request = sanitized_Request( 'https://www.patreon.com/processLogin', compat_urllib_parse.urlencode(login_form).encode('utf-8') ) diff --git a/youtube_dl/extractor/pbs.py b/youtube_dl/extractor/pbs.py index 3448736a2..b787e2a73 100644 --- a/youtube_dl/extractor/pbs.py +++ b/youtube_dl/extractor/pbs.py @@ -8,6 +8,7 @@ from ..utils import ( ExtractorError, determine_ext, int_or_none, + strip_jsonp, unified_strdate, US_RATINGS, ) @@ -21,7 +22,7 @@ class PBSIE(InfoExtractor): # Article with embedded player (or direct video) (?:www\.)?pbs\.org/(?:[^/]+/){2,5}(?P[^/]+?)(?:\.html)?/?(?:$|[?\#]) | # Player - video\.pbs\.org/(?:widget/)?partnerplayer/(?P[^/]+)/ + (?:video|player)\.pbs\.org/(?:widget/)?partnerplayer/(?P[^/]+)/ ) ''' @@ -153,6 +154,26 @@ class PBSIE(InfoExtractor): 'params': { 'skip_download': True, # requires ffmpeg }, + }, + { + # Frontline video embedded via flp2012.js + 'url': 'http://www.pbs.org/wgbh/pages/frontline/the-atomic-artists', + 'info_dict': { + 'id': '2070868960', + 'display_id': 'the-atomic-artists', + 'ext': 'mp4', + 'title': 'FRONTLINE - The Atomic Artists', + 'description': 'md5:f5bfbefadf421e8bb8647602011caf8e', + 'duration': 723, + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'params': { + 'skip_download': True, # requires ffmpeg + }, + }, + { + 'url': 'http://player.pbs.org/widget/partnerplayer/2365297708/?start=0&end=0&chapterbar=false&endscreen=false&topbar=true', + 'only_matching': True, } ] _ERRORS = { @@ -191,9 +212,30 @@ class PBSIE(InfoExtractor): if media_id: return media_id, presumptive_id, upload_date - url = self._search_regex( - r'(?s)]+?(?:[a-z-]+?=["\'].*?["\'][^>]+?)*?\bsrc=["\']([^\'"]+partnerplayer[^\'"]+)["\']', - webpage, 'player URL') + # Fronline video embedded via flp + video_id = self._search_regex( + r'videoid\s*:\s*"([\d+a-z]{7,})"', webpage, 'videoid', default=None) + if video_id: + # pkg_id calculation is reverse engineered from + # http://www.pbs.org/wgbh/pages/frontline/js/flp2012.js + prg_id = self._search_regex( + r'videoid\s*:\s*"([\d+a-z]{7,})"', webpage, 'videoid')[7:] + if 'q' in prg_id: + prg_id = prg_id.split('q')[1] + prg_id = int(prg_id, 16) + getdir = self._download_json( + 'http://www.pbs.org/wgbh/pages/frontline/.json/getdir/getdir%d.json' % prg_id, + presumptive_id, 'Downloading getdir JSON', + transform_source=strip_jsonp) + return getdir['mid'], presumptive_id, upload_date + + for iframe in re.findall(r'(?s)', webpage): + url = self._search_regex( + r'src=(["\'])(?P.+?partnerplayer.+?)\1', iframe, + 'player URL', default=None, group='url') + if url: + break + mobj = re.match(self._VALID_URL, url) player_id = mobj.group('player_id') @@ -221,7 +263,7 @@ class PBSIE(InfoExtractor): return self.playlist_result(entries, display_id) info = self._download_json( - 'http://video.pbs.org/videoInfo/%s?format=json&type=partner' % video_id, + 'http://player.pbs.org/videoInfo/%s?format=json&type=partner' % video_id, display_id) formats = [] diff --git a/youtube_dl/extractor/periscope.py b/youtube_dl/extractor/periscope.py index 8ad936758..63cc764bb 100644 --- a/youtube_dl/extractor/periscope.py +++ b/youtube_dl/extractor/periscope.py @@ -2,17 +2,14 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) from ..utils import parse_iso8601 class PeriscopeIE(InfoExtractor): IE_DESC = 'Periscope' - _VALID_URL = r'https?://(?:www\.)?periscope\.tv/w/(?P[^/?#]+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?periscope\.tv/[^/]+/(?P[^/?#]+)' + # Alive example URLs can be found here http://onperiscope.com/ + _TESTS = [{ 'url': 'https://www.periscope.tv/w/aJUQnjY3MjA3ODF8NTYxMDIyMDl2zCg2pECBgwTqRpQuQD352EMPTKQjT4uqlM3cgWFA-g==', 'md5': '65b57957972e503fcbbaeed8f4fa04ca', 'info_dict': { @@ -25,11 +22,18 @@ class PeriscopeIE(InfoExtractor): 'uploader_id': '1465763', }, 'skip': 'Expires in 24 hours', - } + }, { + 'url': 'https://www.periscope.tv/w/1ZkKzPbMVggJv', + 'only_matching': True, + }, { + 'url': 'https://www.periscope.tv/bastaakanoggano/1OdKrlkZZjOJX', + 'only_matching': True, + }] - def _call_api(self, method, token): + def _call_api(self, method, value): + attribute = 'token' if len(value) > 13 else 'broadcast_id' return self._download_json( - 'https://api.periscope.tv/api/v2/%s?token=%s' % (method, token), token) + 'https://api.periscope.tv/api/v2/%s?%s=%s' % (method, attribute, value), value) def _real_extract(self, url): token = self._match_id(url) @@ -76,24 +80,3 @@ class PeriscopeIE(InfoExtractor): 'thumbnails': thumbnails, 'formats': formats, } - - -class QuickscopeIE(InfoExtractor): - IE_DESC = 'Quick Scope' - _VALID_URL = r'https?://watchonperiscope\.com/broadcast/(?P\d+)' - _TEST = { - 'url': 'https://watchonperiscope.com/broadcast/56180087', - 'only_matching': True, - } - - def _real_extract(self, url): - broadcast_id = self._match_id(url) - request = compat_urllib_request.Request( - 'https://watchonperiscope.com/api/accessChannel', compat_urllib_parse.urlencode({ - 'broadcast_id': broadcast_id, - 'entry_ticket': '', - 'from_push': 'false', - 'uses_sessions': 'true', - }).encode('utf-8')) - return self.url_result( - self._download_json(request, broadcast_id)['share_url'], 'Periscope') diff --git a/youtube_dl/extractor/played.py b/youtube_dl/extractor/played.py index 8a1c296dd..2856af96f 100644 --- a/youtube_dl/extractor/played.py +++ b/youtube_dl/extractor/played.py @@ -5,12 +5,10 @@ import re import os.path from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -46,7 +44,7 @@ class PlayedIE(InfoExtractor): headers = { b'Content-Type': b'application/x-www-form-urlencoded', } - req = compat_urllib_request.Request(url, post, headers) + req = sanitized_Request(url, post, headers) webpage = self._download_webpage( req, video_id, note='Downloading video page ...') diff --git a/youtube_dl/extractor/pluralsight.py b/youtube_dl/extractor/pluralsight.py index fd32836cc..aa7dbcb63 100644 --- a/youtube_dl/extractor/pluralsight.py +++ b/youtube_dl/extractor/pluralsight.py @@ -1,29 +1,35 @@ from __future__ import unicode_literals -import re import json +import random +import collections from .common import InfoExtractor from ..compat import ( compat_str, compat_urllib_parse, - compat_urllib_request, compat_urlparse, ) from ..utils import ( ExtractorError, int_or_none, parse_duration, + sanitized_Request, ) -class PluralsightIE(InfoExtractor): +class PluralsightBaseIE(InfoExtractor): + _API_BASE = 'http://app.pluralsight.com' + + +class PluralsightIE(PluralsightBaseIE): IE_NAME = 'pluralsight' - _VALID_URL = r'https?://(?:www\.)?pluralsight\.com/training/player\?author=(?P[^&]+)&name=(?P[^&]+)(?:&mode=live)?&clip=(?P\d+)&course=(?P[^&]+)' - _LOGIN_URL = 'https://www.pluralsight.com/id/' + _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/training/player\?' + _LOGIN_URL = 'https://app.pluralsight.com/id/' + _NETRC_MACHINE = 'pluralsight' - _TEST = { + _TESTS = [{ 'url': 'http://www.pluralsight.com/training/player?author=mike-mckeown&name=hosting-sql-server-windows-azure-iaas-m7-mgmt&mode=live&clip=3&course=hosting-sql-server-windows-azure-iaas', 'md5': '4d458cf5cf4c593788672419a8dd4cf8', 'info_dict': { @@ -33,7 +39,14 @@ class PluralsightIE(InfoExtractor): 'duration': 338, }, 'skip': 'Requires pluralsight account credentials', - } + }, { + 'url': 'https://app.pluralsight.com/training/player?course=angularjs-get-started&author=scott-allen&name=angularjs-get-started-m1-introduction&clip=0&mode=live', + 'only_matching': True, + }, { + # available without pluralsight account + 'url': 'http://app.pluralsight.com/training/player?author=scott-allen&name=angularjs-get-started-m1-introduction&mode=live&clip=0&course=angularjs-get-started', + 'only_matching': True, + }] def _real_initialize(self): self._login() @@ -41,7 +54,7 @@ class PluralsightIE(InfoExtractor): def _login(self): (username, password) = self._get_login_info() if username is None: - self.raise_login_required('Pluralsight account is required') + return login_page = self._download_webpage( self._LOGIN_URL, None, 'Downloading login page') @@ -60,7 +73,7 @@ class PluralsightIE(InfoExtractor): if not post_url.startswith('http'): post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url) - request = compat_urllib_request.Request( + request = sanitized_Request( post_url, compat_urllib_parse.urlencode(login_form).encode('utf-8')) request.add_header('Content-Type', 'application/x-www-form-urlencoded') @@ -73,30 +86,47 @@ class PluralsightIE(InfoExtractor): if error: raise ExtractorError('Unable to login: %s' % error, expected=True) + if all(p not in response for p in ('__INITIAL_STATE__', '"currentUser"')): + raise ExtractorError('Unable to log in') + def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - author = mobj.group('author') - name = mobj.group('name') - clip_id = mobj.group('clip') - course = mobj.group('course') + qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + + author = qs.get('author', [None])[0] + name = qs.get('name', [None])[0] + clip_id = qs.get('clip', [None])[0] + course = qs.get('course', [None])[0] + + if any(not f for f in (author, name, clip_id, course,)): + raise ExtractorError('Invalid URL', expected=True) display_id = '%s-%s' % (name, clip_id) webpage = self._download_webpage(url, display_id) - collection = self._parse_json( - self._search_regex( - r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)', - webpage, 'modules'), - display_id) + modules = self._search_regex( + r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)', + webpage, 'modules', default=None) + + if modules: + collection = self._parse_json(modules, display_id) + else: + # Webpage may be served in different layout (see + # https://github.com/rg3/youtube-dl/issues/7607) + collection = self._parse_json( + self._search_regex( + r'var\s+initialState\s*=\s*({.+?});\n', webpage, 'initial state'), + display_id)['course']['modules'] module, clip = None, None for module_ in collection: - if module_.get('moduleName') == name: + if name in (module_.get('moduleName'), module_.get('name')): module = module_ for clip_ in module_.get('clips', []): clip_index = clip_.get('clipIndex') + if clip_index is None: + clip_index = clip_.get('index') if clip_index is None: continue if compat_str(clip_index) == clip_id: @@ -112,13 +142,33 @@ class PluralsightIE(InfoExtractor): 'high': {'width': 1024, 'height': 768}, } + AllowedQuality = collections.namedtuple('AllowedQuality', ['ext', 'qualities']) + ALLOWED_QUALITIES = ( - ('webm', ('high',)), - ('mp4', ('low', 'medium', 'high',)), + AllowedQuality('webm', ('high',)), + AllowedQuality('mp4', ('low', 'medium', 'high',)), ) + # In order to minimize the number of calls to ViewClip API and reduce + # the probability of being throttled or banned by Pluralsight we will request + # only single format until formats listing was explicitly requested. + if self._downloader.params.get('listformats', False): + allowed_qualities = ALLOWED_QUALITIES + else: + def guess_allowed_qualities(): + req_format = self._downloader.params.get('format') or 'best' + req_format_split = req_format.split('-') + if len(req_format_split) > 1: + req_ext, req_quality = req_format_split + for allowed_quality in ALLOWED_QUALITIES: + if req_ext == allowed_quality.ext and req_quality in allowed_quality.qualities: + return (AllowedQuality(req_ext, (req_quality, )), ) + req_ext = 'webm' if self._downloader.params.get('prefer_free_formats') else 'mp4' + return (AllowedQuality(req_ext, ('high', )), ) + allowed_qualities = guess_allowed_qualities() + formats = [] - for ext, qualities in ALLOWED_QUALITIES: + for ext, qualities in allowed_qualities: for quality in qualities: f = QUALITIES[quality].copy() clip_post = { @@ -131,13 +181,24 @@ class PluralsightIE(InfoExtractor): 'mt': ext, 'q': '%dx%d' % (f['width'], f['height']), } - request = compat_urllib_request.Request( - 'http://www.pluralsight.com/training/Player/ViewClip', + request = sanitized_Request( + '%s/training/Player/ViewClip' % self._API_BASE, json.dumps(clip_post).encode('utf-8')) request.add_header('Content-Type', 'application/json;charset=utf-8') format_id = '%s-%s' % (ext, quality) clip_url = self._download_webpage( request, display_id, 'Downloading %s URL' % format_id, fatal=False) + + # Pluralsight tracks multiple sequential calls to ViewClip API and start + # to return 429 HTTP errors after some time (see + # https://github.com/rg3/youtube-dl/pull/6989). Moreover it may even lead + # to account ban (see https://github.com/rg3/youtube-dl/issues/6842). + # To somewhat reduce the probability of these consequences + # we will sleep random amount of time before each call to ViewClip. + self._sleep( + random.randint(2, 5), display_id, + '%(video_id)s: Waiting for %(timeout)s seconds to avoid throttling') + if not clip_url: continue f.update({ @@ -163,10 +224,10 @@ class PluralsightIE(InfoExtractor): } -class PluralsightCourseIE(InfoExtractor): +class PluralsightCourseIE(PluralsightBaseIE): IE_NAME = 'pluralsight:course' - _VALID_URL = r'https?://(?:www\.)?pluralsight\.com/courses/(?P[^/]+)' - _TEST = { + _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/(?:library/)?courses/(?P[^/]+)' + _TESTS = [{ # Free course from Pluralsight Starter Subscription for Microsoft TechNet # https://offers.pluralsight.com/technet?loc=zTS3z&prod=zOTprodz&tech=zOttechz&prog=zOTprogz&type=zSOz&media=zOTmediaz&country=zUSz 'url': 'http://www.pluralsight.com/courses/hosting-sql-server-windows-azure-iaas', @@ -176,7 +237,14 @@ class PluralsightCourseIE(InfoExtractor): 'description': 'md5:61b37e60f21c4b2f91dc621a977d0986', }, 'playlist_count': 31, - } + }, { + # available without pluralsight account + 'url': 'https://www.pluralsight.com/courses/angularjs-get-started', + 'only_matching': True, + }, { + 'url': 'https://app.pluralsight.com/library/courses/understanding-microsoft-azure-amazon-aws/table-of-contents', + 'only_matching': True, + }] def _real_extract(self, url): course_id = self._match_id(url) @@ -184,14 +252,14 @@ class PluralsightCourseIE(InfoExtractor): # TODO: PSM cookie course = self._download_json( - 'http://www.pluralsight.com/data/course/%s' % course_id, + '%s/data/course/%s' % (self._API_BASE, course_id), course_id, 'Downloading course JSON') title = course['title'] description = course.get('description') or course.get('shortDescription') course_data = self._download_json( - 'http://www.pluralsight.com/data/course/content/%s' % course_id, + '%s/data/course/content/%s' % (self._API_BASE, course_id), course_id, 'Downloading course data JSON') entries = [] @@ -201,7 +269,7 @@ class PluralsightCourseIE(InfoExtractor): if not player_parameters: continue entries.append(self.url_result( - 'http://www.pluralsight.com/training/player?%s' % player_parameters, + '%s/training/player?%s' % (self._API_BASE, player_parameters), 'Pluralsight')) return self.playlist_result(entries, course_id, title, description) diff --git a/youtube_dl/extractor/pornhd.py b/youtube_dl/extractor/pornhd.py index dbb2c3bd9..57c78ba52 100644 --- a/youtube_dl/extractor/pornhd.py +++ b/youtube_dl/extractor/pornhd.py @@ -36,7 +36,8 @@ class PornHdIE(InfoExtractor): webpage = self._download_webpage(url, display_id or video_id) title = self._html_search_regex( - r'(.+) porn HD.+?', webpage, 'title') + [r']+class=["\']video-name["\'][^>]*>([^<]+)', + r'(.+?) - .*?[Pp]ornHD.*?'], webpage, 'title') description = self._html_search_regex( r'
([^<]+)
', webpage, 'description', fatal=False) view_count = int_or_none(self._html_search_regex( diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index a656ad85a..08275687d 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -8,10 +8,10 @@ from ..compat import ( compat_urllib_parse_unquote, compat_urllib_parse_unquote_plus, compat_urllib_parse_urlparse, - compat_urllib_request, ) from ..utils import ( ExtractorError, + sanitized_Request, str_to_int, ) from ..aes import ( @@ -53,7 +53,7 @@ class PornHubIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - req = compat_urllib_request.Request( + req = sanitized_Request( 'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) @@ -147,7 +147,8 @@ class PornHubPlaylistIE(InfoExtractor): entries = [ self.url_result('http://www.pornhub.com/%s' % video_url, 'PornHub') - for video_url in set(re.findall('href="/?(view_video\.php\?viewkey=\d+[^"]*)"', webpage)) + for video_url in set(re.findall( + r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"', webpage)) ] playlist = self._parse_json( diff --git a/youtube_dl/extractor/pornotube.py b/youtube_dl/extractor/pornotube.py index 34735c51e..5398e708b 100644 --- a/youtube_dl/extractor/pornotube.py +++ b/youtube_dl/extractor/pornotube.py @@ -3,11 +3,9 @@ from __future__ import unicode_literals import json from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( int_or_none, + sanitized_Request, ) @@ -46,7 +44,7 @@ class PornotubeIE(InfoExtractor): 'authenticationSpaceKey': originAuthenticationSpaceKey, 'credentials': 'Clip Application', } - token_req = compat_urllib_request.Request( + token_req = sanitized_Request( 'https://api.aebn.net/auth/v1/token/primal', data=json.dumps(token_req_data).encode('utf-8')) token_req.add_header('Content-Type', 'application/json') @@ -56,7 +54,7 @@ class PornotubeIE(InfoExtractor): token = token_answer['tokenKey'] # Get video URL - delivery_req = compat_urllib_request.Request( + delivery_req = sanitized_Request( 'https://api.aebn.net/delivery/v1/clips/%s/MP4' % video_id) delivery_req.add_header('Authorization', token) delivery_info = self._download_json( @@ -64,7 +62,7 @@ class PornotubeIE(InfoExtractor): video_url = delivery_info['mediaUrl'] # Get additional info (title etc.) - info_req = compat_urllib_request.Request( + info_req = sanitized_Request( 'https://api.aebn.net/content/v1/clips/%s?expand=' 'title,description,primaryImageNumber,startSecond,endSecond,' 'movie.title,movie.MovieId,movie.boxCoverFront,movie.stars,' diff --git a/youtube_dl/extractor/primesharetv.py b/youtube_dl/extractor/primesharetv.py index 304359dc5..85aae9576 100644 --- a/youtube_dl/extractor/primesharetv.py +++ b/youtube_dl/extractor/primesharetv.py @@ -1,11 +1,11 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, +from ..compat import compat_urllib_parse +from ..utils import ( + ExtractorError, + sanitized_Request, ) -from ..utils import ExtractorError class PrimeShareTVIE(InfoExtractor): @@ -41,7 +41,7 @@ class PrimeShareTVIE(InfoExtractor): webpage, 'wait time', default=7)) + 1 self._sleep(wait_time, video_id) - req = compat_urllib_request.Request( + req = sanitized_Request( url, compat_urllib_parse.urlencode(fields), headers) video_page = self._download_webpage( req, video_id, 'Downloading video page') diff --git a/youtube_dl/extractor/promptfile.py b/youtube_dl/extractor/promptfile.py index 8190ed676..d5357283a 100644 --- a/youtube_dl/extractor/promptfile.py +++ b/youtube_dl/extractor/promptfile.py @@ -4,13 +4,11 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( determine_ext, ExtractorError, + sanitized_Request, ) @@ -37,7 +35,7 @@ class PromptFileIE(InfoExtractor): fields = self._hidden_inputs(webpage) post = compat_urllib_parse.urlencode(fields) - req = compat_urllib_request.Request(url, post) + req = sanitized_Request(url, post) req.add_header('Content-type', 'application/x-www-form-urlencoded') webpage = self._download_webpage( req, video_id, 'Downloading video page') diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py index effcf1db3..baa54a3af 100644 --- a/youtube_dl/extractor/prosiebensat1.py +++ b/youtube_dl/extractor/prosiebensat1.py @@ -20,7 +20,7 @@ from ..utils import ( class ProSiebenSat1IE(InfoExtractor): IE_NAME = 'prosiebensat1' IE_DESC = 'ProSiebenSat.1 Digital' - _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany)\.(?:de|at)|ran\.de|fem\.com)/(?P.+)' + _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany)\.(?:de|at|ch)|ran\.de|fem\.com)/(?P.+)' _TESTS = [ { diff --git a/youtube_dl/extractor/qqmusic.py b/youtube_dl/extractor/qqmusic.py index c98539f6a..1ba3bbddf 100644 --- a/youtube_dl/extractor/qqmusic.py +++ b/youtube_dl/extractor/qqmusic.py @@ -7,11 +7,11 @@ import re from .common import InfoExtractor from ..utils import ( + sanitized_Request, strip_jsonp, unescapeHTML, clean_html, ) -from ..compat import compat_urllib_request class QQMusicIE(InfoExtractor): @@ -201,7 +201,7 @@ class QQMusicSingerIE(QQPlaylistBaseIE): singer_desc = None if singer_id: - req = compat_urllib_request.Request( + req = sanitized_Request( 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id) req.add_header( 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html') diff --git a/youtube_dl/extractor/rtbf.py b/youtube_dl/extractor/rtbf.py index 04a66df90..e42b319a3 100644 --- a/youtube_dl/extractor/rtbf.py +++ b/youtube_dl/extractor/rtbf.py @@ -9,8 +9,8 @@ from ..utils import ( class RTBFIE(InfoExtractor): - _VALID_URL = r'https?://www.rtbf.be/video/[^\?]+\?id=(?P\d+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?rtbf\.be/(?:video/[^?]+\?.*\bid=|ouftivi/(?:[^/]+/)*[^?]+\?.*\bvideoId=)(?P\d+)' + _TESTS = [{ 'url': 'https://www.rtbf.be/video/detail_les-diables-au-coeur-episode-2?id=1921274', 'md5': '799f334ddf2c0a582ba80c44655be570', 'info_dict': { @@ -19,7 +19,14 @@ class RTBFIE(InfoExtractor): 'title': 'Les Diables au coeur (épisode 2)', 'duration': 3099, } - } + }, { + # geo restricted + 'url': 'http://www.rtbf.be/ouftivi/heros/detail_scooby-doo-mysteres-associes?id=1097&videoId=2057442', + 'only_matching': True, + }, { + 'url': 'http://www.rtbf.be/ouftivi/niouzz?videoId=2055858', + 'only_matching': True, + }] _QUALITIES = [ ('mobile', 'mobile'), diff --git a/youtube_dl/extractor/rte.py b/youtube_dl/extractor/rte.py index 04158b993..d9cfbf180 100644 --- a/youtube_dl/extractor/rte.py +++ b/youtube_dl/extractor/rte.py @@ -9,16 +9,16 @@ from ..utils import ( class RteIE(InfoExtractor): - _VALID_URL = r'http?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/(?P[0-9]+)/' + _VALID_URL = r'https?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/[^/]+/(?P[0-9]+)' _TEST = { - 'url': 'http://www.rte.ie/player/de/show/10363114/', + 'url': 'http://www.rte.ie/player/ie/show/iwitness-862/10478715/', 'info_dict': { - 'id': '10363114', + 'id': '10478715', 'ext': 'mp4', - 'title': 'One News', + 'title': 'Watch iWitness online', 'thumbnail': 're:^https?://.*\.jpg$', - 'description': 'The One O\'Clock News followed by Weather.', - 'duration': 436.844, + 'description': 'iWitness : The spirit of Ireland, one voice and one minute at a time.', + 'duration': 60.046, }, 'params': { 'skip_download': 'f4m fails with --test atm' diff --git a/youtube_dl/extractor/rtve.py b/youtube_dl/extractor/rtve.py index 5b97d33ca..603d7bd00 100644 --- a/youtube_dl/extractor/rtve.py +++ b/youtube_dl/extractor/rtve.py @@ -6,11 +6,11 @@ import re import time from .common import InfoExtractor -from ..compat import compat_urllib_request, compat_urlparse from ..utils import ( ExtractorError, float_or_none, remove_end, + sanitized_Request, std_headers, struct_unpack, ) @@ -102,20 +102,14 @@ class RTVEALaCartaIE(InfoExtractor): if info['state'] == 'DESPU': raise ExtractorError('The video is no longer available', expected=True) png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/%s/videos/%s.png' % (self._manager, video_id) - png_request = compat_urllib_request.Request(png_url) + png_request = sanitized_Request(png_url) png_request.add_header('Referer', url) png = self._download_webpage(png_request, video_id, 'Downloading url information') video_url = _decrypt_url(png) if not video_url.endswith('.f4m'): - auth_url = video_url.replace( + video_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 mvod1.akcdn instead of flash.akamaihd.multimedia.cdn to get - # the right Content-Length header and the mp4 format - video_url = compat_urlparse.urljoin( - 'http://mvod1.akcdn.rtve.es/', video_path) subtitles = None if info.get('sbtFile') is not None: diff --git a/youtube_dl/extractor/rutube.py b/youtube_dl/extractor/rutube.py index d94dc7399..6b09550b0 100644 --- a/youtube_dl/extractor/rutube.py +++ b/youtube_dl/extractor/rutube.py @@ -9,7 +9,7 @@ from ..compat import ( compat_str, ) from ..utils import ( - ExtractorError, + determine_ext, unified_strdate, ) @@ -51,10 +51,25 @@ class RutubeIE(InfoExtractor): 'http://rutube.ru/api/play/options/%s/?format=json' % video_id, video_id, 'Downloading options JSON') - m3u8_url = options['video_balancer'].get('m3u8') - if m3u8_url is None: - raise ExtractorError('Couldn\'t find m3u8 manifest url') - formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4') + formats = [] + for format_id, format_url in options['video_balancer'].items(): + ext = determine_ext(format_url) + if ext == 'm3u8': + m3u8_formats = self._extract_m3u8_formats( + format_url, video_id, 'mp4', m3u8_id=format_id, fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + elif ext == 'f4m': + f4m_formats = self._extract_f4m_formats( + format_url, video_id, f4m_id=format_id, fatal=False) + if f4m_formats: + formats.extend(f4m_formats) + else: + formats.append({ + 'url': format_url, + 'format_id': format_id, + }) + self._sort_formats(formats) return { 'id': video['id'], @@ -74,9 +89,9 @@ class RutubeIE(InfoExtractor): class RutubeEmbedIE(InfoExtractor): IE_NAME = 'rutube:embed' IE_DESC = 'Rutube embedded videos' - _VALID_URL = 'https?://rutube\.ru/video/embed/(?P[0-9]+)' + _VALID_URL = 'https?://rutube\.ru/(?:video|play)/embed/(?P[0-9]+)' - _TEST = { + _TESTS = [{ 'url': 'http://rutube.ru/video/embed/6722881?vk_puid37=&vk_puid38=', 'info_dict': { 'id': 'a10e53b86e8f349080f718582ce4c661', @@ -90,7 +105,10 @@ class RutubeEmbedIE(InfoExtractor): 'params': { 'skip_download': 'Requires ffmpeg', }, - } + }, { + 'url': 'http://rutube.ru/play/embed/8083783', + 'only_matching': True, + }] def _real_extract(self, url): embed_id = self._match_id(url) diff --git a/youtube_dl/extractor/ruutu.py b/youtube_dl/extractor/ruutu.py index a16b73ff4..e417bf661 100644 --- a/youtube_dl/extractor/ruutu.py +++ b/youtube_dl/extractor/ruutu.py @@ -57,16 +57,21 @@ class RuutuIE(InfoExtractor): extract_formats(child) elif child.tag.endswith('File'): video_url = child.text - if not video_url or video_url in processed_urls or 'NOT_USED' in video_url: + if (not video_url or video_url in processed_urls or + any(p in video_url for p in ('NOT_USED', 'NOT-USED'))): return processed_urls.append(video_url) ext = determine_ext(video_url) if ext == 'm3u8': - formats.extend(self._extract_m3u8_formats( - video_url, video_id, 'mp4', m3u8_id='hls')) + m3u8_formats = self._extract_m3u8_formats( + video_url, video_id, 'mp4', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) elif ext == 'f4m': - formats.extend(self._extract_f4m_formats( - video_url, video_id, f4m_id='hds')) + f4m_formats = self._extract_f4m_formats( + video_url, video_id, f4m_id='hds', fatal=False) + if f4m_formats: + formats.extend(f4m_formats) else: proto = compat_urllib_parse_urlparse(video_url).scheme if not child.tag.startswith('HTTP') and proto != 'rtmp': diff --git a/youtube_dl/extractor/safari.py b/youtube_dl/extractor/safari.py index a602af692..919704261 100644 --- a/youtube_dl/extractor/safari.py +++ b/youtube_dl/extractor/safari.py @@ -4,14 +4,12 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from .brightcove import BrightcoveIE +from .brightcove import BrightcoveLegacyIE -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, + sanitized_Request, smuggle_url, std_headers, ) @@ -58,7 +56,7 @@ class SafariBaseIE(InfoExtractor): 'next': '', } - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form), headers=headers) login_page = self._download_webpage( request, None, 'Logging in as %s' % username) @@ -112,11 +110,11 @@ class SafariIE(SafariBaseIE): '%s/%s/chapter-content/%s.html' % (self._API_BASE, course_id, part), part) - bc_url = BrightcoveIE._extract_brightcove_url(webpage) + bc_url = BrightcoveLegacyIE._extract_brightcove_url(webpage) if not bc_url: raise ExtractorError('Could not extract Brightcove URL from %s' % url, expected=True) - return self.url_result(smuggle_url(bc_url, {'Referer': url}), 'Brightcove') + return self.url_result(smuggle_url(bc_url, {'Referer': url}), 'BrightcoveLegacy') class SafariCourseIE(SafariBaseIE): diff --git a/youtube_dl/extractor/sandia.py b/youtube_dl/extractor/sandia.py index 9c88167f0..759898a49 100644 --- a/youtube_dl/extractor/sandia.py +++ b/youtube_dl/extractor/sandia.py @@ -6,14 +6,12 @@ import json import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( int_or_none, js_to_json, mimetype2ext, + sanitized_Request, unified_strdate, ) @@ -37,7 +35,7 @@ class SandiaIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('Cookie', 'MediasitePlayerCaps=ClientPlugins=4') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/senateisvp.py b/youtube_dl/extractor/senateisvp.py index 9c53704ea..474ebb49b 100644 --- a/youtube_dl/extractor/senateisvp.py +++ b/youtube_dl/extractor/senateisvp.py @@ -121,9 +121,9 @@ class SenateISVPIE(InfoExtractor): 'url': compat_urlparse.urljoin(domain, filename) + '?v=3.1.0&fp=&r=&g=', }] else: - hdcore_sign = '?hdcore=3.1.0' + hdcore_sign = 'hdcore=3.1.0' url_params = (domain, video_id, stream_num) - f4m_url = '%s/z/%s_1@%s/manifest.f4m' % url_params + hdcore_sign + f4m_url = '%s/z/%s_1@%s/manifest.f4m?' % url_params + hdcore_sign m3u8_url = '%s/i/%s_1@%s/master.m3u8' % url_params for entry in self._extract_f4m_formats(f4m_url, video_id, f4m_id='f4m'): # URLs without the extra param induce an 404 error diff --git a/youtube_dl/extractor/shared.py b/youtube_dl/extractor/shared.py index c5636e8e9..8eda3c864 100644 --- a/youtube_dl/extractor/shared.py +++ b/youtube_dl/extractor/shared.py @@ -3,13 +3,11 @@ from __future__ import unicode_literals import base64 from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, int_or_none, + sanitized_Request, ) @@ -46,7 +44,7 @@ class SharedIE(InfoExtractor): 'Video %s does not exist' % video_id, expected=True) download_form = self._hidden_inputs(webpage) - request = compat_urllib_request.Request( + request = sanitized_Request( url, compat_urllib_parse.urlencode(download_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') diff --git a/youtube_dl/extractor/sharesix.py b/youtube_dl/extractor/sharesix.py index ac3e3adf2..f1ea9bdb2 100644 --- a/youtube_dl/extractor/sharesix.py +++ b/youtube_dl/extractor/sharesix.py @@ -4,12 +4,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( parse_duration, + sanitized_Request, ) @@ -50,7 +48,7 @@ class ShareSixIE(InfoExtractor): 'method_free': 'Free' } post = compat_urllib_parse.urlencode(fields) - req = compat_urllib_request.Request(url, post) + req = sanitized_Request(url, post) req.add_header('Content-type', 'application/x-www-form-urlencoded') webpage = self._download_webpage(req, video_id, diff --git a/youtube_dl/extractor/sina.py b/youtube_dl/extractor/sina.py index 0891a441f..b2258a0f6 100644 --- a/youtube_dl/extractor/sina.py +++ b/youtube_dl/extractor/sina.py @@ -4,10 +4,8 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urllib_parse, -) +from ..compat import compat_urllib_parse +from ..utils import sanitized_Request class SinaIE(InfoExtractor): @@ -61,7 +59,7 @@ class SinaIE(InfoExtractor): if mobj.group('token') is not None: # The video id is in the redirected url self.to_screen('Getting video id') - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) request.get_method = lambda: 'HEAD' (_, urlh) = self._download_webpage_handle(request, 'NA', False) return self._real_extract(urlh.geturl()) diff --git a/youtube_dl/extractor/skynewsarabia.py b/youtube_dl/extractor/skynewsarabia.py new file mode 100644 index 000000000..f09fee102 --- /dev/null +++ b/youtube_dl/extractor/skynewsarabia.py @@ -0,0 +1,117 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_str +from ..utils import ( + parse_iso8601, + parse_duration, +) + + +class SkyNewArabiaBaseIE(InfoExtractor): + _IMAGE_BASE_URL = 'http://www.skynewsarabia.com/web/images' + + def _call_api(self, path, value): + return self._download_json('http://api.skynewsarabia.com/web/rest/v2/%s/%s.json' % (path, value), value) + + def _get_limelight_media_id(self, url): + return self._search_regex(r'/media/[^/]+/([a-z0-9]{32})', url, 'limelight media id') + + def _get_image_url(self, image_path_template, width='1600', height='1200'): + return self._IMAGE_BASE_URL + image_path_template.format(width=width, height=height) + + def _extract_video_info(self, video_data): + video_id = compat_str(video_data['id']) + topic = video_data.get('topicTitle') + return { + '_type': 'url_transparent', + 'url': 'limelight:media:%s' % self._get_limelight_media_id(video_data['videoUrl'][0]['url']), + 'id': video_id, + 'title': video_data['headline'], + 'description': video_data.get('summary'), + 'thumbnail': self._get_image_url(video_data['mediaAsset']['imageUrl']), + 'timestamp': parse_iso8601(video_data.get('date')), + 'duration': parse_duration(video_data.get('runTime')), + 'tags': video_data.get('tags', []), + 'categories': [topic] if topic else [], + 'webpage_url': 'http://www.skynewsarabia.com/web/video/%s' % video_id, + 'ie_key': 'LimelightMedia', + } + + +class SkyNewsArabiaIE(SkyNewArabiaBaseIE): + IE_NAME = 'skynewsarabia:video' + _VALID_URL = r'https?://(?:www\.)?skynewsarabia\.com/web/video/(?P[0-9]+)' + _TEST = { + 'url': 'http://www.skynewsarabia.com/web/video/794902/%D9%86%D8%B5%D9%81-%D9%85%D9%84%D9%8A%D9%88%D9%86-%D9%85%D8%B5%D8%A8%D8%A7%D8%AD-%D8%B4%D8%AC%D8%B1%D8%A9-%D9%83%D8%B1%D9%8A%D8%B3%D9%85%D8%A7%D8%B3', + 'info_dict': { + 'id': '794902', + 'ext': 'flv', + 'title': 'نصف مليون مصباح على شجرة كريسماس', + 'description': 'md5:22f1b27f0850eeb10c7e59b1f16eb7c6', + 'upload_date': '20151128', + 'timestamp': 1448697198, + 'duration': 2119, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + video_data = self._call_api('video', video_id) + return self._extract_video_info(video_data) + + +class SkyNewsArabiaArticleIE(SkyNewArabiaBaseIE): + IE_NAME = 'skynewsarabia:video' + _VALID_URL = r'https?://(?:www\.)?skynewsarabia\.com/web/article/(?P[0-9]+)' + _TESTS = [{ + 'url': 'http://www.skynewsarabia.com/web/article/794549/%D8%A7%D9%94%D8%AD%D8%AF%D8%A7%D8%AB-%D8%A7%D9%84%D8%B4%D8%B1%D9%82-%D8%A7%D9%84%D8%A7%D9%94%D9%88%D8%B3%D8%B7-%D8%AE%D8%B1%D9%8A%D8%B7%D8%A9-%D8%A7%D9%84%D8%A7%D9%94%D9%84%D8%B9%D8%A7%D8%A8-%D8%A7%D9%84%D8%B0%D9%83%D9%8A%D8%A9', + 'info_dict': { + 'id': '794549', + 'ext': 'flv', + 'title': 'بالفيديو.. ألعاب ذكية تحاكي واقع المنطقة', + 'description': 'md5:0c373d29919a851e080ee4edd0c5d97f', + 'upload_date': '20151126', + 'timestamp': 1448559336, + 'duration': 281.6, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }, { + 'url': 'http://www.skynewsarabia.com/web/article/794844/%D8%A7%D8%B3%D8%AA%D9%87%D8%AF%D8%A7%D9%81-%D9%82%D9%88%D8%A7%D8%B1%D8%A8-%D8%A7%D9%94%D8%B3%D9%84%D8%AD%D8%A9-%D9%84%D9%85%D9%8A%D9%84%D9%8A%D8%B4%D9%8A%D8%A7%D8%AA-%D8%A7%D9%84%D8%AD%D9%88%D8%AB%D9%8A-%D9%88%D8%B5%D8%A7%D9%84%D8%AD', + 'info_dict': { + 'id': '794844', + 'title': 'إحباط تهريب أسلحة لميليشيات الحوثي وصالح بجنوب اليمن', + 'description': 'md5:5c927b8b2e805796e7f693538d96fc7e', + }, + 'playlist_mincount': 2, + }] + + def _real_extract(self, url): + article_id = self._match_id(url) + article_data = self._call_api('article', article_id) + media_asset = article_data['mediaAsset'] + if media_asset['type'] == 'VIDEO': + topic = article_data.get('topicTitle') + return { + '_type': 'url_transparent', + 'url': 'limelight:media:%s' % self._get_limelight_media_id(media_asset['videoUrl'][0]['url']), + 'id': article_id, + 'title': article_data['headline'], + 'description': article_data.get('summary'), + 'thumbnail': self._get_image_url(media_asset['imageUrl']), + 'timestamp': parse_iso8601(article_data.get('date')), + 'tags': article_data.get('tags', []), + 'categories': [topic] if topic else [], + 'webpage_url': url, + 'ie_key': 'LimelightMedia', + } + entries = [self._extract_video_info(item) for item in article_data.get('inlineItems', []) if item['type'] == 'VIDEO'] + return self.playlist_result(entries, article_id, article_data['headline'], article_data.get('summary')) diff --git a/youtube_dl/extractor/smotri.py b/youtube_dl/extractor/smotri.py index 35a81ee87..30210c8a3 100644 --- a/youtube_dl/extractor/smotri.py +++ b/youtube_dl/extractor/smotri.py @@ -7,13 +7,11 @@ import hashlib import uuid from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, int_or_none, + sanitized_Request, unified_strdate, ) @@ -176,7 +174,7 @@ class SmotriIE(InfoExtractor): if video_password: video_form['pass'] = hashlib.md5(video_password.encode('utf-8')).hexdigest() - request = compat_urllib_request.Request( + request = sanitized_Request( 'http://smotri.com/video/view/url/bot/', compat_urllib_parse.urlencode(video_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') @@ -339,7 +337,7 @@ class SmotriBroadcastIE(InfoExtractor): 'password': password, } - request = compat_urllib_request.Request( + request = sanitized_Request( broadcast_url + '/?no_redirect=1', compat_urllib_parse.urlencode(login_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') broadcast_page = self._download_webpage( diff --git a/youtube_dl/extractor/sohu.py b/youtube_dl/extractor/sohu.py index ba2d5e19b..daf6ad555 100644 --- a/youtube_dl/extractor/sohu.py +++ b/youtube_dl/extractor/sohu.py @@ -6,11 +6,11 @@ import re from .common import InfoExtractor from ..compat import ( compat_str, - compat_urllib_request, compat_urllib_parse, ) from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -96,7 +96,7 @@ class SohuIE(InfoExtractor): else: base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid=' - req = compat_urllib_request.Request(base_data_url + vid_id) + req = sanitized_Request(base_data_url + vid_id) cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') if cn_verification_proxy: diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index 2b60d354a..02e64e094 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -4,13 +4,17 @@ from __future__ import unicode_literals import re import itertools -from .common import InfoExtractor +from .common import ( + InfoExtractor, + SearchInfoExtractor +) from ..compat import ( compat_str, compat_urlparse, compat_urllib_parse, ) from ..utils import ( + encode_dict, ExtractorError, int_or_none, unified_strdate, @@ -469,3 +473,60 @@ class SoundcloudPlaylistIE(SoundcloudIE): 'description': data.get('description'), 'entries': entries, } + + +class SoundcloudSearchIE(SearchInfoExtractor, SoundcloudIE): + IE_NAME = 'soundcloud:search' + IE_DESC = 'Soundcloud search' + _MAX_RESULTS = float('inf') + _TESTS = [{ + 'url': 'scsearch15:post-avant jazzcore', + 'info_dict': { + 'title': 'post-avant jazzcore', + }, + 'playlist_count': 15, + }] + + _SEARCH_KEY = 'scsearch' + _MAX_RESULTS_PER_PAGE = 200 + _DEFAULT_RESULTS_PER_PAGE = 50 + _API_V2_BASE = 'https://api-v2.soundcloud.com' + + def _get_collection(self, endpoint, collection_id, **query): + limit = min( + query.get('limit', self._DEFAULT_RESULTS_PER_PAGE), + self._MAX_RESULTS_PER_PAGE) + query['limit'] = limit + query['client_id'] = self._CLIENT_ID + query['linked_partitioning'] = '1' + query['offset'] = 0 + data = compat_urllib_parse.urlencode(encode_dict(query)) + next_url = '{0}{1}?{2}'.format(self._API_V2_BASE, endpoint, data) + + collected_results = 0 + + for i in itertools.count(1): + response = self._download_json( + next_url, collection_id, 'Downloading page {0}'.format(i), + 'Unable to download API page') + + collection = response.get('collection', []) + if not collection: + break + + collection = list(filter(bool, collection)) + collected_results += len(collection) + + for item in collection: + yield self.url_result(item['uri'], SoundcloudIE.ie_key()) + + if not collection or collected_results >= limit: + break + + next_url = response.get('next_href') + if not next_url: + break + + def _get_n_results(self, query, n): + tracks = self._get_collection('/search/tracks', query, limit=n, q=query) + return self.playlist_result(tracks, playlist_title=query) diff --git a/youtube_dl/extractor/space.py b/youtube_dl/extractor/space.py index c2d0d36a6..ebb5d6ec0 100644 --- a/youtube_dl/extractor/space.py +++ b/youtube_dl/extractor/space.py @@ -3,14 +3,14 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from .brightcove import BrightcoveIE +from .brightcove import BrightcoveLegacyIE from ..utils import RegexNotFoundError, ExtractorError class SpaceIE(InfoExtractor): _VALID_URL = r'https?://(?:(?:www|m)\.)?space\.com/\d+-(?P[^/\.\?]*?)-video\.html' _TEST = { - 'add_ie': ['Brightcove'], + 'add_ie': ['BrightcoveLegacy'], 'url': 'http://www.space.com/23373-huge-martian-landforms-detail-revealed-by-european-probe-video.html', 'info_dict': { 'id': '2780937028001', @@ -31,8 +31,8 @@ class SpaceIE(InfoExtractor): brightcove_url = self._og_search_video_url(webpage) except RegexNotFoundError: # Other videos works fine with the info from the object - brightcove_url = BrightcoveIE._extract_brightcove_url(webpage) + brightcove_url = BrightcoveLegacyIE._extract_brightcove_url(webpage) if brightcove_url is None: raise ExtractorError( 'The webpage does not contain a video', expected=True) - return self.url_result(brightcove_url, BrightcoveIE.ie_key()) + return self.url_result(brightcove_url, BrightcoveLegacyIE.ie_key()) diff --git a/youtube_dl/extractor/spankwire.py b/youtube_dl/extractor/spankwire.py index 9e8fb35b2..692fd78e8 100644 --- a/youtube_dl/extractor/spankwire.py +++ b/youtube_dl/extractor/spankwire.py @@ -6,9 +6,9 @@ from .common import InfoExtractor from ..compat import ( compat_urllib_parse_unquote, compat_urllib_parse_urlparse, - compat_urllib_request, ) from ..utils import ( + sanitized_Request, str_to_int, unified_strdate, ) @@ -51,7 +51,7 @@ class SpankwireIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') - req = compat_urllib_request.Request('http://www.' + mobj.group('url')) + req = sanitized_Request('http://www.' + mobj.group('url')) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/spiegel.py b/youtube_dl/extractor/spiegel.py index 5bd3c0087..39a7aaf9d 100644 --- a/youtube_dl/extractor/spiegel.py +++ b/youtube_dl/extractor/spiegel.py @@ -58,7 +58,8 @@ class SpiegelIE(InfoExtractor): description = self._html_search_meta('description', webpage, 'description') base_url = self._search_regex( - r'var\s+server\s*=\s*"([^"]+)\"', webpage, 'server URL') + [r'server\s*:\s*(["\'])(?P<url>.+?)\1', r'var\s+server\s*=\s*"(?P<url>[^"]+)\"'], + webpage, 'server URL', group='url') xml_url = base_url + video_id + '.xml' idoc = self._download_xml(xml_url, video_id) diff --git a/youtube_dl/extractor/spiegeltv.py b/youtube_dl/extractor/spiegeltv.py index 27f4033c5..034bd47ff 100644 --- a/youtube_dl/extractor/spiegeltv.py +++ b/youtube_dl/extractor/spiegeltv.py @@ -77,17 +77,21 @@ class SpiegeltvIE(InfoExtractor): 'rtmp_live': True, }) elif determine_ext(endpoint) == 'm3u8': - m3u8_formats = self._extract_m3u8_formats( - endpoint.replace('[video]', play_path), - video_id, 'm4v', - preference=1, # Prefer hls since it allows to workaround georestriction - m3u8_id='hls', fatal=False) - if m3u8_formats is not False: - formats.extend(m3u8_formats) + formats.append({ + 'url': endpoint.replace('[video]', play_path), + 'ext': 'm4v', + 'format_id': 'hls', # Prefer hls since it allows to workaround georestriction + 'protocol': 'm3u8', + 'preference': 1, + 'http_headers': { + 'Accept-Encoding': 'deflate', # gzip causes trouble on the server side + }, + }) else: formats.append({ 'url': endpoint, }) + self._check_formats(formats, video_id) thumbnails = [] for image in media_json['images']: diff --git a/youtube_dl/extractor/sportdeutschland.py b/youtube_dl/extractor/sportdeutschland.py index 7ec6c613f..ebb75f059 100644 --- a/youtube_dl/extractor/sportdeutschland.py +++ b/youtube_dl/extractor/sportdeutschland.py @@ -4,11 +4,9 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( parse_iso8601, + sanitized_Request, ) @@ -54,7 +52,7 @@ class SportDeutschlandIE(InfoExtractor): api_url = 'http://proxy.vidibusdynamic.net/sportdeutschland.tv/api/permalinks/%s/%s?access_token=true' % ( sport_id, video_id) - req = compat_urllib_request.Request(api_url, headers={ + req = sanitized_Request(api_url, headers={ 'Accept': 'application/vnd.vidibus.v2.html+json', 'Referer': url, }) diff --git a/youtube_dl/extractor/stitcher.py b/youtube_dl/extractor/stitcher.py new file mode 100644 index 000000000..d5c852f52 --- /dev/null +++ b/youtube_dl/extractor/stitcher.py @@ -0,0 +1,81 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + determine_ext, + int_or_none, + js_to_json, + unescapeHTML, +) + + +class StitcherIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?stitcher\.com/podcast/(?:[^/]+/)+e/(?:(?P<display_id>[^/#?&]+?)-)?(?P<id>\d+)(?:[/#?&]|$)' + _TESTS = [{ + 'url': 'http://www.stitcher.com/podcast/the-talking-machines/e/40789481?autoplay=true', + 'md5': '391dd4e021e6edeb7b8e68fbf2e9e940', + 'info_dict': { + 'id': '40789481', + 'ext': 'mp3', + 'title': 'Machine Learning Mastery and Cancer Clusters', + 'description': 'md5:55163197a44e915a14a1ac3a1de0f2d3', + 'duration': 1604, + 'thumbnail': 're:^https?://.*\.jpg', + }, + }, { + 'url': 'http://www.stitcher.com/podcast/panoply/vulture-tv/e/the-rare-hourlong-comedy-plus-40846275?autoplay=true', + 'info_dict': { + 'id': '40846275', + 'display_id': 'the-rare-hourlong-comedy-plus', + 'ext': 'mp3', + 'title': "The CW's 'Crazy Ex-Girlfriend'", + 'description': 'md5:04f1e2f98eb3f5cbb094cea0f9e19b17', + 'duration': 2235, + 'thumbnail': 're:^https?://.*\.jpg', + }, + 'params': { + 'skip_download': True, + }, + }, { + # escaped title + 'url': 'http://www.stitcher.com/podcast/marketplace-on-stitcher/e/40910226?autoplay=true', + 'only_matching': True, + }, { + 'url': 'http://www.stitcher.com/podcast/panoply/getting-in/e/episode-2a-how-many-extracurriculars-should-i-have-40876278?autoplay=true', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + audio_id = mobj.group('id') + display_id = mobj.group('display_id') or audio_id + + webpage = self._download_webpage(url, display_id) + + episode = self._parse_json( + js_to_json(self._search_regex( + r'(?s)var\s+stitcher\s*=\s*({.+?});\n', webpage, 'episode config')), + display_id)['config']['episode'] + + title = unescapeHTML(episode['title']) + formats = [{ + 'url': episode[episode_key], + 'ext': determine_ext(episode[episode_key]) or 'mp3', + 'vcodec': 'none', + } for episode_key in ('episodeURL',) if episode.get(episode_key)] + description = self._search_regex( + r'Episode Info:\s*</span>([^<]+)<', webpage, 'description', fatal=False) + duration = int_or_none(episode.get('duration')) + thumbnail = episode.get('episodeImage') + + return { + 'id': audio_id, + 'display_id': display_id, + 'title': title, + 'description': description, + 'duration': duration, + 'thumbnail': thumbnail, + 'formats': formats, + } diff --git a/youtube_dl/extractor/streamcloud.py b/youtube_dl/extractor/streamcloud.py index d4e134015..77841b946 100644 --- a/youtube_dl/extractor/streamcloud.py +++ b/youtube_dl/extractor/streamcloud.py @@ -4,10 +4,8 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse +from ..utils import sanitized_Request class StreamcloudIE(InfoExtractor): @@ -43,7 +41,7 @@ class StreamcloudIE(InfoExtractor): headers = { b'Content-Type': b'application/x-www-form-urlencoded', } - req = compat_urllib_request.Request(url, post, headers) + req = sanitized_Request(url, post, headers) webpage = self._download_webpage( req, video_id, note='Downloading video page ...') diff --git a/youtube_dl/extractor/streamcz.py b/youtube_dl/extractor/streamcz.py index e92b93285..d3d2b7eb7 100644 --- a/youtube_dl/extractor/streamcz.py +++ b/youtube_dl/extractor/streamcz.py @@ -5,11 +5,9 @@ import hashlib import time from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( int_or_none, + sanitized_Request, ) @@ -54,7 +52,7 @@ class StreamCZIE(InfoExtractor): video_id = self._match_id(url) api_path = '/episode/%s' % video_id - req = compat_urllib_request.Request(self._API_URL + api_path) + req = sanitized_Request(self._API_URL + api_path) req.add_header('Api-Password', _get_api_key(api_path)) data = self._download_json(req, video_id) diff --git a/youtube_dl/extractor/tapely.py b/youtube_dl/extractor/tapely.py index 744f9db38..ed560bd24 100644 --- a/youtube_dl/extractor/tapely.py +++ b/youtube_dl/extractor/tapely.py @@ -4,14 +4,12 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( clean_html, ExtractorError, float_or_none, parse_iso8601, + sanitized_Request, ) @@ -53,7 +51,7 @@ class TapelyIE(InfoExtractor): display_id = mobj.group('id') playlist_url = self._API_URL.format(display_id) - request = compat_urllib_request.Request(playlist_url) + request = sanitized_Request(playlist_url) request.add_header('X-Requested-With', 'XMLHttpRequest') request.add_header('Accept', 'application/json') request.add_header('Referer', url) diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index 25edc3100..1555aa77c 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -139,6 +139,11 @@ class ThePlatformIE(ThePlatformBaseIE): 'upload_date': '20150701', 'categories': ['Today/Shows/Orange Room', 'Today/Sections/Money', 'Today/Topics/Tech', "Today/Topics/Editor's picks"], }, + }, { + # From http://www.nbc.com/the-blacklist/video/sir-crispin-crandall/2928790?onid=137781#vc137781=1 + # geo-restricted (US), HLS encrypted with AES-128 + 'url': 'http://player.theplatform.com/p/NnzsPC/onsite_universal/select/media/guid/2410887629/2928790?fwsitesection=nbc_the_blacklist_video_library&autoPlay=true&carouselID=137781', + 'only_matching': True, }] @staticmethod @@ -182,8 +187,12 @@ class ThePlatformIE(ThePlatformBaseIE): # Seems there's no pattern for the interested script filename, so # I try one by one for script in reversed(scripts): - feed_script = self._download_webpage(script, video_id, 'Downloading feed script') - feed_id = self._search_regex(r'defaultFeedId\s*:\s*"([^"]+)"', feed_script, 'default feed id', default=None) + feed_script = self._download_webpage( + self._proto_relative_url(script, 'http:'), + video_id, 'Downloading feed script') + feed_id = self._search_regex( + r'defaultFeedId\s*:\s*"([^"]+)"', feed_script, + 'default feed id', default=None) if feed_id is not None: break if feed_id is None: @@ -193,6 +202,15 @@ class ThePlatformIE(ThePlatformBaseIE): if smuggled_data.get('force_smil_url', False): smil_url = url + # Explicitly specified SMIL (see https://github.com/rg3/youtube-dl/issues/7385) + elif '/guid/' in url: + webpage = self._download_webpage(url, video_id) + smil_url = self._search_regex( + r'<link[^>]+href=(["\'])(?P<url>.+?)\1[^>]+type=["\']application/smil\+xml', + webpage, 'smil url', group='url') + path = self._search_regex( + r'link\.theplatform\.com/s/((?:[^/?#&]+/)+[^/?#&]+)', smil_url, 'path') + smil_url += '?' if '?' not in smil_url else '&' + 'formats=m3u,mpeg4&format=SMIL' elif mobj.group('config'): config_url = url + '&form=json' config_url = config_url.replace('swf/', 'config/') diff --git a/youtube_dl/extractor/tlc.py b/youtube_dl/extractor/tlc.py index 13263614c..d6d038a8d 100644 --- a/youtube_dl/extractor/tlc.py +++ b/youtube_dl/extractor/tlc.py @@ -3,7 +3,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from .brightcove import BrightcoveIE +from .brightcove import BrightcoveLegacyIE from .discovery import DiscoveryIE from ..compat import compat_urlparse @@ -66,6 +66,6 @@ class TlcDeIE(InfoExtractor): return { '_type': 'url', - 'url': BrightcoveIE._extract_brightcove_url(iframe), - 'ie': BrightcoveIE.ie_key(), + 'url': BrightcoveLegacyIE._extract_brightcove_url(iframe), + 'ie': BrightcoveLegacyIE.ie_key(), } diff --git a/youtube_dl/extractor/tube8.py b/youtube_dl/extractor/tube8.py index c9cb69333..46ef61ff5 100644 --- a/youtube_dl/extractor/tube8.py +++ b/youtube_dl/extractor/tube8.py @@ -4,12 +4,10 @@ import json import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_urlparse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse_urlparse from ..utils import ( int_or_none, + sanitized_Request, str_to_int, ) from ..aes import aes_decrypt_text @@ -42,7 +40,7 @@ class Tube8IE(InfoExtractor): video_id = mobj.group('id') display_id = mobj.group('display_id') - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, display_id) diff --git a/youtube_dl/extractor/tubitv.py b/youtube_dl/extractor/tubitv.py index 4f86b3ee9..6d78b5dfe 100644 --- a/youtube_dl/extractor/tubitv.py +++ b/youtube_dl/extractor/tubitv.py @@ -5,13 +5,11 @@ import codecs import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, int_or_none, + sanitized_Request, ) @@ -44,7 +42,7 @@ class TubiTvIE(InfoExtractor): 'password': password, } payload = compat_urllib_parse.urlencode(form_data).encode('utf-8') - request = compat_urllib_request.Request(self._LOGIN_URL, payload) + request = sanitized_Request(self._LOGIN_URL, payload) request.add_header('Content-Type', 'application/x-www-form-urlencoded') login_page = self._download_webpage( request, None, False, 'Wrong login info') diff --git a/youtube_dl/extractor/tutv.py b/youtube_dl/extractor/tutv.py index fad720b68..822372ea1 100644 --- a/youtube_dl/extractor/tutv.py +++ b/youtube_dl/extractor/tutv.py @@ -10,10 +10,10 @@ class TutvIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?tu\.tv/videos/(?P<id>[^/?]+)' _TEST = { 'url': 'http://tu.tv/videos/robots-futbolistas', - 'md5': '627c7c124ac2a9b5ab6addb94e0e65f7', + 'md5': '0cd9e28ad270488911b0d2a72323395d', 'info_dict': { 'id': '2973058', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Robots futbolistas', }, } diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py index 023911c41..69882da63 100644 --- a/youtube_dl/extractor/twitch.py +++ b/youtube_dl/extractor/twitch.py @@ -11,14 +11,15 @@ from ..compat import ( compat_str, compat_urllib_parse, compat_urllib_parse_urlparse, - compat_urllib_request, compat_urlparse, ) from ..utils import ( + encode_dict, ExtractorError, int_or_none, parse_duration, parse_iso8601, + sanitized_Request, ) @@ -27,8 +28,7 @@ class TwitchBaseIE(InfoExtractor): _API_BASE = 'https://api.twitch.tv' _USHER_BASE = 'http://usher.twitch.tv' - _LOGIN_URL = 'https://secure.twitch.tv/login' - _LOGIN_POST_URL = 'https://passport.twitch.tv/authentications/new' + _LOGIN_URL = 'http://www.twitch.tv/login' _NETRC_MACHINE = 'twitch' def _handle_error(self, response): @@ -48,7 +48,7 @@ class TwitchBaseIE(InfoExtractor): for cookie in self._downloader.cookiejar: if cookie.name == 'api_token': headers['Twitch-Api-Token'] = cookie.value - request = compat_urllib_request.Request(url, headers=headers) + request = sanitized_Request(url, headers=headers) response = super(TwitchBaseIE, self)._download_json(request, video_id, note) self._handle_error(response) return response @@ -61,26 +61,28 @@ class TwitchBaseIE(InfoExtractor): if username is None: return - login_page = self._download_webpage( + login_page, handle = self._download_webpage_handle( self._LOGIN_URL, None, 'Downloading login page') login_form = self._hidden_inputs(login_page) login_form.update({ - 'login': username.encode('utf-8'), - 'password': password.encode('utf-8'), + 'username': username, + 'password': password, }) + redirect_url = handle.geturl() + post_url = self._search_regex( r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page, - 'post url', default=self._LOGIN_POST_URL, group='url') + 'post url', default=redirect_url, group='url') if not post_url.startswith('http'): - post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url) + post_url = compat_urlparse.urljoin(redirect_url, post_url) - request = compat_urllib_request.Request( - post_url, compat_urllib_parse.urlencode(login_form).encode('utf-8')) - request.add_header('Referer', self._LOGIN_URL) + request = sanitized_Request( + post_url, compat_urllib_parse.urlencode(encode_dict(login_form)).encode('utf-8')) + request.add_header('Referer', redirect_url) response = self._download_webpage( request, None, 'Logging in as %s' % username) @@ -238,14 +240,24 @@ class TwitchVodIE(TwitchItemBaseIE): def _real_extract(self, url): item_id = self._match_id(url) + info = self._download_info(self._ITEM_SHORTCUT, item_id) access_token = self._download_json( '%s/api/vods/%s/access_token' % (self._API_BASE, item_id), item_id, 'Downloading %s access token' % self._ITEM_TYPE) + formats = self._extract_m3u8_formats( - '%s/vod/%s?nauth=%s&nauthsig=%s&allow_source=true' - % (self._USHER_BASE, item_id, access_token['token'], access_token['sig']), + '%s/vod/%s?%s' % ( + self._USHER_BASE, item_id, + compat_urllib_parse.urlencode({ + 'allow_source': 'true', + 'allow_spectre': 'true', + 'player': 'twitchweb', + 'nauth': access_token['token'], + 'nauthsig': access_token['sig'], + })), item_id, 'mp4') + self._prefer_source(formats) info['formats'] = formats diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py index 1aaa06305..a161f046b 100644 --- a/youtube_dl/extractor/twitter.py +++ b/youtube_dl/extractor/twitter.py @@ -1,28 +1,73 @@ +# coding: utf-8 from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_urllib_request from ..utils import ( float_or_none, - unescapeHTML, + xpath_text, + remove_end, + int_or_none, + ExtractorError, + sanitized_Request, ) class TwitterCardIE(InfoExtractor): + IE_NAME = 'twitter:card' _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)' - _TEST = { - 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889', - 'md5': 'a74f50b310c83170319ba16de6955192', - 'info_dict': { - 'id': '560070183650213889', - 'ext': 'mp4', - 'title': 'TwitterCard', - 'thumbnail': 're:^https?://.*\.jpg$', - 'duration': 30.033, + _TESTS = [ + { + 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889', + 'md5': '4fa26a35f9d1bf4b646590ba8e84be19', + 'info_dict': { + 'id': '560070183650213889', + 'ext': 'mp4', + 'title': 'TwitterCard', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 30.033, + } }, - } + { + 'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768', + 'md5': '7ee2a553b63d1bccba97fbed97d9e1c8', + 'info_dict': { + 'id': '623160978427936768', + 'ext': 'mp4', + 'title': 'TwitterCard', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 80.155, + }, + }, + { + 'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977', + 'md5': 'b6f35e8b08a0bec6c8af77a2f4b3a814', + 'info_dict': { + 'id': 'dq4Oj5quskI', + 'ext': 'mp4', + 'title': 'Ubuntu 11.10 Overview', + 'description': 'Take a quick peek at what\'s new and improved in Ubuntu 11.10.\n\nOnce installed take a look at 10 Things to Do After Installing: http://www.omgubuntu.co.uk/2011/10/10-things-to-do-after-installing-ubuntu-11-10/', + 'upload_date': '20111013', + 'uploader': 'OMG! Ubuntu!', + 'uploader_id': 'omgubuntu', + }, + 'add_ie': ['Youtube'], + }, + { + 'url': 'https://twitter.com/i/cards/tfw/v1/665289828897005568', + 'md5': 'ab2745d0b0ce53319a534fccaa986439', + 'info_dict': { + 'id': 'iBb2x00UVlv', + 'ext': 'mp4', + 'upload_date': '20151113', + 'uploader_id': '1189339351084113920', + 'uploader': '@ArsenalTerje', + 'title': 'Vine by @ArsenalTerje', + }, + 'add_ie': ['Vine'], + } + ] def _real_extract(self, url): video_id = self._match_id(url) @@ -36,14 +81,28 @@ class TwitterCardIE(InfoExtractor): config = None formats = [] for user_agent in USER_AGENTS: - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) request.add_header('User-Agent', user_agent) webpage = self._download_webpage(request, video_id) - config = self._parse_json( - unescapeHTML(self._search_regex( - r'data-player-config="([^"]+)"', webpage, 'data player config')), + iframe_url = self._html_search_regex( + r'<iframe[^>]+src="((?:https?:)?//(?:www.youtube.com/embed/[^"]+|(?:www\.)?vine\.co/v/\w+/card))"', + webpage, 'video iframe', default=None) + if iframe_url: + return self.url_result(iframe_url) + + config = self._parse_json(self._html_search_regex( + r'data-player-config="([^"]+)"', webpage, 'data player config'), video_id) + if 'playlist' not in config: + if 'vmapUrl' in config: + vmap_data = self._download_xml(config['vmapUrl'], video_id) + video_url = xpath_text(vmap_data, './/MediaFile').strip() + formats.append({ + 'url': video_url, + }) + break # same video regardless of UA + continue video_url = config['playlist'][0]['source'] @@ -70,3 +129,100 @@ class TwitterCardIE(InfoExtractor): 'duration': duration, 'formats': formats, } + + +class TwitterIE(InfoExtractor): + IE_NAME = 'twitter' + _VALID_URL = r'https?://(?:www\.|m\.|mobile\.)?twitter\.com/(?P<user_id>[^/]+)/status/(?P<id>\d+)' + _TEMPLATE_URL = 'https://twitter.com/%s/status/%s' + + _TESTS = [{ + 'url': 'https://twitter.com/freethenipple/status/643211948184596480', + 'md5': 'db6612ec5d03355953c3ca9250c97e5e', + 'info_dict': { + 'id': '643211948184596480', + 'ext': 'mp4', + 'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 12.922, + 'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"', + 'uploader': 'FREE THE NIPPLE', + 'uploader_id': 'freethenipple', + }, + }, { + 'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1', + 'md5': 'f36dcd5fb92bf7057f155e7d927eeb42', + 'info_dict': { + 'id': '657991469417025536', + 'ext': 'mp4', + 'title': 'Gifs - tu vai cai tu vai cai tu nao eh capaz disso tu vai cai', + 'description': 'Gifs on Twitter: "tu vai cai tu vai cai tu nao eh capaz disso tu vai cai https://t.co/tM46VHFlO5"', + 'thumbnail': 're:^https?://.*\.png', + 'uploader': 'Gifs', + 'uploader_id': 'giphz', + }, + }, { + 'url': 'https://twitter.com/starwars/status/665052190608723968', + 'md5': '39b7199856dee6cd4432e72c74bc69d4', + 'info_dict': { + 'id': '665052190608723968', + 'ext': 'mp4', + 'title': 'Star Wars - A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens.', + 'description': 'Star Wars on Twitter: "A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens."', + 'uploader_id': 'starwars', + 'uploader': 'Star Wars', + }, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + user_id = mobj.group('user_id') + twid = mobj.group('id') + + webpage = self._download_webpage(self._TEMPLATE_URL % (user_id, twid), twid) + + username = remove_end(self._og_search_title(webpage), ' on Twitter') + + title = description = self._og_search_description(webpage).strip('').replace('\n', ' ').strip('“”') + + # strip 'https -_t.co_BJYgOjSeGA' junk from filenames + title = re.sub(r'\s+(https?://[^ ]+)', '', title) + + info = { + 'uploader_id': user_id, + 'uploader': username, + 'webpage_url': url, + 'description': '%s on Twitter: "%s"' % (username, description), + 'title': username + ' - ' + title, + } + + card_id = self._search_regex( + r'["\']/i/cards/tfw/v1/(\d+)', webpage, 'twitter card url', default=None) + if card_id: + card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id + info.update({ + '_type': 'url_transparent', + 'ie_key': 'TwitterCard', + 'url': card_url, + }) + return info + + mobj = re.search(r'''(?x) + <video[^>]+class="animated-gif"[^>]+ + (?:data-height="(?P<height>\d+)")?[^>]+ + (?:data-width="(?P<width>\d+)")?[^>]+ + (?:poster="(?P<poster>[^"]+)")?[^>]*>\s* + <source[^>]+video-src="(?P<url>[^"]+)" + ''', webpage) + + if mobj: + info.update({ + 'id': twid, + 'url': mobj.group('url'), + 'height': int_or_none(mobj.group('height')), + 'width': int_or_none(mobj.group('width')), + 'thumbnail': mobj.group('poster'), + }) + return info + + raise ExtractorError('There\'s not video in this tweet.') diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py index 365d8b4bf..59832b1ec 100644 --- a/youtube_dl/extractor/udemy.py +++ b/youtube_dl/extractor/udemy.py @@ -1,14 +1,16 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..compat import ( + compat_HTTPError, compat_urllib_parse, compat_urllib_request, ) from ..utils import ( ExtractorError, + float_or_none, + int_or_none, + sanitized_Request, ) @@ -17,6 +19,8 @@ class UdemyIE(InfoExtractor): _VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/|lecture/view/?\?lectureId=)(?P<id>\d+)' _LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1' _ORIGIN_URL = 'https://www.udemy.com' + _SUCCESSFULLY_ENROLLED = '>You have enrolled in this course!<' + _ALREADY_ENROLLED = '>You are already taking this course.<' _NETRC_MACHINE = 'udemy' _TESTS = [{ @@ -32,6 +36,29 @@ class UdemyIE(InfoExtractor): 'skip': 'Requires udemy account credentials', }] + def _enroll_course(self, webpage, course_id): + enroll_url = self._search_regex( + r'href=(["\'])(?P<url>https?://(?:www\.)?udemy\.com/course/subscribe/.+?)\1', + webpage, 'enroll url', group='url', + default='https://www.udemy.com/course/subscribe/?courseId=%s' % course_id) + webpage = self._download_webpage(enroll_url, course_id, 'Enrolling in the course') + if self._SUCCESSFULLY_ENROLLED in webpage: + self.to_screen('%s: Successfully enrolled in' % course_id) + elif self._ALREADY_ENROLLED in webpage: + self.to_screen('%s: Already enrolled in' % course_id) + + def _download_lecture(self, course_id, lecture_id): + return self._download_json( + 'https://www.udemy.com/api-2.0/users/me/subscribed-courses/%s/lectures/%s?%s' % ( + course_id, lecture_id, compat_urllib_parse.urlencode({ + 'video_only': '', + 'auto_play': '', + 'fields[lecture]': 'title,description,asset', + 'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,data', + 'instructorPreviewMode': 'False', + })), + lecture_id, 'Downloading lecture JSON') + def _handle_error(self, response): if not isinstance(response, dict): return @@ -53,12 +80,13 @@ class UdemyIE(InfoExtractor): headers['X-Udemy-Client-Id'] = cookie.value elif cookie.name == 'access_token': headers['X-Udemy-Bearer-Token'] = cookie.value + headers['X-Udemy-Authorization'] = 'Bearer %s' % cookie.value if isinstance(url_or_request, compat_urllib_request.Request): for header, value in headers.items(): url_or_request.add_header(header, value) else: - url_or_request = compat_urllib_request.Request(url_or_request, headers=headers) + url_or_request = sanitized_Request(url_or_request, headers=headers) response = super(UdemyIE, self)._download_json(url_or_request, video_id, note) self._handle_error(response) @@ -70,7 +98,7 @@ class UdemyIE(InfoExtractor): def _login(self): (username, password) = self._get_login_info() if username is None: - self.raise_login_required('Udemy account is required') + return login_popup = self._download_webpage( self._LOGIN_URL, None, 'Downloading login popup') @@ -89,7 +117,7 @@ class UdemyIE(InfoExtractor): 'password': password.encode('utf-8'), }) - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) request.add_header('Referer', self._ORIGIN_URL) request.add_header('Origin', self._ORIGIN_URL) @@ -108,44 +136,76 @@ class UdemyIE(InfoExtractor): def _real_extract(self, url): lecture_id = self._match_id(url) - lecture = self._download_json( - 'https://www.udemy.com/api-1.1/lectures/%s' % lecture_id, - lecture_id, 'Downloading lecture JSON') + webpage = self._download_webpage(url, lecture_id) - asset_type = lecture.get('assetType') or lecture.get('asset_type') + course_id = self._search_regex( + r'data-course-id=["\'](\d+)', webpage, 'course id') + + try: + lecture = self._download_lecture(course_id, lecture_id) + except ExtractorError as e: + # Error could possibly mean we are not enrolled in the course + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: + self._enroll_course(webpage, course_id) + lecture_id = self._download_lecture(course_id, lecture_id) + else: + raise + + title = lecture['title'] + description = lecture.get('description') + + asset = lecture['asset'] + + asset_type = asset.get('assetType') or asset.get('asset_type') if asset_type != 'Video': raise ExtractorError( 'Lecture %s is not a video' % lecture_id, expected=True) - asset = lecture['asset'] - stream_url = asset.get('streamUrl') or asset.get('stream_url') - mobj = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url) - if mobj: - return self.url_result(mobj.group(1), 'Youtube') + if stream_url: + youtube_url = self._search_regex( + r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url, 'youtube URL', default=None) + if youtube_url: + return self.url_result(youtube_url, 'Youtube') video_id = asset['id'] thumbnail = asset.get('thumbnailUrl') or asset.get('thumbnail_url') - duration = asset['data']['duration'] + duration = float_or_none(asset.get('data', {}).get('duration')) + outputs = asset.get('data', {}).get('outputs', {}) - download_url = asset.get('downloadUrl') or asset.get('download_url') + formats = [] + for format_ in asset.get('download_urls', {}).get('Video', []): + video_url = format_.get('file') + if not video_url: + continue + format_id = format_.get('label') + f = { + 'url': format_['file'], + 'height': int_or_none(format_id), + } + if format_id: + # Some videos contain additional metadata (e.g. + # https://www.udemy.com/ios9-swift/learn/#/lecture/3383208) + output = outputs.get(format_id) + if isinstance(output, dict): + f.update({ + 'format_id': '%sp' % (output.get('label') or format_id), + 'width': int_or_none(output.get('width')), + 'height': int_or_none(output.get('height')), + 'vbr': int_or_none(output.get('video_bitrate_in_kbps')), + 'vcodec': output.get('video_codec'), + 'fps': int_or_none(output.get('frame_rate')), + 'abr': int_or_none(output.get('audio_bitrate_in_kbps')), + 'acodec': output.get('audio_codec'), + 'asr': int_or_none(output.get('audio_sample_rate')), + 'tbr': int_or_none(output.get('total_bitrate_in_kbps')), + 'filesize': int_or_none(output.get('file_size_in_bytes')), + }) + else: + f['format_id'] = '%sp' % format_id + formats.append(f) - video = download_url.get('Video') or download_url.get('video') - video_480p = download_url.get('Video480p') or download_url.get('video_480p') - - formats = [ - { - 'url': video_480p[0], - 'format_id': '360p', - }, - { - 'url': video[0], - 'format_id': '720p', - }, - ] - - title = lecture['title'] - description = lecture['description'] + self._sort_formats(formats) return { 'id': video_id, @@ -159,9 +219,7 @@ class UdemyIE(InfoExtractor): class UdemyCourseIE(UdemyIE): IE_NAME = 'udemy:course' - _VALID_URL = r'https?://www\.udemy\.com/(?P<coursepath>[\da-z-]+)' - _SUCCESSFULLY_ENROLLED = '>You have enrolled in this course!<' - _ALREADY_ENROLLED = '>You are already taking this course.<' + _VALID_URL = r'https?://www\.udemy\.com/(?P<id>[\da-z-]+)' _TESTS = [] @classmethod @@ -169,24 +227,18 @@ class UdemyCourseIE(UdemyIE): return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url) def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - course_path = mobj.group('coursepath') + course_path = self._match_id(url) + + webpage = self._download_webpage(url, course_path) response = self._download_json( 'https://www.udemy.com/api-1.1/courses/%s' % course_path, course_path, 'Downloading course JSON') - course_id = int(response['id']) - course_title = response['title'] + course_id = response['id'] + course_title = response.get('title') - webpage = self._download_webpage( - 'https://www.udemy.com/course/subscribe/?courseId=%s' % course_id, - course_id, 'Enrolling in the course') - - if self._SUCCESSFULLY_ENROLLED in webpage: - self.to_screen('%s: Successfully enrolled in' % course_id) - elif self._ALREADY_ENROLLED in webpage: - self.to_screen('%s: Already enrolled in' % course_id) + self._enroll_course(webpage, course_id) response = self._download_json( 'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id, diff --git a/youtube_dl/extractor/udn.py b/youtube_dl/extractor/udn.py index 2151f8338..ee35b7227 100644 --- a/youtube_dl/extractor/udn.py +++ b/youtube_dl/extractor/udn.py @@ -12,7 +12,8 @@ from ..compat import compat_urlparse class UDNEmbedIE(InfoExtractor): IE_DESC = '聯合影音' - _VALID_URL = r'https?://video\.udn\.com/(?:embed|play)/news/(?P<id>\d+)' + _PROTOCOL_RELATIVE_VALID_URL = r'//video\.udn\.com/(?:embed|play)/news/(?P<id>\d+)' + _VALID_URL = r'https?:' + _PROTOCOL_RELATIVE_VALID_URL _TESTS = [{ 'url': 'http://video.udn.com/embed/news/300040', 'md5': 'de06b4c90b042c128395a88f0384817e', diff --git a/youtube_dl/extractor/vbox7.py b/youtube_dl/extractor/vbox7.py index 722eb5236..1e740fbe6 100644 --- a/youtube_dl/extractor/vbox7.py +++ b/youtube_dl/extractor/vbox7.py @@ -4,11 +4,11 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..compat import ( compat_urllib_parse, - compat_urllib_request, compat_urlparse, ) from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -49,7 +49,7 @@ class Vbox7IE(InfoExtractor): info_url = "http://vbox7.com/play/magare.do" data = compat_urllib_parse.urlencode({'as3': '1', 'vid': video_id}) - info_request = compat_urllib_request.Request(info_url, data) + info_request = sanitized_Request(info_url, data) info_request.add_header('Content-Type', 'application/x-www-form-urlencoded') info_response = self._download_webpage(info_request, video_id, 'Downloading info webpage') if info_response is None: diff --git a/youtube_dl/extractor/veoh.py b/youtube_dl/extractor/veoh.py index 01e258e32..9633f7ffe 100644 --- a/youtube_dl/extractor/veoh.py +++ b/youtube_dl/extractor/veoh.py @@ -4,12 +4,10 @@ import re import json from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( int_or_none, ExtractorError, + sanitized_Request, ) @@ -110,7 +108,7 @@ class VeohIE(InfoExtractor): if 'class="adultwarning-container"' in webpage: self.report_age_confirmation() age_limit = 18 - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) request.add_header('Cookie', 'confirmedAdult=true') webpage = self._download_webpage(request, video_id) diff --git a/youtube_dl/extractor/vessel.py b/youtube_dl/extractor/vessel.py index 3c8d2a943..1a0ff3395 100644 --- a/youtube_dl/extractor/vessel.py +++ b/youtube_dl/extractor/vessel.py @@ -4,10 +4,10 @@ from __future__ import unicode_literals import json from .common import InfoExtractor -from ..compat import compat_urllib_request from ..utils import ( ExtractorError, parse_iso8601, + sanitized_Request, ) @@ -33,7 +33,7 @@ class VesselIE(InfoExtractor): @staticmethod def make_json_request(url, data): payload = json.dumps(data).encode('utf-8') - req = compat_urllib_request.Request(url, payload) + req = sanitized_Request(url, payload) req.add_header('Content-Type', 'application/json; charset=utf-8') return req diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py index c17094f81..571289421 100644 --- a/youtube_dl/extractor/vevo.py +++ b/youtube_dl/extractor/vevo.py @@ -1,15 +1,13 @@ from __future__ import unicode_literals import re -import xml.etree.ElementTree from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) +from ..compat import compat_etree_fromstring from ..utils import ( ExtractorError, int_or_none, + sanitized_Request, ) @@ -73,7 +71,7 @@ class VevoIE(InfoExtractor): _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com/' def _real_initialize(self): - req = compat_urllib_request.Request( + req = sanitized_Request( 'http://www.vevo.com/auth', data=b'') webpage = self._download_webpage( req, None, @@ -97,7 +95,7 @@ class VevoIE(InfoExtractor): if last_version['version'] == -1: raise ExtractorError('Unable to extract last version of the video') - renditions = xml.etree.ElementTree.fromstring(last_version['data']) + renditions = compat_etree_fromstring(last_version['data']) formats = [] # Already sorted from worst to best quality for rend in renditions.findall('rendition'): @@ -114,7 +112,7 @@ class VevoIE(InfoExtractor): def _formats_from_smil(self, smil_xml): formats = [] - smil_doc = xml.etree.ElementTree.fromstring(smil_xml.encode('utf-8')) + smil_doc = compat_etree_fromstring(smil_xml.encode('utf-8')) els = smil_doc.findall('.//{http://www.w3.org/2001/SMIL20/Language}video') for el in els: src = el.attrib['src'] diff --git a/youtube_dl/extractor/viddler.py b/youtube_dl/extractor/viddler.py index 8516a2940..40ffbad2a 100644 --- a/youtube_dl/extractor/viddler.py +++ b/youtube_dl/extractor/viddler.py @@ -4,9 +4,7 @@ from .common import InfoExtractor from ..utils import ( float_or_none, int_or_none, -) -from ..compat import ( - compat_urllib_request + sanitized_Request, ) @@ -65,7 +63,7 @@ class ViddlerIE(InfoExtractor): 'http://api.viddler.com/api/v2/viddler.videos.getPlaybackDetails.json?video_id=%s&key=v0vhrt7bg2xq1vyxhkct' % video_id) headers = {'Referer': 'http://static.cdn-ec.viddler.com/js/arpeggio/v2/embed.html'} - request = compat_urllib_request.Request(json_url, None, headers) + request = sanitized_Request(json_url, None, headers) data = self._download_json(request, video_id)['video'] formats = [] diff --git a/youtube_dl/extractor/videofyme.py b/youtube_dl/extractor/videofyme.py index 94f9e9be9..cd3f50a63 100644 --- a/youtube_dl/extractor/videofyme.py +++ b/youtube_dl/extractor/videofyme.py @@ -2,8 +2,8 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( - find_xpath_attr, int_or_none, + parse_iso8601, ) @@ -18,33 +18,35 @@ class VideofyMeIE(InfoExtractor): 'id': '1100701', 'ext': 'mp4', 'title': 'This is VideofyMe', - 'description': None, + 'description': '', + 'upload_date': '20130326', + 'timestamp': 1364288959, 'uploader': 'VideofyMe', 'uploader_id': 'thisisvideofyme', 'view_count': int, + 'likes': int, + 'comment_count': int, }, - } def _real_extract(self, url): video_id = self._match_id(url) - config = self._download_xml('http://sunshine.videofy.me/?videoId=%s' % video_id, - video_id) - video = config.find('video') - sources = video.find('sources') - url_node = next(node for node in [find_xpath_attr(sources, 'source', 'id', 'HQ %s' % key) - for key in ['on', 'av', 'off']] if node is not None) - video_url = url_node.find('url').text - view_count = int_or_none(self._search_regex( - r'([0-9]+)', video.find('views').text, 'view count', fatal=False)) + + config = self._download_json('http://vf-player-info-loader.herokuapp.com/%s.json' % video_id, video_id)['videoinfo'] + + video = config.get('video') + blog = config.get('blog', {}) return { 'id': video_id, - 'title': video.find('title').text, - 'url': video_url, - 'thumbnail': video.find('thumb').text, - 'description': video.find('description').text, - 'uploader': config.find('blog/name').text, - 'uploader_id': video.find('identifier').text, - 'view_count': view_count, + 'title': video['title'], + 'url': video['sources']['source']['url'], + 'thumbnail': video.get('thumb'), + 'description': video.get('description'), + 'timestamp': parse_iso8601(video.get('date')), + 'uploader': blog.get('name'), + 'uploader_id': blog.get('identifier'), + 'view_count': int_or_none(self._search_regex(r'([0-9]+)', video.get('views'), 'view count', fatal=False)), + 'likes': int_or_none(video.get('likes')), + 'comment_count': int_or_none(video.get('nrOfComments')), } diff --git a/youtube_dl/extractor/videolecturesnet.py b/youtube_dl/extractor/videolecturesnet.py deleted file mode 100644 index 649ac9433..000000000 --- a/youtube_dl/extractor/videolecturesnet.py +++ /dev/null @@ -1,82 +0,0 @@ -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..compat import ( - compat_HTTPError, - compat_urlparse, -) -from ..utils import ( - ExtractorError, - parse_duration, -) - - -class VideoLecturesNetIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?videolectures\.net/(?P<id>[^/#?]+)/*(?:[#?].*)?$' - IE_NAME = 'videolectures.net' - - _TESTS = [{ - 'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/', - 'info_dict': { - 'id': 'promogram_igor_mekjavic_eng', - 'ext': 'mp4', - 'title': 'Automatics, robotics and biocybernetics', - 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482', - 'upload_date': '20130627', - 'duration': 565, - 'thumbnail': 're:http://.*\.jpg', - }, - }, { - # video with invalid direct format links (HTTP 403) - 'url': 'http://videolectures.net/russir2010_filippova_nlp/', - 'info_dict': { - 'id': 'russir2010_filippova_nlp', - 'ext': 'flv', - 'title': 'NLP at Google', - 'description': 'md5:fc7a6d9bf0302d7cc0e53f7ca23747b3', - 'duration': 5352, - 'thumbnail': 're:http://.*\.jpg', - }, - 'params': { - # rtmp download - 'skip_download': True, - }, - }, { - 'url': 'http://videolectures.net/deeplearning2015_montreal/', - 'info_dict': { - 'id': 'deeplearning2015_montreal', - 'title': 'Deep Learning Summer School, Montreal 2015', - 'description': 'md5:90121a40cc6926df1bf04dcd8563ed3b', - }, - 'playlist_count': 30, - }] - - def _real_extract(self, url): - video_id = self._match_id(url) - - smil_url = 'http://videolectures.net/%s/video/1/smil.xml' % video_id - - try: - smil = self._download_smil(smil_url, video_id) - except ExtractorError as e: - if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: - # Probably a playlist - webpage = self._download_webpage(url, video_id) - entries = [ - self.url_result(compat_urlparse.urljoin(url, video_url), 'VideoLecturesNet') - for _, video_url in re.findall(r'<a[^>]+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', webpage)] - playlist_title = self._html_search_meta('title', webpage, 'title', fatal=True) - playlist_description = self._html_search_meta('description', webpage, 'description') - return self.playlist_result(entries, video_id, playlist_title, playlist_description) - - info = self._parse_smil(smil, smil_url, video_id) - - info['id'] = video_id - - switch = smil.find('.//switch') - if switch is not None: - info['duration'] = parse_duration(switch.attrib.get('dur')) - - return info diff --git a/youtube_dl/extractor/videomega.py b/youtube_dl/extractor/videomega.py index 78ff6310a..87aca327b 100644 --- a/youtube_dl/extractor/videomega.py +++ b/youtube_dl/extractor/videomega.py @@ -4,7 +4,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_urllib_request +from ..utils import sanitized_Request class VideoMegaIE(InfoExtractor): @@ -30,7 +30,7 @@ class VideoMegaIE(InfoExtractor): video_id = self._match_id(url) iframe_url = 'http://videomega.tv/cdn.php?ref=%s' % video_id - req = compat_urllib_request.Request(iframe_url) + req = sanitized_Request(iframe_url) req.add_header('Referer', url) req.add_header('Cookie', 'noadvtday=0') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/vidme.py b/youtube_dl/extractor/vidme.py index 078d283b2..3d63ed4f0 100644 --- a/youtube_dl/extractor/vidme.py +++ b/youtube_dl/extractor/vidme.py @@ -14,7 +14,7 @@ class VidmeIE(InfoExtractor): _VALID_URL = r'https?://vid\.me/(?:e/)?(?P<id>[\da-zA-Z]+)' _TESTS = [{ 'url': 'https://vid.me/QNB', - 'md5': 'c62f1156138dc3323902188c5b5a8bd6', + 'md5': 'f42d05e7149aeaec5c037b17e5d3dc82', 'info_dict': { 'id': 'QNB', 'ext': 'mp4', @@ -93,6 +93,39 @@ class VidmeIE(InfoExtractor): 'params': { 'skip_download': True, }, + }, { + # nsfw, user-disabled + 'url': 'https://vid.me/dzGJ', + 'only_matching': True, + }, { + # suspended + 'url': 'https://vid.me/Ox3G', + 'only_matching': True, + }, { + # deleted + 'url': 'https://vid.me/KTPm', + 'only_matching': True, + }, { + # no formats in the API response + 'url': 'https://vid.me/e5g', + 'info_dict': { + 'id': 'e5g', + 'ext': 'mp4', + 'title': 'Video upload (e5g)', + 'thumbnail': 're:^https?://.*\.jpg', + 'timestamp': 1401480195, + 'upload_date': '20140530', + 'uploader': None, + 'uploader_id': None, + 'age_limit': 0, + 'duration': 483, + 'view_count': int, + 'like_count': int, + 'comment_count': int, + }, + 'params': { + 'skip_download': True, + }, }] def _real_extract(self, url): @@ -114,6 +147,17 @@ class VidmeIE(InfoExtractor): video = response['video'] + if video.get('state') == 'deleted': + raise ExtractorError( + 'Vidme said: Sorry, this video has been deleted.', + expected=True) + + if video.get('state') in ('user-disabled', 'suspended'): + raise ExtractorError( + 'Vidme said: This video has been suspended either due to a copyright claim, ' + 'or for violating the terms of use.', + expected=True) + formats = [{ 'format_id': f.get('type'), 'url': f['uri'], @@ -121,6 +165,14 @@ class VidmeIE(InfoExtractor): 'height': int_or_none(f.get('height')), 'preference': 0 if f.get('type', '').endswith('clip') else 1, } for f in video.get('formats', []) if f.get('uri')] + + if not formats and video.get('complete_url'): + formats.append({ + 'url': video.get('complete_url'), + 'width': int_or_none(video.get('width')), + 'height': int_or_none(video.get('height')), + }) + self._sort_formats(formats) title = video['title'] @@ -137,7 +189,7 @@ class VidmeIE(InfoExtractor): return { 'id': video_id, - 'title': title, + 'title': title or 'Video upload (%s)' % video_id, 'description': description, 'thumbnail': thumbnail, 'uploader': uploader, diff --git a/youtube_dl/extractor/vidzi.py b/youtube_dl/extractor/vidzi.py index 08a5a7b8d..2ba9f31df 100644 --- a/youtube_dl/extractor/vidzi.py +++ b/youtube_dl/extractor/vidzi.py @@ -20,8 +20,14 @@ class VidziIE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - video_url = self._html_search_regex( - r'{\s*file\s*:\s*"([^"]+)"\s*}', webpage, 'video url') + video_host = self._html_search_regex( + r'id=\'vplayer\'><img src="http://(.*?)/i', webpage, + 'video host') + video_hash = self._html_search_regex( + r'\|([a-z0-9]+)\|hls\|type', webpage, 'video_hash') + ext = self._html_search_regex( + r'\|tracks\|([a-z0-9]+)\|', webpage, 'video ext') + video_url = 'http://' + video_host + '/' + video_hash + '/v.' + ext title = self._html_search_regex( r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title') diff --git a/youtube_dl/extractor/viewster.py b/youtube_dl/extractor/viewster.py index 632e57fb4..185b1c119 100644 --- a/youtube_dl/extractor/viewster.py +++ b/youtube_dl/extractor/viewster.py @@ -4,7 +4,6 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..compat import ( compat_HTTPError, - compat_urllib_request, compat_urllib_parse, compat_urllib_parse_unquote, ) @@ -13,6 +12,7 @@ from ..utils import ( ExtractorError, int_or_none, parse_iso8601, + sanitized_Request, HEADRequest, ) @@ -76,7 +76,7 @@ class ViewsterIE(InfoExtractor): _ACCEPT_HEADER = 'application/json, text/javascript, */*; q=0.01' def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True): - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) request.add_header('Accept', self._ACCEPT_HEADER) request.add_header('Auth-token', self._AUTH_TOKEN) return super(ViewsterIE, self)._download_json(request, video_id, note, fatal=fatal) @@ -131,10 +131,11 @@ class ViewsterIE(InfoExtractor): formats.extend(self._extract_f4m_formats( video_url, video_id, f4m_id='hds')) elif ext == 'm3u8': - formats.extend(self._extract_m3u8_formats( + m3u8_formats = self._extract_m3u8_formats( video_url, video_id, 'mp4', m3u8_id='hls', - fatal=False # m3u8 sometimes fail - )) + fatal=False) # m3u8 sometimes fail + if m3u8_formats: + formats.extend(m3u8_formats) else: format_id = media.get('Bitrate') f = { diff --git a/youtube_dl/extractor/viidea.py b/youtube_dl/extractor/viidea.py new file mode 100644 index 000000000..525e303d4 --- /dev/null +++ b/youtube_dl/extractor/viidea.py @@ -0,0 +1,188 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..compat import ( + compat_urlparse, + compat_str, +) +from ..utils import ( + parse_duration, + js_to_json, + parse_iso8601, +) + + +class ViideaIE(InfoExtractor): + _VALID_URL = r'''(?x)http://(?:www\.)?(?: + videolectures\.net| + flexilearn\.viidea\.net| + presentations\.ocwconsortium\.org| + video\.travel-zoom\.si| + video\.pomp-forum\.si| + tv\.nil\.si| + video\.hekovnik.com| + video\.szko\.si| + kpk\.viidea\.com| + inside\.viidea\.net| + video\.kiberpipa\.org| + bvvideo\.si| + kongres\.viidea\.net| + edemokracija\.viidea\.com + )(?:/lecture)?/(?P<id>[^/]+)(?:/video/(?P<part>\d+))?/*(?:[#?].*)?$''' + + _TESTS = [{ + 'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/', + 'info_dict': { + 'id': '20171', + 'display_id': 'promogram_igor_mekjavic_eng', + 'ext': 'mp4', + 'title': 'Automatics, robotics and biocybernetics', + 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482', + 'thumbnail': 're:http://.*\.jpg', + 'timestamp': 1372349289, + 'upload_date': '20130627', + 'duration': 565, + }, + }, { + # video with invalid direct format links (HTTP 403) + 'url': 'http://videolectures.net/russir2010_filippova_nlp/', + 'info_dict': { + 'id': '14891', + 'display_id': 'russir2010_filippova_nlp', + 'ext': 'flv', + 'title': 'NLP at Google', + 'description': 'md5:fc7a6d9bf0302d7cc0e53f7ca23747b3', + 'thumbnail': 're:http://.*\.jpg', + 'timestamp': 1284375600, + 'upload_date': '20100913', + 'duration': 5352, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }, { + # event playlist + 'url': 'http://videolectures.net/deeplearning2015_montreal/', + 'info_dict': { + 'id': '23181', + 'title': 'Deep Learning Summer School, Montreal 2015', + 'description': 'md5:0533a85e4bd918df52a01f0e1ebe87b7', + 'thumbnail': 're:http://.*\.jpg', + 'timestamp': 1438560000, + }, + 'playlist_count': 30, + }, { + # multi part lecture + 'url': 'http://videolectures.net/mlss09uk_bishop_ibi/', + 'info_dict': { + 'id': '9737', + 'display_id': 'mlss09uk_bishop_ibi', + 'title': 'Introduction To Bayesian Inference', + 'thumbnail': 're:http://.*\.jpg', + 'timestamp': 1251622800, + }, + 'playlist': [{ + 'info_dict': { + 'id': '9737_part1', + 'display_id': 'mlss09uk_bishop_ibi_part1', + 'ext': 'wmv', + 'title': 'Introduction To Bayesian Inference (Part 1)', + 'thumbnail': 're:http://.*\.jpg', + 'duration': 4622, + 'timestamp': 1251622800, + 'upload_date': '20090830', + }, + }, { + 'info_dict': { + 'id': '9737_part2', + 'display_id': 'mlss09uk_bishop_ibi_part2', + 'ext': 'wmv', + 'title': 'Introduction To Bayesian Inference (Part 2)', + 'thumbnail': 're:http://.*\.jpg', + 'duration': 5641, + 'timestamp': 1251622800, + 'upload_date': '20090830', + }, + }], + 'playlist_count': 2, + }] + + def _real_extract(self, url): + lecture_slug, explicit_part_id = re.match(self._VALID_URL, url).groups() + + webpage = self._download_webpage(url, lecture_slug) + + cfg = self._parse_json(self._search_regex( + [r'cfg\s*:\s*({.+?})\s*,\s*[\da-zA-Z_]+\s*:\s*\(?\s*function', + r'cfg\s*:\s*({[^}]+})'], + webpage, 'cfg'), lecture_slug, js_to_json) + + lecture_id = compat_str(cfg['obj_id']) + + base_url = self._proto_relative_url(cfg['livepipe'], 'http:') + + lecture_data = self._download_json( + '%s/site/api/lecture/%s?format=json' % (base_url, lecture_id), + lecture_id)['lecture'][0] + + lecture_info = { + 'id': lecture_id, + 'display_id': lecture_slug, + 'title': lecture_data['title'], + 'timestamp': parse_iso8601(lecture_data.get('time')), + 'description': lecture_data.get('description_wiki'), + 'thumbnail': lecture_data.get('thumb'), + } + + playlist_entries = [] + lecture_type = lecture_data.get('type') + parts = [compat_str(video) for video in cfg.get('videos', [])] + if parts: + multipart = len(parts) > 1 + + def extract_part(part_id): + smil_url = '%s/%s/video/%s/smil.xml' % (base_url, lecture_slug, part_id) + smil = self._download_smil(smil_url, lecture_id) + info = self._parse_smil(smil, smil_url, lecture_id) + info['id'] = lecture_id if not multipart else '%s_part%s' % (lecture_id, part_id) + info['display_id'] = lecture_slug if not multipart else '%s_part%s' % (lecture_slug, part_id) + if multipart: + info['title'] += ' (Part %s)' % part_id + switch = smil.find('.//switch') + if switch is not None: + info['duration'] = parse_duration(switch.attrib.get('dur')) + item_info = lecture_info.copy() + item_info.update(info) + return item_info + + if explicit_part_id or not multipart: + result = extract_part(explicit_part_id or parts[0]) + else: + result = { + '_type': 'multi_video', + 'entries': [extract_part(part) for part in parts], + } + result.update(lecture_info) + + # Immediately return explicitly requested part or non event item + if explicit_part_id or lecture_type != 'evt': + return result + + playlist_entries.append(result) + + # It's probably a playlist + if not parts or lecture_type == 'evt': + playlist_webpage = self._download_webpage( + '%s/site/ajax/drilldown/?id=%s' % (base_url, lecture_id), lecture_id) + entries = [ + self.url_result(compat_urlparse.urljoin(url, video_url), 'Viidea') + for _, video_url in re.findall( + r'<a[^>]+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', playlist_webpage)] + playlist_entries.extend(entries) + + playlist = self.playlist_result(playlist_entries, lecture_id) + playlist.update(lecture_info) + return playlist diff --git a/youtube_dl/extractor/viki.py b/youtube_dl/extractor/viki.py index ddbd395c8..a63c23617 100644 --- a/youtube_dl/extractor/viki.py +++ b/youtube_dl/extractor/viki.py @@ -7,14 +7,14 @@ import hmac import hashlib import itertools +from .common import InfoExtractor from ..utils import ( ExtractorError, int_or_none, parse_age_limit, parse_iso8601, + sanitized_Request, ) -from ..compat import compat_urllib_request -from .common import InfoExtractor class VikiBaseIE(InfoExtractor): @@ -43,7 +43,7 @@ class VikiBaseIE(InfoExtractor): hashlib.sha1 ).hexdigest() url = self._API_URL_TEMPLATE % (query, sig) - return compat_urllib_request.Request( + return sanitized_Request( url, json.dumps(post_data).encode('utf-8')) if post_data else url def _call_api(self, path, video_id, note, timestamp=None, post_data=None): diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index fa1b22049..f392ccf1c 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -8,15 +8,15 @@ import itertools from .common import InfoExtractor from ..compat import ( compat_HTTPError, - compat_urllib_parse, - compat_urllib_request, compat_urlparse, ) from ..utils import ( + encode_dict, ExtractorError, InAdvancePagedList, int_or_none, RegexNotFoundError, + sanitized_Request, smuggle_url, std_headers, unified_strdate, @@ -40,17 +40,17 @@ class VimeoBaseInfoExtractor(InfoExtractor): self.report_login() webpage = self._download_webpage(self._LOGIN_URL, None, False) token, vuid = self._extract_xsrft_and_vuid(webpage) - data = urlencode_postdata({ + data = urlencode_postdata(encode_dict({ 'action': 'login', 'email': username, 'password': password, 'service': 'vimeo', 'token': token, - }) - login_request = compat_urllib_request.Request(self._LOGIN_URL, data) + })) + login_request = sanitized_Request(self._LOGIN_URL, data) login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - login_request.add_header('Cookie', 'vuid=%s' % vuid) login_request.add_header('Referer', self._LOGIN_URL) + self._set_vimeo_cookie('vuid', vuid) self._download_webpage(login_request, None, False, 'Wrong login info') def _extract_xsrft_and_vuid(self, webpage): @@ -62,6 +62,9 @@ class VimeoBaseInfoExtractor(InfoExtractor): webpage, 'vuid', group='vuid') return xsrft, vuid + def _set_vimeo_cookie(self, name, value): + self._set_cookie('vimeo.com', name, value) + class VimeoIE(VimeoBaseInfoExtractor): """Information extractor for vimeo.com.""" @@ -133,7 +136,7 @@ class VimeoIE(VimeoBaseInfoExtractor): 'uploader_id': 'user18948128', 'uploader': 'Jaime Marquínez Ferrándiz', 'duration': 10, - 'description': 'This is "youtube-dl password protected test video" by Jaime Marquínez Ferrándiz on Vimeo, the home for high quality videos and the people who love them.', + 'description': 'This is "youtube-dl password protected test video" by Jaime Marquínez Ferrándiz on Vimeo, the home for high quality videos and the people\u2026', }, 'params': { 'videopassword': 'youtube-dl', @@ -181,6 +184,15 @@ class VimeoIE(VimeoBaseInfoExtractor): 'uploader_id': 'user28849593', }, }, + { + 'url': 'https://vimeo.com/109815029', + 'note': 'Video not completely processed, "failed" seed status', + 'only_matching': True, + }, + { + 'url': 'https://vimeo.com/groups/travelhd/videos/22439234', + 'only_matching': True, + }, ] @staticmethod @@ -203,17 +215,17 @@ class VimeoIE(VimeoBaseInfoExtractor): if password is None: raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) token, vuid = self._extract_xsrft_and_vuid(webpage) - data = urlencode_postdata({ + data = urlencode_postdata(encode_dict({ 'password': password, 'token': token, - }) + })) if url.startswith('http://'): # vimeo only supports https now, but the user can give an http url url = url.replace('http://', 'https://') - password_request = compat_urllib_request.Request(url + '/password', data) + password_request = sanitized_Request(url + '/password', data) password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - password_request.add_header('Cookie', 'clip_test2=1; vuid=%s' % vuid) password_request.add_header('Referer', url) + self._set_vimeo_cookie('vuid', vuid) return self._download_webpage( password_request, video_id, 'Verifying the password', 'Wrong password') @@ -222,9 +234,9 @@ class VimeoIE(VimeoBaseInfoExtractor): password = self._downloader.params.get('videopassword', None) if password is None: raise ExtractorError('This video is protected by a password, use the --video-password option') - data = compat_urllib_parse.urlencode({'password': password}) + data = urlencode_postdata(encode_dict({'password': password})) pass_url = url + '/check-password' - password_request = compat_urllib_request.Request(pass_url, data) + password_request = sanitized_Request(pass_url, data) password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') return self._download_json( password_request, video_id, @@ -253,7 +265,7 @@ class VimeoIE(VimeoBaseInfoExtractor): url = 'https://vimeo.com/' + video_id # Retrieve video webpage to extract further information - request = compat_urllib_request.Request(url, None, headers) + request = sanitized_Request(url, None, headers) try: webpage = self._download_webpage(request, video_id) except ExtractorError as ee: @@ -273,20 +285,30 @@ class VimeoIE(VimeoBaseInfoExtractor): self.report_extraction(video_id) vimeo_config = self._search_regex( - r'vimeo\.config\s*=\s*({.+?});', webpage, + r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage, 'vimeo config', default=None) if vimeo_config: seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {}) if seed_status.get('state') == 'failed': raise ExtractorError( - '%s returned error: %s' % (self.IE_NAME, seed_status['title']), + '%s said: %s' % (self.IE_NAME, seed_status['title']), expected=True) # Extract the config JSON try: try: config_url = self._html_search_regex( - r' data-config-url="(.+?)"', webpage, 'config URL') + r' data-config-url="(.+?)"', webpage, + 'config URL', default=None) + if not config_url: + # Sometimes new react-based page is served instead of old one that require + # different config URL extraction approach (see + # https://github.com/rg3/youtube-dl/pull/7209) + vimeo_clip_page_config = self._search_regex( + r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage, + 'vimeo clip page config') + config_url = self._parse_json( + vimeo_clip_page_config, video_id)['player']['config_url'] config_json = self._download_webpage(config_url, video_id) config = json.loads(config_json) except RegexNotFoundError: @@ -369,47 +391,29 @@ class VimeoIE(VimeoBaseInfoExtractor): like_count = None comment_count = None - # Vimeo specific: extract request signature and timestamp - sig = config['request']['signature'] - timestamp = config['request']['timestamp'] - - # Vimeo specific: extract video codec and quality information - # First consider quality, then codecs, then take everything - codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')] - files = {'hd': [], 'sd': [], 'other': []} - config_files = config["video"].get("files") or config["request"].get("files") - for codec_name, codec_extension in codecs: - for quality in config_files.get(codec_name, []): - format_id = '-'.join((codec_name, quality)).lower() - key = quality if quality in files else 'other' - video_url = None - if isinstance(config_files[codec_name], dict): - file_info = config_files[codec_name][quality] - video_url = file_info.get('url') - else: - file_info = {} - if video_url is None: - video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \ - % (video_id, sig, timestamp, quality, codec_name.upper()) - - files[key].append({ - 'ext': codec_extension, - 'url': video_url, - 'format_id': format_id, - 'width': int_or_none(file_info.get('width')), - 'height': int_or_none(file_info.get('height')), - 'tbr': int_or_none(file_info.get('bitrate')), - }) formats = [] - m3u8_url = config_files.get('hls', {}).get('all') + config_files = config['video'].get('files') or config['request'].get('files', {}) + for f in config_files.get('progressive', []): + video_url = f.get('url') + if not video_url: + continue + formats.append({ + 'url': video_url, + 'format_id': 'http-%s' % f.get('quality'), + 'width': int_or_none(f.get('width')), + 'height': int_or_none(f.get('height')), + 'fps': int_or_none(f.get('fps')), + 'tbr': int_or_none(f.get('bitrate')), + }) + m3u8_url = config_files.get('hls', {}).get('url') if m3u8_url: m3u8_formats = self._extract_m3u8_formats( m3u8_url, video_id, 'mp4', 'm3u8_native', 0, 'hls', fatal=False) if m3u8_formats: formats.extend(m3u8_formats) - for key in ('other', 'sd', 'hd'): - formats += files[key] - self._sort_formats(formats) + # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps + # at the same time without actual units specified. This lead to wrong sorting. + self._sort_formats(formats, field_preference=('height', 'width', 'fps', 'format_id')) subtitles = {} text_tracks = config['request'].get('text_tracks') @@ -473,21 +477,20 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): token, vuid = self._extract_xsrft_and_vuid(webpage) fields['token'] = token fields['password'] = password - post = urlencode_postdata(fields) + post = urlencode_postdata(encode_dict(fields)) password_path = self._search_regex( r'action="([^"]+)"', login_form, 'password URL') password_url = compat_urlparse.urljoin(page_url, password_path) - password_request = compat_urllib_request.Request(password_url, post) + password_request = sanitized_Request(password_url, post) password_request.add_header('Content-type', 'application/x-www-form-urlencoded') - password_request.add_header('Cookie', 'vuid=%s' % vuid) - self._set_cookie('vimeo.com', 'xsrft', token) + self._set_vimeo_cookie('vuid', vuid) + self._set_vimeo_cookie('xsrft', token) return self._download_webpage( password_request, list_id, 'Verifying the password', 'Wrong password') - def _extract_videos(self, list_id, base_url): - video_ids = [] + def _title_and_entries(self, list_id, base_url): for pagenum in itertools.count(1): page_url = self._page_url(base_url, pagenum) webpage = self._download_webpage( @@ -496,18 +499,18 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): if pagenum == 1: webpage = self._login_list_password(page_url, list_id, webpage) + yield self._extract_list_title(webpage) + + for video_id in re.findall(r'id="clip_(\d+?)"', webpage): + yield self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo') - video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage)) if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None: break - entries = [self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo') - for video_id in video_ids] - return {'_type': 'playlist', - 'id': list_id, - 'title': self._extract_list_title(webpage), - 'entries': entries, - } + def _extract_videos(self, list_id, base_url): + title_and_entries = self._title_and_entries(list_id, base_url) + list_title = next(title_and_entries) + return self.playlist_result(title_and_entries, list_id, list_title) def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -568,7 +571,7 @@ class VimeoAlbumIE(VimeoChannelIE): class VimeoGroupsIE(VimeoAlbumIE): IE_NAME = 'vimeo:group' - _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)' + _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)' _TESTS = [{ 'url': 'https://vimeo.com/groups/rolexawards', 'info_dict': { @@ -637,7 +640,7 @@ class VimeoWatchLaterIE(VimeoChannelIE): def _page_url(self, base_url, pagenum): url = '%s/page:%d/' % (base_url, pagenum) - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) # Set the header to get a partial html page with the ids, # the normal page doesn't contain them. request.add_header('X-Requested-With', 'XMLHttpRequest') diff --git a/youtube_dl/extractor/vine.py b/youtube_dl/extractor/vine.py index c733a48fa..cb2a4b0b5 100644 --- a/youtube_dl/extractor/vine.py +++ b/youtube_dl/extractor/vine.py @@ -1,10 +1,14 @@ +# coding: utf-8 from __future__ import unicode_literals import re import itertools from .common import InfoExtractor -from ..utils import unified_strdate +from ..utils import ( + int_or_none, + unified_strdate, +) class VineIE(InfoExtractor): @@ -17,10 +21,12 @@ class VineIE(InfoExtractor): 'ext': 'mp4', 'title': 'Chicken.', 'alt_title': 'Vine by Jack Dorsey', - 'description': 'Chicken.', 'upload_date': '20130519', 'uploader': 'Jack Dorsey', 'uploader_id': '76', + 'like_count': int, + 'comment_count': int, + 'repost_count': int, }, }, { 'url': 'https://vine.co/v/MYxVapFvz2z', @@ -29,11 +35,13 @@ class VineIE(InfoExtractor): 'id': 'MYxVapFvz2z', 'ext': 'mp4', 'title': 'Fuck Da Police #Mikebrown #justice #ferguson #prayforferguson #protesting #NMOS14', - 'alt_title': 'Vine by Luna', - 'description': 'Fuck Da Police #Mikebrown #justice #ferguson #prayforferguson #protesting #NMOS14', + 'alt_title': 'Vine by Mars Ruiz', 'upload_date': '20140815', - 'uploader': 'Luna', + 'uploader': 'Mars Ruiz', 'uploader_id': '1102363502380728320', + 'like_count': int, + 'comment_count': int, + 'repost_count': int, }, }, { 'url': 'https://vine.co/v/bxVjBbZlPUH', @@ -43,14 +51,33 @@ class VineIE(InfoExtractor): 'ext': 'mp4', 'title': '#mw3 #ac130 #killcam #angelofdeath', 'alt_title': 'Vine by Z3k3', - 'description': '#mw3 #ac130 #killcam #angelofdeath', 'upload_date': '20130430', 'uploader': 'Z3k3', 'uploader_id': '936470460173008896', + 'like_count': int, + 'comment_count': int, + 'repost_count': int, }, }, { 'url': 'https://vine.co/oembed/MYxVapFvz2z.json', 'only_matching': True, + }, { + 'url': 'https://vine.co/v/e192BnZnZ9V', + 'info_dict': { + 'id': 'e192BnZnZ9V', + 'ext': 'mp4', + 'title': 'ยิ้ม~ เขิน~ อาย~ น่าร้ากอ้ะ >//< @n_whitewo @orlameena #lovesicktheseries #lovesickseason2', + 'alt_title': 'Vine by Pimry_zaa', + 'upload_date': '20150705', + 'uploader': 'Pimry_zaa', + 'uploader_id': '1135760698325307392', + 'like_count': int, + 'comment_count': int, + 'repost_count': int, + }, + 'params': { + 'skip_download': True, + }, }] def _real_extract(self, url): @@ -58,32 +85,33 @@ class VineIE(InfoExtractor): webpage = self._download_webpage('https://vine.co/v/' + video_id, video_id) data = self._parse_json( - self._html_search_regex( - r'window\.POST_DATA = { %s: ({.+?}) };\s*</script>' % video_id, + self._search_regex( + r'window\.POST_DATA\s*=\s*{\s*%s\s*:\s*({.+?})\s*};\s*</script>' % video_id, webpage, 'vine data'), video_id) formats = [{ 'format_id': '%(format)s-%(rate)s' % f, - 'vcodec': f['format'], - 'quality': f['rate'], + 'vcodec': f.get('format'), + 'quality': f.get('rate'), 'url': f['videoUrl'], - } for f in data['videoUrls']] + } for f in data['videoUrls'] if f.get('videoUrl')] self._sort_formats(formats) + username = data.get('username') + return { 'id': video_id, - 'title': self._og_search_title(webpage), - 'alt_title': self._og_search_description(webpage, default=None), - 'description': data['description'], - 'thumbnail': data['thumbnailUrl'], - 'upload_date': unified_strdate(data['created']), - 'uploader': data['username'], - 'uploader_id': data['userIdStr'], - 'like_count': data['likes']['count'], - 'comment_count': data['comments']['count'], - 'repost_count': data['reposts']['count'], + 'title': data.get('description') or self._og_search_title(webpage), + 'alt_title': 'Vine by %s' % username if username else self._og_search_description(webpage, default=None), + 'thumbnail': data.get('thumbnailUrl'), + 'upload_date': unified_strdate(data.get('created')), + 'uploader': username, + 'uploader_id': data.get('userIdStr'), + 'like_count': int_or_none(data.get('likes', {}).get('count')), + 'comment_count': int_or_none(data.get('comments', {}).get('count')), + 'repost_count': int_or_none(data.get('reposts', {}).get('count')), 'formats': formats, } diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 765e9e6fd..d99a42a9f 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -8,11 +8,11 @@ from .common import InfoExtractor from ..compat import ( compat_str, compat_urllib_parse, - compat_urllib_request, ) from ..utils import ( ExtractorError, orderedSet, + sanitized_Request, str_to_int, unescapeHTML, unified_strdate, @@ -182,7 +182,7 @@ class VKIE(InfoExtractor): 'pass': password.encode('cp1251'), }) - request = compat_urllib_request.Request( + request = sanitized_Request( 'https://login.vk.com/?act=login', compat_urllib_parse.urlencode(login_form).encode('utf-8')) login_page = self._download_webpage( @@ -281,9 +281,13 @@ class VKIE(InfoExtractor): mobj.group(1) + ' ' + mobj.group(2) upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2)) - view_count = str_to_int(self._search_regex( - r'"mv_views_count_number"[^>]*>([\d,.]+) views<', - info_page, 'view count', fatal=False)) + view_count = None + views = self._html_search_regex( + r'"mv_views_count_number"[^>]*>(.+?\bviews?)<', + info_page, 'view count', fatal=False) + if views: + view_count = str_to_int(self._search_regex( + r'([\d,.]+)', views, 'view count', fatal=False)) formats = [{ 'format_id': k, diff --git a/youtube_dl/extractor/vodlocker.py b/youtube_dl/extractor/vodlocker.py index ccf1928b5..357594a11 100644 --- a/youtube_dl/extractor/vodlocker.py +++ b/youtube_dl/extractor/vodlocker.py @@ -2,14 +2,15 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, +from ..compat import compat_urllib_parse +from ..utils import ( + ExtractorError, + sanitized_Request, ) class VodlockerIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?vodlocker\.com/(?P<id>[0-9a-zA-Z]+)(?:\..*?)?' + _VALID_URL = r'https?://(?:www\.)?vodlocker\.com/(?:embed-)?(?P<id>[0-9a-zA-Z]+)(?:\..*?)?' _TESTS = [{ 'url': 'http://vodlocker.com/e8wvyzz4sl42', @@ -26,12 +27,18 @@ class VodlockerIE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) + if any(p in webpage for p in ( + '>THIS FILE WAS DELETED<', + '>File Not Found<', + 'The file you were looking for could not be found, sorry for any inconvenience.<')): + raise ExtractorError('Video %s does not exist' % video_id, expected=True) + fields = self._hidden_inputs(webpage) if fields['op'] == 'download1': self._sleep(3, video_id) # they do detect when requests happen too fast! post = compat_urllib_parse.urlencode(fields) - req = compat_urllib_request.Request(url, post) + req = sanitized_Request(url, post) req.add_header('Content-type', 'application/x-www-form-urlencoded') webpage = self._download_webpage( req, video_id, 'Downloading video page') diff --git a/youtube_dl/extractor/voicerepublic.py b/youtube_dl/extractor/voicerepublic.py index 254383d6c..93d15a556 100644 --- a/youtube_dl/extractor/voicerepublic.py +++ b/youtube_dl/extractor/voicerepublic.py @@ -3,14 +3,12 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( ExtractorError, determine_ext, int_or_none, + sanitized_Request, ) @@ -37,7 +35,7 @@ class VoiceRepublicIE(InfoExtractor): def _real_extract(self, url): display_id = self._match_id(url) - req = compat_urllib_request.Request( + req = sanitized_Request( compat_urlparse.urljoin(url, '/talks/%s' % display_id)) # Older versions of Firefox get redirected to an "upgrade browser" page req.add_header('User-Agent', 'youtube-dl') diff --git a/youtube_dl/extractor/wistia.py b/youtube_dl/extractor/wistia.py index 13a079151..fdb16d91c 100644 --- a/youtube_dl/extractor/wistia.py +++ b/youtube_dl/extractor/wistia.py @@ -1,8 +1,10 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import compat_urllib_request -from ..utils import ExtractorError +from ..utils import ( + ExtractorError, + sanitized_Request, +) class WistiaIE(InfoExtractor): @@ -23,7 +25,7 @@ class WistiaIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - request = compat_urllib_request.Request(self._API_URL.format(video_id)) + request = sanitized_Request(self._API_URL.format(video_id)) request.add_header('Referer', url) # Some videos require this. data_json = self._download_json(request, video_id) if data_json.get('error'): diff --git a/youtube_dl/extractor/wsj.py b/youtube_dl/extractor/wsj.py index 2ddf29a69..5a897371d 100644 --- a/youtube_dl/extractor/wsj.py +++ b/youtube_dl/extractor/wsj.py @@ -84,6 +84,5 @@ class WSJIE(InfoExtractor): 'duration': duration, 'upload_date': upload_date, 'title': title, - 'formats': formats, 'categories': categories, } diff --git a/youtube_dl/extractor/gorillavid.py b/youtube_dl/extractor/xfileshare.py similarity index 79% rename from youtube_dl/extractor/gorillavid.py rename to youtube_dl/extractor/xfileshare.py index d23e3eac1..a3236e66c 100644 --- a/youtube_dl/extractor/gorillavid.py +++ b/youtube_dl/extractor/xfileshare.py @@ -1,25 +1,23 @@ -# -*- coding: utf-8 -*- +# coding: utf-8 from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, encode_dict, int_or_none, + sanitized_Request, ) -class GorillaVidIE(InfoExtractor): - IE_DESC = 'GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net and filehoot.com' +class XFileShareIE(InfoExtractor): + IE_DESC = 'XFileShare based sites: GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net, filehoot.com and vidto.me' _VALID_URL = r'''(?x) https?://(?P<host>(?:www\.)? - (?:daclips\.in|gorillavid\.in|movpod\.in|fastvideo\.in|realvid\.net|filehoot\.com))/ + (?:daclips\.in|gorillavid\.in|movpod\.in|fastvideo\.in|realvid\.net|filehoot\.com|vidto\.me))/ (?:embed-)?(?P<id>[0-9a-zA-Z]+)(?:-[0-9]+x[0-9]+\.html)? ''' @@ -76,6 +74,13 @@ class GorillaVidIE(InfoExtractor): 'title': 'youtube-dl test video \'äBaW_jenozKc.mp4.mp4', 'thumbnail': 're:http://.*\.jpg', } + }, { + 'url': 'http://vidto.me/ku5glz52nqe1.html', + 'info_dict': { + 'id': 'ku5glz52nqe1', + 'ext': 'mp4', + 'title': 'test' + } }] def _real_extract(self, url): @@ -99,18 +104,23 @@ class GorillaVidIE(InfoExtractor): post = compat_urllib_parse.urlencode(encode_dict(fields)) - req = compat_urllib_request.Request(url, post) + req = sanitized_Request(url, post) req.add_header('Content-type', 'application/x-www-form-urlencoded') webpage = self._download_webpage(req, video_id, 'Downloading video page') - title = self._search_regex( - [r'style="z-index: [0-9]+;">([^<]+)</span>', r'<td nowrap>([^<]+)</td>', r'>Watch (.+) '], - webpage, 'title', default=None) or self._og_search_title(webpage) + title = (self._search_regex( + [r'style="z-index: [0-9]+;">([^<]+)</span>', + r'<td nowrap>([^<]+)</td>', + r'>Watch (.+) ', + r'<h2 class="video-page-head">([^<]+)</h2>'], + webpage, 'title', default=None) or self._og_search_title(webpage)).strip() video_url = self._search_regex( - r'file\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'file url') + [r'file\s*:\s*["\'](http[^"\']+)["\'],', + r'file_link\s*=\s*\'(https?:\/\/[0-9a-zA-z.\/\-_]+)'], + webpage, 'file url') thumbnail = self._search_regex( - r'image\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'thumbnail', fatal=False) + r'image\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'thumbnail', default=None) formats = [{ 'format_id': 'sd', diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py index 779e4f46a..a1fe24050 100644 --- a/youtube_dl/extractor/xtube.py +++ b/youtube_dl/extractor/xtube.py @@ -3,12 +3,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urllib_parse_unquote, -) +from ..compat import compat_urllib_parse_unquote from ..utils import ( parse_duration, + sanitized_Request, str_to_int, ) @@ -32,7 +30,7 @@ class XTubeIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/xvideos.py b/youtube_dl/extractor/xvideos.py index 5dcf2fdd1..710ad5041 100644 --- a/youtube_dl/extractor/xvideos.py +++ b/youtube_dl/extractor/xvideos.py @@ -3,14 +3,12 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_unquote, - compat_urllib_request, -) +from ..compat import compat_urllib_parse_unquote from ..utils import ( clean_html, ExtractorError, determine_ext, + sanitized_Request, ) @@ -48,7 +46,7 @@ class XVideosIE(InfoExtractor): 'url': video_url, }] - android_req = compat_urllib_request.Request(url) + android_req = sanitized_Request(url) android_req.add_header('User-Agent', self._ANDROID_USER_AGENT) android_webpage = self._download_webpage(android_req, video_id, fatal=False) diff --git a/youtube_dl/extractor/yandexmusic.py b/youtube_dl/extractor/yandexmusic.py index 08dc81f3a..d3cc1a29f 100644 --- a/youtube_dl/extractor/yandexmusic.py +++ b/youtube_dl/extractor/yandexmusic.py @@ -8,11 +8,11 @@ from .common import InfoExtractor from ..compat import ( compat_str, compat_urllib_parse, - compat_urllib_request, ) from ..utils import ( int_or_none, float_or_none, + sanitized_Request, ) @@ -154,7 +154,7 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE): if len(tracks) < len(track_ids): present_track_ids = set([compat_str(track['id']) for track in tracks if track.get('id')]) missing_track_ids = set(map(compat_str, track_ids)) - set(present_track_ids) - request = compat_urllib_request.Request( + request = sanitized_Request( 'https://music.yandex.ru/handlers/track-entries.jsx', compat_urllib_parse.urlencode({ 'entries': ','.join(missing_track_ids), diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 2e81d9223..69ecc837a 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -4,12 +4,13 @@ from __future__ import unicode_literals import base64 from .common import InfoExtractor -from ..utils import ExtractorError - from ..compat import ( compat_urllib_parse, compat_ord, - compat_urllib_request, +) +from ..utils import ( + ExtractorError, + sanitized_Request, ) @@ -187,7 +188,7 @@ class YoukuIE(InfoExtractor): video_id = self._match_id(url) def retrieve_data(req_url, note): - req = compat_urllib_request.Request(req_url) + req = sanitized_Request(req_url) cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') if cn_verification_proxy: diff --git a/youtube_dl/extractor/youporn.py b/youtube_dl/extractor/youporn.py index 4ba7c36db..dd724085a 100644 --- a/youtube_dl/extractor/youporn.py +++ b/youtube_dl/extractor/youporn.py @@ -1,121 +1,171 @@ from __future__ import unicode_literals - -import json import re -import sys from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_urlparse, - compat_urllib_request, -) from ..utils import ( - ExtractorError, + int_or_none, + sanitized_Request, + str_to_int, unescapeHTML, unified_strdate, ) -from ..aes import ( - aes_decrypt_text -) +from ..aes import aes_decrypt_text class YouPornIE(InfoExtractor): - _VALID_URL = r'^(?P<proto>https?://)(?:www\.)?(?P<url>youporn\.com/watch/(?P<videoid>[0-9]+)/(?P<title>[^/]+))' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?youporn\.com/watch/(?P<id>\d+)/(?P<display_id>[^/?#&]+)' + _TESTS = [{ 'url': 'http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/', + 'md5': '71ec5fcfddacf80f495efa8b6a8d9a89', 'info_dict': { 'id': '505835', + 'display_id': 'sex-ed-is-it-safe-to-masturbate-daily', 'ext': 'mp4', - 'upload_date': '20101221', - 'description': 'Love & Sex Answers: http://bit.ly/DanAndJenn -- Is It Unhealthy To Masturbate Daily?', - 'uploader': 'Ask Dan And Jennifer', 'title': 'Sex Ed: Is It Safe To Masturbate Daily?', + 'description': 'Love & Sex Answers: http://bit.ly/DanAndJenn -- Is It Unhealthy To Masturbate Daily?', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'Ask Dan And Jennifer', + 'upload_date': '20101221', + 'average_rating': int, + 'view_count': int, + 'comment_count': int, + 'categories': list, + 'tags': list, 'age_limit': 18, - } - } + }, + }, { + # Anonymous User uploader + 'url': 'http://www.youporn.com/watch/561726/big-tits-awesome-brunette-on-amazing-webcam-show/?from=related3&al=2&from_id=561726&pos=4', + 'info_dict': { + 'id': '561726', + 'display_id': 'big-tits-awesome-brunette-on-amazing-webcam-show', + 'ext': 'mp4', + 'title': 'Big Tits Awesome Brunette On amazing webcam show', + 'description': 'http://sweetlivegirls.com Big Tits Awesome Brunette On amazing webcam show.mp4', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'Anonymous User', + 'upload_date': '20111125', + 'average_rating': int, + 'view_count': int, + 'comment_count': int, + 'categories': list, + 'tags': list, + 'age_limit': 18, + }, + 'params': { + 'skip_download': True, + }, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('videoid') - url = mobj.group('proto') + 'www.' + mobj.group('url') + video_id = mobj.group('id') + display_id = mobj.group('display_id') - req = compat_urllib_request.Request(url) - req.add_header('Cookie', 'age_verified=1') - webpage = self._download_webpage(req, video_id) - age_limit = self._rta_search(webpage) + request = sanitized_Request(url) + request.add_header('Cookie', 'age_verified=1') + webpage = self._download_webpage(request, display_id) - # Get JSON parameters - json_params = self._search_regex( - [r'videoJa?son\s*=\s*({.+})', - r'var\s+currentVideo\s*=\s*new\s+Video\((.+?)\)[,;]'], - webpage, 'JSON parameters') - try: - params = json.loads(json_params) - except ValueError: - raise ExtractorError('Invalid JSON') + title = self._search_regex( + [r'(?:video_titles|videoTitle)\s*[:=]\s*(["\'])(?P<title>.+?)\1', + r'<h1[^>]+class=["\']heading\d?["\'][^>]*>([^<])<'], + webpage, 'title', group='title') - self.report_extraction(video_id) - try: - video_title = params['title'] - upload_date = unified_strdate(params['release_date_f']) - video_description = params['description'] - video_uploader = params['submitted_by'] - thumbnail = params['thumbnails'][0]['image'] - except KeyError: - raise ExtractorError('Missing JSON parameter: ' + sys.exc_info()[1]) + links = [] - # Get all of the links from the page - DOWNLOAD_LIST_RE = r'(?s)<ul class="downloadList">(?P<download_list>.*?)</ul>' - download_list_html = self._search_regex(DOWNLOAD_LIST_RE, - webpage, 'download list').strip() - LINK_RE = r'<a href="([^"]+)">' - links = re.findall(LINK_RE, download_list_html) + sources = self._search_regex( + r'sources\s*:\s*({.+?})', webpage, 'sources', default=None) + if sources: + for _, link in re.findall(r'[^:]+\s*:\s*(["\'])(http.+?)\1', sources): + links.append(link) - # Get all encrypted links - encrypted_links = re.findall(r'var encryptedQuality[0-9]{3}URL = \'([a-zA-Z0-9+/]+={0,2})\';', webpage) - for encrypted_link in encrypted_links: - link = aes_decrypt_text(encrypted_link, video_title, 32).decode('utf-8') + # Fallback #1 + for _, link in re.findall( + r'(?:videoUrl|videoSrc|videoIpadUrl|html5PlayerSrc)\s*[:=]\s*(["\'])(http.+?)\1', webpage): links.append(link) + # Fallback #2, this also contains extra low quality 180p format + for _, link in re.findall(r'<a[^>]+href=(["\'])(http.+?)\1[^>]+title=["\']Download [Vv]ideo', webpage): + links.append(link) + + # Fallback #3, encrypted links + for _, encrypted_link in re.findall( + r'encryptedQuality\d{3,4}URL\s*=\s*(["\'])([\da-zA-Z+/=]+)\1', webpage): + links.append(aes_decrypt_text(encrypted_link, title, 32).decode('utf-8')) + formats = [] - for link in links: - # A link looks like this: - # http://cdn1.download.youporn.phncdn.com/201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4?nvb=20121113051249&nva=20121114051249&ir=1200&sr=1200&hash=014b882080310e95fb6a0 - # A path looks like this: - # /201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4 - video_url = unescapeHTML(link) - path = compat_urllib_parse_urlparse(video_url).path - format_parts = path.split('/')[4].split('_')[:2] - - dn = compat_urllib_parse_urlparse(video_url).netloc.partition('.')[0] - - resolution = format_parts[0] - height = int(resolution[:-len('p')]) - bitrate = int(format_parts[1][:-len('k')]) - format = '-'.join(format_parts) + '-' + dn - - formats.append({ + for video_url in set(unescapeHTML(link) for link in links): + f = { 'url': video_url, - 'format': format, - 'format_id': format, - 'height': height, - 'tbr': bitrate, - 'resolution': resolution, - }) - + } + # Video URL's path looks like this: + # /201012/17/505835/720p_1500k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4 + # We will benefit from it by extracting some metadata + mobj = re.search(r'/(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+/', video_url) + if mobj: + height = int(mobj.group('height')) + bitrate = int(mobj.group('bitrate')) + f.update({ + 'format_id': '%dp-%dk' % (height, bitrate), + 'height': height, + 'tbr': bitrate, + }) + formats.append(f) self._sort_formats(formats) - if not formats: - raise ExtractorError('ERROR: no known formats available for video') + description = self._html_search_regex( + r'(?s)<div[^>]+class=["\']video-description["\'][^>]*>(.+?)</div>', + webpage, 'description', default=None) + thumbnail = self._search_regex( + r'(?:imageurl\s*=|poster\s*:)\s*(["\'])(?P<thumbnail>.+?)\1', + webpage, 'thumbnail', fatal=False, group='thumbnail') + + uploader = self._html_search_regex( + r'(?s)<div[^>]+class=["\']videoInfoBy["\'][^>]*>\s*By:\s*</div>(.+?)</(?:a|div)>', + webpage, 'uploader', fatal=False) + upload_date = unified_strdate(self._html_search_regex( + r'(?s)<div[^>]+class=["\']videoInfoTime["\'][^>]*>(.+?)</div>', + webpage, 'upload date', fatal=False)) + + age_limit = self._rta_search(webpage) + + average_rating = int_or_none(self._search_regex( + r'<div[^>]+class=["\']videoInfoRating["\'][^>]*>\s*<div[^>]+class=["\']videoRatingPercentage["\'][^>]*>(\d+)%</div>', + webpage, 'average rating', fatal=False)) + + view_count = str_to_int(self._search_regex( + r'(?s)<div[^>]+class=["\']videoInfoViews["\'][^>]*>.*?([\d,.]+)\s*</div>', + webpage, 'view count', fatal=False)) + comment_count = str_to_int(self._search_regex( + r'>All [Cc]omments? \(([\d,.]+)\)', + webpage, 'comment count', fatal=False)) + + def extract_tag_box(title): + tag_box = self._search_regex( + (r'<div[^>]+class=["\']tagBoxTitle["\'][^>]*>\s*%s\b.*?</div>\s*' + '<div[^>]+class=["\']tagBoxContent["\']>(.+?)</div>') % re.escape(title), + webpage, '%s tag box' % title, default=None) + if not tag_box: + return [] + return re.findall(r'<a[^>]+href=[^>]+>([^<]+)', tag_box) + + categories = extract_tag_box('Category') + tags = extract_tag_box('Tags') return { 'id': video_id, - 'uploader': video_uploader, - 'upload_date': upload_date, - 'title': video_title, + 'display_id': display_id, + 'title': title, + 'description': description, 'thumbnail': thumbnail, - 'description': video_description, + 'uploader': uploader, + 'upload_date': upload_date, + 'average_rating': average_rating, + 'view_count': view_count, + 'comment_count': comment_count, + 'categories': categories, + 'tags': tags, 'age_limit': age_limit, 'formats': formats, } diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index b252e36e1..9b39505ba 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -20,7 +20,6 @@ from ..compat import ( compat_urllib_parse_unquote, compat_urllib_parse_unquote_plus, compat_urllib_parse_urlparse, - compat_urllib_request, compat_urlparse, compat_str, ) @@ -35,6 +34,7 @@ from ..utils import ( orderedSet, parse_duration, remove_start, + sanitized_Request, smuggle_url, str_to_int, unescapeHTML, @@ -114,7 +114,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor): login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('ascii') - req = compat_urllib_request.Request(self._LOGIN_URL, login_data) + req = sanitized_Request(self._LOGIN_URL, login_data) login_results = self._download_webpage( req, None, note='Logging in', errnote='unable to log in', fatal=False) @@ -147,7 +147,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor): tfa_data = compat_urllib_parse.urlencode(encode_dict(tfa_form_strs)).encode('ascii') - tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data) + tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data) tfa_results = self._download_webpage( tfa_req, None, note='Submitting TFA code', errnote='unable to submit tfa', fatal=False) @@ -178,6 +178,69 @@ class YoutubeBaseInfoExtractor(InfoExtractor): return +class YoutubeEntryListBaseInfoExtractor(InfoExtractor): + # Extract entries from page with "Load more" button + def _entries(self, page, playlist_id): + more_widget_html = content_html = page + for page_num in itertools.count(1): + for entry in self._process_page(content_html): + yield entry + + mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) + if not mobj: + break + + more = self._download_json( + 'https://youtube.com/%s' % mobj.group('more'), playlist_id, + 'Downloading page #%s' % page_num, + transform_source=uppercase_escape) + content_html = more['content_html'] + if not content_html.strip(): + # Some webpages show a "Load more" button but they don't + # have more videos + break + more_widget_html = more['load_more_widget_html'] + + +class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): + def _process_page(self, content): + for video_id, video_title in self.extract_videos_from_page(content): + yield self.url_result(video_id, 'Youtube', video_id, video_title) + + def extract_videos_from_page(self, page): + ids_in_page = [] + titles_in_page = [] + for mobj in re.finditer(self._VIDEO_RE, page): + # The link with index 0 is not the first video of the playlist (not sure if still actual) + if 'index' in mobj.groupdict() and mobj.group('id') == '0': + continue + video_id = mobj.group('id') + video_title = unescapeHTML(mobj.group('title')) + if video_title: + video_title = video_title.strip() + try: + idx = ids_in_page.index(video_id) + if video_title and not titles_in_page[idx]: + titles_in_page[idx] = video_title + except ValueError: + ids_in_page.append(video_id) + titles_in_page.append(video_title) + return zip(ids_in_page, titles_in_page) + + +class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): + def _process_page(self, content): + for playlist_id in re.findall(r'href="/?playlist\?list=(.+?)"', content): + yield self.url_result( + 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist') + + def _real_extract(self, url): + playlist_id = self._match_id(url) + webpage = self._download_webpage(url, playlist_id) + title = self._og_search_title(webpage, fatal=False) + return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title) + + class YoutubeIE(YoutubeBaseInfoExtractor): IE_DESC = 'YouTube.com' _VALID_URL = r"""(?x)^ @@ -195,7 +258,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): |(?: # or the v= param in all its forms (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) (?:\?|\#!?) # the params delimiter ? or # or #! - (?:.*?&)?? # any other preceding param (like /?s=tuff&v=xxxx) + (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY) v= ) )) @@ -283,6 +346,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug) '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}, '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}, @@ -363,7 +427,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012', 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7', 'uploader': 'SET India', - 'uploader_id': 'setindia' + 'uploader_id': 'setindia', + 'age_limit': 18, } }, { @@ -500,7 +565,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'info_dict': { 'id': 'lqQg6PlCWgI', 'ext': 'mp4', - 'upload_date': '20120724', + 'upload_date': '20150827', 'uploader_id': 'olympic', 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games', 'uploader': 'Olympics', @@ -628,6 +693,47 @@ class YoutubeIE(YoutubeBaseInfoExtractor): { 'url': 'http://vid.plus/FlRa-iH7PGw', 'only_matching': True, + }, + { + # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468) + 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg', + 'info_dict': { + 'id': 'lsguqyKfVQg', + 'ext': 'mp4', + 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21', + 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a', + 'upload_date': '20151119', + 'uploader_id': 'IronSoulElf', + 'uploader': 'IronSoulElf', + }, + 'params': { + 'skip_download': True, + }, + }, + { + # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468) + 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8', + 'only_matching': True, + }, + { + # Video with yt:stretch=17:0 + 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM', + 'info_dict': { + 'id': 'Q39EVAstoRM', + 'ext': 'mp4', + 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4', + 'description': 'md5:ee18a25c350637c8faff806845bddee9', + 'upload_date': '20151107', + 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA', + 'uploader': 'CH GAMER DROID', + }, + 'params': { + 'skip_download': True, + }, + }, + { + 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY', + 'only_matching': True, } ] @@ -657,7 +763,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): def _extract_signature_function(self, video_id, player_url, example_sig): id_m = re.match( - r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?)?\.(?P<ext>[a-z]+)$', + r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$', player_url) if not id_m: raise ExtractorError('Cannot identify player %r' % player_url) @@ -812,16 +918,33 @@ class YoutubeIE(YoutubeBaseInfoExtractor): return {} return sub_lang_list + def _get_ytplayer_config(self, video_id, webpage): + patterns = ( + # User data may contain arbitrary character sequences that may affect + # JSON extraction with regex, e.g. when '};' is contained the second + # regex won't capture the whole JSON. Yet working around by trying more + # concrete regex first keeping in mind proper quoted string handling + # to be implemented in future that will replace this workaround (see + # https://github.com/rg3/youtube-dl/issues/7468, + # https://github.com/rg3/youtube-dl/pull/7599) + r';ytplayer\.config\s*=\s*({.+?});ytplayer', + r';ytplayer\.config\s*=\s*({.+?});', + ) + config = self._search_regex( + patterns, webpage, 'ytplayer.config', default=None) + if config: + return self._parse_json( + uppercase_escape(config), video_id, fatal=False) + def _get_automatic_captions(self, video_id, webpage): """We need the webpage for getting the captions url, pass it as an argument to speed up the process.""" self.to_screen('%s: Looking for automatic captions' % video_id) - mobj = re.search(r';ytplayer.config = ({.*?});', webpage) + player_config = self._get_ytplayer_config(video_id, webpage) err_msg = 'Couldn\'t find automatic captions for %s' % video_id - if mobj is None: + if not player_config: self._downloader.report_warning(err_msg) return {} - player_config = json.loads(mobj.group(1)) try: args = player_config['args'] caption_url = args['ttsurl'] @@ -1028,10 +1151,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): age_gate = False video_info = None # Try looking directly into the video webpage - mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage) - if mobj: - json_code = uppercase_escape(mobj.group(1)) - ytplayer_config = json.loads(json_code) + ytplayer_config = self._get_ytplayer_config(video_id, video_webpage) + if ytplayer_config: args = ytplayer_config['args'] if args.get('url_encoded_fmt_stream_map'): # Convert to the same format returned by compat_parse_qs @@ -1061,6 +1182,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if not video_info: video_info = get_video_info if 'token' in get_video_info: + # Different get_video_info requests may report different results, e.g. + # some may report video unavailability, but some may serve it without + # any complaint (see https://github.com/rg3/youtube-dl/issues/7362, + # the original webpage as well as el=info and el=embedded get_video_info + # requests report video unavailability due to geo restriction while + # el=detailpage succeeds and returns valid data). This is probably + # due to YouTube measures against IP ranges of hosting providers. + # Working around by preferring the first succeeded video_info containing + # the token if no such video_info yet was found. + if 'token' not in video_info: + video_info = get_video_info break if 'token' not in video_info: if 'reason' in video_info: @@ -1286,7 +1418,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): player_desc = 'flash player %s' % player_version else: player_version = self._search_regex( - r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', + [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'], player_url, 'html5 player', fatal=False) player_desc = 'html5 player %s' % player_version @@ -1348,6 +1480,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): manifest_url = video_info['hlsvp'][0] url_map = self._extract_from_m3u8(manifest_url, video_id) formats = _map_to_format_list(url_map) + # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming + for a_format in formats: + a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True' else: raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') @@ -1385,10 +1520,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor): r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">', video_webpage) if stretched_m: - ratio = float(stretched_m.group('w')) / float(stretched_m.group('h')) - for f in formats: - if f.get('vcodec') != 'none': - f['stretched_ratio'] = ratio + w = float(stretched_m.group('w')) + h = float(stretched_m.group('h')) + # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0). + # We will only process correct ratios. + if w > 0 and h > 0: + ratio = w / h + for f in formats: + if f.get('vcodec') != 'none': + f['stretched_ratio'] = ratio self._sort_formats(formats) @@ -1419,7 +1559,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): } -class YoutubePlaylistIE(YoutubeBaseInfoExtractor): +class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com playlists' _VALID_URL = r"""(?x)(?: (?:https?://)? @@ -1427,7 +1567,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): youtube\.com/ (?: (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries) - \? (?:.*?&)*? (?:p|a|list)= + \? (?:.*?[&;])*? (?:p|a|list)= | p/ ) ( @@ -1440,7 +1580,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,}) )""" _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s' - _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)' + _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?' IE_NAME = 'youtube:playlist' _TESTS = [{ 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', @@ -1557,37 +1697,11 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): else: self.report_warning('Youtube gives an alert message: ' + match) - # Extract the video ids from the playlist pages - def _entries(): - more_widget_html = content_html = page - for page_num in itertools.count(1): - matches = re.finditer(self._VIDEO_RE, content_html) - # We remove the duplicates and the link with index 0 - # (it's not the first video of the playlist) - new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') - for vid_id in new_ids: - yield self.url_result(vid_id, 'Youtube', video_id=vid_id) - - mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) - if not mobj: - break - - more = self._download_json( - 'https://youtube.com/%s' % mobj.group('more'), playlist_id, - 'Downloading page #%s' % page_num, - transform_source=uppercase_escape) - content_html = more['content_html'] - if not content_html.strip(): - # Some webpages show a "Load more" button but they don't - # have more videos - break - more_widget_html = more['load_more_widget_html'] - playlist_title = self._html_search_regex( - r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>', + r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>', page, 'title') - return self.playlist_result(_entries(), playlist_id, playlist_title) + return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title) def _real_extract(self, url): # Extract playlist id @@ -1613,36 +1727,31 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): return self._extract_playlist(playlist_id) -class YoutubeChannelIE(InfoExtractor): +class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com channels' _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)' _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos' + _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?' IE_NAME = 'youtube:channel' _TESTS = [{ 'note': 'paginated channel', 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', 'playlist_mincount': 91, 'info_dict': { - 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w', + 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w', + 'title': 'Uploads from lex will', } + }, { + 'note': 'Age restricted channel', + # from https://www.youtube.com/user/DeusExOfficial + 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w', + 'playlist_mincount': 64, + 'info_dict': { + 'id': 'UUs0ifCMCm1icqRbqhUINa0w', + 'title': 'Uploads from Deus Ex', + }, }] - @staticmethod - def extract_videos_from_page(page): - ids_in_page = [] - titles_in_page = [] - for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page): - video_id = mobj.group('id') - video_title = unescapeHTML(mobj.group('title')) - try: - idx = ids_in_page.index(video_id) - if video_title and not titles_in_page[idx]: - titles_in_page[idx] = video_title - except ValueError: - ids_in_page.append(video_id) - titles_in_page.append(video_title) - return zip(ids_in_page, titles_in_page) - def _real_extract(self, url): channel_id = self._match_id(url) @@ -1661,7 +1770,7 @@ class YoutubeChannelIE(InfoExtractor): 'channelId', channel_page, 'channel id', default=None) if not channel_playlist_id: channel_playlist_id = self._search_regex( - r'data-channel-external-id="([^"]+)"', + r'data-(?:channel-external-|yt)id="([^"]+)"', channel_page, 'channel id', default=None) if channel_playlist_id and channel_playlist_id.startswith('UC'): playlist_id = 'UU' + channel_playlist_id[2:] @@ -1685,29 +1794,7 @@ class YoutubeChannelIE(InfoExtractor): for video_id, video_title in self.extract_videos_from_page(channel_page)] return self.playlist_result(entries, channel_id) - def _entries(): - more_widget_html = content_html = channel_page - for pagenum in itertools.count(1): - - for video_id, video_title in self.extract_videos_from_page(content_html): - yield self.url_result( - video_id, 'Youtube', video_id=video_id, - video_title=video_title) - - mobj = re.search( - r'data-uix-load-more-href="/?(?P<more>[^"]+)"', - more_widget_html) - if not mobj: - break - - more = self._download_json( - 'https://youtube.com/%s' % mobj.group('more'), channel_id, - 'Downloading page #%s' % (pagenum + 1), - transform_source=uppercase_escape) - content_html = more['content_html'] - more_widget_html = more['load_more_widget_html'] - - return self.playlist_result(_entries(), channel_id) + return self.playlist_result(self._entries(channel_page, channel_id), channel_id) class YoutubeUserIE(YoutubeChannelIE): @@ -1738,6 +1825,29 @@ class YoutubeUserIE(YoutubeChannelIE): return super(YoutubeUserIE, cls).suitable(url) +class YoutubeUserPlaylistsIE(YoutubePlaylistsBaseInfoExtractor): + IE_DESC = 'YouTube.com user playlists' + _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/user/(?P<id>[^/]+)/playlists' + IE_NAME = 'youtube:user:playlists' + + _TESTS = [{ + 'url': 'http://www.youtube.com/user/ThirstForScience/playlists', + 'playlist_mincount': 4, + 'info_dict': { + 'id': 'ThirstForScience', + 'title': 'Thirst for Science', + }, + }, { + # with "Load more" button + 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd', + 'playlist_mincount': 70, + 'info_dict': { + 'id': 'igorkle1', + 'title': 'Игорь Клейнер', + }, + }] + + class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE): IE_DESC = 'YouTube.com searches' # there doesn't appear to be a real limit, for example if you search for @@ -1833,7 +1943,7 @@ class YoutubeSearchURLIE(InfoExtractor): } -class YoutubeShowIE(InfoExtractor): +class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor): IE_DESC = 'YouTube.com (multi-season) shows' _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)' IE_NAME = 'youtube:show' @@ -1847,26 +1957,9 @@ class YoutubeShowIE(InfoExtractor): }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - playlist_id = mobj.group('id') - webpage = self._download_webpage( - 'https://www.youtube.com/show/%s/playlists' % playlist_id, playlist_id, 'Downloading show webpage') - # There's one playlist for each season of the show - m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) - self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons))) - entries = [ - self.url_result( - 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist') - for season in m_seasons - ] - title = self._og_search_title(webpage, fatal=False) - - return { - '_type': 'playlist', - 'id': playlist_id, - 'title': title, - 'entries': entries, - } + playlist_id = self._match_id(url) + return super(YoutubeShowIE, self)._real_extract( + 'https://www.youtube.com/show/%s/playlists' % playlist_id) class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py index 98f15177b..a795f56b3 100644 --- a/youtube_dl/extractor/zdf.py +++ b/youtube_dl/extractor/zdf.py @@ -9,6 +9,7 @@ from ..utils import ( int_or_none, unified_strdate, OnDemandPagedList, + xpath_text, ) @@ -19,13 +20,11 @@ def extract_from_xml_url(ie, video_id, xml_url): errnote='Failed to download video info') title = doc.find('.//information/title').text - description = doc.find('.//information/detail').text - duration = int(doc.find('.//details/lengthSec').text) - uploader_node = doc.find('.//details/originChannelTitle') - uploader = None if uploader_node is None else uploader_node.text - uploader_id_node = doc.find('.//details/originChannelId') - uploader_id = None if uploader_id_node is None else uploader_id_node.text - upload_date = unified_strdate(doc.find('.//details/airtime').text) + description = xpath_text(doc, './/information/detail', 'description') + duration = int_or_none(xpath_text(doc, './/details/lengthSec', 'duration')) + uploader = xpath_text(doc, './/details/originChannelTitle', 'uploader') + uploader_id = xpath_text(doc, './/details/originChannelId', 'uploader id') + upload_date = unified_strdate(xpath_text(doc, './/details/airtime', 'upload date')) def xml_to_format(fnode): video_url = fnode.find('url').text @@ -40,15 +39,14 @@ def extract_from_xml_url(ie, video_id, xml_url): ext = format_m.group('container') proto = format_m.group('proto').lower() - quality = fnode.find('./quality').text - abr = int(fnode.find('./audioBitrate').text) // 1000 - vbr_node = fnode.find('./videoBitrate') - vbr = None if vbr_node is None else int(vbr_node.text) // 1000 + quality = xpath_text(fnode, './quality', 'quality') + abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000) + vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000) - width_node = fnode.find('./width') - width = None if width_node is None else int_or_none(width_node.text) - height_node = fnode.find('./height') - height = None if height_node is None else int_or_none(height_node.text) + width = int_or_none(xpath_text(fnode, './width', 'width')) + height = int_or_none(xpath_text(fnode, './height', 'height')) + + filesize = int_or_none(xpath_text(fnode, './filesize', 'filesize')) format_note = '' if not format_note: @@ -64,12 +62,31 @@ def extract_from_xml_url(ie, video_id, xml_url): 'vbr': vbr, 'width': width, 'height': height, - 'filesize': int_or_none(fnode.find('./filesize').text), + 'filesize': filesize, 'format_note': format_note, 'protocol': proto, '_available': is_available, } + def xml_to_thumbnails(fnode): + thumbnails = [] + for node in fnode: + thumbnail_url = node.text + if not thumbnail_url: + continue + thumbnail = { + 'url': thumbnail_url, + } + if 'key' in node.attrib: + m = re.match('^([0-9]+)x([0-9]+)$', node.attrib['key']) + if m: + thumbnail['width'] = int(m.group(1)) + thumbnail['height'] = int(m.group(2)) + thumbnails.append(thumbnail) + return thumbnails + + thumbnails = xml_to_thumbnails(doc.findall('.//teaserimages/teaserimage')) + format_nodes = doc.findall('.//formitaeten/formitaet') formats = list(filter( lambda f: f['_available'], @@ -81,6 +98,7 @@ def extract_from_xml_url(ie, video_id, xml_url): 'title': title, 'description': description, 'duration': duration, + 'thumbnails': thumbnails, 'uploader': uploader, 'uploader_id': uploader_id, 'upload_date': upload_date, diff --git a/youtube_dl/jsinterp.py b/youtube_dl/jsinterp.py index 0e0c7d90d..2191e8b89 100644 --- a/youtube_dl/jsinterp.py +++ b/youtube_dl/jsinterp.py @@ -214,7 +214,7 @@ class JSInterpreter(object): obj = {} obj_m = re.search( (r'(?:var\s+)?%s\s*=\s*\{' % re.escape(objname)) + - r'\s*(?P<fields>([a-zA-Z$0-9]+\s*:\s*function\(.*?\)\s*\{.*?\})*)' + + r'\s*(?P<fields>([a-zA-Z$0-9]+\s*:\s*function\(.*?\)\s*\{.*?\}(?:,\s*)?)*)' + r'\}\s*;', self.code) fields = obj_m.group('fields') @@ -232,10 +232,10 @@ class JSInterpreter(object): def extract_function(self, funcname): func_m = re.search( r'''(?x) - (?:function\s+%s|[{;]%s\s*=\s*function)\s* + (?:function\s+%s|[{;]%s\s*=\s*function|var\s+%s\s*=\s*function)\s* \((?P<args>[^)]*)\)\s* \{(?P<code>[^}]+)\}''' % ( - re.escape(funcname), re.escape(funcname)), + re.escape(funcname), re.escape(funcname), re.escape(funcname)), self.code) if func_m is None: raise ExtractorError('Could not find JS function %r' % funcname) diff --git a/youtube_dl/options.py b/youtube_dl/options.py index 3dd6d290b..c46e136bf 100644 --- a/youtube_dl/options.py +++ b/youtube_dl/options.py @@ -338,7 +338,7 @@ def parseOpts(overrideArguments=None): video_format.add_option( '-F', '--list-formats', action='store_true', dest='listformats', - help='List all available formats') + help='List all available formats of requested videos') video_format.add_option( '--youtube-include-dash-manifest', action='store_true', dest='youtube_include_dash_manifest', default=True, @@ -363,7 +363,7 @@ def parseOpts(overrideArguments=None): subtitles.add_option( '--write-auto-sub', '--write-automatic-sub', action='store_true', dest='writeautomaticsub', default=False, - help='Write automatic subtitle file (YouTube only)') + help='Write automatically generated subtitle file (YouTube only)') subtitles.add_option( '--all-subs', action='store_true', dest='allsubtitles', default=False, diff --git a/youtube_dl/postprocessor/ffmpeg.py b/youtube_dl/postprocessor/ffmpeg.py index 4f320e124..5ed723bc6 100644 --- a/youtube_dl/postprocessor/ffmpeg.py +++ b/youtube_dl/postprocessor/ffmpeg.py @@ -272,7 +272,7 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor): return [], information try: - self._downloader.to_screen('[' + self.basename + '] Destination: ' + new_path) + self._downloader.to_screen('[ffmpeg] Destination: ' + new_path) self.run_ffmpeg(path, new_path, acodec, more_opts) except AudioConversionError as e: raise PostProcessingError( diff --git a/youtube_dl/update.py b/youtube_dl/update.py index fc7ac8305..074eb64a7 100644 --- a/youtube_dl/update.py +++ b/youtube_dl/update.py @@ -9,11 +9,8 @@ import subprocess import sys from zipimport import zipimporter -from .compat import ( - compat_str, - compat_urllib_request, -) -from .utils import make_HTTPS_handler +from .compat import compat_str + from .version import __version__ @@ -47,7 +44,7 @@ def rsa_verify(message, signature, key): return True -def update_self(to_screen, verbose): +def update_self(to_screen, verbose, opener): """Update the program file with the latest version from the repository""" UPDATE_URL = "https://rg3.github.io/youtube-dl/update/" @@ -59,9 +56,6 @@ def update_self(to_screen, verbose): to_screen('It looks like you installed youtube-dl with a package manager, pip, setup.py or a tarball. Please use that to update.') return - https_handler = make_HTTPS_handler({}) - opener = compat_urllib_request.build_opener(https_handler) - # Check if there is a new version try: newversion = opener.open(VERSION_URL).read().decode('utf-8').strip() diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 7dbe25661..d0606b4bc 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -3,6 +3,7 @@ from __future__ import unicode_literals +import base64 import calendar import codecs import contextlib @@ -35,6 +36,7 @@ import zlib from .compat import ( compat_basestring, compat_chr, + compat_etree_fromstring, compat_html_entities, compat_http_client, compat_kwargs, @@ -177,10 +179,19 @@ def xpath_with_ns(path, ns_map): def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT): - if sys.version_info < (2, 7): # Crazy 2.6 - xpath = xpath.encode('ascii') + def _find_xpath(xpath): + if sys.version_info < (2, 7): # Crazy 2.6 + xpath = xpath.encode('ascii') + return node.find(xpath) + + if isinstance(xpath, (str, compat_str)): + n = _find_xpath(xpath) + else: + for xp in xpath: + n = _find_xpath(xp) + if n is not None: + break - n = node.find(xpath) if n is None: if default is not NO_DEFAULT: return default @@ -355,13 +366,20 @@ def sanitize_path(s): if drive_or_unc: norm_path.pop(0) sanitized_path = [ - path_part if path_part in ['.', '..'] else re.sub('(?:[/<>:"\\|\\\\?\\*]|\.$)', '#', path_part) + path_part if path_part in ['.', '..'] else re.sub('(?:[/<>:"\\|\\\\?\\*]|[\s.]$)', '#', path_part) for path_part in norm_path] if drive_or_unc: sanitized_path.insert(0, drive_or_unc + os.path.sep) return os.path.join(*sanitized_path) +# Prepend protocol-less URLs with `http:` scheme in order to mitigate the number of +# unwanted failures due to missing protocol +def sanitized_Request(url, *args, **kwargs): + return compat_urllib_request.Request( + 'http:%s' % url if url.startswith('//') else url, *args, **kwargs) + + def orderedSet(iterable): """ Remove all duplicates from the input iterable """ res = [] @@ -385,10 +403,14 @@ def _htmlentity_transform(entity): numstr = '0%s' % numstr else: base = 10 - return compat_chr(int(numstr, base)) + # See https://github.com/rg3/youtube-dl/issues/7518 + try: + return compat_chr(int(numstr, base)) + except ValueError: + pass # Unknown entity in name, return its literal representation - return ('&%s;' % entity) + return '&%s;' % entity def unescapeHTML(s): @@ -641,6 +663,16 @@ def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs): return hc +def handle_youtubedl_headers(headers): + filtered_headers = headers + + if 'Youtubedl-no-compression' in filtered_headers: + filtered_headers = dict((k, v) for k, v in filtered_headers.items() if k.lower() != 'accept-encoding') + del filtered_headers['Youtubedl-no-compression'] + + return filtered_headers + + class YoutubeDLHandler(compat_urllib_request.HTTPHandler): """Handler for HTTP requests and responses. @@ -648,7 +680,7 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler): the standard headers to every HTTP request and handles gzipped and deflated responses from web servers. If compression is to be avoided in a particular request, the original request in the program code only has - to include the HTTP header "Youtubedl-No-Compression", which will be + to include the HTTP header "Youtubedl-no-compression", which will be removed before making the real request. Part of this code was copied from: @@ -709,10 +741,8 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler): # 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'] + + req.headers = handle_youtubedl_headers(req.headers) if sys.version_info < (2, 7) and '#' in req.get_full_url(): # Python 2.6 is brain-dead when it comes to fragments @@ -813,9 +843,11 @@ def parse_iso8601(date_str, delimiter='T', timezone=None): if date_str is None: return None + date_str = re.sub(r'\.[0-9]+', '', date_str) + if timezone is None: m = re.search( - r'(\.[0-9]+)?(?:Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)', + r'(?:Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)', date_str) if not m: timezone = datetime.timedelta() @@ -828,9 +860,12 @@ def parse_iso8601(date_str, delimiter='T', timezone=None): timezone = datetime.timedelta( hours=sign * int(m.group('hours')), minutes=sign * int(m.group('minutes'))) - date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) - dt = datetime.datetime.strptime(date_str, date_format) - timezone - return calendar.timegm(dt.timetuple()) + try: + date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) + dt = datetime.datetime.strptime(date_str, date_format) - timezone + return calendar.timegm(dt.timetuple()) + except ValueError: + pass def unified_strdate(date_str, day_first=True): @@ -895,7 +930,8 @@ def unified_strdate(date_str, day_first=True): timetuple = email.utils.parsedate_tz(date_str) if timetuple: upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d') - return upload_date + if upload_date is not None: + return compat_str(upload_date) def determine_ext(url, default_ext='unknown_video'): @@ -904,6 +940,21 @@ def determine_ext(url, default_ext='unknown_video'): guess = url.partition('?')[0].rpartition('.')[2] if re.match(r'^[A-Za-z0-9]+$', guess): return guess + elif guess.rstrip('/') in ( + 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac', + 'flv', 'f4v', 'f4a', 'f4b', + 'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus', + 'mkv', 'mka', 'mk3d', + 'avi', 'divx', + 'mov', + 'asf', 'wmv', 'wma', + '3gp', '3g2', + 'mp3', + 'flac', + 'ape', + 'wav', + 'f4f', 'f4m', 'm3u8', 'smil'): + return guess.rstrip('/') else: return default_ext @@ -1647,30 +1698,9 @@ def urlencode_postdata(*args, **kargs): def encode_dict(d, encoding='utf-8'): - return dict((k.encode(encoding), v.encode(encoding)) for k, v in d.items()) - - -try: - etree_iter = xml.etree.ElementTree.Element.iter -except AttributeError: # Python <=2.6 - etree_iter = lambda n: n.findall('.//*') - - -def parse_xml(s): - class TreeBuilder(xml.etree.ElementTree.TreeBuilder): - def doctype(self, name, pubid, system): - pass # Ignore doctypes - - parser = xml.etree.ElementTree.XMLParser(target=TreeBuilder()) - kwargs = {'parser': parser} if sys.version_info >= (2, 7) else {} - tree = xml.etree.ElementTree.XML(s.encode('utf-8'), **kwargs) - # Fix up XML parser in Python 2.x - if sys.version_info < (3, 0): - for n in etree_iter(tree): - if n.text is not None: - if not isinstance(n.text, compat_str): - n.text = n.text.decode('utf-8') - return tree + def encode(v): + return v.encode(encoding) if isinstance(v, compat_basestring) else v + return dict((encode(k), encode(v)) for k, v in d.items()) US_RATINGS = { @@ -1700,8 +1730,8 @@ def js_to_json(code): if v in ('true', 'false', 'null'): return v if v.startswith('"'): - return v - if v.startswith("'"): + v = re.sub(r"\\'", "'", v[1:-1]) + elif v.startswith("'"): v = v[1:-1] v = re.sub(r"\\\\|\\'|\"", lambda m: { '\\\\': '\\\\', @@ -1795,6 +1825,10 @@ def urlhandle_detect_ext(url_handle): return mimetype2ext(getheader('Content-Type')) +def encode_data_uri(data, mime_type): + return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii')) + + def age_restricted(content_limit, age_limit): """ Returns True iff the content should be blocked """ @@ -1969,7 +2003,7 @@ def dfxp2srt(dfxp_data): return out - dfxp = xml.etree.ElementTree.fromstring(dfxp_data.encode('utf-8')) + dfxp = compat_etree_fromstring(dfxp_data.encode('utf-8')) out = [] paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall(_x('.//ttaf1:p')) or dfxp.findall('.//p') diff --git a/youtube_dl/version.py b/youtube_dl/version.py index aaa43d315..bd0de9f53 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,3 +1,3 @@ from __future__ import unicode_literals -__version__ = '2015.10.13' +__version__ = '2015.11.27.1'