diff options
33 files changed, 1255 insertions, 315 deletions
@@ -103,3 +103,4 @@ Christopher Krooss Ondřej Caletka Dinesh S Johan K. Jensen +Yen Chi Hsuan @@ -267,10 +267,22 @@ which means you can modify it, redistribute it or use it however you like. by extension for the extensions aac, m4a, mp3, mp4, ogg, wav, webm. You can also use the special names "best", "bestvideo", - "bestaudio", "worst". By default, youtube- - dl will pick the best quality. Use commas - to download multiple audio formats, such as - -f + "bestaudio", "worst". You can filter the + video results by putting a condition in + brackets, as in -f "best[height=720]" (or + -f "[filesize>10M]"). This works for + filesize, height, width, tbr, abr, and vbr + and the comparisons <, <=, >, >=, =, != . + Formats for which the value is not known + are excluded unless you put a question mark + (?) after the operator. You can combine + format filters, so -f "[height <=? + 720][tbr>500]" selects up to 720p videos + (or videos where the height is not known) + with a bitrate of at least 500 KBit/s. By + default, youtube-dl will pick the best + quality. Use commas to download multiple + audio formats, such as -f 136/137/mp4/bestvideo,140/m4a/bestaudio. You can merge the video and audio of two formats into a single file using -f <video- @@ -304,7 +316,8 @@ which means you can modify it, redistribute it or use it however you like. ## Authentication Options: -u, --username USERNAME login with this account ID - -p, --password PASSWORD account password + -p, --password PASSWORD account password. If this option is left + out, youtube-dl will ask interactively. -2, --twofactor TWOFACTOR two-factor auth code -n, --netrc use .netrc authentication data --video-password PASSWORD video password (vimeo, smotri) @@ -444,9 +457,15 @@ Apparently YouTube requires you to pass a CAPTCHA test if you download too much. Once the video is fully downloaded, use any video player, such as [vlc](http://www.videolan.org) or [mplayer](http://www.mplayerhq.hu/). -### The links provided by youtube-dl -g are not working anymore +### I extracted a video URL with -g, but it does not play on another machine / in my webbrowser. -The URLs youtube-dl outputs require the downloader to have the correct cookies. Use the `--cookies` option to write the required cookies into a file, and advise your downloader to read cookies from that file. Some sites also require a common user agent to be used, use `--dump-user-agent` to see the one in use by youtube-dl. +It depends a lot on the service. In many cases, requests for the video (to download/play it) must come from the same IP address and with the same cookies. Use the `--cookies` option to write the required cookies into a file, and advise your downloader to read cookies from that file. Some sites also require a common user agent to be used, use `--dump-user-agent` to see the one in use by youtube-dl. + +It may be beneficial to use IPv6; in some cases, the restrictions are only applied to IPv4. Some services (sometimes only for a subset of videos) do not restrict the video URL by IP address, cookie, or user-agent, but these are the exception rather than the rule. + +Please bear in mind that some URL protocols are **not** supported by browsers out of the box, including RTMP. If you are using -g, your own downloader must support these as well. + +If you want to play the video on a machine that is not running youtube-dl, you can relay the video content from the machine that runs youtube-dl. You can use `-o -` to let youtube-dl stream a video to stdout, or simply allow the player to download the files written by youtube-dl in turn. ### ERROR: no fmt_url_map or conn information found in video info @@ -481,6 +500,10 @@ To make a different directory work - either for ffmpeg, or for youtube-dl, or fo From then on, after restarting your shell, you will be able to access both youtube-dl and ffmpeg (and youtube-dl will be able to find ffmpeg) by simply typing `youtube-dl` or `ffmpeg`, no matter what directory you're in. +### How do I put downloads into a specific folder? + +Use the `-o` to specify an [output template](#output-template), for example `-o "/home/user/videos/%(title)s-%(id)s.%(ext)s"`. If you want this for all of your downloads, put the option into your [configuration file](#configuration). + ### How can I detect whether a given URL is supported by youtube-dl? For one, have a look at the [list of supported sites](docs/supportedsites). Note that it can sometimes happen that the site changes its URL scheme (say, from http://example.com/v/1234567 to http://example.com/v/1234567 ) and youtube-dl reports an URL of a service in that list as unsupported. In that case, simply report a bug. diff --git a/test/test_YoutubeDL.py b/test/test_YoutubeDL.py index 85d87f2c3..678b9f7d1 100644 --- a/test/test_YoutubeDL.py +++ b/test/test_YoutubeDL.py @@ -281,6 +281,61 @@ class TestFormatSelection(unittest.TestCase): downloaded = ydl.downloaded_info_dicts[0] self.assertEqual(downloaded['format_id'], f1id) + def test_format_filtering(self): + formats = [ + {'format_id': 'A', 'filesize': 500, 'width': 1000}, + {'format_id': 'B', 'filesize': 1000, 'width': 500}, + {'format_id': 'C', 'filesize': 1000, 'width': 400}, + {'format_id': 'D', 'filesize': 2000, 'width': 600}, + {'format_id': 'E', 'filesize': 3000}, + {'format_id': 'F'}, + {'format_id': 'G', 'filesize': 1000000}, + ] + for f in formats: + f['url'] = 'http://_/' + f['ext'] = 'unknown' + info_dict = _make_result(formats) + + ydl = YDL({'format': 'best[filesize<3000]'}) + ydl.process_ie_result(info_dict) + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['format_id'], 'D') + + ydl = YDL({'format': 'best[filesize<=3000]'}) + ydl.process_ie_result(info_dict) + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['format_id'], 'E') + + ydl = YDL({'format': 'best[filesize <= ? 3000]'}) + ydl.process_ie_result(info_dict) + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['format_id'], 'F') + + ydl = YDL({'format': 'best [filesize = 1000] [width>450]'}) + ydl.process_ie_result(info_dict) + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['format_id'], 'B') + + ydl = YDL({'format': 'best [filesize = 1000] [width!=450]'}) + ydl.process_ie_result(info_dict) + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['format_id'], 'C') + + ydl = YDL({'format': '[filesize>?1]'}) + ydl.process_ie_result(info_dict) + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['format_id'], 'G') + + ydl = YDL({'format': '[filesize<1M]'}) + ydl.process_ie_result(info_dict) + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['format_id'], 'E') + + ydl = YDL({'format': '[filesize<1MiB]'}) + ydl.process_ie_result(info_dict) + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['format_id'], 'G') + def test_add_extra_info(self): test_dict = { 'extractor': 'Foo', diff --git a/test/test_all_urls.py b/test/test_all_urls.py index bd4fe17bf..e66264b4b 100644 --- a/test/test_all_urls.py +++ b/test/test_all_urls.py @@ -14,7 +14,6 @@ from test.helper import gettestcases from youtube_dl.extractor import ( FacebookIE, gen_extractors, - TwitchIE, YoutubeIE, ) @@ -72,18 +71,6 @@ class TestAllURLsMatching(unittest.TestCase): self.assertMatch('http://www.youtube.com/results?search_query=making+mustard', ['youtube:search_url']) self.assertMatch('https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video', ['youtube:search_url']) - def test_twitch_channelid_matching(self): - self.assertTrue(TwitchIE.suitable('twitch.tv/vanillatv')) - self.assertTrue(TwitchIE.suitable('www.twitch.tv/vanillatv')) - self.assertTrue(TwitchIE.suitable('http://www.twitch.tv/vanillatv')) - self.assertTrue(TwitchIE.suitable('http://www.twitch.tv/vanillatv/')) - - def test_twitch_videoid_matching(self): - self.assertTrue(TwitchIE.suitable('http://www.twitch.tv/vanillatv/b/328087483')) - - def test_twitch_chapterid_matching(self): - self.assertTrue(TwitchIE.suitable('http://www.twitch.tv/tsm_theoddone/c/2349361')) - def test_youtube_extract(self): assertExtractId = lambda url, id: self.assertEqual(YoutubeIE.extract_id(url), id) assertExtractId('http://www.youtube.com/watch?&v=BaW_jenozKc', 'BaW_jenozKc') @@ -115,8 +102,6 @@ class TestAllURLsMatching(unittest.TestCase): self.assertMatch(':ythistory', ['youtube:history']) self.assertMatch(':thedailyshow', ['ComedyCentralShows']) self.assertMatch(':tds', ['ComedyCentralShows']) - self.assertMatch(':colbertreport', ['ComedyCentralShows']) - self.assertMatch(':cr', ['ComedyCentralShows']) def test_vimeo_matching(self): self.assertMatch('http://vimeo.com/channels/tributes', ['vimeo:channel']) diff --git a/test/test_utils.py b/test/test_utils.py index 399cdfdbd..bdd7f268a 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -28,6 +28,7 @@ from youtube_dl.utils import ( fix_xml_ampersands, InAdvancePagedList, intlist_to_bytes, + is_html, js_to_json, limit_length, OnDemandPagedList, @@ -148,6 +149,7 @@ class TestUtil(unittest.TestCase): self.assertEqual(unified_strdate('8/7/2009'), '20090708') self.assertEqual(unified_strdate('Dec 14, 2012'), '20121214') self.assertEqual(unified_strdate('2012/10/11 01:56:38 +0000'), '20121011') + self.assertEqual(unified_strdate('1968 12 10'), '19681210') self.assertEqual(unified_strdate('1968-12-10'), '19681210') self.assertEqual(unified_strdate('28/01/2014 21:00:00 +0100'), '20140128') self.assertEqual( @@ -416,5 +418,21 @@ ffmpeg version 2.4.4 Copyright (c) 2000-2014 the FFmpeg ...'''), '2.4.4') self.assertTrue(age_restricted(18, 14)) self.assertFalse(age_restricted(18, 18)) + def test_is_html(self): + self.assertFalse(is_html(b'\x49\x44\x43<html')) + self.assertTrue(is_html(b'<!DOCTYPE foo>\xaaa')) + self.assertTrue(is_html( # UTF-8 with BOM + b'\xef\xbb\xbf<!DOCTYPE foo>\xaaa')) + self.assertTrue(is_html( # UTF-16-LE + b'\xff\xfe<\x00h\x00t\x00m\x00l\x00>\x00\xe4\x00' + )) + self.assertTrue(is_html( # UTF-16-BE + b'\xfe\xff\x00<\x00h\x00t\x00m\x00l\x00>\x00\xe4' + )) + self.assertTrue(is_html( # UTF-32-BE + b'\x00\x00\xFE\xFF\x00\x00\x00<\x00\x00\x00h\x00\x00\x00t\x00\x00\x00m\x00\x00\x00l\x00\x00\x00>\x00\x00\x00\xe4')) + self.assertTrue(is_html( # UTF-32-LE + b'\xFF\xFE\x00\x00<\x00\x00\x00h\x00\x00\x00t\x00\x00\x00m\x00\x00\x00l\x00\x00\x00>\x00\x00\x00\xe4\x00\x00\x00')) + if __name__ == '__main__': unittest.main() diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 772fddd45..8ef74e414 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -10,6 +10,7 @@ import io import itertools import json import locale +import operator import os import platform import re @@ -49,6 +50,7 @@ from .utils import ( make_HTTPS_handler, MaxDownloadsReached, PagedList, + parse_filesize, PostProcessingError, platform_name, preferredencoding, @@ -768,7 +770,59 @@ class YoutubeDL(object): else: raise Exception('Invalid result type: %s' % result_type) + def _apply_format_filter(self, format_spec, available_formats): + " Returns a tuple of the remaining format_spec and filtered formats " + + OPERATORS = { + '<': operator.lt, + '<=': operator.le, + '>': operator.gt, + '>=': operator.ge, + '=': operator.eq, + '!=': operator.ne, + } + operator_rex = re.compile(r'''(?x)\s*\[ + (?P<key>width|height|tbr|abr|vbr|filesize) + \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* + (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?) + \]$ + ''' % '|'.join(map(re.escape, OPERATORS.keys()))) + m = operator_rex.search(format_spec) + if not m: + raise ValueError('Invalid format specification %r' % format_spec) + + try: + comparison_value = int(m.group('value')) + except ValueError: + comparison_value = parse_filesize(m.group('value')) + if comparison_value is None: + comparison_value = parse_filesize(m.group('value') + 'B') + if comparison_value is None: + raise ValueError( + 'Invalid value %r in format specification %r' % ( + m.group('value'), format_spec)) + op = OPERATORS[m.group('op')] + + def _filter(f): + actual_value = f.get(m.group('key')) + if actual_value is None: + return m.group('none_inclusive') + return op(actual_value, comparison_value) + new_formats = [f for f in available_formats if _filter(f)] + + new_format_spec = format_spec[:-len(m.group(0))] + if not new_format_spec: + new_format_spec = 'best' + + return (new_format_spec, new_formats) + def select_format(self, format_spec, available_formats): + while format_spec.endswith(']'): + format_spec, available_formats = self._apply_format_filter( + format_spec, available_formats) + if not available_formats: + return None + if format_spec == 'best' or format_spec is None: return available_formats[-1] elif format_spec == 'worst': diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 5a1c2f69b..9ab90ac62 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -1,6 +1,7 @@ from __future__ import unicode_literals from .abc import ABCIE +from .abc7news import Abc7NewsIE from .academicearth import AcademicEarthCourseIE from .addanime import AddAnimeIE from .adobetv import AdobeTVIE @@ -175,6 +176,7 @@ from .goshgay import GoshgayIE from .grooveshark import GroovesharkIE from .groupon import GrouponIE from .hark import HarkIE +from .hearthisat import HearThisAtIE from .heise import HeiseIE from .hellporno import HellPornoIE from .helsinki import HelsinkiIE @@ -226,6 +228,7 @@ from .livestream import ( LivestreamOriginalIE, LivestreamShortenerIE, ) +from .lnkgo import LnkGoIE from .lrt import LRTIE from .lynda import ( LyndaIE, @@ -295,6 +298,7 @@ from .nowness import NownessIE from .nowvideo import NowVideoIE from .npo import ( NPOIE, + NPOLiveIE, TegenlichtVproIE, ) from .nrk import ( @@ -406,6 +410,7 @@ from .stanfordoc import StanfordOpenClassroomIE from .steam import SteamIE from .streamcloud import StreamcloudIE from .streamcz import StreamCZIE +from .streetvoice import StreetVoiceIE from .sunporno import SunPornoIE from .swrmediathek import SWRMediathekIE from .syfy import SyfyIE @@ -427,6 +432,7 @@ from .telemb import TeleMBIE from .teletask import TeleTaskIE from .tenplay import TenPlayIE from .testurl import TestURLIE +from .testtube import TestTubeIE from .tf1 import TF1IE from .theonion import TheOnionIE from .theplatform import ThePlatformIE @@ -455,7 +461,14 @@ from .tvigle import TvigleIE from .tvp import TvpIE, TvpSeriesIE from .tvplay import TVPlayIE from .twentyfourvideo import TwentyFourVideoIE -from .twitch import TwitchIE +from .twitch import ( + TwitchVideoIE, + TwitchChapterIE, + TwitchVodIE, + TwitchProfileIE, + TwitchPastBroadcastsIE, + TwitchStreamIE, +) from .ubu import UbuIE from .udemy import ( UdemyIE, diff --git a/youtube_dl/extractor/abc7news.py b/youtube_dl/extractor/abc7news.py new file mode 100644 index 000000000..c04949c21 --- /dev/null +++ b/youtube_dl/extractor/abc7news.py @@ -0,0 +1,68 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import parse_iso8601 + + +class Abc7NewsIE(InfoExtractor): + _VALID_URL = r'https?://abc7news\.com(?:/[^/]+/(?P<display_id>[^/]+))?/(?P<id>\d+)' + _TESTS = [ + { + 'url': 'http://abc7news.com/entertainment/east-bay-museum-celebrates-vintage-synthesizers/472581/', + 'info_dict': { + 'id': '472581', + 'display_id': 'east-bay-museum-celebrates-vintage-synthesizers', + 'ext': 'mp4', + 'title': 'East Bay museum celebrates history of synthesized music', + 'description': 'md5:a4f10fb2f2a02565c1749d4adbab4b10', + 'thumbnail': 're:^https?://.*\.jpg$', + 'timestamp': 1421123075, + 'upload_date': '20150113', + 'uploader': 'Jonathan Bloom', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, + { + 'url': 'http://abc7news.com/472581', + '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') or video_id + + webpage = self._download_webpage(url, display_id) + + m3u8 = self._html_search_meta( + 'contentURL', webpage, 'm3u8 url', fatal=True) + + formats = self._extract_m3u8_formats(m3u8, display_id, 'mp4') + self._sort_formats(formats) + + title = self._og_search_title(webpage).strip() + description = self._og_search_description(webpage).strip() + thumbnail = self._og_search_thumbnail(webpage) + timestamp = parse_iso8601(self._search_regex( + r'<div class="meta">\s*<time class="timeago" datetime="([^"]+)">', + webpage, 'upload date', fatal=False)) + uploader = self._search_regex( + r'rel="author">([^<]+)</a>', + webpage, 'uploader', default=None) + + return { + 'id': video_id, + 'display_id': display_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'timestamp': timestamp, + 'uploader': uploader, + 'formats': formats, + } diff --git a/youtube_dl/extractor/atresplayer.py b/youtube_dl/extractor/atresplayer.py index 72e83bfc2..5db1941b3 100644 --- a/youtube_dl/extractor/atresplayer.py +++ b/youtube_dl/extractor/atresplayer.py @@ -4,9 +4,12 @@ import time import hmac from .common import InfoExtractor -from ..utils import ( +from ..compat import ( compat_str, + compat_urllib_parse, compat_urllib_request, +) +from ..utils import ( int_or_none, float_or_none, xpath_text, @@ -44,6 +47,33 @@ class AtresPlayerIE(InfoExtractor): _PLAYER_URL_TEMPLATE = 'https://servicios.atresplayer.com/episode/getplayer.json?episodePk=%s' _EPISODE_URL_TEMPLATE = 'http://www.atresplayer.com/episodexml/%s' + _LOGIN_URL = 'https://servicios.atresplayer.com/j_spring_security_check' + + def _real_initialize(self): + self._login() + + def _login(self): + (username, password) = self._get_login_info() + if username is None: + return + + login_form = { + 'j_username': username, + 'j_password': password, + } + + request = compat_urllib_request.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( + request, None, 'Logging in as %s' % username) + + error = self._html_search_regex( + r'(?s)<ul class="list_error">(.+?)</ul>', response, 'error', default=None) + if error: + raise ExtractorError( + 'Unable to login: %s' % error, expected=True) + def _real_extract(self, url): video_id = self._match_id(url) diff --git a/youtube_dl/extractor/bandcamp.py b/youtube_dl/extractor/bandcamp.py index b45d68a61..aea0263d6 100644 --- a/youtube_dl/extractor/bandcamp.py +++ b/youtube_dl/extractor/bandcamp.py @@ -161,7 +161,8 @@ class BandcampAlbumIE(InfoExtractor): entries = [ self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key()) for t_path in tracks_paths] - title = self._search_regex(r'album_title : "(.*?)"', webpage, 'title') + title = self._search_regex( + r'album_title\s*:\s*"(.*?)"', webpage, 'title', fatal=False) return { '_type': 'playlist', 'id': playlist_id, diff --git a/youtube_dl/extractor/cnn.py b/youtube_dl/extractor/cnn.py index 93e8d0de3..90ea07438 100644 --- a/youtube_dl/extractor/cnn.py +++ b/youtube_dl/extractor/cnn.py @@ -51,7 +51,7 @@ class CNNIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) path = mobj.group('path') page_title = mobj.group('title') - info_url = 'http://cnn.com/video/data/3.0/%s/index.xml' % path + info_url = 'http://edition.cnn.com/video/data/3.0/%s/index.xml' % path info = self._download_xml(info_url, page_title) formats = [] @@ -143,13 +143,13 @@ class CNNArticleIE(InfoExtractor): _VALID_URL = r'https?://(?:(?:edition|www)\.)?cnn\.com/(?!video/)' _TEST = { 'url': 'http://www.cnn.com/2014/12/21/politics/obama-north-koreas-hack-not-war-but-cyber-vandalism/', - 'md5': '275b326f85d80dff7592a9820f5dc887', + 'md5': '689034c2a3d9c6dc4aa72d65a81efd01', 'info_dict': { - 'id': 'bestoftv/2014/12/21/sotu-crowley-president-obama-north-korea-not-going-to-be-intimidated.cnn', + 'id': 'bestoftv/2014/12/21/ip-north-korea-obama.cnn', 'ext': 'mp4', - 'title': 'Obama: We\'re not going to be intimidated', - 'description': 'md5:e735586f3dc936075fa654a4d91b21f9', - 'upload_date': '20141220', + 'title': 'Obama: Cyberattack not an act of war', + 'description': 'md5:51ce6750450603795cad0cdfbd7d05c5', + 'upload_date': '20141221', }, 'add_ie': ['CNN'], } diff --git a/youtube_dl/extractor/comedycentral.py b/youtube_dl/extractor/comedycentral.py index 8d27af5e5..15ca361f0 100644 --- a/youtube_dl/extractor/comedycentral.py +++ b/youtube_dl/extractor/comedycentral.py @@ -34,12 +34,12 @@ class ComedyCentralIE(MTVServicesInfoExtractor): class ComedyCentralShowsIE(MTVServicesInfoExtractor): IE_DESC = 'The Daily Show / The Colbert Report' - # urls can be abbreviations like :thedailyshow or :colbert + # urls can be abbreviations like :thedailyshow # urls for episodes like: # or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day # or: http://www.colbertnation.com/the-colbert-report-videos/421667/november-29-2012/moon-shattering-news # or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524 - _VALID_URL = r'''(?x)^(:(?P<shortname>tds|thedailyshow|cr|colbert|colbertnation|colbertreport) + _VALID_URL = r'''(?x)^(:(?P<shortname>tds|thedailyshow) |https?://(:www\.)? (?P<showname>thedailyshow|thecolbertreport)\.(?:cc\.)?com/ ((?:full-)?episodes/(?:[0-9a-z]{6}/)?(?P<episode>.*)| diff --git a/youtube_dl/extractor/fourtube.py b/youtube_dl/extractor/fourtube.py index 7187e0752..b2284ab01 100644 --- a/youtube_dl/extractor/fourtube.py +++ b/youtube_dl/extractor/fourtube.py @@ -7,10 +7,9 @@ from ..compat import ( compat_urllib_request, ) from ..utils import ( - clean_html, parse_duration, + parse_iso8601, str_to_int, - unified_strdate, ) @@ -28,68 +27,81 @@ class FourTubeIE(InfoExtractor): 'uploader': 'WCP Club', 'uploader_id': 'wcp-club', 'upload_date': '20131031', + 'timestamp': 1383263892, 'duration': 583, + 'view_count': int, + 'like_count': int, + 'categories': list, } } def _real_extract(self, url): video_id = self._match_id(url) - webpage_url = 'http://www.4tube.com/videos/' + video_id - webpage = self._download_webpage(webpage_url, video_id) + webpage = self._download_webpage(url, video_id) - self.report_extraction(video_id) + title = self._html_search_meta('name', webpage) + timestamp = parse_iso8601(self._html_search_meta( + 'uploadDate', webpage)) + thumbnail = self._html_search_meta('thumbnailUrl', webpage) + uploader_id = self._html_search_regex( + r'<a class="img-avatar" href="[^"]+/channels/([^/"]+)" title="Go to [^"]+ page">', + webpage, 'uploader id') + uploader = self._html_search_regex( + r'<a class="img-avatar" href="[^"]+/channels/[^/"]+" title="Go to ([^"]+) page">', + webpage, 'uploader') - playlist_json = self._html_search_regex(r'var playerConfigPlaylist\s+=\s+([^;]+)', webpage, 'Playlist') - media_id = self._search_regex(r'idMedia:\s*(\d+)', playlist_json, 'Media Id') - sources = self._search_regex(r'sources:\s*\[([^\]]*)\]', playlist_json, 'Sources').split(',') - title = self._search_regex(r'title:\s*"([^"]*)', playlist_json, 'Title') - thumbnail_url = self._search_regex(r'image:\s*"([^"]*)', playlist_json, 'Thumbnail', fatal=False) + categories_html = self._search_regex( + r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="list">(.*?)</ul>', + webpage, 'categories', fatal=False) + categories = None + if categories_html: + categories = [ + c.strip() for c in re.findall( + r'(?s)<li><a.*?>(.*?)</a>', categories_html)] - uploader_str = self._search_regex(r'<span>Uploaded by</span>(.*?)<span>', webpage, 'uploader', fatal=False) - mobj = re.search(r'<a href="/sites/(?P<id>[^"]+)"><strong>(?P<name>[^<]+)</strong></a>', uploader_str) - (uploader, uploader_id) = (mobj.group('name'), mobj.group('id')) if mobj else (clean_html(uploader_str), None) + view_count = str_to_int(self._search_regex( + r'<meta itemprop="interactionCount" content="UserPlays:([0-9,]+)">', + webpage, 'view count', fatal=False)) + like_count = str_to_int(self._search_regex( + r'<meta itemprop="interactionCount" content="UserLikes:([0-9,]+)">', + webpage, 'like count', fatal=False)) + duration = parse_duration(self._html_search_meta('duration', webpage)) - upload_date = None - view_count = None - duration = None - description = self._html_search_meta('description', webpage, 'description') - if description: - upload_date = self._search_regex(r'Published Date: (\d{2} [a-zA-Z]{3} \d{4})', description, 'upload date', - fatal=False) - if upload_date: - upload_date = unified_strdate(upload_date) - view_count = self._search_regex(r'Views: ([\d,\.]+)', description, 'view count', fatal=False) - if view_count: - view_count = str_to_int(view_count) - duration = parse_duration(self._search_regex(r'Length: (\d+m\d+s)', description, 'duration', fatal=False)) + params_js = self._search_regex( + r'\$\.ajax\(url,\ opts\);\s*\}\s*\}\)\(([0-9,\[\] ]+)\)', + webpage, 'initialization parameters' + ) + params = self._parse_json('[%s]' % params_js, video_id) + media_id = params[0] + sources = ['%s' % p for p in params[2]] - token_url = "http://tkn.4tube.com/{0}/desktop/{1}".format(media_id, "+".join(sources)) + token_url = 'http://tkn.4tube.com/{0}/desktop/{1}'.format( + media_id, '+'.join(sources)) headers = { 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) tokens = self._download_json(token_req, video_id) - formats = [{ 'url': tokens[format]['token'], 'format_id': format + 'p', 'resolution': format + 'p', 'quality': int(format), } for format in sources] - self._sort_formats(formats) return { 'id': video_id, 'title': title, 'formats': formats, - 'thumbnail': thumbnail_url, + 'categories': categories, + 'thumbnail': thumbnail, 'uploader': uploader, 'uploader_id': uploader_id, - 'upload_date': upload_date, + 'timestamp': timestamp, + 'like_count': like_count, 'view_count': view_count, 'duration': duration, 'age_limit': 18, - 'webpage_url': webpage_url, } diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 7a5bf9392..b893d8149 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -17,6 +17,7 @@ from ..utils import ( ExtractorError, float_or_none, HEADRequest, + is_html, orderedSet, parse_xml, smuggle_url, @@ -647,7 +648,7 @@ class GenericIE(InfoExtractor): # Maybe it's a direct link to a video? # Be careful not to download the whole thing! first_bytes = full_response.read(512) - if not re.match(r'^\s*<', first_bytes.decode('utf-8', 'replace')): + if not is_html(first_bytes): self._downloader.report_warning( 'URL could be a direct video link, returning it as such.') upload_date = unified_strdate( diff --git a/youtube_dl/extractor/hearthisat.py b/youtube_dl/extractor/hearthisat.py new file mode 100644 index 000000000..a19b31ac0 --- /dev/null +++ b/youtube_dl/extractor/hearthisat.py @@ -0,0 +1,117 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..compat import ( + compat_urllib_request, + compat_urlparse, +) +from ..utils import ( + HEADRequest, + str_to_int, + urlencode_postdata, + urlhandle_detect_ext, +) + + +class HearThisAtIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?hearthis\.at/(?P<artist>[^/]+)/(?P<title>[A-Za-z0-9\-]+)/?$' + _PLAYLIST_URL = 'https://hearthis.at/playlist.php' + _TEST = { + 'url': 'https://hearthis.at/moofi/dr-kreep', + 'md5': 'ab6ec33c8fed6556029337c7885eb4e0', + 'info_dict': { + 'id': '150939', + 'ext': 'wav', + 'title': 'Moofi - Dr. Kreep', + 'thumbnail': 're:^https?://.*\.jpg$', + 'timestamp': 1421564134, + 'description': 'Creepy Patch. Mutable Instruments Braids Vowel + Formant Mode.', + 'upload_date': '20150118', + 'comment_count': int, + 'view_count': int, + 'like_count': int, + 'duration': 71, + 'categories': ['Experimental'], + } + } + + def _real_extract(self, url): + m = re.match(self._VALID_URL, url) + display_id = '{artist:s} - {title:s}'.format(**m.groupdict()) + + webpage = self._download_webpage(url, display_id) + track_id = self._search_regex( + r'intTrackId\s*=\s*(\d+)', webpage, 'track ID') + + payload = urlencode_postdata({'tracks[]': track_id}) + req = compat_urllib_request.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] + title = '{artist:s} - {title:s}'.format(**track) + + categories = None + if track.get('category'): + categories = [track['category']] + + description = self._og_search_description(webpage) + thumbnail = self._og_search_thumbnail(webpage) + + meta_span = r'<span[^>]+class="%s".*?</i>([^<]+)</span>' + view_count = str_to_int(self._search_regex( + meta_span % 'plays_count', webpage, 'view count', fatal=False)) + like_count = str_to_int(self._search_regex( + meta_span % 'likes_count', webpage, 'like count', fatal=False)) + comment_count = str_to_int(self._search_regex( + meta_span % 'comment_count', webpage, 'comment count', fatal=False)) + duration = str_to_int(self._search_regex( + r'data-length="(\d+)', webpage, 'duration', fatal=False)) + timestamp = str_to_int(self._search_regex( + r'<span[^>]+class="calctime"[^>]+data-time="(\d+)', webpage, 'timestamp', fatal=False)) + + formats = [] + mp3_url = self._search_regex( + r'(?s)<a class="player-link"\s+(?:[a-zA-Z0-9_:-]+="[^"]+"\s+)*?data-mp3="([^"]+)"', + webpage, 'mp3 URL', fatal=False) + if mp3_url: + formats.append({ + 'format_id': 'mp3', + 'vcodec': 'none', + 'acodec': 'mp3', + 'url': mp3_url, + }) + download_path = self._search_regex( + r'<a class="[^"]*download_fct[^"]*"\s+href="([^"]+)"', + webpage, 'download URL', default=None) + if download_path: + download_url = compat_urlparse.urljoin(url, download_path) + ext_req = HEADRequest(download_url) + ext_handle = self._request_webpage( + ext_req, display_id, note='Determining extension') + ext = urlhandle_detect_ext(ext_handle) + formats.append({ + 'format_id': 'download', + 'vcodec': 'none', + 'ext': ext, + 'url': download_url, + 'preference': 2, # Usually better quality + }) + self._sort_formats(formats) + + return { + 'id': track_id, + 'display_id': display_id, + 'title': title, + 'formats': formats, + 'thumbnail': thumbnail, + 'description': description, + 'duration': duration, + 'timestamp': timestamp, + 'view_count': view_count, + 'comment_count': comment_count, + 'like_count': like_count, + 'categories': categories, + } diff --git a/youtube_dl/extractor/lnkgo.py b/youtube_dl/extractor/lnkgo.py new file mode 100644 index 000000000..a8e357859 --- /dev/null +++ b/youtube_dl/extractor/lnkgo.py @@ -0,0 +1,124 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + int_or_none, + js_to_json, + unified_strdate, +) + + +class LnkGoIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?lnkgo\.alfa\.lt/visi\-video/(?P<show>[^/]+)/ziurek\-(?P<display_id>[A-Za-z0-9\-]+)' + _TESTS = [{ + 'url': 'http://lnkgo.alfa.lt/visi-video/yra-kaip-yra/ziurek-yra-kaip-yra-162', + 'info_dict': { + 'id': '46712', + 'ext': 'mp4', + 'title': 'Yra kaip yra', + 'upload_date': '20150107', + 'description': 'md5:d82a5e36b775b7048617f263a0e3475e', + 'age_limit': 7, + 'duration': 3019, + 'thumbnail': 're:^https?://.*\.jpg$' + }, + 'params': { + 'skip_download': True, # HLS download + }, + }, { + 'url': 'http://lnkgo.alfa.lt/visi-video/aktualai-pratesimas/ziurek-nerdas-taiso-kompiuteri-2', + 'info_dict': { + 'id': '47289', + 'ext': 'mp4', + 'title': 'Nėrdas: Kompiuterio Valymas', + 'upload_date': '20150113', + 'description': 'md5:7352d113a242a808676ff17e69db6a69', + 'age_limit': 18, + 'duration': 346, + 'thumbnail': 're:^https?://.*\.jpg$' + }, + 'params': { + 'skip_download': True, # HLS download + }, + }] + _AGE_LIMITS = { + 'N-7': 7, + 'N-14': 14, + 'S': 18, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + display_id = mobj.group('display_id') + + webpage = self._download_webpage( + url, display_id, 'Downloading player webpage') + + video_id = self._search_regex( + r'data-ep="([^"]+)"', webpage, 'video ID') + title = self._og_search_title(webpage) + description = self._og_search_description(webpage) + + thumbnail_w = int_or_none( + self._og_search_property('image:width', webpage, 'thumbnail width', fatal=False)) + thumbnail_h = int_or_none( + self._og_search_property('image:height', webpage, 'thumbnail height', fatal=False)) + thumbnail = { + 'url': self._og_search_thumbnail(webpage), + } + if thumbnail_w and thumbnail_h: + thumbnail.update({ + 'width': thumbnail_w, + 'height': thumbnail_h, + }) + + upload_date = unified_strdate(self._search_regex( + r'class="meta-item\sair-time">.*?<strong>([^<]+)</strong>', webpage, 'upload date', fatal=False)) + duration = int_or_none(self._search_regex( + r'VideoDuration = "([^"]+)"', webpage, 'duration', fatal=False)) + + pg_rating = self._search_regex( + r'pgrating="([^"]+)"', webpage, 'PG rating', fatal=False, default='') + age_limit = self._AGE_LIMITS.get(pg_rating.upper(), 0) + + sources_js = self._search_regex( + r'(?s)sources:\s(\[.*?\]),', webpage, 'sources') + sources = self._parse_json( + sources_js, video_id, transform_source=js_to_json) + + formats = [] + for source in sources: + if source.get('provider') == 'rtmp': + m = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<play_path>.+)$', source['file']) + if not m: + continue + formats.append({ + 'format_id': 'rtmp', + 'ext': 'flv', + 'url': m.group('url'), + 'play_path': m.group('play_path'), + 'page_url': url, + }) + elif source.get('file').endswith('.m3u8'): + formats.append({ + 'format_id': 'hls', + 'ext': source.get('type', 'mp4'), + 'url': source['file'], + }) + + self._sort_formats(formats) + + return { + 'id': video_id, + 'display_id': display_id, + 'title': title, + 'formats': formats, + 'thumbnails': [thumbnail], + 'duration': duration, + 'description': description, + 'age_limit': age_limit, + 'upload_date': upload_date, + } diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py index 690c46b6a..f840f6532 100644 --- a/youtube_dl/extractor/nbc.py +++ b/youtube_dl/extractor/nbc.py @@ -6,6 +6,7 @@ import json from .common import InfoExtractor from ..compat import ( compat_str, + compat_HTTPError, ) from ..utils import ( ExtractorError, @@ -78,6 +79,16 @@ class NBCNewsIE(InfoExtractor): }, 'add_ie': ['ThePlatform'], }, + { + 'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156', + 'md5': 'fdbf39ab73a72df5896b6234ff98518a', + 'info_dict': { + 'id': 'Wjf9EDR3A_60', + 'ext': 'mp4', + 'title': 'FULL EPISODE: Family Business', + 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04', + }, + }, ] def _real_extract(self, url): @@ -115,10 +126,19 @@ class NBCNewsIE(InfoExtractor): if not base_url: continue playlist_url = base_url + '?form=MPXNBCNewsAPI' - all_videos = self._download_json(playlist_url, title)['videos'] try: - info = next(v for v in all_videos if v['mpxId'] == mpxid) + all_videos = self._download_json(playlist_url, title) + except ExtractorError as ee: + if isinstance(ee.cause, compat_HTTPError): + continue + raise + + if not all_videos or 'videos' not in all_videos: + continue + + try: + info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid) break except StopIteration: continue diff --git a/youtube_dl/extractor/ndtv.py b/youtube_dl/extractor/ndtv.py index 95e7d63aa..2a1ca80df 100644 --- a/youtube_dl/extractor/ndtv.py +++ b/youtube_dl/extractor/ndtv.py @@ -27,9 +27,7 @@ class NDTVIE(InfoExtractor): } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - + video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) filename = self._search_regex( diff --git a/youtube_dl/extractor/npo.py b/youtube_dl/extractor/npo.py index 6d20f755a..175b14583 100644 --- a/youtube_dl/extractor/npo.py +++ b/youtube_dl/extractor/npo.py @@ -1,7 +1,5 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..utils import ( fix_xml_ampersands, @@ -13,7 +11,16 @@ from ..utils import ( ) -class NPOIE(InfoExtractor): +class NPOBaseIE(InfoExtractor): + def _get_token(self, video_id): + token_page = self._download_webpage( + 'http://ida.omroep.nl/npoplayer/i.js', + video_id, note='Downloading token') + return self._search_regex( + r'npoplayer\.token = "(.+?)"', token_page, 'token') + + +class NPOIE(NPOBaseIE): IE_NAME = 'npo.nl' _VALID_URL = r'https?://www\.npo\.nl/[^/]+/[^/]+/(?P<id>[^/?]+)' @@ -80,8 +87,7 @@ class NPOIE(InfoExtractor): ] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = self._match_id(url) return self._get_info(video_id) def _get_info(self, video_id): @@ -91,12 +97,8 @@ class NPOIE(InfoExtractor): # We have to remove the javascript callback transform_source=strip_jsonp, ) - token_page = self._download_webpage( - 'http://ida.omroep.nl/npoplayer/i.js', - video_id, - note='Downloading token' - ) - token = self._search_regex(r'npoplayer\.token = "(.+?)"', token_page, 'token') + + token = self._get_token(video_id) formats = [] @@ -170,6 +172,83 @@ class NPOIE(InfoExtractor): } +class NPOLiveIE(NPOBaseIE): + IE_NAME = 'npo.nl:live' + _VALID_URL = r'https?://www\.npo\.nl/live/(?P<id>.+)' + + _TEST = { + 'url': 'http://www.npo.nl/live/npo-1', + 'info_dict': { + 'id': 'LI_NEDERLAND1_136692', + 'display_id': 'npo-1', + 'ext': 'mp4', + 'title': 're:^Nederland 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'description': 'Livestream', + 'is_live': True, + }, + 'params': { + 'skip_download': True, + } + } + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + live_id = self._search_regex( + r'data-prid="([^"]+)"', webpage, 'live id') + + metadata = self._download_json( + 'http://e.omroep.nl/metadata/%s' % live_id, + display_id, transform_source=strip_jsonp) + + token = self._get_token(display_id) + + formats = [] + + streams = metadata.get('streams') + if streams: + for stream in streams: + stream_type = stream.get('type').lower() + if stream_type == 'ss': + continue + stream_info = self._download_json( + 'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp' + % (stream.get('url'), token), + display_id, 'Downloading %s JSON' % stream_type) + if stream_info.get('error_code', 0) or stream_info.get('errorcode', 0): + continue + stream_url = self._download_json( + stream_info['stream'], display_id, + 'Downloading %s URL' % stream_type, + transform_source=strip_jsonp) + if stream_type == 'hds': + f4m_formats = self._extract_f4m_formats(stream_url, display_id) + # f4m downloader downloads only piece of live stream + for f4m_format in f4m_formats: + f4m_format['preference'] = -1 + formats.extend(f4m_formats) + elif stream_type == 'hls': + formats.extend(self._extract_m3u8_formats(stream_url, display_id, 'mp4')) + else: + formats.append({ + 'url': stream_url, + }) + + self._sort_formats(formats) + + return { + 'id': live_id, + 'display_id': display_id, + 'title': self._live_title(metadata['titel']), + 'description': metadata['info'], + 'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'], + 'formats': formats, + 'is_live': True, + } + + class TegenlichtVproIE(NPOIE): IE_NAME = 'tegenlicht.vpro.nl' _VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?' diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 634142d0d..fb2032832 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -10,6 +10,7 @@ from ..compat import ( compat_urllib_request, ) from ..utils import ( + ExtractorError, str_to_int, ) from ..aes import ( @@ -44,6 +45,15 @@ class PornHubIE(InfoExtractor): req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) + error_msg = self._html_search_regex( + r'(?s)<div class="userMessageSection[^"]*".*?>(.*?)</div>', + webpage, 'error message', default=None) + if error_msg: + error_msg = re.sub(r'\s+', ' ', error_msg) + raise ExtractorError( + 'PornHub said: %s' % error_msg, + expected=True, video_id=video_id) + video_title = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, 'title') video_uploader = self._html_search_regex( r'(?s)From: .+?<(?:a href="/users/|a href="/channels/|<span class="username)[^>]+>(.+?)<', diff --git a/youtube_dl/extractor/smotri.py b/youtube_dl/extractor/smotri.py index baef3daa0..26f361c93 100644 --- a/youtube_dl/extractor/smotri.py +++ b/youtube_dl/extractor/smotri.py @@ -90,6 +90,20 @@ class SmotriIE(InfoExtractor): }, 'skip': 'Video is not approved by moderator', }, + # not approved by moderator, but available + { + 'url': 'http://smotri.com/video/view/?id=v28888533b73', + 'md5': 'f44bc7adac90af518ef1ecf04893bb34', + 'info_dict': { + 'id': 'v28888533b73', + 'ext': 'mp4', + 'title': 'Russian Spies Killed By ISIL Child Soldier', + 'uploader': 'Mopeder', + 'uploader_id': 'mopeder', + 'duration': 71, + 'thumbnail': 'http://frame9.loadup.ru/d7/32/2888853.2.3.jpg', + }, + }, # swf player { 'url': 'http://pics.smotri.com/scrubber_custom8.swf?file=v9188090500', @@ -146,13 +160,16 @@ class SmotriIE(InfoExtractor): video = self._download_json(request, video_id, 'Downloading video JSON') - if video.get('_moderate_no') or not video.get('moderated'): - raise ExtractorError('Video %s has not been approved by moderator' % video_id, expected=True) + video_url = video.get('_vidURL') or video.get('_vidURL_mp4') + + if not video_url: + if video.get('_moderate_no') or not video.get('moderated'): + raise ExtractorError( + 'Video %s has not been approved by moderator' % video_id, expected=True) - if video.get('error'): - raise ExtractorError('Video %s does not exist' % video_id, expected=True) + if video.get('error'): + raise ExtractorError('Video %s does not exist' % video_id, expected=True) - video_url = video.get('_vidURL') or video.get('_vidURL_mp4') title = video['title'] thumbnail = video['_imgURL'] upload_date = unified_strdate(video['added']) diff --git a/youtube_dl/extractor/spiegel.py b/youtube_dl/extractor/spiegel.py index 1e55a9ffb..f345883c7 100644 --- a/youtube_dl/extractor/spiegel.py +++ b/youtube_dl/extractor/spiegel.py @@ -4,7 +4,14 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_urlparse +from ..compat import ( + compat_urlparse, + compat_HTTPError, +) +from ..utils import ( + HEADRequest, + ExtractorError, +) from .spiegeltv import SpiegeltvIE @@ -60,21 +67,31 @@ class SpiegelIE(InfoExtractor): xml_url = base_url + video_id + '.xml' idoc = self._download_xml(xml_url, video_id) - formats = [ - { - 'format_id': n.tag.rpartition('type')[2], - 'url': base_url + n.find('./filename').text, - 'width': int(n.find('./width').text), - 'height': int(n.find('./height').text), - 'abr': int(n.find('./audiobitrate').text), - 'vbr': int(n.find('./videobitrate').text), - 'vcodec': n.find('./codec').text, - 'acodec': 'MP4A', - } - for n in list(idoc) - # Blacklist type 6, it's extremely LQ and not available on the same server - if n.tag.startswith('type') and n.tag != 'type6' - ] + formats = [] + for n in list(idoc): + if n.tag.startswith('type') and n.tag != 'type6': + format_id = n.tag.rpartition('type')[2] + video_url = base_url + n.find('./filename').text + # Test video URLs beforehand as some of them are invalid + try: + self._request_webpage( + HEADRequest(video_url), video_id, + 'Checking %s video URL' % format_id) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: + self.report_warning( + '%s video URL is invalid, skipping' % format_id, video_id) + continue + formats.append({ + 'format_id': format_id, + 'url': video_url, + 'width': int(n.find('./width').text), + 'height': int(n.find('./height').text), + 'abr': int(n.find('./audiobitrate').text), + 'vbr': int(n.find('./videobitrate').text), + 'vcodec': n.find('./codec').text, + 'acodec': 'MP4A', + }) duration = float(idoc[0].findall('./duration')[0].text) self._sort_formats(formats) diff --git a/youtube_dl/extractor/streetvoice.py b/youtube_dl/extractor/streetvoice.py new file mode 100644 index 000000000..6a57fa60a --- /dev/null +++ b/youtube_dl/extractor/streetvoice.py @@ -0,0 +1,51 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_str +from ..utils import unified_strdate + + +class StreetVoiceIE(InfoExtractor): + _VALID_URL = r'https?://(?:.+?\.)?streetvoice\.com/[^/]+/songs/(?P<id>[0-9]+)' + _TESTS = [{ + 'url': 'http://streetvoice.com/skippylu/songs/94440/', + 'md5': '15974627fc01a29e492c98593c2fd472', + 'info_dict': { + 'id': '94440', + 'ext': 'mp3', + 'filesize': 4167053, + 'title': '輸', + 'description': 'Crispy脆樂團 - 輸', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 260, + 'upload_date': '20091018', + 'uploader': 'Crispy脆樂團', + 'uploader_id': '627810', + } + }, { + 'url': 'http://tw.streetvoice.com/skippylu/songs/94440/', + 'only_matching': True, + }] + + def _real_extract(self, url): + song_id = self._match_id(url) + + song = self._download_json( + 'http://streetvoice.com/music/api/song/%s' % song_id, song_id) + + title = song['name'] + author = song['musician']['name'] + + return { + 'id': song_id, + 'url': song['file'], + 'filesize': song.get('size'), + 'title': title, + 'description': '%s - %s' % (author, title), + 'thumbnail': self._proto_relative_url(song.get('image'), 'http:'), + 'duration': song.get('length'), + 'upload_date': unified_strdate(song.get('created_at')), + 'uploader': author, + 'uploader_id': compat_str(song['musician']['id']), + } diff --git a/youtube_dl/extractor/testtube.py b/youtube_dl/extractor/testtube.py new file mode 100644 index 000000000..fd47e71a2 --- /dev/null +++ b/youtube_dl/extractor/testtube.py @@ -0,0 +1,60 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import int_or_none + + +class TestTubeIE(InfoExtractor): + _VALID_URL = r'https?://testtube\.com/[^/?#]+/(?P<id>[^/?#]+)' + _TESTS = [{ + 'url': 'https://testtube.com/dnews/5-weird-ways-plants-can-eat-animals?utm_source=FB&utm_medium=DNews&utm_campaign=DNewsSocial', + 'info_dict': { + 'id': '60163', + 'display_id': '5-weird-ways-plants-can-eat-animals', + 'duration': 275, + 'ext': 'mp4', + 'title': '5 Weird Ways Plants Can Eat Animals', + 'description': 'Why have some plants evolved to eat meat?', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'DNews', + 'uploader_id': 'dnews', + }, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + video_id = self._search_regex( + r"player\.loadRevision3Item\('video_id',\s*([0-9]+)\);", + webpage, 'video ID') + + all_info = self._download_json( + 'https://testtube.com/api/getPlaylist.json?api_key=ba9c741bce1b9d8e3defcc22193f3651b8867e62&codecs=h264,vp8,theora&video_id=%s' % video_id, + video_id) + info = all_info['items'][0] + + formats = [] + for vcodec, fdatas in info['media'].items(): + for name, fdata in fdatas.items(): + formats.append({ + 'format_id': '%s-%s' % (vcodec, name), + 'url': fdata['url'], + 'vcodec': vcodec, + 'tbr': fdata.get('bitrate'), + }) + self._sort_formats(formats) + + duration = int_or_none(info.get('duration')) + + return { + 'id': video_id, + 'display_id': display_id, + 'title': info['title'], + 'description': info.get('summary'), + 'thumbnail': info.get('images', {}).get('large'), + 'uploader': info.get('show', {}).get('name'), + 'uploader_id': info.get('show', {}).get('slug'), + 'duration': duration, + 'formats': formats, + } diff --git a/youtube_dl/extractor/tinypic.py b/youtube_dl/extractor/tinypic.py index 4fe89dbe5..e036b8cdf 100644 --- a/youtube_dl/extractor/tinypic.py +++ b/youtube_dl/extractor/tinypic.py @@ -9,17 +9,23 @@ from ..utils import ExtractorError class TinyPicIE(InfoExtractor): IE_NAME = 'tinypic' IE_DESC = 'tinypic.com videos' - _VALID_URL = r'http://tinypic\.com/player\.php\?v=(?P<id>[^&]+)&s=\d+' - - _TEST = { - 'url': 'http://tinypic.com/player.php?v=6xw7tc%3E&s=5#.UtqZmbRFCM8', - 'md5': '609b74432465364e72727ebc6203f044', - 'info_dict': { - 'id': '6xw7tc', - 'ext': 'flv', - 'title': 'shadow phenomenon weird', + _VALID_URL = r'http://(?:.+?\.)?tinypic\.com/player\.php\?v=(?P<id>[^&]+)&s=\d+' + + _TESTS = [ + { + 'url': 'http://tinypic.com/player.php?v=6xw7tc%3E&s=5#.UtqZmbRFCM8', + 'md5': '609b74432465364e72727ebc6203f044', + 'info_dict': { + 'id': '6xw7tc', + 'ext': 'flv', + 'title': 'shadow phenomenon weird', + }, + }, + { + 'url': 'http://de.tinypic.com/player.php?v=dy90yh&s=8', + 'only_matching': True, } - } + ] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) diff --git a/youtube_dl/extractor/tvp.py b/youtube_dl/extractor/tvp.py index cc26f417a..f57d609d4 100644 --- a/youtube_dl/extractor/tvp.py +++ b/youtube_dl/extractor/tvp.py @@ -12,61 +12,59 @@ class TvpIE(InfoExtractor): _TESTS = [{ 'url': 'http://vod.tvp.pl/filmy-fabularne/filmy-za-darmo/ogniem-i-mieczem/wideo/odc-2/4278035', + 'md5': 'cdd98303338b8a7f7abab5cd14092bf2', 'info_dict': { 'id': '4278035', 'ext': 'wmv', 'title': 'Ogniem i mieczem, odc. 2', - 'description': 'Bohun dowiaduje się o złamaniu przez kniahinię danego mu słowa i wyrusza do Rozłogów. Helenie w ostatniej chwili udaje się uciec dzięki pomocy Zagłoby.', }, }, { 'url': 'http://vod.tvp.pl/seriale/obyczajowe/czas-honoru/sezon-1-1-13/i-seria-odc-13/194536', + 'md5': '8aa518c15e5cc32dfe8db400dc921fbb', 'info_dict': { 'id': '194536', 'ext': 'mp4', 'title': 'Czas honoru, I seria – odc. 13', - # 'description': 'WŁADEK\nCzesław prosi Marię o dostarczenie Władkowi zarazki tyfusu. Jeśli zachoruje zostanie przewieziony do szpitala skąd łatwiej będzie go odbić. Czy matka zdecyduje się zarazić syna? Karol odwiedza Wandę przyznaje się, że ją oszukiwał, ale ostrzega też, że grozi jej aresztowanie i nalega, żeby wyjechała z Warszawy. Czy dziewczyna zdecyduje się znów oddalić od ukochanego? Rozpoczyna się akcja odbicia Władka.', }, }, { 'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176', + 'md5': 'c3b15ed1af288131115ff17a17c19dda', 'info_dict': { 'id': '17916176', 'ext': 'mp4', 'title': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata', }, - 'params': { - # m3u8 download - 'skip_download': 'true', - }, }, { 'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272', + 'md5': 'c3b15ed1af288131115ff17a17c19dda', 'info_dict': { 'id': '17834272', 'ext': 'mp4', 'title': 'Na sygnale, odc. 39', - 'description': 'Ekipa Wiktora ratuje młodą matkę, która spadła ze schodów trzymając na rękach noworodka. Okazuje się, że dziewczyna jest surogatką, a biologiczni rodzice dziecka próbują zmusić ją do oddania synka…', - }, - 'params': { - # m3u8 download - 'skip_download': 'true', }, }] def _real_extract(self, url): video_id = self._match_id(url) + webpage = self._download_webpage( 'http://www.tvp.pl/sess/tvplayer.php?object_id=%s' % video_id, video_id) - title = self._og_search_title(webpage) - series = self._search_regex( - r'{name:\s*([\'"])SeriesTitle\1,\s*value:\s*\1(?P<series>.*?)\1},', + title = self._search_regex( + r'name\s*:\s*([\'"])Title\1\s*,\s*value\s*:\s*\1(?P<title>.+?)\1', + webpage, 'title', group='title') + series_title = self._search_regex( + r'name\s*:\s*([\'"])SeriesTitle\1\s*,\s*value\s*:\s*\1(?P<series>.+?)\1', webpage, 'series', group='series', default=None) - if series is not None and series not in title: - title = '%s, %s' % (series, title) - description = self._og_search_description(webpage, default=None) + if series_title: + title = '%s, %s' % (series_title, title) + + thumbnail = self._search_regex( + r"poster\s*:\s*'([^']+)'", webpage, 'thumbnail', default=None) video_url = self._search_regex( r'0:{src:([\'"])(?P<url>.*?)\1', webpage, 'formats', group='url', default=None) - if video_url is None: + if not video_url: video_url = self._download_json( 'http://www.tvp.pl/pub/stat/videofileinfo?video_id=%s' % video_id, video_id)['video_url'] @@ -89,8 +87,7 @@ class TvpIE(InfoExtractor): return { 'id': video_id, 'title': title, - 'thumbnail': self._og_search_thumbnail(webpage), - 'description': description, + 'thumbnail': thumbnail, 'formats': formats, } diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py index b11a1d561..340cadcf5 100644 --- a/youtube_dl/extractor/twitch.py +++ b/youtube_dl/extractor/twitch.py @@ -3,9 +3,11 @@ from __future__ import unicode_literals import itertools import re +import random from .common import InfoExtractor from ..compat import ( + compat_str, compat_urllib_parse, compat_urllib_request, ) @@ -15,44 +17,12 @@ from ..utils import ( ) -class TwitchIE(InfoExtractor): - # TODO: One broadcast may be split into multiple videos. The key - # 'broadcast_id' is the same for all parts, and 'broadcast_part' - # starts at 1 and increases. Can we treat all parts as one video? - _VALID_URL = r"""(?x)^(?:http://)?(?:www\.)?twitch\.tv/ - (?: - (?P<channelid>[^/]+)| - (?:(?:[^/]+)/v/(?P<vodid>[^/]+))| - (?:(?:[^/]+)/b/(?P<videoid>[^/]+))| - (?:(?:[^/]+)/c/(?P<chapterid>[^/]+)) - ) - /?(?:\#.*)?$ - """ - _PAGE_LIMIT = 100 +class TwitchBaseIE(InfoExtractor): + _VALID_URL_BASE = r'https?://(?:www\.)?twitch\.tv' + _API_BASE = 'https://api.twitch.tv' + _USHER_BASE = 'http://usher.twitch.tv' _LOGIN_URL = 'https://secure.twitch.tv/user/login' - _TESTS = [{ - 'url': 'http://www.twitch.tv/riotgames/b/577357806', - 'info_dict': { - 'id': 'a577357806', - 'title': 'Worlds Semifinals - Star Horn Royal Club vs. OMG', - }, - 'playlist_mincount': 12, - }, { - 'url': 'http://www.twitch.tv/acracingleague/c/5285812', - 'info_dict': { - 'id': 'c5285812', - 'title': 'ACRL Off Season - Sports Cars @ Nordschleife', - }, - 'playlist_mincount': 3, - }, { - 'url': 'http://www.twitch.tv/vanillatv', - 'info_dict': { - 'id': 'vanillatv', - 'title': 'VanillaTV', - }, - 'playlist_mincount': 412, - }] def _handle_error(self, response): if not isinstance(response, dict): @@ -64,34 +34,60 @@ class TwitchIE(InfoExtractor): expected=True) def _download_json(self, url, video_id, note='Downloading JSON metadata'): - response = super(TwitchIE, self)._download_json(url, video_id, note) + response = super(TwitchBaseIE, self)._download_json(url, video_id, note) self._handle_error(response) return response - def _extract_media(self, item, item_id): - ITEMS = { - 'a': 'video', - 'v': 'vod', - 'c': 'chapter', + def _real_initialize(self): + self._login() + + def _login(self): + (username, password) = self._get_login_info() + if username is None: + return + + login_page = self._download_webpage( + self._LOGIN_URL, None, 'Downloading login page') + + authenticity_token = self._search_regex( + r'<input name="authenticity_token" type="hidden" value="([^"]+)"', + login_page, 'authenticity token') + + login_form = { + 'utf8': '✓'.encode('utf-8'), + 'authenticity_token': authenticity_token, + 'redirect_on_login': '', + 'embed_form': 'false', + 'mp_source_action': '', + 'follow': '', + 'user[login]': username, + 'user[password]': password, } - info = self._extract_info(self._download_json( + + request = compat_urllib_request.Request( + self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) + request.add_header('Referer', self._LOGIN_URL) + response = self._download_webpage( + request, None, 'Logging in as %s' % username) + + m = re.search( + r"id=([\"'])login_error_message\1[^>]*>(?P<msg>[^<]+)", response) + if m: + raise ExtractorError( + 'Unable to login: %s' % m.group('msg').strip(), expected=True) + + +class TwitchItemBaseIE(TwitchBaseIE): + def _download_info(self, item, item_id): + return self._extract_info(self._download_json( '%s/kraken/videos/%s%s' % (self._API_BASE, item, item_id), item_id, - 'Downloading %s info JSON' % ITEMS[item])) - - if item == 'v': - access_token = self._download_json( - '%s/api/vods/%s/access_token' % (self._API_BASE, item_id), item_id, - 'Downloading %s access token' % ITEMS[item]) - formats = self._extract_m3u8_formats( - 'http://usher.twitch.tv/vod/%s?nauth=%s&nauthsig=%s' - % (item_id, access_token['token'], access_token['sig']), - item_id, 'mp4') - info['formats'] = formats - return info + 'Downloading %s info JSON' % self._ITEM_TYPE)) + def _extract_media(self, item_id): + info = self._download_info(self._ITEM_SHORTCUT, item_id) response = self._download_json( - '%s/api/videos/%s%s' % (self._API_BASE, item, item_id), item_id, - 'Downloading %s playlist JSON' % ITEMS[item]) + '%s/api/videos/%s%s' % (self._API_BASE, self._ITEM_SHORTCUT, item_id), item_id, + 'Downloading %s playlist JSON' % self._ITEM_TYPE) entries = [] chunks = response['chunks'] qualities = list(chunks.keys()) @@ -129,119 +125,227 @@ class TwitchIE(InfoExtractor): 'view_count': info['views'], } - def _real_initialize(self): - self._login() + def _real_extract(self, url): + return self._extract_media(self._match_id(url)) - def _login(self): - (username, password) = self._get_login_info() - if username is None: - return - login_page = self._download_webpage( - self._LOGIN_URL, None, 'Downloading login page') +class TwitchVideoIE(TwitchItemBaseIE): + IE_NAME = 'twitch:video' + _VALID_URL = r'%s/[^/]+/b/(?P<id>[^/]+)' % TwitchBaseIE._VALID_URL_BASE + _ITEM_TYPE = 'video' + _ITEM_SHORTCUT = 'a' - authenticity_token = self._search_regex( - r'<input name="authenticity_token" type="hidden" value="([^"]+)"', - login_page, 'authenticity token') + _TEST = { + 'url': 'http://www.twitch.tv/riotgames/b/577357806', + 'info_dict': { + 'id': 'a577357806', + 'title': 'Worlds Semifinals - Star Horn Royal Club vs. OMG', + }, + 'playlist_mincount': 12, + } - login_form = { - 'utf8': '✓'.encode('utf-8'), - 'authenticity_token': authenticity_token, - 'redirect_on_login': '', - 'embed_form': 'false', - 'mp_source_action': '', - 'follow': '', - 'user[login]': username, - 'user[password]': password, - } - request = compat_urllib_request.Request( - self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) - request.add_header('Referer', self._LOGIN_URL) - response = self._download_webpage( - request, None, 'Logging in as %s' % username) +class TwitchChapterIE(TwitchItemBaseIE): + IE_NAME = 'twitch:chapter' + _VALID_URL = r'%s/[^/]+/c/(?P<id>[^/]+)' % TwitchBaseIE._VALID_URL_BASE + _ITEM_TYPE = 'chapter' + _ITEM_SHORTCUT = 'c' - m = re.search( - r"id=([\"'])login_error_message\1[^>]*>(?P<msg>[^<]+)", response) - if m: - raise ExtractorError( - 'Unable to login: %s' % m.group('msg').strip(), expected=True) + _TESTS = [{ + 'url': 'http://www.twitch.tv/acracingleague/c/5285812', + 'info_dict': { + 'id': 'c5285812', + 'title': 'ACRL Off Season - Sports Cars @ Nordschleife', + }, + 'playlist_mincount': 3, + }, { + 'url': 'http://www.twitch.tv/tsm_theoddone/c/2349361', + 'only_matching': True, + }] + + +class TwitchVodIE(TwitchItemBaseIE): + IE_NAME = 'twitch:vod' + _VALID_URL = r'%s/[^/]+/v/(?P<id>[^/]+)' % TwitchBaseIE._VALID_URL_BASE + _ITEM_TYPE = 'vod' + _ITEM_SHORTCUT = 'v' + + _TEST = { + 'url': 'http://www.twitch.tv/ksptv/v/3622000', + 'info_dict': { + 'id': 'v3622000', + 'ext': 'mp4', + 'title': '''KSPTV: Squadcast: "Everyone's on vacation so here's Dahud" Edition!''', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 6951, + 'timestamp': 1419028564, + 'upload_date': '20141219', + 'uploader': 'KSPTV', + 'uploader_id': 'ksptv', + 'view_count': int, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + } + + def _real_extract(self, url): + item_id = self._match_id(url) + info = self._download_info(self._ITEM_SHORTCUT, item_id) + access_token = self._download_json( + '%s/api/vods/%s/access_token' % (self._API_BASE, item_id), item_id, + 'Downloading %s access token' % self._ITEM_TYPE) + formats = self._extract_m3u8_formats( + '%s/vod/%s?nauth=%s&nauthsig=%s' + % (self._USHER_BASE, item_id, access_token['token'], access_token['sig']), + item_id, 'mp4') + info['formats'] = formats + return info + + +class TwitchPlaylistBaseIE(TwitchBaseIE): + _PLAYLIST_URL = '%s/kraken/channels/%%s/videos/?offset=%%d&limit=%%d' % TwitchBaseIE._API_BASE + _PAGE_LIMIT = 100 + + def _extract_playlist(self, channel_id): + info = self._download_json( + '%s/kraken/channels/%s' % (self._API_BASE, channel_id), + channel_id, 'Downloading channel info JSON') + channel_name = info.get('display_name') or info.get('name') + entries = [] + offset = 0 + limit = self._PAGE_LIMIT + for counter in itertools.count(1): + response = self._download_json( + self._PLAYLIST_URL % (channel_id, offset, limit), + channel_id, 'Downloading %s videos JSON page %d' % (self._PLAYLIST_TYPE, counter)) + videos = response['videos'] + if not videos: + break + entries.extend([self.url_result(video['url']) for video in videos]) + offset += limit + return self.playlist_result(entries, channel_id, channel_name) + + def _real_extract(self, url): + return self._extract_playlist(self._match_id(url)) + + +class TwitchProfileIE(TwitchPlaylistBaseIE): + IE_NAME = 'twitch:profile' + _VALID_URL = r'%s/(?P<id>[^/]+)/profile/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE + _PLAYLIST_TYPE = 'profile' + + _TEST = { + 'url': 'http://www.twitch.tv/vanillatv/profile', + 'info_dict': { + 'id': 'vanillatv', + 'title': 'VanillaTV', + }, + 'playlist_mincount': 412, + } + + +class TwitchPastBroadcastsIE(TwitchPlaylistBaseIE): + IE_NAME = 'twitch:past_broadcasts' + _VALID_URL = r'%s/(?P<id>[^/]+)/profile/past_broadcasts/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE + _PLAYLIST_URL = TwitchPlaylistBaseIE._PLAYLIST_URL + '&broadcasts=true' + _PLAYLIST_TYPE = 'past broadcasts' + + _TEST = { + 'url': 'http://www.twitch.tv/spamfish/profile/past_broadcasts', + 'info_dict': { + 'id': 'spamfish', + 'title': 'Spamfish', + }, + 'playlist_mincount': 54, + } + + +class TwitchStreamIE(TwitchBaseIE): + IE_NAME = 'twitch:stream' + _VALID_URL = r'%s/(?P<id>[^/]+)/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE + + _TEST = { + 'url': 'http://www.twitch.tv/shroomztv', + 'info_dict': { + 'id': '12772022048', + 'display_id': 'shroomztv', + 'ext': 'mp4', + 'title': 're:^ShroomzTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'description': 'H1Z1 - lonewolfing with ShroomzTV | A3 Battle Royale later - @ShroomzTV', + 'is_live': True, + 'timestamp': 1421928037, + 'upload_date': '20150122', + 'uploader': 'ShroomzTV', + 'uploader_id': 'shroomztv', + 'view_count': int, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - if mobj.group('chapterid'): - return self._extract_media('c', mobj.group('chapterid')) + channel_id = self._match_id(url) + + stream = self._download_json( + '%s/kraken/streams/%s' % (self._API_BASE, channel_id), channel_id, + 'Downloading stream JSON').get('stream') + + # Fallback on profile extraction if stream is offline + if not stream: + return self.url_result( + 'http://www.twitch.tv/%s/profile' % channel_id, + 'TwitchProfile', channel_id) - """ - webpage = self._download_webpage(url, chapter_id) - m = re.search(r'PP\.archive_id = "([0-9]+)";', webpage) + access_token = self._download_json( + '%s/api/channels/%s/access_token' % (self._API_BASE, channel_id), channel_id, + 'Downloading channel access token') + + query = { + 'allow_source': 'true', + 'p': random.randint(1000000, 10000000), + 'player': 'twitchweb', + 'segment_preference': '4', + 'sig': access_token['sig'], + 'token': access_token['token'], + } + + formats = self._extract_m3u8_formats( + '%s/api/channel/hls/%s.m3u8?%s' + % (self._USHER_BASE, channel_id, compat_urllib_parse.urlencode(query).encode('utf-8')), + channel_id, 'mp4') + + view_count = stream.get('viewers') + timestamp = parse_iso8601(stream.get('created_at')) + + channel = stream['channel'] + title = self._live_title(channel.get('display_name') or channel.get('name')) + description = channel.get('status') + + thumbnails = [] + for thumbnail_key, thumbnail_url in stream['preview'].items(): + m = re.search(r'(?P<width>\d+)x(?P<height>\d+)\.jpg$', thumbnail_key) if not m: - raise ExtractorError('Cannot find archive of a chapter') - archive_id = m.group(1) - - api = api_base + '/broadcast/by_chapter/%s.xml' % chapter_id - doc = self._download_xml( - api, chapter_id, - note='Downloading chapter information', - errnote='Chapter information download failed') - for a in doc.findall('.//archive'): - if archive_id == a.find('./id').text: - break - else: - raise ExtractorError('Could not find chapter in chapter information') - - video_url = a.find('./video_file_url').text - video_ext = video_url.rpartition('.')[2] or 'flv' - - chapter_api_url = 'https://api.twitch.tv/kraken/videos/c' + chapter_id - chapter_info = self._download_json( - chapter_api_url, 'c' + chapter_id, - note='Downloading chapter metadata', - errnote='Download of chapter metadata failed') - - bracket_start = int(doc.find('.//bracket_start').text) - bracket_end = int(doc.find('.//bracket_end').text) - - # TODO determine start (and probably fix up file) - # youtube-dl -v http://www.twitch.tv/firmbelief/c/1757457 - #video_url += '?start=' + TODO:start_timestamp - # bracket_start is 13290, but we want 51670615 - self._downloader.report_warning('Chapter detected, but we can just download the whole file. ' - 'Chapter starts at %s and ends at %s' % (formatSeconds(bracket_start), formatSeconds(bracket_end))) - - info = { - 'id': 'c' + chapter_id, - 'url': video_url, - 'ext': video_ext, - 'title': chapter_info['title'], - 'thumbnail': chapter_info['preview'], - 'description': chapter_info['description'], - 'uploader': chapter_info['channel']['display_name'], - 'uploader_id': chapter_info['channel']['name'], - } - return info - """ - elif mobj.group('videoid'): - return self._extract_media('a', mobj.group('videoid')) - elif mobj.group('vodid'): - return self._extract_media('v', mobj.group('vodid')) - elif mobj.group('channelid'): - channel_id = mobj.group('channelid') - info = self._download_json( - '%s/kraken/channels/%s' % (self._API_BASE, channel_id), - channel_id, 'Downloading channel info JSON') - channel_name = info.get('display_name') or info.get('name') - entries = [] - offset = 0 - limit = self._PAGE_LIMIT - for counter in itertools.count(1): - response = self._download_json( - '%s/kraken/channels/%s/videos/?offset=%d&limit=%d' - % (self._API_BASE, channel_id, offset, limit), - channel_id, 'Downloading channel videos JSON page %d' % counter) - videos = response['videos'] - if not videos: - break - entries.extend([self.url_result(video['url'], 'Twitch') for video in videos]) - offset += limit - return self.playlist_result(entries, channel_id, channel_name) + continue + thumbnails.append({ + 'url': thumbnail_url, + 'width': int(m.group('width')), + 'height': int(m.group('height')), + }) + + return { + 'id': compat_str(stream['_id']), + 'display_id': channel_id, + 'title': title, + 'description': description, + 'thumbnails': thumbnails, + 'uploader': channel.get('display_name'), + 'uploader_id': channel.get('name'), + 'timestamp': timestamp, + 'view_count': view_count, + 'formats': formats, + 'is_live': True, + } diff --git a/youtube_dl/extractor/videomega.py b/youtube_dl/extractor/videomega.py index 7a78f0d26..fc6e05fe0 100644 --- a/youtube_dl/extractor/videomega.py +++ b/youtube_dl/extractor/videomega.py @@ -1,11 +1,15 @@ # coding: utf-8 from __future__ import unicode_literals +import re + from .common import InfoExtractor from ..compat import ( compat_urllib_parse, + compat_urllib_request, ) from ..utils import ( + ExtractorError, remove_start, ) @@ -16,34 +20,40 @@ class VideoMegaIE(InfoExtractor): (?:iframe\.php)?\?ref=(?P<id>[A-Za-z0-9]+) ''' _TEST = { - 'url': 'http://videomega.tv/?ref=GKeGPVedBe', - 'md5': '240fb5bcf9199961f48eb17839b084d6', + 'url': 'http://videomega.tv/?ref=QR0HCUHI1661IHUCH0RQ', + 'md5': 'bf5c2f95c4c917536e80936af7bc51e1', 'info_dict': { - 'id': 'GKeGPVedBe', + 'id': 'QR0HCUHI1661IHUCH0RQ', 'ext': 'mp4', - 'title': 'XXL - All Sports United', + 'title': 'Big Buck Bunny', 'thumbnail': 're:^https?://.*\.jpg$', } } def _real_extract(self, url): video_id = self._match_id(url) - url = 'http://videomega.tv/iframe.php?ref={0:}'.format(video_id) - webpage = self._download_webpage(url, video_id) - escaped_data = self._search_regex( - r'unescape\("([^"]+)"\)', webpage, 'escaped data') + iframe_url = 'http://videomega.tv/iframe.php?ref={0:}'.format(video_id) + req = compat_urllib_request.Request(iframe_url) + req.add_header('Referer', url) + webpage = self._download_webpage(req, video_id) + + try: + escaped_data = re.findall(r'unescape\("([^"]+)"\)', webpage)[-1] + except IndexError: + raise ExtractorError('Unable to extract escaped data') + playlist = compat_urllib_parse.unquote(escaped_data) thumbnail = self._search_regex( r'image:\s*"([^"]+)"', playlist, 'thumbnail', fatal=False) - url = self._search_regex(r'file:\s*"([^"]+)"', playlist, 'URL') + video_url = self._search_regex(r'file:\s*"([^"]+)"', playlist, 'URL') title = remove_start(self._html_search_regex( r'<title>(.*?)</title>', webpage, 'title'), 'VideoMega.tv - ') formats = [{ 'format_id': 'sd', - 'url': url, + 'url': video_url, }] self._sort_formats(formats) @@ -52,4 +62,5 @@ class VideoMegaIE(InfoExtractor): 'title': title, 'formats': formats, 'thumbnail': thumbnail, + 'http_referer': iframe_url, } diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index c7611a3a0..2a1f8be0a 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -394,6 +394,23 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 'format': '141', }, }, + # JS player signature function name containing $ + { + 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM', + 'info_dict': { + 'id': 'nfWlot6h_JM', + 'ext': 'm4a', + 'title': 'Taylor Swift - Shake It Off', + 'description': 'md5:2acfda1b285bdd478ccec22f9918199d', + 'uploader': 'TaylorSwiftVEVO', + 'uploader_id': 'TaylorSwiftVEVO', + 'upload_date': '20140818', + }, + 'params': { + 'youtube_include_dash_manifest': True, + 'format': '141', + }, + }, # Controversy video { 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8', @@ -588,7 +605,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): def _parse_sig_js(self, jscode): funcname = self._search_regex( - r'\.sig\|\|([a-zA-Z0-9]+)\(', jscode, + r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode, 'Initial JS player signature function name') jsi = JSInterpreter(jscode) diff --git a/youtube_dl/options.py b/youtube_dl/options.py index a30974efd..fd7b400b2 100644 --- a/youtube_dl/options.py +++ b/youtube_dl/options.py @@ -264,7 +264,7 @@ def parseOpts(overrideArguments=None): authentication.add_option( '-p', '--password', dest='password', metavar='PASSWORD', - help='account password') + help='account password. If this option is left out, youtube-dl will ask interactively.') authentication.add_option( '-2', '--twofactor', dest='twofactor', metavar='TWOFACTOR', @@ -289,6 +289,17 @@ def parseOpts(overrideArguments=None): 'extensions aac, m4a, mp3, mp4, ogg, wav, webm. ' 'You can also use the special names "best",' ' "bestvideo", "bestaudio", "worst". ' + ' You can filter the video results by putting a condition in' + ' brackets, as in -f "best[height=720]"' + ' (or -f "[filesize>10M]"). ' + ' This works for filesize, height, width, tbr, abr, and vbr' + ' and the comparisons <, <=, >, >=, =, != .' + ' Formats for which the value is not known are excluded unless you' + ' put a question mark (?) after the operator.' + ' You can combine format filters, so ' + '-f "[height <=? 720][tbr>500]" ' + 'selects up to 720p videos (or videos where the height is not ' + 'known) with a bitrate of at least 500 KBit/s.' ' By default, youtube-dl will pick the best quality.' ' Use commas to download multiple audio formats, such as' ' -f 136/137/mp4/bestvideo,140/m4a/bestaudio.' diff --git a/youtube_dl/postprocessor/ffmpeg.py b/youtube_dl/postprocessor/ffmpeg.py index 26b99e43c..faccdc43d 100644 --- a/youtube_dl/postprocessor/ffmpeg.py +++ b/youtube_dl/postprocessor/ffmpeg.py @@ -475,15 +475,21 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor): filename = information['filepath'] input_files = [filename] + [subtitles_filename(filename, lang, self._subformat) for lang in sub_langs] - opts = ['-map', '0:0', '-map', '0:1', '-c:v', 'copy', '-c:a', 'copy'] + opts = [ + '-map', '0', + '-c', 'copy', + # Don't copy the existing subtitles, we may be running the + # postprocessor a second time + '-map', '-0:s', + '-c:s', 'mov_text', + ] for (i, lang) in enumerate(sub_langs): - opts.extend(['-map', '%d:0' % (i + 1), '-c:s:%d' % i, 'mov_text']) + opts.extend(['-map', '%d:0' % (i + 1)]) lang_code = self._conver_lang_code(lang) if lang_code is not None: opts.extend(['-metadata:s:s:%d' % i, 'language=%s' % lang_code]) - opts.extend(['-f', 'mp4']) - temp_filename = filename + '.temp' + temp_filename = prepend_extension(filename, 'temp') self._downloader.to_screen('[ffmpeg] Embedding subtitles in \'%s\'' % filename) self.run_ffmpeg_multiple_files(input_files, temp_filename, opts) os.remove(encodeFilename(filename)) @@ -503,6 +509,10 @@ class FFmpegMetadataPP(FFmpegPostProcessor): metadata['artist'] = info['uploader'] elif info.get('uploader_id') is not None: metadata['artist'] = info['uploader_id'] + if info.get('description') is not None: + metadata['description'] = info['description'] + if info.get('webpage_url') is not None: + metadata['comment'] = info['webpage_url'] if not metadata: self._downloader.to_screen('[ffmpeg] There isn\'t any metadata to add') diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index daf94abd1..3536a5bd6 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -399,8 +399,9 @@ def formatSeconds(secs): def make_HTTPS_handler(params, **kwargs): opts_no_check_certificate = params.get('nocheckcertificate', False) if hasattr(ssl, 'create_default_context'): # Python >= 3.4 or 2.7.9 - context = ssl.create_default_context(ssl.Purpose.CLIENT_AUTH) + context = ssl.create_default_context(ssl.Purpose.SERVER_AUTH) if opts_no_check_certificate: + context.check_hostname = False context.verify_mode = ssl.CERT_NONE try: return YoutubeDLHTTPSHandler(params, context=context, **kwargs) @@ -723,6 +724,7 @@ def unified_strdate(date_str, day_first=True): '%b %dst %Y %I:%M%p', '%b %dnd %Y %I:%M%p', '%b %dth %Y %I:%M%p', + '%Y %m %d', '%Y-%m-%d', '%Y/%m/%d', '%Y/%m/%d %H:%M:%S', @@ -1277,7 +1279,7 @@ def parse_duration(s): s = s.strip() m = re.match( - r'''(?ix)T? + r'''(?ix)(?:P?T)? (?: (?P<only_mins>[0-9.]+)\s*(?:mins?|minutes?)\s*| (?P<only_hours>[0-9.]+)\s*(?:hours?)| @@ -1612,6 +1614,14 @@ def urlhandle_detect_ext(url_handle): except AttributeError: # Python < 3 getheader = url_handle.info().getheader + cd = getheader('Content-Disposition') + if cd: + m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd) + if m: + e = determine_ext(m.group('filename'), default_ext=None) + if e: + return e + return getheader('Content-Type').split("/")[1] @@ -1623,3 +1633,23 @@ def age_restricted(content_limit, age_limit): if content_limit is None: return False # Content available for everyone return age_limit < content_limit + + +def is_html(first_bytes): + """ Detect whether a file contains HTML by examining its first bytes. """ + + BOMS = [ + (b'\xef\xbb\xbf', 'utf-8'), + (b'\x00\x00\xfe\xff', 'utf-32-be'), + (b'\xff\xfe\x00\x00', 'utf-32-le'), + (b'\xff\xfe', 'utf-16-le'), + (b'\xfe\xff', 'utf-16-be'), + ] + for bom, enc in BOMS: + if first_bytes.startswith(bom): + s = first_bytes[len(bom):].decode(enc, 'replace') + break + else: + s = first_bytes.decode('utf-8', 'replace') + + return re.match(r'^\s*<', s) diff --git a/youtube_dl/version.py b/youtube_dl/version.py index 0a184ad08..28458fd69 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,3 +1,3 @@ from __future__ import unicode_literals -__version__ = '2015.01.11' +__version__ = '2015.01.23.1' |