From: Philipp Hagemeister Date: Fri, 20 Dec 2013 16:13:26 +0000 (+0100) Subject: Merge remote-tracking branch 'rzhxeo/crunchyroll' X-Git-Url: http://git.bitcoin.ninja/index.cgi?p=youtube-dl;a=commitdiff_plain;h=304cbe981e9f3b62288abc54bd84b0eaef286795;hp=c8434e83163fc90007eb5b501ea0e827f8b5e127 Merge remote-tracking branch 'rzhxeo/crunchyroll' --- diff --git a/README.md b/README.md index a2b296613..caed94846 100644 --- a/README.md +++ b/README.md @@ -30,13 +30,16 @@ which means you can modify it, redistribute it or use it however you like. --list-extractors List all supported extractors and the URLs they would handle --extractor-descriptions Output descriptions of all supported extractors - --proxy URL Use the specified HTTP/HTTPS proxy + --proxy URL Use the specified HTTP/HTTPS proxy. Pass in an + empty string (--proxy "") for direct connection --no-check-certificate Suppress HTTPS certificate validation. --cache-dir DIR Location in the filesystem where youtube-dl can store downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache /youtube-dl . --no-cache-dir Disable filesystem caching + --bidi-workaround Work around terminals that lack bidirectional + text support. Requires fribidi executable in PATH ## Video Selection: --playlist-start NUMBER playlist video to start at (default is 1) @@ -53,10 +56,15 @@ which means you can modify it, redistribute it or use it however you like. --date DATE download only videos uploaded in this date --datebefore DATE download only videos uploaded before this date --dateafter DATE download only videos uploaded after this date + --min-views COUNT Do not download any videos with less than COUNT + views + --max-views COUNT Do not download any videos with more than COUNT + views --no-playlist download only the currently playing video --age-limit YEARS download only videos suitable for the given age - --download-archive FILE Download only videos not present in the archive - file. Record all downloaded videos in it. + --download-archive FILE Download only videos not listed in the archive + file. Record the IDs of all downloaded videos in + it. ## Download Options: -r, --rate-limit LIMIT maximum download rate in bytes per second (e.g. @@ -92,12 +100,16 @@ which means you can modify it, redistribute it or use it however you like. ownloads/%(uploader)s/%(title)s-%(id)s.%(ext)s' . --autonumber-size NUMBER Specifies the number of digits in %(autonumber)s when it is present in output filename template or - --autonumber option is given + --auto-number option is given --restrict-filenames Restrict filenames to only ASCII characters, and avoid "&" and spaces in filenames -a, --batch-file FILE file containing URLs to download ('-' for stdin) + --load-info FILE json file containing the video information + (created with the "--write-json" option -w, --no-overwrites do not overwrite files - -c, --continue resume partially downloaded files + -c, --continue force resume of partially downloaded files. By + default, youtube-dl will resume downloads if + possible. --no-continue do not resume partially downloaded files (restart from beginning) --cookies FILE file to read cookies from and dump cookie jar in @@ -119,19 +131,21 @@ which means you can modify it, redistribute it or use it however you like. --get-id simulate, quiet but print id --get-thumbnail simulate, quiet but print thumbnail URL --get-description simulate, quiet but print video description + --get-duration simulate, quiet but print video length --get-filename simulate, quiet but print output filename --get-format simulate, quiet but print output format + -j, --dump-json simulate, quiet but print JSON information --newline output progress bar as new lines --no-progress do not print progress bar --console-title display progress in console titlebar -v, --verbose print various debugging information --dump-intermediate-pages print downloaded pages to debug problems(very verbose) - --write-pages Write downloaded pages to files in the current - directory + --write-pages Write downloaded intermediary pages to files in + the current directory to debug problems ## Video Format Options: - -f, --format FORMAT video format code, specifiy the order of + -f, --format FORMAT video format code, specify the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported --all-formats download all available video formats @@ -179,7 +193,7 @@ which means you can modify it, redistribute it or use it however you like. # CONFIGURATION -You can configure youtube-dl by placing default arguments (such as `--extract-audio --no-mtime` to always extract the audio and not copy the mtime) into `/etc/youtube-dl.conf` and/or `~/.config/youtube-dl.conf`. +You can configure youtube-dl by placing default arguments (such as `--extract-audio --no-mtime` to always extract the audio and not copy the mtime) into `/etc/youtube-dl.conf` and/or `~/.config/youtube-dl.conf`. On Windows, the configuration file locations are `%APPDATA%\youtube-dl\config.txt` and `C:\Users\\youtube-dl.conf`. # OUTPUT TEMPLATE @@ -269,14 +283,54 @@ This README file was originally written by Daniel Bolton ( +Bugs and suggestions should be reported at: . Unless you were prompted so or there is another pertinent reason (e.g. GitHub fails to accept the bug report), please do not send bug reports via personal email. -Please include: - -* Your exact command line, like `youtube-dl -t "http://www.youtube.com/watch?v=uHlDtZ6Oc3s&feature=channel_video_title"`. A common mistake is not to escape the `&`. Putting URLs in quotes should solve this problem. -* If possible re-run the command with `--verbose`, and include the full output, it is really helpful to us. -* The output of `youtube-dl --version` -* The output of `python --version` -* The name and version of your Operating System ("Ubuntu 11.04 x64" or "Windows 7 x64" is usually enough). +Please include the full output of the command when run with `--verbose`. 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. For discussions, join us in the irc channel #youtube-dl on freenode. + +When you submit a request, please re-read it once to avoid a couple of mistakes (you can and should use this as a checklist): + +### Is the description of the issue itself sufficient? + +We often get issue reports that we cannot really decipher. While in most cases we eventually get the required information after asking back multiple times, this poses an unnecessary drain on our resources. Many contributors, including myself, are also not native speakers, so we may misread some parts. + +So please elaborate on what feature you are requesting, or what bug you want to be fixed. Make sure that it's obvious + +- What the problem is +- How it could be fixed +- How your proposed solution would look like + +If your report is shorter than two lines, it is almost certainly missing some of these, which makes it hard for us to respond to it. We're often too polite to close the issue outright, but the missing info makes misinterpretation likely. As a commiter myself, I often get frustrated by these issues, since the only possible way for me to move forward on them is to ask for clarification over and over. + +For bug reports, this means that your report should contain the *complete* output of youtube-dl when called with the -v flag. The error message you get for (most) bugs even says so, but you would not believe how many of our bug reports do not contain this information. + +Site support requests must contain an example URL. An example URL is a URL you might want to download, like http://www.youtube.com/watch?v=BaW_jenozKc . There should be an obvious video present. Except under very special circumstances, the main page of a video service (e.g. http://www.youtube.com/ ) is *not* an example URL. + +### Are you using the latest version? + +Before reporting any issue, type youtube-dl -U. This should report that you're up-to-date. Ábout 20% of the reports we receive are already fixed, but people are using outdated versions. This goes for feature requests as well. + +### Is the issue already documented? + +Make sure that someone has not already opened the issue you're trying to open. Search at the top of the window or at https://github.com/rg3/youtube-dl/search?type=Issues . If there is an issue, feel free to write something along the lines of "This affects me as well, with version 2015.01.01. Here is some more information on the issue: ...". While some issues may be old, a new post into them often spurs rapid activity. + +### Why are existing options not enough? + +Before requesting a new feature, please have a quick peek at [the list of supported options](https://github.com/rg3/youtube-dl/blob/master/README.md#synopsis). Many feature requests are for features that actually exist already! Please, absolutely do show off your work in the issue report and detail how the existing similar options do *not* solve your problem. + +### Is there enough context in your bug report? + +People want to solve problems, and often think they do us a favor by breaking down their larger problems (e.g. wanting to skip already downloaded files) to a specific request (e.g. requesting us to look whether the file exists before downloading the info page). However, what often happens is that they break down the problem into two steps: One simple, and one impossible (or extremely complicated one). + +We are then presented with a very complicated request when the original problem could be solved far easier, e.g. by recording the downloaded video IDs in a separate file. To avoid this, you must include the greater context where it is non-obvious. In particular, every feature request that does not consist of adding support for a new site should contain a use case scenario that explains in what situation the missing feature would be useful. + +### Does the issue involve one problem, and one problem only? + +Some of our users seem to think there is a limit of issues they can or should open. There is no limit of issues they can or should open. While it may seem appealing to be able to dump all your issues into one ticket, that means that someone who solves one of your issues cannot mark the issue as closed. Typically, reporting a bunch of issues leads to the ticket lingering since nobody wants to attack that behemoth, until someone mercifully splits the issue into multiple ones. + +In particular, every site support request issue should only pertain to services at one site (generally under a common domain, but always using the same backend technology). Do not request support for vimeo user videos, Whitehouse podcasts, and Google Plus pages in the same issue. Also, make sure that you don't post bug reports alongside feature requests. As a rule of thumb, a feature request does not include outputs of youtube-dl that are not immediately related to the feature at hand. Do not post reports of a network error alongside the request for a new video service. + +### Is anyone going to need the feature? + +Only post features that you (or an incapicated friend you can personally talk to) require. Do not post features because they seem like a good idea. If they are really useful, they will be requested by someone who requires them. diff --git a/devscripts/bash-completion.in b/devscripts/bash-completion.in index ce893fcbe..3af87a378 100644 --- a/devscripts/bash-completion.in +++ b/devscripts/bash-completion.in @@ -1,10 +1,21 @@ __youtube_dl() { - local cur prev opts + local cur prev opts fileopts diropts keywords COMPREPLY=() cur="${COMP_WORDS[COMP_CWORD]}" + prev="${COMP_WORDS[COMP_CWORD-1]}" opts="{{flags}}" - keywords=":ytfavorites :ytrecommended :ytsubscriptions :ytwatchlater" + keywords=":ytfavorites :ytrecommended :ytsubscriptions :ytwatchlater :ythistory" + fileopts="-a|--batch-file|--download-archive|--cookies" + diropts="--cache-dir" + + if [[ ${prev} =~ ${fileopts} ]]; then + COMPREPLY=( $(compgen -f -- ${cur}) ) + return 0 + elif [[ ${prev} =~ ${diropts} ]]; then + COMPREPLY=( $(compgen -d -- ${cur}) ) + return 0 + fi if [[ ${cur} =~ : ]]; then COMPREPLY=( $(compgen -W "${keywords}" -- ${cur}) ) diff --git a/setup.py b/setup.py index aa7cfca08..8e24fe679 100644 --- a/setup.py +++ b/setup.py @@ -48,7 +48,7 @@ else: 'data_files': [ # Installing system-wide would require sudo... ('etc/bash_completion.d', ['youtube-dl.bash-completion']), ('share/doc/youtube_dl', ['README.txt']), - ('share/man/man1/', ['youtube-dl.1']) + ('share/man/man1', ['youtube-dl.1']) ] } if setuptools_available: diff --git a/test/helper.py b/test/helper.py index d7bf7a828..b1f421ac5 100644 --- a/test/helper.py +++ b/test/helper.py @@ -12,10 +12,6 @@ from youtube_dl import YoutubeDL from youtube_dl.utils import preferredencoding -def global_setup(): - youtube_dl._setup_opener(timeout=10) - - def get_params(override=None): PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json") diff --git a/test/parameters.json b/test/parameters.json index f042880ed..487a46d56 100644 --- a/test/parameters.json +++ b/test/parameters.json @@ -39,5 +39,6 @@ "writeinfojson": true, "writesubtitles": false, "allsubtitles": false, - "listssubtitles": false + "listssubtitles": false, + "socket_timeout": 20 } diff --git a/test/test_YoutubeDL.py b/test/test_YoutubeDL.py index ffebb4ae5..3100c362a 100644 --- a/test/test_YoutubeDL.py +++ b/test/test_YoutubeDL.py @@ -7,6 +7,7 @@ import unittest sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) from test.helper import FakeYDL +from youtube_dl import YoutubeDL class YDL(FakeYDL): @@ -128,6 +129,32 @@ class TestFormatSelection(unittest.TestCase): downloaded = ydl.downloaded_info_dicts[0] self.assertEqual(downloaded['format_id'], u'35') + def test_add_extra_info(self): + test_dict = { + 'extractor': 'Foo', + } + extra_info = { + 'extractor': 'Bar', + 'playlist': 'funny videos', + } + YDL.add_extra_info(test_dict, extra_info) + self.assertEqual(test_dict['extractor'], 'Foo') + self.assertEqual(test_dict['playlist'], 'funny videos') + + def test_prepare_filename(self): + info = { + u'id': u'1234', + u'ext': u'mp4', + u'width': None, + } + def fname(templ): + ydl = YoutubeDL({'outtmpl': templ}) + return ydl.prepare_filename(info) + self.assertEqual(fname(u'%(id)s.%(ext)s'), u'1234.mp4') + self.assertEqual(fname(u'%(id)s-%(width)s.%(ext)s'), u'1234-NA.mp4') + # Replace missing fields with 'NA' + self.assertEqual(fname(u'%(uploader_date)s-%(id)s.%(ext)s'), u'NA-1234.mp4') + if __name__ == '__main__': unittest.main() diff --git a/test/test_age_restriction.py b/test/test_age_restriction.py index d500c6edc..c9cdb96cb 100644 --- a/test/test_age_restriction.py +++ b/test/test_age_restriction.py @@ -6,8 +6,7 @@ import sys import unittest sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) -from test.helper import global_setup, try_rm -global_setup() +from test.helper import try_rm from youtube_dl import YoutubeDL @@ -24,7 +23,7 @@ def _download_restricted(url, filename, age): } ydl = YoutubeDL(params) ydl.add_default_info_extractors() - json_filename = filename + '.info.json' + json_filename = os.path.splitext(filename)[0] + '.info.json' try_rm(json_filename) ydl.download([url]) res = os.path.exists(json_filename) diff --git a/test/test_all_urls.py b/test/test_all_urls.py index 56e5f80e1..bd77b7c30 100644 --- a/test/test_all_urls.py +++ b/test/test_all_urls.py @@ -10,6 +10,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) from test.helper import get_testcases from youtube_dl.extractor import ( + FacebookIE, gen_extractors, JustinTVIE, YoutubeIE, @@ -87,12 +88,15 @@ class TestAllURLsMatching(unittest.TestCase): assertExtractId('http://www.youtube.com/watch?v=BaW_jenozKcsharePLED17F32AD9753930', 'BaW_jenozKc') assertExtractId('BaW_jenozKc', 'BaW_jenozKc') + def test_facebook_matching(self): + self.assertTrue(FacebookIE.suitable(u'https://www.facebook.com/Shiniknoh#!/photo.php?v=10153317450565268')) + def test_no_duplicates(self): ies = gen_extractors() for tc in get_testcases(): url = tc['url'] for ie in ies: - if type(ie).__name__ in ['GenericIE', tc['name'] + 'IE']: + if type(ie).__name__ in ('GenericIE', tc['name'] + 'IE'): self.assertTrue(ie.suitable(url), '%s should match URL %r' % (type(ie).__name__, url)) else: self.assertFalse(ie.suitable(url), '%s should not match URL %r' % (type(ie).__name__, url)) @@ -100,11 +104,19 @@ class TestAllURLsMatching(unittest.TestCase): def test_keywords(self): self.assertMatch(':ytsubs', ['youtube:subscriptions']) self.assertMatch(':ytsubscriptions', ['youtube:subscriptions']) - self.assertMatch(':thedailyshow', ['ComedyCentral']) - self.assertMatch(':tds', ['ComedyCentral']) - self.assertMatch(':colbertreport', ['ComedyCentral']) - self.assertMatch(':cr', ['ComedyCentral']) - + 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']) + self.assertMatch('http://vimeo.com/user7108434', ['vimeo:user']) + + # https://github.com/rg3/youtube-dl/issues/1930 + def test_soundcloud_not_matching_sets(self): + self.assertMatch('http://soundcloud.com/floex/sets/gone-ep', ['soundcloud:set']) if __name__ == '__main__': unittest.main() diff --git a/test/test_dailymotion_subtitles.py b/test/test_dailymotion_subtitles.py deleted file mode 100644 index ba3580ea4..000000000 --- a/test/test_dailymotion_subtitles.py +++ /dev/null @@ -1,70 +0,0 @@ -#!/usr/bin/env python - -# Allow direct execution -import os -import sys -import unittest -sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) - -from test.helper import FakeYDL, global_setup, md5 -global_setup() - - -from youtube_dl.extractor import DailymotionIE - -class TestDailymotionSubtitles(unittest.TestCase): - def setUp(self): - self.DL = FakeYDL() - self.url = 'http://www.dailymotion.com/video/xczg00' - def getInfoDict(self): - IE = DailymotionIE(self.DL) - info_dict = IE.extract(self.url) - return info_dict - def getSubtitles(self): - info_dict = self.getInfoDict() - return info_dict['subtitles'] - def test_no_writesubtitles(self): - subtitles = self.getSubtitles() - self.assertEqual(subtitles, None) - def test_subtitles(self): - self.DL.params['writesubtitles'] = True - subtitles = self.getSubtitles() - self.assertEqual(md5(subtitles['en']), '976553874490cba125086bbfea3ff76f') - def test_subtitles_lang(self): - self.DL.params['writesubtitles'] = True - self.DL.params['subtitleslangs'] = ['fr'] - subtitles = self.getSubtitles() - self.assertEqual(md5(subtitles['fr']), '594564ec7d588942e384e920e5341792') - def test_allsubtitles(self): - self.DL.params['writesubtitles'] = True - self.DL.params['allsubtitles'] = True - subtitles = self.getSubtitles() - self.assertEqual(len(subtitles.keys()), 5) - def test_list_subtitles(self): - self.DL.expect_warning(u'Automatic Captions not supported by this server') - self.DL.params['listsubtitles'] = True - info_dict = self.getInfoDict() - self.assertEqual(info_dict, None) - def test_automatic_captions(self): - self.DL.expect_warning(u'Automatic Captions not supported by this server') - self.DL.params['writeautomaticsub'] = True - self.DL.params['subtitleslang'] = ['en'] - subtitles = self.getSubtitles() - self.assertTrue(len(subtitles.keys()) == 0) - def test_nosubtitles(self): - self.DL.expect_warning(u'video doesn\'t have subtitles') - self.url = 'http://www.dailymotion.com/video/x12u166_le-zapping-tele-star-du-08-aout-2013_tv' - self.DL.params['writesubtitles'] = True - self.DL.params['allsubtitles'] = True - subtitles = self.getSubtitles() - self.assertEqual(len(subtitles), 0) - def test_multiple_langs(self): - self.DL.params['writesubtitles'] = True - langs = ['es', 'fr', 'de'] - self.DL.params['subtitleslangs'] = langs - subtitles = self.getSubtitles() - for lang in langs: - self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang) - -if __name__ == '__main__': - unittest.main() diff --git a/test/test_download.py b/test/test_download.py index dfb04d010..dd5818dba 100644 --- a/test/test_download.py +++ b/test/test_download.py @@ -9,12 +9,10 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) from test.helper import ( get_params, get_testcases, - global_setup, try_rm, md5, report_warning ) -global_setup() import hashlib @@ -31,6 +29,7 @@ from youtube_dl.utils import ( ExtractorError, UnavailableVideoError, ) +from youtube_dl.extractor import get_info_extractor RETRIES = 3 @@ -63,9 +62,10 @@ def generator(test_case): def test_template(self): ie = youtube_dl.extractor.get_info_extractor(test_case['name']) + other_ies = [get_info_extractor(ie_key) for ie_key in test_case.get('add_ie', [])] def print_skipping(reason): print('Skipping %s: %s' % (test_case['name'], reason)) - if not ie._WORKING: + if not ie.working(): print_skipping('IE marked as not _WORKING') return if 'playlist' not in test_case: @@ -77,6 +77,10 @@ def generator(test_case): if 'skip' in test_case: print_skipping(test_case['skip']) return + for other_ie in other_ies: + if not other_ie.working(): + print_skipping(u'test depends on %sIE, marked as not WORKING' % other_ie.ie_key()) + return params = get_params(test_case.get('params', {})) @@ -97,7 +101,7 @@ def generator(test_case): tc_filename = get_tc_filename(tc) try_rm(tc_filename) try_rm(tc_filename + '.part') - try_rm(tc_filename + '.info.json') + try_rm(os.path.splitext(tc_filename)[0] + '.info.json') try_rm_tcs_files() try: try_num = 1 @@ -124,11 +128,12 @@ def generator(test_case): if not test_case.get('params', {}).get('skip_download', False): self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename) self.assertTrue(tc_filename in finished_hook_called) - self.assertTrue(os.path.exists(tc_filename + '.info.json')) + info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json' + self.assertTrue(os.path.exists(info_json_fn)) if 'md5' in tc: md5_for_file = _file_md5(tc_filename) self.assertEqual(md5_for_file, tc['md5']) - with io.open(tc_filename + '.info.json', encoding='utf-8') as infof: + with io.open(info_json_fn, encoding='utf-8') as infof: info_dict = json.load(infof) for (info_field, expected) in tc.get('info_dict', {}).items(): if isinstance(expected, compat_str) and expected.startswith('md5:'): @@ -148,6 +153,9 @@ def generator(test_case): # Check for the presence of mandatory fields for key in ('id', 'url', 'title', 'ext'): self.assertTrue(key in info_dict.keys() and info_dict[key]) + # Check for mandatory fields that are automatically set by YoutubeDL + for key in ['webpage_url', 'extractor', 'extractor_key']: + self.assertTrue(info_dict.get(key), u'Missing field: %s' % key) finally: try_rm_tcs_files() diff --git a/test/test_playlists.py b/test/test_playlists.py index d6a8d56df..1b7b4e3d8 100644 --- a/test/test_playlists.py +++ b/test/test_playlists.py @@ -8,18 +8,27 @@ import sys import unittest sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) -from test.helper import FakeYDL, global_setup -global_setup() +from test.helper import FakeYDL from youtube_dl.extractor import ( + AcademicEarthCourseIE, DailymotionPlaylistIE, DailymotionUserIE, VimeoChannelIE, + VimeoUserIE, + VimeoAlbumIE, + VimeoGroupsIE, UstreamChannelIE, + SoundcloudSetIE, SoundcloudUserIE, LivestreamIE, NHLVideocenterIE, + BambuserChannelIE, + BandcampAlbumIE, + SmotriCommunityIE, + SmotriUserIE, + IviCompilationIE ) @@ -52,6 +61,30 @@ class TestPlaylists(unittest.TestCase): self.assertEqual(result['title'], u'Vimeo Tributes') self.assertTrue(len(result['entries']) > 24) + def test_vimeo_user(self): + dl = FakeYDL() + ie = VimeoUserIE(dl) + result = ie.extract('http://vimeo.com/nkistudio/videos') + self.assertIsPlaylist(result) + self.assertEqual(result['title'], u'Nki') + self.assertTrue(len(result['entries']) > 65) + + def test_vimeo_album(self): + dl = FakeYDL() + ie = VimeoAlbumIE(dl) + result = ie.extract('http://vimeo.com/album/2632481') + self.assertIsPlaylist(result) + self.assertEqual(result['title'], u'Staff Favorites: November 2013') + self.assertTrue(len(result['entries']) > 12) + + def test_vimeo_groups(self): + dl = FakeYDL() + ie = VimeoGroupsIE(dl) + result = ie.extract('http://vimeo.com/groups/rolexawards') + self.assertIsPlaylist(result) + self.assertEqual(result['title'], u'Rolex Awards for Enterprise') + self.assertTrue(len(result['entries']) > 72) + def test_ustream_channel(self): dl = FakeYDL() ie = UstreamChannelIE(dl) @@ -60,6 +93,14 @@ class TestPlaylists(unittest.TestCase): self.assertEqual(result['id'], u'5124905') self.assertTrue(len(result['entries']) >= 11) + def test_soundcloud_set(self): + dl = FakeYDL() + ie = SoundcloudSetIE(dl) + result = ie.extract('https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep') + self.assertIsPlaylist(result) + self.assertEqual(result['title'], u'The Royal Concept EP') + self.assertTrue(len(result['entries']) >= 6) + def test_soundcloud_user(self): dl = FakeYDL() ie = SoundcloudUserIE(dl) @@ -85,5 +126,68 @@ class TestPlaylists(unittest.TestCase): self.assertEqual(result['title'], u'Highlights') self.assertEqual(len(result['entries']), 12) + def test_bambuser_channel(self): + dl = FakeYDL() + ie = BambuserChannelIE(dl) + result = ie.extract('http://bambuser.com/channel/pixelversity') + self.assertIsPlaylist(result) + self.assertEqual(result['title'], u'pixelversity') + self.assertTrue(len(result['entries']) >= 60) + + def test_bandcamp_album(self): + dl = FakeYDL() + ie = BandcampAlbumIE(dl) + result = ie.extract('http://mpallante.bandcamp.com/album/nightmare-night-ep') + self.assertIsPlaylist(result) + self.assertEqual(result['title'], u'Nightmare Night EP') + self.assertTrue(len(result['entries']) >= 4) + + def test_smotri_community(self): + dl = FakeYDL() + ie = SmotriCommunityIE(dl) + result = ie.extract('http://smotri.com/community/video/kommuna') + self.assertIsPlaylist(result) + self.assertEqual(result['id'], u'kommuna') + self.assertEqual(result['title'], u'КПРФ') + self.assertTrue(len(result['entries']) >= 4) + + def test_smotri_user(self): + dl = FakeYDL() + ie = SmotriUserIE(dl) + result = ie.extract('http://smotri.com/user/inspector') + self.assertIsPlaylist(result) + self.assertEqual(result['id'], u'inspector') + self.assertEqual(result['title'], u'Inspector') + self.assertTrue(len(result['entries']) >= 9) + + def test_AcademicEarthCourse(self): + dl = FakeYDL() + ie = AcademicEarthCourseIE(dl) + result = ie.extract(u'http://academicearth.org/courses/building-dynamic-websites/') + self.assertIsPlaylist(result) + self.assertEqual(result['id'], u'building-dynamic-websites') + self.assertEqual(result['title'], u'Building Dynamic Websites') + self.assertEqual(result['description'], u"Today's websites are increasingly dynamic. Pages are no longer static HTML files but instead generated by scripts and database calls. User interfaces are more seamless, with technologies like Ajax replacing traditional page reloads. This course teaches students how to build dynamic websites with Ajax and with Linux, Apache, MySQL, and PHP (LAMP), one of today's most popular frameworks. Students learn how to set up domain names with DNS, how to structure pages with XHTML and CSS, how to program in JavaScript and PHP, how to configure Apache and MySQL, how to design and query databases with SQL, how to use Ajax with both XML and JSON, and how to build mashups. The course explores issues of security, scalability, and cross-browser support and also discusses enterprise-level deployments of websites, including third-party hosting, virtualization, colocation in data centers, firewalling, and load-balancing.") + self.assertEqual(len(result['entries']), 10) + + def test_ivi_compilation(self): + dl = FakeYDL() + ie = IviCompilationIE(dl) + result = ie.extract('http://www.ivi.ru/watch/dezhurnyi_angel') + self.assertIsPlaylist(result) + self.assertEqual(result['id'], u'dezhurnyi_angel') + self.assertEqual(result['title'], u'Дежурный ангел (2010 - 2012)') + self.assertTrue(len(result['entries']) >= 36) + + def test_ivi_compilation_season(self): + dl = FakeYDL() + ie = IviCompilationIE(dl) + result = ie.extract('http://www.ivi.ru/watch/dezhurnyi_angel/season2') + self.assertIsPlaylist(result) + self.assertEqual(result['id'], u'dezhurnyi_angel/season2') + self.assertEqual(result['title'], u'Дежурный ангел (2010 - 2012) 2 сезон') + self.assertTrue(len(result['entries']) >= 20) + + if __name__ == '__main__': unittest.main() diff --git a/test/test_subtitles.py b/test/test_subtitles.py new file mode 100644 index 000000000..23a653124 --- /dev/null +++ b/test/test_subtitles.py @@ -0,0 +1,210 @@ +#!/usr/bin/env python + +# Allow direct execution +import os +import sys +import unittest +sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) + +from test.helper import FakeYDL, md5 + + +from youtube_dl.extractor import ( + YoutubeIE, + DailymotionIE, + TEDIE, +) + + +class BaseTestSubtitles(unittest.TestCase): + url = None + IE = None + def setUp(self): + self.DL = FakeYDL() + self.ie = self.IE(self.DL) + + def getInfoDict(self): + info_dict = self.ie.extract(self.url) + return info_dict + + def getSubtitles(self): + info_dict = self.getInfoDict() + return info_dict['subtitles'] + + +class TestYoutubeSubtitles(BaseTestSubtitles): + url = 'QRS8MkLhQmM' + IE = YoutubeIE + + def getSubtitles(self): + info_dict = self.getInfoDict() + return info_dict[0]['subtitles'] + + def test_youtube_no_writesubtitles(self): + self.DL.params['writesubtitles'] = False + subtitles = self.getSubtitles() + self.assertEqual(subtitles, None) + + def test_youtube_subtitles(self): + self.DL.params['writesubtitles'] = True + subtitles = self.getSubtitles() + self.assertEqual(md5(subtitles['en']), '4cd9278a35ba2305f47354ee13472260') + + def test_youtube_subtitles_lang(self): + self.DL.params['writesubtitles'] = True + self.DL.params['subtitleslangs'] = ['it'] + subtitles = self.getSubtitles() + self.assertEqual(md5(subtitles['it']), '164a51f16f260476a05b50fe4c2f161d') + + def test_youtube_allsubtitles(self): + self.DL.params['writesubtitles'] = True + self.DL.params['allsubtitles'] = True + subtitles = self.getSubtitles() + self.assertEqual(len(subtitles.keys()), 13) + + def test_youtube_subtitles_sbv_format(self): + self.DL.params['writesubtitles'] = True + self.DL.params['subtitlesformat'] = 'sbv' + subtitles = self.getSubtitles() + self.assertEqual(md5(subtitles['en']), '13aeaa0c245a8bed9a451cb643e3ad8b') + + def test_youtube_subtitles_vtt_format(self): + self.DL.params['writesubtitles'] = True + self.DL.params['subtitlesformat'] = 'vtt' + subtitles = self.getSubtitles() + self.assertEqual(md5(subtitles['en']), '3cb210999d3e021bd6c7f0ea751eab06') + + def test_youtube_list_subtitles(self): + self.DL.expect_warning(u'Video doesn\'t have automatic captions') + self.DL.params['listsubtitles'] = True + info_dict = self.getInfoDict() + self.assertEqual(info_dict, None) + + def test_youtube_automatic_captions(self): + self.url = '8YoUxe5ncPo' + self.DL.params['writeautomaticsub'] = True + self.DL.params['subtitleslangs'] = ['it'] + subtitles = self.getSubtitles() + self.assertTrue(subtitles['it'] is not None) + + def test_youtube_nosubtitles(self): + self.DL.expect_warning(u'video doesn\'t have subtitles') + self.url = 'sAjKT8FhjI8' + self.DL.params['writesubtitles'] = True + self.DL.params['allsubtitles'] = True + subtitles = self.getSubtitles() + self.assertEqual(len(subtitles), 0) + + def test_youtube_multiple_langs(self): + self.url = 'QRS8MkLhQmM' + self.DL.params['writesubtitles'] = True + langs = ['it', 'fr', 'de'] + self.DL.params['subtitleslangs'] = langs + subtitles = self.getSubtitles() + for lang in langs: + self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang) + + +class TestDailymotionSubtitles(BaseTestSubtitles): + url = 'http://www.dailymotion.com/video/xczg00' + IE = DailymotionIE + + def test_no_writesubtitles(self): + subtitles = self.getSubtitles() + self.assertEqual(subtitles, None) + + def test_subtitles(self): + self.DL.params['writesubtitles'] = True + subtitles = self.getSubtitles() + self.assertEqual(md5(subtitles['en']), '976553874490cba125086bbfea3ff76f') + + def test_subtitles_lang(self): + self.DL.params['writesubtitles'] = True + self.DL.params['subtitleslangs'] = ['fr'] + subtitles = self.getSubtitles() + self.assertEqual(md5(subtitles['fr']), '594564ec7d588942e384e920e5341792') + + def test_allsubtitles(self): + self.DL.params['writesubtitles'] = True + self.DL.params['allsubtitles'] = True + subtitles = self.getSubtitles() + self.assertEqual(len(subtitles.keys()), 5) + + def test_list_subtitles(self): + self.DL.expect_warning(u'Automatic Captions not supported by this server') + self.DL.params['listsubtitles'] = True + info_dict = self.getInfoDict() + self.assertEqual(info_dict, None) + + def test_automatic_captions(self): + self.DL.expect_warning(u'Automatic Captions not supported by this server') + self.DL.params['writeautomaticsub'] = True + self.DL.params['subtitleslang'] = ['en'] + subtitles = self.getSubtitles() + self.assertTrue(len(subtitles.keys()) == 0) + + def test_nosubtitles(self): + self.DL.expect_warning(u'video doesn\'t have subtitles') + self.url = 'http://www.dailymotion.com/video/x12u166_le-zapping-tele-star-du-08-aout-2013_tv' + self.DL.params['writesubtitles'] = True + self.DL.params['allsubtitles'] = True + subtitles = self.getSubtitles() + self.assertEqual(len(subtitles), 0) + + def test_multiple_langs(self): + self.DL.params['writesubtitles'] = True + langs = ['es', 'fr', 'de'] + self.DL.params['subtitleslangs'] = langs + subtitles = self.getSubtitles() + for lang in langs: + self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang) + + +class TestTedSubtitles(BaseTestSubtitles): + url = 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html' + IE = TEDIE + + def test_no_writesubtitles(self): + subtitles = self.getSubtitles() + self.assertEqual(subtitles, None) + + def test_subtitles(self): + self.DL.params['writesubtitles'] = True + subtitles = self.getSubtitles() + self.assertEqual(md5(subtitles['en']), '2154f31ff9b9f89a0aa671537559c21d') + + def test_subtitles_lang(self): + self.DL.params['writesubtitles'] = True + self.DL.params['subtitleslangs'] = ['fr'] + subtitles = self.getSubtitles() + self.assertEqual(md5(subtitles['fr']), '7616cbc6df20ec2c1204083c83871cf6') + + def test_allsubtitles(self): + self.DL.params['writesubtitles'] = True + self.DL.params['allsubtitles'] = True + subtitles = self.getSubtitles() + self.assertEqual(len(subtitles.keys()), 28) + + def test_list_subtitles(self): + self.DL.expect_warning(u'Automatic Captions not supported by this server') + self.DL.params['listsubtitles'] = True + info_dict = self.getInfoDict() + self.assertEqual(info_dict, None) + + def test_automatic_captions(self): + self.DL.expect_warning(u'Automatic Captions not supported by this server') + self.DL.params['writeautomaticsub'] = True + self.DL.params['subtitleslang'] = ['en'] + subtitles = self.getSubtitles() + self.assertTrue(len(subtitles.keys()) == 0) + + def test_multiple_langs(self): + self.DL.params['writesubtitles'] = True + langs = ['es', 'fr', 'de'] + self.DL.params['subtitleslangs'] = langs + subtitles = self.getSubtitles() + for lang in langs: + self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang) + +if __name__ == '__main__': + unittest.main() diff --git a/test/test_utils.py b/test/test_utils.py index f3fbff042..e5778cd83 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -13,17 +13,21 @@ import xml.etree.ElementTree #from youtube_dl.utils import htmlentity_transform from youtube_dl.utils import ( - timeconvert, - sanitize_filename, - unescapeHTML, - orderedSet, DateRange, - unified_strdate, + encodeFilename, find_xpath_attr, get_meta_content, - xpath_with_ns, + orderedSet, + sanitize_filename, + shell_quote, smuggle_url, + str_to_int, + timeconvert, + unescapeHTML, + unified_strdate, unsmuggle_url, + url_basename, + xpath_with_ns, ) if sys.version_info < (3, 0): @@ -170,6 +174,23 @@ class TestUtil(unittest.TestCase): self.assertEqual(res_url, url) self.assertEqual(res_data, None) + def test_shell_quote(self): + args = ['ffmpeg', '-i', encodeFilename(u'ñ€ß\'.mp4')] + self.assertEqual(shell_quote(args), u"""ffmpeg -i 'ñ€ß'"'"'.mp4'""") + + def test_str_to_int(self): + self.assertEqual(str_to_int('123,456'), 123456) + self.assertEqual(str_to_int('123.456'), 123456) + + def test_url_basename(self): + self.assertEqual(url_basename(u'http://foo.de/'), u'') + self.assertEqual(url_basename(u'http://foo.de/bar/baz'), u'baz') + self.assertEqual(url_basename(u'http://foo.de/bar/baz?x=y'), u'baz') + self.assertEqual(url_basename(u'http://foo.de/bar/baz#x=y'), u'baz') + self.assertEqual(url_basename(u'http://foo.de/bar/baz/'), u'baz') + self.assertEqual( + url_basename(u'http://media.w3.org/2010/05/sintel/trailer.mp4'), + u'trailer.mp4') if __name__ == '__main__': unittest.main() diff --git a/test/test_write_annotations.py b/test/test_write_annotations.py index 35defb895..eac53b285 100644 --- a/test/test_write_annotations.py +++ b/test/test_write_annotations.py @@ -7,8 +7,7 @@ import sys import unittest sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) -from test.helper import get_params, global_setup, try_rm -global_setup() +from test.helper import get_params, try_rm import io diff --git a/test/test_write_info_json.py b/test/test_write_info_json.py index a5b6f6972..90426a559 100644 --- a/test/test_write_info_json.py +++ b/test/test_write_info_json.py @@ -7,8 +7,7 @@ import sys import unittest sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) -from test.helper import get_params, global_setup -global_setup() +from test.helper import get_params import io @@ -31,9 +30,10 @@ params = get_params({ TEST_ID = 'BaW_jenozKc' -INFO_JSON_FILE = TEST_ID + '.mp4.info.json' +INFO_JSON_FILE = TEST_ID + '.info.json' DESCRIPTION_FILE = TEST_ID + '.mp4.description' EXPECTED_DESCRIPTION = u'''test chars: "'/\ä↭𝕐 +test URL: https://github.com/rg3/youtube-dl/issues/1892 This is a test video for youtube-dl. diff --git a/test/test_youtube_lists.py b/test/test_youtube_lists.py index 4b7a7847b..d9fe5af4e 100644 --- a/test/test_youtube_lists.py +++ b/test/test_youtube_lists.py @@ -6,8 +6,7 @@ import sys import unittest sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) -from test.helper import FakeYDL, global_setup -global_setup() +from test.helper import FakeYDL from youtube_dl.extractor import ( @@ -16,6 +15,7 @@ from youtube_dl.extractor import ( YoutubeIE, YoutubeChannelIE, YoutubeShowIE, + YoutubeTopListIE, ) @@ -27,7 +27,7 @@ class TestYoutubeLists(unittest.TestCase): def test_youtube_playlist(self): dl = FakeYDL() ie = YoutubePlaylistIE(dl) - result = ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')[0] + result = ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re') self.assertIsPlaylist(result) self.assertEqual(result['title'], 'ytdl test PL') ytie_results = [YoutubeIE()._extract_id(url['url']) for url in result['entries']] @@ -44,13 +44,13 @@ class TestYoutubeLists(unittest.TestCase): def test_issue_673(self): dl = FakeYDL() ie = YoutubePlaylistIE(dl) - result = ie.extract('PLBB231211A4F62143')[0] + result = ie.extract('PLBB231211A4F62143') self.assertTrue(len(result['entries']) > 25) def test_youtube_playlist_long(self): dl = FakeYDL() ie = YoutubePlaylistIE(dl) - result = ie.extract('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')[0] + result = ie.extract('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q') self.assertIsPlaylist(result) self.assertTrue(len(result['entries']) >= 799) @@ -58,7 +58,7 @@ class TestYoutubeLists(unittest.TestCase): #651 dl = FakeYDL() ie = YoutubePlaylistIE(dl) - result = ie.extract('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')[0] + result = ie.extract('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC') ytie_results = [YoutubeIE()._extract_id(url['url']) for url in result['entries']] self.assertFalse('pElCt5oNDuI' in ytie_results) self.assertFalse('KdPEApIVdWM' in ytie_results) @@ -66,7 +66,7 @@ class TestYoutubeLists(unittest.TestCase): def test_youtube_playlist_empty(self): dl = FakeYDL() ie = YoutubePlaylistIE(dl) - result = ie.extract('https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx')[0] + result = ie.extract('https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx') self.assertIsPlaylist(result) self.assertEqual(len(result['entries']), 0) @@ -74,7 +74,7 @@ class TestYoutubeLists(unittest.TestCase): dl = FakeYDL() ie = YoutubePlaylistIE(dl) # TODO find a > 100 (paginating?) videos course - result = ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')[0] + result = ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8') entries = result['entries'] self.assertEqual(YoutubeIE()._extract_id(entries[0]['url']), 'j9WZyLZCBzs') self.assertEqual(len(entries), 25) @@ -84,22 +84,22 @@ class TestYoutubeLists(unittest.TestCase): dl = FakeYDL() ie = YoutubeChannelIE(dl) #test paginated channel - result = ie.extract('https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w')[0] + result = ie.extract('https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w') self.assertTrue(len(result['entries']) > 90) #test autogenerated channel - result = ie.extract('https://www.youtube.com/channel/HCtnHdj3df7iM/videos')[0] + result = ie.extract('https://www.youtube.com/channel/HCtnHdj3df7iM/videos') self.assertTrue(len(result['entries']) >= 18) def test_youtube_user(self): dl = FakeYDL() ie = YoutubeUserIE(dl) - result = ie.extract('https://www.youtube.com/user/TheLinuxFoundation')[0] + result = ie.extract('https://www.youtube.com/user/TheLinuxFoundation') self.assertTrue(len(result['entries']) >= 320) def test_youtube_safe_search(self): dl = FakeYDL() ie = YoutubePlaylistIE(dl) - result = ie.extract('PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl')[0] + result = ie.extract('PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl') self.assertEqual(len(result['entries']), 2) def test_youtube_show(self): @@ -108,5 +108,21 @@ class TestYoutubeLists(unittest.TestCase): result = ie.extract('http://www.youtube.com/show/airdisasters') self.assertTrue(len(result) >= 3) + def test_youtube_mix(self): + dl = FakeYDL() + ie = YoutubePlaylistIE(dl) + result = ie.extract('http://www.youtube.com/watch?v=lLJf9qJHR3E&list=RDrjFaenf1T-Y') + entries = result['entries'] + self.assertTrue(len(entries) >= 20) + original_video = entries[0] + self.assertEqual(original_video['id'], 'rjFaenf1T-Y') + + def test_youtube_toplist(self): + dl = FakeYDL() + ie = YoutubeTopListIE(dl) + result = ie.extract('yttoplist:music:Top Tracks') + entries = result['entries'] + self.assertTrue(len(entries) >= 5) + if __name__ == '__main__': unittest.main() diff --git a/test/test_youtube_signature.py b/test/test_youtube_signature.py index 5e1ff5eb0..056700614 100644 --- a/test/test_youtube_signature.py +++ b/test/test_youtube_signature.py @@ -6,9 +6,6 @@ import sys import unittest sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) -from test.helper import global_setup -global_setup() - import io import re diff --git a/test/test_youtube_subtitles.py b/test/test_youtube_subtitles.py deleted file mode 100644 index 00430a338..000000000 --- a/test/test_youtube_subtitles.py +++ /dev/null @@ -1,95 +0,0 @@ -#!/usr/bin/env python - -# Allow direct execution -import os -import sys -import unittest -sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) - -from test.helper import FakeYDL, global_setup, md5 -global_setup() - - -from youtube_dl.extractor import YoutubeIE - - -class TestYoutubeSubtitles(unittest.TestCase): - def setUp(self): - self.DL = FakeYDL() - self.url = 'QRS8MkLhQmM' - - def getInfoDict(self): - IE = YoutubeIE(self.DL) - info_dict = IE.extract(self.url) - return info_dict - - def getSubtitles(self): - info_dict = self.getInfoDict() - return info_dict[0]['subtitles'] - - def test_youtube_no_writesubtitles(self): - self.DL.params['writesubtitles'] = False - subtitles = self.getSubtitles() - self.assertEqual(subtitles, None) - - def test_youtube_subtitles(self): - self.DL.params['writesubtitles'] = True - subtitles = self.getSubtitles() - self.assertEqual(md5(subtitles['en']), '4cd9278a35ba2305f47354ee13472260') - - def test_youtube_subtitles_lang(self): - self.DL.params['writesubtitles'] = True - self.DL.params['subtitleslangs'] = ['it'] - subtitles = self.getSubtitles() - self.assertEqual(md5(subtitles['it']), '164a51f16f260476a05b50fe4c2f161d') - - def test_youtube_allsubtitles(self): - self.DL.params['writesubtitles'] = True - self.DL.params['allsubtitles'] = True - subtitles = self.getSubtitles() - self.assertEqual(len(subtitles.keys()), 13) - - def test_youtube_subtitles_sbv_format(self): - self.DL.params['writesubtitles'] = True - self.DL.params['subtitlesformat'] = 'sbv' - subtitles = self.getSubtitles() - self.assertEqual(md5(subtitles['en']), '13aeaa0c245a8bed9a451cb643e3ad8b') - - def test_youtube_subtitles_vtt_format(self): - self.DL.params['writesubtitles'] = True - self.DL.params['subtitlesformat'] = 'vtt' - subtitles = self.getSubtitles() - self.assertEqual(md5(subtitles['en']), '356cdc577fde0c6783b9b822e7206ff7') - - def test_youtube_list_subtitles(self): - self.DL.expect_warning(u'Video doesn\'t have automatic captions') - self.DL.params['listsubtitles'] = True - info_dict = self.getInfoDict() - self.assertEqual(info_dict, None) - - def test_youtube_automatic_captions(self): - self.url = '8YoUxe5ncPo' - self.DL.params['writeautomaticsub'] = True - self.DL.params['subtitleslangs'] = ['it'] - subtitles = self.getSubtitles() - self.assertTrue(subtitles['it'] is not None) - - def test_youtube_nosubtitles(self): - self.DL.expect_warning(u'video doesn\'t have subtitles') - self.url = 'sAjKT8FhjI8' - self.DL.params['writesubtitles'] = True - self.DL.params['allsubtitles'] = True - subtitles = self.getSubtitles() - self.assertEqual(len(subtitles), 0) - - def test_youtube_multiple_langs(self): - self.url = 'QRS8MkLhQmM' - self.DL.params['writesubtitles'] = True - langs = ['it', 'fr', 'de'] - self.DL.params['subtitleslangs'] = langs - subtitles = self.getSubtitles() - for lang in langs: - self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang) - -if __name__ == '__main__': - unittest.main() diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py index 8ecabab1a..47124932f 100644 --- a/youtube_dl/FileDownloader.py +++ b/youtube_dl/FileDownloader.py @@ -1,15 +1,19 @@ -import math import os import re import subprocess import sys import time -import traceback -if os.name == 'nt': - import ctypes - -from .utils import * +from .utils import ( + compat_urllib_error, + compat_urllib_request, + ContentTooShortError, + determine_ext, + encodeFilename, + format_bytes, + sanitize_open, + timeconvert, +) class FileDownloader(object): @@ -49,20 +53,6 @@ class FileDownloader(object): self._progress_hooks = [] self.params = params - @staticmethod - def format_bytes(bytes): - if bytes is None: - return 'N/A' - if type(bytes) is str: - bytes = float(bytes) - if bytes == 0.0: - exponent = 0 - else: - exponent = int(math.log(bytes, 1024.0)) - suffix = ['B','KiB','MiB','GiB','TiB','PiB','EiB','ZiB','YiB'][exponent] - converted = float(bytes) / float(1024 ** exponent) - return '%.2f%s' % (converted, suffix) - @staticmethod def format_seconds(seconds): (mins, secs) = divmod(seconds, 60) @@ -113,7 +103,7 @@ class FileDownloader(object): def format_speed(speed): if speed is None: return '%10s' % '---b/s' - return '%10s' % ('%s/s' % FileDownloader.format_bytes(speed)) + return '%10s' % ('%s/s' % format_bytes(speed)) @staticmethod def best_block_size(elapsed_time, bytes): @@ -144,16 +134,8 @@ class FileDownloader(object): def to_stderr(self, message): self.ydl.to_screen(message) - def to_cons_title(self, message): - """Set console/terminal window title to message.""" - if not self.params.get('consoletitle', False): - return - if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow(): - # c_wchar_p() might not be necessary if `message` is - # already of type unicode() - ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) - elif 'TERM' in os.environ: - self.to_screen('\033]0;%s\007' % message, skip_eol=True) + def to_console_title(self, message): + self.ydl.to_console_title(message) def trouble(self, *args, **kargs): self.ydl.trouble(*args, **kargs) @@ -194,7 +176,7 @@ class FileDownloader(object): if old_filename == new_filename: return os.rename(encodeFilename(old_filename), encodeFilename(new_filename)) - except (IOError, OSError) as err: + except (IOError, OSError): self.report_error(u'unable to rename file') def try_utime(self, filename, last_modified_hdr): @@ -222,22 +204,59 @@ class FileDownloader(object): """Report destination filename.""" self.to_screen(u'[download] Destination: ' + filename) + def _report_progress_status(self, msg, is_last_line=False): + fullmsg = u'[download] ' + msg + if self.params.get('progress_with_newline', False): + self.to_screen(fullmsg) + else: + if os.name == 'nt': + prev_len = getattr(self, '_report_progress_prev_line_length', + 0) + if prev_len > len(fullmsg): + fullmsg += u' ' * (prev_len - len(fullmsg)) + self._report_progress_prev_line_length = len(fullmsg) + clear_line = u'\r' + else: + clear_line = (u'\r\x1b[K' if sys.stderr.isatty() else u'\r') + self.to_screen(clear_line + fullmsg, skip_eol=not is_last_line) + self.to_console_title(u'youtube-dl ' + msg) + def report_progress(self, percent, data_len_str, speed, eta): """Report download progress.""" if self.params.get('noprogress', False): return - clear_line = (u'\x1b[K' if sys.stderr.isatty() and os.name != 'nt' else u'') - eta_str = self.format_eta(eta) - percent_str = self.format_percent(percent) + if eta is not None: + eta_str = self.format_eta(eta) + else: + eta_str = 'Unknown ETA' + if percent is not None: + percent_str = self.format_percent(percent) + else: + percent_str = 'Unknown %' speed_str = self.format_speed(speed) - if self.params.get('progress_with_newline', False): - self.to_screen(u'[download] %s of %s at %s ETA %s' % - (percent_str, data_len_str, speed_str, eta_str)) + + msg = (u'%s of %s at %s ETA %s' % + (percent_str, data_len_str, speed_str, eta_str)) + self._report_progress_status(msg) + + def report_progress_live_stream(self, downloaded_data_len, speed, elapsed): + if self.params.get('noprogress', False): + return + downloaded_str = format_bytes(downloaded_data_len) + speed_str = self.format_speed(speed) + elapsed_str = FileDownloader.format_seconds(elapsed) + msg = u'%s at %s (%s)' % (downloaded_str, speed_str, elapsed_str) + self._report_progress_status(msg) + + def report_finish(self, data_len_str, tot_time): + """Report download finished.""" + if self.params.get('noprogress', False): + self.to_screen(u'[download] Download completed') else: - self.to_screen(u'\r%s[download] %s of %s at %s ETA %s' % - (clear_line, percent_str, data_len_str, speed_str, eta_str), skip_eol=True) - self.to_cons_title(u'youtube-dl - %s of %s at %s ETA %s' % - (percent_str.strip(), data_len_str.strip(), speed_str.strip(), eta_str.strip())) + self._report_progress_status( + (u'100%% of %s in %s' % + (data_len_str, self.format_seconds(tot_time))), + is_last_line=True) def report_resuming_byte(self, resume_len): """Report attempt to resume at given byte.""" @@ -251,23 +270,85 @@ class FileDownloader(object): """Report file has already been fully downloaded.""" try: self.to_screen(u'[download] %s has already been downloaded' % file_name) - except (UnicodeEncodeError) as err: + except UnicodeEncodeError: self.to_screen(u'[download] The file has already been downloaded') def report_unable_to_resume(self): """Report it was impossible to resume download.""" self.to_screen(u'[download] Unable to resume') - def report_finish(self, data_len_str, tot_time): - """Report download finished.""" - if self.params.get('noprogress', False): - self.to_screen(u'[download] Download completed') - else: - clear_line = (u'\x1b[K' if sys.stderr.isatty() and os.name != 'nt' else u'') - self.to_screen(u'\r%s[download] 100%% of %s in %s' % - (clear_line, data_len_str, self.format_seconds(tot_time))) + def _download_with_rtmpdump(self, filename, url, player_url, page_url, play_path, tc_url, live, conn): + def run_rtmpdump(args): + start = time.time() + resume_percent = None + resume_downloaded_data_len = None + proc = subprocess.Popen(args, stderr=subprocess.PIPE) + cursor_in_new_line = True + proc_stderr_closed = False + while not proc_stderr_closed: + # read line from stderr + line = u'' + while True: + char = proc.stderr.read(1) + if not char: + proc_stderr_closed = True + break + if char in [b'\r', b'\n']: + break + line += char.decode('ascii', 'replace') + if not line: + # proc_stderr_closed is True + continue + mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec \(([0-9]{1,2}\.[0-9])%\)', line) + if mobj: + downloaded_data_len = int(float(mobj.group(1))*1024) + percent = float(mobj.group(2)) + if not resume_percent: + resume_percent = percent + resume_downloaded_data_len = downloaded_data_len + eta = self.calc_eta(start, time.time(), 100-resume_percent, percent-resume_percent) + speed = self.calc_speed(start, time.time(), downloaded_data_len-resume_downloaded_data_len) + data_len = None + if percent > 0: + data_len = int(downloaded_data_len * 100 / percent) + data_len_str = u'~' + format_bytes(data_len) + self.report_progress(percent, data_len_str, speed, eta) + cursor_in_new_line = False + self._hook_progress({ + 'downloaded_bytes': downloaded_data_len, + 'total_bytes': data_len, + 'tmpfilename': tmpfilename, + 'filename': filename, + 'status': 'downloading', + 'eta': eta, + 'speed': speed, + }) + else: + # no percent for live streams + mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec', line) + if mobj: + downloaded_data_len = int(float(mobj.group(1))*1024) + time_now = time.time() + speed = self.calc_speed(start, time_now, downloaded_data_len) + self.report_progress_live_stream(downloaded_data_len, speed, time_now - start) + cursor_in_new_line = False + self._hook_progress({ + 'downloaded_bytes': downloaded_data_len, + 'tmpfilename': tmpfilename, + 'filename': filename, + 'status': 'downloading', + 'speed': speed, + }) + elif self.params.get('verbose', False): + if not cursor_in_new_line: + self.to_screen(u'') + cursor_in_new_line = True + self.to_screen(u'[rtmpdump] '+line) + proc.wait() + if not cursor_in_new_line: + self.to_screen(u'') + return proc.returncode - def _download_with_rtmpdump(self, filename, url, player_url, page_url, play_path, tc_url): self.report_destination(filename) tmpfilename = self.temp_name(filename) test = self.params.get('test', False) @@ -278,12 +359,11 @@ class FileDownloader(object): except (OSError, IOError): self.report_error(u'RTMP download detected but "rtmpdump" could not be run') return False - verbosity_option = '--verbose' if self.params.get('verbose', False) else '--quiet' # Download using rtmpdump. rtmpdump returns exit code 2 when # the connection was interrumpted and resuming appears to be # possible. This is part of rtmpdump's normal usage, AFAIK. - basic_args = ['rtmpdump', verbosity_option, '-r', url, '-o', tmpfilename] + basic_args = ['rtmpdump', '--verbose', '-r', url, '-o', tmpfilename] if player_url is not None: basic_args += ['--swfVfy', player_url] if page_url is not None: @@ -294,31 +374,53 @@ class FileDownloader(object): basic_args += ['--tcUrl', url] if test: basic_args += ['--stop', '1'] + if live: + basic_args += ['--live'] + if conn: + basic_args += ['--conn', conn] args = basic_args + [[], ['--resume', '--skip', '1']][self.params.get('continuedl', False)] + + if sys.platform == 'win32' and sys.version_info < (3, 0): + # Windows subprocess module does not actually support Unicode + # on Python 2.x + # See http://stackoverflow.com/a/9951851/35070 + subprocess_encoding = sys.getfilesystemencoding() + args = [a.encode(subprocess_encoding, 'ignore') for a in args] + else: + subprocess_encoding = None + if self.params.get('verbose', False): + if subprocess_encoding: + str_args = [ + a.decode(subprocess_encoding) if isinstance(a, bytes) else a + for a in args] + else: + str_args = args try: import pipes - shell_quote = lambda args: ' '.join(map(pipes.quote, args)) + shell_quote = lambda args: ' '.join(map(pipes.quote, str_args)) except ImportError: shell_quote = repr - self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(args)) - retval = subprocess.call(args) + self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(str_args)) + + retval = run_rtmpdump(args) + while (retval == 2 or retval == 1) and not test: prevsize = os.path.getsize(encodeFilename(tmpfilename)) - self.to_screen(u'\r[rtmpdump] %s bytes' % prevsize, skip_eol=True) + self.to_screen(u'[rtmpdump] %s bytes' % prevsize) time.sleep(5.0) # This seems to be needed - retval = subprocess.call(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1]) + retval = run_rtmpdump(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1]) cursize = os.path.getsize(encodeFilename(tmpfilename)) if prevsize == cursize and retval == 1: break # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those if prevsize == cursize and retval == 2 and cursize > 1024: - self.to_screen(u'\r[rtmpdump] Could not download the whole video. This can happen for some advertisements.') + self.to_screen(u'[rtmpdump] Could not download the whole video. This can happen for some advertisements.') retval = 0 break if retval == 0 or (test and retval == 2): fsize = os.path.getsize(encodeFilename(tmpfilename)) - self.to_screen(u'\r[rtmpdump] %s bytes' % fsize) + self.to_screen(u'[rtmpdump] %s bytes' % fsize) self.try_rename(tmpfilename, filename) self._hook_progress({ 'downloaded_bytes': fsize, @@ -366,15 +468,20 @@ class FileDownloader(object): self.report_destination(filename) tmpfilename = self.temp_name(filename) - args = ['ffmpeg', '-y', '-i', url, '-f', 'mp4', tmpfilename] - # Check for ffmpeg first - try: - subprocess.call(['ffmpeg', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT) - except (OSError, IOError): - self.report_error(u'm3u8 download detected but "%s" could not be run' % args[0] ) - return False + args = ['-y', '-i', url, '-f', 'mp4', '-c', 'copy', + '-bsf:a', 'aac_adtstoasc', tmpfilename] - retval = subprocess.call(args) + for program in ['avconv', 'ffmpeg']: + try: + subprocess.call([program, '-version'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT) + break + except (OSError, IOError): + pass + else: + self.report_error(u'm3u8 download detected but ffmpeg or avconv could not be found') + cmd = [program] + args + + retval = subprocess.call(cmd) if retval == 0: fsize = os.path.getsize(encodeFilename(tmpfilename)) self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize)) @@ -411,7 +518,9 @@ class FileDownloader(object): info_dict.get('player_url', None), info_dict.get('page_url', None), info_dict.get('play_path', None), - info_dict.get('tc_url', None)) + info_dict.get('tc_url', None), + info_dict.get('rtmp_live', False), + info_dict.get('rtmp_conn', None)) # Attempt to download using mplayer if url.startswith('mms') or url.startswith('rtsp'): @@ -515,7 +624,7 @@ class FileDownloader(object): self.to_screen(u'\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len, max_data_len)) return False - data_len_str = self.format_bytes(data_len) + data_len_str = format_bytes(data_len) byte_counter = 0 + resume_len block_size = self.params.get('buffersize', 1024) start = time.time() @@ -550,12 +659,11 @@ class FileDownloader(object): # Progress message speed = self.calc_speed(start, time.time(), byte_counter - resume_len) if data_len is None: - self.report_progress('Unknown %', data_len_str, speed_str, 'Unknown ETA') - eta = None + eta = percent = None else: percent = self.calc_percent(byte_counter, data_len) eta = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len) - self.report_progress(percent, data_len_str, speed, eta) + self.report_progress(percent, data_len_str, speed, eta) self._hook_progress({ 'downloaded_bytes': byte_counter, diff --git a/youtube_dl/PostProcessor.py b/youtube_dl/PostProcessor.py index 13b56ede5..69aedf87a 100644 --- a/youtube_dl/PostProcessor.py +++ b/youtube_dl/PostProcessor.py @@ -501,7 +501,7 @@ class FFmpegMetadataPP(FFmpegPostProcessor): options = ['-c', 'copy'] for (name, value) in metadata.items(): - options.extend(['-metadata', '%s="%s"' % (name, value)]) + options.extend(['-metadata', '%s=%s' % (name, value)]) options.extend(['-f', ext]) self._downloader.to_screen(u'[ffmpeg] Adding metadata to \'%s\'' % filename) diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 7f73ea360..2a078adfb 100644 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -3,19 +3,58 @@ from __future__ import absolute_import +import collections import errno import io +import json import os +import platform import re import shutil +import subprocess import socket import sys import time import traceback -from .utils import * +if os.name == 'nt': + import ctypes + +from .utils import ( + compat_cookiejar, + compat_http_client, + compat_str, + compat_urllib_error, + compat_urllib_request, + ContentTooShortError, + date_from_str, + DateRange, + determine_ext, + DownloadError, + encodeFilename, + ExtractorError, + format_bytes, + formatSeconds, + get_term_width, + locked_file, + make_HTTPS_handler, + MaxDownloadsReached, + PostProcessingError, + platform_name, + preferredencoding, + SameFileError, + sanitize_filename, + subtitles_filename, + takewhile_inclusive, + UnavailableVideoError, + url_basename, + write_json_file, + write_string, + YoutubeDLHandler, +) from .extractor import get_info_extractor, gen_extractors from .FileDownloader import FileDownloader +from .version import __version__ class YoutubeDL(object): @@ -57,6 +96,8 @@ class YoutubeDL(object): forcethumbnail: Force printing thumbnail URL. forcedescription: Force printing description. forcefilename: Force printing final filename. + forceduration: Force printing duration. + forcejson: Force printing info_dict as JSON. simulate: Do not download the video files. format: Video format code. format_limit: Highest quality format to try. @@ -68,6 +109,7 @@ class YoutubeDL(object): playlistend: Playlist item to end at. matchtitle: Download only matching titles. rejecttitle: Reject downloads for matching titles. + logger: Log messages to a logging.Logger instance. logtostderr: Log messages to stderr instead of stdout. writedescription: Write the video description to a .description file writeinfojson: Write the video description to a .info.json file @@ -88,9 +130,24 @@ class YoutubeDL(object): noplaylist: Download single video instead of a playlist if in doubt. age_limit: An integer representing the user's age in years. Unsuitable videos for the given age are skipped. - downloadarchive: File name of a file where all downloads are recorded. + min_views: An integer representing the minimum view count the video + must have in order to not be skipped. + Videos without view count information are always + downloaded. None for no limit. + max_views: An integer representing the maximum view count. + Videos that are more popular than that are not + downloaded. + Videos without view count information are always + downloaded. None for no limit. + download_archive: File name of a file where all downloads are recorded. Videos already present in the file are not downloaded again. + cookiefile: File name where cookies should be read from and dumped to. + nocheckcertificate:Do not verify SSL certificates + proxy: URL of the proxy server to use + socket_timeout: Time to wait for unresponsive hosts, in seconds + bidi_workaround: Work around buggy terminals without bidirectional text + support, using fridibi The following parameters are not used by YoutubeDL itself, they are used by the FileDownloader: @@ -105,7 +162,7 @@ class YoutubeDL(object): _num_downloads = None _screen_file = None - def __init__(self, params): + def __init__(self, params=None): """Create a FileDownloader object with the given options.""" self._ies = [] self._ies_instances = {} @@ -114,6 +171,29 @@ class YoutubeDL(object): self._download_retcode = 0 self._num_downloads = 0 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] + self._err_file = sys.stderr + self.params = {} if params is None else params + + if params.get('bidi_workaround', False): + try: + import pty + master, slave = pty.openpty() + width = get_term_width() + if width is None: + width_args = [] + else: + width_args = ['-w', str(width)] + self._fribidi = subprocess.Popen( + ['fribidi', '-c', 'UTF-8'] + width_args, + stdin=subprocess.PIPE, + stdout=slave, + stderr=self._err_file) + self._fribidi_channel = os.fdopen(master, 'rb') + except OSError as ose: + if ose.errno == 2: + self.report_warning(u'Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.') + else: + raise if (sys.version_info >= (3,) and sys.platform != 'win32' and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] @@ -123,14 +203,15 @@ class YoutubeDL(object): u'Assuming --restrict-filenames since file system encoding ' u'cannot encode all charactes. ' u'Set the LC_ALL environment variable to fix this.') - params['restrictfilenames'] = True + self.params['restrictfilenames'] = True - self.params = params self.fd = FileDownloader(self, self.params) - if '%(stitle)s' in self.params['outtmpl']: + if '%(stitle)s' in self.params.get('outtmpl', ''): self.report_warning(u'%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.') + self._setup_opener() + def add_info_extractor(self, ie): """Add an InfoExtractor object to the end of the list.""" self._ies.append(ie) @@ -161,24 +242,76 @@ class YoutubeDL(object): self._pps.append(pp) pp.set_downloader(self) + def _bidi_workaround(self, message): + if not hasattr(self, '_fribidi_channel'): + return message + + assert type(message) == type(u'') + line_count = message.count(u'\n') + 1 + self._fribidi.stdin.write((message + u'\n').encode('utf-8')) + self._fribidi.stdin.flush() + res = u''.join(self._fribidi_channel.readline().decode('utf-8') + for _ in range(line_count)) + return res[:-len(u'\n')] + def to_screen(self, message, skip_eol=False): """Print message to stdout if not in quiet mode.""" - if not self.params.get('quiet', False): + return self.to_stdout(message, skip_eol, check_quiet=True) + + def to_stdout(self, message, skip_eol=False, check_quiet=False): + """Print message to stdout if not in quiet mode.""" + if self.params.get('logger'): + self.params['logger'].debug(message) + elif not check_quiet or not self.params.get('quiet', False): + message = self._bidi_workaround(message) terminator = [u'\n', u''][skip_eol] output = message + terminator + write_string(output, self._screen_file) def to_stderr(self, message): """Print message to stderr.""" assert type(message) == type(u'') - output = message + u'\n' - if 'b' in getattr(self._screen_file, 'mode', '') or sys.version_info[0] < 3: # Python 2 lies about the mode of sys.stdout/sys.stderr - output = output.encode(preferredencoding()) - sys.stderr.write(output) + if self.params.get('logger'): + self.params['logger'].error(message) + else: + message = self._bidi_workaround(message) + output = message + u'\n' + write_string(output, self._err_file) + + def to_console_title(self, message): + if not self.params.get('consoletitle', False): + return + if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow(): + # c_wchar_p() might not be necessary if `message` is + # already of type unicode() + ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) + elif 'TERM' in os.environ: + write_string(u'\033]0;%s\007' % message, self._screen_file) + + def save_console_title(self): + if not self.params.get('consoletitle', False): + return + if 'TERM' in os.environ: + # Save the title on stack + write_string(u'\033[22;0t', self._screen_file) + + def restore_console_title(self): + if not self.params.get('consoletitle', False): + return + if 'TERM' in os.environ: + # Restore the title from stack + write_string(u'\033[23;0t', self._screen_file) + + def __enter__(self): + self.save_console_title() + return self - def fixed_template(self): - """Checks if the output template is fixed.""" - return (re.search(u'(?u)%\\(.+?\\)s', self.params['outtmpl']) is None) + def __exit__(self, *args): + self.restore_console_title() + + if self.params.get('cookiefile') is not None: + self.cookiejar.save() def trouble(self, message=None, tb=None): """Determine action to take when a download problem appears. @@ -215,7 +348,7 @@ class YoutubeDL(object): Print the message to stderr, it will be prefixed with 'WARNING:' If stderr is a tty file the 'WARNING:' will be colored ''' - if sys.stderr.isatty() and os.name != 'nt': + if self._err_file.isatty() and os.name != 'nt': _msg_header = u'\033[0;33mWARNING:\033[0m' else: _msg_header = u'WARNING:' @@ -227,34 +360,18 @@ class YoutubeDL(object): Do the same as trouble, but prefixes the message with 'ERROR:', colored in red if stderr is a tty file. ''' - if sys.stderr.isatty() and os.name != 'nt': + if self._err_file.isatty() and os.name != 'nt': _msg_header = u'\033[0;31mERROR:\033[0m' else: _msg_header = u'ERROR:' error_message = u'%s %s' % (_msg_header, message) self.trouble(error_message, tb) - def report_writedescription(self, descfn): - """ Report that the description file is being written """ - self.to_screen(u'[info] Writing video description to: ' + descfn) - - def report_writesubtitles(self, sub_filename): - """ Report that the subtitles file is being written """ - self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename) - - def report_writeinfojson(self, infofn): - """ Report that the metadata file has been written """ - self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn) - - def report_writeannotations(self, annofn): - """ Report that the annotations file has been written. """ - self.to_screen(u'[info] Writing video annotations to: ' + annofn) - def report_file_already_downloaded(self, file_name): """Report file has already been fully downloaded.""" try: self.to_screen(u'[download] %s has already been downloaded' % file_name) - except (UnicodeEncodeError) as err: + except UnicodeEncodeError: self.to_screen(u'[download] The file has already been downloaded') def increment_downloads(self): @@ -276,18 +393,17 @@ class YoutubeDL(object): template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index'] sanitize = lambda k, v: sanitize_filename( - u'NA' if v is None else compat_str(v), + compat_str(v), restricted=self.params.get('restrictfilenames'), is_id=(k == u'id')) template_dict = dict((k, sanitize(k, v)) - for k, v in template_dict.items()) + for k, v in template_dict.items() + if v is not None) + template_dict = collections.defaultdict(lambda: u'NA', template_dict) tmpl = os.path.expanduser(self.params['outtmpl']) filename = tmpl % template_dict return filename - except KeyError as err: - self.report_error(u'Erroneous output template') - return None except ValueError as err: self.report_error(u'Error in output template: ' + str(err) + u' (encoding: ' + repr(preferredencoding()) + ')') return None @@ -295,30 +411,47 @@ class YoutubeDL(object): def _match_entry(self, info_dict): """ Returns None iff the file should be downloaded """ - title = info_dict['title'] - matchtitle = self.params.get('matchtitle', False) - if matchtitle: - if not re.search(matchtitle, title, re.IGNORECASE): - return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"' - rejecttitle = self.params.get('rejecttitle', False) - if rejecttitle: - if re.search(rejecttitle, title, re.IGNORECASE): - return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"' + video_title = info_dict.get('title', info_dict.get('id', u'video')) + if 'title' in info_dict: + # This can happen when we're just evaluating the playlist + title = info_dict['title'] + matchtitle = self.params.get('matchtitle', False) + if matchtitle: + if not re.search(matchtitle, title, re.IGNORECASE): + return u'"' + title + '" title did not match pattern "' + matchtitle + '"' + rejecttitle = self.params.get('rejecttitle', False) + if rejecttitle: + if re.search(rejecttitle, title, re.IGNORECASE): + return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"' date = info_dict.get('upload_date', None) if date is not None: dateRange = self.params.get('daterange', DateRange()) if date not in dateRange: - return u'[download] %s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange) + return u'%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange) + view_count = info_dict.get('view_count', None) + if view_count is not None: + min_views = self.params.get('min_views') + if min_views is not None and view_count < min_views: + return u'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views) + max_views = self.params.get('max_views') + if max_views is not None and view_count > max_views: + return u'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views) age_limit = self.params.get('age_limit') if age_limit is not None: if age_limit < info_dict.get('age_limit', 0): return u'Skipping "' + title + '" because it is age restricted' if self.in_download_archive(info_dict): - return (u'%(title)s has already been recorded in archive' - % info_dict) + return u'%s has already been recorded in archive' % video_title return None - def extract_info(self, url, download=True, ie_key=None, extra_info={}): + @staticmethod + def add_extra_info(info_dict, extra_info): + '''Set the keys from extra_info in info dict if they are missing''' + for key, value in extra_info.items(): + info_dict.setdefault(key, value) + + def extract_info(self, url, download=True, ie_key=None, extra_info={}, + process=True): ''' Returns a list with a dictionary for each video we find. If 'download', also downloads the videos. @@ -344,17 +477,21 @@ class YoutubeDL(object): break if isinstance(ie_result, list): # Backwards compatibility: old IE result format - for result in ie_result: - result.update(extra_info) ie_result = { '_type': 'compat_list', 'entries': ie_result, } + self.add_extra_info(ie_result, + { + 'extractor': ie.IE_NAME, + 'webpage_url': url, + 'webpage_url_basename': url_basename(url), + 'extractor_key': ie.ie_key(), + }) + if process: + return self.process_ie_result(ie_result, download, extra_info) else: - ie_result.update(extra_info) - if 'extractor' not in ie_result: - ie_result['extractor'] = ie.IE_NAME - return self.process_ie_result(ie_result, download=download) + return ie_result except ExtractorError as de: # An error we somewhat expected self.report_error(compat_str(de), de.format_traceback()) break @@ -378,8 +515,8 @@ class YoutubeDL(object): result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system if result_type == 'video': - ie_result.update(extra_info) - return self.process_video_result(ie_result) + self.add_extra_info(ie_result, extra_info) + return self.process_video_result(ie_result, download=download) elif result_type == 'url': # We have to add extra_info to the results because it may be # contained in a playlist @@ -387,6 +524,32 @@ class YoutubeDL(object): download, ie_key=ie_result.get('ie_key'), extra_info=extra_info) + elif result_type == 'url_transparent': + # Use the information from the embedding page + info = self.extract_info( + ie_result['url'], ie_key=ie_result.get('ie_key'), + extra_info=extra_info, download=False, process=False) + + def make_result(embedded_info): + new_result = ie_result.copy() + for f in ('_type', 'url', 'ext', 'player_url', 'formats', + 'entries', 'urlhandle', 'ie_key', 'duration', + 'subtitles', 'annotations', 'format', + 'thumbnail', 'thumbnails'): + if f in new_result: + del new_result[f] + if f in embedded_info: + new_result[f] = embedded_info[f] + return new_result + new_result = make_result(info) + + assert new_result.get('_type') != 'url_transparent' + if new_result.get('_type') == 'compat_list': + new_result['entries'] = [ + make_result(e) for e in new_result['entries']] + + return self.process_ie_result( + new_result, download=download, extra_info=extra_info) elif result_type == 'playlist': # We process each entry in the playlist playlist = ie_result.get('title', None) or ie_result.get('id', None) @@ -396,16 +559,16 @@ class YoutubeDL(object): n_all_entries = len(ie_result['entries']) playliststart = self.params.get('playliststart', 1) - 1 - playlistend = self.params.get('playlistend', -1) - + playlistend = self.params.get('playlistend', None) + # For backwards compatibility, interpret -1 as whole list if playlistend == -1: - entries = ie_result['entries'][playliststart:] - else: - entries = ie_result['entries'][playliststart:playlistend] + playlistend = None + entries = ie_result['entries'][playliststart:playlistend] n_entries = len(entries) - self.to_screen(u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" % + self.to_screen( + u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" % (ie_result['extractor'], playlist, n_all_entries, n_entries)) for i, entry in enumerate(entries, 1): @@ -413,12 +576,17 @@ class YoutubeDL(object): extra = { 'playlist': playlist, 'playlist_index': i + playliststart, + 'extractor': ie_result['extractor'], + 'webpage_url': ie_result['webpage_url'], + 'webpage_url_basename': url_basename(ie_result['webpage_url']), + 'extractor_key': ie_result['extractor_key'], } - if not 'extractor' in entry: - # We set the extractor, if it's an url it will be set then to - # the new extractor, but if it's already a video we must make - # sure it's present: see issue #877 - entry['extractor'] = ie_result['extractor'] + + reason = self._match_entry(entry) + if reason is not None: + self.to_screen(u'[download] ' + reason) + continue + entry_result = self.process_ie_result(entry, download=download, extra_info=extra) @@ -427,10 +595,16 @@ class YoutubeDL(object): return ie_result elif result_type == 'compat_list': def _fixup(r): - r.setdefault('extractor', ie_result['extractor']) + self.add_extra_info(r, + { + 'extractor': ie_result['extractor'], + 'webpage_url': ie_result['webpage_url'], + 'webpage_url_basename': url_basename(ie_result['webpage_url']), + 'extractor_key': ie_result['extractor_key'], + }) return r ie_result['entries'] = [ - self.process_ie_result(_fixup(r), download=download) + self.process_ie_result(_fixup(r), download, extra_info) for r in ie_result['entries'] ] return ie_result @@ -569,20 +743,25 @@ class YoutubeDL(object): # Forced printings if self.params.get('forcetitle', False): - compat_print(info_dict['title']) + self.to_stdout(info_dict['fulltitle']) if self.params.get('forceid', False): - compat_print(info_dict['id']) + self.to_stdout(info_dict['id']) if self.params.get('forceurl', False): # For RTMP URLs, also include the playpath - compat_print(info_dict['url'] + info_dict.get('play_path', u'')) + self.to_stdout(info_dict['url'] + info_dict.get('play_path', u'')) if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None: - compat_print(info_dict['thumbnail']) + self.to_stdout(info_dict['thumbnail']) if self.params.get('forcedescription', False) and info_dict.get('description') is not None: - compat_print(info_dict['description']) + self.to_stdout(info_dict['description']) if self.params.get('forcefilename', False) and filename is not None: - compat_print(filename) + self.to_stdout(filename) + if self.params.get('forceduration', False) and info_dict.get('duration') is not None: + self.to_stdout(formatSeconds(info_dict['duration'])) if self.params.get('forceformat', False): - compat_print(info_dict['format']) + self.to_stdout(info_dict['format']) + if self.params.get('forcejson', False): + info_dict['_filename'] = filename + self.to_stdout(json.dumps(info_dict)) # Do nothing else if in simulate mode if self.params.get('simulate', False): @@ -600,28 +779,34 @@ class YoutubeDL(object): return if self.params.get('writedescription', False): - try: - descfn = filename + u'.description' - self.report_writedescription(descfn) - with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: - descfile.write(info_dict['description']) - except (KeyError, TypeError): - self.report_warning(u'There\'s no description to write.') - except (OSError, IOError): - self.report_error(u'Cannot write description file ' + descfn) - return + descfn = filename + u'.description' + if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)): + self.to_screen(u'[info] Video description is already present') + else: + try: + self.to_screen(u'[info] Writing video description to: ' + descfn) + with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: + descfile.write(info_dict['description']) + except (KeyError, TypeError): + self.report_warning(u'There\'s no description to write.') + except (OSError, IOError): + self.report_error(u'Cannot write description file ' + descfn) + return if self.params.get('writeannotations', False): - try: - annofn = filename + u'.annotations.xml' - self.report_writeannotations(annofn) - with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: - annofile.write(info_dict['annotations']) - except (KeyError, TypeError): - self.report_warning(u'There are no annotations to write.') - except (OSError, IOError): - self.report_error(u'Cannot write annotations file: ' + annofn) - return + annofn = filename + u'.annotations.xml' + if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)): + self.to_screen(u'[info] Video annotations are already present') + else: + try: + self.to_screen(u'[info] Writing video annotations to: ' + annofn) + with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: + annofile.write(info_dict['annotations']) + except (KeyError, TypeError): + self.report_warning(u'There are no annotations to write.') + except (OSError, IOError): + self.report_error(u'Cannot write annotations file: ' + annofn) + return subtitles_are_requested = any([self.params.get('writesubtitles', False), self.params.get('writeautomaticsub')]) @@ -630,45 +815,55 @@ class YoutubeDL(object): # subtitles download errors are already managed as troubles in relevant IE # that way it will silently go on when used with unsupporting IE subtitles = info_dict['subtitles'] - sub_format = self.params.get('subtitlesformat') + sub_format = self.params.get('subtitlesformat', 'srt') for sub_lang in subtitles.keys(): sub = subtitles[sub_lang] if sub is None: continue try: sub_filename = subtitles_filename(filename, sub_lang, sub_format) - self.report_writesubtitles(sub_filename) - with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile: - subfile.write(sub) + if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)): + self.to_screen(u'[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format)) + else: + self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename) + with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile: + subfile.write(sub) except (OSError, IOError): self.report_error(u'Cannot write subtitles file ' + descfn) return if self.params.get('writeinfojson', False): - infofn = filename + u'.info.json' - self.report_writeinfojson(infofn) - try: - json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle']) - write_json_file(json_info_dict, encodeFilename(infofn)) - except (OSError, IOError): - self.report_error(u'Cannot write metadata to JSON file ' + infofn) - return + infofn = os.path.splitext(filename)[0] + u'.info.json' + if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)): + self.to_screen(u'[info] Video description metadata is already present') + else: + self.to_screen(u'[info] Writing video description metadata as JSON to: ' + infofn) + try: + json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle']) + write_json_file(json_info_dict, encodeFilename(infofn)) + except (OSError, IOError): + self.report_error(u'Cannot write metadata to JSON file ' + infofn) + return if self.params.get('writethumbnail', False): if info_dict.get('thumbnail') is not None: thumb_format = determine_ext(info_dict['thumbnail'], u'jpg') - thumb_filename = filename.rpartition('.')[0] + u'.' + thumb_format - self.to_screen(u'[%s] %s: Downloading thumbnail ...' % - (info_dict['extractor'], info_dict['id'])) - try: - uf = compat_urllib_request.urlopen(info_dict['thumbnail']) - with open(thumb_filename, 'wb') as thumbf: - shutil.copyfileobj(uf, thumbf) - self.to_screen(u'[%s] %s: Writing thumbnail to: %s' % - (info_dict['extractor'], info_dict['id'], thumb_filename)) - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self.report_warning(u'Unable to download thumbnail "%s": %s' % - (info_dict['thumbnail'], compat_str(err))) + thumb_filename = os.path.splitext(filename)[0] + u'.' + thumb_format + if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)): + self.to_screen(u'[%s] %s: Thumbnail is already present' % + (info_dict['extractor'], info_dict['id'])) + else: + self.to_screen(u'[%s] %s: Downloading thumbnail ...' % + (info_dict['extractor'], info_dict['id'])) + try: + uf = compat_urllib_request.urlopen(info_dict['thumbnail']) + with open(thumb_filename, 'wb') as thumbf: + shutil.copyfileobj(uf, thumbf) + self.to_screen(u'[%s] %s: Writing thumbnail to: %s' % + (info_dict['extractor'], info_dict['id'], thumb_filename)) + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + self.report_warning(u'Unable to download thumbnail "%s": %s' % + (info_dict['thumbnail'], compat_str(err))) if not self.params.get('skip_download', False): if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)): @@ -696,13 +891,15 @@ class YoutubeDL(object): def download(self, url_list): """Download a given list of URLs.""" - if len(url_list) > 1 and self.fixed_template(): + if (len(url_list) > 1 and + '%' not in self.params['outtmpl'] + and self.params.get('max_downloads') != 1): raise SameFileError(self.params['outtmpl']) for url in url_list: try: #It also downloads the videos - videos = self.extract_info(url) + self.extract_info(url) except UnavailableVideoError: self.report_error(u'unable to download video') except MaxDownloadsReached: @@ -711,6 +908,20 @@ class YoutubeDL(object): return self._download_retcode + def download_with_info_file(self, info_filename): + with io.open(info_filename, 'r', encoding='utf-8') as f: + info = json.load(f) + try: + self.process_ie_result(info, download=True) + except DownloadError: + webpage_url = info.get('webpage_url') + if webpage_url is not None: + self.report_warning(u'The info failed to download, trying with "%s"' % webpage_url) + return self.download([webpage_url]) + else: + raise + return self._download_retcode + def post_process(self, filename, ie_info): """Run all the postprocessors on the given file.""" info = dict(ie_info) @@ -734,11 +945,26 @@ class YoutubeDL(object): except (IOError, OSError): self.report_warning(u'Unable to remove downloaded video file') + def _make_archive_id(self, info_dict): + # Future-proof against any change in case + # and backwards compatibility with prior versions + extractor = info_dict.get('extractor_key') + if extractor is None: + if 'id' in info_dict: + extractor = info_dict.get('ie_key') # key in a playlist + if extractor is None: + return None # Incomplete video information + return extractor.lower() + u' ' + info_dict['id'] + def in_download_archive(self, info_dict): fn = self.params.get('download_archive') if fn is None: return False - vid_id = info_dict['extractor'] + u' ' + info_dict['id'] + + vid_id = self._make_archive_id(info_dict) + if vid_id is None: + return False # Incomplete video information + try: with locked_file(fn, 'r', encoding='utf-8') as archive_file: for line in archive_file: @@ -753,12 +979,15 @@ class YoutubeDL(object): fn = self.params.get('download_archive') if fn is None: return - vid_id = info_dict['extractor'] + u' ' + info_dict['id'] + vid_id = self._make_archive_id(info_dict) + assert vid_id with locked_file(fn, 'a', encoding='utf-8') as archive_file: archive_file.write(vid_id + u'\n') @staticmethod def format_resolution(format, default='unknown'): + if format.get('vcodec') == 'none': + return 'audio only' if format.get('_resolution') is not None: return format['_resolution'] if format.get('height') is not None: @@ -771,23 +1000,124 @@ class YoutubeDL(object): return res def list_formats(self, info_dict): - def line(format): - return (u'%-15s%-10s%-12s%s' % ( + def format_note(fdict): + res = u'' + if fdict.get('format_note') is not None: + res += fdict['format_note'] + u' ' + if (fdict.get('vcodec') is not None and + fdict.get('vcodec') != 'none'): + res += u'%-5s' % fdict['vcodec'] + elif fdict.get('vbr') is not None: + res += u'video' + if fdict.get('vbr') is not None: + res += u'@%4dk' % fdict['vbr'] + if fdict.get('acodec') is not None: + if res: + res += u', ' + res += u'%-5s' % fdict['acodec'] + elif fdict.get('abr') is not None: + if res: + res += u', ' + res += 'audio' + if fdict.get('abr') is not None: + res += u'@%3dk' % fdict['abr'] + if fdict.get('filesize') is not None: + if res: + res += u', ' + res += format_bytes(fdict['filesize']) + return res + + def line(format, idlen=20): + return ((u'%-' + compat_str(idlen + 1) + u's%-10s%-12s%s') % ( format['format_id'], format['ext'], self.format_resolution(format), - format.get('format_note', ''), - ) - ) + format_note(format), + )) formats = info_dict.get('formats', [info_dict]) - formats_s = list(map(line, formats)) + idlen = max(len(u'format code'), + max(len(f['format_id']) for f in formats)) + formats_s = [line(f, idlen) for f in formats] if len(formats) > 1: - formats_s[0] += (' ' if formats[0].get('format_note') else '') + '(worst)' - formats_s[-1] += (' ' if formats[-1].get('format_note') else '') + '(best)' + formats_s[0] += (' ' if format_note(formats[0]) else '') + '(worst)' + formats_s[-1] += (' ' if format_note(formats[-1]) else '') + '(best)' header_line = line({ 'format_id': u'format code', 'ext': u'extension', - '_resolution': u'resolution', 'format_note': u'note'}) + '_resolution': u'resolution', 'format_note': u'note'}, idlen=idlen) self.to_screen(u'[info] Available formats for %s:\n%s\n%s' % (info_dict['id'], header_line, u"\n".join(formats_s))) + + def urlopen(self, req): + """ Start an HTTP download """ + return self._opener.open(req) + + def print_debug_header(self): + if not self.params.get('verbose'): + return + write_string(u'[debug] youtube-dl version ' + __version__ + u'\n') + try: + sp = subprocess.Popen( + ['git', 'rev-parse', '--short', 'HEAD'], + stdout=subprocess.PIPE, stderr=subprocess.PIPE, + cwd=os.path.dirname(os.path.abspath(__file__))) + out, err = sp.communicate() + out = out.decode().strip() + if re.match('[0-9a-f]+', out): + write_string(u'[debug] Git HEAD: ' + out + u'\n') + except: + try: + sys.exc_clear() + except: + pass + write_string(u'[debug] Python version %s - %s' % + (platform.python_version(), platform_name()) + u'\n') + + proxy_map = {} + for handler in self._opener.handlers: + if hasattr(handler, 'proxies'): + proxy_map.update(handler.proxies) + write_string(u'[debug] Proxy map: ' + compat_str(proxy_map) + u'\n') + + def _setup_opener(self): + timeout_val = self.params.get('socket_timeout') + timeout = 600 if timeout_val is None else float(timeout_val) + + opts_cookiefile = self.params.get('cookiefile') + opts_proxy = self.params.get('proxy') + + if opts_cookiefile is None: + self.cookiejar = compat_cookiejar.CookieJar() + else: + self.cookiejar = compat_cookiejar.MozillaCookieJar( + opts_cookiefile) + if os.access(opts_cookiefile, os.R_OK): + self.cookiejar.load() + + cookie_processor = compat_urllib_request.HTTPCookieProcessor( + self.cookiejar) + if opts_proxy is not None: + if opts_proxy == '': + proxies = {} + else: + proxies = {'http': opts_proxy, 'https': opts_proxy} + else: + proxies = compat_urllib_request.getproxies() + # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805) + if 'http' in proxies and 'https' not in proxies: + proxies['https'] = proxies['http'] + proxy_handler = compat_urllib_request.ProxyHandler(proxies) + https_handler = make_HTTPS_handler( + self.params.get('nocheckcertificate', False)) + opener = compat_urllib_request.build_opener( + https_handler, proxy_handler, cookie_processor, YoutubeDLHandler()) + # Delete the default user-agent header, which would otherwise apply in + # cases where our custom HTTP handler doesn't come into play + # (See https://github.com/rg3/youtube-dl/issues/1309 for details) + opener.addheaders = [] + self._opener = opener + + # TODO remove this global modification + compat_urllib_request.install_opener(opener) + socket.setdefaulttimeout(timeout) diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index 48ffcbf8e..6df44020b 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -32,50 +32,47 @@ __authors__ = ( 'Ismael Mejía', 'Steffan \'Ruirize\' James', 'Andras Elso', + 'Jelle van der Waa', + 'Marcin Cieślak', + 'Anton Larionov', + 'Takuya Tsuchida', + 'Sergey M.', + 'Michael Orlitzky', ) __license__ = 'Public Domain' import codecs -import collections import getpass import optparse import os import random import re import shlex -import socket -import subprocess import sys -import traceback -import platform from .utils import ( - compat_cookiejar, compat_print, - compat_str, - compat_urllib_request, DateRange, decodeOption, determine_ext, + get_term_width, DownloadError, get_cachedir, - make_HTTPS_handler, MaxDownloadsReached, - platform_name, preferredencoding, SameFileError, + setproctitle, std_headers, write_string, - YoutubeDLHandler, ) from .update import update_self -from .version import __version__ from .FileDownloader import ( FileDownloader, ) from .extractor import gen_extractors +from .version import __version__ from .YoutubeDL import YoutubeDL from .PostProcessor import ( FFmpegMetadataPP, @@ -86,11 +83,11 @@ from .PostProcessor import ( def parseOpts(overrideArguments=None): - def _readOptions(filename_bytes): + def _readOptions(filename_bytes, default=[]): try: optionf = open(filename_bytes) except IOError: - return [] # silently skip if file is not present + return default # silently skip if file is not present try: res = [] for l in optionf: @@ -118,19 +115,6 @@ def parseOpts(overrideArguments=None): def _comma_separated_values_options_callback(option, opt_str, value, parser): setattr(parser.values, option.dest, value.split(',')) - def _find_term_columns(): - columns = os.environ.get('COLUMNS', None) - if columns: - return int(columns) - - try: - sp = subprocess.Popen(['stty', 'size'], stdout=subprocess.PIPE, stderr=subprocess.PIPE) - out,err = sp.communicate() - return int(out.split()[1]) - except: - pass - return None - def _hide_login_info(opts): opts = list(opts) for private_opt in ['-p', '--password', '-u', '--username', '--video-password']: @@ -145,7 +129,7 @@ def parseOpts(overrideArguments=None): max_help_position = 80 # No need to wrap help messages if we're on a wide console - columns = _find_term_columns() + columns = get_term_width() if columns: max_width = columns fmt = optparse.IndentedHelpFormatter(width=max_width, max_help_position=max_help_position) @@ -196,7 +180,9 @@ def parseOpts(overrideArguments=None): general.add_option('--extractor-descriptions', action='store_true', dest='list_extractor_descriptions', help='Output descriptions of all supported extractors', default=False) - general.add_option('--proxy', dest='proxy', default=None, help='Use the specified HTTP/HTTPS proxy', metavar='URL') + general.add_option( + '--proxy', dest='proxy', default=None, metavar='URL', + help='Use the specified HTTP/HTTPS proxy. Pass in an empty string (--proxy "") for direct connection') general.add_option('--no-check-certificate', action='store_true', dest='no_check_certificate', default=False, help='Suppress HTTPS certificate validation.') general.add_option( '--cache-dir', dest='cachedir', default=get_cachedir(), metavar='DIR', @@ -204,27 +190,47 @@ def parseOpts(overrideArguments=None): general.add_option( '--no-cache-dir', action='store_const', const=None, dest='cachedir', help='Disable filesystem caching') - - - selection.add_option('--playlist-start', - dest='playliststart', metavar='NUMBER', help='playlist video to start at (default is %default)', default=1) - selection.add_option('--playlist-end', - dest='playlistend', metavar='NUMBER', help='playlist video to end at (default is last)', default=-1) + general.add_option( + '--socket-timeout', dest='socket_timeout', + type=float, default=None, help=optparse.SUPPRESS_HELP) + general.add_option( + '--bidi-workaround', dest='bidi_workaround', action='store_true', + help=u'Work around terminals that lack bidirectional text support. Requires fribidi executable in PATH') + + + selection.add_option( + '--playlist-start', + dest='playliststart', metavar='NUMBER', default=1, type=int, + help='playlist video to start at (default is %default)') + selection.add_option( + '--playlist-end', + dest='playlistend', metavar='NUMBER', default=None, type=int, + help='playlist video to end at (default is last)') selection.add_option('--match-title', dest='matchtitle', metavar='REGEX',help='download only matching titles (regex or caseless sub-string)') selection.add_option('--reject-title', dest='rejecttitle', metavar='REGEX',help='skip download for matching titles (regex or caseless sub-string)') - selection.add_option('--max-downloads', metavar='NUMBER', dest='max_downloads', help='Abort after downloading NUMBER files', default=None) + selection.add_option('--max-downloads', metavar='NUMBER', + dest='max_downloads', type=int, default=None, + help='Abort after downloading NUMBER files') selection.add_option('--min-filesize', metavar='SIZE', dest='min_filesize', help="Do not download any videos smaller than SIZE (e.g. 50k or 44.6m)", default=None) selection.add_option('--max-filesize', metavar='SIZE', dest='max_filesize', help="Do not download any videos larger than SIZE (e.g. 50k or 44.6m)", default=None) selection.add_option('--date', metavar='DATE', dest='date', help='download only videos uploaded in this date', default=None) selection.add_option('--datebefore', metavar='DATE', dest='datebefore', help='download only videos uploaded before this date', default=None) selection.add_option('--dateafter', metavar='DATE', dest='dateafter', help='download only videos uploaded after this date', default=None) + selection.add_option( + '--min-views', metavar='COUNT', dest='min_views', + default=None, type=int, + help="Do not download any videos with less than COUNT views",) + selection.add_option( + '--max-views', metavar='COUNT', dest='max_views', + default=None, type=int, + help="Do not download any videos with more than COUNT views",) selection.add_option('--no-playlist', action='store_true', dest='noplaylist', help='download only the currently playing video', default=False) selection.add_option('--age-limit', metavar='YEARS', dest='age_limit', help='download only videos suitable for the given age', default=None, type=int) selection.add_option('--download-archive', metavar='FILE', dest='download_archive', - help='Download only videos not present in the archive file. Record all downloaded videos in it.') + help='Download only videos not listed in the archive file. Record the IDs of all downloaded videos in it.') authentication.add_option('-u', '--username', @@ -239,7 +245,7 @@ def parseOpts(overrideArguments=None): video_format.add_option('-f', '--format', action='store', dest='format', metavar='FORMAT', default='best', - help='video format code, specifiy the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported') + help='video format code, specify the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported') video_format.add_option('--all-formats', action='store_const', dest='format', help='download all available video formats', const='all') video_format.add_option('--prefer-free-formats', @@ -298,12 +304,18 @@ def parseOpts(overrideArguments=None): verbosity.add_option('--get-description', action='store_true', dest='getdescription', help='simulate, quiet but print video description', default=False) + verbosity.add_option('--get-duration', + action='store_true', dest='getduration', + help='simulate, quiet but print video length', default=False) verbosity.add_option('--get-filename', action='store_true', dest='getfilename', help='simulate, quiet but print output filename', default=False) verbosity.add_option('--get-format', action='store_true', dest='getformat', help='simulate, quiet but print output format', default=False) + verbosity.add_option('-j', '--dump-json', + action='store_true', dest='dumpjson', + help='simulate, quiet but print JSON information', default=False) verbosity.add_option('--newline', action='store_true', dest='progress_with_newline', help='output progress bar as new lines', default=False) verbosity.add_option('--no-progress', @@ -318,7 +330,7 @@ def parseOpts(overrideArguments=None): help='print downloaded pages to debug problems(very verbose)') verbosity.add_option('--write-pages', action='store_true', dest='write_pages', default=False, - help='Write downloaded pages to files in the current directory') + help='Write downloaded intermediary pages to files in the current directory to debug problems') verbosity.add_option('--youtube-print-sig-code', action='store_true', dest='youtube_print_sig_code', default=False, help=optparse.SUPPRESS_HELP) @@ -349,16 +361,19 @@ def parseOpts(overrideArguments=None): 'for example with -o \'/my/downloads/%(uploader)s/%(title)s-%(id)s.%(ext)s\' .')) filesystem.add_option('--autonumber-size', dest='autonumber_size', metavar='NUMBER', - help='Specifies the number of digits in %(autonumber)s when it is present in output filename template or --autonumber option is given') + help='Specifies the number of digits in %(autonumber)s when it is present in output filename template or --auto-number option is given') filesystem.add_option('--restrict-filenames', action='store_true', dest='restrictfilenames', help='Restrict filenames to only ASCII characters, and avoid "&" and spaces in filenames', default=False) filesystem.add_option('-a', '--batch-file', dest='batchfile', metavar='FILE', help='file containing URLs to download (\'-\' for stdin)') + filesystem.add_option('--load-info', + dest='load_info_filename', metavar='FILE', + help='json file containing the video information (created with the "--write-json" option') filesystem.add_option('-w', '--no-overwrites', action='store_true', dest='nooverwrites', help='do not overwrite files', default=False) filesystem.add_option('-c', '--continue', - action='store_true', dest='continue_dl', help='resume partially downloaded files', default=True) + action='store_true', dest='continue_dl', help='force resume of partially downloaded files. By default, youtube-dl will resume downloads if possible.', default=True) filesystem.add_option('--no-continue', action='store_false', dest='continue_dl', help='do not resume partially downloaded files (restart from beginning)') @@ -416,6 +431,8 @@ def parseOpts(overrideArguments=None): if opts.verbose: write_string(u'[debug] Override config: ' + repr(overrideArguments) + '\n') else: + systemConf = _readOptions('/etc/youtube-dl.conf') + xdg_config_home = os.environ.get('XDG_CONFIG_HOME') if xdg_config_home: userConfFile = os.path.join(xdg_config_home, 'youtube-dl', 'config') @@ -425,8 +442,31 @@ def parseOpts(overrideArguments=None): userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl', 'config') if not os.path.isfile(userConfFile): userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl.conf') - systemConf = _readOptions('/etc/youtube-dl.conf') - userConf = _readOptions(userConfFile) + userConf = _readOptions(userConfFile, None) + + if userConf is None: + appdata_dir = os.environ.get('appdata') + if appdata_dir: + userConf = _readOptions( + os.path.join(appdata_dir, 'youtube-dl', 'config'), + default=None) + if userConf is None: + userConf = _readOptions( + os.path.join(appdata_dir, 'youtube-dl', 'config.txt'), + default=None) + + if userConf is None: + userConf = _readOptions( + os.path.join(os.path.expanduser('~'), 'youtube-dl.conf'), + default=None) + if userConf is None: + userConf = _readOptions( + os.path.join(os.path.expanduser('~'), 'youtube-dl.conf.txt'), + default=None) + + if userConf is None: + userConf = [] + commandLineConf = sys.argv[1:] argv = systemConf + userConf + commandLineConf opts, args = parser.parse_args(argv) @@ -437,27 +477,17 @@ def parseOpts(overrideArguments=None): return parser, opts, args + def _real_main(argv=None): # Compatibility fixes for Windows if sys.platform == 'win32': # https://github.com/rg3/youtube-dl/issues/820 codecs.register(lambda name: codecs.lookup('utf-8') if name == 'cp65001' else None) + setproctitle(u'youtube-dl') + parser, opts, args = parseOpts(argv) - # Open appropriate CookieJar - if opts.cookiefile is None: - jar = compat_cookiejar.CookieJar() - else: - try: - jar = compat_cookiejar.MozillaCookieJar(opts.cookiefile) - if os.access(opts.cookiefile, os.R_OK): - jar.load() - except (IOError, OSError) as err: - if opts.verbose: - traceback.print_exc() - write_string(u'ERROR: unable to open cookie file\n') - sys.exit(101) # Set user agent if opts.user_agent is not None: std_headers['User-Agent'] = opts.user_agent @@ -489,8 +519,6 @@ def _real_main(argv=None): all_urls = batchurls + args all_urls = [url.strip() for url in all_urls] - opener = _setup_opener(jar=jar, opts=opts) - extractors = gen_extractors() if opts.list_extractors: @@ -545,25 +573,17 @@ def _real_main(argv=None): if opts.retries is not None: try: opts.retries = int(opts.retries) - except (TypeError, ValueError) as err: + except (TypeError, ValueError): parser.error(u'invalid retry count specified') if opts.buffersize is not None: numeric_buffersize = FileDownloader.parse_bytes(opts.buffersize) if numeric_buffersize is None: parser.error(u'invalid buffer size specified') opts.buffersize = numeric_buffersize - try: - opts.playliststart = int(opts.playliststart) - if opts.playliststart <= 0: - raise ValueError(u'Playlist start must be positive') - except (TypeError, ValueError) as err: - parser.error(u'invalid playlist start number specified') - try: - opts.playlistend = int(opts.playlistend) - if opts.playlistend != -1 and (opts.playlistend <= 0 or opts.playlistend < opts.playliststart): - raise ValueError(u'Playlist end must be greater than playlist start') - except (TypeError, ValueError) as err: - parser.error(u'invalid playlist end number specified') + if opts.playliststart <= 0: + raise ValueError(u'Playlist start must be positive') + if opts.playlistend not in (-1, None) and opts.playlistend < opts.playliststart: + raise ValueError(u'Playlist end must be greater than playlist start') if opts.extractaudio: if opts.audioformat not in ['best', 'aac', 'mp3', 'm4a', 'opus', 'vorbis', 'wav']: parser.error(u'invalid audio format specified') @@ -596,27 +616,30 @@ def _real_main(argv=None): or (opts.useid and u'%(id)s.%(ext)s') or (opts.autonumber and u'%(autonumber)s-%(id)s.%(ext)s') or u'%(title)s-%(id)s.%(ext)s') - if '%(ext)s' not in outtmpl and opts.extractaudio: + if not os.path.splitext(outtmpl)[1] and opts.extractaudio: parser.error(u'Cannot download a video and extract audio into the same' - u' file! Use "%%(ext)s" instead of %r' % - determine_ext(outtmpl, u'')) + u' file! Use "{0}.%(ext)s" instead of "{0}" as the output' + u' template'.format(outtmpl)) + + any_printing = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson - # YoutubeDL - ydl = YoutubeDL({ + ydl_opts = { 'usenetrc': opts.usenetrc, 'username': opts.username, 'password': opts.password, 'videopassword': opts.videopassword, - 'quiet': (opts.quiet or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat), + 'quiet': (opts.quiet or any_printing), 'forceurl': opts.geturl, 'forcetitle': opts.gettitle, 'forceid': opts.getid, 'forcethumbnail': opts.getthumbnail, 'forcedescription': opts.getdescription, + 'forceduration': opts.getduration, 'forcefilename': opts.getfilename, 'forceformat': opts.getformat, + 'forcejson': opts.dumpjson, 'simulate': opts.simulate, - 'skip_download': (opts.skip_download or opts.simulate or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat), + 'skip_download': (opts.skip_download or opts.simulate or any_printing), 'format': opts.format, 'format_limit': opts.format_limit, 'listformats': opts.listformats, @@ -660,107 +683,58 @@ def _real_main(argv=None): 'keepvideo': opts.keepvideo, 'min_filesize': opts.min_filesize, 'max_filesize': opts.max_filesize, + 'min_views': opts.min_views, + 'max_views': opts.max_views, 'daterange': date, 'cachedir': opts.cachedir, 'youtube_print_sig_code': opts.youtube_print_sig_code, 'age_limit': opts.age_limit, 'download_archive': opts.download_archive, - }) - - if opts.verbose: - write_string(u'[debug] youtube-dl version ' + __version__ + u'\n') - try: - sp = subprocess.Popen( - ['git', 'rev-parse', '--short', 'HEAD'], - stdout=subprocess.PIPE, stderr=subprocess.PIPE, - cwd=os.path.dirname(os.path.abspath(__file__))) - out, err = sp.communicate() - out = out.decode().strip() - if re.match('[0-9a-f]+', out): - write_string(u'[debug] Git HEAD: ' + out + u'\n') - except: - try: - sys.exc_clear() - except: - pass - write_string(u'[debug] Python version %s - %s' %(platform.python_version(), platform_name()) + u'\n') - - proxy_map = {} - for handler in opener.handlers: - if hasattr(handler, 'proxies'): - proxy_map.update(handler.proxies) - write_string(u'[debug] Proxy map: ' + compat_str(proxy_map) + u'\n') - - ydl.add_default_info_extractors() - - # PostProcessors - # Add the metadata pp first, the other pps will copy it - if opts.addmetadata: - ydl.add_post_processor(FFmpegMetadataPP()) - if opts.extractaudio: - ydl.add_post_processor(FFmpegExtractAudioPP(preferredcodec=opts.audioformat, preferredquality=opts.audioquality, nopostoverwrites=opts.nopostoverwrites)) - if opts.recodevideo: - ydl.add_post_processor(FFmpegVideoConvertor(preferedformat=opts.recodevideo)) - if opts.embedsubtitles: - ydl.add_post_processor(FFmpegEmbedSubtitlePP(subtitlesformat=opts.subtitlesformat)) - - # Update version - if opts.update_self: - update_self(ydl.to_screen, opts.verbose) - - # Maybe do nothing - if len(all_urls) < 1: - if not opts.update_self: - parser.error(u'you must provide at least one URL') - else: - sys.exit() + 'cookiefile': opts.cookiefile, + 'nocheckcertificate': opts.no_check_certificate, + 'proxy': opts.proxy, + 'socket_timeout': opts.socket_timeout, + 'bidi_workaround': opts.bidi_workaround, + } - try: - retcode = ydl.download(all_urls) - except MaxDownloadsReached: - ydl.to_screen(u'--max-download limit reached, aborting.') - retcode = 101 + with YoutubeDL(ydl_opts) as ydl: + ydl.print_debug_header() + ydl.add_default_info_extractors() + + # PostProcessors + # Add the metadata pp first, the other pps will copy it + if opts.addmetadata: + ydl.add_post_processor(FFmpegMetadataPP()) + if opts.extractaudio: + ydl.add_post_processor(FFmpegExtractAudioPP(preferredcodec=opts.audioformat, preferredquality=opts.audioquality, nopostoverwrites=opts.nopostoverwrites)) + if opts.recodevideo: + ydl.add_post_processor(FFmpegVideoConvertor(preferedformat=opts.recodevideo)) + if opts.embedsubtitles: + ydl.add_post_processor(FFmpegEmbedSubtitlePP(subtitlesformat=opts.subtitlesformat)) + + # Update version + if opts.update_self: + update_self(ydl.to_screen, opts.verbose) + + # Maybe do nothing + if (len(all_urls) < 1) and (opts.load_info_filename is None): + if not opts.update_self: + parser.error(u'you must provide at least one URL') + else: + sys.exit() - # Dump cookie jar if requested - if opts.cookiefile is not None: try: - jar.save() - except (IOError, OSError): - sys.exit(u'ERROR: unable to save cookie jar') + if opts.load_info_filename is not None: + retcode = ydl.download_with_info_file(opts.load_info_filename) + else: + retcode = ydl.download(all_urls) + except MaxDownloadsReached: + ydl.to_screen(u'--max-download limit reached, aborting.') + retcode = 101 sys.exit(retcode) -def _setup_opener(jar=None, opts=None, timeout=300): - if opts is None: - FakeOptions = collections.namedtuple( - 'FakeOptions', ['proxy', 'no_check_certificate']) - opts = FakeOptions(proxy=None, no_check_certificate=False) - - cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar) - if opts.proxy is not None: - if opts.proxy == '': - proxies = {} - else: - proxies = {'http': opts.proxy, 'https': opts.proxy} - else: - proxies = compat_urllib_request.getproxies() - # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805) - if 'http' in proxies and 'https' not in proxies: - proxies['https'] = proxies['http'] - proxy_handler = compat_urllib_request.ProxyHandler(proxies) - https_handler = make_HTTPS_handler(opts) - opener = compat_urllib_request.build_opener( - https_handler, proxy_handler, cookie_processor, YoutubeDLHandler()) - # Delete the default user-agent header, which would otherwise apply in - # cases where our custom HTTP handler doesn't come into play - # (See https://github.com/rg3/youtube-dl/issues/1309 for details) - opener.addheaders = [] - compat_urllib_request.install_opener(opener) - socket.setdefaulttimeout(timeout) - return opener - - def main(argv=None): try: _real_main(argv) diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index a61e17ea1..a39a1e2f4 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -1,5 +1,8 @@ -from .appletrailers import AppleTrailersIE +from .academicearth import AcademicEarthCourseIE from .addanime import AddAnimeIE +from .anitube import AnitubeIE +from .aparat import AparatIE +from .appletrailers import AppleTrailersIE from .archiveorg import ArchiveOrgIE from .ard import ARDIE from .arte import ( @@ -7,9 +10,12 @@ from .arte import ( ArteTVPlus7IE, ArteTVCreativeIE, ArteTVFutureIE, + ArteTVDDCIE, ) from .auengine import AUEngineIE -from .bandcamp import BandcampIE +from .bambuser import BambuserIE, BambuserChannelIE +from .bandcamp import BandcampIE, BandcampAlbumIE +from .blinkx import BlinkxIE from .bliptv import BlipTVIE, BlipTVUserIE from .bloomberg import BloombergIE from .breakcom import BreakIE @@ -17,14 +23,19 @@ from .brightcove import BrightcoveIE from .c56 import C56IE from .canalplus import CanalplusIE from .canalc2 import Canalc2IE +from .cbs import CBSIE +from .channel9 import Channel9IE from .cinemassacre import CinemassacreIE +from .clipfish import ClipfishIE +from .clipsyndicate import ClipsyndicateIE from .cnn import CNNIE from .collegehumor import CollegeHumorIE -from .comedycentral import ComedyCentralIE +from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE from .condenast import CondeNastIE from .criterion import CriterionIE from .crunchyroll import CrunchyrollIE from .cspan import CSpanIE +from .d8 import D8IE from .dailymotion import ( DailymotionIE, DailymotionPlaylistIE, @@ -38,8 +49,10 @@ from .defense import DefenseGouvFrIE from .ebaumsworld import EbaumsWorldIE from .ehow import EHowIE from .eighttracks import EightTracksIE +from .eitb import EitbIE from .escapist import EscapistIE from .exfm import ExfmIE +from .extremetube import ExtremeTubeIE from .facebook import FacebookIE from .faz import FazIE from .fktv import ( @@ -50,11 +63,12 @@ from .flickr import FlickrIE from .francetv import ( PluzzIE, FranceTvInfoIE, - France2IE, + FranceTVIE, GenerationQuoiIE ) from .freesound import FreesoundIE from .funnyordie import FunnyOrDieIE +from .gamekings import GamekingsIE from .gamespot import GameSpotIE from .gametrailers import GametrailersIE from .generic import GenericIE @@ -65,10 +79,15 @@ from .hotnewhiphop import HotNewHipHopIE from .howcast import HowcastIE from .hypem import HypemIE from .ign import IGNIE, OneUPIE +from .imdb import ImdbIE from .ina import InaIE from .infoq import InfoQIE from .instagram import InstagramIE from .internetvideoarchive import InternetVideoArchiveIE +from .ivi import ( + IviIE, + IviCompilationIE +) from .jeuxvideo import JeuxVideoIE from .jukebox import JukeboxIE from .justintv import JustinTVIE @@ -77,11 +96,13 @@ from .keezmovies import KeezMoviesIE from .kickstarter import KickStarterIE from .keek import KeekIE from .liveleak import LiveLeakIE -from .livestream import LivestreamIE +from .livestream import LivestreamIE, LivestreamOriginalIE +from .mdr import MDRIE from .metacafe import MetacafeIE from .metacritic import MetacriticIE from .mit import TechTVMITIE, MITIE from .mixcloud import MixcloudIE +from .mofosex import MofosexIE from .mtv import MTVIE from .muzu import MuzuTVIE from .myspace import MySpaceIE @@ -90,15 +111,22 @@ from .myvideo import MyVideoIE from .naver import NaverIE from .nba import NBAIE from .nbc import NBCNewsIE +from .ndtv import NDTVIE from .newgrounds import NewgroundsIE from .nhl import NHLIE, NHLVideocenterIE +from .niconico import NiconicoIE +from .ninegag import NineGagIE from .nowvideo import NowVideoIE from .ooyala import OoyalaIE from .orf import ORFIE from .pbs import PBSIE from .photobucket import PhotobucketIE +from .podomatic import PodomaticIE +from .pornhd import PornHdIE from .pornhub import PornHubIE from .pornotube import PornotubeIE +from .pyvideo import PyvideoIE +from .radiofrance import RadioFranceIE from .rbmaradio import RBMARadioIE from .redtube import RedTubeIE from .ringtv import RingTVIE @@ -110,26 +138,40 @@ from .rutube import RutubeIE from .sina import SinaIE from .slashdot import SlashdotIE from .slideshare import SlideshareIE +from .smotri import ( + SmotriIE, + SmotriCommunityIE, + SmotriUserIE, + SmotriBroadcastIE, +) from .sohu import SohuIE from .soundcloud import SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE -from .southparkstudios import SouthParkStudiosIE +from .southparkstudios import ( + SouthParkStudiosIE, + SouthparkDeIE, +) +from .space import SpaceIE from .spankwire import SpankwireIE from .spiegel import SpiegelIE from .stanfordoc import StanfordOpenClassroomIE from .statigram import StatigramIE from .steam import SteamIE +from .streamcloud import StreamcloudIE from .sztvhu import SztvHuIE from .teamcoco import TeamcocoIE from .techtalks import TechTalksIE from .ted import TEDIE from .tf1 import TF1IE +from .theplatform import ThePlatformIE from .thisav import ThisAVIE +from .toutv import TouTvIE from .traileraddict import TrailerAddictIE from .trilulilu import TriluliluIE from .tube8 import Tube8IE from .tudou import TudouIE from .tumblr import TumblrIE from .tutv import TutvIE +from .tvp import TvpIE from .unistra import UnistraIE from .ustream import UstreamIE, UstreamChannelIE from .vbox7 import Vbox7IE @@ -141,18 +183,31 @@ from .viddler import ViddlerIE from .videodetective import VideoDetectiveIE from .videofyme import VideofyMeIE from .videopremium import VideoPremiumIE -from .vimeo import VimeoIE, VimeoChannelIE +from .vimeo import ( + VimeoIE, + VimeoChannelIE, + VimeoUserIE, + VimeoAlbumIE, + VimeoGroupsIE, +) from .vine import VineIE +from .viki import VikiIE from .vk import VKIE from .wat import WatIE from .websurg import WeBSurgIE from .weibo import WeiboIE from .wimp import WimpIE +from .wistia import WistiaIE from .worldstarhiphop import WorldStarHipHopIE from .xhamster import XHamsterIE from .xnxx import XNXXIE from .xvideos import XVideosIE -from .yahoo import YahooIE, YahooSearchIE +from .xtube import XTubeIE +from .yahoo import ( + YahooIE, + YahooNewsIE, + YahooSearchIE, +) from .youjizz import YouJizzIE from .youku import YoukuIE from .youporn import YouPornIE @@ -160,6 +215,7 @@ from .youtube import ( YoutubeIE, YoutubePlaylistIE, YoutubeSearchIE, + YoutubeSearchDateIE, YoutubeUserIE, YoutubeChannelIE, YoutubeShowIE, @@ -168,6 +224,8 @@ from .youtube import ( YoutubeTruncatedURLIE, YoutubeWatchLaterIE, YoutubeFavouritesIE, + YoutubeHistoryIE, + YoutubeTopListIE, ) from .zdf import ZDFIE diff --git a/youtube_dl/extractor/academicearth.py b/youtube_dl/extractor/academicearth.py new file mode 100644 index 000000000..ac05f8246 --- /dev/null +++ b/youtube_dl/extractor/academicearth.py @@ -0,0 +1,31 @@ +import re + +from .common import InfoExtractor + + +class AcademicEarthCourseIE(InfoExtractor): + _VALID_URL = r'^https?://(?:www\.)?academicearth\.org/(?:courses|playlists)/(?P[^?#/]+)' + IE_NAME = u'AcademicEarth:Course' + + def _real_extract(self, url): + m = re.match(self._VALID_URL, url) + playlist_id = m.group('id') + + webpage = self._download_webpage(url, playlist_id) + title = self._html_search_regex( + r'

(.*?)

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

(.*?)

', + webpage, u'description', fatal=False) + urls = re.findall( + r'

', + webpage) + entries = [self.url_result(u) for u in urls] + + return { + '_type': 'playlist', + 'id': playlist_id, + 'title': title, + 'description': description, + 'entries': entries, + } diff --git a/youtube_dl/extractor/addanime.py b/youtube_dl/extractor/addanime.py index b99d4b966..a3a1b999d 100644 --- a/youtube_dl/extractor/addanime.py +++ b/youtube_dl/extractor/addanime.py @@ -13,7 +13,7 @@ from ..utils import ( class AddAnimeIE(InfoExtractor): - _VALID_URL = r'^http://(?:\w+\.)?add-anime\.net/watch_video.php\?(?:.*?)v=(?P[\w_]+)(?:.*)' + _VALID_URL = r'^http://(?:\w+\.)?add-anime\.net/watch_video\.php\?(?:.*?)v=(?P[\w_]+)(?:.*)' IE_NAME = u'AddAnime' _TEST = { u'url': u'http://www.add-anime.net/watch_video.php?v=24MR3YO5SAS9', diff --git a/youtube_dl/extractor/anitube.py b/youtube_dl/extractor/anitube.py new file mode 100644 index 000000000..2b019daa9 --- /dev/null +++ b/youtube_dl/extractor/anitube.py @@ -0,0 +1,53 @@ +import re + +from .common import InfoExtractor + + +class AnitubeIE(InfoExtractor): + IE_NAME = u'anitube.se' + _VALID_URL = r'https?://(?:www\.)?anitube\.se/video/(?P\d+)' + + _TEST = { + u'url': u'http://www.anitube.se/video/36621', + u'md5': u'59d0eeae28ea0bc8c05e7af429998d43', + u'file': u'36621.mp4', + u'info_dict': { + u'id': u'36621', + u'ext': u'mp4', + u'title': u'Recorder to Randoseru 01', + }, + u'skip': u'Blocked in the US', + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + 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, u'key') + + config_xml = self._download_xml('http://www.anitube.se/nuevo/econfig.php?key=%s' % key, + key) + + video_title = config_xml.find('title').text + + formats = [] + video_url = config_xml.find('file') + if video_url is not None: + formats.append({ + 'format_id': 'sd', + 'url': video_url.text, + }) + video_url = config_xml.find('filehd') + if video_url is not None: + formats.append({ + 'format_id': 'hd', + 'url': video_url.text, + }) + + return { + 'id': video_id, + 'title': video_title, + 'formats': formats + } diff --git a/youtube_dl/extractor/aparat.py b/youtube_dl/extractor/aparat.py new file mode 100644 index 000000000..7e93bc4df --- /dev/null +++ b/youtube_dl/extractor/aparat.py @@ -0,0 +1,56 @@ +#coding: utf-8 + +import re + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + HEADRequest, +) + + +class AparatIE(InfoExtractor): + _VALID_URL = r'^https?://(?:www\.)?aparat\.com/(?:v/|video/video/embed/videohash/)(?P[a-zA-Z0-9]+)' + + _TEST = { + u'url': u'http://www.aparat.com/v/wP8On', + u'file': u'wP8On.mp4', + u'md5': u'6714e0af7e0d875c5a39c4dc4ab46ad1', + u'info_dict': { + u"title": u"تیم گلکسی 11 - زومیت", + }, + #u'skip': u'Extremely unreliable', + } + + def _real_extract(self, url): + m = re.match(self._VALID_URL, url) + video_id = m.group('id') + + # Note: There is an easier-to-parse configuration at + # http://www.aparat.com/video/video/config/videohash/%video_id + # but the URL in there does not work + embed_url = (u'http://www.aparat.com/video/video/embed/videohash/' + + video_id + u'/vt/frame') + webpage = self._download_webpage(embed_url, video_id) + + video_urls = re.findall(r'fileList\[[0-9]+\]\s*=\s*"([^"]+)"', webpage) + for i, video_url in enumerate(video_urls): + req = HEADRequest(video_url) + res = self._request_webpage( + req, video_id, note=u'Testing video URL %d' % i, errnote=False) + if res: + break + else: + raise ExtractorError(u'No working video URLs found') + + title = self._search_regex(r'\s+title:\s*"([^"]+)"', webpage, u'title') + thumbnail = self._search_regex( + r'\s+image:\s*"([^"]+)"', webpage, u'thumbnail', fatal=False) + + return { + 'id': video_id, + 'title': title, + 'url': video_url, + 'ext': 'mp4', + 'thumbnail': thumbnail, + } diff --git a/youtube_dl/extractor/appletrailers.py b/youtube_dl/extractor/appletrailers.py index 6d6237f8a..ef5644aa5 100644 --- a/youtube_dl/extractor/appletrailers.py +++ b/youtube_dl/extractor/appletrailers.py @@ -1,5 +1,4 @@ import re -import xml.etree.ElementTree import json from .common import InfoExtractor @@ -10,7 +9,7 @@ from ..utils import ( class AppleTrailersIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?trailers.apple.com/trailers/(?P[^/]+)/(?P[^/]+)' + _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/trailers/(?P[^/]+)/(?P[^/]+)' _TEST = { u"url": u"http://trailers.apple.com/trailers/wb/manofsteel/", u"playlist": [ @@ -65,18 +64,18 @@ class AppleTrailersIE(InfoExtractor): uploader_id = mobj.group('company') playlist_url = compat_urlparse.urljoin(url, u'includes/playlists/itunes.inc') - playlist_snippet = self._download_webpage(playlist_url, movie) - playlist_cleaned = re.sub(r'(?s).*?', u'', playlist_snippet) - playlist_cleaned = re.sub(r'', r'', playlist_cleaned) - # The ' in the onClick attributes are not escaped, it couldn't be parsed - # with xml.etree.ElementTree.fromstring - # like: http://trailers.apple.com/trailers/wb/gravity/ - def _clean_json(m): - return u'iTunes.playURL(%s);' % m.group(1).replace('\'', ''') - playlist_cleaned = re.sub(self._JSON_RE, _clean_json, playlist_cleaned) - playlist_html = u'' + playlist_cleaned + u'' + def fix_html(s): + s = re.sub(r'(?s).*?', u'', s) + s = re.sub(r'', r'', s) + # The ' in the onClick attributes are not escaped, it couldn't be parsed + # like: http://trailers.apple.com/trailers/wb/gravity/ + def _clean_json(m): + return u'iTunes.playURL(%s);' % m.group(1).replace('\'', ''') + s = re.sub(self._JSON_RE, _clean_json, s) + s = u'' + s + u'' + return s + doc = self._download_xml(playlist_url, movie, transform_source=fix_html) - doc = xml.etree.ElementTree.fromstring(playlist_html) playlist = [] for li in doc.findall('./div/ul/li'): on_click = li.find('.//a').attrib['onClick'] @@ -113,7 +112,7 @@ class AppleTrailersIE(InfoExtractor): }) formats = sorted(formats, key=lambda f: (f['height'], f['width'])) - info = { + playlist.append({ '_type': 'video', 'id': video_id, 'title': title, @@ -124,12 +123,7 @@ class AppleTrailersIE(InfoExtractor): 'upload_date': upload_date, 'uploader_id': uploader_id, 'user_agent': 'QuickTime compatible (youtube-dl)', - } - # TODO: Remove when #980 has been merged - info['url'] = formats[-1]['url'] - info['ext'] = formats[-1]['ext'] - - playlist.append(info) + }) return { '_type': 'playlist', diff --git a/youtube_dl/extractor/archiveorg.py b/youtube_dl/extractor/archiveorg.py index 61ce4469a..8bb546410 100644 --- a/youtube_dl/extractor/archiveorg.py +++ b/youtube_dl/extractor/archiveorg.py @@ -11,7 +11,7 @@ from ..utils import ( class ArchiveOrgIE(InfoExtractor): IE_NAME = 'archive.org' IE_DESC = 'archive.org videos' - _VALID_URL = r'(?:https?://)?(?:www\.)?archive.org/details/(?P[^?/]+)(?:[?].*)?$' + _VALID_URL = r'(?:https?://)?(?:www\.)?archive\.org/details/(?P[^?/]+)(?:[?].*)?$' _TEST = { u"url": u"http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect", u'file': u'XD300-23_68HighlightsAResearchCntAugHumanIntellect.ogv', @@ -49,7 +49,7 @@ class ArchiveOrgIE(InfoExtractor): for f in formats: f['ext'] = determine_ext(f['url']) - info = { + return { '_type': 'video', 'id': video_id, 'title': title, @@ -57,12 +57,5 @@ class ArchiveOrgIE(InfoExtractor): 'description': description, 'uploader': uploader, 'upload_date': upload_date, + 'thumbnail': data.get('misc', {}).get('image'), } - thumbnail = data.get('misc', {}).get('image') - if thumbnail: - info['thumbnail'] = thumbnail - - # TODO: Remove when #980 has been merged - info.update(formats[-1]) - - return info diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index e10c74c11..9254fbfe0 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -1,7 +1,6 @@ # encoding: utf-8 import re import json -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( @@ -10,6 +9,8 @@ from ..utils import ( unified_strdate, determine_ext, get_element_by_id, + compat_str, + get_element_by_attribute, ) # There are different sources of video in arte.tv, the extraction process @@ -17,8 +18,8 @@ from ..utils import ( # add tests. class ArteTvIE(InfoExtractor): - _VIDEOS_URL = r'(?:http://)?videos.arte.tv/(?Pfr|de)/.*-(?P.*?).html' - _LIVEWEB_URL = r'(?:http://)?liveweb.arte.tv/(?Pfr|de)/(?P.+?)/(?P.+)' + _VIDEOS_URL = r'(?:http://)?videos\.arte\.tv/(?Pfr|de)/.*-(?P.*?)\.html' + _LIVEWEB_URL = r'(?:http://)?liveweb\.arte\.tv/(?Pfr|de)/(?P.+?)/(?P.+)' _LIVE_URL = r'index-[0-9]+\.html$' IE_NAME = u'arte.tv' @@ -68,7 +69,7 @@ class ArteTvIE(InfoExtractor): lang = mobj.group('lang') return self._extract_liveweb(url, name, lang) - if re.search(self._LIVE_URL, video_id) is not None: + if re.search(self._LIVE_URL, url) is not None: raise ExtractorError(u'Arte live streams are not yet supported, sorry') # self.extractLiveStream(url) # return @@ -77,8 +78,7 @@ class ArteTvIE(InfoExtractor): """Extract from videos.arte.tv""" ref_xml_url = url.replace('/videos/', '/do_delegate/videos/') ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml') - ref_xml = self._download_webpage(ref_xml_url, video_id, note=u'Downloading metadata') - ref_xml_doc = xml.etree.ElementTree.fromstring(ref_xml) + ref_xml_doc = self._download_xml(ref_xml_url, video_id, note=u'Downloading metadata') config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang) config_xml_url = config_node.attrib['ref'] config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration') @@ -108,13 +108,12 @@ class ArteTvIE(InfoExtractor): """Extract form http://liveweb.arte.tv/""" webpage = self._download_webpage(url, name) video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, u'event id') - config_xml = self._download_webpage('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id, + config_doc = self._download_xml('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id, video_id, u'Downloading information') - config_doc = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8')) event_doc = config_doc.find('event') url_node = event_doc.find('video').find('urlHd') if url_node is None: - url_node = video_doc.find('urlSd') + url_node = event_doc.find('urlSd') return {'id': video_id, 'title': event_doc.find('name%s' % lang.capitalize()).text, @@ -144,7 +143,9 @@ class ArteTVPlus7IE(InfoExtractor): def _extract_from_webpage(self, webpage, video_id, lang): json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url') + return self._extract_from_json_url(json_url, video_id, lang) + def _extract_from_json_url(self, json_url, video_id, lang): json_info = self._download_webpage(json_url, video_id, 'Downloading info json') self.report_extraction(video_id) info = json.loads(json_info) @@ -181,20 +182,30 @@ class ArteTVPlus7IE(InfoExtractor): formats = all_formats else: raise ExtractorError(u'The formats list is empty') - # We order the formats by quality + if re.match(r'[A-Z]Q', formats[0]['quality']) is not None: - sort_key = lambda f: ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality']) + def sort_key(f): + return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality']) else: - sort_key = lambda f: int(f.get('height',-1)) + def sort_key(f): + return ( + # Sort first by quality + int(f.get('height',-1)), + int(f.get('bitrate',-1)), + # The original version with subtitles has lower relevance + re.match(r'VO-ST(F|A)', f.get('versionCode', '')) is None, + # The version with sourds/mal subtitles has also lower relevance + re.match(r'VO?(F|A)-STM\1', f.get('versionCode', '')) is None, + ) formats = sorted(formats, key=sort_key) - # Prefer videos without subtitles in the same language - formats = sorted(formats, key=lambda f: re.match(r'VO(F|A)-STM\1', f.get('versionCode', '')) is None) - # Pick the best quality def _format(format_info): - quality = format_info['quality'] - m_quality = re.match(r'\w*? - (\d*)p', quality) - if m_quality is not None: - quality = m_quality.group(1) + quality = '' + height = format_info.get('height') + if height is not None: + quality = compat_str(height) + bitrate = format_info.get('bitrate') + if bitrate is not None: + quality += '-%d' % bitrate if format_info.get('versionCode') is not None: format_id = u'%s-%s' % (quality, format_info['versionCode']) else: @@ -203,7 +214,7 @@ class ArteTVPlus7IE(InfoExtractor): 'format_id': format_id, 'format_note': format_info.get('versionLibelle'), 'width': format_info.get('width'), - 'height': format_info.get('height'), + 'height': height, } if format_info['mediaType'] == u'rtmp': info['url'] = format_info['streamer'] @@ -249,3 +260,21 @@ class ArteTVFutureIE(ArteTVPlus7IE): webpage = self._download_webpage(url, anchor_id) row = get_element_by_id(anchor_id, webpage) return self._extract_from_webpage(row, anchor_id, lang) + + +class ArteTVDDCIE(ArteTVPlus7IE): + IE_NAME = u'arte.tv:ddc' + _VALID_URL = r'http?://ddc\.arte\.tv/(?Pemission|folge)/(?P.+)' + + def _real_extract(self, url): + video_id, lang = self._extract_url_info(url) + if lang == 'folge': + lang = 'de' + elif lang == 'emission': + lang = 'fr' + webpage = self._download_webpage(url, video_id) + scriptElement = get_element_by_attribute('class', 'visu_video_block', webpage) + script_url = self._html_search_regex(r'src="(.*?)"', scriptElement, 'script url') + javascriptPlayerGenerator = self._download_webpage(script_url, video_id, 'Download javascript player generator') + json_url = self._search_regex(r"json_url=(.*)&rendering_place.*", javascriptPlayerGenerator, 'json url') + return self._extract_from_json_url(json_url, video_id, lang) diff --git a/youtube_dl/extractor/auengine.py b/youtube_dl/extractor/auengine.py index 0febbff4f..bcccc0b7a 100644 --- a/youtube_dl/extractor/auengine.py +++ b/youtube_dl/extractor/auengine.py @@ -1,10 +1,10 @@ -import os.path import re from .common import InfoExtractor from ..utils import ( compat_urllib_parse, - compat_urllib_parse_urlparse, + determine_ext, + ExtractorError, ) class AUEngineIE(InfoExtractor): @@ -16,7 +16,7 @@ class AUEngineIE(InfoExtractor): u"title": u"[Commie]The Legend of the Legendary Heroes - 03 - Replication Eye (Alpha Stigma)[F9410F5A]" } } - _VALID_URL = r'(?:http://)?(?:www\.)?auengine\.com/embed.php\?.*?file=([^&]+).*?' + _VALID_URL = r'(?:http://)?(?:www\.)?auengine\.com/embed\.php\?.*?file=([^&]+).*?' def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -25,22 +25,25 @@ class AUEngineIE(InfoExtractor): title = self._html_search_regex(r'(?P<title>.+?)', webpage, u'title') title = title.strip() - links = re.findall(r'[^A-Za-z0-9]?(?:file|url):\s*["\'](http[^\'"&]*)', webpage) - links = [compat_urllib_parse.unquote(l) for l in links] + links = re.findall(r'\s(?:file|url):\s*["\']([^\'"]+)["\']', webpage) + links = map(compat_urllib_parse.unquote, links) + + thumbnail = None + video_url = None for link in links: - root, pathext = os.path.splitext(compat_urllib_parse_urlparse(link).path) - if pathext == '.png': + if link.endswith('.png'): thumbnail = link - elif pathext == '.mp4': - url = link - ext = pathext + elif '/videos/' in link: + video_url = link + if not video_url: + raise ExtractorError(u'Could not find video URL') + ext = u'.' + determine_ext(video_url) if ext == title[-len(ext):]: title = title[:-len(ext)] - ext = ext[1:] - return [{ + + return { 'id': video_id, - 'url': url, - 'ext': ext, + 'url': video_url, 'title': title, 'thumbnail': thumbnail, - }] + } diff --git a/youtube_dl/extractor/bambuser.py b/youtube_dl/extractor/bambuser.py new file mode 100644 index 000000000..d48c0c38d --- /dev/null +++ b/youtube_dl/extractor/bambuser.py @@ -0,0 +1,86 @@ +import re +import json +import itertools + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_request, +) + + +class BambuserIE(InfoExtractor): + IE_NAME = u'bambuser' + _VALID_URL = r'https?://bambuser\.com/v/(?P\d+)' + _API_KEY = '005f64509e19a868399060af746a00aa' + + _TEST = { + u'url': u'http://bambuser.com/v/4050584', + # MD5 seems to be flaky, see https://travis-ci.org/rg3/youtube-dl/jobs/14051016#L388 + #u'md5': u'fba8f7693e48fd4e8641b3fd5539a641', + u'info_dict': { + u'id': u'4050584', + u'ext': u'flv', + u'title': u'Education engineering days - lightning talks', + u'duration': 3741, + u'uploader': u'pixelversity', + u'uploader_id': u'344706', + }, + u'params': { + # It doesn't respect the 'Range' header, it would download the whole video + # caused the travis builds to fail: https://travis-ci.org/rg3/youtube-dl/jobs/14493845#L59 + u'skip_download': True, + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + info_url = ('http://player-c.api.bambuser.com/getVideo.json?' + '&api_key=%s&vid=%s' % (self._API_KEY, video_id)) + info_json = self._download_webpage(info_url, video_id) + info = json.loads(info_json)['result'] + + return { + 'id': video_id, + 'title': info['title'], + 'url': info['url'], + 'thumbnail': info.get('preview'), + 'duration': int(info['length']), + 'view_count': int(info['views_total']), + 'uploader': info['username'], + 'uploader_id': info['uid'], + } + + +class BambuserChannelIE(InfoExtractor): + IE_NAME = u'bambuser:channel' + _VALID_URL = r'https?://bambuser\.com/channel/(?P.*?)(?:/|#|\?|$)' + # The maximum number we can get with each request + _STEP = 50 + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + user = mobj.group('user') + urls = [] + last_id = '' + for i in itertools.count(1): + req_url = ('http://bambuser.com/xhr-api/index.php?username={user}' + '&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) + # Without setting this header, we wouldn't get any result + req.add_header('Referer', 'http://bambuser.com/channel/%s' % user) + info_json = self._download_webpage(req, user, + u'Downloading page %d' % i) + results = json.loads(info_json)['result'] + if len(results) == 0: + break + last_id = results[-1]['vid'] + urls.extend(self.url_result(v['page'], 'Bambuser') for v in results) + + return { + '_type': 'playlist', + 'title': user, + 'entries': urls, + } diff --git a/youtube_dl/extractor/bandcamp.py b/youtube_dl/extractor/bandcamp.py index 129a20f44..3a32c14c5 100644 --- a/youtube_dl/extractor/bandcamp.py +++ b/youtube_dl/extractor/bandcamp.py @@ -3,13 +3,16 @@ import re from .common import InfoExtractor from ..utils import ( + compat_str, + compat_urlparse, ExtractorError, ) class BandcampIE(InfoExtractor): + IE_NAME = u'Bandcamp' _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P.*)' - _TEST = { + _TESTS = [{ u'url': u'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song', u'file': u'1812978515.mp3', u'md5': u'cdeb30cdae1921719a3cbcab696ef53c', @@ -17,7 +20,7 @@ class BandcampIE(InfoExtractor): u"title": u"youtube-dl test song \"'/\\\u00e4\u21ad" }, u'skip': u'There is a limit of 200 free downloads / month for the test song' - } + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -26,6 +29,23 @@ class BandcampIE(InfoExtractor): # We get the link to the free download page m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage) if m_download is None: + m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage) + if m_trackinfo: + json_code = m_trackinfo.group(1) + data = json.loads(json_code) + + for d in data: + formats = [{ + 'format_id': 'format_id', + 'url': format_url, + 'ext': format_id.partition('-')[0] + } for format_id, format_url in sorted(d['file'].items())] + return { + 'id': compat_str(d['id']), + 'title': d['title'], + 'formats': formats, + } + else: raise ExtractorError(u'No free songs found') download_link = m_download.group(1) @@ -61,3 +81,49 @@ class BandcampIE(InfoExtractor): } return [track_info] + + +class BandcampAlbumIE(InfoExtractor): + IE_NAME = u'Bandcamp:album' + _VALID_URL = r'http://.*?\.bandcamp\.com/album/(?P<title>.*)' + + _TEST = { + u'url': u'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1', + u'playlist': [ + { + u'file': u'1353101989.mp3', + u'md5': u'39bc1eded3476e927c724321ddf116cf', + u'info_dict': { + u'title': u'Intro', + } + }, + { + u'file': u'38097443.mp3', + u'md5': u'1a2c32e2691474643e912cc6cd4bffaa', + u'info_dict': { + u'title': u'Kero One - Keep It Alive (Blazo remix)', + } + }, + ], + u'params': { + u'playlistend': 2 + }, + u'skip': u'Bancamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test' + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + title = mobj.group('title') + webpage = self._download_webpage(url, title) + tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage) + if not tracks_paths: + raise ExtractorError(u'The page doesn\'t contain any track') + 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, u'title') + return { + '_type': 'playlist', + 'title': title, + 'entries': entries, + } diff --git a/youtube_dl/extractor/blinkx.py b/youtube_dl/extractor/blinkx.py new file mode 100644 index 000000000..e118f2e9f --- /dev/null +++ b/youtube_dl/extractor/blinkx.py @@ -0,0 +1,90 @@ +import datetime +import json +import re + +from .common import InfoExtractor +from ..utils import ( + remove_start, +) + + +class BlinkxIE(InfoExtractor): + _VALID_URL = r'^(?:https?://(?:www\.)blinkx\.com/ce/|blinkx:)(?P<id>[^?]+)' + _IE_NAME = u'blinkx' + + _TEST = { + u'url': u'http://www.blinkx.com/ce/8aQUy7GVFYgFzpKhT0oqsilwOGFRVXk3R1ZGWWdGenBLaFQwb3FzaWx3OGFRVXk3R1ZGWWdGenB', + u'file': u'8aQUy7GV.mp4', + u'md5': u'2e9a07364af40163a908edbf10bb2492', + u'info_dict': { + u"title": u"Police Car Rolls Away", + u"uploader": u"stupidvideos.com", + u"upload_date": u"20131215", + u"description": u"A police car gently rolls away from a fight. Maybe it felt weird being around a confrontation and just had to get out of there!", + u"duration": 14.886, + u"thumbnails": [{ + "width": 100, + "height": 76, + "url": "http://cdn.blinkx.com/stream/b/41/StupidVideos/20131215/1873969261/1873969261_tn_0.jpg", + }], + }, + } + + def _real_extract(self, url): + m = re.match(self._VALID_URL, url) + video_id = m.group('id') + display_id = video_id[:8] + + api_url = (u'https://apib4.blinkx.com/api.php?action=play_video&' + + u'video=%s' % video_id) + data_json = self._download_webpage(api_url, display_id) + data = json.loads(data_json)['api']['results'][0] + dt = datetime.datetime.fromtimestamp(data['pubdate_epoch']) + upload_date = dt.strftime('%Y%m%d') + + duration = None + thumbnails = [] + formats = [] + for m in data['media']: + if m['type'] == 'jpg': + thumbnails.append({ + 'url': m['link'], + 'width': int(m['w']), + 'height': int(m['h']), + }) + elif m['type'] == 'original': + duration = m['d'] + elif m['type'] == 'youtube': + yt_id = m['link'] + self.to_screen(u'Youtube video detected: %s' % yt_id) + return self.url_result(yt_id, 'Youtube', video_id=yt_id) + elif m['type'] in ('flv', 'mp4'): + vcodec = remove_start(m['vcodec'], 'ff') + acodec = remove_start(m['acodec'], 'ff') + format_id = (u'%s-%sk-%s' % + (vcodec, + (int(m['vbr']) + int(m['abr'])) // 1000, + m['w'])) + formats.append({ + 'format_id': format_id, + 'url': m['link'], + 'vcodec': vcodec, + 'acodec': acodec, + 'abr': int(m['abr']) // 1000, + 'vbr': int(m['vbr']) // 1000, + 'width': int(m['w']), + 'height': int(m['h']), + }) + formats.sort(key=lambda f: (f['width'], f['vbr'], f['abr'])) + + return { + 'id': display_id, + 'fullid': video_id, + 'title': data['title'], + 'formats': formats, + 'uploader': data['channel_name'], + 'upload_date': upload_date, + 'description': data.get('description'), + 'thumbnails': thumbnails, + 'duration': duration, + } diff --git a/youtube_dl/extractor/bliptv.py b/youtube_dl/extractor/bliptv.py index 493504f75..5e33a69df 100644 --- a/youtube_dl/extractor/bliptv.py +++ b/youtube_dl/extractor/bliptv.py @@ -51,8 +51,7 @@ class BlipTVIE(InfoExtractor): url = 'http://blip.tv/play/g_%s' % api_mobj.group('video_id') urlp = compat_urllib_parse_urlparse(url) if urlp.path.startswith('/play/'): - request = compat_urllib_request.Request(url) - response = compat_urllib_request.urlopen(request) + response = self._request_webpage(url, None, False) redirecturl = response.geturl() rurlp = compat_urllib_parse_urlparse(redirecturl) file_id = compat_parse_qs(rurlp.fragment)['file'][0].rpartition('/')[2] @@ -69,25 +68,23 @@ class BlipTVIE(InfoExtractor): request.add_header('User-Agent', 'iTunes/10.6.1') self.report_extraction(mobj.group(1)) info = None - try: - urlh = compat_urllib_request.urlopen(request) - if urlh.headers.get('Content-Type', '').startswith('video/'): # Direct download - basename = url.split('/')[-1] - title,ext = os.path.splitext(basename) - title = title.decode('UTF-8') - ext = ext.replace('.', '') - self.report_direct_download(title) - info = { - 'id': title, - 'url': url, - 'uploader': None, - 'upload_date': None, - 'title': title, - 'ext': ext, - 'urlhandle': urlh - } - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'ERROR: unable to download video info webpage: %s' % compat_str(err)) + urlh = self._request_webpage(request, None, False, + u'unable to download video info webpage') + if urlh.headers.get('Content-Type', '').startswith('video/'): # Direct download + basename = url.split('/')[-1] + title,ext = os.path.splitext(basename) + title = title.decode('UTF-8') + ext = ext.replace('.', '') + self.report_direct_download(title) + info = { + 'id': title, + 'url': url, + 'uploader': None, + 'upload_date': None, + 'title': title, + 'ext': ext, + 'urlhandle': urlh + } if info is None: # Regular URL try: json_code_bytes = urlh.read() diff --git a/youtube_dl/extractor/bloomberg.py b/youtube_dl/extractor/bloomberg.py index 3666a780b..755d9c9ef 100644 --- a/youtube_dl/extractor/bloomberg.py +++ b/youtube_dl/extractor/bloomberg.py @@ -4,7 +4,7 @@ from .common import InfoExtractor class BloombergIE(InfoExtractor): - _VALID_URL = r'https?://www\.bloomberg\.com/video/(?P<name>.+?).html' + _VALID_URL = r'https?://www\.bloomberg\.com/video/(?P<name>.+?)\.html' _TEST = { u'url': u'http://www.bloomberg.com/video/shah-s-presentation-on-foreign-exchange-strategies-qurhIVlJSB6hzkVi229d8g.html', diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py index 1392f382a..b1b7526ca 100644 --- a/youtube_dl/extractor/brightcove.py +++ b/youtube_dl/extractor/brightcove.py @@ -9,10 +9,13 @@ from ..utils import ( compat_urllib_parse, find_xpath_attr, compat_urlparse, + compat_str, + compat_urllib_request, ExtractorError, ) + class BrightcoveIE(InfoExtractor): _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)' _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s' @@ -23,7 +26,7 @@ class BrightcoveIE(InfoExtractor): # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/ u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001', u'file': u'2371591881001.mp4', - u'md5': u'9e80619e0a94663f0bdc849b4566af19', + u'md5': u'8eccab865181d29ec2958f32a6a754f5', u'note': u'Test Brightcove downloads and detection in GenericIE', u'info_dict': { u'title': u'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”', @@ -41,6 +44,29 @@ class BrightcoveIE(InfoExtractor): u'uploader': u'Oracle', }, }, + { + # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/ + u'url': u'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001', + u'info_dict': { + u'id': u'2750934548001', + u'ext': u'mp4', + u'title': u'This Bracelet Acts as a Personal Thermostat', + u'description': u'md5:547b78c64f4112766ccf4e151c20b6a0', + u'uploader': u'Mashable', + }, + }, + { + # test that the default referer works + # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/ + u'url': u'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001', + u'info_dict': { + u'id': u'2878862109001', + u'ext': u'mp4', + u'title': u'Lost in Motion II', + u'description': u'md5:363109c02998fee92ec02211bd8000df', + u'uploader': u'National Ballet of Canada', + }, + }, ] @classmethod @@ -61,31 +87,65 @@ class BrightcoveIE(InfoExtractor): params = {'flashID': object_doc.attrib['id'], 'playerID': find_xpath_attr(object_doc, './param', 'name', 'playerID').attrib['value'], } - playerKey = find_xpath_attr(object_doc, './param', 'name', 'playerKey') + def find_param(name): + node = find_xpath_attr(object_doc, './param', 'name', name) + if node is not None: + return node.attrib['value'] + return None + playerKey = find_param('playerKey') # Not all pages define this value if playerKey is not None: - params['playerKey'] = playerKey.attrib['value'] - videoPlayer = find_xpath_attr(object_doc, './param', 'name', '@videoPlayer') + params['playerKey'] = playerKey + # The three fields hold the id of the video + videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') if videoPlayer is not None: - params['@videoPlayer'] = videoPlayer.attrib['value'] + params['@videoPlayer'] = videoPlayer + linkBase = find_param('linkBaseURL') + if linkBase is not None: + params['linkBaseURL'] = linkBase data = compat_urllib_parse.urlencode(params) return cls._FEDERATED_URL_TEMPLATE % data + @classmethod + def _extract_brightcove_url(cls, webpage): + """Try to extract the brightcove url from the wepbage, returns None + if it can't be found + """ + m_brightcove = re.search( + r'<object[^>]+?class=([\'"])[^>]*?BrightcoveExperience.*?\1.+?</object>', + webpage, re.DOTALL) + if m_brightcove is not None: + return cls._build_brighcove_url(m_brightcove.group()) + else: + return None + def _real_extract(self, url): + # Change the 'videoId' and others field to '@videoPlayer' + url = re.sub(r'(?<=[?&])(videoI(d|D)|bctid)', '%40videoPlayer', url) + # Change bckey (used by bcove.me urls) to playerKey + url = re.sub(r'(?<=[?&])bckey', 'playerKey', url) mobj = re.match(self._VALID_URL, url) query_str = mobj.group('query') query = compat_urlparse.parse_qs(query_str) videoPlayer = query.get('@videoPlayer') if videoPlayer: - return self._get_video_info(videoPlayer[0], query_str) + return self._get_video_info(videoPlayer[0], query_str, query, + # We set the original url as the default 'Referer' header + referer=url) else: player_key = query['playerKey'] return self._get_playlist_info(player_key[0]) - def _get_video_info(self, video_id, query): - request_url = self._FEDERATED_URL_TEMPLATE % query - webpage = self._download_webpage(request_url, video_id) + 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) + linkBase = query.get('linkBaseURL') + if linkBase is not None: + referer = linkBase[0] + if referer is not None: + req.add_header('Referer', referer) + webpage = self._download_webpage(req, video_id) self.report_extraction(video_id) info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json') @@ -109,7 +169,7 @@ class BrightcoveIE(InfoExtractor): def _extract_video_info(self, video_info): info = { - 'id': video_info['id'], + 'id': compat_str(video_info['id']), 'title': video_info['displayName'], 'description': video_info.get('shortDescription'), 'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'), @@ -119,15 +179,14 @@ class BrightcoveIE(InfoExtractor): renditions = video_info.get('renditions') if renditions: renditions = sorted(renditions, key=lambda r: r['size']) - best_format = renditions[-1] - info.update({ - 'url': best_format['defaultURL'], - 'ext': 'mp4', - }) + info['formats'] = [{ + 'url': rend['defaultURL'], + 'height': rend.get('frameHeight'), + 'width': rend.get('frameWidth'), + } for rend in renditions] elif video_info.get('FLVFullLengthURL') is not None: info.update({ 'url': video_info['FLVFullLengthURL'], - 'ext': 'flv', }) else: raise ExtractorError(u'Unable to extract video url for %s' % info['id']) diff --git a/youtube_dl/extractor/canalc2.py b/youtube_dl/extractor/canalc2.py index e7f4fa9fd..3d8d7f9d2 100644 --- a/youtube_dl/extractor/canalc2.py +++ b/youtube_dl/extractor/canalc2.py @@ -6,7 +6,7 @@ from .common import InfoExtractor class Canalc2IE(InfoExtractor): IE_NAME = 'canalc2.tv' - _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?idVideo=(\d+)&voir=oui' + _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?.*?idVideo=(?P<id>\d+)' _TEST = { u'url': u'http://www.canalc2.tv/video.asp?idVideo=12163&voir=oui', @@ -18,7 +18,9 @@ class Canalc2IE(InfoExtractor): } def _real_extract(self, url): - video_id = re.match(self._VALID_URL, url).group(1) + 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 webpage = self._download_webpage(url, video_id) file_name = self._search_regex( r"so\.addVariable\('file','(.*?)'\);", diff --git a/youtube_dl/extractor/canalplus.py b/youtube_dl/extractor/canalplus.py index 1db9b24cf..7cdcd8399 100644 --- a/youtube_dl/extractor/canalplus.py +++ b/youtube_dl/extractor/canalplus.py @@ -1,10 +1,10 @@ # encoding: utf-8 import re -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import unified_strdate + class CanalplusIE(InfoExtractor): _VALID_URL = r'https?://(www\.canalplus\.fr/.*?/(?P<path>.*)|player\.canalplus\.fr/#/(?P<id>\d+))' _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/cplus/%s' @@ -25,16 +25,15 @@ class CanalplusIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = mobj.groupdict().get('id') if video_id is None: webpage = self._download_webpage(url, mobj.group('path')) video_id = self._search_regex(r'videoId = "(\d+)";', webpage, u'video id') info_url = self._VIDEO_INFO_TEMPLATE % video_id - info_page = self._download_webpage(info_url,video_id, + doc = self._download_xml(info_url,video_id, u'Downloading video info') self.report_extraction(video_id) - doc = xml.etree.ElementTree.fromstring(info_page.encode('utf-8')) video_info = [video for video in doc if video.find('ID').text == video_id][0] infos = video_info.find('INFOS') media = video_info.find('MEDIA') diff --git a/youtube_dl/extractor/cbs.py b/youtube_dl/extractor/cbs.py new file mode 100644 index 000000000..ac0315853 --- /dev/null +++ b/youtube_dl/extractor/cbs.py @@ -0,0 +1,30 @@ +import re + +from .common import InfoExtractor + + +class CBSIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?cbs\.com/shows/[^/]+/video/(?P<id>[^/]+)/.*' + + _TEST = { + u'url': u'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/', + u'file': u'4JUVEwq3wUT7.flv', + u'info_dict': { + u'title': u'Connect Chat feat. Garth Brooks', + u'description': u'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!', + u'duration': 1495, + }, + u'params': { + # rtmp download + u'skip_download': True, + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + webpage = self._download_webpage(url, video_id) + real_id = self._search_regex( + r"video\.settings\.pid\s*=\s*'([^']+)';", + webpage, u'real video ID') + return self.url_result(u'theplatform:%s' % real_id) diff --git a/youtube_dl/extractor/channel9.py b/youtube_dl/extractor/channel9.py new file mode 100644 index 000000000..ae70ea229 --- /dev/null +++ b/youtube_dl/extractor/channel9.py @@ -0,0 +1,267 @@ +# encoding: utf-8 + +import re + +from .common import InfoExtractor +from ..utils import ExtractorError + +class Channel9IE(InfoExtractor): + ''' + Common extractor for channel9.msdn.com. + + The type of provided URL (video or playlist) is determined according to + meta Search.PageType from web page HTML rather than URL itself, as it is + not always possible to do. + ''' + IE_DESC = u'Channel 9' + IE_NAME = u'channel9' + _VALID_URL = r'^https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+)/?' + + _TESTS = [ + { + u'url': u'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002', + u'file': u'Events_TechEd_Australia_2013_KOS002.mp4', + u'md5': u'bbd75296ba47916b754e73c3a4bbdf10', + u'info_dict': { + u'title': u'Developer Kick-Off Session: Stuff We Love', + u'description': u'md5:c08d72240b7c87fcecafe2692f80e35f', + u'duration': 4576, + u'thumbnail': u'http://media.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg', + u'session_code': u'KOS002', + u'session_day': u'Day 1', + u'session_room': u'Arena 1A', + u'session_speakers': [ u'Ed Blankenship', u'Andrew Coates', u'Brady Gaster', u'Patrick Klug', u'Mads Kristensen' ], + }, + }, + { + u'url': u'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing', + u'file': u'posts_Self-service-BI-with-Power-BI-nuclear-testing.mp4', + u'md5': u'b43ee4529d111bc37ba7ee4f34813e68', + u'info_dict': { + u'title': u'Self-service BI with Power BI - nuclear testing', + u'description': u'md5:d1e6ecaafa7fb52a2cacdf9599829f5b', + u'duration': 1540, + u'thumbnail': u'http://media.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg', + u'authors': [ u'Mike Wilmot' ], + }, + } + ] + + _RSS_URL = 'http://channel9.msdn.com/%s/RSS' + + # Sorted by quality + _known_formats = ['MP3', 'MP4', 'Mid Quality WMV', 'Mid Quality MP4', 'High Quality WMV', 'High Quality MP4'] + + def _restore_bytes(self, formatted_size): + if not formatted_size: + return 0 + m = re.match(r'^(?P<size>\d+(?:\.\d+)?)\s+(?P<units>[a-zA-Z]+)', formatted_size) + if not m: + return 0 + units = m.group('units') + try: + exponent = [u'B', u'KB', u'MB', u'GB', u'TB', u'PB', u'EB', u'ZB', u'YB'].index(units.upper()) + except ValueError: + return 0 + size = float(m.group('size')) + return int(size * (1024 ** exponent)) + + def _formats_from_html(self, html): + FORMAT_REGEX = r''' + (?x) + <a\s+href="(?P<url>[^"]+)">(?P<quality>[^<]+)</a>\s* + <span\s+class="usage">\((?P<note>[^\)]+)\)</span>\s* + (?:<div\s+class="popup\s+rounded">\s* + <h3>File\s+size</h3>\s*(?P<filesize>.*?)\s* + </div>)? # File size part may be missing + ''' + # Extract known formats + formats = [{'url': x.group('url'), + 'format_id': x.group('quality'), + 'format_note': x.group('note'), + 'format': '%s (%s)' % (x.group('quality'), x.group('note')), + 'filesize': self._restore_bytes(x.group('filesize')), # File size is approximate + } for x in list(re.finditer(FORMAT_REGEX, html)) if x.group('quality') in self._known_formats] + # Sort according to known formats list + formats.sort(key=lambda fmt: self._known_formats.index(fmt['format_id'])) + return formats + + def _extract_title(self, html): + title = self._html_search_meta(u'title', html, u'title') + if title is None: + title = self._og_search_title(html) + TITLE_SUFFIX = u' (Channel 9)' + if title is not None and title.endswith(TITLE_SUFFIX): + title = title[:-len(TITLE_SUFFIX)] + return title + + def _extract_description(self, html): + DESCRIPTION_REGEX = r'''(?sx) + <div\s+class="entry-content">\s* + <div\s+id="entry-body">\s* + (?P<description>.+?)\s* + </div>\s* + </div> + ''' + m = re.search(DESCRIPTION_REGEX, html) + if m is not None: + return m.group('description') + return self._html_search_meta(u'description', html, u'description') + + def _extract_duration(self, html): + m = re.search(r'data-video_duration="(?P<hours>\d{2}):(?P<minutes>\d{2}):(?P<seconds>\d{2})"', html) + return ((int(m.group('hours')) * 60 * 60) + (int(m.group('minutes')) * 60) + int(m.group('seconds'))) if m else None + + def _extract_slides(self, html): + m = re.search(r'<a href="(?P<slidesurl>[^"]+)" class="slides">Slides</a>', html) + return m.group('slidesurl') if m is not None else None + + def _extract_zip(self, html): + m = re.search(r'<a href="(?P<zipurl>[^"]+)" class="zip">Zip</a>', html) + return m.group('zipurl') if m is not None else None + + def _extract_avg_rating(self, html): + m = re.search(r'<p class="avg-rating">Avg Rating: <span>(?P<avgrating>[^<]+)</span></p>', html) + return float(m.group('avgrating')) if m is not None else 0 + + def _extract_rating_count(self, html): + m = re.search(r'<div class="rating-count">\((?P<ratingcount>[^<]+)\)</div>', html) + return int(self._fix_count(m.group('ratingcount'))) if m is not None else 0 + + def _extract_view_count(self, html): + m = re.search(r'<li class="views">\s*<span class="count">(?P<viewcount>[^<]+)</span> Views\s*</li>', html) + return int(self._fix_count(m.group('viewcount'))) if m is not None else 0 + + def _extract_comment_count(self, html): + m = re.search(r'<li class="comments">\s*<a href="#comments">\s*<span class="count">(?P<commentcount>[^<]+)</span> Comments\s*</a>\s*</li>', html) + return int(self._fix_count(m.group('commentcount'))) if m is not None else 0 + + def _fix_count(self, count): + return int(str(count).replace(',', '')) if count is not None else None + + def _extract_authors(self, html): + m = re.search(r'(?s)<li class="author">(.*?)</li>', html) + if m is None: + return None + return re.findall(r'<a href="/Niners/[^"]+">([^<]+)</a>', m.group(1)) + + def _extract_session_code(self, html): + m = re.search(r'<li class="code">\s*(?P<code>.+?)\s*</li>', html) + return m.group('code') if m is not None else None + + def _extract_session_day(self, html): + m = re.search(r'<li class="day">\s*<a href="/Events/[^"]+">(?P<day>[^<]+)</a>\s*</li>', html) + return m.group('day') if m is not None else None + + def _extract_session_room(self, html): + m = re.search(r'<li class="room">\s*(?P<room>.+?)\s*</li>', html) + return m.group('room') if m is not None else None + + def _extract_session_speakers(self, html): + return re.findall(r'<a href="/Events/Speakers/[^"]+">([^<]+)</a>', html) + + def _extract_content(self, html, content_path): + # Look for downloadable content + formats = self._formats_from_html(html) + slides = self._extract_slides(html) + zip_ = self._extract_zip(html) + + # Nothing to download + if len(formats) == 0 and slides is None and zip_ is None: + self._downloader.report_warning(u'None of recording, slides or zip are available for %s' % content_path) + return + + # Extract meta + title = self._extract_title(html) + description = self._extract_description(html) + thumbnail = self._og_search_thumbnail(html) + duration = self._extract_duration(html) + avg_rating = self._extract_avg_rating(html) + rating_count = self._extract_rating_count(html) + view_count = self._extract_view_count(html) + comment_count = self._extract_comment_count(html) + + common = {'_type': 'video', + 'id': content_path, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'avg_rating': avg_rating, + 'rating_count': rating_count, + 'view_count': view_count, + 'comment_count': comment_count, + } + + result = [] + + if slides is not None: + d = common.copy() + d.update({ 'title': title + '-Slides', 'url': slides }) + result.append(d) + + if zip_ is not None: + d = common.copy() + d.update({ 'title': title + '-Zip', 'url': zip_ }) + result.append(d) + + if len(formats) > 0: + d = common.copy() + d.update({ 'title': title, 'formats': formats }) + result.append(d) + + return result + + def _extract_entry_item(self, html, content_path): + contents = self._extract_content(html, content_path) + if contents is None: + return contents + + authors = self._extract_authors(html) + + for content in contents: + content['authors'] = authors + + return contents + + def _extract_session(self, html, content_path): + contents = self._extract_content(html, content_path) + if contents is None: + return contents + + session_meta = {'session_code': self._extract_session_code(html), + 'session_day': self._extract_session_day(html), + 'session_room': self._extract_session_room(html), + 'session_speakers': self._extract_session_speakers(html), + } + + for content in contents: + content.update(session_meta) + + return contents + + def _extract_list(self, content_path): + rss = self._download_xml(self._RSS_URL % content_path, content_path, u'Downloading RSS') + entries = [self.url_result(session_url.text, 'Channel9') + for session_url in rss.findall('./channel/item/link')] + title_text = rss.find('./channel/title').text + return self.playlist_result(entries, content_path, title_text) + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + content_path = mobj.group('contentpath') + + webpage = self._download_webpage(url, content_path, u'Downloading web page') + + page_type_m = re.search(r'<meta name="Search.PageType" content="(?P<pagetype>[^"]+)"/>', webpage) + if page_type_m is None: + raise ExtractorError(u'Search.PageType not found, don\'t know how to process this page', expected=True) + + page_type = page_type_m.group('pagetype') + if page_type == 'List': # List page, may contain list of 'item'-like objects + return self._extract_list(content_path) + elif page_type == 'Entry.Item': # Any 'item'-like page, may contain downloadable content + return self._extract_entry_item(webpage, content_path) + elif page_type == 'Session': # Event session page, may contain downloadable content + return self._extract_session(webpage, content_path) + else: + raise ExtractorError(u'Unexpected Search.PageType %s' % page_type, expected=True) \ No newline at end of file diff --git a/youtube_dl/extractor/cinemassacre.py b/youtube_dl/extractor/cinemassacre.py index 2fe1033f0..f0d08cebf 100644 --- a/youtube_dl/extractor/cinemassacre.py +++ b/youtube_dl/extractor/cinemassacre.py @@ -41,7 +41,7 @@ class CinemassacreIE(InfoExtractor): webpage_url = u'http://' + mobj.group('url') webpage = self._download_webpage(webpage_url, None) # Don't know video id yet video_date = mobj.group('date_Y') + mobj.group('date_m') + mobj.group('date_d') - mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/(?:embed|player)\.php\?id=(?:Cinemassacre-)?(?P<video_id>.+?))"', webpage) + mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?id=(?:Cinemassacre-)?(?P<video_id>.+?))"', webpage) if not mobj: raise ExtractorError(u'Can\'t extract embed url and video id') playerdata_url = mobj.group(u'embed_url') @@ -65,6 +65,7 @@ class CinemassacreIE(InfoExtractor): { 'url': url, 'play_path': 'mp4:' + sd_file, + 'rtmp_live': True, # workaround 'ext': 'flv', 'format': 'sd', 'format_id': 'sd', @@ -72,6 +73,7 @@ class CinemassacreIE(InfoExtractor): { 'url': url, 'play_path': 'mp4:' + hd_file, + 'rtmp_live': True, # workaround 'ext': 'flv', 'format': 'hd', 'format_id': 'hd', diff --git a/youtube_dl/extractor/clipfish.py b/youtube_dl/extractor/clipfish.py new file mode 100644 index 000000000..43efb08bf --- /dev/null +++ b/youtube_dl/extractor/clipfish.py @@ -0,0 +1,58 @@ +import re +import time +import xml.etree.ElementTree + +from .common import InfoExtractor +from ..utils import ExtractorError + + +class ClipfishIE(InfoExtractor): + IE_NAME = u'clipfish' + + _VALID_URL = r'^https?://(?:www\.)?clipfish\.de/.*?/video/(?P<id>[0-9]+)/' + _TEST = { + u'url': u'http://www.clipfish.de/special/game-trailer/video/3966754/fifa-14-e3-2013-trailer/', + u'file': u'3966754.mp4', + u'md5': u'2521cd644e862936cf2e698206e47385', + u'info_dict': { + u'title': u'FIFA 14 - E3 2013 Trailer', + u'duration': 82, + }, + u'skip': 'Blocked in the US' + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group(1) + + info_url = ('http://www.clipfish.de/devxml/videoinfo/%s?ts=%d' % + (video_id, int(time.time()))) + doc = self._download_xml( + info_url, video_id, note=u'Downloading info page') + title = doc.find('title').text + video_url = doc.find('filename').text + if video_url is None: + xml_bytes = xml.etree.ElementTree.tostring(doc) + raise ExtractorError(u'Cannot find video URL in document %r' % + xml_bytes) + thumbnail = doc.find('imageurl').text + duration_str = doc.find('duration').text + m = re.match( + r'^(?P<hours>[0-9]+):(?P<minutes>[0-9]{2}):(?P<seconds>[0-9]{2}):(?P<ms>[0-9]*)$', + duration_str) + if m: + duration = ( + (int(m.group('hours')) * 60 * 60) + + (int(m.group('minutes')) * 60) + + (int(m.group('seconds'))) + ) + else: + duration = None + + return { + 'id': video_id, + 'title': title, + 'url': video_url, + 'thumbnail': thumbnail, + 'duration': duration, + } diff --git a/youtube_dl/extractor/clipsyndicate.py b/youtube_dl/extractor/clipsyndicate.py new file mode 100644 index 000000000..c60089ad3 --- /dev/null +++ b/youtube_dl/extractor/clipsyndicate.py @@ -0,0 +1,50 @@ +import re + +from .common import InfoExtractor +from ..utils import ( + find_xpath_attr, + fix_xml_all_ampersand, +) + + +class ClipsyndicateIE(InfoExtractor): + _VALID_URL = r'http://www\.clipsyndicate\.com/video/play(list/\d+)?/(?P<id>\d+)' + + _TEST = { + u'url': u'http://www.clipsyndicate.com/video/play/4629301/brick_briscoe', + u'md5': u'4d7d549451bad625e0ff3d7bd56d776c', + u'info_dict': { + u'id': u'4629301', + u'ext': u'mp4', + u'title': u'Brick Briscoe', + u'duration': 612, + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + js_player = self._download_webpage( + 'http://eplayer.clipsyndicate.com/embed/player.js?va_id=%s' % video_id, + video_id, u'Downlaoding player') + # it includes a required token + flvars = self._search_regex(r'flvars: "(.*?)"', js_player, u'flvars') + + pdoc = self._download_xml( + 'http://eplayer.clipsyndicate.com/osmf/playlist?%s' % flvars, + video_id, u'Downloading video info', + transform_source=fix_xml_all_ampersand) + + track_doc = pdoc.find('trackList/track') + def find_param(name): + node = find_xpath_attr(track_doc, './/param', 'name', name) + if node is not None: + return node.attrib['value'] + + return { + 'id': video_id, + 'title': find_param('title'), + 'url': track_doc.find('location').text, + 'thumbnail': find_param('thumbnail'), + 'duration': int(find_param('duration')), + } diff --git a/youtube_dl/extractor/cnn.py b/youtube_dl/extractor/cnn.py index a79f881cd..a034bb2fb 100644 --- a/youtube_dl/extractor/cnn.py +++ b/youtube_dl/extractor/cnn.py @@ -1,12 +1,11 @@ import re -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import determine_ext class CNNIE(InfoExtractor): - _VALID_URL = r'''(?x)https?://(edition\.)?cnn\.com/video/(data/.+?|\?)/ + _VALID_URL = r'''(?x)https?://((edition|www)\.)?cnn\.com/video/(data/.+?|\?)/ (?P<path>.+?/(?P<title>[^/]+?)(?:\.cnn|(?=&)))''' _TESTS = [{ @@ -33,8 +32,7 @@ class CNNIE(InfoExtractor): path = mobj.group('path') page_title = mobj.group('title') info_url = u'http://cnn.com/video/data/3.0/%s/index.xml' % path - info_xml = self._download_webpage(info_url, page_title) - info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8')) + info = self._download_xml(info_url, page_title) formats = [] for f in info.findall('files/file'): diff --git a/youtube_dl/extractor/collegehumor.py b/youtube_dl/extractor/collegehumor.py index 8d4c93d6d..b27c1dfc5 100644 --- a/youtube_dl/extractor/collegehumor.py +++ b/youtube_dl/extractor/collegehumor.py @@ -1,5 +1,4 @@ import re -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( @@ -46,11 +45,10 @@ class CollegeHumorIE(InfoExtractor): self.report_extraction(video_id) xmlUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id - metaXml = self._download_webpage(xmlUrl, video_id, + mdoc = self._download_xml(xmlUrl, video_id, u'Downloading info XML', u'Unable to download video info XML') - mdoc = xml.etree.ElementTree.fromstring(metaXml) try: videoNode = mdoc.findall('./video')[0] youtubeIdNode = videoNode.find('./youtubeID') @@ -65,16 +63,13 @@ class CollegeHumorIE(InfoExtractor): if next_url.endswith(u'manifest.f4m'): manifest_url = next_url + '?hdcore=2.10.3' - manifestXml = self._download_webpage(manifest_url, video_id, + adoc = self._download_xml(manifest_url, video_id, u'Downloading XML manifest', u'Unable to download video info XML') - adoc = xml.etree.ElementTree.fromstring(manifestXml) try: - media_node = adoc.findall('./{http://ns.adobe.com/f4m/1.0}media')[0] - node_id = media_node.attrib['url'] video_id = adoc.findall('./{http://ns.adobe.com/f4m/1.0}id')[0].text - except IndexError as err: + except IndexError: raise ExtractorError(u'Invalid manifest file') url_pr = compat_urllib_parse_urlparse(info['thumbnail']) info['url'] = url_pr.scheme + '://' + url_pr.netloc + video_id[:-2].replace('.csmil','').replace(',','') diff --git a/youtube_dl/extractor/comedycentral.py b/youtube_dl/extractor/comedycentral.py index 69b2beece..a54ce3ee7 100644 --- a/youtube_dl/extractor/comedycentral.py +++ b/youtube_dl/extractor/comedycentral.py @@ -1,7 +1,7 @@ import re -import xml.etree.ElementTree from .common import InfoExtractor +from .mtv import MTVServicesInfoExtractor from ..utils import ( compat_str, compat_urllib_parse, @@ -11,7 +11,31 @@ from ..utils import ( ) -class ComedyCentralIE(InfoExtractor): +class ComedyCentralIE(MTVServicesInfoExtractor): + _VALID_URL = r'https?://(?:www.)?comedycentral.com/(video-clips|episodes|cc-studios)/(?P<title>.*)' + _FEED_URL = u'http://comedycentral.com/feeds/mrss/' + + _TEST = { + u'url': u'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother', + u'md5': u'4167875aae411f903b751a21f357f1ee', + u'info_dict': { + u'id': u'cef0cbb3-e776-4bc9-b62e-8016deccb354', + u'ext': u'mp4', + u'title': u'Uncensored - Greg Fitzsimmons - Too Good of a Mother', + u'description': u'After a certain point, breastfeeding becomes c**kblocking.', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + title = mobj.group('title') + webpage = self._download_webpage(url, title) + mgid = self._search_regex(r'data-mgid="(?P<mgid>mgid:.*?)"', + webpage, u'mgid') + return self._get_videos_info(mgid) + + +class ComedyCentralShowsIE(InfoExtractor): IE_DESC = u'The Daily Show / Colbert Report' # urls can be abbreviations like :thedailyshow or :colbert # urls for episodes like: @@ -127,13 +151,12 @@ class ComedyCentralIE(InfoExtractor): uri = mMovieParams[0][1] indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri}) - indexXml = self._download_webpage(indexUrl, epTitle, + idoc = self._download_xml(indexUrl, epTitle, u'Downloading show index', u'unable to download episode index') results = [] - idoc = xml.etree.ElementTree.fromstring(indexXml) itemEls = idoc.findall('.//item') for partNum,itemEl in enumerate(itemEls): mediaId = itemEl.findall('./guid')[0].text @@ -144,10 +167,9 @@ class ComedyCentralIE(InfoExtractor): configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' + compat_urllib_parse.urlencode({'uri': mediaId})) - configXml = self._download_webpage(configUrl, epTitle, + cdoc = self._download_xml(configUrl, epTitle, u'Downloading configuration for %s' % shortMediaId) - cdoc = xml.etree.ElementTree.fromstring(configXml) turls = [] for rendition in cdoc.findall('.//rendition'): finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text) @@ -169,7 +191,7 @@ class ComedyCentralIE(InfoExtractor): }) effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1) - info = { + results.append({ 'id': shortMediaId, 'formats': formats, 'uploader': showId, @@ -177,11 +199,6 @@ class ComedyCentralIE(InfoExtractor): 'title': effTitle, 'thumbnail': None, 'description': compat_str(officialTitle), - } - - # TODO: Remove when #980 has been merged - info.update(info['formats'][-1]) - - results.append(info) + }) return results diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index cef4dce85..ba46a7bc7 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -4,11 +4,11 @@ import re import socket import sys import netrc +import xml.etree.ElementTree from ..utils import ( compat_http_client, compat_urllib_error, - compat_urllib_request, compat_str, clean_html, @@ -18,6 +18,8 @@ from ..utils import ( sanitize_filename, unescapeHTML, ) +_NO_DEFAULT = object() + class InfoExtractor(object): """Information Extractor class. @@ -33,15 +35,39 @@ class InfoExtractor(object): The dictionaries must include the following fields: id: Video identifier. - url: Final video URL. title: Video title, unescaped. - ext: Video filename extension. - Instead of url and ext, formats can also specified. + Additionally, it must contain either a formats entry or url and ext: + + formats: A list of dictionaries for each format available, it must + be ordered from worst to best quality. Potential fields: + * url Mandatory. The URL of the video file + * ext Will be calculated from url if missing + * format A human-readable description of the format + ("mp4 container with h264/opus"). + Calculated from the format_id, width, height. + and format_note fields if missing. + * format_id A short description of the format + ("mp4_h264_opus" or "19") + * format_note Additional info about the format + ("3D" or "DASH video") + * width Width of the video, if known + * height Height of the video, if known + * abr Average audio bitrate in KBit/s + * acodec Name of the audio codec in use + * vbr Average video bitrate in KBit/s + * vcodec Name of the video codec in use + * filesize The number of bytes, if known in advance + * player_url SWF Player URL (used for rtmpdump). + url: Final video URL. + ext: Video filename extension. + format: The video format, defaults to ext (used for --get-format) + player_url: SWF Player URL (used for rtmpdump). + urlhandle: [internal] The urlHandle to be used to download the file, + like returned by urllib.request.urlopen The following fields are optional: - format: The video format, defaults to ext (used for --get-format) thumbnails: A list of dictionaries (with the entries "resolution" and "url") for the varying thumbnails thumbnail: Full URL to a video thumbnail image. @@ -50,27 +76,17 @@ class InfoExtractor(object): upload_date: Video upload date (YYYYMMDD). uploader_id: Nickname or id of the video uploader. location: Physical location of the video. - player_url: SWF Player URL (used for rtmpdump). subtitles: The subtitle file contents as a dictionary in the format {language: subtitles}. + duration: Length of the video in seconds, as an integer. view_count: How many users have watched the video on the platform. - urlhandle: [internal] The urlHandle to be used to download the file, - like returned by urllib.request.urlopen + like_count: Number of positive ratings of the video + dislike_count: Number of negative ratings of the video + comment_count: Number of comments on the video age_limit: Age restriction for the video, as an integer (years) - formats: A list of dictionaries for each format available, it must - be ordered from worst to best quality. Potential fields: - * url Mandatory. The URL of the video file - * ext Will be calculated from url if missing - * format A human-readable description of the format - ("mp4 container with h264/opus"). - Calculated from the format_id, width, height. - and format_note fields if missing. - * format_id A short description of the format - ("mp4_h264_opus" or "19") - * format_note Additional info about the format - ("3D" or "DASH video") - * width Width of the video, if known - * height Height of the video, if known + webpage_url: The url to the video webpage, if given to youtube-dl it + should allow to get the same result again. (It will be set + by YoutubeDL if it's missing) Unless mentioned otherwise, the fields should be Unicode strings. @@ -142,27 +158,40 @@ class InfoExtractor(object): def IE_NAME(self): return type(self).__name__[:-2] - def _request_webpage(self, url_or_request, video_id, note=None, errnote=None): + def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): """ Returns the response handle """ if note is None: self.report_download_webpage(video_id) elif note is not False: - self.to_screen(u'%s: %s' % (video_id, note)) + if video_id is None: + self.to_screen(u'%s' % (note,)) + else: + self.to_screen(u'%s: %s' % (video_id, note)) try: - return compat_urllib_request.urlopen(url_or_request) + return self._downloader.urlopen(url_or_request) except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + if errnote is False: + return False if errnote is None: errnote = u'Unable to download webpage' - raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2], cause=err) + errmsg = u'%s: %s' % (errnote, compat_str(err)) + if fatal: + raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) + else: + self._downloader.report_warning(errmsg) + return False - def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None): + def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True): """ Returns a tuple (page content as string, URL handle) """ # Strip hashes from the URL (#1038) if isinstance(url_or_request, (compat_str, str)): url_or_request = url_or_request.partition('#')[0] - urlh = self._request_webpage(url_or_request, video_id, note, errnote) + urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal) + if urlh is False: + assert not fatal + return False content_type = urlh.headers.get('Content-Type', '') webpage_bytes = urlh.read() m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) @@ -197,9 +226,23 @@ class InfoExtractor(object): content = webpage_bytes.decode(encoding, 'replace') return (content, urlh) - def _download_webpage(self, url_or_request, video_id, note=None, errnote=None): + def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): """ Returns the data of the page as a string """ - return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0] + res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal) + if res is False: + return res + else: + content, _ = res + return content + + def _download_xml(self, url_or_request, video_id, + note=u'Downloading XML', errnote=u'Unable to download XML', + transform_source=None): + """Return the xml as an xml.etree.ElementTree.Element""" + xml_string = self._download_webpage(url_or_request, video_id, note, errnote) + if transform_source: + xml_string = transform_source(xml_string) + return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8')) def to_screen(self, msg): """Print msg to screen, prefixing it with '[ie_name]'""" @@ -222,14 +265,18 @@ class InfoExtractor(object): self.to_screen(u'Logging in') #Methods for following #608 - def url_result(self, url, ie=None): + @staticmethod + def url_result(url, ie=None, video_id=None): """Returns a url that points to a page that should be processed""" #TODO: ie should be the class used for getting the info video_info = {'_type': 'url', 'url': url, 'ie_key': ie} + if video_id is not None: + video_info['id'] = video_id return video_info - def playlist_result(self, entries, playlist_id=None, playlist_title=None): + @staticmethod + def playlist_result(entries, playlist_id=None, playlist_title=None): """Returns a playlist""" video_info = {'_type': 'playlist', 'entries': entries} @@ -239,7 +286,7 @@ class InfoExtractor(object): video_info['title'] = playlist_title return video_info - def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0): + def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0): """ Perform a regex search on the given string, using a single or a list of patterns returning the first matching group. @@ -253,7 +300,7 @@ class InfoExtractor(object): mobj = re.search(p, string, flags) if mobj: break - if sys.stderr.isatty() and os.name != 'nt': + if os.name != 'nt' and sys.stderr.isatty(): _name = u'\033[0;34m%s\033[0m' % name else: _name = name @@ -261,7 +308,7 @@ class InfoExtractor(object): if mobj: # return the first matching group return next(g for g in mobj.groups() if g is not None) - elif default is not None: + elif default is not _NO_DEFAULT: return default elif fatal: raise RegexNotFoundError(u'Unable to extract %s' % _name) @@ -270,7 +317,7 @@ class InfoExtractor(object): u'please report this issue on http://yt-dl.org/bug' % _name) return None - def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0): + def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0): """ Like _search_regex, but strips HTML tags and unescapes entities. """ @@ -312,13 +359,21 @@ class InfoExtractor(object): # Helper functions for extracting OpenGraph info @staticmethod - def _og_regex(prop): - return r'<meta.+?property=[\'"]og:%s[\'"].+?content=(?:"(.+?)"|\'(.+?)\')' % re.escape(prop) + def _og_regexes(prop): + content_re = r'content=(?:"([^>]+?)"|\'(.+?)\')' + property_re = r'property=[\'"]og:%s[\'"]' % re.escape(prop) + template = r'<meta[^>]+?%s[^>]+?%s' + return [ + template % (property_re, content_re), + template % (content_re, property_re), + ] def _og_search_property(self, prop, html, name=None, **kargs): if name is None: name = 'OpenGraph %s' % prop - escaped = self._search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs) + escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) + if escaped is None: + return None return unescapeHTML(escaped) def _og_search_thumbnail(self, html, **kargs): @@ -331,10 +386,22 @@ class InfoExtractor(object): return self._og_search_property('title', html, **kargs) def _og_search_video_url(self, html, name='video url', secure=True, **kargs): - regexes = [self._og_regex('video')] - if secure: regexes.insert(0, self._og_regex('video:secure_url')) + regexes = self._og_regexes('video') + if secure: regexes = self._og_regexes('video:secure_url') + regexes return self._html_search_regex(regexes, html, name, **kargs) + def _html_search_meta(self, name, html, display_name=None): + if display_name is None: + display_name = name + return self._html_search_regex( + r'''(?ix)<meta + (?=[^>]+(?:itemprop|name|property)=["\']%s["\']) + [^>]+content=["\']([^"\']+)["\']''' % re.escape(name), + html, display_name, fatal=False) + + def _dc_search_uploader(self, html): + return self._html_search_meta('dc.creator', html, 'uploader') + def _rta_search(self, html): # See http://www.rtalabel.org/index.php?content=howtofaq#single if re.search(r'(?ix)<meta\s+name="rating"\s+' @@ -343,6 +410,23 @@ class InfoExtractor(object): return 18 return 0 + def _media_rating_search(self, html): + # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ + rating = self._html_search_meta('rating', html) + + if not rating: + return None + + RATING_TABLE = { + 'safe for kids': 0, + 'general': 8, + '14 years': 14, + 'mature': 17, + 'restricted': 19, + } + return RATING_TABLE.get(rating.lower(), None) + + class SearchInfoExtractor(InfoExtractor): """ diff --git a/youtube_dl/extractor/cspan.py b/youtube_dl/extractor/cspan.py index 7bf03c584..d5730684d 100644 --- a/youtube_dl/extractor/cspan.py +++ b/youtube_dl/extractor/cspan.py @@ -6,7 +6,7 @@ from ..utils import ( ) class CSpanIE(InfoExtractor): - _VALID_URL = r'http://www.c-spanvideo.org/program/(.*)' + _VALID_URL = r'http://www\.c-spanvideo\.org/program/(.*)' _TEST = { u'url': u'http://www.c-spanvideo.org/program/HolderonV', u'file': u'315139.flv', diff --git a/youtube_dl/extractor/d8.py b/youtube_dl/extractor/d8.py new file mode 100644 index 000000000..a56842b16 --- /dev/null +++ b/youtube_dl/extractor/d8.py @@ -0,0 +1,22 @@ +# encoding: utf-8 +from .canalplus import CanalplusIE + + +class D8IE(CanalplusIE): + _VALID_URL = r'https?://www\.d8\.tv/.*?/(?P<path>.*)' + _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/d8/%s' + IE_NAME = u'd8.tv' + + _TEST = { + u'url': u'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html', + u'file': u'966289.flv', + u'info_dict': { + u'title': u'Campagne intime - Documentaire exceptionnel', + u'description': u'md5:d2643b799fb190846ae09c61e59a859f', + u'upload_date': u'20131108', + }, + u'params': { + # rtmp + u'skip_download': True, + }, + } diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 355b4ed0a..6685c94a3 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -11,6 +11,7 @@ from ..utils import ( get_element_by_attribute, get_element_by_id, orderedSet, + str_to_int, ExtractorError, ) @@ -27,7 +28,7 @@ class DailymotionBaseInfoExtractor(InfoExtractor): class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): """Information Extractor for Dailymotion""" - _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)' + _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)' IE_NAME = u'dailymotion' _FORMATS = [ @@ -80,7 +81,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): # Extract id and simplified title from URL mobj = re.match(self._VALID_URL, url) - video_id = mobj.group(1).split('_')[0].split('?')[0] + video_id = mobj.group('id') url = 'http://www.dailymotion.com/video/%s' % video_id @@ -100,10 +101,6 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): self.to_screen(u'Vevo video detected: %s' % vevo_id) return self.url_result(u'vevo:%s' % vevo_id, ie='Vevo') - video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>', - # Looking for official user - r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'], - webpage, 'video uploader', fatal=False) age_limit = self._rta_search(webpage) video_upload_date = None @@ -141,23 +138,29 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): raise ExtractorError(u'Unable to extract video URL') # subtitles - video_subtitles = self.extract_subtitles(video_id) + video_subtitles = self.extract_subtitles(video_id, webpage) if self._downloader.params.get('listsubtitles', False): - self._list_available_subtitles(video_id) + self._list_available_subtitles(video_id, webpage) return + view_count = self._search_regex( + r'video_views_count[^>]+>\s+([\d\.,]+)', webpage, u'view count', fatal=False) + if view_count is not None: + view_count = str_to_int(view_count) + return { 'id': video_id, 'formats': formats, - 'uploader': video_uploader, + 'uploader': info['owner_screenname'], 'upload_date': video_upload_date, 'title': self._og_search_title(webpage), 'subtitles': video_subtitles, 'thumbnail': info['thumbnail_url'], 'age_limit': age_limit, + 'view_count': view_count, } - def _get_available_subtitles(self, video_id): + def _get_available_subtitles(self, video_id, webpage): try: sub_list = self._download_webpage( 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id, @@ -186,7 +189,7 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor): webpage = self._download_webpage(request, id, u'Downloading page %s' % pagenum) - playlist_el = get_element_by_attribute(u'class', u'video_list', webpage) + playlist_el = get_element_by_attribute(u'class', u'row video_list', webpage) video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el)) if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None: diff --git a/youtube_dl/extractor/daum.py b/youtube_dl/extractor/daum.py index a804e83bd..4876ecb48 100644 --- a/youtube_dl/extractor/daum.py +++ b/youtube_dl/extractor/daum.py @@ -1,6 +1,5 @@ # encoding: utf-8 import re -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( @@ -10,7 +9,7 @@ from ..utils import ( class DaumIE(InfoExtractor): - _VALID_URL = r'https?://tvpot\.daum\.net/.*?clipid=(?P<id>\d+)' + _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/.*?clipid=(?P<id>\d+)' IE_NAME = u'daum.net' _TEST = { @@ -29,17 +28,16 @@ class DaumIE(InfoExtractor): video_id = mobj.group(1) canonical_url = 'http://tvpot.daum.net/v/%s' % video_id webpage = self._download_webpage(canonical_url, video_id) - full_id = self._search_regex(r'<link rel="video_src" href=".+?vid=(.+?)"', + full_id = self._search_regex( + r'<iframe src="http://videofarm.daum.net/controller/video/viewer/Video.html\?.*?vid=(.+?)[&"]', webpage, u'full id') query = compat_urllib_parse.urlencode({'vid': full_id}) - info_xml = self._download_webpage( + info = self._download_xml( 'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id, u'Downloading video info') - urls_xml = self._download_webpage( + urls = self._download_xml( 'http://videofarm.daum.net/controller/api/open/v1_2/MovieData.apixml?' + query, video_id, u'Downloading video formats info') - info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8')) - urls = xml.etree.ElementTree.fromstring(urls_xml.encode('utf-8')) self.to_screen(u'%s: Getting video urls' % video_id) formats = [] @@ -49,10 +47,9 @@ class DaumIE(InfoExtractor): 'vid': full_id, 'profile': profile, }) - url_xml = self._download_webpage( + url_doc = self._download_xml( 'http://videofarm.daum.net/controller/api/open/v1_2/MovieLocation.apixml?' + format_query, video_id, note=False) - url_doc = xml.etree.ElementTree.fromstring(url_xml.encode('utf-8')) format_url = url_doc.find('result/url').text formats.append({ 'url': format_url, @@ -60,7 +57,7 @@ class DaumIE(InfoExtractor): 'format_id': profile, }) - info = { + return { 'id': video_id, 'title': info.find('TITLE').text, 'formats': formats, @@ -69,6 +66,3 @@ class DaumIE(InfoExtractor): 'duration': int(info.find('DURATION').text), 'upload_date': info.find('REGDTTM').text[:8], } - # TODO: Remove when #980 has been merged - info.update(formats[-1]) - return info diff --git a/youtube_dl/extractor/depositfiles.py b/youtube_dl/extractor/depositfiles.py index d43348955..2c9fb5f2e 100644 --- a/youtube_dl/extractor/depositfiles.py +++ b/youtube_dl/extractor/depositfiles.py @@ -25,7 +25,7 @@ class DepositFilesIE(InfoExtractor): url = 'http://depositfiles.com/en/files/' + file_id # Retrieve file webpage with 'Free download' button pressed - free_download_indication = { 'gateway_result' : '1' } + free_download_indication = {'gateway_result' : '1'} request = compat_urllib_request.Request(url, compat_urllib_parse.urlencode(free_download_indication)) try: self.report_download_webpage(file_id) diff --git a/youtube_dl/extractor/dreisat.py b/youtube_dl/extractor/dreisat.py index 765cb1f37..cb7226f82 100644 --- a/youtube_dl/extractor/dreisat.py +++ b/youtube_dl/extractor/dreisat.py @@ -1,7 +1,6 @@ # coding: utf-8 import re -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( @@ -12,7 +11,7 @@ from ..utils import ( class DreiSatIE(InfoExtractor): IE_NAME = '3sat' - _VALID_URL = r'(?:http://)?(?:www\.)?3sat.de/mediathek/index.php\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$' + _VALID_URL = r'(?:http://)?(?:www\.)?3sat\.de/mediathek/index\.php\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$' _TEST = { u"url": u"http://www.3sat.de/mediathek/index.php?obj=36983", u'file': u'36983.webm', @@ -30,8 +29,7 @@ class DreiSatIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id - details_xml = self._download_webpage(details_url, video_id, note=u'Downloading video details') - details_doc = xml.etree.ElementTree.fromstring(details_xml.encode('utf-8')) + details_doc = self._download_xml(details_url, video_id, note=u'Downloading video details') thumbnail_els = details_doc.findall('.//teaserimage') thumbnails = [{ @@ -67,7 +65,7 @@ class DreiSatIE(InfoExtractor): return (qidx, prefer_http, format['video_bitrate']) formats.sort(key=_sortkey) - info = { + return { '_type': 'video', 'id': video_id, 'title': video_title, @@ -78,8 +76,3 @@ class DreiSatIE(InfoExtractor): 'uploader': video_uploader, 'upload_date': upload_date, } - - # TODO: Remove when #980 has been merged - info.update(formats[-1]) - - return info diff --git a/youtube_dl/extractor/ebaumsworld.py b/youtube_dl/extractor/ebaumsworld.py index f02c6998b..877113d63 100644 --- a/youtube_dl/extractor/ebaumsworld.py +++ b/youtube_dl/extractor/ebaumsworld.py @@ -1,5 +1,4 @@ import re -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import determine_ext @@ -21,9 +20,8 @@ class EbaumsWorldIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') - config_xml = self._download_webpage( + config = self._download_xml( 'http://www.ebaumsworld.com/video/player/%s' % video_id, video_id) - config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8')) video_url = config.find('file').text return { diff --git a/youtube_dl/extractor/eighttracks.py b/youtube_dl/extractor/eighttracks.py index 2cfbcd363..88f5526b8 100644 --- a/youtube_dl/extractor/eighttracks.py +++ b/youtube_dl/extractor/eighttracks.py @@ -1,4 +1,3 @@ -import itertools import json import random import re @@ -11,7 +10,7 @@ from ..utils import ( class EightTracksIE(InfoExtractor): IE_NAME = '8tracks' - _VALID_URL = r'https?://8tracks.com/(?P<user>[^/]+)/(?P<id>[^/#]+)(?:#.*)?$' + _VALID_URL = r'https?://8tracks\.com/(?P<user>[^/]+)/(?P<id>[^/#]+)(?:#.*)?$' _TEST = { u"name": u"EightTracks", u"url": u"http://8tracks.com/ytdl/youtube-dl-test-tracks-a", diff --git a/youtube_dl/extractor/eitb.py b/youtube_dl/extractor/eitb.py new file mode 100644 index 000000000..4ba323148 --- /dev/null +++ b/youtube_dl/extractor/eitb.py @@ -0,0 +1,37 @@ +# encoding: utf-8 +import re + +from .common import InfoExtractor +from .brightcove import BrightcoveIE +from ..utils import ExtractorError + + +class EitbIE(InfoExtractor): + IE_NAME = u'eitb.tv' + _VALID_URL = r'https?://www\.eitb\.tv/(eu/bideoa|es/video)/[^/]+/(?P<playlist_id>\d+)/(?P<chapter_id>\d+)' + + _TEST = { + u'add_ie': ['Brightcove'], + u'url': u'http://www.eitb.tv/es/video/60-minutos-60-minutos-2013-2014/2677100210001/2743577154001/lasa-y-zabala-30-anos/', + u'md5': u'edf4436247185adee3ea18ce64c47998', + u'info_dict': { + u'id': u'2743577154001', + u'ext': u'mp4', + u'title': u'60 minutos (Lasa y Zabala, 30 años)', + # All videos from eitb has this description in the brightcove info + u'description': u'.', + u'uploader': u'Euskal Telebista', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + chapter_id = mobj.group('chapter_id') + webpage = self._download_webpage(url, chapter_id) + bc_url = BrightcoveIE._extract_brightcove_url(webpage) + if bc_url is None: + raise ExtractorError(u'Could not extract the Brightcove url') + # The BrightcoveExperience object doesn't contain the video id, we set + # it manually + bc_url += '&%40videoPlayer={0}'.format(chapter_id) + return self.url_result(bc_url, BrightcoveIE.ie_key()) diff --git a/youtube_dl/extractor/escapist.py b/youtube_dl/extractor/escapist.py index 3aa2da52c..b1242f6bc 100644 --- a/youtube_dl/extractor/escapist.py +++ b/youtube_dl/extractor/escapist.py @@ -11,11 +11,11 @@ from ..utils import ( class EscapistIE(InfoExtractor): - _VALID_URL = r'^(https?://)?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<episode>[^/?]+)[/?]?.*$' + _VALID_URL = r'^https?://?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<episode>[^/?]+)[/?]?.*$' _TEST = { u'url': u'http://www.escapistmagazine.com/videos/view/the-escapist-presents/6618-Breaking-Down-Baldurs-Gate', u'file': u'6618-Breaking-Down-Baldurs-Gate.mp4', - u'md5': u'c6793dbda81388f4264c1ba18684a74d', + u'md5': u'ab3a706c681efca53f0a35f1415cf0d1', u'info_dict': { u"description": u"Baldur's Gate: Original, Modded or Enhanced Edition? I'll break down what you can expect from the new Baldur's Gate: Enhanced Edition.", u"uploader": u"the-escapist-presents", @@ -25,50 +25,60 @@ class EscapistIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) showName = mobj.group('showname') videoId = mobj.group('episode') self.report_extraction(videoId) webpage = self._download_webpage(url, videoId) - videoDesc = self._html_search_regex('<meta name="description" content="([^"]*)"', + videoDesc = self._html_search_regex( + r'<meta name="description" content="([^"]*)"', webpage, u'description', fatal=False) - playerUrl = self._og_search_video_url(webpage, name='player url') + playerUrl = self._og_search_video_url(webpage, name=u'player URL') - title = self._html_search_regex('<meta name="title" content="([^"]*)"', - webpage, u'player url').split(' : ')[-1] + title = self._html_search_regex( + r'<meta name="title" content="([^"]*)"', + webpage, u'title').split(' : ')[-1] - configUrl = self._search_regex('config=(.*)$', playerUrl, u'config url') + configUrl = self._search_regex('config=(.*)$', playerUrl, u'config URL') configUrl = compat_urllib_parse.unquote(configUrl) - configJSON = self._download_webpage(configUrl, videoId, - u'Downloading configuration', - u'unable to download configuration') - - # Technically, it's JavaScript, not JSON - configJSON = configJSON.replace("'", '"') - + formats = [] + + def _add_format(name, cfgurl): + configJSON = self._download_webpage( + cfgurl, videoId, + u'Downloading ' + name + ' configuration', + u'Unable to download ' + name + ' configuration') + + # Technically, it's JavaScript, not JSON + configJSON = configJSON.replace("'", '"') + + try: + config = json.loads(configJSON) + except (ValueError,) as err: + raise ExtractorError(u'Invalid JSON in configuration file: ' + compat_str(err)) + playlist = config['playlist'] + formats.append({ + 'url': playlist[1]['url'], + 'format_id': name, + }) + + _add_format(u'normal', configUrl) + hq_url = (configUrl + + ('&hq=1' if '?' in configUrl else configUrl + '?hq=1')) try: - config = json.loads(configJSON) - except (ValueError,) as err: - raise ExtractorError(u'Invalid JSON in configuration file: ' + compat_str(err)) + _add_format(u'hq', hq_url) + except ExtractorError: + pass # That's fine, we'll just use normal quality - playlist = config['playlist'] - videoUrl = playlist[1]['url'] - - info = { + return { 'id': videoId, - 'url': videoUrl, + 'formats': formats, 'uploader': showName, - 'upload_date': None, 'title': title, - 'ext': 'mp4', 'thumbnail': self._og_search_thumbnail(webpage), 'description': videoDesc, 'player_url': playerUrl, } - - return [info] diff --git a/youtube_dl/extractor/exfm.py b/youtube_dl/extractor/exfm.py index c74556579..682901d16 100644 --- a/youtube_dl/extractor/exfm.py +++ b/youtube_dl/extractor/exfm.py @@ -8,7 +8,7 @@ class ExfmIE(InfoExtractor): IE_NAME = u'exfm' IE_DESC = u'ex.fm' _VALID_URL = r'(?:http://)?(?:www\.)?ex\.fm/song/([^/]+)' - _SOUNDCLOUD_URL = r'(?:http://)?(?:www\.)?api\.soundcloud.com/tracks/([^/]+)/stream' + _SOUNDCLOUD_URL = r'(?:http://)?(?:www\.)?api\.soundcloud\.com/tracks/([^/]+)/stream' _TESTS = [ { u'url': u'http://ex.fm/song/eh359', @@ -21,6 +21,7 @@ class ExfmIE(InfoExtractor): u'description': u'Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive', }, u'note': u'Soundcloud song', + u'skip': u'The site is down too often', }, { u'url': u'http://ex.fm/song/wddt8', @@ -30,6 +31,7 @@ class ExfmIE(InfoExtractor): u'title': u'Safe and Sound', u'uploader': u'Capital Cities', }, + u'skip': u'The site is down too often', }, ] diff --git a/youtube_dl/extractor/extremetube.py b/youtube_dl/extractor/extremetube.py new file mode 100644 index 000000000..1c20e4364 --- /dev/null +++ b/youtube_dl/extractor/extremetube.py @@ -0,0 +1,50 @@ +import os +import re + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_parse_urlparse, + compat_urllib_request, + compat_urllib_parse, +) + +class ExtremeTubeIE(InfoExtractor): + _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>extremetube\.com/video/.+?(?P<videoid>[0-9]+))(?:[/?&]|$)' + _TEST = { + u'url': u'http://www.extremetube.com/video/music-video-14-british-euro-brit-european-cumshots-swallow-652431', + u'file': u'652431.mp4', + u'md5': u'1fb9228f5e3332ec8c057d6ac36f33e0', + u'info_dict': { + u"title": u"Music Video 14 british euro brit european cumshots swallow", + u"uploader": u"unknown", + u"age_limit": 18, + } + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('videoid') + url = 'http://www.' + mobj.group('url') + + req = compat_urllib_request.Request(url) + req.add_header('Cookie', 'age_verified=1') + webpage = self._download_webpage(req, video_id) + + video_title = self._html_search_regex(r'<h1 [^>]*?title="([^"]+)"[^>]*>\1<', webpage, u'title') + uploader = self._html_search_regex(r'>Posted by:(?=<)(?:\s|<[^>]*>)*(.+?)\|', webpage, u'uploader', fatal=False) + video_url = compat_urllib_parse.unquote(self._html_search_regex(r'video_url=(.+?)&', webpage, u'video_url')) + path = compat_urllib_parse_urlparse(video_url).path + extension = os.path.splitext(path)[1][1:] + format = path.split('/')[5].split('_')[:2] + format = "-".join(format) + + return { + 'id': video_id, + 'title': video_title, + 'uploader': uploader, + 'url': video_url, + 'ext': extension, + 'format': format, + 'format_id': format, + 'age_limit': 18, + } diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py index f8bdfc2d3..4556079c8 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -1,5 +1,4 @@ import json -import netrc import re import socket @@ -18,7 +17,7 @@ from ..utils import ( class FacebookIE(InfoExtractor): """Information Extractor for Facebook""" - _VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)' + _VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:[^#?]*#!/)?(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)' _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1' _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1' _NETRC_MACHINE = 'facebook' @@ -28,7 +27,7 @@ class FacebookIE(InfoExtractor): u'file': u'120708114770723.mp4', u'md5': u'48975a41ccc4b7a581abd68651c1a5a8', u'info_dict': { - u"duration": 279, + u"duration": 279, u"title": u"PEOPLE ARE AWESOME 2013" } } diff --git a/youtube_dl/extractor/faz.py b/youtube_dl/extractor/faz.py index 89ed08db4..c6ab6952e 100644 --- a/youtube_dl/extractor/faz.py +++ b/youtube_dl/extractor/faz.py @@ -1,6 +1,5 @@ # encoding: utf-8 import re -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( @@ -10,7 +9,7 @@ from ..utils import ( class FazIE(InfoExtractor): IE_NAME = u'faz.net' - _VALID_URL = r'https?://www\.faz\.net/multimedia/videos/.*?-(?P<id>\d+).html' + _VALID_URL = r'https?://www\.faz\.net/multimedia/videos/.*?-(?P<id>\d+)\.html' _TEST = { u'url': u'http://www.faz.net/multimedia/videos/stockholm-chemie-nobelpreis-fuer-drei-amerikanische-forscher-12610585.html', @@ -28,9 +27,8 @@ class FazIE(InfoExtractor): webpage = self._download_webpage(url, video_id) config_xml_url = self._search_regex(r'writeFLV\(\'(.+?)\',', webpage, u'config xml url') - config_xml = self._download_webpage(config_xml_url, video_id, + config = self._download_xml(config_xml_url, video_id, u'Downloading config xml') - config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8')) encodings = config.find('ENCODINGS') formats = [] @@ -46,13 +44,10 @@ class FazIE(InfoExtractor): }) descr = self._html_search_regex(r'<p class="Content Copy">(.*?)</p>', webpage, u'description') - info = { + return { 'id': video_id, 'title': self._og_search_title(webpage), 'formats': formats, 'description': descr, 'thumbnail': config.find('STILL/STILL_BIG').text, } - # TODO: Remove when #980 has been merged - info.update(formats[-1]) - return info diff --git a/youtube_dl/extractor/fktv.py b/youtube_dl/extractor/fktv.py index 9c89362ef..d7048c8c1 100644 --- a/youtube_dl/extractor/fktv.py +++ b/youtube_dl/extractor/fktv.py @@ -12,7 +12,7 @@ from ..utils import ( class FKTVIE(InfoExtractor): IE_NAME = u'fernsehkritik.tv' - _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik.tv/folge-(?P<ep>[0-9]+)(?:/.*)?' + _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik\.tv/folge-(?P<ep>[0-9]+)(?:/.*)?' _TEST = { u'url': u'http://fernsehkritik.tv/folge-1', @@ -39,7 +39,6 @@ class FKTVIE(InfoExtractor): for i, _ in enumerate(files, 1): video_id = '%04d%d' % (episode, i) video_url = 'http://dl%d.fernsehkritik.tv/fernsehkritik%d%s.flv' % (server, episode, '' if i == 1 else '-%d' % i) - video_title = 'Fernsehkritik %d.%d' % (episode, i) videos.append({ 'id': video_id, 'url': video_url, @@ -53,7 +52,7 @@ class FKTVIE(InfoExtractor): class FKTVPosteckeIE(InfoExtractor): IE_NAME = u'fernsehkritik.tv:postecke' - _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik.tv/inline-video/postecke.php\?(.*&)?ep=(?P<ep>[0-9]+)(&|$)' + _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik\.tv/inline-video/postecke\.php\?(.*&)?ep=(?P<ep>[0-9]+)(&|$)' _TEST = { u'url': u'http://fernsehkritik.tv/inline-video/postecke.php?iframe=true&width=625&height=440&ep=120', u'file': u'0120.flv', diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index 086cafca0..ad85bc16d 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -1,6 +1,5 @@ # encoding: utf-8 import re -import xml.etree.ElementTree import json from .common import InfoExtractor @@ -11,11 +10,10 @@ from ..utils import ( class FranceTVBaseInfoExtractor(InfoExtractor): def _extract_video(self, video_id): - xml_desc = self._download_webpage( + info = self._download_xml( 'http://www.francetvinfo.fr/appftv/webservices/video/' 'getInfosOeuvre.php?id-diffusion=' + video_id, video_id, 'Downloading XML config') - info = xml.etree.ElementTree.fromstring(xml_desc.encode('utf-8')) manifest_url = info.find('videos/video/url').text video_url = manifest_url.replace('manifest.f4m', 'index_2_av.m3u8') @@ -23,7 +21,7 @@ class FranceTVBaseInfoExtractor(InfoExtractor): thumbnail_path = info.find('image').text return {'id': video_id, - 'ext': 'mp4', + 'ext': 'flv' if video_url.startswith('rtmp') else 'mp4', 'url': video_url, 'title': info.find('titre').text, 'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', thumbnail_path), @@ -47,7 +45,7 @@ class PluzzIE(FranceTVBaseInfoExtractor): class FranceTvInfoIE(FranceTVBaseInfoExtractor): IE_NAME = u'francetvinfo.fr' - _VALID_URL = r'https?://www\.francetvinfo\.fr/replay.*/(?P<title>.+).html' + _VALID_URL = r'https?://www\.francetvinfo\.fr/replay.*/(?P<title>.+)\.html' _TEST = { u'url': u'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html', @@ -68,35 +66,101 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): return self._extract_video(video_id) -class France2IE(FranceTVBaseInfoExtractor): - IE_NAME = u'france2.fr' - _VALID_URL = r'''(?x)https?://www\.france2\.fr/ +class FranceTVIE(FranceTVBaseInfoExtractor): + IE_NAME = u'francetv' + IE_DESC = u'France 2, 3, 4, 5 and Ô' + _VALID_URL = r'''(?x)https?://www\.france[2345o]\.fr/ (?: - emissions/.*?/videos/(?P<id>\d+) - | emission/(?P<key>[^/?]+) + emissions/.*?/(videos|emissions)/(?P<id>[^/?]+) + | (emissions?|jt)/(?P<key>[^/?]+) )''' - _TEST = { - u'url': u'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104', - u'file': u'75540104.mp4', - u'info_dict': { - u'title': u'13h15, le samedi...', - u'description': u'md5:2e5b58ba7a2d3692b35c792be081a03d', + _TESTS = [ + # france2 + { + u'url': u'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104', + u'file': u'75540104.mp4', + u'info_dict': { + u'title': u'13h15, le samedi...', + u'description': u'md5:2e5b58ba7a2d3692b35c792be081a03d', + }, + u'params': { + # m3u8 download + u'skip_download': True, + }, }, - u'params': { - u'skip_download': True, + # france3 + { + u'url': u'http://www.france3.fr/emissions/pieces-a-conviction/diffusions/13-11-2013_145575', + u'info_dict': { + u'id': u'000702326_CAPP_PicesconvictionExtrait313022013_120220131722_Au', + u'ext': u'flv', + u'title': u'Le scandale du prix des médicaments', + u'description': u'md5:1384089fbee2f04fc6c9de025ee2e9ce', + }, + u'params': { + # rtmp download + u'skip_download': True, + }, }, - } + # france4 + { + u'url': u'http://www.france4.fr/emissions/hero-corp/videos/rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4', + u'info_dict': { + u'id': u'rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4', + u'ext': u'flv', + u'title': u'Hero Corp Making of - Extrait 1', + u'description': u'md5:c87d54871b1790679aec1197e73d650a', + }, + u'params': { + # rtmp download + u'skip_download': True, + }, + }, + # france5 + { + u'url': u'http://www.france5.fr/emissions/c-a-dire/videos/92837968', + u'info_dict': { + u'id': u'92837968', + u'ext': u'mp4', + u'title': u'C à dire ?!', + u'description': u'md5:fb1db1cbad784dcce7c7a7bd177c8e2f', + }, + u'params': { + # m3u8 download + u'skip_download': True, + }, + }, + # franceo + { + u'url': u'http://www.franceo.fr/jt/info-afrique/04-12-2013', + u'info_dict': { + u'id': u'92327925', + u'ext': u'mp4', + u'title': u'Infô-Afrique', + u'description': u'md5:ebf346da789428841bee0fd2a935ea55', + }, + u'params': { + # m3u8 download + u'skip_download': True, + }, + u'skip': u'The id changes frequently', + }, + ] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) if mobj.group('key'): webpage = self._download_webpage(url, mobj.group('key')) - video_id = self._html_search_regex( - r'''(?x)<div\s+class="video-player">\s* + id_res = [ + (r'''(?x)<div\s+class="video-player">\s* <a\s+href="http://videos.francetv.fr/video/([0-9]+)"\s+ - class="francetv-video-player">''', - webpage, u'video ID') + class="francetv-video-player">'''), + (r'<a id="player_direct" href="http://info\.francetelevisions' + '\.fr/\?id-video=([^"/&]+)'), + (r'<a class="video" id="ftv_player_(.+?)"'), + ] + video_id = self._html_search_regex(id_res, webpage, u'video ID') else: video_id = mobj.group('id') return self._extract_video(video_id) diff --git a/youtube_dl/extractor/gamekings.py b/youtube_dl/extractor/gamekings.py new file mode 100644 index 000000000..a3a5251fe --- /dev/null +++ b/youtube_dl/extractor/gamekings.py @@ -0,0 +1,38 @@ +import re + +from .common import InfoExtractor + + +class GamekingsIE(InfoExtractor): + _VALID_URL = r'http://www\.gamekings\.tv/videos/(?P<name>[0-9a-z\-]+)' + _TEST = { + u"url": u"http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/", + u'file': u'20130811.mp4', + # MD5 is flaky, seems to change regularly + #u'md5': u'2f32b1f7b80fdc5cb616efb4f387f8a3', + u'info_dict': { + u"title": u"Phoenix Wright: Ace Attorney \u2013 Dual Destinies Review", + u"description": u"Melle en Steven hebben voor de review een week in de rechtbank doorbracht met Phoenix Wright: Ace Attorney - Dual Destinies.", + } + } + + def _real_extract(self, url): + + mobj = re.match(self._VALID_URL, url) + name = mobj.group('name') + webpage = self._download_webpage(url, name) + video_url = self._og_search_video_url(webpage) + + video = re.search(r'[0-9]+', video_url) + video_id = video.group(0) + + # Todo: add medium format + video_url = video_url.replace(video_id, 'large/' + video_id) + + return { + 'id': video_id, + 'ext': 'mp4', + 'url': video_url, + 'title': self._og_search_title(webpage), + 'description': self._og_search_description(webpage), + } diff --git a/youtube_dl/extractor/gamespot.py b/youtube_dl/extractor/gamespot.py index 098768361..26b7d2ae5 100644 --- a/youtube_dl/extractor/gamespot.py +++ b/youtube_dl/extractor/gamespot.py @@ -24,7 +24,7 @@ class GameSpotIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - page_id = video_id = mobj.group('page_id') + page_id = mobj.group('page_id') webpage = self._download_webpage(url, page_id) data_video_json = self._search_regex(r'data-video=\'(.*?)\'', webpage, u'data video') data_video = json.loads(unescapeHTML(data_video_json)) @@ -47,13 +47,10 @@ class GameSpotIE(InfoExtractor): 'format_id': q, }) - info = { + return { 'id': data_video['guid'], 'title': compat_urllib_parse.unquote(data_video['title']), 'formats': formats, 'description': get_meta_content('description', webpage), 'thumbnail': self._og_search_thumbnail(webpage), } - # TODO: Remove when #980 has been merged - info.update(formats[-1]) - return info diff --git a/youtube_dl/extractor/gametrailers.py b/youtube_dl/extractor/gametrailers.py index 3cc02d97e..d82a5d4b2 100644 --- a/youtube_dl/extractor/gametrailers.py +++ b/youtube_dl/extractor/gametrailers.py @@ -1,13 +1,10 @@ import re -from .mtv import MTVIE, _media_xml_tag +from .mtv import MTVServicesInfoExtractor -class GametrailersIE(MTVIE): - """ - Gametrailers use the same videos system as MTVIE, it just changes the feed - url, where the uri is and the method to get the thumbnails. - """ - _VALID_URL = r'http://www.gametrailers.com/(?P<type>videos|reviews|full-episodes)/(?P<id>.*?)/(?P<title>.*)' + +class GametrailersIE(MTVServicesInfoExtractor): + _VALID_URL = r'http://www\.gametrailers\.com/(?P<type>videos|reviews|full-episodes)/(?P<id>.*?)/(?P<title>.*)' _TEST = { u'url': u'http://www.gametrailers.com/videos/zbvr8i/mirror-s-edge-2-e3-2013--debut-trailer', u'file': u'70e9a5d7-cf25-4a10-9104-6f3e7342ae0d.mp4', @@ -17,15 +14,9 @@ class GametrailersIE(MTVIE): u'description': u'Faith is back! Check out the World Premiere trailer for Mirror\'s Edge 2 straight from the EA Press Conference at E3 2013!', }, } - # Overwrite MTVIE properties we don't want - _TESTS = [] _FEED_URL = 'http://www.gametrailers.com/feeds/mrss' - def _get_thumbnail_url(self, uri, itemdoc): - search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail')) - return itemdoc.find(search_path).attrib['url'] - def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 2c8fcf5ae..bdb4f58d6 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -11,10 +11,14 @@ from ..utils import ( compat_urlparse, ExtractorError, + HEADRequest, smuggle_url, unescapeHTML, + unified_strdate, + url_basename, ) from .brightcove import BrightcoveIE +from .ooyala import OoyalaIE class GenericIE(InfoExtractor): @@ -33,6 +37,7 @@ class GenericIE(InfoExtractor): }, # embedded vimeo video { + u'add_ie': ['Vimeo'], u'url': u'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references', u'file': u'22444065.mp4', u'md5': u'2903896e23df39722c33f015af0666e2', @@ -44,6 +49,7 @@ class GenericIE(InfoExtractor): }, # bandcamp page with custom domain { + u'add_ie': ['Bandcamp'], u'url': u'http://bronyrock.com/track/the-pony-mash', u'file': u'3235767654.mp3', u'info_dict': { @@ -52,6 +58,44 @@ class GenericIE(InfoExtractor): }, u'skip': u'There is a limit of 200 free downloads / month for the test song', }, + # embedded brightcove video + # it also tests brightcove videos that need to set the 'Referer' in the + # http requests + { + u'add_ie': ['Brightcove'], + u'url': u'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/', + u'info_dict': { + u'id': u'2765128793001', + u'ext': u'mp4', + u'title': u'Le cours de bourse : l’analyse technique', + u'description': u'md5:7e9ad046e968cb2d1114004aba466fd9', + u'uploader': u'BFM BUSINESS', + }, + u'params': { + u'skip_download': True, + }, + }, + # Direct link to a video + { + u'url': u'http://media.w3.org/2010/05/sintel/trailer.mp4', + u'file': u'trailer.mp4', + u'md5': u'67d406c2bcb6af27fa886f31aa934bbe', + u'info_dict': { + u'id': u'trailer', + u'title': u'trailer', + u'upload_date': u'20100513', + } + }, + # ooyala video + { + u'url': u'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219', + u'md5': u'5644c6ca5d5782c1d0d350dad9bd840c', + u'info_dict': { + u'id': u'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ', + u'ext': u'mp4', + u'title': u'2cc213299525360.mov', #that's what we get + }, + }, ] def report_download_webpage(self, video_id): @@ -64,23 +108,20 @@ class GenericIE(InfoExtractor): """Report information extraction.""" self._downloader.to_screen(u'[redirect] Following redirect to %s' % new_url) - def _test_redirect(self, url): + def _send_head(self, url): """Check if it is a redirect, like url shorteners, in case return the new url.""" - class HeadRequest(compat_urllib_request.Request): - def get_method(self): - return "HEAD" class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler): """ Subclass the HTTPRedirectHandler to make it use our - HeadRequest also on the redirected URL + HEADRequest also on the redirected URL """ def redirect_request(self, req, fp, code, msg, headers, newurl): if code in (301, 302, 303, 307): newurl = newurl.replace(' ', '%20') newheaders = dict((k,v) for k,v in req.headers.items() if k.lower() not in ("content-length", "content-type")) - return HeadRequest(newurl, + return HEADRequest(newurl, headers=newheaders, origin_req_host=req.get_origin_req_host(), unverifiable=True) @@ -109,32 +150,49 @@ class GenericIE(InfoExtractor): compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]: opener.add_handler(handler()) - response = opener.open(HeadRequest(url)) + response = opener.open(HEADRequest(url)) if response is None: raise ExtractorError(u'Invalid URL protocol') - new_url = response.geturl() - - if url == new_url: - return False - - self.report_following_redirect(new_url) - return new_url + return response def _real_extract(self, url): parsed_url = compat_urlparse.urlparse(url) if not parsed_url.scheme: self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http') return self.url_result('http://' + url) + video_id = os.path.splitext(url.split('/')[-1])[0] try: - new_url = self._test_redirect(url) - if new_url: - return [self.url_result(new_url)] + response = self._send_head(url) + + # Check for redirect + new_url = response.geturl() + if url != new_url: + self.report_following_redirect(new_url) + return self.url_result(new_url) + + # Check for direct link to a video + content_type = response.headers.get('Content-Type', '') + m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type) + if m: + upload_date = response.headers.get('Last-Modified') + if upload_date: + upload_date = unified_strdate(upload_date) + return { + 'id': video_id, + 'title': os.path.splitext(url_basename(url))[0], + 'formats': [{ + 'format_id': m.group('format_id'), + 'url': url, + 'vcodec': u'none' if m.group('type') == 'audio' else None + }], + 'upload_date': upload_date, + } + except compat_urllib_error.HTTPError: # This may be a stupid server that doesn't like HEAD, our UA, or so pass - video_id = url.split('/')[-1] try: webpage = self._download_webpage(url, video_id) except ValueError: @@ -143,11 +201,25 @@ class GenericIE(InfoExtractor): raise ExtractorError(u'Failed to download URL: %s' % url) self.report_extraction(video_id) + + # it's tempting to parse this further, but you would + # have to take into account all the variations like + # Video Title - Site Name + # Site Name | Video Title + # Video Title - Tagline | Site Name + # and so on and so forth; it's just not practical + video_title = self._html_search_regex( + r'(?s)<title>(.*?)', webpage, u'video title', + default=u'video') + + # video uploader is domain name + video_uploader = self._search_regex( + r'^(?:https?://)?([^/]*)/.*', url, u'video uploader') + # Look for BrightCove: - m_brightcove = re.search(r']+?class=([\'"])[^>]*?BrightcoveExperience.*?\1.+?', webpage, re.DOTALL) - if m_brightcove is not None: + bc_url = BrightcoveIE._extract_brightcove_url(webpage) + if bc_url is not None: self.to_screen(u'Brightcove video detected.') - bc_url = BrightcoveIE._build_brighcove_url(m_brightcove.group()) return self.url_result(bc_url, 'Brightcove') # Look for embedded Vimeo player @@ -159,17 +231,73 @@ class GenericIE(InfoExtractor): return self.url_result(surl, 'Vimeo') # Look for embedded YouTube player - mobj = re.search( - r']+?src="(https?://(?:www\.)?youtube.com/embed/.+?)"', webpage) + matches = re.findall(r'''(?x) + (?:]+?src=|embedSWF\(\s*) + (["\'])(?P(?:https?:)?//(?:www\.)?youtube\.com/ + (?:embed|v)/.+?) + \1''', webpage) + if matches: + urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube') + for tuppl in matches] + return self.playlist_result( + urlrs, playlist_id=video_id, playlist_title=video_title) + + # Look for embedded Dailymotion player + matches = re.findall( + r']+?src=(["\'])(?P(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage) + if matches: + urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Dailymotion') + for tuppl in matches] + return self.playlist_result( + urlrs, playlist_id=video_id, playlist_title=video_title) + + # Look for embedded Wistia player + match = re.search( + r']+?src=(["\'])(?P(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage) + if match: + return { + '_type': 'url_transparent', + 'url': unescapeHTML(match.group('url')), + 'ie_key': 'Wistia', + 'uploader': video_uploader, + 'title': video_title, + 'id': video_id, + } + + # Look for embedded blip.tv player + mobj = re.search(r']*https?://api.blip.tv/\w+/redirect/\w+/(\d+)', webpage) + if mobj: + return self.url_result('http://blip.tv/seo/-'+mobj.group(1), 'BlipTV') + mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*https?://(?:\w+\.)?blip.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', webpage) if mobj: - surl = unescapeHTML(mobj.group(1)) - return self.url_result(surl, 'Youtube') + player_url = 'http://blip.tv/play/%s.x?p=1' % mobj.group(1) + player_page = self._download_webpage(player_url, mobj.group(1)) + blip_video_id = self._search_regex(r'data-episode-id="(\d+)', player_page, u'blip_video_id', fatal=False) + if blip_video_id: + return self.url_result('http://blip.tv/seo/-'+blip_video_id, 'BlipTV') # Look for Bandcamp pages with custom domain mobj = re.search(r']*?content="(.*?bandcamp\.com.*?)"', webpage) if mobj is not None: burl = unescapeHTML(mobj.group(1)) - return self.url_result(burl, 'Bandcamp') + # Don't set the extractor because it can be a track url or an album + return self.url_result(burl) + + # Look for embedded Vevo player + mobj = re.search( + r']+?src=(["\'])(?P(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage) + if mobj is not None: + return self.url_result(mobj.group('url')) + + # Look for Ooyala videos + mobj = re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=([^"&]+)', webpage) + if mobj is not None: + return OoyalaIE._build_url_result(mobj.group(1)) + + # Look for Aparat videos + mobj = re.search(r'