diff --git a/.github/ISSUE_TEMPLATE/1_broken_site.md b/.github/ISSUE_TEMPLATE/1_broken_site.md index ccd033716..4c75c8d5d 100644 --- a/.github/ISSUE_TEMPLATE/1_broken_site.md +++ b/.github/ISSUE_TEMPLATE/1_broken_site.md @@ -18,7 +18,7 @@ title: '' - [ ] I'm reporting a broken site support -- [ ] I've verified that I'm running youtube-dl version **2019.07.30** +- [ ] I've verified that I'm running youtube-dl version **2019.08.13** - [ ] I've checked that all provided URLs are alive and playable in a browser - [ ] I've checked that all URLs and arguments with special characters are properly quoted or escaped - [ ] I've searched the bugtracker for similar issues including closed ones @@ -41,7 +41,7 @@ Add the `-v` flag to your command line you run youtube-dl with (`youtube-dl -v < [debug] User config: [] [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj'] [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251 - [debug] youtube-dl version 2019.07.30 + [debug] youtube-dl version 2019.08.13 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4 [debug] Proxy map: {} diff --git a/.github/ISSUE_TEMPLATE/2_site_support_request.md b/.github/ISSUE_TEMPLATE/2_site_support_request.md index 8709937ad..8e8c43c47 100644 --- a/.github/ISSUE_TEMPLATE/2_site_support_request.md +++ b/.github/ISSUE_TEMPLATE/2_site_support_request.md @@ -19,7 +19,7 @@ labels: 'site-support-request' - [ ] I'm reporting a new site support request -- [ ] I've verified that I'm running youtube-dl version **2019.07.30** +- [ ] I've verified that I'm running youtube-dl version **2019.08.13** - [ ] I've checked that all provided URLs are alive and playable in a browser - [ ] I've checked that none of provided URLs violate any copyrights - [ ] I've searched the bugtracker for similar site support requests including closed ones diff --git a/.github/ISSUE_TEMPLATE/3_site_feature_request.md b/.github/ISSUE_TEMPLATE/3_site_feature_request.md index c3a555ed3..df719a29c 100644 --- a/.github/ISSUE_TEMPLATE/3_site_feature_request.md +++ b/.github/ISSUE_TEMPLATE/3_site_feature_request.md @@ -18,13 +18,13 @@ title: '' - [ ] I'm reporting a site feature request -- [ ] I've verified that I'm running youtube-dl version **2019.07.30** +- [ ] I've verified that I'm running youtube-dl version **2019.08.13** - [ ] I've searched the bugtracker for similar site feature requests including closed ones diff --git a/.github/ISSUE_TEMPLATE/4_bug_report.md b/.github/ISSUE_TEMPLATE/4_bug_report.md index 07042a466..3616db1a7 100644 --- a/.github/ISSUE_TEMPLATE/4_bug_report.md +++ b/.github/ISSUE_TEMPLATE/4_bug_report.md @@ -18,7 +18,7 @@ title: '' - [ ] I'm reporting a broken site support issue -- [ ] I've verified that I'm running youtube-dl version **2019.07.30** +- [ ] I've verified that I'm running youtube-dl version **2019.08.13** - [ ] I've checked that all provided URLs are alive and playable in a browser - [ ] I've checked that all URLs and arguments with special characters are properly quoted or escaped - [ ] I've searched the bugtracker for similar bug reports including closed ones @@ -43,7 +43,7 @@ Add the `-v` flag to your command line you run youtube-dl with (`youtube-dl -v < [debug] User config: [] [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj'] [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251 - [debug] youtube-dl version 2019.07.30 + [debug] youtube-dl version 2019.08.13 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4 [debug] Proxy map: {} diff --git a/.github/ISSUE_TEMPLATE/5_feature_request.md b/.github/ISSUE_TEMPLATE/5_feature_request.md index 4cf75a2eb..0fa37aef1 100644 --- a/.github/ISSUE_TEMPLATE/5_feature_request.md +++ b/.github/ISSUE_TEMPLATE/5_feature_request.md @@ -19,13 +19,13 @@ labels: 'request' - [ ] I'm reporting a feature request -- [ ] I've verified that I'm running youtube-dl version **2019.07.30** +- [ ] I've verified that I'm running youtube-dl version **2019.08.13** - [ ] I've searched the bugtracker for similar feature requests including closed ones diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index cd9ccbe96..ac759ddc4 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -339,6 +339,72 @@ Incorrect: 'PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4' ``` +### Inline values + +Extracting variables is acceptable for reducing code duplication and improving readability of complex expressions. However, you should avoid extracting variables used only once and moving them to opposite parts of the extractor file, which makes reading the linear flow difficult. + +#### Example + +Correct: + +```python +title = self._html_search_regex(r'([^<]+)', webpage, 'title') +``` + +Incorrect: + +```python +TITLE_RE = r'([^<]+)' +# ...some lines of code... +title = self._html_search_regex(TITLE_RE, webpage, 'title') +``` + +### Collapse fallbacks + +Multiple fallback values can quickly become unwieldy. Collapse multiple fallback values into a single expression via a list of patterns. + +#### Example + +Good: + +```python +description = self._html_search_meta( + ['og:description', 'description', 'twitter:description'], + webpage, 'description', default=None) +``` + +Unwieldy: + +```python +description = ( + self._og_search_description(webpage, default=None) + or self._html_search_meta('description', webpage, default=None) + or self._html_search_meta('twitter:description', webpage, default=None)) +``` + +Methods supporting list of patterns are: `_search_regex`, `_html_search_regex`, `_og_search_property`, `_html_search_meta`. + +### Trailing parentheses + +Always move trailing parentheses after the last argument. + +#### Example + +Correct: + +```python + lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'], + list) +``` + +Incorrect: + +```python + lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'], + list, +) +``` + ### Use convenience conversion and parsing functions Wrap all extracted numeric data into safe functions from [`youtube_dl/utils.py`](https://github.com/ytdl-org/youtube-dl/blob/master/youtube_dl/utils.py): `int_or_none`, `float_or_none`. Use them for string to number conversions as well. diff --git a/ChangeLog b/ChangeLog index f6f1f7e38..9b9e2e149 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,33 @@ +version 2019.08.13 + +Core +* [downloader/fragment] Fix ETA calculation of resumed download (#21992) +* [YoutubeDL] Check annotations availability (#18582) + +Extractors +* [youtube:playlist] Improve flat extraction (#21927) +* [youtube] Fix annotations extraction (#22045) ++ [discovery] Extract series meta field (#21808) +* [youtube] Improve error detection (#16445) +* [vimeo] Fix album extraction (#1933, #15704, #15855, #18967, #21986) ++ [roosterteeth] Add support for watch URLs +* [discovery] Limit video data by show slug (#21980) + + +version 2019.08.02 + +Extractors ++ [tvigle] Add support for HLS and DASH formats (#21967) +* [tvigle] Fix extraction (#21967) ++ [yandexvideo] Add support for DASH formats (#21971) +* [discovery] Use API call for video data extraction (#21808) ++ [mgtv] Extract format_note (#21881) +* [tvn24] Fix metadata extraction (#21833, #21834) +* [dlive] Relax URL regular expression (#21909) ++ [openload] Add support for oload.best (#21913) +* [youtube] Improve metadata extraction for age gate content (#21943) + + version 2019.07.30 Extractors diff --git a/README.md b/README.md index 8c48a3012..c39b13616 100644 --- a/README.md +++ b/README.md @@ -1216,6 +1216,72 @@ Incorrect: 'PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4' ``` +### Inline values + +Extracting variables is acceptable for reducing code duplication and improving readability of complex expressions. However, you should avoid extracting variables used only once and moving them to opposite parts of the extractor file, which makes reading the linear flow difficult. + +#### Example + +Correct: + +```python +title = self._html_search_regex(r'([^<]+)', webpage, 'title') +``` + +Incorrect: + +```python +TITLE_RE = r'([^<]+)' +# ...some lines of code... +title = self._html_search_regex(TITLE_RE, webpage, 'title') +``` + +### Collapse fallbacks + +Multiple fallback values can quickly become unwieldy. Collapse multiple fallback values into a single expression via a list of patterns. + +#### Example + +Good: + +```python +description = self._html_search_meta( + ['og:description', 'description', 'twitter:description'], + webpage, 'description', default=None) +``` + +Unwieldy: + +```python +description = ( + self._og_search_description(webpage, default=None) + or self._html_search_meta('description', webpage, default=None) + or self._html_search_meta('twitter:description', webpage, default=None)) +``` + +Methods supporting list of patterns are: `_search_regex`, `_html_search_regex`, `_og_search_property`, `_html_search_meta`. + +### Trailing parentheses + +Always move trailing parentheses after the last argument. + +#### Example + +Correct: + +```python + lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'], + list) +``` + +Incorrect: + +```python + lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'], + list, +) +``` + ### Use convenience conversion and parsing functions Wrap all extracted numeric data into safe functions from [`youtube_dl/utils.py`](https://github.com/ytdl-org/youtube-dl/blob/master/youtube_dl/utils.py): `int_or_none`, `float_or_none`. Use them for string to number conversions as well. diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 3e832fec2..6a44bc7ba 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -1783,6 +1783,8 @@ class YoutubeDL(object): annofn = replace_extension(filename, 'annotations.xml', info_dict.get('ext')) if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)): self.to_screen('[info] Video annotations are already present') + elif not info_dict.get('annotations'): + self.report_warning('There are no annotations to write.') else: try: self.to_screen('[info] Writing video annotations to: ' + annofn) diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index 165c975dd..9a659fc65 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -94,7 +94,7 @@ def _real_main(argv=None): if opts.verbose: write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n') except IOError: - sys.exit('ERROR: batch file could not be read') + sys.exit('ERROR: batch file %s could not be read' % opts.batchfile) all_urls = batch_urls + [url.strip() for url in args] # batch_urls are already striped in read_batch_urls _enc = preferredencoding() all_urls = [url.decode(_enc, 'ignore') if isinstance(url, bytes) else url for url in all_urls] diff --git a/youtube_dl/downloader/fragment.py b/youtube_dl/downloader/fragment.py index f2e5733b6..02f35459e 100644 --- a/youtube_dl/downloader/fragment.py +++ b/youtube_dl/downloader/fragment.py @@ -190,12 +190,13 @@ class FragmentFD(FileDownloader): }) def _start_frag_download(self, ctx): + resume_len = ctx['complete_frags_downloaded_bytes'] total_frags = ctx['total_frags'] # This dict stores the download progress, it's updated by the progress # hook state = { 'status': 'downloading', - 'downloaded_bytes': ctx['complete_frags_downloaded_bytes'], + 'downloaded_bytes': resume_len, 'fragment_index': ctx['fragment_index'], 'fragment_count': total_frags, 'filename': ctx['filename'], @@ -234,8 +235,8 @@ class FragmentFD(FileDownloader): state['downloaded_bytes'] += frag_downloaded_bytes - ctx['prev_frag_downloaded_bytes'] if not ctx['live']: state['eta'] = self.calc_eta( - start, time_now, estimated_size, - state['downloaded_bytes']) + start, time_now, estimated_size - resume_len, + state['downloaded_bytes'] - resume_len) state['speed'] = s.get('speed') or ctx.get('speed') ctx['speed'] = state['speed'] ctx['prev_frag_downloaded_bytes'] = frag_downloaded_bytes diff --git a/youtube_dl/extractor/discovery.py b/youtube_dl/extractor/discovery.py index 9003545ce..6a2712cc5 100644 --- a/youtube_dl/extractor/discovery.py +++ b/youtube_dl/extractor/discovery.py @@ -5,14 +5,8 @@ import re import string from .discoverygo import DiscoveryGoBaseIE -from ..compat import ( - compat_str, - compat_urllib_parse_unquote, -) -from ..utils import ( - ExtractorError, - try_get, -) +from ..compat import compat_urllib_parse_unquote +from ..utils import ExtractorError from ..compat import compat_HTTPError @@ -40,15 +34,15 @@ class DiscoveryIE(DiscoveryGoBaseIE): cookingchanneltv| motortrend ) - )\.com(?P/tv-shows/[^/]+/(?:video|full-episode)s/(?P[^./?#]+))''' + )\.com/tv-shows/(?P[^/]+)/(?:video|full-episode)s/(?P[^./?#]+)''' _TESTS = [{ - 'url': 'https://www.discovery.com/tv-shows/cash-cab/videos/dave-foley', + 'url': 'https://go.discovery.com/tv-shows/cash-cab/videos/riding-with-matthew-perry', 'info_dict': { - 'id': '5a2d9b4d6b66d17a5026e1fd', + 'id': '5a2f35ce6b66d17a5026e29e', 'ext': 'mp4', - 'title': 'Dave Foley', - 'description': 'md5:4b39bcafccf9167ca42810eb5f28b01f', - 'duration': 608, + 'title': 'Riding with Matthew Perry', + 'description': 'md5:a34333153e79bc4526019a5129e7f878', + 'duration': 84, }, 'params': { 'skip_download': True, # requires ffmpeg @@ -59,20 +53,17 @@ class DiscoveryIE(DiscoveryGoBaseIE): }, { 'url': 'https://go.discovery.com/tv-shows/alaskan-bush-people/videos/follow-your-own-road', 'only_matching': True, + }, { + # using `show_slug` is important to get the correct video data + 'url': 'https://www.sciencechannel.com/tv-shows/mythbusters-on-science/full-episodes/christmas-special', + 'only_matching': True, }] _GEO_COUNTRIES = ['US'] _GEO_BYPASS = False + _API_BASE_URL = 'https://api.discovery.com/v1/' def _real_extract(self, url): - site, path, display_id = re.match(self._VALID_URL, url).groups() - webpage = self._download_webpage(url, display_id) - - react_data = self._parse_json(self._search_regex( - r'window\.__reactTransmitPacket\s*=\s*({.+?});', - webpage, 'react data'), display_id) - content_blocks = react_data['layout'][path]['contentBlocks'] - video = next(cb for cb in content_blocks if cb.get('type') == 'video')['content']['items'][0] - video_id = video['id'] + site, show_slug, display_id = re.match(self._VALID_URL, url).groups() access_token = None cookies = self._get_cookies(url) @@ -82,27 +73,36 @@ class DiscoveryIE(DiscoveryGoBaseIE): if auth_storage_cookie and auth_storage_cookie.value: auth_storage = self._parse_json(compat_urllib_parse_unquote( compat_urllib_parse_unquote(auth_storage_cookie.value)), - video_id, fatal=False) or {} + display_id, fatal=False) or {} access_token = auth_storage.get('a') or auth_storage.get('access_token') if not access_token: access_token = self._download_json( - 'https://%s.com/anonymous' % site, display_id, query={ + 'https://%s.com/anonymous' % site, display_id, + 'Downloading token JSON metadata', query={ 'authRel': 'authorization', - 'client_id': try_get( - react_data, lambda x: x['application']['apiClientId'], - compat_str) or '3020a40c2356a645b4b4', + 'client_id': '3020a40c2356a645b4b4', 'nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]), 'redirectUri': 'https://fusion.ddmcdn.com/app/mercury-sdk/180/redirectHandler.html?https://www.%s.com' % site, })['access_token'] - try: - headers = self.geo_verification_headers() - headers['Authorization'] = 'Bearer ' + access_token + headers = self.geo_verification_headers() + headers['Authorization'] = 'Bearer ' + access_token + try: + video = self._download_json( + self._API_BASE_URL + 'content/videos', + display_id, 'Downloading content JSON metadata', + headers=headers, query={ + 'embed': 'show.name', + 'fields': 'authenticated,description.detailed,duration,episodeNumber,id,name,parental.rating,season.number,show,tags', + 'slug': display_id, + 'show_slug': show_slug, + })[0] + video_id = video['id'] stream = self._download_json( - 'https://api.discovery.com/v1/streaming/video/' + video_id, - display_id, headers=headers) + self._API_BASE_URL + 'streaming/video/' + video_id, + display_id, 'Downloading streaming JSON metadata', headers=headers) except ExtractorError as e: if isinstance(e.cause, compat_HTTPError) and e.cause.code in (401, 403): e_description = self._parse_json( diff --git a/youtube_dl/extractor/dlive.py b/youtube_dl/extractor/dlive.py index 8787f15a6..d95c67a5b 100644 --- a/youtube_dl/extractor/dlive.py +++ b/youtube_dl/extractor/dlive.py @@ -9,8 +9,8 @@ from ..utils import int_or_none class DLiveVODIE(InfoExtractor): IE_NAME = 'dlive:vod' - _VALID_URL = r'https?://(?:www\.)?dlive\.tv/p/(?P.+?)\+(?P[a-zA-Z0-9]+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?dlive\.tv/p/(?P.+?)\+(?P[^/?#&]+)' + _TESTS = [{ 'url': 'https://dlive.tv/p/pdp+3mTzOl4WR', 'info_dict': { 'id': '3mTzOl4WR', @@ -20,7 +20,10 @@ class DLiveVODIE(InfoExtractor): 'timestamp': 1562011015, 'uploader_id': 'pdp', } - } + }, { + 'url': 'https://dlive.tv/p/pdpreplay+D-RD-xSZg', + 'only_matching': True, + }] def _real_extract(self, url): uploader_id, vod_id = re.match(self._VALID_URL, url).groups() diff --git a/youtube_dl/extractor/mgtv.py b/youtube_dl/extractor/mgtv.py index 7ae2e3c3b..71fc3ec56 100644 --- a/youtube_dl/extractor/mgtv.py +++ b/youtube_dl/extractor/mgtv.py @@ -82,6 +82,7 @@ class MGTVIE(InfoExtractor): 'http_headers': { 'Referer': url, }, + 'format_note': stream.get('name'), }) self._sort_formats(formats) diff --git a/youtube_dl/extractor/openload.py b/youtube_dl/extractor/openload.py index 11e92e471..030355257 100644 --- a/youtube_dl/extractor/openload.py +++ b/youtube_dl/extractor/openload.py @@ -243,7 +243,7 @@ class PhantomJSwrapper(object): class OpenloadIE(InfoExtractor): - _DOMAINS = r'(?:openload\.(?:co|io|link|pw)|oload\.(?:tv|biz|stream|site|xyz|win|download|cloud|cc|icu|fun|club|info|press|pw|life|live|space|services|website)|oladblock\.(?:services|xyz|me)|openloed\.co)' + _DOMAINS = r'(?:openload\.(?:co|io|link|pw)|oload\.(?:tv|best|biz|stream|site|xyz|win|download|cloud|cc|icu|fun|club|info|press|pw|life|live|space|services|website)|oladblock\.(?:services|xyz|me)|openloed\.co)' _VALID_URL = r'''(?x) https?:// (?P @@ -368,6 +368,9 @@ class OpenloadIE(InfoExtractor): }, { 'url': 'https://oload.biz/f/bEk3Gp8ARr4/', 'only_matching': True, + }, { + 'url': 'https://oload.best/embed/kkz9JgVZeWc/', + 'only_matching': True, }, { 'url': 'https://oladblock.services/f/b8NWEgkqNLI/', 'only_matching': True, diff --git a/youtube_dl/extractor/piksel.py b/youtube_dl/extractor/piksel.py index c0c276a50..401298cb8 100644 --- a/youtube_dl/extractor/piksel.py +++ b/youtube_dl/extractor/piksel.py @@ -18,15 +18,14 @@ class PikselIE(InfoExtractor): _VALID_URL = r'https?://player\.piksel\.com/v/(?P[a-z0-9]+)' _TESTS = [ { - 'url': 'http://player.piksel.com/v/nv60p12f', - 'md5': 'd9c17bbe9c3386344f9cfd32fad8d235', + 'url': 'http://player.piksel.com/v/ums2867l', + 'md5': '34e34c8d89dc2559976a6079db531e85', 'info_dict': { - 'id': 'nv60p12f', + 'id': 'ums2867l', 'ext': 'mp4', - 'title': 'فن الحياة - الحلقة 1', - 'description': 'احدث برامج الداعية الاسلامي " مصطفي حسني " فى رمضان 2016علي النهار نور', - 'timestamp': 1465231790, - 'upload_date': '20160606', + 'title': 'GX-005 with Caption', + 'timestamp': 1481335659, + 'upload_date': '20161210' } }, { @@ -39,7 +38,7 @@ class PikselIE(InfoExtractor): 'title': 'WAW- State of Washington vs. Donald J. Trump, et al', 'description': 'State of Washington vs. Donald J. Trump, et al, Case Number 17-CV-00141-JLR, TRO Hearing, Civil Rights Case, 02/3/2017, 1:00 PM (PST), Seattle Federal Courthouse, Seattle, WA, Judge James L. Robart presiding.', 'timestamp': 1486171129, - 'upload_date': '20170204', + 'upload_date': '20170204' } } ] @@ -113,6 +112,13 @@ class PikselIE(InfoExtractor): }) self._sort_formats(formats) + subtitles = {} + for caption in video_data.get('captions', []): + caption_url = caption.get('url') + if caption_url: + subtitles.setdefault(caption.get('locale', 'en'), []).append({ + 'url': caption_url}) + return { 'id': video_id, 'title': title, @@ -120,4 +126,5 @@ class PikselIE(InfoExtractor): 'thumbnail': video_data.get('thumbnailUrl'), 'timestamp': parse_iso8601(video_data.get('dateadd')), 'formats': formats, + 'subtitles': subtitles, } diff --git a/youtube_dl/extractor/roosterteeth.py b/youtube_dl/extractor/roosterteeth.py index d3eeeba62..8d88ee499 100644 --- a/youtube_dl/extractor/roosterteeth.py +++ b/youtube_dl/extractor/roosterteeth.py @@ -17,7 +17,7 @@ from ..utils import ( class RoosterTeethIE(InfoExtractor): - _VALID_URL = r'https?://(?:.+?\.)?roosterteeth\.com/episode/(?P[^/?#&]+)' + _VALID_URL = r'https?://(?:.+?\.)?roosterteeth\.com/(?:episode|watch)/(?P[^/?#&]+)' _LOGIN_URL = 'https://roosterteeth.com/login' _NETRC_MACHINE = 'roosterteeth' _TESTS = [{ @@ -49,6 +49,9 @@ class RoosterTeethIE(InfoExtractor): # only available for FIRST members 'url': 'http://roosterteeth.com/episode/rt-docs-the-world-s-greatest-head-massage-the-world-s-greatest-head-massage-an-asmr-journey-part-one', 'only_matching': True, + }, { + 'url': 'https://roosterteeth.com/watch/million-dollars-but-season-2-million-dollars-but-the-game-announcement', + 'only_matching': True, }] def _login(self): diff --git a/youtube_dl/extractor/tvigle.py b/youtube_dl/extractor/tvigle.py index 3475ef4c3..180259aba 100644 --- a/youtube_dl/extractor/tvigle.py +++ b/youtube_dl/extractor/tvigle.py @@ -9,6 +9,8 @@ from ..utils import ( float_or_none, int_or_none, parse_age_limit, + try_get, + url_or_none, ) @@ -23,11 +25,10 @@ class TvigleIE(InfoExtractor): _TESTS = [ { 'url': 'http://www.tvigle.ru/video/sokrat/', - 'md5': '36514aed3657d4f70b4b2cef8eb520cd', 'info_dict': { 'id': '1848932', 'display_id': 'sokrat', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Сократ', 'description': 'md5:d6b92ffb7217b4b8ebad2e7665253c17', 'duration': 6586, @@ -37,7 +38,6 @@ class TvigleIE(InfoExtractor): }, { 'url': 'http://www.tvigle.ru/video/vladimir-vysotskii/vedushchii-teleprogrammy-60-minut-ssha-o-vladimire-vysotskom/', - 'md5': 'e7efe5350dd5011d0de6550b53c3ba7b', 'info_dict': { 'id': '5142516', 'ext': 'flv', @@ -62,7 +62,7 @@ class TvigleIE(InfoExtractor): webpage = self._download_webpage(url, display_id) video_id = self._html_search_regex( (r']+class=["\']player["\'][^>]+id=["\'](\d+)', - r'var\s+cloudId\s*=\s*["\'](\d+)', + r'cloudId\s*=\s*["\'](\d+)', r'class="video-preview current_playing" id="(\d+)"'), webpage, 'video id') @@ -90,21 +90,40 @@ class TvigleIE(InfoExtractor): age_limit = parse_age_limit(item.get('ageRestrictions')) formats = [] - for vcodec, fmts in item['videos'].items(): + for vcodec, url_or_fmts in item['videos'].items(): if vcodec == 'hls': - continue - for format_id, video_url in fmts.items(): - if format_id == 'm3u8': + m3u8_url = url_or_none(url_or_fmts) + if not m3u8_url: continue - height = self._search_regex( - r'^(\d+)[pP]$', format_id, 'height', default=None) - formats.append({ - 'url': video_url, - 'format_id': '%s-%s' % (vcodec, format_id), - 'vcodec': vcodec, - 'height': int_or_none(height), - 'filesize': int_or_none(item.get('video_files_size', {}).get(vcodec, {}).get(format_id)), - }) + formats.extend(self._extract_m3u8_formats( + m3u8_url, video_id, ext='mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False)) + elif vcodec == 'dash': + mpd_url = url_or_none(url_or_fmts) + if not mpd_url: + continue + formats.extend(self._extract_mpd_formats( + mpd_url, video_id, mpd_id='dash', fatal=False)) + else: + if not isinstance(url_or_fmts, dict): + continue + for format_id, video_url in url_or_fmts.items(): + if format_id == 'm3u8': + continue + video_url = url_or_none(video_url) + if not video_url: + continue + height = self._search_regex( + r'^(\d+)[pP]$', format_id, 'height', default=None) + filesize = int_or_none(try_get( + item, lambda x: x['video_files_size'][vcodec][format_id])) + formats.append({ + 'url': video_url, + 'format_id': '%s-%s' % (vcodec, format_id), + 'vcodec': vcodec, + 'height': int_or_none(height), + 'filesize': filesize, + }) self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/tvn24.py b/youtube_dl/extractor/tvn24.py index 6590e1fd0..de0fb5063 100644 --- a/youtube_dl/extractor/tvn24.py +++ b/youtube_dl/extractor/tvn24.py @@ -4,6 +4,7 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( int_or_none, + NO_DEFAULT, unescapeHTML, ) @@ -17,9 +18,21 @@ class TVN24IE(InfoExtractor): 'id': '1584444', 'ext': 'mp4', 'title': '"Święta mają być wesołe, dlatego, ludziska, wszyscy pod jemiołę"', - 'description': 'Wyjątkowe orędzie Artura Andrusa, jednego z gości "Szkła kontaktowego".', + 'description': 'Wyjątkowe orędzie Artura Andrusa, jednego z gości Szkła kontaktowego.', 'thumbnail': 're:https?://.*[.]jpeg', } + }, { + # different layout + 'url': 'https://tvnmeteo.tvn24.pl/magazyny/maja-w-ogrodzie,13/odcinki-online,1,4,1,0/pnacza-ptaki-i-iglaki-odc-691-hgtv-odc-29,1771763.html', + 'info_dict': { + 'id': '1771763', + 'ext': 'mp4', + 'title': 'Pnącza, ptaki i iglaki (odc. 691 /HGTV odc. 29)', + 'thumbnail': 're:https?://.*', + }, + 'params': { + 'skip_download': True, + }, }, { 'url': 'http://fakty.tvn24.pl/ogladaj-online,60/53-konferencja-bezpieczenstwa-w-monachium,716431.html', 'only_matching': True, @@ -35,18 +48,21 @@ class TVN24IE(InfoExtractor): }] def _real_extract(self, url): - video_id = self._match_id(url) + display_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) + webpage = self._download_webpage(url, display_id) - title = self._og_search_title(webpage) + title = self._og_search_title( + webpage, default=None) or self._search_regex( + r']+class=["\']magazineItemHeader[^>]+>(.+?)(?!\1).+?)\1' % attr, webpage, - name, group='json', fatal=fatal) or '{}', - video_id, transform_source=unescapeHTML, fatal=fatal) + name, group='json', default=default, fatal=fatal) or '{}', + display_id, transform_source=unescapeHTML, fatal=fatal) quality_data = extract_json('data-quality', 'formats') @@ -59,16 +75,24 @@ class TVN24IE(InfoExtractor): }) self._sort_formats(formats) - description = self._og_search_description(webpage) + description = self._og_search_description(webpage, default=None) thumbnail = self._og_search_thumbnail( webpage, default=None) or self._html_search_regex( r'\bdata-poster=(["\'])(?P(?!\1).+?)\1', webpage, 'thumbnail', group='url') + video_id = None + share_params = extract_json( - 'data-share-params', 'share params', fatal=False) + 'data-share-params', 'share params', default=None) if isinstance(share_params, dict): - video_id = share_params.get('id') or video_id + video_id = share_params.get('id') + + if not video_id: + video_id = self._search_regex( + r'data-vid-id=["\'](\d+)', webpage, 'video id', + default=None) or self._search_regex( + r',(\d+)\.html', url, 'video id', default=display_id) return { 'id': video_id, diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index b5b44a79a..ddf375c6c 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -2,12 +2,14 @@ from __future__ import unicode_literals import base64 +import functools import json import re import itertools from .common import InfoExtractor from ..compat import ( + compat_kwargs, compat_HTTPError, compat_str, compat_urlparse, @@ -19,6 +21,7 @@ from ..utils import ( int_or_none, merge_dicts, NO_DEFAULT, + OnDemandPagedList, parse_filesize, qualities, RegexNotFoundError, @@ -98,6 +101,13 @@ class VimeoBaseInfoExtractor(InfoExtractor): webpage, 'vuid', group='vuid') return xsrft, vuid + def _extract_vimeo_config(self, webpage, video_id, *args, **kwargs): + vimeo_config = self._search_regex( + r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', + webpage, 'vimeo config', *args, **compat_kwargs(kwargs)) + if vimeo_config: + return self._parse_json(vimeo_config, video_id) + def _set_vimeo_cookie(self, name, value): self._set_cookie('vimeo.com', name, value) @@ -253,7 +263,7 @@ class VimeoIE(VimeoBaseInfoExtractor): \. )? vimeo(?Ppro)?\.com/ - (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/) + (?!(?:channels|album|showcase)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/) (?:.*?/)? (?: (?: @@ -580,11 +590,9 @@ class VimeoIE(VimeoBaseInfoExtractor): # and latter we extract those that are Vimeo specific. self.report_extraction(video_id) - vimeo_config = self._search_regex( - r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage, - 'vimeo config', default=None) + vimeo_config = self._extract_vimeo_config(webpage, video_id, default=None) if vimeo_config: - seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {}) + seed_status = vimeo_config.get('seed_status', {}) if seed_status.get('state') == 'failed': raise ExtractorError( '%s said: %s' % (self.IE_NAME, seed_status['title']), @@ -905,7 +913,7 @@ class VimeoUserIE(VimeoChannelIE): class VimeoAlbumIE(VimeoChannelIE): IE_NAME = 'vimeo:album' - _VALID_URL = r'https://vimeo\.com/album/(?P\d+)(?:$|[?#]|/(?!video))' + _VALID_URL = r'https://vimeo\.com/(?:album|showcase)/(?P\d+)(?:$|[?#]|/(?!video))' _TITLE_RE = r'
\n\s*

(.*?)

' _TESTS = [{ 'url': 'https://vimeo.com/album/2632481', @@ -925,21 +933,39 @@ class VimeoAlbumIE(VimeoChannelIE): 'params': { 'videopassword': 'youtube-dl', } - }, { - 'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail', - 'only_matching': True, - }, { - # TODO: respect page number - 'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail', - 'only_matching': True, }] + _PAGE_SIZE = 100 - def _page_url(self, base_url, pagenum): - return '%s/page:%d/' % (base_url, pagenum) + def _fetch_page(self, album_id, authorizaion, hashed_pass, page): + api_page = page + 1 + query = { + 'fields': 'link', + 'page': api_page, + 'per_page': self._PAGE_SIZE, + } + if hashed_pass: + query['_hashed_pass'] = hashed_pass + videos = self._download_json( + 'https://api.vimeo.com/albums/%s/videos' % album_id, + album_id, 'Downloading page %d' % api_page, query=query, headers={ + 'Authorization': 'jwt ' + authorizaion, + })['data'] + for video in videos: + link = video.get('link') + if not link: + continue + yield self.url_result(link, VimeoIE.ie_key(), VimeoIE._match_id(link)) def _real_extract(self, url): album_id = self._match_id(url) - return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id) + webpage = self._download_webpage(url, album_id) + webpage = self._login_list_password(url, album_id, webpage) + api_config = self._extract_vimeo_config(webpage, album_id)['api'] + entries = OnDemandPagedList(functools.partial( + self._fetch_page, album_id, api_config['jwt'], + api_config.get('hashed_pass')), self._PAGE_SIZE) + return self.playlist_result(entries, album_id, self._html_search_regex( + r'\s*(.+?)(?:\s+on Vimeo)?', webpage, 'title', fatal=False)) class VimeoGroupsIE(VimeoAlbumIE): diff --git a/youtube_dl/extractor/yandexvideo.py b/youtube_dl/extractor/yandexvideo.py index 1aea95383..46529be05 100644 --- a/youtube_dl/extractor/yandexvideo.py +++ b/youtube_dl/extractor/yandexvideo.py @@ -3,6 +3,7 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( + determine_ext, int_or_none, url_or_none, ) @@ -47,6 +48,10 @@ class YandexVideoIE(InfoExtractor): # episode, sports 'url': 'https://yandex.ru/?stream_channel=1538487871&stream_id=4132a07f71fb0396be93d74b3477131d', 'only_matching': True, + }, { + # DASH with DRM + 'url': 'https://yandex.ru/portal/video?from=morda&stream_id=485a92d94518d73a9d0ff778e13505f8', + 'only_matching': True, }] def _real_extract(self, url): @@ -59,13 +64,22 @@ class YandexVideoIE(InfoExtractor): 'disable_trackings': 1, })['content'] - m3u8_url = url_or_none(content.get('content_url')) or url_or_none( + content_url = url_or_none(content.get('content_url')) or url_or_none( content['streams'][0]['url']) title = content.get('title') or content.get('computed_title') - formats = self._extract_m3u8_formats( - m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native', - m3u8_id='hls') + ext = determine_ext(content_url) + + if ext == 'm3u8': + formats = self._extract_m3u8_formats( + content_url, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls') + elif ext == 'mpd': + formats = self._extract_mpd_formats( + content_url, video_id, mpd_id='dash') + else: + formats = [{'url': content_url}] + self._sort_formats(formats) description = content.get('description') diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 9a182fcf6..57d76a5a2 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -31,6 +31,7 @@ from ..utils import ( clean_html, dict_get, error_to_compat_str, + extract_attributes, ExtractorError, float_or_none, get_element_by_attribute, @@ -324,17 +325,18 @@ class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): for video_id, video_title in self.extract_videos_from_page(content): yield self.url_result(video_id, 'Youtube', video_id, video_title) - def extract_videos_from_page(self, page): - ids_in_page = [] - titles_in_page = [] - for mobj in re.finditer(self._VIDEO_RE, page): + def extract_videos_from_page_impl(self, video_re, page, ids_in_page, titles_in_page): + for mobj in re.finditer(video_re, page): # The link with index 0 is not the first video of the playlist (not sure if still actual) if 'index' in mobj.groupdict() and mobj.group('id') == '0': continue video_id = mobj.group('id') - video_title = unescapeHTML(mobj.group('title')) + video_title = unescapeHTML( + mobj.group('title')) if 'title' in mobj.groupdict() else None if video_title: video_title = video_title.strip() + if video_title == '► Play all': + video_title = None try: idx = ids_in_page.index(video_id) if video_title and not titles_in_page[idx]: @@ -342,6 +344,12 @@ class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): except ValueError: ids_in_page.append(video_id) titles_in_page.append(video_title) + + def extract_videos_from_page(self, page): + ids_in_page = [] + titles_in_page = [] + self.extract_videos_from_page_impl( + self._VIDEO_RE, page, ids_in_page, titles_in_page) return zip(ids_in_page, titles_in_page) @@ -1595,17 +1603,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor): video_id = mobj.group(2) return video_id - def _extract_annotations(self, video_id): - return self._download_webpage( - 'https://www.youtube.com/annotations_invideo', video_id, - note='Downloading annotations', - errnote='Unable to download video annotations', fatal=False, - query={ - 'features': 1, - 'legacy': 1, - 'video_id': video_id, - }) - @staticmethod def _extract_chapters(description, duration): if not description: @@ -1700,6 +1697,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor): def extract_token(v_info): return dict_get(v_info, ('account_playback_token', 'accountPlaybackToken', 'token')) + def extract_player_response(player_response, video_id): + pl_response = str_or_none(player_response) + if not pl_response: + return + pl_response = self._parse_json(pl_response, video_id, fatal=False) + if isinstance(pl_response, dict): + add_dash_mpd_pr(pl_response) + return pl_response + player_response = {} # Get video info @@ -1722,7 +1728,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): note='Refetching age-gated info webpage', errnote='unable to download video info webpage') video_info = compat_parse_qs(video_info_webpage) + pl_response = video_info.get('player_response', [None])[0] + player_response = extract_player_response(pl_response, video_id) add_dash_mpd(video_info) + view_count = extract_view_count(video_info) else: age_gate = False video_info = None @@ -1745,11 +1754,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): is_live = True sts = ytplayer_config.get('sts') if not player_response: - pl_response = str_or_none(args.get('player_response')) - if pl_response: - pl_response = self._parse_json(pl_response, video_id, fatal=False) - if isinstance(pl_response, dict): - player_response = pl_response + player_response = extract_player_response(args.get('player_response'), video_id) if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True): add_dash_mpd_pr(player_response) # We also try looking in get_video_info since it may contain different dashmpd @@ -1781,9 +1786,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): get_video_info = compat_parse_qs(video_info_webpage) if not player_response: pl_response = get_video_info.get('player_response', [None])[0] - if isinstance(pl_response, dict): - player_response = pl_response - add_dash_mpd_pr(player_response) + player_response = extract_player_response(pl_response, video_id) add_dash_mpd(get_video_info) if view_count is None: view_count = extract_view_count(get_video_info) @@ -1806,9 +1809,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor): break def extract_unavailable_message(): - return self._html_search_regex( - r'(?s)]+id="unavailable-message"[^>]*>(.+?)', - video_webpage, 'unavailable message', default=None) + messages = [] + for tag, kind in (('h1', 'message'), ('div', 'submessage')): + msg = self._html_search_regex( + r'(?s)<{tag}[^>]+id=["\']unavailable-{kind}["\'][^>]*>(.+?)'.format(tag=tag, kind=kind), + video_webpage, 'unavailable %s' % kind, default=None) + if msg: + messages.append(msg) + if messages: + return '\n'.join(messages) if not video_info: unavailable_message = extract_unavailable_message() @@ -2092,9 +2101,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor): a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True' formats.append(a_format) else: - error_message = clean_html(video_info.get('reason', [None])[0]) + error_message = extract_unavailable_message() if not error_message: - error_message = extract_unavailable_message() + error_message = clean_html(try_get( + player_response, lambda x: x['playabilityStatus']['reason'], + compat_str)) + if not error_message: + error_message = clean_html( + try_get(video_info, lambda x: x['reason'][0], compat_str)) if error_message: raise ExtractorError(error_message, expected=True) raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info') @@ -2265,7 +2279,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # annotations video_annotations = None if self._downloader.params.get('writeannotations', False): - video_annotations = self._extract_annotations(video_id) + xsrf_token = self._search_regex( + r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P[A-Za-z0-9+/=]+)\2', + video_webpage, 'xsrf token', group='xsrf_token', fatal=False) + invideo_url = try_get( + player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str) + if xsrf_token and invideo_url: + xsrf_field_name = self._search_regex( + r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P\w+)\2', + video_webpage, 'xsrf field name', + group='xsrf_field_name', default='session_token') + video_annotations = self._download_webpage( + self._proto_relative_url(invideo_url), + video_id, note='Downloading annotations', + errnote='Unable to download video annotations', fatal=False, + data=urlencode_postdata({xsrf_field_name: xsrf_token})) chapters = self._extract_chapters(description_original, video_duration) @@ -2423,7 +2451,8 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): (%(playlist_id)s) )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE} _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s' - _VIDEO_RE = r'href="\s*/watch\?v=(?P[0-9A-Za-z_-]{11})(?:&(?:[^"]*?index=(?P\d+))?(?:[^>]+>(?P[^<]+))?)?' + _VIDEO_RE_TPL = r'href="\s*/watch\?v=%s(?:&(?:[^"]*?index=(?P<index>\d+))?(?:[^>]+>(?P<title>[^<]+))?)?' + _VIDEO_RE = _VIDEO_RE_TPL % r'(?P<id>[0-9A-Za-z_-]{11})' IE_NAME = 'youtube:playlist' _TESTS = [{ 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', @@ -2588,6 +2617,34 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): def _real_initialize(self): self._login() + def extract_videos_from_page(self, page): + ids_in_page = [] + titles_in_page = [] + + for item in re.findall( + r'(<[^>]*\bdata-video-id\s*=\s*["\'][0-9A-Za-z_-]{11}[^>]+>)', page): + attrs = extract_attributes(item) + video_id = attrs['data-video-id'] + video_title = unescapeHTML(attrs.get('data-title')) + if video_title: + video_title = video_title.strip() + ids_in_page.append(video_id) + titles_in_page.append(video_title) + + # Fallback with old _VIDEO_RE + self.extract_videos_from_page_impl( + self._VIDEO_RE, page, ids_in_page, titles_in_page) + + # Relaxed fallbacks + self.extract_videos_from_page_impl( + r'href="\s*/watch\?v\s*=\s*(?P<id>[0-9A-Za-z_-]{11})', page, + ids_in_page, titles_in_page) + self.extract_videos_from_page_impl( + r'data-video-ids\s*=\s*["\'](?P<id>[0-9A-Za-z_-]{11})', page, + ids_in_page, titles_in_page) + + return zip(ids_in_page, titles_in_page) + def _extract_mix(self, playlist_id): # The mixes are generated from a single video # the id of the playlist is just 'RD' + video_id diff --git a/youtube_dl/version.py b/youtube_dl/version.py index 04dc83605..b53a08cae 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,3 +1,3 @@ from __future__ import unicode_literals -__version__ = '2019.07.30' +__version__ = '2019.08.13'