From be306d6a313903a3ebdb8a8ff055bb6b58c9f818 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Fri, 2 Aug 2019 05:25:01 +0700 Subject: [PATCH 1/4] [tvigle] Fix extraction and add support for HLS and DASH formats (closes #21967) --- youtube_dl/extractor/tvigle.py | 53 +++++++++++++++++++++++----------- 1 file changed, 36 insertions(+), 17 deletions(-) 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 { From 2e9522b06173f2c5cfb2ba020958242d2a93feb7 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Fri, 2 Aug 2019 05:36:32 +0700 Subject: [PATCH 2/4] [ChangeLog] Actualize [ci skip] --- ChangeLog | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/ChangeLog b/ChangeLog index f6f1f7e38..c650e25d5 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,17 @@ +version + +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 From 4f2d735803f723a8d8d6ffbbb1dd6b203f71af58 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Fri, 2 Aug 2019 05:37:54 +0700 Subject: [PATCH 3/4] release 2019.08.02 --- .github/ISSUE_TEMPLATE/1_broken_site.md | 6 +- .../ISSUE_TEMPLATE/2_site_support_request.md | 4 +- .../ISSUE_TEMPLATE/3_site_feature_request.md | 4 +- .github/ISSUE_TEMPLATE/4_bug_report.md | 6 +- .github/ISSUE_TEMPLATE/5_feature_request.md | 4 +- CONTRIBUTING.md | 64 ------------------- ChangeLog | 2 +- youtube_dl/version.py | 2 +- 8 files changed, 14 insertions(+), 78 deletions(-) diff --git a/.github/ISSUE_TEMPLATE/1_broken_site.md b/.github/ISSUE_TEMPLATE/1_broken_site.md index ccd033716..4d3894ad3 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 **2019.08.02** - [ ] 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 2019.08.02 [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..796e11e54 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 **2019.08.02** - [ ] 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..aa2348548 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 **2019.08.02** - [ ] 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..5b2501a65 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 **2019.08.02** - [ ] 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 2019.08.02 [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..d1758a95c 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 **2019.08.02** - [ ] I've searched the bugtracker for similar feature requests including closed ones diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index d0e0a5637..cd9ccbe96 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -366,67 +366,3 @@ duration = float_or_none(video.get('durationMs'), scale=1000) view_count = int_or_none(video.get('views')) ``` -### 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 meta values. - -#### 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)) -``` - -### 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, -) -``` - diff --git a/ChangeLog b/ChangeLog index c650e25d5..7db147498 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,4 +1,4 @@ -version +version 2019.08.02 Extractors + [tvigle] Add support for HLS and DASH formats (#21967) diff --git a/youtube_dl/version.py b/youtube_dl/version.py index 04dc83605..0f7fdb23d 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__ = '2019.08.02' From d9d3a5a816253f14ee33623662690293365013e0 Mon Sep 17 00:00:00 2001 From: Sergey M Date: Fri, 2 Aug 2019 05:54:56 +0700 Subject: [PATCH 4/4] [README.md] Move code from #21939 to the right place --- README.md | 66 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 66 insertions(+) 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.