1
0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2026-01-13 10:21:30 +00:00

Compare commits

...

41 Commits

Author SHA1 Message Date
pukkandan
6aecd87106 Release 2021.06.23 2021-06-23 07:34:55 +05:30
pukkandan
ed807c1837 Update to ytdl-commit-379f52a
[liveleak] Remove extractor
379f52a495
2021-06-23 07:34:55 +05:30
Mevious
29f63c9672 [funimation] Extract subtitles (#434)
Closes #420, https://github.com/ytdl-org/youtube-dl/issues/25645
Related: https://github.com/ytdl-org/youtube-dl/pull/24906

Authored by: Mevious
2021-06-23 07:27:53 +05:30
pukkandan
9fc0de5796 [hotstar] Use server time for authentication instead of local time
Closes #396
2021-06-23 06:04:42 +05:30
siikamiika
c60ee3a218 [youtube_live_chat] Support ongoing live chat (#422)
Authored by: siikamiika
2021-06-23 05:42:39 +05:30
pukkandan
8a77e5e6bc [cleanup] Revert unnecessary changes in 51d9739f80 2021-06-23 05:34:40 +05:30
pukkandan
51d9739f80 Add option --throttled-rate below which video data is re-extracted
Currently only for HTTP downloads

Closes #430, workaround for https://github.com/ytdl-org/youtube-dl/issues/29326
2021-06-23 05:29:58 +05:30
pukkandan
4c7853de14 [fragment] Merge during download for -N, and refactor hls/dash (#364) 2021-06-22 00:29:50 +05:30
pukkandan
e6779b9400 [twitcasting] Websocket support (#399)
Closes #392
Authored by: nao20010128nao
2021-06-21 22:56:45 +05:30
pukkandan
e36d50c5dd [websockets] Add WebSocketFragmentFD (#399)
Necessary for #392

Co-authored by: nao20010128nao, pukkandan
2021-06-21 22:56:36 +05:30
pukkandan
ff0f78e1fe [aria2c] Lower --min-split-size for HTTP downloads
This makes downloading smaller files much faster
2021-06-20 19:28:54 +05:30
pukkandan
7e067091e8 [options] Rename --add-metadata to --embed-metadata
and clarify that it embeds chapter markers
2021-06-20 04:59:35 +05:30
pukkandan
f89b3e2d7a Skip fixup of existing files and add --fixup force to force it 2021-06-20 04:59:34 +05:30
pukkandan
fd7cfb6444 [cleanup] Refactor fixup 2021-06-20 04:26:11 +05:30
pukkandan
4e6767b5f2 [youtube] Temporary fix for age-gate
Related:
https://stackoverflow.com/a/67629882
https://github.com/yt-dlp/yt-dlp/issues/319
https://github.com/ytdl-org/youtube-dl/issues/29333
https://github.com/ytdl-org/youtube-dl/issues/29086
2021-06-18 20:32:52 +05:30
pukkandan
9fea350f0d Fix id sanitization in filenames
Closes #415
2021-06-17 02:32:24 +05:30
pukkandan
e858a9d6d3 [EmbedThumbnail] Add compat-option embed-thumbnail-atomicparsley
to force use of atomicparsley for embedding thumbnails in mp4

Related: #411
2021-06-16 22:33:32 +05:30
pukkandan
7e87e27c52 [postprocessor] Fix _restrict_to when a codec is not set 2021-06-14 14:09:22 +05:30
pukkandan
d0fb4bd16f [pornhub] Extract cast
Closes #406, https://github.com/ytdl-org/youtube-dl/pull/27384
2021-06-13 21:38:08 +05:30
felix
3fd4c2a543 [mediasite] Extract slides (#343)
Fixes:
https://github.com/ytdl-org/youtube-dl/issues/4974#issue-58006762
https://github.com/ytdl-org/youtube-dl/issues/4540#issuecomment-69574231
https://github.com/ytdl-org/youtube-dl/pull/11185#issuecomment-335554239

Authored by: fstirlitz
2021-06-13 20:36:40 +05:30
felix
cdb19aa4c2 [downloader/mhtml] Add new downloader (#343)
This downloader is intended to be used for streams that consist of a
timed sequence of stand-alone images, such as slideshows or thumbnail
streams

This can be used for implementing:

https://github.com/ytdl-org/youtube-dl/issues/4974#issue-58006762
https://github.com/ytdl-org/youtube-dl/issues/4540#issuecomment-69574231
https://github.com/ytdl-org/youtube-dl/pull/11185#issuecomment-335554239

https://github.com/ytdl-org/youtube-dl/issues/9868
https://github.com/ytdl-org/youtube-dl/pull/14951


Authored by: fstirlitz
2021-06-13 20:36:40 +05:30
pukkandan
4d85fbbdbb Fix bug in 8326b00aab 2021-06-13 14:36:13 +05:30
pukkandan
551f93885e Ignore images formats from merge 2021-06-13 04:16:42 +05:30
pukkandan
8326b00aab Allow images formats
Necessary for #343.

* They are identified by `vcodec=acodec='none'`
* These formats show as the worst in `-F`
* Any postprocessor that expects audio/video will be skipped
* `b*` and all related selectors will skip such formats
* This commit also does not add any selector for downloading such formats. They have to be explicitly requested by the `format_id`. Implementation of a selector is left for when #389 is resolved
2021-06-13 03:45:53 +05:30
pukkandan
b0249bcaf0 Expand --check-formats to thumbnails
Closes #402
2021-06-13 03:45:53 +05:30
pukkandan
21cd8fae49 Use NamedTemporaryFile for --check-formats 2021-06-13 03:45:53 +05:30
pukkandan
45db527fa6 [youtube] Login is not needed for :ytrec 2021-06-13 03:45:53 +05:30
pukkandan
28419ca2c8 [utils] Improve LazyList
* Add `repr` and `str` that mimics `list`
* Add `reversed`. Unlike `[::-1]`, reversed does not exhaust the iterable and modifies the `LazyList` in-place
* Add tests
2021-06-13 03:45:53 +05:30
pukkandan
8ba8714880 [EmbedThumbnail] Fix for already downloaded thumbnail 2021-06-11 19:13:24 +05:30
pukkandan
187986a857 Better error handling of syntax errors in -f 2021-06-11 19:13:22 +05:30
coletdjnz
4ba001080f [youtube] Non-fatal alert reporting for unavailable videos page (#401)
Co-Authored by: colethedj, pukkandan
2021-06-10 21:12:56 +00:00
coletdjnz
1974e99f4b [youtube] Improve SAPISID cookie handling (closes #393) (#395)
Author: colethedj
2021-06-10 21:02:57 +00:00
pukkandan
0181adefc6 [build] Build Windows x86 version with py3.7
and remove redundant tests
Closes #390

:ci skip

Co-authored by: pukkandan, shirt-dev
2021-06-10 01:41:04 +05:30
pukkandan
fd3c633d26 [version] update
:ci skip all
2021-06-10 01:36:46 +05:30
pukkandan
0d47c278d1 Release 2021.06.09 2021-06-10 00:49:48 +05:30
pukkandan
385a27fad1 Improve offset parsing in outtmpl 2021-06-09 20:01:57 +05:30
pukkandan
5c6542ce69 [test] More rigorous tests for prepare_filename
All tests of `prepare_outtmpl` is now also run on `prepare_filename`
2021-06-09 20:01:56 +05:30
pukkandan
639f1cea92 Fix %d and empty default in outtmpl
Closes #388
2021-06-09 15:37:15 +05:30
pukkandan
b5c5d84f60 Revert "[build] Build Windows x86 version with py3.8"
This reverts commit aa75e51f99.

See #390

This is being reverted instead of modified due to #388
2021-06-09 15:37:15 +05:30
pukkandan
aa75e51f99 [build] Build Windows x86 version with py3.8
and remove redundant tests
:ci skip

Ao-authored by: pukkandan, shirt-dev
2021-06-09 02:18:55 +05:30
pukkandan
884ce9d05d [version] update :ci skip all 2021-06-09 02:18:55 +05:30
59 changed files with 1432 additions and 991 deletions

3
.gitattributes vendored
View File

@@ -1 +1,4 @@
* text=auto
Makefile* text whitespace=-tab-in-indent
*.sh text eol=lf

View File

@@ -21,7 +21,7 @@ assignees: ''
<!--
Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.06.01. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.06.09. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
- Make sure that all provided video/audio/playlist URLs (if any) are alive and playable in a browser.
- Make sure that all URLs and arguments with special characters are properly quoted or escaped as explained in https://github.com/yt-dlp/yt-dlp.
- Search the bugtracker for similar issues: https://github.com/yt-dlp/yt-dlp. DO NOT post duplicates.
@@ -29,7 +29,7 @@ Carefully read and work through this check list in order to prevent the most com
-->
- [ ] I'm reporting a broken site support
- [ ] I've verified that I'm running yt-dlp version **2021.06.01**
- [ ] I've verified that I'm running yt-dlp version **2021.06.09**
- [ ] 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
@@ -44,7 +44,7 @@ Add the `-v` flag to your command line you run yt-dlp with (`yt-dlp -v <your com
[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] yt-dlp version 2021.06.01
[debug] yt-dlp version 2021.06.09
[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: {}

View File

@@ -21,7 +21,7 @@ assignees: ''
<!--
Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.06.01. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.06.09. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
- Make sure that all provided video/audio/playlist URLs (if any) are alive and playable in a browser.
- Make sure that site you are requesting is not dedicated to copyright infringement, see https://github.com/yt-dlp/yt-dlp. yt-dlp does not support such sites. In order for site support request to be accepted all provided example URLs should not violate any copyrights.
- Search the bugtracker for similar site support requests: https://github.com/yt-dlp/yt-dlp. DO NOT post duplicates.
@@ -29,7 +29,7 @@ Carefully read and work through this check list in order to prevent the most com
-->
- [ ] I'm reporting a new site support request
- [ ] I've verified that I'm running yt-dlp version **2021.06.01**
- [ ] I've verified that I'm running yt-dlp version **2021.06.09**
- [ ] 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

View File

@@ -21,13 +21,13 @@ assignees: ''
<!--
Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.06.01. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.06.09. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
- Search the bugtracker for similar site feature requests: https://github.com/yt-dlp/yt-dlp. DO NOT post duplicates.
- Finally, put x into all relevant boxes like this [x] (Dont forget to delete the empty space)
-->
- [ ] I'm reporting a site feature request
- [ ] I've verified that I'm running yt-dlp version **2021.06.01**
- [ ] I've verified that I'm running yt-dlp version **2021.06.09**
- [ ] I've searched the bugtracker for similar site feature requests including closed ones

View File

@@ -21,7 +21,7 @@ assignees: ''
<!--
Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.06.01. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.06.09. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
- Make sure that all provided video/audio/playlist URLs (if any) are alive and playable in a browser.
- Make sure that all URLs and arguments with special characters are properly quoted or escaped as explained in https://github.com/yt-dlp/yt-dlp.
- Search the bugtracker for similar issues: https://github.com/yt-dlp/yt-dlp. DO NOT post duplicates.
@@ -30,7 +30,7 @@ Carefully read and work through this check list in order to prevent the most com
-->
- [ ] I'm reporting a broken site support issue
- [ ] I've verified that I'm running yt-dlp version **2021.06.01**
- [ ] I've verified that I'm running yt-dlp version **2021.06.09**
- [ ] 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
@@ -46,7 +46,7 @@ Add the `-v` flag to your command line you run yt-dlp with (`yt-dlp -v <your com
[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] yt-dlp version 2021.06.01
[debug] yt-dlp version 2021.06.09
[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: {}

View File

@@ -21,13 +21,13 @@ assignees: ''
<!--
Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.06.01. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.06.09. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
- Search the bugtracker for similar feature requests: https://github.com/yt-dlp/yt-dlp. DO NOT post duplicates.
- Finally, put x into all relevant boxes like this [x] (Dont forget to delete the empty space)
-->
- [ ] I'm reporting a feature request
- [ ] I've verified that I'm running yt-dlp version **2021.06.01**
- [ ] I've verified that I'm running yt-dlp version **2021.06.09**
- [ ] I've searched the bugtracker for similar feature requests including closed ones

View File

@@ -95,14 +95,15 @@ jobs:
steps:
- uses: actions/checkout@v2
- name: Set up Python
# 3.8 is used for Win7 support
- name: Set up Python 3.8
uses: actions/setup-python@v2
with:
python-version: '3.8'
- name: Upgrade pip and enable wheel support
run: python -m pip install --upgrade pip setuptools wheel
- name: Install Requirements
run: pip install pyinstaller mutagen pycryptodome
run: pip install pyinstaller mutagen pycryptodome websockets
- name: Bump version
id: bump_version
run: python devscripts/update-version.py
@@ -137,15 +138,16 @@ jobs:
steps:
- uses: actions/checkout@v2
- name: Set up Python 3.4.4 32-Bit
# 3.7 is used for Vista support. See https://github.com/yt-dlp/yt-dlp/issues/390
- name: Set up Python 3.7 32-Bit
uses: actions/setup-python@v2
with:
python-version: '3.4.4'
python-version: '3.7'
architecture: 'x86'
- name: Upgrade pip and enable wheel support
run: python -m pip install pip==19.1.1 setuptools==43.0.0 wheel==0.33.6
- name: Install Requirements for 32 Bit
run: pip install pyinstaller==3.5 mutagen==1.42.0 pycryptodome==3.9.4 pefile==2019.4.18
run: python -m pip install --upgrade pip setuptools wheel
- name: Install Requirements
run: pip install pyinstaller mutagen pycryptodome websockets
- name: Bump version
id: bump_version
run: python devscripts/update-version.py

View File

@@ -9,11 +9,13 @@ jobs:
fail-fast: false
matrix:
os: [ubuntu-18.04]
python-version: [3.6, 3.7, 3.8, 3.9, pypy-3.6, pypy-3.7]
# py3.9 is in quick-test
python-version: [3.7, 3.8, pypy-3.6, pypy-3.7]
run-tests-ext: [sh]
include:
# atleast one of the tests must be in windows
- os: windows-latest
python-version: 3.4 # Windows x86 build is still in 3.4
python-version: 3.6
run-tests-ext: bat
steps:
- uses: actions/checkout@v2

View File

@@ -9,11 +9,11 @@ jobs:
fail-fast: true
matrix:
os: [ubuntu-18.04]
python-version: [3.6, 3.7, 3.8, 3.9, pypy-3.6, pypy-3.7]
python-version: [3.7, 3.8, 3.9, pypy-3.6, pypy-3.7]
run-tests-ext: [sh]
include:
- os: windows-latest
python-version: 3.4 # Windows x86 build is still in 3.4
python-version: 3.6
run-tests-ext: bat
steps:
- uses: actions/checkout@v2

View File

@@ -19,15 +19,54 @@
-->
### 2021.06.23
* Merge youtube-dl: Upto [commit/379f52a](https://github.com/ytdl-org/youtube-dl/commit/379f52a4954013767219d25099cce9e0f9401961)
* **Add option `--throttled-rate`** below which video data is re-extracted
* [fragment] **Merge during download for `-N`**, and refactor `hls`/`dash`
* [websockets] Add `WebSocketFragmentFD`by [nao20010128nao](https://github.com/nao20010128nao), [pukkandan](https://github.com/pukkandan)
* Allow `images` formats in addition to video/audio
* [downloader/mhtml] Add new downloader for slideshows/storyboards by [fstirlitz](https://github.com/fstirlitz)
* [youtube] Temporary **fix for age-gate**
* [youtube] Support ongoing live chat by [siikamiika](https://github.com/siikamiika)
* [youtube] Improve SAPISID cookie handling by [colethedj](https://github.com/colethedj)
* [youtube] Login is not needed for `:ytrec`
* [youtube] Non-fatal alert reporting for unavailable videos page by [colethedj](https://github.com/colethedj)
* [twitcasting] Websocket support by [nao20010128nao](https://github.com/nao20010128nao)
* [mediasite] Extract slides by [fstirlitz](https://github.com/fstirlitz)
* [funimation] Extract subtitles
* [pornhub] Extract `cast`
* [hotstar] Use server time for authentication instead of local time
* [EmbedThumbnail] Fix for already downloaded thumbnail
* [EmbedThumbnail] Add compat-option `embed-thumbnail-atomicparsley`
* Expand `--check-formats` to thumbnails
* Fix id sanitization in filenames
* Skip fixup of existing files and add `--fixup force` to force it
* Better error handling of syntax errors in `-f`
* Use `NamedTemporaryFile` for `--check-formats`
* [aria2c] Lower `--min-split-size` for HTTP downloads
* [options] Rename `--add-metadata` to `--embed-metadata`
* [utils] Improve `LazyList` and add tests
* [build] Build Windows x86 version with py3.7 and remove redundant tests by [pukkandan](https://github.com/pukkandan), [shirt](https://github.com/shirt-dev)
* [docs] Clarify that `--embed-metadata` embeds chapter markers
* [cleanup] Refactor fixup
### 2021.06.09
* Fix bug where `%(field)d` in filename template throws error
* Improve offset parsing in outtmpl
* [test] More rigorous tests for `prepare_filename`
### 2021.06.08
* Remove support for obsolete Python versions: Only 3.6+ is now supported
* Merge youtube-dl: Upto [commit/c2350ca](https://github.com/ytdl-org/youtube-dl/commit/c2350cac243ba1ec1586fe85b0d62d1b700047a2)
* [hls] Fix decryption for multithreaded downloader
* [extractor] Fix pre-checking archive for some extractors
* [extractor] Fix FourCC fallback when parsing ISM [fstirlitz](https://github.com/fstirlitz)
* [twitcasting] Add TwitCastingUserIE, TwitCastingLiveIE [pukkandan](https://github.com/pukkandan), [nao20010128nao](https://github.com/nao20010128nao)
* [vidio] Add VidioPremierIE and VidioLiveIE [minEplaYerspe](Https://github.com/MinePlayersPE)
* [extractor] Fix FourCC fallback when parsing ISM by [fstirlitz](https://github.com/fstirlitz)
* [twitcasting] Add TwitCastingUserIE, TwitCastingLiveIE by [pukkandan](https://github.com/pukkandan), [nao20010128nao](https://github.com/nao20010128nao)
* [vidio] Add VidioPremierIE and VidioLiveIE by [MinePlayersPE](Https://github.com/MinePlayersPE)
* [viki] Fix extraction from [ytdl-org/youtube-dl@59e583f](https://github.com/ytdl-org/youtube-dl/commit/59e583f7e8530ca92776c866897d895c072e2a82)
* [youtube] Support shorts URL
* [zoom] Extract transcripts as subtitles
@@ -35,13 +74,13 @@
* Fix and refactor `prepare_outtmpl`
* Make more fields available for `--print` when used with `--flat-playlist`
* [utils] Generalize `traverse_dict` to `traverse_obj`
* [downloader/ffmpeg] Hide FFmpeg banner unless in verbose mode [fstirlitz](https://github.com/fstirlitz)
* [downloader/ffmpeg] Hide FFmpeg banner unless in verbose mode by [fstirlitz](https://github.com/fstirlitz)
* [build] Release `yt-dlp.tar.gz`
* [build,update] Add GNU-style SHA512 and prepare updater for simlar SHA256 [nihil-admirari](https://github.com/nihil-admirari)
* [pyinst] Show Python version in exe metadata [nihil-admirari](https://github.com/nihil-admirari)
* [build,update] Add GNU-style SHA512 and prepare updater for simlar SHA256 by [nihil-admirari](https://github.com/nihil-admirari)
* [pyinst] Show Python version in exe metadata by [nihil-admirari](https://github.com/nihil-admirari)
* [docs] Improve documentation of dependencies
* [cleanup] Mark unused files
* [cleanup] Point all shebang to `python3` [fstirlitz](https://github.com/fstirlitz)
* [cleanup] Point all shebang to `python3` by [fstirlitz](https://github.com/fstirlitz)
* [cleanup] Remove duplicate file `trovolive.py`

View File

@@ -66,7 +66,7 @@ The major new features from the latest release of [blackjack4494/yt-dlc](https:/
* **[Format Sorting](#sorting-formats)**: The default format sorting options have been changed so that higher resolution and better codecs will be now preferred instead of simply using larger bitrate. Furthermore, you can now specify the sort order using `-S`. This allows for much easier format selection that what is possible by simply using `--format` ([examples](#format-selection-examples))
* **Merged with youtube-dl [commit/c2350ca](https://github.com/ytdl-org/youtube-dl/commit/c2350cac243ba1ec1586fe85b0d62d1b700047a2)**: (v2021.06.06) You get all the latest features and patches of [youtube-dl](https://github.com/ytdl-org/youtube-dl) in addition to all the features of [youtube-dlc](https://github.com/blackjack4494/yt-dlc)
* **Merged with youtube-dl [commit/379f52a](https://github.com/ytdl-org/youtube-dl/commit/379f52a4954013767219d25099cce9e0f9401961)**: (v2021.06.06) You get all the latest features and patches of [youtube-dl](https://github.com/ytdl-org/youtube-dl) in addition to all the features of [youtube-dlc](https://github.com/blackjack4494/yt-dlc)
* **Merged with animelover1984/youtube-dl**: You get most of the features and improvements from [animelover1984/youtube-dl](https://github.com/animelover1984/youtube-dl) including `--write-comments`, `BiliBiliSearch`, `BilibiliChannel`, Embedding thumbnail in mp4/ogg/opus, playlist infojson etc. Note that the NicoNico improvements are not available. See [#31](https://github.com/yt-dlp/yt-dlp/pull/31) for details.
@@ -131,6 +131,7 @@ Some of yt-dlp's default options are different from that of youtube-dl and youtu
* Youtube channel URLs are automatically redirected to `/video`. Append a `/featured` to the URL to download only the videos in the home page. If the channel does not have a videos tab, we try to download the equivalent `UU` playlist instead. Also, `/live` URLs raise an error if there are no live videos instead of silently downloading the entire channel. You may use `--compat-options no-youtube-channel-redirect` to revert all these redirections
* Unavailable videos are also listed for youtube playlists. Use `--compat-options no-youtube-unavailable-videos` to remove this
* If `ffmpeg` is used as the downloader, the downloading and merging of formats happen in a single step when possible. Use `--compat-options no-direct-merge` to revert this
* Thumbnail embedding in `mp4` is done with mutagen if possible. Use `--compat-options embed-thumbnail-atomicparsley` to force the use of AtomicParsley instead
For ease of use, a few more compat options are available:
* `--compat-options all`: Use all compat options
@@ -181,6 +182,7 @@ While all the other dependancies are optional, `ffmpeg` and `ffprobe` are highly
* [**sponskrub**](https://github.com/faissaloo/SponSkrub) - For using the [sponskrub options](#sponskrub-sponsorblock-options). Licenced under [GPLv3+](https://github.com/faissaloo/SponSkrub/blob/master/LICENCE.md)
* [**mutagen**](https://github.com/quodlibet/mutagen) - For embedding thumbnail in certain formats. Licenced under [GPLv2+](https://github.com/quodlibet/mutagen/blob/master/COPYING)
* [**pycryptodome**](https://github.com/Legrandin/pycryptodome) - For decrypting various data. Licenced under [BSD2](https://github.com/Legrandin/pycryptodome/blob/master/LICENSE.rst)
* [**websockets**](https://github.com/aaugustin/websockets) - For downloading over websocket. Licenced under [BSD3](https://github.com/aaugustin/websockets/blob/main/LICENSE)
* [**AtomicParsley**](https://github.com/wez/atomicparsley) - For embedding thumbnail in mp4/m4a if mutagen is not present. Licenced under [GPLv2+](https://github.com/wez/atomicparsley/blob/master/COPYING)
* [**rtmpdump**](http://rtmpdump.mplayerhq.hu) - For downloading `rtmp` streams. ffmpeg will be used as a fallback. Licenced under [GPLv2+](http://rtmpdump.mplayerhq.hu)
* [**mplayer**](http://mplayerhq.hu/design7/info.html) or [**mpv**](https://mpv.io) - For downloading `rstp` streams. ffmpeg will be used as a fallback. Licenced under [GPLv2+](https://github.com/mpv-player/mpv/blob/master/Copyright)
@@ -189,14 +191,14 @@ While all the other dependancies are optional, `ffmpeg` and `ffprobe` are highly
To use or redistribute the dependencies, you must agree to their respective licensing terms.
Note that the windows releases are already built with the python interpreter, mutagen and pycryptodome included.
Note that the windows releases are already built with the python interpreter, mutagen, pycryptodome and websockets included.
### COMPILE
**For Windows**:
To build the Windows executable, you must have pyinstaller (and optionally mutagen and pycryptodome)
To build the Windows executable, you must have pyinstaller (and optionally mutagen, pycryptodome, websockets)
python3 -m pip install --upgrade pyinstaller mutagen pycryptodome
python3 -m pip install --upgrade pyinstaller mutagen pycryptodome websockets
Once you have all the necessary dependencies installed, just run `py pyinst.py`. The executable will be built for the same architecture (32/64 bit) as the python used to build it.
@@ -372,6 +374,9 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
(default is 1)
-r, --limit-rate RATE Maximum download rate in bytes per second
(e.g. 50K or 4.2M)
--throttled-rate RATE Minimum download rate in bytes per second
below which throttling is assumed and the
video data is re-extracted (e.g. 100K)
-R, --retries RETRIES Number of retries (default is 10), or
"infinite"
--fragment-retries RETRIES Number of retries for a fragment (default
@@ -710,7 +715,8 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
Metadata, EmbedSubtitle, EmbedThumbnail,
SubtitlesConvertor, ThumbnailsConvertor,
VideoRemuxer, VideoConvertor, SponSkrub,
FixupStretched, FixupM4a and FixupM3u8. The
FixupStretched, FixupM4a, FixupM3u8,
FixupTimestamp and FixupDuration. The
supported executables are: AtomicParsley,
FFmpeg, FFprobe, and SponSkrub. You can
also specify "PP+EXE:ARGS" to give the
@@ -734,10 +740,13 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
--embed-subs Embed subtitles in the video (only for mp4,
webm and mkv videos)
--no-embed-subs Do not embed subtitles (default)
--embed-thumbnail Embed thumbnail in the audio as cover art
--embed-thumbnail Embed thumbnail in the video as cover art
--no-embed-thumbnail Do not embed thumbnail (default)
--add-metadata Write metadata to the video file
--no-add-metadata Do not write metadata (default)
--embed-metadata Embed metadata including chapter markers
(if supported by the format) to the video
file (Alias: --add-metadata)
--no-embed-metadata Do not write metadata (default)
(Alias: --no-add-metadata)
--parse-metadata FROM:TO Parse additional metadata like title/artist
from other fields; see "MODIFYING METADATA"
for details
@@ -747,7 +756,8 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
file. One of never (do nothing), warn (only
emit a warning), detect_or_warn (the
default; fix file if we can, warn
otherwise)
otherwise), force (try fixing even if file
already exists
--ffmpeg-location PATH Location of the ffmpeg binary; either the
path to the binary or its containing
directory
@@ -1140,7 +1150,7 @@ You can change the criteria for being considered the `best` by using `-S` (`--fo
- `lang`: Language preference as given by the extractor
- `quality`: The quality of the format as given by the extractor
- `source`: Preference of the source as given by the extractor
- `proto`: Protocol used for download (`https`/`ftps` > `http`/`ftp` > `m3u8_native` > `m3u8` > `http_dash_segments` > other > `mms`/`rtsp` > unknown > `f4f`/`f4m`)
- `proto`: Protocol used for download (`https`/`ftps` > `http`/`ftp` > `m3u8_native`/`m3u8` > `http_dash_segments`> `websocket_frag` > other > `mms`/`rtsp` > unknown > `f4f`/`f4m`)
- `vcodec`: Video Codec (`av01` > `vp9.2` > `vp9` > `h265` > `h264` > `vp8` > `h263` > `theora` > other > unknown)
- `acodec`: Audio Codec (`opus` > `vorbis` > `aac` > `mp4a` > `mp3` > `ac3` > `dts` > other > unknown)
- `codec`: Equivalent to `vcodec,acodec`

View File

@@ -6,6 +6,7 @@ import sys
# import os
import platform
from PyInstaller.utils.hooks import collect_submodules
from PyInstaller.utils.win32.versioninfo import (
VarStruct, VarFileInfo, StringStruct, StringTable,
StringFileInfo, FixedFileInfo, VSVersionInfo, SetVersion,
@@ -66,16 +67,15 @@ VERSION_FILE = VSVersionInfo(
]
)
dependancies = ['Crypto', 'mutagen'] + collect_submodules('websockets')
excluded_modules = ['test', 'ytdlp_plugins', 'youtube-dl', 'youtube-dlc']
PyInstaller.__main__.run([
'--name=yt-dlp%s' % _x86,
'--onefile',
'--icon=devscripts/cloud.ico',
'--exclude-module=youtube_dl',
'--exclude-module=youtube_dlc',
'--exclude-module=test',
'--exclude-module=ytdlp_plugins',
'--hidden-import=mutagen',
'--hidden-import=Crypto',
*[f'--exclude-module={module}' for module in excluded_modules],
*[f'--hidden-import={module}' for module in dependancies],
'--upx-exclude=vcruntime140.dll',
'yt_dlp/__main__.py',
])

View File

@@ -1,2 +1,3 @@
mutagen
pycryptodome
websockets

View File

@@ -19,7 +19,7 @@ LONG_DESCRIPTION = '\n\n'.join((
'**PS**: Some links in this document will not work since this is a copy of the README.md from Github',
open('README.md', 'r', encoding='utf-8').read()))
REQUIREMENTS = ['mutagen', 'pycryptodome']
REQUIREMENTS = ['mutagen', 'pycryptodome', 'websockets']
if sys.argv[1:2] == ['py2exe']:
raise NotImplementedError('py2exe is not currently supported; instead, use "pyinst.py" to build with pyinstaller')

View File

@@ -225,8 +225,7 @@
- **Culturebox**
- **CultureUnplugged**
- **curiositystream**
- **curiositystream:collections**
- **curiositystream:series**
- **curiositystream:collection**
- **CWTV**
- **DagelijkseKost**: dagelijksekost.een.be
- **DailyMail**
@@ -497,8 +496,6 @@
- **LinuxAcademy**
- **LiTV**
- **LiveJournal**
- **LiveLeak**
- **LiveLeakEmbed**
- **livestream**
- **livestream:original**
- **LnkGo**

View File

@@ -17,7 +17,7 @@ from yt_dlp.compat import compat_str, compat_urllib_error
from yt_dlp.extractor import YoutubeIE
from yt_dlp.extractor.common import InfoExtractor
from yt_dlp.postprocessor.common import PostProcessor
from yt_dlp.utils import ExtractorError, float_or_none, match_filter_func
from yt_dlp.utils import ExtractorError, int_or_none, match_filter_func
TEST_URL = 'http://localhost/sample.mp4'
@@ -461,14 +461,13 @@ class TestFormatSelection(unittest.TestCase):
def test_invalid_format_specs(self):
def assert_syntax_error(format_spec):
ydl = YDL({'format': format_spec})
info_dict = _make_result([{'format_id': 'foo', 'url': TEST_URL}])
self.assertRaises(SyntaxError, ydl.process_ie_result, info_dict)
self.assertRaises(SyntaxError, YDL, {'format': format_spec})
assert_syntax_error('bestvideo,,best')
assert_syntax_error('+bestaudio')
assert_syntax_error('bestvideo+')
assert_syntax_error('/')
assert_syntax_error('[720<height]')
def test_format_filtering(self):
formats = [
@@ -664,92 +663,115 @@ class TestYoutubeDL(unittest.TestCase):
'formats': [{'id': 'id1'}, {'id': 'id2'}, {'id': 'id3'}]
}
def test_prepare_outtmpl(self):
def out(tmpl, **params):
def test_prepare_outtmpl_and_filename(self):
def test(tmpl, expected, *, info=None, **params):
params['outtmpl'] = tmpl
ydl = YoutubeDL(params)
ydl._num_downloads = 1
err = ydl.validate_outtmpl(tmpl)
if err:
raise err
outtmpl, tmpl_dict = ydl.prepare_outtmpl(tmpl, self.outtmpl_info)
return outtmpl % tmpl_dict
self.assertEqual(ydl.validate_outtmpl(tmpl), None)
self.assertEqual(out('%(id)s.%(ext)s'), '1234.mp4')
self.assertEqual(out('%(duration_string)s'), '27:46:40')
self.assertTrue(float_or_none(out('%(epoch)d')))
self.assertEqual(out('%(resolution)s'), '1080p')
self.assertEqual(out('%(playlist_index)s'), '001')
self.assertEqual(out('%(autonumber)s'), '00001')
self.assertEqual(out('%(autonumber+2)03d', autonumber_start=3), '005')
self.assertEqual(out('%(autonumber)s', autonumber_size=3), '001')
outtmpl, tmpl_dict = ydl.prepare_outtmpl(tmpl, info or self.outtmpl_info)
out = outtmpl % tmpl_dict
fname = ydl.prepare_filename(info or self.outtmpl_info)
self.assertEqual(out('%%'), '%')
self.assertEqual(out('%%%%'), '%%')
self.assertEqual(out('%(invalid@tmpl|def)s', outtmpl_na_placeholder='none'), 'none')
self.assertEqual(out('%()s'), 'NA')
self.assertEqual(out('%s'), '%s')
self.assertEqual(out('%d'), '%d')
self.assertRaises(ValueError, out, '%')
self.assertRaises(ValueError, out, '%(title)')
if callable(expected):
self.assertTrue(expected(out))
self.assertTrue(expected(fname))
elif isinstance(expected, compat_str):
self.assertEqual((out, fname), (expected, expected))
else:
self.assertEqual((out, fname), expected)
# Auto-generated fields
test('%(id)s.%(ext)s', '1234.mp4')
test('%(duration_string)s', ('27:46:40', '27-46-40'))
test('%(epoch)d', int_or_none)
test('%(resolution)s', '1080p')
test('%(playlist_index)s', '001')
test('%(autonumber)s', '00001')
test('%(autonumber+2)03d', '005', autonumber_start=3)
test('%(autonumber)s', '001', autonumber_size=3)
# Escaping %
test('%%', '%')
test('%%%%', '%%')
test('%%(width)06d.%(ext)s', '%(width)06d.mp4')
test('%(width)06d.%(ext)s', 'NA.mp4')
test('%(width)06d.%%(ext)s', 'NA.%(ext)s')
test('%%(width)06d.%(ext)s', '%(width)06d.mp4')
# ID sanitization
test('%(id)s', '_abcd', info={'id': '_abcd'})
test('%(some_id)s', '_abcd', info={'some_id': '_abcd'})
test('%(formats.0.id)s', '_abcd', info={'formats': [{'id': '_abcd'}]})
test('%(id)s', '-abcd', info={'id': '-abcd'})
test('%(id)s', '.abcd', info={'id': '.abcd'})
test('%(id)s', 'ab__cd', info={'id': 'ab__cd'})
test('%(id)s', ('ab:cd', 'ab -cd'), info={'id': 'ab:cd'})
# Invalid templates
self.assertTrue(isinstance(YoutubeDL.validate_outtmpl('%'), ValueError))
self.assertTrue(isinstance(YoutubeDL.validate_outtmpl('%(title)'), ValueError))
test('%(invalid@tmpl|def)s', 'none', outtmpl_na_placeholder='none')
test('%()s', 'NA')
test('%s', '%s')
test('%d', '%d')
# NA placeholder
NA_TEST_OUTTMPL = '%(uploader_date)s-%(width)d-%(x|def)s-%(id)s.%(ext)s'
self.assertEqual(out(NA_TEST_OUTTMPL), 'NA-NA-def-1234.mp4')
self.assertEqual(out(NA_TEST_OUTTMPL, outtmpl_na_placeholder='none'), 'none-none-def-1234.mp4')
self.assertEqual(out(NA_TEST_OUTTMPL, outtmpl_na_placeholder=''), '--def-1234.mp4')
test(NA_TEST_OUTTMPL, 'NA-NA-def-1234.mp4')
test(NA_TEST_OUTTMPL, 'none-none-def-1234.mp4', outtmpl_na_placeholder='none')
test(NA_TEST_OUTTMPL, '--def-1234.mp4', outtmpl_na_placeholder='')
# String formatting
FMT_TEST_OUTTMPL = '%%(height)%s.%%(ext)s'
self.assertEqual(out(FMT_TEST_OUTTMPL % 's'), '1080.mp4')
self.assertEqual(out(FMT_TEST_OUTTMPL % 'd'), '1080.mp4')
self.assertEqual(out(FMT_TEST_OUTTMPL % '6d'), ' 1080.mp4')
self.assertEqual(out(FMT_TEST_OUTTMPL % '-6d'), '1080 .mp4')
self.assertEqual(out(FMT_TEST_OUTTMPL % '06d'), '001080.mp4')
self.assertEqual(out(FMT_TEST_OUTTMPL % ' 06d'), ' 01080.mp4')
self.assertEqual(out(FMT_TEST_OUTTMPL % ' 06d'), ' 01080.mp4')
self.assertEqual(out(FMT_TEST_OUTTMPL % '0 6d'), ' 01080.mp4')
self.assertEqual(out(FMT_TEST_OUTTMPL % '0 6d'), ' 01080.mp4')
self.assertEqual(out(FMT_TEST_OUTTMPL % ' 0 6d'), ' 01080.mp4')
test(FMT_TEST_OUTTMPL % 's', '1080.mp4')
test(FMT_TEST_OUTTMPL % 'd', '1080.mp4')
test(FMT_TEST_OUTTMPL % '6d', ' 1080.mp4')
test(FMT_TEST_OUTTMPL % '-6d', '1080 .mp4')
test(FMT_TEST_OUTTMPL % '06d', '001080.mp4')
test(FMT_TEST_OUTTMPL % ' 06d', ' 01080.mp4')
test(FMT_TEST_OUTTMPL % ' 06d', ' 01080.mp4')
test(FMT_TEST_OUTTMPL % '0 6d', ' 01080.mp4')
test(FMT_TEST_OUTTMPL % '0 6d', ' 01080.mp4')
test(FMT_TEST_OUTTMPL % ' 0 6d', ' 01080.mp4')
self.assertEqual(out('%(id)d'), '1234')
self.assertEqual(out('%(height)c'), '1')
self.assertEqual(out('%(ext)c'), 'm')
self.assertEqual(out('%(id)d %(id)r'), "1234 '1234'")
self.assertEqual(out('%(ext)s-%(ext|def)d'), 'mp4-def')
self.assertEqual(out('%(width|0)04d'), '0000')
self.assertEqual(out('%(width|)d', outtmpl_na_placeholder='none'), '')
# Type casting
test('%(id)d', '1234')
test('%(height)c', '1')
test('%(ext)c', 'm')
test('%(id)d %(id)r', "1234 '1234'")
test('%(id)r %(height)r', "'1234' 1080")
test('%(ext)s-%(ext|def)d', 'mp4-def')
test('%(width|0)04d', '0000')
test('a%(width|)d', 'a', outtmpl_na_placeholder='none')
# Internal formatting
FORMATS = self.outtmpl_info['formats']
self.assertEqual(out('%(timestamp+-1000>%H-%M-%S)s'), '11-43-20')
self.assertEqual(out('%(id+1-height+3)05d'), '00158')
self.assertEqual(out('%(width+100)05d'), 'NA')
self.assertEqual(out('%(formats.0)s'), str(FORMATS[0]))
self.assertEqual(out('%(height.0)03d'), '001')
self.assertEqual(out('%(formats.-1.id)s'), str(FORMATS[-1]['id']))
self.assertEqual(out('%(formats.3)s'), 'NA')
self.assertEqual(out('%(formats.:2:-1)r'), repr(FORMATS[:2:-1]))
self.assertEqual(out('%(formats.0.id.-1+id)f'), '1235.000000')
test('%(timestamp-1000>%H-%M-%S)s', '11-43-20')
test('%(id+1-height+3)05d', '00158')
test('%(width+100)05d', 'NA')
test('%(formats.0) 15s', ('% 15s' % FORMATS[0], '% 15s' % str(FORMATS[0]).replace(':', ' -')))
test('%(formats.0)r', (repr(FORMATS[0]), repr(FORMATS[0]).replace(':', ' -')))
test('%(height.0)03d', '001')
test('%(-height.0)04d', '-001')
test('%(formats.-1.id)s', FORMATS[-1]['id'])
test('%(formats.0.id.-1)d', FORMATS[0]['id'][-1])
test('%(formats.3)s', 'NA')
test('%(formats.:2:-1)r', repr(FORMATS[:2:-1]))
test('%(formats.0.id.-1+id)f', '1235.000000')
test('%(formats.0.id.-1+formats.1.id.-1)d', '3')
def test_prepare_filename(self):
def fname(templ):
params = {'outtmpl': templ}
ydl = YoutubeDL(params)
return ydl.prepare_filename(self.outtmpl_info)
# Empty filename
test('%(foo|)s-%(bar|)s.%(ext)s', '-.mp4')
# test('%(foo|)s.%(ext)s', ('.mp4', '_.mp4')) # fixme
# test('%(foo|)s', ('', '_')) # fixme
self.assertEqual(fname('%%'), '%')
self.assertEqual(fname('%%%%'), '%%')
self.assertEqual(fname('%%(width)06d.%(ext)s'), '%(width)06d.mp4')
self.assertEqual(fname('%(width)06d.%(ext)s'), 'NA.mp4')
self.assertEqual(fname('%(width)06d.%%(ext)s'), 'NA.%(ext)s')
self.assertEqual(fname('%%(width)06d.%(ext)s'), '%(width)06d.mp4')
self.assertEqual(fname('Hello %(title1)s'), 'Hello $PATH')
self.assertEqual(fname('Hello %(title2)s'), 'Hello %PATH%')
self.assertEqual(fname('%(title3)s'), 'foo_bar_test')
self.assertEqual(fname('%(formats.0)s'), "{'id' - 'id1'}")
self.assertEqual(fname('%(id)r %(height)r'), "'1234' 1080")
self.assertEqual(fname('%(formats.0)r'), "{'id' - 'id1'}")
# Path expansion and escaping
test('Hello %(title1)s', 'Hello $PATH')
test('Hello %(title2)s', 'Hello %PATH%')
test('%(title3)s', ('foo/bar\\test', 'foo_bar_test'))
test('folder/%(title3)s', ('folder/foo/bar\\test', 'folder%sfoo_bar_test' % os.path.sep))
def test_format_note(self):
ydl = YoutubeDL()

View File

@@ -12,6 +12,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
# Various small unit tests
import io
import itertools
import json
import xml.etree.ElementTree
@@ -108,6 +109,7 @@ from yt_dlp.utils import (
cli_bool_option,
parse_codecs,
iri_to_uri,
LazyList,
)
from yt_dlp.compat import (
compat_chr,
@@ -126,6 +128,7 @@ class TestUtil(unittest.TestCase):
self.assertTrue(timeconvert('bougrg') is None)
def test_sanitize_filename(self):
self.assertEqual(sanitize_filename(''), '')
self.assertEqual(sanitize_filename('abc'), 'abc')
self.assertEqual(sanitize_filename('abc_d-e'), 'abc_d-e')
@@ -1524,6 +1527,47 @@ Line 1
self.assertEqual(clean_podcast_url('https://www.podtrac.com/pts/redirect.mp3/chtbl.com/track/5899E/traffic.megaphone.fm/HSW7835899191.mp3'), 'https://traffic.megaphone.fm/HSW7835899191.mp3')
self.assertEqual(clean_podcast_url('https://play.podtrac.com/npr-344098539/edge1.pod.npr.org/anon.npr-podcasts/podcast/npr/waitwait/2020/10/20201003_waitwait_wwdtmpodcast201003-015621a5-f035-4eca-a9a1-7c118d90bc3c.mp3'), 'https://edge1.pod.npr.org/anon.npr-podcasts/podcast/npr/waitwait/2020/10/20201003_waitwait_wwdtmpodcast201003-015621a5-f035-4eca-a9a1-7c118d90bc3c.mp3')
def test_LazyList(self):
it = list(range(10))
self.assertEqual(list(LazyList(it)), it)
self.assertEqual(LazyList(it).exhaust(), it)
self.assertEqual(LazyList(it)[5], it[5])
self.assertEqual(LazyList(it)[::2], it[::2])
self.assertEqual(LazyList(it)[1::2], it[1::2])
self.assertEqual(LazyList(it)[6:2:-2], it[6:2:-2])
self.assertEqual(LazyList(it)[::-1], it[::-1])
self.assertTrue(LazyList(it))
self.assertFalse(LazyList(range(0)))
self.assertEqual(len(LazyList(it)), len(it))
self.assertEqual(repr(LazyList(it)), repr(it))
self.assertEqual(str(LazyList(it)), str(it))
self.assertEqual(list(reversed(LazyList(it))), it[::-1])
self.assertEqual(list(reversed(LazyList(it))[1:3:7]), it[::-1][1:3:7])
def test_LazyList_laziness(self):
def test(ll, idx, val, cache):
self.assertEqual(ll[idx], val)
self.assertEqual(getattr(ll, '_LazyList__cache'), list(cache))
ll = LazyList(range(10))
test(ll, 0, 0, range(1))
test(ll, 5, 5, range(6))
test(ll, -3, 7, range(10))
ll = reversed(LazyList(range(10)))
test(ll, -1, 0, range(1))
test(ll, 3, 6, range(10))
ll = LazyList(itertools.count())
test(ll, 10, 10, range(11))
reversed(ll)
test(ll, -15, 14, range(15))
if __name__ == '__main__':
unittest.main()

View File

@@ -20,6 +20,7 @@ import re
import shutil
import subprocess
import sys
import tempfile
import time
import tokenize
import traceback
@@ -67,6 +68,7 @@ from .utils import (
STR_FORMAT_RE,
formatSeconds,
GeoRestrictedError,
HEADRequest,
int_or_none,
iri_to_uri,
ISO3166Utils,
@@ -86,7 +88,6 @@ from .utils import (
preferredencoding,
prepend_extension,
process_communicate_or_kill,
random_uuidv4,
register_socks_protocols,
RejectedVideoReached,
render_table,
@@ -100,6 +101,7 @@ from .utils import (
str_or_none,
strftime_or_none,
subtitles_filename,
ThrottledDownload,
to_high_limit_path,
traverse_obj,
UnavailableVideoError,
@@ -126,13 +128,14 @@ from .downloader import (
)
from .downloader.rtmp import rtmpdump_version
from .postprocessor import (
get_postprocessor,
FFmpegFixupDurationPP,
FFmpegFixupM3u8PP,
FFmpegFixupM4aPP,
FFmpegFixupStretchedPP,
FFmpegFixupTimestampPP,
FFmpegMergerPP,
FFmpegPostProcessor,
# FFmpegSubtitlesConvertorPP,
get_postprocessor,
MoveFilesAfterDownloadPP,
)
from .version import __version__
@@ -390,15 +393,15 @@ class YoutubeDL(object):
compat_opts: Compatibility options. See "Differences in default behavior".
Note that only format-sort, format-spec, no-live-chat,
no-attach-info-json, playlist-index, list-formats,
no-direct-merge, no-youtube-channel-redirect,
and no-youtube-unavailable-videos works when used via the API
no-direct-merge, embed-thumbnail-atomicparsley,
no-youtube-unavailable-videos, no-youtube-channel-redirect,
works when used via the API
The following parameters are not used by YoutubeDL itself, they are used by
the downloader (see yt_dlp/downloader/common.py):
nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
noresizebuffer, retries, continuedl, noprogress, consoletitle,
xattr_set_filesize, external_downloader_args, hls_use_mpegts,
http_chunk_size.
nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
max_filesize, test, noresizebuffer, retries, continuedl, noprogress, consoletitle,
xattr_set_filesize, external_downloader_args, hls_use_mpegts, http_chunk_size.
The following options are used by the post processors:
prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
@@ -472,8 +475,7 @@ class YoutubeDL(object):
if sys.version_info < (3, 6):
self.report_warning(
'Support for Python version %d.%d have been deprecated and will break in future versions of yt-dlp! '
'Update to Python 3.6 or above' % sys.version_info[:2])
'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
def check_deprecated(param, option, suggestion):
if self.params.get(param) is not None:
@@ -539,6 +541,11 @@ class YoutubeDL(object):
self.outtmpl_dict = self.parse_outtmpl()
# Creating format selector here allows us to catch syntax errors before the extraction
self.format_selector = (
None if self.params.get('format') is None
else self.build_format_selector(self.params['format']))
self._setup_opener()
"""Preload the archive, if any is specified"""
@@ -564,14 +571,9 @@ class YoutubeDL(object):
self.add_default_info_extractors()
for pp_def_raw in self.params.get('postprocessors', []):
pp_class = get_postprocessor(pp_def_raw['key'])
pp_def = dict(pp_def_raw)
del pp_def['key']
if 'when' in pp_def:
when = pp_def['when']
del pp_def['when']
else:
when = 'post_process'
when = pp_def.pop('when', 'post_process')
pp_class = get_postprocessor(pp_def.pop('key'))
pp = pp_class(self, **compat_kwargs(pp_def))
self.add_post_processor(pp, when=when)
@@ -813,6 +815,21 @@ class YoutubeDL(object):
'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
return outtmpl_dict
def get_output_path(self, dir_type='', filename=None):
paths = self.params.get('paths', {})
assert isinstance(paths, dict)
path = os.path.join(
expand_path(paths.get('home', '').strip()),
expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
filename or '')
# Temporary fix for #4787
# 'Treat' all problem characters by passing filename through preferredencoding
# to workaround encoding issues with subprocess on python2 @ Windows
if sys.version_info < (3, 0) and sys.platform == 'win32':
path = encodeFilename(path, True).decode(preferredencoding())
return sanitize_path(path, force=self.params.get('windowsfilenames'))
@staticmethod
def validate_outtmpl(tmpl):
''' @return None or Exception object '''
@@ -847,23 +864,24 @@ class YoutubeDL(object):
'autonumber': self.params.get('autonumber_size') or 5,
}
EXTERNAL_FORMAT_RE = STR_FORMAT_RE.format('[^)]*')
# Field is of the form key1.key2...
# where keys (except first) can be string, int or slice
FIELD_RE = r'\w+(?:\.(?:\w+|[-\d]*(?::[-\d]*){0,2}))*'
INTERNAL_FORMAT_RE = re.compile(r'''(?x)
(?P<negate>-)?
(?P<fields>{0})
(?P<maths>(?:[-+]-?(?:\d+(?:\.\d+)?|{0}))*)
(?:>(?P<strf_format>.+?))?
(?:\|(?P<default>.*?))?
$'''.format(FIELD_RE))
MATH_OPERATORS_RE = re.compile(r'(?<![-+])([-+])')
TMPL_DICT = {}
EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE.format('[^)]*'))
MATH_FUNCTIONS = {
'+': float.__add__,
'-': float.__sub__,
}
tmpl_dict = {}
# Field is of the form key1.key2...
# where keys (except first) can be string, int or slice
FIELD_RE = r'\w+(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
MATH_FIELD_RE = r'''{field}|{num}'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
INTERNAL_FORMAT_RE = re.compile(r'''(?x)
(?P<negate>-)?
(?P<fields>{field})
(?P<maths>(?:{math_op}{math_field})*)
(?:>(?P<strf_format>.+?))?
(?:\|(?P<default>.*?))?
$'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
get_key = lambda k: traverse_obj(
info_dict, k.split('.'), is_user_input=True, traverse_string=True)
@@ -877,24 +895,27 @@ class YoutubeDL(object):
if value is not None:
value *= -1
# Do maths
if mdict['maths']:
offset_key = mdict['maths']
if offset_key:
value = float_or_none(value)
operator = None
for item in MATH_OPERATORS_RE.split(mdict['maths'])[1:]:
if item == '' or value is None:
return None
if operator:
item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
offset = float_or_none(item)
if offset is None:
offset = float_or_none(get_key(item))
try:
value = operator(value, multiplier * offset)
except (TypeError, ZeroDivisionError):
return None
operator = None
else:
while offset_key:
item = re.match(
MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
offset_key).group(0)
offset_key = offset_key[len(item):]
if operator is None:
operator = MATH_FUNCTIONS[item]
continue
item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
offset = float_or_none(item)
if offset is None:
offset = float_or_none(get_key(item))
try:
value = operator(value, multiplier * offset)
except (TypeError, ZeroDivisionError):
return None
operator = None
# Datetime formatting
if mdict['strf_format']:
value = strftime_or_none(value, mdict['strf_format'])
@@ -909,7 +930,7 @@ class YoutubeDL(object):
fmt = outer_mobj.group('format')
mobj = re.match(INTERNAL_FORMAT_RE, key)
if mobj is None:
value, default = None, na
value, default, mobj = None, na, {'fields': ''}
else:
mobj = mobj.groupdict()
default = mobj['default'] if mobj['default'] is not None else na
@@ -919,7 +940,6 @@ class YoutubeDL(object):
fmt = '0{:d}d'.format(field_size_compat_map[key])
value = default if value is None else value
key += '\0%s' % fmt
if fmt == 'c':
value = compat_str(value)
@@ -936,11 +956,13 @@ class YoutubeDL(object):
# If value is an object, sanitize might convert it to a string
# So we convert it to repr first
value, fmt = repr(value), '%ss' % fmt[:-1]
value = sanitize(key, value)
tmpl_dict[key] = value
if fmt[-1] in 'csr':
value = sanitize(mobj['fields'].split('.')[-1], value)
key += '\0%s' % fmt
TMPL_DICT[key] = value
return '%({key}){fmt}'.format(key=key, fmt=fmt)
return re.sub(EXTERNAL_FORMAT_RE, create_key, outtmpl), tmpl_dict
return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
def _prepare_filename(self, info_dict, tmpl_type='default'):
try:
@@ -985,12 +1007,11 @@ class YoutubeDL(object):
def prepare_filename(self, info_dict, dir_type='', warn=False):
"""Generate the output filename."""
paths = self.params.get('paths', {})
assert isinstance(paths, dict)
filename = self._prepare_filename(info_dict, dir_type or 'default')
if warn and not self.__prepare_filename_warned:
if not paths:
if not self.params.get('paths'):
pass
elif filename == '-':
self.report_warning('--paths is ignored when an outputting to stdout')
@@ -1000,18 +1021,7 @@ class YoutubeDL(object):
if filename == '-' or not filename:
return filename
homepath = expand_path(paths.get('home', '').strip())
assert isinstance(homepath, compat_str)
subdir = expand_path(paths.get(dir_type, '').strip()) if dir_type else ''
assert isinstance(subdir, compat_str)
path = os.path.join(homepath, subdir, filename)
# Temporary fix for #4787
# 'Treat' all problem characters by passing filename through preferredencoding
# to workaround encoding issues with subprocess on python2 @ Windows
if sys.version_info < (3, 0) and sys.platform == 'win32':
path = encodeFilename(path, True).decode(preferredencoding())
return sanitize_path(path, force=self.params.get('windowsfilenames'))
return self.get_output_path(dir_type, filename)
def _match_entry(self, info_dict, incomplete=False, silent=False):
""" Returns None if the file should be downloaded """
@@ -1135,6 +1145,10 @@ class YoutubeDL(object):
self.report_error(msg)
except ExtractorError as e: # An error we somewhat expected
self.report_error(compat_str(e), e.format_traceback())
except ThrottledDownload:
self.to_stderr('\r')
self.report_warning('The download speed is below throttle limit. Re-extracting data')
return wrapper(self, *args, **kwargs)
except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
raise
except Exception as e:
@@ -1483,12 +1497,11 @@ class YoutubeDL(object):
'!=': operator.ne,
}
operator_rex = re.compile(r'''(?x)\s*
(?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
\s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
(?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
$
(?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
(?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
''' % '|'.join(map(re.escape, OPERATORS.keys())))
m = operator_rex.search(filter_spec)
m = operator_rex.fullmatch(filter_spec)
if m:
try:
comparison_value = int(m.group('value'))
@@ -1509,13 +1522,12 @@ class YoutubeDL(object):
'$=': lambda attr, value: attr.endswith(value),
'*=': lambda attr, value: value in attr,
}
str_operator_rex = re.compile(r'''(?x)
\s*(?P<key>[a-zA-Z0-9._-]+)
\s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
\s*(?P<value>[a-zA-Z0-9._-]+)
\s*$
str_operator_rex = re.compile(r'''(?x)\s*
(?P<key>[a-zA-Z0-9._-]+)\s*
(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
(?P<value>[a-zA-Z0-9._-]+)\s*
''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
m = str_operator_rex.search(filter_spec)
m = str_operator_rex.fullmatch(filter_spec)
if m:
comparison_value = m.group('value')
str_op = STR_OPERATORS[m.group('op')]
@@ -1525,7 +1537,7 @@ class YoutubeDL(object):
op = str_op
if not m:
raise ValueError('Invalid filter specification %r' % filter_spec)
raise SyntaxError('Invalid filter specification %r' % filter_spec)
def _filter(f):
actual_value = f.get(m.group('key'))
@@ -1680,9 +1692,12 @@ class YoutubeDL(object):
formats_info.extend(format_2.get('requested_formats', (format_2,)))
if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
get_no_more = {"video": False, "audio": False}
get_no_more = {'video': False, 'audio': False}
for (i, fmt_info) in enumerate(formats_info):
for aud_vid in ["audio", "video"]:
if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
formats_info.pop(i)
continue
for aud_vid in ['audio', 'video']:
if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
if get_no_more[aud_vid]:
formats_info.pop(i)
@@ -1735,18 +1750,20 @@ class YoutubeDL(object):
def _check_formats(formats):
for f in formats:
self.to_screen('[info] Testing format %s' % f['format_id'])
paths = self.params.get('paths', {})
temp_file = os.path.join(
expand_path(paths.get('home', '').strip()),
expand_path(paths.get('temp', '').strip()),
'ytdl.%s.f%s.check-format' % (random_uuidv4(), f['format_id']))
temp_file = tempfile.NamedTemporaryFile(
suffix='.tmp', delete=False,
dir=self.get_output_path('temp') or None)
temp_file.close()
try:
dl, _ = self.dl(temp_file, f, test=True)
dl, _ = self.dl(temp_file.name, f, test=True)
except (ExtractorError, IOError, OSError, ValueError) + network_exceptions:
dl = False
finally:
if os.path.exists(temp_file):
os.remove(temp_file)
if os.path.exists(temp_file.name):
try:
os.remove(temp_file.name)
except OSError:
self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
if dl:
yield f
else:
@@ -1788,7 +1805,9 @@ class YoutubeDL(object):
yield f
elif format_spec == 'mergeall':
def selector_function(ctx):
formats = list(_check_formats(ctx['formats']))
formats = ctx['formats']
if check_formats:
formats = list(_check_formats(formats))
if not formats:
return
merged_format = formats[-1]
@@ -1809,14 +1828,16 @@ class YoutubeDL(object):
format_modified = mobj.group('mod') is not None
format_fallback = not format_type and not format_modified # for b, w
filter_f = (
_filter_f = (
(lambda f: f.get('%scodec' % format_type) != 'none')
if format_type and format_modified # bv*, ba*, wv*, wa*
else (lambda f: f.get('%scodec' % not_format_type) == 'none')
if format_type # bv, ba, wv, wa
else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
if not format_modified # b, w
else None) # b*, w*
else lambda f: True) # b*, w*
filter_f = lambda f: _filter_f(f) and (
f.get('vcodec') != 'none' or f.get('acodec') != 'none')
else:
filter_f = ((lambda f: f.get('ext') == format_spec)
if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
@@ -1909,8 +1930,7 @@ class YoutubeDL(object):
self.cookiejar.add_cookie_header(pr)
return pr.get_header('Cookie')
@staticmethod
def _sanitize_thumbnails(info_dict):
def _sanitize_thumbnails(self, info_dict):
thumbnails = info_dict.get('thumbnails')
if thumbnails is None:
thumbnail = info_dict.get('thumbnail')
@@ -1923,12 +1943,25 @@ class YoutubeDL(object):
t.get('height') if t.get('height') is not None else -1,
t.get('id') if t.get('id') is not None else '',
t.get('url')))
def test_thumbnail(t):
self.to_screen('[info] Testing thumbnail %s' % t['id'])
try:
self.urlopen(HEADRequest(t['url']))
except network_exceptions as err:
self.to_screen('[info] Unable to connect to thumbnail %s URL "%s" - %s. Skipping...' % (
t['id'], t['url'], error_to_compat_str(err)))
return False
return True
for i, t in enumerate(thumbnails):
t['url'] = sanitize_url(t['url'])
if t.get('width') and t.get('height'):
t['resolution'] = '%dx%d' % (t['width'], t['height'])
if t.get('id') is None:
t['id'] = '%d' % i
if t.get('width') and t.get('height'):
t['resolution'] = '%dx%d' % (t['width'], t['height'])
t['url'] = sanitize_url(t['url'])
if self.params.get('check_formats'):
info_dict['thumbnails'] = reversed(LazyList(filter(test_thumbnail, thumbnails[::-1])))
def process_video_result(self, info_dict, download=True):
assert info_dict.get('_type', 'video') == 'video'
@@ -2114,12 +2147,11 @@ class YoutubeDL(object):
self.list_formats(info_dict)
return
req_format = self.params.get('format')
if req_format is None:
format_selector = self.format_selector
if format_selector is None:
req_format = self._default_format_spec(info_dict, download=download)
self.write_debug('Default format spec: %s' % req_format)
format_selector = self.build_format_selector(req_format)
format_selector = self.build_format_selector(req_format)
# While in format selection we may need to have an access to the original
# format set in order to calculate some metrics or do some processing.
@@ -2653,65 +2685,53 @@ class YoutubeDL(object):
return
if success and full_filename != '-':
# Fixup content
fixup_policy = self.params.get('fixup')
if fixup_policy is None:
fixup_policy = 'detect_or_warn'
INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg to fix this automatically.'
def fixup():
do_fixup = True
fixup_policy = self.params.get('fixup')
vid = info_dict['id']
stretched_ratio = info_dict.get('stretched_ratio')
if stretched_ratio is not None and stretched_ratio != 1:
if fixup_policy == 'warn':
self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
info_dict['id'], stretched_ratio))
elif fixup_policy == 'detect_or_warn':
stretched_pp = FFmpegFixupStretchedPP(self)
if stretched_pp.available:
info_dict['__postprocessors'].append(stretched_pp)
if fixup_policy in ('ignore', 'never'):
return
elif fixup_policy == 'warn':
do_fixup = False
elif fixup_policy != 'force':
assert fixup_policy in ('detect_or_warn', None)
if not info_dict.get('__real_download'):
do_fixup = False
def ffmpeg_fixup(cndn, msg, cls):
if not cndn:
return
if not do_fixup:
self.report_warning(f'{vid}: {msg}')
return
pp = cls(self)
if pp.available:
info_dict['__postprocessors'].append(pp)
else:
self.report_warning(
'%s: Non-uniform pixel ratio (%s). %s'
% (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE))
else:
assert fixup_policy in ('ignore', 'never')
self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
if (info_dict.get('requested_formats') is None
and info_dict.get('container') == 'm4a_dash'
and info_dict.get('ext') == 'm4a'):
if fixup_policy == 'warn':
self.report_warning(
'%s: writing DASH m4a. '
'Only some players support this container.'
% info_dict['id'])
elif fixup_policy == 'detect_or_warn':
fixup_pp = FFmpegFixupM4aPP(self)
if fixup_pp.available:
info_dict['__postprocessors'].append(fixup_pp)
else:
self.report_warning(
'%s: writing DASH m4a. '
'Only some players support this container. %s'
% (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
else:
assert fixup_policy in ('ignore', 'never')
stretched_ratio = info_dict.get('stretched_ratio')
ffmpeg_fixup(
stretched_ratio not in (1, None),
f'Non-uniform pixel ratio {stretched_ratio}',
FFmpegFixupStretchedPP)
if ('protocol' in info_dict
and get_suitable_downloader(info_dict, self.params).__name__ == 'HlsFD'):
if fixup_policy == 'warn':
self.report_warning('%s: malformed AAC bitstream detected.' % (
info_dict['id']))
elif fixup_policy == 'detect_or_warn':
fixup_pp = FFmpegFixupM3u8PP(self)
if fixup_pp.available:
info_dict['__postprocessors'].append(fixup_pp)
else:
self.report_warning(
'%s: malformed AAC bitstream detected. %s'
% (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
else:
assert fixup_policy in ('ignore', 'never')
ffmpeg_fixup(
(info_dict.get('requested_formats') is None
and info_dict.get('container') == 'm4a_dash'
and info_dict.get('ext') == 'm4a'),
'writing DASH m4a. Only some players support this container',
FFmpegFixupM4aPP)
downloader = (get_suitable_downloader(info_dict, self.params).__name__
if 'protocol' in info_dict else None)
ffmpeg_fixup(downloader == 'HlsFD', 'malformed AAC bitstream detected', FFmpegFixupM3u8PP)
ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP)
ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP)
fixup()
try:
info_dict = self.post_process(dl_filename, info_dict, files_to_move)
except PostProcessingError as err:
@@ -2793,7 +2813,7 @@ class YoutubeDL(object):
info_dict['epoch'] = int(time.time())
reject = lambda k, v: k in remove_keys
filter_fn = lambda obj: (
list(map(filter_fn, obj)) if isinstance(obj, (list, tuple, set))
list(map(filter_fn, obj)) if isinstance(obj, (LazyList, list, tuple, set))
else obj if not isinstance(obj, dict)
else dict((k, filter_fn(v)) for k, v in obj.items() if not reject(k, v)))
return filter_fn(info_dict)
@@ -2904,6 +2924,8 @@ class YoutubeDL(object):
@staticmethod
def format_resolution(format, default='unknown'):
if format.get('vcodec') == 'none':
if format.get('acodec') == 'none':
return 'images'
return 'audio only'
if format.get('resolution') is not None:
return format['resolution']
@@ -3031,7 +3053,7 @@ class YoutubeDL(object):
hideEmpty=new_format)))
def list_thumbnails(self, info_dict):
thumbnails = info_dict.get('thumbnails')
thumbnails = list(info_dict.get('thumbnails'))
if not thumbnails:
self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
return
@@ -3241,6 +3263,7 @@ class YoutubeDL(object):
if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
ret.append(suffix + thumb_ext)
t['filepath'] = thumb_filename
self.to_screen('[%s] %s: Thumbnail %sis already present' %
(info_dict['extractor'], info_dict['id'], thumb_display_id))
else:

View File

@@ -151,6 +151,11 @@ def _real_main(argv=None):
if numeric_limit is None:
parser.error('invalid rate limit specified')
opts.ratelimit = numeric_limit
if opts.throttledratelimit is not None:
numeric_limit = FileDownloader.parse_bytes(opts.throttledratelimit)
if numeric_limit is None:
parser.error('invalid rate limit specified')
opts.throttledratelimit = numeric_limit
if opts.min_filesize is not None:
numeric_limit = FileDownloader.parse_bytes(opts.min_filesize)
if numeric_limit is None:
@@ -268,6 +273,7 @@ def _real_main(argv=None):
'filename', 'format-sort', 'abort-on-error', 'format-spec', 'no-playlist-metafiles',
'multistreams', 'no-live-chat', 'playlist-index', 'list-formats', 'no-direct-merge',
'no-youtube-channel-redirect', 'no-youtube-unavailable-videos', 'no-attach-info-json',
'embed-thumbnail-atomicparsley',
]
compat_opts = parse_compat_opts()
@@ -551,6 +557,7 @@ def _real_main(argv=None):
'ignoreerrors': opts.ignoreerrors,
'force_generic_extractor': opts.force_generic_extractor,
'ratelimit': opts.ratelimit,
'throttledratelimit': opts.throttledratelimit,
'overwrites': opts.overwrites,
'retries': opts.retries,
'fragment_retries': opts.fragment_retries,

View File

@@ -3030,6 +3030,21 @@ except AttributeError:
compat_Match = type(re.compile('').match(''))
import asyncio
try:
compat_asyncio_run = asyncio.run
except AttributeError:
def compat_asyncio_run(coro):
try:
loop = asyncio.get_event_loop()
except RuntimeError:
loop = asyncio.new_event_loop()
asyncio.set_event_loop(loop)
loop.run_until_complete(coro)
asyncio.run = compat_asyncio_run
__all__ = [
'compat_HTMLParseError',
'compat_HTMLParser',
@@ -3037,6 +3052,7 @@ __all__ = [
'compat_Match',
'compat_Pattern',
'compat_Struct',
'compat_asyncio_run',
'compat_b64decode',
'compat_basestring',
'compat_chr',

View File

@@ -22,8 +22,10 @@ from .http import HttpFD
from .rtmp import RtmpFD
from .rtsp import RtspFD
from .ism import IsmFD
from .mhtml import MhtmlFD
from .niconico import NiconicoDmcFD
from .youtube_live_chat import YoutubeLiveChatReplayFD
from .websocket import WebSocketFragmentFD
from .youtube_live_chat import YoutubeLiveChatFD
from .external import (
get_external_downloader,
FFmpegFD,
@@ -39,8 +41,11 @@ PROTOCOL_MAP = {
'f4m': F4mFD,
'http_dash_segments': DashSegmentsFD,
'ism': IsmFD,
'mhtml': MhtmlFD,
'niconico_dmc': NiconicoDmcFD,
'youtube_live_chat_replay': YoutubeLiveChatReplayFD,
'websocket_frag': WebSocketFragmentFD,
'youtube_live_chat': YoutubeLiveChatFD,
'youtube_live_chat_replay': YoutubeLiveChatFD,
}
@@ -50,6 +55,7 @@ def shorten_protocol_name(proto, simplify=False):
'rtmp_ffmpeg': 'rtmp_f',
'http_dash_segments': 'dash',
'niconico_dmc': 'dmc',
'websocket_frag': 'WSfrag',
}
if simplify:
short_protocol_names.update({

View File

@@ -32,6 +32,7 @@ class FileDownloader(object):
verbose: Print additional info to stdout.
quiet: Do not print messages to stdout.
ratelimit: Download speed limit, in bytes/sec.
throttledratelimit: Assume the download is being throttled below this speed (bytes/sec)
retries: Number of times to retry for HTTP error 5xx
buffersize: Size of download buffer in bytes.
noresizebuffer: Do not automatically resize the download buffer.

View File

@@ -1,21 +1,9 @@
from __future__ import unicode_literals
import errno
try:
import concurrent.futures
can_threaded_download = True
except ImportError:
can_threaded_download = False
from ..downloader import _get_real_downloader
from .fragment import FragmentFD
from ..compat import compat_urllib_error
from ..utils import (
DownloadError,
sanitize_open,
urljoin,
)
from ..utils import urljoin
class DashSegmentsFD(FragmentFD):
@@ -43,9 +31,6 @@ class DashSegmentsFD(FragmentFD):
else:
self._prepare_and_start_frag_download(ctx)
fragment_retries = self.params.get('fragment_retries', 0)
skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
fragments_to_download = []
frag_index = 0
for i, fragment in enumerate(fragments):
@@ -76,116 +61,5 @@ class DashSegmentsFD(FragmentFD):
if not success:
return False
else:
def download_fragment(fragment):
i = fragment['index']
frag_index = fragment['frag_index']
fragment_url = fragment['url']
ctx['fragment_index'] = frag_index
# In DASH, the first segment contains necessary headers to
# generate a valid MP4 file, so always abort for the first segment
fatal = i == 0 or not skip_unavailable_fragments
count = 0
while count <= fragment_retries:
try:
success, frag_content = self._download_fragment(ctx, fragment_url, info_dict)
if not success:
return False, frag_index
break
except compat_urllib_error.HTTPError as err:
# YouTube may often return 404 HTTP error for a fragment causing the
# whole download to fail. However if the same fragment is immediately
# retried with the same request data this usually succeeds (1-2 attempts
# is usually enough) thus allowing to download the whole file successfully.
# To be future-proof we will retry all fragments that fail with any
# HTTP error.
count += 1
if count <= fragment_retries:
self.report_retry_fragment(err, frag_index, count, fragment_retries)
except DownloadError:
# Don't retry fragment if error occurred during HTTP downloading
# itself since it has own retry settings
if not fatal:
break
raise
if count > fragment_retries:
if not fatal:
return False, frag_index
ctx['dest_stream'].close()
self.report_error('Giving up after %s fragment retries' % fragment_retries)
return False, frag_index
return frag_content, frag_index
def append_fragment(frag_content, frag_index):
fatal = frag_index == 1 or not skip_unavailable_fragments
if frag_content:
fragment_filename = '%s-Frag%d' % (ctx['tmpfilename'], frag_index)
try:
file, frag_sanitized = sanitize_open(fragment_filename, 'rb')
ctx['fragment_filename_sanitized'] = frag_sanitized
file.close()
self._append_fragment(ctx, frag_content)
return True
except EnvironmentError as ose:
if ose.errno != errno.ENOENT:
raise
# FileNotFoundError
if not fatal:
self.report_skip_fragment(frag_index)
return True
else:
ctx['dest_stream'].close()
self.report_error(
'fragment %s not found, unable to continue' % frag_index)
return False
else:
if not fatal:
self.report_skip_fragment(frag_index)
return True
else:
ctx['dest_stream'].close()
self.report_error(
'fragment %s not found, unable to continue' % frag_index)
return False
max_workers = self.params.get('concurrent_fragment_downloads', 1)
if can_threaded_download and max_workers > 1:
self.report_warning('The download speed shown is only of one thread. This is a known issue')
_download_fragment = lambda f: (f, download_fragment(f)[1])
with concurrent.futures.ThreadPoolExecutor(max_workers) as pool:
futures = [pool.submit(_download_fragment, fragment) for fragment in fragments_to_download]
# timeout must be 0 to return instantly
done, not_done = concurrent.futures.wait(futures, timeout=0)
try:
while not_done:
# Check every 1 second for KeyboardInterrupt
freshly_done, not_done = concurrent.futures.wait(not_done, timeout=1)
done |= freshly_done
except KeyboardInterrupt:
for future in not_done:
future.cancel()
# timeout must be none to cancel
concurrent.futures.wait(not_done, timeout=None)
raise KeyboardInterrupt
for fragment, frag_index in map(lambda x: x.result(), futures):
fragment_filename = '%s-Frag%d' % (ctx['tmpfilename'], frag_index)
down, frag_sanitized = sanitize_open(fragment_filename, 'rb')
fragment['fragment_filename_sanitized'] = frag_sanitized
frag_content = down.read()
down.close()
result = append_fragment(frag_content, frag_index)
if not result:
return False
else:
for fragment in fragments_to_download:
frag_content, frag_index = download_fragment(fragment)
result = append_fragment(frag_content, frag_index)
if not result:
return False
self._finish_frag_download(ctx)
self.download_and_append_fragments(ctx, fragments_to_download, info_dict)
return True

View File

@@ -280,6 +280,8 @@ class Aria2cFD(ExternalFD):
'--file-allocation=none', '-x16', '-j16', '-s16']
if 'fragments' in info_dict:
cmd += ['--allow-overwrite=true', '--allow-piece-length-change=true']
else:
cmd += ['--min-split-size', '1M']
if info_dict.get('http_headers') is not None:
for key, val in info_dict['http_headers'].items():
@@ -345,6 +347,10 @@ class FFmpegFD(ExternalFD):
# TODO: Fix path for ffmpeg
return FFmpegPostProcessor().available
def on_process_started(self, proc, stdin):
""" Override this in subclasses """
pass
def _call_downloader(self, tmpfilename, info_dict):
urls = [f['url'] for f in info_dict.get('requested_formats', [])] or [info_dict['url']]
ffpp = FFmpegPostProcessor(downloader=self)
@@ -472,6 +478,8 @@ class FFmpegFD(ExternalFD):
self._debug_cmd(args)
proc = subprocess.Popen(args, stdin=subprocess.PIPE, env=env)
if url in ('-', 'pipe:'):
self.on_process_started(proc, proc.stdin)
try:
retval = proc.wait()
except BaseException as e:
@@ -480,7 +488,7 @@ class FFmpegFD(ExternalFD):
# produces a file that is playable (this is mostly useful for live
# streams). Note that Windows is not affected and produces playable
# files (see https://github.com/ytdl-org/youtube-dl/issues/8300).
if isinstance(e, KeyboardInterrupt) and sys.platform != 'win32':
if isinstance(e, KeyboardInterrupt) and sys.platform != 'win32' and url not in ('-', 'pipe:'):
process_communicate_or_kill(proc, b'q')
else:
proc.kill()

View File

@@ -4,9 +4,26 @@ import os
import time
import json
try:
from Crypto.Cipher import AES
can_decrypt_frag = True
except ImportError:
can_decrypt_frag = False
try:
import concurrent.futures
can_threaded_download = True
except ImportError:
can_threaded_download = False
from .common import FileDownloader
from .http import HttpFD
from ..compat import (
compat_urllib_error,
compat_struct_pack,
)
from ..utils import (
DownloadError,
error_to_compat_str,
encodeFilename,
sanitize_open,
@@ -56,7 +73,7 @@ class FragmentFD(FileDownloader):
def report_retry_fragment(self, err, frag_index, count, retries):
self.to_screen(
'[download] Got server HTTP error: %s. Retrying fragment %d (attempt %d of %s) ...'
'\r[download] Got server HTTP error: %s. Retrying fragment %d (attempt %d of %s) ...'
% (error_to_compat_str(err), frag_index, count, self.format_retries(retries)))
def report_skip_fragment(self, frag_index):
@@ -112,11 +129,15 @@ class FragmentFD(FileDownloader):
return False, None
if fragment_info_dict.get('filetime'):
ctx['fragment_filetime'] = fragment_info_dict.get('filetime')
down, frag_sanitized = sanitize_open(fragment_filename, 'rb')
ctx['fragment_filename_sanitized'] = fragment_filename
return True, self._read_fragment(ctx)
def _read_fragment(self, ctx):
down, frag_sanitized = sanitize_open(ctx['fragment_filename_sanitized'], 'rb')
ctx['fragment_filename_sanitized'] = frag_sanitized
frag_content = down.read()
down.close()
return True, frag_content
return frag_content
def _append_fragment(self, ctx, frag_content):
try:
@@ -304,3 +325,106 @@ class FragmentFD(FileDownloader):
'tmpfilename': tmpfilename,
'fragment_index': 0,
})
def download_and_append_fragments(self, ctx, fragments, info_dict, pack_func=None):
fragment_retries = self.params.get('fragment_retries', 0)
skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
test = self.params.get('test', False)
if not pack_func:
pack_func = lambda frag_content, _: frag_content
def download_fragment(fragment, ctx):
frag_index = ctx['fragment_index'] = fragment['frag_index']
headers = info_dict.get('http_headers', {})
byte_range = fragment.get('byte_range')
if byte_range:
headers['Range'] = 'bytes=%d-%d' % (byte_range['start'], byte_range['end'] - 1)
# Never skip the first fragment
fatal = (fragment.get('index') or frag_index) == 0 or not skip_unavailable_fragments
count, frag_content = 0, None
while count <= fragment_retries:
try:
success, frag_content = self._download_fragment(ctx, fragment['url'], info_dict, headers)
if not success:
return False, frag_index
break
except compat_urllib_error.HTTPError as err:
# Unavailable (possibly temporary) fragments may be served.
# First we try to retry then either skip or abort.
# See https://github.com/ytdl-org/youtube-dl/issues/10165,
# https://github.com/ytdl-org/youtube-dl/issues/10448).
count += 1
if count <= fragment_retries:
self.report_retry_fragment(err, frag_index, count, fragment_retries)
except DownloadError:
# Don't retry fragment if error occurred during HTTP downloading
# itself since it has own retry settings
if not fatal:
break
raise
if count > fragment_retries:
if not fatal:
return False, frag_index
ctx['dest_stream'].close()
self.report_error('Giving up after %s fragment retries' % fragment_retries)
return False, frag_index
return frag_content, frag_index
def decrypt_fragment(fragment, frag_content):
decrypt_info = fragment.get('decrypt_info')
if not decrypt_info or decrypt_info['METHOD'] != 'AES-128':
return frag_content
iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', fragment['media_sequence'])
decrypt_info['KEY'] = decrypt_info.get('KEY') or self.ydl.urlopen(
self._prepare_url(info_dict, info_dict.get('_decryption_key_url') or decrypt_info['URI'])).read()
# Don't decrypt the content in tests since the data is explicitly truncated and it's not to a valid block
# size (see https://github.com/ytdl-org/youtube-dl/pull/27660). Tests only care that the correct data downloaded,
# not what it decrypts to.
if test:
return frag_content
return AES.new(decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
def append_fragment(frag_content, frag_index, ctx):
if not frag_content:
fatal = frag_index == 1 or not skip_unavailable_fragments
if not fatal:
self.report_skip_fragment(frag_index)
return True
else:
ctx['dest_stream'].close()
self.report_error(
'fragment %s not found, unable to continue' % frag_index)
return False
self._append_fragment(ctx, pack_func(frag_content, frag_index))
return True
max_workers = self.params.get('concurrent_fragment_downloads', 1)
if can_threaded_download and max_workers > 1:
def _download_fragment(fragment):
try:
ctx_copy = ctx.copy()
frag_content, frag_index = download_fragment(fragment, ctx_copy)
return fragment, frag_content, frag_index, ctx_copy.get('fragment_filename_sanitized')
except Exception:
# Return immediately on exception so that it is raised in the main thread
return
self.report_warning('The download speed shown is only of one thread. This is a known issue and patches are welcome')
with concurrent.futures.ThreadPoolExecutor(max_workers) as pool:
for fragment, frag_content, frag_index, frag_filename in pool.map(_download_fragment, fragments):
ctx['fragment_filename_sanitized'] = frag_filename
ctx['fragment_index'] = frag_index
result = append_fragment(decrypt_fragment(fragment, frag_content), frag_index, ctx)
if not result:
return False
else:
for fragment in fragments:
frag_content, frag_index = download_fragment(fragment, ctx)
result = append_fragment(decrypt_fragment(fragment, frag_content), frag_index, ctx)
if not result:
return False
self._finish_frag_download(ctx)

View File

@@ -1,32 +1,18 @@
from __future__ import unicode_literals
import errno
import re
import io
import binascii
try:
from Crypto.Cipher import AES
can_decrypt_frag = True
except ImportError:
can_decrypt_frag = False
try:
import concurrent.futures
can_threaded_download = True
except ImportError:
can_threaded_download = False
from ..downloader import _get_real_downloader
from .fragment import FragmentFD
from .fragment import FragmentFD, can_decrypt_frag
from .external import FFmpegFD
from ..compat import (
compat_urllib_error,
compat_urlparse,
compat_struct_pack,
)
from ..utils import (
parse_m3u8_attributes,
sanitize_open,
update_url_query,
bug_reports_message,
)
@@ -151,10 +137,6 @@ class HlsFD(FragmentFD):
extra_state = ctx.setdefault('extra_state', {})
fragment_retries = self.params.get('fragment_retries', 0)
skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
test = self.params.get('test', False)
format_index = info_dict.get('format_index')
extra_query = None
extra_param_to_segment_url = info_dict.get('extra_param_to_segment_url')
@@ -258,7 +240,7 @@ class HlsFD(FragmentFD):
media_sequence += 1
# We only download the first fragment during the test
if test:
if self.params.get('test', False):
fragments = [fragments[0] if fragments else None]
if real_downloader:
@@ -272,55 +254,6 @@ class HlsFD(FragmentFD):
if not success:
return False
else:
def decrypt_fragment(fragment, frag_content):
decrypt_info = fragment['decrypt_info']
if decrypt_info['METHOD'] != 'AES-128':
return frag_content
iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', fragment['media_sequence'])
decrypt_info['KEY'] = decrypt_info.get('KEY') or self.ydl.urlopen(
self._prepare_url(info_dict, info_dict.get('_decryption_key_url') or decrypt_info['URI'])).read()
# Don't decrypt the content in tests since the data is explicitly truncated and it's not to a valid block
# size (see https://github.com/ytdl-org/youtube-dl/pull/27660). Tests only care that the correct data downloaded,
# not what it decrypts to.
if test:
return frag_content
return AES.new(decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
def download_fragment(fragment):
frag_index = fragment['frag_index']
frag_url = fragment['url']
byte_range = fragment['byte_range']
ctx['fragment_index'] = frag_index
count = 0
headers = info_dict.get('http_headers', {})
if byte_range:
headers['Range'] = 'bytes=%d-%d' % (byte_range['start'], byte_range['end'] - 1)
while count <= fragment_retries:
try:
success, frag_content = self._download_fragment(
ctx, frag_url, info_dict, headers)
if not success:
return False, frag_index
break
except compat_urllib_error.HTTPError as err:
# Unavailable (possibly temporary) fragments may be served.
# First we try to retry then either skip or abort.
# See https://github.com/ytdl-org/youtube-dl/issues/10165,
# https://github.com/ytdl-org/youtube-dl/issues/10448).
count += 1
if count <= fragment_retries:
self.report_retry_fragment(err, frag_index, count, fragment_retries)
if count > fragment_retries:
ctx['dest_stream'].close()
self.report_error('Giving up after %s fragment retries' % fragment_retries)
return False, frag_index
return decrypt_fragment(fragment, frag_content), frag_index
pack_fragment = lambda frag_content, _: frag_content
if is_webvtt:
def pack_fragment(frag_content, frag_index):
output = io.StringIO()
@@ -388,75 +321,7 @@ class HlsFD(FragmentFD):
block.write_into(output)
return output.getvalue().encode('utf-8')
def append_fragment(frag_content, frag_index):
fatal = frag_index == 1 or not skip_unavailable_fragments
if frag_content:
fragment_filename = '%s-Frag%d' % (ctx['tmpfilename'], frag_index)
try:
file, frag_sanitized = sanitize_open(fragment_filename, 'rb')
ctx['fragment_filename_sanitized'] = frag_sanitized
file.close()
frag_content = pack_fragment(frag_content, frag_index)
self._append_fragment(ctx, frag_content)
return True
except EnvironmentError as ose:
if ose.errno != errno.ENOENT:
raise
# FileNotFoundError
if not fatal:
self.report_skip_fragment(frag_index)
return True
else:
ctx['dest_stream'].close()
self.report_error(
'fragment %s not found, unable to continue' % frag_index)
return False
else:
if not fatal:
self.report_skip_fragment(frag_index)
return True
else:
ctx['dest_stream'].close()
self.report_error(
'fragment %s not found, unable to continue' % frag_index)
return False
max_workers = self.params.get('concurrent_fragment_downloads', 1)
if can_threaded_download and max_workers > 1:
self.report_warning('The download speed shown is only of one thread. This is a known issue')
_download_fragment = lambda f: (f, download_fragment(f)[1])
with concurrent.futures.ThreadPoolExecutor(max_workers) as pool:
futures = [pool.submit(_download_fragment, fragment) for fragment in fragments]
# timeout must be 0 to return instantly
done, not_done = concurrent.futures.wait(futures, timeout=0)
try:
while not_done:
# Check every 1 second for KeyboardInterrupt
freshly_done, not_done = concurrent.futures.wait(not_done, timeout=1)
done |= freshly_done
except KeyboardInterrupt:
for future in not_done:
future.cancel()
# timeout must be none to cancel
concurrent.futures.wait(not_done, timeout=None)
raise KeyboardInterrupt
for fragment, frag_index in map(lambda x: x.result(), futures):
fragment_filename = '%s-Frag%d' % (ctx['tmpfilename'], frag_index)
down, frag_sanitized = sanitize_open(fragment_filename, 'rb')
fragment['fragment_filename_sanitized'] = frag_sanitized
frag_content = down.read()
down.close()
result = append_fragment(decrypt_fragment(fragment, frag_content), frag_index)
if not result:
return False
else:
for fragment in fragments:
frag_content, frag_index = download_fragment(fragment)
result = append_fragment(frag_content, frag_index)
if not result:
return False
self._finish_frag_download(ctx)
pack_fragment = None
self.download_and_append_fragments(ctx, fragments, info_dict, pack_fragment)
return True

View File

@@ -18,6 +18,7 @@ from ..utils import (
int_or_none,
sanitize_open,
sanitized_Request,
ThrottledDownload,
write_xattr,
XAttrMetadataError,
XAttrUnavailableError,
@@ -223,6 +224,7 @@ class HttpFD(FileDownloader):
# measure time over whole while-loop, so slow_down() and best_block_size() work together properly
now = None # needed for slow_down() in the first loop run
before = start # start measuring
throttle_start = None
def retry(e):
to_stdout = ctx.tmpfilename == '-'
@@ -313,6 +315,18 @@ class HttpFD(FileDownloader):
if data_len is not None and byte_counter == data_len:
break
if speed and speed < (self.params.get('throttledratelimit') or 0):
# The speed must stay below the limit for 3 seconds
# This prevents raising error when the speed temporarily goes down
if throttle_start is None:
throttle_start = now
elif now - throttle_start > 3:
if ctx.stream is not None and ctx.tmpfilename != '-':
ctx.stream.close()
raise ThrottledDownload()
else:
throttle_start = None
if not is_test and ctx.chunk_size and ctx.data_len is not None and byte_counter < ctx.data_len:
ctx.resume_len = byte_counter
# ctx.block_size = block_size

202
yt_dlp/downloader/mhtml.py Normal file
View File

@@ -0,0 +1,202 @@
# coding: utf-8
from __future__ import unicode_literals
import io
import quopri
import re
import uuid
from .fragment import FragmentFD
from ..utils import (
escapeHTML,
formatSeconds,
srt_subtitles_timecode,
urljoin,
)
from ..version import __version__ as YT_DLP_VERSION
class MhtmlFD(FragmentFD):
FD_NAME = 'mhtml'
_STYLESHEET = """\
html, body {
margin: 0;
padding: 0;
height: 100vh;
}
html {
overflow-y: scroll;
scroll-snap-type: y mandatory;
}
body {
scroll-snap-type: y mandatory;
display: flex;
flex-flow: column;
}
body > figure {
max-width: 100vw;
max-height: 100vh;
scroll-snap-align: center;
}
body > figure > figcaption {
text-align: center;
height: 2.5em;
}
body > figure > img {
display: block;
margin: auto;
max-width: 100%;
max-height: calc(100vh - 5em);
}
"""
_STYLESHEET = re.sub(r'\s+', ' ', _STYLESHEET)
_STYLESHEET = re.sub(r'\B \B|(?<=[\w\-]) (?=[^\w\-])|(?<=[^\w\-]) (?=[\w\-])', '', _STYLESHEET)
@staticmethod
def _escape_mime(s):
return '=?utf-8?Q?' + (b''.join(
bytes((b,)) if b >= 0x20 else b'=%02X' % b
for b in quopri.encodestring(s.encode('utf-8'), header=True)
)).decode('us-ascii') + '?='
def _gen_cid(self, i, fragment, frag_boundary):
return '%u.%s@yt-dlp.github.io.invalid' % (i, frag_boundary)
def _gen_stub(self, *, fragments, frag_boundary, title):
output = io.StringIO()
output.write((
'<!DOCTYPE html>'
'<html>'
'<head>'
'' '<meta name="generator" content="yt-dlp {version}">'
'' '<title>{title}</title>'
'' '<style>{styles}</style>'
'<body>'
).format(
version=escapeHTML(YT_DLP_VERSION),
styles=self._STYLESHEET,
title=escapeHTML(title)
))
t0 = 0
for i, frag in enumerate(fragments):
output.write('<figure>')
try:
t1 = t0 + frag['duration']
output.write((
'<figcaption>Slide #{num}: {t0} {t1} (duration: {duration})</figcaption>'
).format(
num=i + 1,
t0=srt_subtitles_timecode(t0),
t1=srt_subtitles_timecode(t1),
duration=formatSeconds(frag['duration'], msec=True)
))
except (KeyError, ValueError, TypeError):
t1 = None
output.write((
'<figcaption>Slide #{num}</figcaption>'
).format(num=i + 1))
output.write('<img src="cid:{cid}">'.format(
cid=self._gen_cid(i, frag, frag_boundary)))
output.write('</figure>')
t0 = t1
return output.getvalue()
def real_download(self, filename, info_dict):
fragment_base_url = info_dict.get('fragment_base_url')
fragments = info_dict['fragments'][:1] if self.params.get(
'test', False) else info_dict['fragments']
title = info_dict['title']
origin = info_dict['webpage_url']
ctx = {
'filename': filename,
'total_frags': len(fragments),
}
self._prepare_and_start_frag_download(ctx)
extra_state = ctx.setdefault('extra_state', {
'header_written': False,
'mime_boundary': str(uuid.uuid4()).replace('-', ''),
})
frag_boundary = extra_state['mime_boundary']
if not extra_state['header_written']:
stub = self._gen_stub(
fragments=fragments,
frag_boundary=frag_boundary,
title=title
)
ctx['dest_stream'].write((
'MIME-Version: 1.0\r\n'
'From: <nowhere@yt-dlp.github.io.invalid>\r\n'
'To: <nowhere@yt-dlp.github.io.invalid>\r\n'
'Subject: {title}\r\n'
'Content-type: multipart/related; '
'' 'boundary="{boundary}"; '
'' 'type="text/html"\r\n'
'X.yt-dlp.Origin: {origin}\r\n'
'\r\n'
'--{boundary}\r\n'
'Content-Type: text/html; charset=utf-8\r\n'
'Content-Length: {length}\r\n'
'\r\n'
'{stub}\r\n'
).format(
origin=origin,
boundary=frag_boundary,
length=len(stub),
title=self._escape_mime(title),
stub=stub
).encode('utf-8'))
extra_state['header_written'] = True
for i, fragment in enumerate(fragments):
if (i + 1) <= ctx['fragment_index']:
continue
fragment_url = urljoin(fragment_base_url, fragment['path'])
success, frag_content = self._download_fragment(ctx, fragment_url, info_dict)
if not success:
continue
mime_type = b'image/jpeg'
if frag_content.startswith(b'\x89PNG\r\n\x1a\n'):
mime_type = b'image/png'
if frag_content.startswith((b'GIF87a', b'GIF89a')):
mime_type = b'image/gif'
if frag_content.startswith(b'RIFF') and frag_content[8:12] == 'WEBP':
mime_type = b'image/webp'
frag_header = io.BytesIO()
frag_header.write(
b'--%b\r\n' % frag_boundary.encode('us-ascii'))
frag_header.write(
b'Content-ID: <%b>\r\n' % self._gen_cid(i, fragment, frag_boundary).encode('us-ascii'))
frag_header.write(
b'Content-type: %b\r\n' % mime_type)
frag_header.write(
b'Content-length: %u\r\n' % len(frag_content))
frag_header.write(
b'Content-location: %b\r\n' % fragment_url.encode('us-ascii'))
frag_header.write(
b'X.yt-dlp.Duration: %f\r\n' % fragment['duration'])
frag_header.write(b'\r\n')
self._append_fragment(
ctx, frag_header.getvalue() + frag_content + b'\r\n')
ctx['dest_stream'].write(
b'--%b--\r\n\r\n' % frag_boundary.encode('us-ascii'))
self._finish_frag_download(ctx)
return True

View File

@@ -0,0 +1,59 @@
import os
import signal
import asyncio
import threading
try:
import websockets
has_websockets = True
except ImportError:
has_websockets = False
from .common import FileDownloader
from .external import FFmpegFD
class FFmpegSinkFD(FileDownloader):
""" A sink to ffmpeg for downloading fragments in any form """
def real_download(self, filename, info_dict):
info_copy = info_dict.copy()
info_copy['url'] = '-'
async def call_conn(proc, stdin):
try:
await self.real_connection(stdin, info_dict)
except (BrokenPipeError, OSError):
pass
finally:
try:
stdin.flush()
stdin.close()
except OSError:
pass
os.kill(os.getpid(), signal.SIGINT)
class FFmpegStdinFD(FFmpegFD):
@classmethod
def get_basename(cls):
return FFmpegFD.get_basename()
def on_process_started(self, proc, stdin):
thread = threading.Thread(target=asyncio.run, daemon=True, args=(call_conn(proc, stdin), ))
thread.start()
return FFmpegStdinFD(self.ydl, self.params or {}).download(filename, info_copy)
async def real_connection(self, sink, info_dict):
""" Override this in subclasses """
raise NotImplementedError('This method must be implemented by subclasses')
class WebSocketFragmentFD(FFmpegSinkFD):
async def real_connection(self, sink, info_dict):
async with websockets.connect(info_dict['url'], extra_headers=info_dict.get('http_headers', {})) as ws:
while True:
recv = await ws.recv()
if isinstance(recv, str):
recv = recv.encode('utf8')
sink.write(recv)

View File

@@ -1,20 +1,23 @@
from __future__ import division, unicode_literals
import json
import time
from .fragment import FragmentFD
from ..compat import compat_urllib_error
from ..utils import (
try_get,
dict_get,
int_or_none,
RegexNotFoundError,
)
from ..extractor.youtube import YoutubeBaseInfoExtractor as YT_BaseIE
class YoutubeLiveChatReplayFD(FragmentFD):
""" Downloads YouTube live chat replays fragment by fragment """
class YoutubeLiveChatFD(FragmentFD):
""" Downloads YouTube live chats fragment by fragment """
FD_NAME = 'youtube_live_chat_replay'
FD_NAME = 'youtube_live_chat'
def real_download(self, filename, info_dict):
video_id = info_dict['video_id']
@@ -31,6 +34,8 @@ class YoutubeLiveChatReplayFD(FragmentFD):
ie = YT_BaseIE(self.ydl)
start_time = int(time.time() * 1000)
def dl_fragment(url, data=None, headers=None):
http_headers = info_dict.get('http_headers', {})
if headers:
@@ -38,36 +43,70 @@ class YoutubeLiveChatReplayFD(FragmentFD):
http_headers.update(headers)
return self._download_fragment(ctx, url, info_dict, http_headers, data)
def download_and_parse_fragment(url, frag_index, request_data):
def parse_actions_replay(live_chat_continuation):
offset = continuation_id = None
processed_fragment = bytearray()
for action in live_chat_continuation.get('actions', []):
if 'replayChatItemAction' in action:
replay_chat_item_action = action['replayChatItemAction']
offset = int(replay_chat_item_action['videoOffsetTimeMsec'])
processed_fragment.extend(
json.dumps(action, ensure_ascii=False).encode('utf-8') + b'\n')
if offset is not None:
continuation_id = try_get(
live_chat_continuation,
lambda x: x['continuations'][0]['liveChatReplayContinuationData']['continuation'])
self._append_fragment(ctx, processed_fragment)
return continuation_id, offset
live_offset = 0
def parse_actions_live(live_chat_continuation):
nonlocal live_offset
continuation_id = None
processed_fragment = bytearray()
for action in live_chat_continuation.get('actions', []):
timestamp = self.parse_live_timestamp(action)
if timestamp is not None:
live_offset = timestamp - start_time
# compatibility with replay format
pseudo_action = {
'replayChatItemAction': {'actions': [action]},
'videoOffsetTimeMsec': str(live_offset),
'isLive': True,
}
processed_fragment.extend(
json.dumps(pseudo_action, ensure_ascii=False).encode('utf-8') + b'\n')
continuation_data_getters = [
lambda x: x['continuations'][0]['invalidationContinuationData'],
lambda x: x['continuations'][0]['timedContinuationData'],
]
continuation_data = try_get(live_chat_continuation, continuation_data_getters, dict)
if continuation_data:
continuation_id = continuation_data.get('continuation')
timeout_ms = int_or_none(continuation_data.get('timeoutMs'))
if timeout_ms is not None:
time.sleep(timeout_ms / 1000)
self._append_fragment(ctx, processed_fragment)
return continuation_id, live_offset
if info_dict['protocol'] == 'youtube_live_chat_replay':
parse_actions = parse_actions_replay
elif info_dict['protocol'] == 'youtube_live_chat':
parse_actions = parse_actions_live
def download_and_parse_fragment(url, frag_index, request_data, headers):
count = 0
while count <= fragment_retries:
try:
success, raw_fragment = dl_fragment(url, request_data, {'content-type': 'application/json'})
success, raw_fragment = dl_fragment(url, request_data, headers)
if not success:
return False, None, None
try:
data = ie._extract_yt_initial_data(video_id, raw_fragment.decode('utf-8', 'replace'))
except RegexNotFoundError:
data = None
if not data:
data = json.loads(raw_fragment)
data = json.loads(raw_fragment)
live_chat_continuation = try_get(
data,
lambda x: x['continuationContents']['liveChatContinuation'], dict) or {}
offset = continuation_id = None
processed_fragment = bytearray()
for action in live_chat_continuation.get('actions', []):
if 'replayChatItemAction' in action:
replay_chat_item_action = action['replayChatItemAction']
offset = int(replay_chat_item_action['videoOffsetTimeMsec'])
processed_fragment.extend(
json.dumps(action, ensure_ascii=False).encode('utf-8') + b'\n')
if offset is not None:
continuation_id = try_get(
live_chat_continuation,
lambda x: x['continuations'][0]['liveChatReplayContinuationData']['continuation'])
self._append_fragment(ctx, processed_fragment)
continuation_id, offset = parse_actions(live_chat_continuation)
return True, continuation_id, offset
except compat_urllib_error.HTTPError as err:
count += 1
@@ -100,7 +139,11 @@ class YoutubeLiveChatReplayFD(FragmentFD):
innertube_context = try_get(ytcfg, lambda x: x['INNERTUBE_CONTEXT'])
if not api_key or not innertube_context:
return False
url = 'https://www.youtube.com/youtubei/v1/live_chat/get_live_chat_replay?key=' + api_key
visitor_data = try_get(innertube_context, lambda x: x['client']['visitorData'], str)
if info_dict['protocol'] == 'youtube_live_chat_replay':
url = 'https://www.youtube.com/youtubei/v1/live_chat/get_live_chat_replay?key=' + api_key
elif info_dict['protocol'] == 'youtube_live_chat':
url = 'https://www.youtube.com/youtubei/v1/live_chat/get_live_chat?key=' + api_key
frag_index = offset = 0
while continuation_id is not None:
@@ -111,8 +154,11 @@ class YoutubeLiveChatReplayFD(FragmentFD):
}
if frag_index > 1:
request_data['currentPlayerState'] = {'playerOffsetMs': str(max(offset - 5000, 0))}
headers = ie._generate_api_headers(ytcfg, visitor_data=visitor_data)
headers.update({'content-type': 'application/json'})
fragment_request_data = json.dumps(request_data, ensure_ascii=False).encode('utf-8') + b'\n'
success, continuation_id, offset = download_and_parse_fragment(
url, frag_index, json.dumps(request_data, ensure_ascii=False).encode('utf-8') + b'\n')
url, frag_index, fragment_request_data, headers)
if not success:
return False
if test:
@@ -120,3 +166,39 @@ class YoutubeLiveChatReplayFD(FragmentFD):
self._finish_frag_download(ctx)
return True
@staticmethod
def parse_live_timestamp(action):
action_content = dict_get(
action,
['addChatItemAction', 'addLiveChatTickerItemAction', 'addBannerToLiveChatCommand'])
if not isinstance(action_content, dict):
return None
item = dict_get(action_content, ['item', 'bannerRenderer'])
if not isinstance(item, dict):
return None
renderer = dict_get(item, [
# text
'liveChatTextMessageRenderer', 'liveChatPaidMessageRenderer',
'liveChatMembershipItemRenderer', 'liveChatPaidStickerRenderer',
# ticker
'liveChatTickerPaidMessageItemRenderer',
'liveChatTickerSponsorItemRenderer',
# banner
'liveChatBannerRenderer',
])
if not isinstance(renderer, dict):
return None
parent_item_getters = [
lambda x: x['showItemEndpoint']['showLiveChatItemEndpoint']['renderer'],
lambda x: x['contents'],
]
parent_item = try_get(renderer, parent_item_getters, dict)
if parent_item:
renderer = dict_get(parent_item, [
'liveChatTextMessageRenderer', 'liveChatPaidMessageRenderer',
'liveChatMembershipItemRenderer', 'liveChatPaidStickerRenderer',
])
if not isinstance(renderer, dict):
return None
return int_or_none(renderer.get('timestampUsec'), 1000)

View File

@@ -9,10 +9,10 @@ from ..utils import (
class AppleConnectIE(InfoExtractor):
_VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/idsa\.(?P<id>[\w-]+)'
_TEST = {
_VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/(?:id)?sa\.(?P<id>[\w-]+)'
_TESTS = [{
'url': 'https://itunes.apple.com/us/post/idsa.4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
'md5': 'e7c38568a01ea45402570e6029206723',
'md5': 'c1d41f72c8bcaf222e089434619316e4',
'info_dict': {
'id': '4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
'ext': 'm4v',
@@ -22,7 +22,10 @@ class AppleConnectIE(InfoExtractor):
'upload_date': '20150710',
'timestamp': 1436545535,
},
}
}, {
'url': 'https://itunes.apple.com/us/post/sa.0fe0229f-2457-11e5-9f40-1bb645f2d5d9',
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
@@ -36,7 +39,7 @@ class AppleConnectIE(InfoExtractor):
video_data = self._parse_json(video_json, video_id)
timestamp = str_to_int(self._html_search_regex(r'data-timestamp="(\d+)"', webpage, 'timestamp'))
like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count'))
like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count', default=None))
return {
'id': video_id,

View File

@@ -281,7 +281,7 @@ class BiliBiliIE(InfoExtractor):
webpage)
if uploader_mobj:
info.update({
'uploader': uploader_mobj.group('name'),
'uploader': uploader_mobj.group('name').strip(),
'uploader_id': uploader_mobj.group('id'),
})

View File

@@ -24,7 +24,7 @@ class CanvasIE(InfoExtractor):
_VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?P<site_id>canvas|een|ketnet|vrt(?:video|nieuws)|sporza|dako)/assets/(?P<id>[^/?#&]+)'
_TESTS = [{
'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
'md5': '68993eda72ef62386a15ea2cf3c93107',
'md5': '37b2b7bb9b3dcaa05b67058dc3a714a9',
'info_dict': {
'id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
'display_id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
@@ -32,9 +32,9 @@ class CanvasIE(InfoExtractor):
'title': 'Nachtwacht: De Greystook',
'description': 'Nachtwacht: De Greystook',
'thumbnail': r're:^https?://.*\.jpg$',
'duration': 1468.04,
'duration': 1468.02,
},
'expected_warnings': ['is not a supported codec', 'Unknown MIME type'],
'expected_warnings': ['is not a supported codec'],
}, {
'url': 'https://mediazone.vrt.be/api/v1/canvas/assets/mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
'only_matching': True,

View File

@@ -290,6 +290,7 @@ class InfoExtractor(object):
categories: A list of categories that the video falls in, for example
["Sports", "Berlin"]
tags: A list of tags assigned to the video, e.g. ["sweden", "pop music"]
cast: A list of the video cast
is_live: True, False, or None (=unknown). Whether this video is a
live stream that goes on instead of a fixed-length video.
was_live: True, False, or None (=unknown). Whether this video was
@@ -1473,7 +1474,7 @@ class InfoExtractor(object):
class FormatSort:
regex = r' *((?P<reverse>\+)?(?P<field>[a-zA-Z0-9_]+)((?P<separator>[~:])(?P<limit>.*?))?)? *$'
default = ('hidden', 'hasvid', 'ie_pref', 'lang', 'quality',
default = ('hidden', 'aud_or_vid', 'hasvid', 'ie_pref', 'lang', 'quality',
'res', 'fps', 'codec:vp9.2', 'size', 'br', 'asr',
'proto', 'ext', 'hasaud', 'source', 'format_id') # These must not be aliases
ytdl_default = ('hasaud', 'quality', 'tbr', 'filesize', 'vbr',
@@ -1486,7 +1487,7 @@ class InfoExtractor(object):
'acodec': {'type': 'ordered', 'regex': True,
'order': ['opus', 'vorbis', 'aac', 'mp?4a?', 'mp3', 'e?a?c-?3', 'dts', '', None, 'none']},
'proto': {'type': 'ordered', 'regex': True, 'field': 'protocol',
'order': ['(ht|f)tps', '(ht|f)tp$', 'm3u8.+', 'm3u8', '.*dash', '', 'mms|rtsp', 'none', 'f4']},
'order': ['(ht|f)tps', '(ht|f)tp$', 'm3u8.+', '.*dash', 'ws|websocket', '', 'mms|rtsp', 'none', 'f4']},
'vext': {'type': 'ordered', 'field': 'video_ext',
'order': ('mp4', 'webm', 'flv', '', 'none'),
'order_free': ('webm', 'mp4', 'flv', '', 'none')},
@@ -1494,6 +1495,9 @@ class InfoExtractor(object):
'order': ('m4a', 'aac', 'mp3', 'ogg', 'opus', 'webm', '', 'none'),
'order_free': ('opus', 'ogg', 'webm', 'm4a', 'mp3', 'aac', '', 'none')},
'hidden': {'visible': False, 'forced': True, 'type': 'extractor', 'max': -1000},
'aud_or_vid': {'visible': False, 'forced': True, 'type': 'multiple', 'default': 1,
'field': ('vcodec', 'acodec'),
'function': lambda it: int(any(v != 'none' for v in it))},
'ie_pref': {'priority': True, 'type': 'extractor'},
'hasvid': {'priority': True, 'field': 'vcodec', 'type': 'boolean', 'not_in_list': ('none',)},
'hasaud': {'field': 'acodec', 'type': 'boolean', 'not_in_list': ('none',)},
@@ -1701,9 +1705,7 @@ class InfoExtractor(object):
def wrapped_function(values):
values = tuple(filter(lambda x: x is not None, values))
return (self._get_field_setting(field, 'function')(*values) if len(values) > 1
else values[0] if values
else None)
return self._get_field_setting(field, 'function')(values) if values else None
value = wrapped_function((get_value(f) for f in actual_fields))
else:
@@ -1719,7 +1721,7 @@ class InfoExtractor(object):
if not format.get('ext') and 'url' in format:
format['ext'] = determine_ext(format['url'])
if format.get('vcodec') == 'none':
format['audio_ext'] = format['ext']
format['audio_ext'] = format['ext'] if format.get('acodec') != 'none' else 'none'
format['video_ext'] = 'none'
else:
format['video_ext'] = format['ext']
@@ -2125,6 +2127,7 @@ class InfoExtractor(object):
format_id.append(str(format_index))
f = {
'format_id': '-'.join(format_id),
'format_note': name,
'format_index': format_index,
'url': manifest_url,
'manifest_url': m3u8_url,
@@ -2636,7 +2639,7 @@ class InfoExtractor(object):
mime_type = representation_attrib['mimeType']
content_type = representation_attrib.get('contentType', mime_type.split('/')[0])
if content_type in ('video', 'audio', 'text'):
if content_type in ('video', 'audio', 'text') or mime_type == 'image/jpeg':
base_url = ''
for element in (representation, adaptation_set, period, mpd_doc):
base_url_e = element.find(_add_ns('BaseURL'))
@@ -2653,9 +2656,15 @@ class InfoExtractor(object):
url_el = representation.find(_add_ns('BaseURL'))
filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el is not None else None)
bandwidth = int_or_none(representation_attrib.get('bandwidth'))
if representation_id is not None:
format_id = representation_id
else:
format_id = content_type
if mpd_id:
format_id = mpd_id + '-' + format_id
if content_type in ('video', 'audio'):
f = {
'format_id': '%s-%s' % (mpd_id, representation_id) if mpd_id else representation_id,
'format_id': format_id,
'manifest_url': mpd_url,
'ext': mimetype2ext(mime_type),
'width': int_or_none(representation_attrib.get('width')),
@@ -2675,6 +2684,17 @@ class InfoExtractor(object):
'manifest_url': mpd_url,
'filesize': filesize,
}
elif mime_type == 'image/jpeg':
# See test case in VikiIE
# https://www.viki.com/videos/1175236v-choosing-spouse-by-lottery-episode-1
f = {
'format_id': format_id,
'ext': 'mhtml',
'manifest_url': mpd_url,
'format_note': 'DASH storyboards (jpeg)',
'acodec': 'none',
'vcodec': 'none',
}
representation_ms_info = extract_multisegment_info(representation, adaption_set_ms_info)
def prepare_template(template_name, identifiers):
@@ -2693,7 +2713,8 @@ class InfoExtractor(object):
t += c
# Next, $...$ templates are translated to their
# %(...) counterparts to be used with % operator
t = t.replace('$RepresentationID$', representation_id)
if representation_id is not None:
t = t.replace('$RepresentationID$', representation_id)
t = re.sub(r'\$(%s)\$' % '|'.join(identifiers), r'%(\1)d', t)
t = re.sub(r'\$(%s)%%([^$]+)\$' % '|'.join(identifiers), r'%(\1)\2', t)
t.replace('$$', '$')
@@ -2810,7 +2831,7 @@ class InfoExtractor(object):
'url': mpd_url or base_url,
'fragment_base_url': base_url,
'fragments': [],
'protocol': 'http_dash_segments',
'protocol': 'http_dash_segments' if mime_type != 'image/jpeg' else 'mhtml',
})
if 'initialization_url' in representation_ms_info:
initialization_url = representation_ms_info['initialization_url']
@@ -2821,7 +2842,7 @@ class InfoExtractor(object):
else:
# Assuming direct URL to unfragmented media.
f['url'] = base_url
if content_type in ('video', 'audio'):
if content_type in ('video', 'audio') or mime_type == 'image/jpeg':
formats.append(f)
elif content_type == 'text':
subtitles.setdefault(lang or 'und', []).append(f)

View File

@@ -143,9 +143,9 @@ class CuriosityStreamIE(CuriosityStreamBaseIE):
}
class CuriosityStreamCollectionsIE(CuriosityStreamBaseIE):
IE_NAME = 'curiositystream:collections'
_VALID_URL = r'https?://(?:app\.)?curiositystream\.com/collections/(?P<id>\d+)'
class CuriosityStreamCollectionIE(CuriosityStreamBaseIE):
IE_NAME = 'curiositystream:collection'
_VALID_URL = r'https?://(?:app\.)?curiositystream\.com/(?:collections?|series)/(?P<id>\d+)'
_API_BASE_URL = 'https://api.curiositystream.com/v2/collections/'
_TESTS = [{
'url': 'https://curiositystream.com/collections/86',
@@ -155,6 +155,20 @@ class CuriosityStreamCollectionsIE(CuriosityStreamBaseIE):
'description': 'Wondering where to start? Here are a few of our favorite series and films... from our couch to yours.',
},
'playlist_mincount': 7,
}, {
'url': 'https://app.curiositystream.com/collection/2',
'info_dict': {
'id': '2',
'title': 'Curious Minds: The Internet',
'description': 'How is the internet shaping our lives in the 21st Century?',
},
'playlist_mincount': 16,
}, {
'url': 'https://curiositystream.com/series/2',
'only_matching': True,
}, {
'url': 'https://curiositystream.com/collections/36',
'only_matching': True,
}]
def _real_extract(self, url):
@@ -163,25 +177,10 @@ class CuriosityStreamCollectionsIE(CuriosityStreamBaseIE):
entries = []
for media in collection.get('media', []):
media_id = compat_str(media.get('id'))
media_type, ie = ('series', CuriosityStreamSeriesIE) if media.get('is_collection') else ('video', CuriosityStreamIE)
media_type, ie = ('series', CuriosityStreamCollectionIE) if media.get('is_collection') else ('video', CuriosityStreamIE)
entries.append(self.url_result(
'https://curiositystream.com/%s/%s' % (media_type, media_id),
ie=ie.ie_key(), video_id=media_id))
return self.playlist_result(
entries, collection_id,
collection.get('title'), collection.get('description'))
class CuriosityStreamSeriesIE(CuriosityStreamCollectionsIE):
IE_NAME = 'curiositystream:series'
_VALID_URL = r'https?://(?:app\.)?curiositystream\.com/series/(?P<id>\d+)'
_API_BASE_URL = 'https://api.curiositystream.com/v2/series/'
_TESTS = [{
'url': 'https://app.curiositystream.com/series/2',
'info_dict': {
'id': '2',
'title': 'Curious Minds: The Internet',
'description': 'How is the internet shaping our lives in the 21st Century?',
},
'playlist_mincount': 16,
}]

View File

@@ -22,16 +22,19 @@ class EggheadBaseIE(InfoExtractor):
class EggheadCourseIE(EggheadBaseIE):
IE_DESC = 'egghead.io course'
IE_NAME = 'egghead:course'
_VALID_URL = r'https://egghead\.io/courses/(?P<id>[^/?#&]+)'
_TEST = {
_VALID_URL = r'https://(?:app\.)?egghead\.io/(?:course|playlist)s/(?P<id>[^/?#&]+)'
_TESTS = [{
'url': 'https://egghead.io/courses/professor-frisby-introduces-composable-functional-javascript',
'playlist_count': 29,
'info_dict': {
'id': '72',
'id': '432655',
'title': 'Professor Frisby Introduces Composable Functional JavaScript',
'description': 're:(?s)^This course teaches the ubiquitous.*You\'ll start composing functionality before you know it.$',
},
}
}, {
'url': 'https://app.egghead.io/playlists/professor-frisby-introduces-composable-functional-javascript',
'only_matching': True,
}]
def _real_extract(self, url):
playlist_id = self._match_id(url)
@@ -65,7 +68,7 @@ class EggheadCourseIE(EggheadBaseIE):
class EggheadLessonIE(EggheadBaseIE):
IE_DESC = 'egghead.io lesson'
IE_NAME = 'egghead:lesson'
_VALID_URL = r'https://egghead\.io/(?:api/v1/)?lessons/(?P<id>[^/?#&]+)'
_VALID_URL = r'https://(?:app\.)?egghead\.io/(?:api/v1/)?lessons/(?P<id>[^/?#&]+)'
_TESTS = [{
'url': 'https://egghead.io/lessons/javascript-linear-data-flow-with-container-style-types-box',
'info_dict': {
@@ -88,6 +91,9 @@ class EggheadLessonIE(EggheadBaseIE):
}, {
'url': 'https://egghead.io/api/v1/lessons/react-add-redux-to-a-react-application',
'only_matching': True,
}, {
'url': 'https://app.egghead.io/lessons/javascript-linear-data-flow-with-container-style-types-box',
'only_matching': True,
}]
def _real_extract(self, url):

View File

@@ -291,8 +291,7 @@ from .ctvnews import CTVNewsIE
from .cultureunplugged import CultureUnpluggedIE
from .curiositystream import (
CuriosityStreamIE,
CuriosityStreamCollectionsIE,
CuriosityStreamSeriesIE,
CuriosityStreamCollectionIE,
)
from .cwtv import CWTVIE
from .dailymail import DailyMailIE
@@ -655,10 +654,6 @@ from .linkedin import (
from .linuxacademy import LinuxAcademyIE
from .litv import LiTVIE
from .livejournal import LiveJournalIE
from .liveleak import (
LiveLeakIE,
LiveLeakEmbedIE,
)
from .livestream import (
LivestreamIE,
LivestreamOriginalIE,

View File

@@ -10,8 +10,9 @@ from ..utils import (
determine_ext,
int_or_none,
js_to_json,
urlencode_postdata,
urljoin,
ExtractorError,
urlencode_postdata
)
@@ -109,6 +110,7 @@ class FunimationIE(InfoExtractor):
if series:
title = '%s - %s' % (series, title)
description = self._html_search_meta(['description', 'og:description'], webpage, fatal=True)
subtitles = self.extract_subtitles(url, video_id, display_id)
try:
headers = {}
@@ -153,6 +155,24 @@ class FunimationIE(InfoExtractor):
'season_number': int_or_none(title_data.get('seasonNum') or _search_kane('season')),
'episode_number': int_or_none(title_data.get('episodeNum')),
'episode': episode,
'subtitles': subtitles,
'season_id': title_data.get('seriesId'),
'formats': formats,
}
def _get_subtitles(self, url, video_id, display_id):
player_url = urljoin(url, '/player/' + video_id)
player_page = self._download_webpage(player_url, display_id)
text_tracks_json_string = self._search_regex(
r'"textTracks": (\[{.+?}\])',
player_page, 'subtitles data', default='')
text_tracks = self._parse_json(
text_tracks_json_string, display_id, js_to_json, fatal=False) or []
subtitles = {}
for text_track in text_tracks:
url_element = {'url': text_track.get('src')}
language = text_track.get('language')
if text_track.get('type') == 'CC':
language += '_CC'
subtitles.setdefault(language, []).append(url_element)
return subtitles

View File

@@ -84,7 +84,6 @@ from .jwplatform import JWPlatformIE
from .digiteka import DigitekaIE
from .arkena import ArkenaIE
from .instagram import InstagramIE
from .liveleak import LiveLeakIE
from .threeqsdn import ThreeQSDNIE
from .theplatform import ThePlatformIE
from .kaltura import KalturaIE
@@ -1632,31 +1631,6 @@ class GenericIE(InfoExtractor):
'upload_date': '20160409',
},
},
# LiveLeak embed
{
'url': 'http://www.wykop.pl/link/3088787/',
'md5': '7619da8c820e835bef21a1efa2a0fc71',
'info_dict': {
'id': '874_1459135191',
'ext': 'mp4',
'title': 'Man shows poor quality of new apartment building',
'description': 'The wall is like a sand pile.',
'uploader': 'Lake8737',
},
'add_ie': [LiveLeakIE.ie_key()],
},
# Another LiveLeak embed pattern (#13336)
{
'url': 'https://milo.yiannopoulos.net/2017/06/concealed-carry-robbery/',
'info_dict': {
'id': '2eb_1496309988',
'ext': 'mp4',
'title': 'Thief robs place where everyone was armed',
'description': 'md5:694d73ee79e535953cf2488562288eee',
'uploader': 'brazilwtf',
},
'add_ie': [LiveLeakIE.ie_key()],
},
# Duplicated embedded video URLs
{
'url': 'http://www.hudl.com/athlete/2538180/highlights/149298443',
@@ -3204,11 +3178,6 @@ class GenericIE(InfoExtractor):
return self.url_result(
self._proto_relative_url(instagram_embed_url), InstagramIE.ie_key())
# Look for LiveLeak embeds
liveleak_urls = LiveLeakIE._extract_urls(webpage)
if liveleak_urls:
return self.playlist_from_matches(liveleak_urls, video_id, video_title)
# Look for 3Q SDN embeds
threeqsdn_url = ThreeQSDNIE._extract_url(webpage)
if threeqsdn_url:

View File

@@ -27,8 +27,8 @@ from ..utils import (
class HotStarBaseIE(InfoExtractor):
_AKAMAI_ENCRYPTION_KEY = b'\x05\xfc\x1a\x01\xca\xc9\x4b\xc4\x12\xfc\x53\x12\x07\x75\xf9\xee'
def _call_api_impl(self, path, video_id, query):
st = int(time.time())
def _call_api_impl(self, path, video_id, query, st=None):
st = int_or_none(st) or int(time.time())
exp = st + 6000
auth = 'st=%d~exp=%d~acl=/*' % (st, exp)
auth += '~hmac=' + hmac.new(self._AKAMAI_ENCRYPTION_KEY, auth.encode(), hashlib.sha256).hexdigest()
@@ -75,9 +75,9 @@ class HotStarBaseIE(InfoExtractor):
'tas': 10000,
})
def _call_api_v2(self, path, video_id):
def _call_api_v2(self, path, video_id, st=None):
return self._call_api_impl(
'%s/content/%s' % (path, video_id), video_id, {
'%s/content/%s' % (path, video_id), video_id, st=st, query={
'desired-config': 'audio_channel:stereo|dynamic_range:sdr|encryption:plain|ladder:tv|package:dash|resolution:hd|subs-tag:HotstarVIP|video_codec:vp9',
'device-id': compat_str(uuid.uuid4()),
'os-name': 'Windows',
@@ -131,7 +131,8 @@ class HotStarIE(HotStarBaseIE):
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
webpage, urlh = self._download_webpage_handle(url, video_id)
st = urlh.headers.get('x-origin-date')
app_state = self._parse_json(self._search_regex(
r'<script>window\.APP_STATE\s*=\s*({.+?})</script>',
webpage, 'app state'), video_id)
@@ -155,7 +156,7 @@ class HotStarIE(HotStarBaseIE):
formats = []
geo_restricted = False
# change to v2 in the future
playback_sets = self._call_api_v2('play/v1/playback', video_id)['playBackSets']
playback_sets = self._call_api_v2('play/v1/playback', video_id, st=st)['playBackSets']
for playback_set in playback_sets:
if not isinstance(playback_set, dict):
continue

View File

@@ -1,191 +0,0 @@
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..utils import int_or_none
class LiveLeakIE(InfoExtractor):
_VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?.*?\b[it]=(?P<id>[\w_]+)'
_TESTS = [{
'url': 'http://www.liveleak.com/view?i=757_1364311680',
'md5': '0813c2430bea7a46bf13acf3406992f4',
'info_dict': {
'id': '757_1364311680',
'ext': 'mp4',
'description': 'extremely bad day for this guy..!',
'uploader': 'ljfriel2',
'title': 'Most unlucky car accident',
'thumbnail': r're:^https?://.*\.jpg$'
}
}, {
'url': 'http://www.liveleak.com/view?i=f93_1390833151',
'md5': 'd3f1367d14cc3c15bf24fbfbe04b9abf',
'info_dict': {
'id': 'f93_1390833151',
'ext': 'mp4',
'description': 'German Television Channel NDR does an exclusive interview with Edward Snowden.\r\nUploaded on LiveLeak cause German Television thinks the rest of the world isn\'t intereseted in Edward Snowden.',
'uploader': 'ARD_Stinkt',
'title': 'German Television does first Edward Snowden Interview (ENGLISH)',
'thumbnail': r're:^https?://.*\.jpg$'
}
}, {
# Prochan embed
'url': 'http://www.liveleak.com/view?i=4f7_1392687779',
'md5': '42c6d97d54f1db107958760788c5f48f',
'info_dict': {
'id': '4f7_1392687779',
'ext': 'mp4',
'description': "The guy with the cigarette seems amazingly nonchalant about the whole thing... I really hope my friends' reactions would be a bit stronger.\r\n\r\nAction-go to 0:55.",
'uploader': 'CapObveus',
'title': 'Man is Fatally Struck by Reckless Car While Packing up a Moving Truck',
'age_limit': 18,
},
'skip': 'Video is dead',
}, {
# Covers https://github.com/ytdl-org/youtube-dl/pull/5983
# Multiple resolutions
'url': 'http://www.liveleak.com/view?i=801_1409392012',
'md5': 'c3a449dbaca5c0d1825caecd52a57d7b',
'info_dict': {
'id': '801_1409392012',
'ext': 'mp4',
'description': 'Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.',
'uploader': 'bony333',
'title': 'Crazy Hungarian tourist films close call waterspout in Croatia',
'thumbnail': r're:^https?://.*\.jpg$'
}
}, {
# Covers https://github.com/ytdl-org/youtube-dl/pull/10664#issuecomment-247439521
'url': 'http://m.liveleak.com/view?i=763_1473349649',
'add_ie': ['Youtube'],
'info_dict': {
'id': '763_1473349649',
'ext': 'mp4',
'title': 'Reporters and public officials ignore epidemic of black on asian violence in Sacramento | Colin Flaherty',
'description': 'Colin being the warrior he is and showing the injustice Asians in Sacramento are being subjected to.',
'uploader': 'Ziz',
'upload_date': '20160908',
'uploader_id': 'UCEbta5E_jqlZmEJsriTEtnw'
},
'params': {
'skip_download': True,
},
}, {
'url': 'https://www.liveleak.com/view?i=677_1439397581',
'info_dict': {
'id': '677_1439397581',
'title': 'Fuel Depot in China Explosion caught on video',
},
'playlist_count': 3,
}, {
'url': 'https://www.liveleak.com/view?t=HvHi_1523016227',
'only_matching': True,
}, {
# No original video
'url': 'https://www.liveleak.com/view?t=C26ZZ_1558612804',
'only_matching': True,
}]
@staticmethod
def _extract_urls(webpage):
return re.findall(
r'<iframe[^>]+src="(https?://(?:\w+\.)?liveleak\.com/ll_embed\?[^"]*[ift]=[\w_]+[^"]+)"',
webpage)
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
video_description = self._og_search_description(webpage)
video_uploader = self._html_search_regex(
r'By:.*?(\w+)</a>', webpage, 'uploader', fatal=False)
age_limit = int_or_none(self._search_regex(
r'you confirm that you are ([0-9]+) years and over.',
webpage, 'age limit', default=None))
video_thumbnail = self._og_search_thumbnail(webpage)
entries = self._parse_html5_media_entries(url, webpage, video_id)
if not entries:
# Maybe an embed?
embed_url = self._search_regex(
r'<iframe[^>]+src="((?:https?:)?//(?:www\.)?(?:prochan|youtube)\.com/embed[^"]+)"',
webpage, 'embed URL')
return {
'_type': 'url_transparent',
'url': embed_url,
'id': video_id,
'title': video_title,
'description': video_description,
'uploader': video_uploader,
'age_limit': age_limit,
}
for idx, info_dict in enumerate(entries):
formats = []
for a_format in info_dict['formats']:
if not a_format.get('height'):
a_format['height'] = int_or_none(self._search_regex(
r'([0-9]+)p\.mp4', a_format['url'], 'height label',
default=None))
formats.append(a_format)
# Removing '.*.mp4' gives the raw video, which is essentially
# the same video without the LiveLeak logo at the top (see
# https://github.com/ytdl-org/youtube-dl/pull/4768)
orig_url = re.sub(r'\.mp4\.[^.]+', '', a_format['url'])
if a_format['url'] != orig_url:
format_id = a_format.get('format_id')
format_id = 'original' + ('-' + format_id if format_id else '')
if self._is_valid_url(orig_url, video_id, format_id):
formats.append({
'format_id': format_id,
'url': orig_url,
'quality': 1,
})
self._sort_formats(formats)
info_dict['formats'] = formats
# Don't append entry ID for one-video pages to keep backward compatibility
if len(entries) > 1:
info_dict['id'] = '%s_%s' % (video_id, idx + 1)
else:
info_dict['id'] = video_id
info_dict.update({
'title': video_title,
'description': video_description,
'uploader': video_uploader,
'age_limit': age_limit,
'thumbnail': video_thumbnail,
})
return self.playlist_result(entries, video_id, video_title)
class LiveLeakEmbedIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?liveleak\.com/ll_embed\?.*?\b(?P<kind>[ift])=(?P<id>[\w_]+)'
# See generic.py for actual test cases
_TESTS = [{
'url': 'https://www.liveleak.com/ll_embed?i=874_1459135191',
'only_matching': True,
}, {
'url': 'https://www.liveleak.com/ll_embed?f=ab065df993c1',
'only_matching': True,
}]
def _real_extract(self, url):
kind, video_id = re.match(self._VALID_URL, url).groups()
if kind == 'f':
webpage = self._download_webpage(url, video_id)
liveleak_url = self._search_regex(
r'(?:logourl\s*:\s*|window\.open\()(?P<q1>[\'"])(?P<url>%s)(?P=q1)' % LiveLeakIE._VALID_URL,
webpage, 'LiveLeak URL', group='url')
else:
liveleak_url = 'http://www.liveleak.com/view?%s=%s' % (kind, video_id)
return self.url_result(liveleak_url, ie=LiveLeakIE.ie_key())

View File

@@ -122,6 +122,52 @@ class MediasiteIE(InfoExtractor):
r'(?xi)<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:(?:https?:)?//[^/]+)?/Mediasite/Play/%s(?:\?.*?)?)\1' % _ID_RE,
webpage)]
def __extract_slides(self, *, stream_id, snum, Stream, duration, images):
slide_base_url = Stream['SlideBaseUrl']
fname_template = Stream['SlideImageFileNameTemplate']
if fname_template != 'slide_{0:D4}.jpg':
self.report_warning('Unusual slide file name template; report a bug if slide downloading fails')
fname_template = re.sub(r'\{0:D([0-9]+)\}', r'{0:0\1}', fname_template)
fragments = []
for i, slide in enumerate(Stream['Slides']):
if i == 0:
if slide['Time'] > 0:
default_slide = images.get('DefaultSlide')
if default_slide is None:
default_slide = images.get('DefaultStreamImage')
if default_slide is not None:
default_slide = default_slide['ImageFilename']
if default_slide is not None:
fragments.append({
'path': default_slide,
'duration': slide['Time'] / 1000,
})
next_time = try_get(None, [
lambda _: Stream['Slides'][i + 1]['Time'],
lambda _: duration,
lambda _: slide['Time'],
], expected_type=(int, float))
fragments.append({
'path': fname_template.format(slide.get('Number', i + 1)),
'duration': (next_time - slide['Time']) / 1000
})
return {
'format_id': '%s-%u.slides' % (stream_id, snum),
'ext': 'mhtml',
'url': slide_base_url,
'protocol': 'mhtml',
'acodec': 'none',
'vcodec': 'none',
'format_note': 'Slides',
'fragments': fragments,
'fragment_base_url': slide_base_url,
}
def _real_extract(self, url):
url, data = unsmuggle_url(url, {})
mobj = re.match(self._VALID_URL, url)
@@ -198,10 +244,15 @@ class MediasiteIE(InfoExtractor):
'ext': mimetype2ext(VideoUrl.get('MimeType')),
})
# TODO: if Stream['HasSlideContent']:
# synthesise an MJPEG video stream '%s-%u.slides' % (stream_type, snum)
# from Stream['Slides']
# this will require writing a custom downloader...
if Stream.get('HasSlideContent', False):
images = player_options['PlayerLayoutOptions']['Images']
stream_formats.append(self.__extract_slides(
stream_id=stream_id,
snum=snum,
Stream=Stream,
duration=presentation.get('Duration'),
images=images,
))
# disprefer 'secondary' streams
if stream_type != 0:

View File

@@ -58,7 +58,7 @@ class NRKBaseIE(InfoExtractor):
def _call_api(self, path, video_id, item=None, note=None, fatal=True, query=None):
return self._download_json(
urljoin('http://psapi.nrk.no/', path),
urljoin('https://psapi.nrk.no/', path),
video_id, note or 'Downloading %s JSON' % item,
fatal=fatal, query=query,
headers={'Accept-Encoding': 'gzip, deflate, br'})

View File

@@ -98,6 +98,9 @@ class ORFTVthekIE(InfoExtractor):
elif ext == 'f4m':
formats.extend(self._extract_f4m_formats(
src, video_id, f4m_id=format_id, fatal=False))
elif ext == 'mpd':
formats.extend(self._extract_mpd_formats(
src, video_id, mpd_id=format_id, fatal=False))
else:
formats.append({
'format_id': format_id,

View File

@@ -14,6 +14,7 @@ from ..compat import (
)
from .openload import PhantomJSwrapper
from ..utils import (
clean_html,
determine_ext,
ExtractorError,
int_or_none,
@@ -30,6 +31,7 @@ from ..utils import (
class PornHubBaseIE(InfoExtractor):
_NETRC_MACHINE = 'pornhub'
_PORNHUB_HOST_RE = r'(?:(?P<host>pornhub(?:premium)?\.(?:com|net|org))|pornhubthbh7ap3u\.onion)'
def _download_webpage_handle(self, *args, **kwargs):
def dl(*args, **kwargs):
@@ -122,11 +124,13 @@ class PornHubIE(PornHubBaseIE):
_VALID_URL = r'''(?x)
https?://
(?:
(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
(?:[^/]+\.)?
%s
/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
(?:www\.)?thumbzilla\.com/video/
)
(?P<id>[\da-z]+)
'''
''' % PornHubBaseIE._PORNHUB_HOST_RE
_TESTS = [{
'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
'md5': 'a6391306d050e4547f62b3f485dd9ba9',
@@ -145,6 +149,7 @@ class PornHubIE(PornHubBaseIE):
'age_limit': 18,
'tags': list,
'categories': list,
'cast': list,
},
}, {
# non-ASCII title
@@ -236,6 +241,13 @@ class PornHubIE(PornHubBaseIE):
}, {
'url': 'https://www.pornhubpremium.com/view_video.php?viewkey=ph5f75b0f4b18e3',
'only_matching': True,
}, {
# geo restricted
'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5a9813bfa7156',
'only_matching': True,
}, {
'url': 'http://pornhubthbh7ap3u.onion/view_video.php?viewkey=ph5a9813bfa7156',
'only_matching': True,
}]
@staticmethod
@@ -275,6 +287,11 @@ class PornHubIE(PornHubBaseIE):
'PornHub said: %s' % error_msg,
expected=True, video_id=video_id)
if any(re.search(p, webpage) for p in (
r'class=["\']geoBlocked["\']',
r'>\s*This content is unavailable in your country')):
self.raise_geo_restricted()
# video_title from flashvars contains whitespace instead of non-ASCII (see
# http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
# on that anymore.
@@ -408,17 +425,14 @@ class PornHubIE(PornHubBaseIE):
format_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False))
return
tbr = None
mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', format_url)
if mobj:
if not height:
height = int(mobj.group('height'))
tbr = int(mobj.group('tbr'))
if not height:
height = int_or_none(self._search_regex(
r'(?P<height>\d+)[pP]?_\d+[kK]', format_url, 'height',
default=None))
formats.append({
'url': format_url,
'format_id': '%dp' % height if height else None,
'height': height,
'tbr': tbr,
})
for video_url, height in video_urls:
@@ -440,7 +454,10 @@ class PornHubIE(PornHubBaseIE):
add_format(video_url, height)
continue
add_format(video_url)
self._sort_formats(formats)
# field_preference is unnecessary here, but kept for code-similarity with youtube-dl
self._sort_formats(
formats, field_preference=('height', 'width', 'fps', 'format_id'))
video_uploader = self._html_search_regex(
r'(?s)From:&nbsp;.+?<(?:a\b[^>]+\bhref=["\']/(?:(?:user|channel)s|model|pornstar)/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<',
@@ -464,7 +481,7 @@ class PornHubIE(PornHubBaseIE):
r'(?s)<div[^>]+\bclass=["\'].*?\b%sWrapper[^>]*>(.+?)</div>'
% meta_key, webpage, meta_key, default=None)
if div:
return re.findall(r'<a[^>]+\bhref=[^>]+>([^<]+)', div)
return [clean_html(x).strip() for x in re.findall(r'(?s)<a[^>]+\bhref=[^>]+>.+?</a>', div)]
info = self._search_json_ld(webpage, video_id, default={})
# description provided in JSON-LD is irrelevant
@@ -485,6 +502,7 @@ class PornHubIE(PornHubBaseIE):
'age_limit': 18,
'tags': extract_list('tags'),
'categories': extract_list('categories'),
'cast': extract_list('pornstars'),
'subtitles': subtitles,
}, info)
@@ -513,7 +531,7 @@ class PornHubPlaylistBaseIE(PornHubBaseIE):
class PornHubUserIE(PornHubPlaylistBaseIE):
_VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)'
_VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?%s/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)' % PornHubBaseIE._PORNHUB_HOST_RE
_TESTS = [{
'url': 'https://www.pornhub.com/model/zoe_ph',
'playlist_mincount': 118,
@@ -542,6 +560,9 @@ class PornHubUserIE(PornHubPlaylistBaseIE):
# Same as before, multi page
'url': 'https://www.pornhubpremium.com/pornstar/lily-labeau',
'only_matching': True,
}, {
'url': 'https://pornhubthbh7ap3u.onion/model/zoe_ph',
'only_matching': True,
}]
def _real_extract(self, url):
@@ -617,7 +638,7 @@ class PornHubPagedPlaylistBaseIE(PornHubPlaylistBaseIE):
class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE):
_VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?P<id>(?:[^/]+/)*[^/?#&]+)'
_VALID_URL = r'https?://(?:[^/]+\.)?%s/(?P<id>(?:[^/]+/)*[^/?#&]+)' % PornHubBaseIE._PORNHUB_HOST_RE
_TESTS = [{
'url': 'https://www.pornhub.com/model/zoe_ph/videos',
'only_matching': True,
@@ -722,6 +743,9 @@ class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE):
}, {
'url': 'https://de.pornhub.com/playlist/4667351',
'only_matching': True,
}, {
'url': 'https://pornhubthbh7ap3u.onion/model/zoe_ph/videos',
'only_matching': True,
}]
@classmethod
@@ -732,7 +756,7 @@ class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE):
class PornHubUserVideosUploadIE(PornHubPagedPlaylistBaseIE):
_VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)'
_VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?%s/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)' % PornHubBaseIE._PORNHUB_HOST_RE
_TESTS = [{
'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos/upload',
'info_dict': {
@@ -742,4 +766,7 @@ class PornHubUserVideosUploadIE(PornHubPagedPlaylistBaseIE):
}, {
'url': 'https://www.pornhub.com/model/zoe_ph/videos/upload',
'only_matching': True,
}, {
'url': 'http://pornhubthbh7ap3u.onion/pornstar/jenny-blighe/videos/upload',
'only_matching': True,
}]

View File

@@ -5,12 +5,14 @@ import itertools
import re
from .common import InfoExtractor
from ..downloader.websocket import has_websockets
from ..utils import (
clean_html,
float_or_none,
get_element_by_class,
get_element_by_id,
parse_duration,
qualities,
str_to_int,
try_get,
unified_timestamp,
@@ -89,9 +91,24 @@ class TwitCastingIE(InfoExtractor):
video_js_data = video_js_data[0]
m3u8_url = try_get(video_js_data, lambda x: x['source']['url'])
stream_server_data = self._download_json(
'https://twitcasting.tv/streamserver.php?target=%s&mode=client' % uploader_id, video_id,
'Downloading live info', fatal=False)
is_live = 'data-status="online"' in webpage
formats = []
if is_live and not m3u8_url:
m3u8_url = 'https://twitcasting.tv/%s/metastream.m3u8' % uploader_id
if is_live and has_websockets and stream_server_data:
qq = qualities(['base', 'mobilesource', 'main'])
for mode, ws_url in stream_server_data['llfmp4']['streams'].items():
formats.append({
'url': ws_url,
'format_id': 'ws-%s' % mode,
'ext': 'mp4',
'quality': qq(mode),
'protocol': 'websocket_frag', # TwitCasting simply sends moof atom directly over WS
})
thumbnail = video_js_data.get('thumbnailUrl') or self._og_search_thumbnail(webpage)
description = clean_html(get_element_by_id(
@@ -106,10 +123,9 @@ class TwitCastingIE(InfoExtractor):
r'data-toggle="true"[^>]+datetime="([^"]+)"',
webpage, 'datetime', None))
formats = None
if m3u8_url:
formats = self._extract_m3u8_formats(
m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', live=is_live)
formats.extend(self._extract_m3u8_formats(
m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', live=is_live))
self._sort_formats(formats)
return {

View File

@@ -28,7 +28,7 @@ class UMGDeIE(InfoExtractor):
def _real_extract(self, url):
video_id = self._match_id(url)
video_data = self._download_json(
'https://api.universal-music.de/graphql',
'https://graphql.universal-music.de/',
video_id, query={
'query': '''{
universalMusic(channel:16) {
@@ -56,11 +56,9 @@ class UMGDeIE(InfoExtractor):
formats = []
def add_m3u8_format(format_id):
m3u8_formats = self._extract_m3u8_formats(
formats.extend(self._extract_m3u8_formats(
hls_url_template % format_id, video_id, 'mp4',
'm3u8_native', m3u8_id='hls', fatal='False')
if m3u8_formats and m3u8_formats[0].get('height'):
formats.extend(m3u8_formats)
'm3u8_native', m3u8_id='hls', fatal=False))
for f in video_data.get('formats', []):
f_url = f.get('url')

View File

@@ -142,6 +142,7 @@ class VikiIE(VikiBaseIE):
IE_NAME = 'viki'
_VALID_URL = r'%s(?:videos|player)/(?P<id>[0-9]+v)' % VikiBaseIE._VALID_URL_BASE
_TESTS = [{
'note': 'Free non-DRM video with storyboards in MPD',
'url': 'https://www.viki.com/videos/1175236v-choosing-spouse-by-lottery-episode-1',
'info_dict': {
'id': '1175236v',
@@ -155,7 +156,6 @@ class VikiIE(VikiBaseIE):
'params': {
'format': 'bestvideo',
},
'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'],
}, {
'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
'info_dict': {
@@ -173,7 +173,6 @@ class VikiIE(VikiBaseIE):
'format': 'bestvideo',
},
'skip': 'Blocked in the US',
'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'],
}, {
# clip
'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference',
@@ -225,7 +224,6 @@ class VikiIE(VikiBaseIE):
'params': {
'format': 'bestvideo',
},
'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'],
}, {
# youtube external
'url': 'http://www.viki.com/videos/50562v-poor-nastya-complete-episode-1',
@@ -264,7 +262,6 @@ class VikiIE(VikiBaseIE):
'params': {
'format': 'bestvideo',
},
'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'],
}]
def _real_extract(self, url):

View File

@@ -301,12 +301,22 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
_YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
def _generate_sapisidhash_header(self):
sapisid_cookie = self._get_cookies('https://www.youtube.com').get('SAPISID')
# Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
# See: https://github.com/yt-dlp/yt-dlp/issues/393
yt_cookies = self._get_cookies('https://www.youtube.com')
sapisid_cookie = dict_get(
yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
if sapisid_cookie is None:
return
time_now = round(time.time())
sapisidhash = hashlib.sha1((str(time_now) + " " + sapisid_cookie.value + " " + "https://www.youtube.com").encode("utf-8")).hexdigest()
return "SAPISIDHASH %s_%s" % (time_now, sapisidhash)
# SAPISID cookie is required if not already present
if not yt_cookies.get('SAPISID'):
self._set_cookie(
'.youtube.com', 'SAPISID', sapisid_cookie.value, secure=True, expire_time=time_now + 3600)
# SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
sapisidhash = hashlib.sha1(
f'{time_now} {sapisid_cookie.value} https://www.youtube.com'.encode('utf-8')).hexdigest()
return f'SAPISIDHASH {time_now}_{sapisidhash}'
def _call_api(self, ep, query, video_id, fatal=True, headers=None,
note='Downloading API JSON', errnote='Unable to download API page',
@@ -454,20 +464,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# Invidious instances taken from https://github.com/iv-org/documentation/blob/master/Invidious-Instances.md
r'(?:www\.)?invidious\.pussthecat\.org',
r'(?:www\.)?invidious\.zee\.li',
r'(?:(?:www|au)\.)?ytprivate\.com',
r'(?:www\.)?invidious\.namazso\.eu',
r'(?:www\.)?invidious\.ethibox\.fr',
r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
# youtube-dl invidious instances list
r'(?:(?:www|no)\.)?invidiou\.sh',
r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
r'(?:www\.)?invidious\.kabi\.tk',
r'(?:www\.)?invidious\.mastodon\.host',
r'(?:www\.)?invidious\.zapashcanon\.fr',
r'(?:www\.)?invidious\.kavin\.rocks',
r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
r'(?:www\.)?invidious\.tinfoil-hat\.net',
r'(?:www\.)?invidious\.himiko\.cloud',
r'(?:www\.)?invidious\.reallyancient\.tech',
@@ -494,6 +499,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
r'(?:www\.)?invidious\.toot\.koeln',
r'(?:www\.)?invidious\.fdn\.fr',
r'(?:www\.)?watch\.nettohikari\.com',
r'(?:www\.)?invidious\.namazso\.eu',
r'(?:www\.)?invidious\.silkky\.cloud',
r'(?:www\.)?invidious\.exonip\.de',
r'(?:www\.)?invidious\.riverside\.rocks',
r'(?:www\.)?invidious\.blamefran\.net',
r'(?:www\.)?invidious\.moomoo\.de',
r'(?:www\.)?ytb\.trom\.tf',
r'(?:www\.)?yt\.cyberhost\.uk',
r'(?:www\.)?kgg2m7yk5aybusll\.onion',
r'(?:www\.)?qklhadlycap4cnod\.onion',
r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
@@ -502,6 +515,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
)
_VALID_URL = r"""(?x)^
(
@@ -1866,6 +1883,16 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'comment_count': len(comments),
}
@staticmethod
def _get_video_info_params(video_id):
return {
'video_id': video_id,
'eurl': 'https://youtube.googleapis.com/v/' + video_id,
'html5': '1',
'c': 'TVHTML5',
'cver': '6.20180913',
}
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})
video_id = self._match_id(url)
@@ -1898,16 +1925,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
base_url + 'get_video_info', video_id,
'Fetching youtube music info webpage',
'unable to download youtube music info webpage', query={
'video_id': video_id,
'eurl': 'https://youtube.googleapis.com/v/' + video_id,
**self._get_video_info_params(video_id),
'el': 'detailpage',
'c': 'WEB_REMIX',
'cver': '0.1',
'cplayer': 'UNIPLAYER',
'html5': '1',
}, fatal=False)),
}, fatal=False) or ''),
lambda x: x['player_response'][0],
compat_str) or '{}', video_id)
compat_str) or '{}', video_id, fatal=False)
ytm_streaming_data = ytm_player_response.get('streamingData') or {}
player_response = None
@@ -1926,12 +1951,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
pr = self._parse_json(try_get(compat_parse_qs(
self._download_webpage(
base_url + 'get_video_info', video_id,
'Refetching age-gated info webpage',
'unable to download video info webpage', query={
'video_id': video_id,
'eurl': 'https://youtube.googleapis.com/v/' + video_id,
'html5': '1',
}, fatal=False)),
'Refetching age-gated info webpage', 'unable to download video info webpage',
query=self._get_video_info_params(video_id), fatal=False)),
lambda x: x['player_response'][0],
compat_str) or '{}', video_id)
if pr:
@@ -2325,18 +2346,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
initial_data = self._call_api(
'next', {'videoId': video_id}, video_id, fatal=False, api_key=self._extract_api_key(ytcfg))
if not is_live:
try:
# This will error if there is no livechat
initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
info['subtitles']['live_chat'] = [{
'url': 'https://www.youtube.com/watch?v=%s' % video_id, # url is needed to set cookies
'video_id': video_id,
'ext': 'json',
'protocol': 'youtube_live_chat_replay',
}]
except (KeyError, IndexError, TypeError):
pass
try:
# This will error if there is no livechat
initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
info['subtitles']['live_chat'] = [{
'url': 'https://www.youtube.com/watch?v=%s' % video_id, # url is needed to set cookies
'video_id': video_id,
'ext': 'json',
'protocol': 'youtube_live_chat' if is_live else 'youtube_live_chat_replay',
}]
except (KeyError, IndexError, TypeError):
pass
if initial_data:
chapters = self._extract_chapters_from_json(
@@ -3592,7 +3612,13 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
else:
# Youtube may send alerts if there was an issue with the continuation page
self._extract_and_report_alerts(response, expected=False)
try:
self._extract_and_report_alerts(response, expected=False)
except ExtractorError as e:
if fatal:
raise
self.report_warning(error_to_compat_str(e))
return
if not check_get_keys or dict_get(response, check_get_keys):
break
# Youtube sometimes sends incomplete data
@@ -4062,6 +4088,7 @@ class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
_VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
_FEED_NAME = 'recommended'
_LOGIN_REQUIRED = False
_TESTS = [{
'url': ':ytrec',
'only_matching': True,

View File

@@ -599,6 +599,10 @@ def parseOpts(overrideArguments=None):
'-r', '--limit-rate', '--rate-limit',
dest='ratelimit', metavar='RATE',
help='Maximum download rate in bytes per second (e.g. 50K or 4.2M)')
downloader.add_option(
'--throttled-rate',
dest='throttledratelimit', metavar='RATE',
help='Minimum download rate in bytes per second below which throttling is assumed and the video data is re-extracted (e.g. 100K)')
downloader.add_option(
'-R', '--retries',
dest='retries', metavar='RETRIES', default=10,
@@ -1165,7 +1169,7 @@ def parseOpts(overrideArguments=None):
'to give the argument to the specified postprocessor/executable. Supported PP are: '
'Merger, ExtractAudio, SplitChapters, Metadata, EmbedSubtitle, EmbedThumbnail, '
'SubtitlesConvertor, ThumbnailsConvertor, VideoRemuxer, VideoConvertor, '
'SponSkrub, FixupStretched, FixupM4a and FixupM3u8. '
'SponSkrub, FixupStretched, FixupM4a, FixupM3u8, FixupTimestamp and FixupDuration. '
'The supported executables are: AtomicParsley, FFmpeg, FFprobe, and SponSkrub. '
'You can also specify "PP+EXE:ARGS" to give the arguments to the specified executable '
'only when being used by the specified postprocessor. Additionally, for ffmpeg/ffprobe, '
@@ -1200,19 +1204,19 @@ def parseOpts(overrideArguments=None):
postproc.add_option(
'--embed-thumbnail',
action='store_true', dest='embedthumbnail', default=False,
help='Embed thumbnail in the audio as cover art')
help='Embed thumbnail in the video as cover art')
postproc.add_option(
'--no-embed-thumbnail',
action='store_false', dest='embedthumbnail',
help='Do not embed thumbnail (default)')
postproc.add_option(
'--add-metadata',
'--embed-metadata', '--add-metadata',
action='store_true', dest='addmetadata', default=False,
help='Write metadata to the video file')
help='Embed metadata including chapter markers (if supported by the format) to the video file (Alias: --add-metadata)')
postproc.add_option(
'--no-add-metadata',
'--no-embed-metadata', '--no-add-metadata',
action='store_false', dest='addmetadata',
help='Do not write metadata (default)')
help='Do not write metadata (default) (Alias: --no-add-metadata)')
postproc.add_option(
'--metadata-from-title',
metavar='FORMAT', dest='metafromtitle',
@@ -1230,10 +1234,12 @@ def parseOpts(overrideArguments=None):
postproc.add_option(
'--fixup',
metavar='POLICY', dest='fixup', default=None,
choices=('never', 'ignore', 'warn', 'detect_or_warn', 'force'),
help=(
'Automatically correct known faults of the file. '
'One of never (do nothing), warn (only emit a warning), '
'detect_or_warn (the default; fix file if we can, warn otherwise)'))
'detect_or_warn (the default; fix file if we can, warn otherwise), '
'force (try fixing even if file already exists'))
postproc.add_option(
'--prefer-avconv', '--no-prefer-ffmpeg',
action='store_false', dest='prefer_ffmpeg',

View File

@@ -5,7 +5,9 @@ from .ffmpeg import (
FFmpegPostProcessor,
FFmpegEmbedSubtitlePP,
FFmpegExtractAudioPP,
FFmpegFixupDurationPP,
FFmpegFixupStretchedPP,
FFmpegFixupTimestampPP,
FFmpegFixupM3u8PP,
FFmpegFixupM4aPP,
FFmpegMergerPP,
@@ -35,9 +37,11 @@ __all__ = [
'FFmpegEmbedSubtitlePP',
'FFmpegExtractAudioPP',
'FFmpegSplitChaptersPP',
'FFmpegFixupDurationPP',
'FFmpegFixupM3u8PP',
'FFmpegFixupM4aPP',
'FFmpegFixupStretchedPP',
'FFmpegFixupTimestampPP',
'FFmpegMergerPP',
'FFmpegMetadataPP',
'FFmpegSubtitlesConvertorPP',

View File

@@ -1,5 +1,6 @@
from __future__ import unicode_literals
import functools
import os
from ..compat import compat_str
@@ -67,6 +68,25 @@ class PostProcessor(object):
"""Sets the downloader for this PP."""
self._downloader = downloader
@staticmethod
def _restrict_to(*, video=True, audio=True, images=True):
allowed = {'video': video, 'audio': audio, 'images': images}
def decorator(func):
@functools.wraps(func)
def wrapper(self, info):
format_type = (
'video' if info.get('vcodec') != 'none'
else 'audio' if info.get('acodec') != 'none'
else 'images')
if allowed[format_type]:
return func(self, info)
else:
self.to_screen('Skipping %s' % format_type)
return [], info
return wrapper
return decorator
def run(self, information):
"""Run the PostProcessor.

View File

@@ -16,6 +16,7 @@ try:
except ImportError:
has_mutagen = False
from .common import PostProcessor
from .ffmpeg import (
FFmpegPostProcessor,
FFmpegThumbnailsConvertorPP,
@@ -62,6 +63,7 @@ class EmbedThumbnailPP(FFmpegPostProcessor):
def _report_run(self, exe, filename):
self.to_screen('%s: Adding thumbnail to "%s"' % (exe, filename))
@PostProcessor._restrict_to(images=False)
def run(self, info):
filename = info['filepath']
temp_filename = prepend_extension(filename, 'temp')
@@ -123,8 +125,9 @@ class EmbedThumbnailPP(FFmpegPostProcessor):
self.run_ffmpeg(filename, temp_filename, options)
elif info['ext'] in ['m4a', 'mp4', 'mov']:
prefer_atomicparsley = 'embed-thumbnail-atomicparsley' in self.get_param('compat_opts', [])
# Method 1: Use mutagen
if not has_mutagen:
if not has_mutagen or prefer_atomicparsley:
success = False
else:
try:
@@ -143,7 +146,7 @@ class EmbedThumbnailPP(FFmpegPostProcessor):
success = False
# Method 2: Use ffmpeg+ffprobe
if not success:
if not success and not prefer_atomicparsley:
success = True
try:
options = ['-c', 'copy', '-map', '0', '-dn', '-map', '1']

View File

@@ -310,6 +310,7 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
except FFmpegPostProcessorError as err:
raise AudioConversionError(err.msg)
@PostProcessor._restrict_to(images=False)
def run(self, information):
path = information['filepath']
orig_ext = information['ext']
@@ -419,6 +420,7 @@ class FFmpegVideoConvertorPP(FFmpegPostProcessor):
return ['-c:v', 'libxvid', '-vtag', 'XVID']
return []
@PostProcessor._restrict_to(images=False)
def run(self, information):
path, source_ext = information['filepath'], information['ext'].lower()
target_ext = self._target_ext(source_ext)
@@ -456,6 +458,7 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
super(FFmpegEmbedSubtitlePP, self).__init__(downloader)
self._already_have_subtitle = already_have_subtitle
@PostProcessor._restrict_to(images=False)
def run(self, information):
if information['ext'] not in ('mp4', 'webm', 'mkv'):
self.to_screen('Subtitles can only be embedded in mp4, webm or mkv files')
@@ -523,6 +526,7 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
class FFmpegMetadataPP(FFmpegPostProcessor):
@PostProcessor._restrict_to(images=False)
def run(self, info):
metadata = {}
@@ -625,6 +629,7 @@ class FFmpegMetadataPP(FFmpegPostProcessor):
class FFmpegMergerPP(FFmpegPostProcessor):
@PostProcessor._restrict_to(images=False)
def run(self, info):
filename = info['filepath']
temp_filename = prepend_extension(filename, 'temp')
@@ -656,55 +661,71 @@ class FFmpegMergerPP(FFmpegPostProcessor):
return True
class FFmpegFixupStretchedPP(FFmpegPostProcessor):
class FFmpegFixupPostProcessor(FFmpegPostProcessor):
def _fixup(self, msg, filename, options):
temp_filename = prepend_extension(filename, 'temp')
self.to_screen(f'{msg} of "{filename}"')
self.run_ffmpeg(filename, temp_filename, options)
os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
class FFmpegFixupStretchedPP(FFmpegFixupPostProcessor):
@PostProcessor._restrict_to(images=False, audio=False)
def run(self, info):
stretched_ratio = info.get('stretched_ratio')
if stretched_ratio is None or stretched_ratio == 1:
return [], info
filename = info['filepath']
temp_filename = prepend_extension(filename, 'temp')
options = ['-c', 'copy', '-map', '0', '-dn', '-aspect', '%f' % stretched_ratio]
self.to_screen('Fixing aspect ratio in "%s"' % filename)
self.run_ffmpeg(filename, temp_filename, options)
os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
if stretched_ratio not in (None, 1):
self._fixup('Fixing aspect ratio', info['filepath'], [
'-c', 'copy', '-map', '0', '-dn', '-aspect', '%f' % stretched_ratio])
return [], info
class FFmpegFixupM4aPP(FFmpegPostProcessor):
class FFmpegFixupM4aPP(FFmpegFixupPostProcessor):
@PostProcessor._restrict_to(images=False, video=False)
def run(self, info):
if info.get('container') != 'm4a_dash':
return [], info
filename = info['filepath']
temp_filename = prepend_extension(filename, 'temp')
options = ['-c', 'copy', '-map', '0', '-dn', '-f', 'mp4']
self.to_screen('Correcting container in "%s"' % filename)
self.run_ffmpeg(filename, temp_filename, options)
os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
if info.get('container') == 'm4a_dash':
self._fixup('Correcting container', info['filepath'], [
'-c', 'copy', '-map', '0', '-dn', '-f', 'mp4'])
return [], info
class FFmpegFixupM3u8PP(FFmpegPostProcessor):
class FFmpegFixupM3u8PP(FFmpegFixupPostProcessor):
@PostProcessor._restrict_to(images=False)
def run(self, info):
filename = info['filepath']
if self.get_audio_codec(filename) == 'aac':
temp_filename = prepend_extension(filename, 'temp')
if self.get_audio_codec(info['filepath']) == 'aac':
self._fixup('Fixing malformed AAC bitstream', info['filepath'], [
'-c', 'copy', '-map', '0', '-dn', '-f', 'mp4', '-bsf:a', 'aac_adtstoasc'])
return [], info
options = ['-c', 'copy', '-map', '0', '-dn', '-f', 'mp4', '-bsf:a', 'aac_adtstoasc']
self.to_screen('Fixing malformed AAC bitstream in "%s"' % filename)
self.run_ffmpeg(filename, temp_filename, options)
os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
class FFmpegFixupTimestampPP(FFmpegFixupPostProcessor):
def __init__(self, downloader=None, trim=0.001):
# "trim" should be used when the video contains unintended packets
super(FFmpegFixupTimestampPP, self).__init__(downloader)
assert isinstance(trim, (int, float))
self.trim = str(trim)
@PostProcessor._restrict_to(images=False)
def run(self, info):
required_version = '4.4'
if is_outdated_version(self._versions[self.basename], required_version):
self.report_warning(
'A re-encode is needed to fix timestamps in older versions of ffmpeg. '
f'Please install ffmpeg {required_version} or later to fixup without re-encoding')
opts = ['-vf', 'setpts=PTS-STARTPTS']
else:
opts = ['-c', 'copy', '-bsf', 'setts=ts=TS-STARTPTS']
self._fixup('Fixing frame timestamp', info['filepath'], opts + ['-map', '0', '-dn', '-ss', self.trim])
return [], info
class FFmpegFixupDurationPP(FFmpegFixupPostProcessor):
@PostProcessor._restrict_to(images=False)
def run(self, info):
self._fixup('Fixing video duration', info['filepath'], ['-c', 'copy', '-map', '0', '-dn'])
return [], info
@@ -805,6 +826,7 @@ class FFmpegSplitChaptersPP(FFmpegPostProcessor):
['-ss', compat_str(chapter['start_time']),
'-t', compat_str(chapter['end_time'] - chapter['start_time'])])
@PostProcessor._restrict_to(images=False)
def run(self, info):
chapters = info.get('chapters') or []
if not chapters:

View File

@@ -41,6 +41,7 @@ class SponSkrubPP(PostProcessor):
return None
return path
@PostProcessor._restrict_to(images=False)
def run(self, information):
if self.path is None:
return [], information

View File

@@ -89,13 +89,9 @@ def run_update(ydl):
err = None
if isinstance(globals().get('__loader__'), zipimporter):
# We only support python 3.6 or above
if sys.version_info < (3, 6):
err = 'This is the last release of yt-dlp for Python version %d.%d! Please update to Python 3.6 or above' % sys.version_info[:2]
pass
elif hasattr(sys, 'frozen'):
# Python 3.6 supports only vista and above
if sys.getwindowsversion()[0] < 6:
err = 'This is the last release of yt-dlp for your version of Windows. Please update to Windows Vista or above'
pass
else:
err = 'It looks like you installed yt-dlp with a package manager, pip, setup.py or a tarball. Please use that to update'
if err:

View File

@@ -2107,6 +2107,8 @@ def sanitize_filename(s, restricted=False, is_id=False):
return '_'
return char
if s == '':
return ''
# Handle timestamps
s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s)
result = ''.join(map(replace_insane, s))
@@ -2242,6 +2244,17 @@ def unescapeHTML(s):
r'&([^&;]+;)', lambda m: _htmlentity_transform(m.group(1)), s)
def escapeHTML(text):
return (
text
.replace('&', '&amp;')
.replace('<', '&lt;')
.replace('>', '&gt;')
.replace('"', '&quot;')
.replace("'", '&#39;')
)
def process_communicate_or_kill(p, *args, **kwargs):
try:
return p.communicate(*args, **kwargs)
@@ -2321,13 +2334,14 @@ def decodeOption(optval):
return optval
def formatSeconds(secs, delim=':'):
def formatSeconds(secs, delim=':', msec=False):
if secs > 3600:
return '%d%s%02d%s%02d' % (secs // 3600, delim, (secs % 3600) // 60, delim, secs % 60)
ret = '%d%s%02d%s%02d' % (secs // 3600, delim, (secs % 3600) // 60, delim, secs % 60)
elif secs > 60:
return '%d%s%02d' % (secs // 60, delim, secs % 60)
ret = '%d%s%02d' % (secs // 60, delim, secs % 60)
else:
return '%d' % secs
ret = '%d' % secs
return '%s.%03d' % (ret, secs % 1) if msec else ret
def make_HTTPS_handler(params, **kwargs):
@@ -2490,6 +2504,11 @@ class RejectedVideoReached(YoutubeDLError):
pass
class ThrottledDownload(YoutubeDLError):
""" Download speed below --throttled-rate. """
pass
class MaxDownloadsReached(YoutubeDLError):
""" --max-downloads limit has been reached. """
pass
@@ -3952,10 +3971,14 @@ class LazyList(collections.Sequence):
def __init__(self, iterable):
self.__iterable = iter(iterable)
self.__cache = []
self.__reversed = False
def __iter__(self):
for item in self.__cache:
yield item
if self.__reversed:
# We need to consume the entire iterable to iterate in reverse
yield from self.exhaust()[::-1]
return
yield from self.__cache
for item in self.__iterable:
self.__cache.append(item)
yield item
@@ -3963,29 +3986,39 @@ class LazyList(collections.Sequence):
def exhaust(self):
''' Evaluate the entire iterable '''
self.__cache.extend(self.__iterable)
return self.__cache
@staticmethod
def _reverse_index(x):
return -(x + 1)
def __getitem__(self, idx):
if isinstance(idx, slice):
step = idx.step or 1
start = idx.start if idx.start is not None else 1 if step > 0 else -1
start = idx.start if idx.start is not None else 0 if step > 0 else -1
stop = idx.stop if idx.stop is not None else -1 if step > 0 else 0
if self.__reversed:
start, stop, step = map(self._reverse_index, (start, stop, step))
idx = slice(start, stop, step)
elif isinstance(idx, int):
if self.__reversed:
idx = self._reverse_index(idx)
start = stop = idx
else:
raise TypeError('indices must be integers or slices')
if start < 0 or stop < 0:
# We need to consume the entire iterable to be able to slice from the end
# Obviously, never use this with infinite iterables
self.exhaust()
else:
n = max(start, stop) - len(self.__cache) + 1
if n > 0:
self.__cache.extend(itertools.islice(self.__iterable, n))
return self.exhaust()[idx]
n = max(start, stop) - len(self.__cache) + 1
if n > 0:
self.__cache.extend(itertools.islice(self.__iterable, n))
return self.__cache[idx]
def __bool__(self):
try:
self[0]
self[-1] if self.__reversed else self[0]
except IndexError:
return False
return True
@@ -3994,6 +4027,17 @@ class LazyList(collections.Sequence):
self.exhaust()
return len(self.__cache)
def __reversed__(self):
self.__reversed = not self.__reversed
return self
def __repr__(self):
# repr and str should mimic a list. So we exhaust the iterable
return repr(self.exhaust())
def __str__(self):
return repr(self.exhaust())
class PagedList(object):
def __len__(self):
@@ -6202,6 +6246,8 @@ def traverse_obj(obj, keys, *, casesense=True, is_user_input=False, traverse_str
if is_user_input:
key = (int_or_none(key) if ':' not in key
else slice(*map(int_or_none, key.split(':'))))
if key is None:
return None
if not isinstance(obj, (list, tuple)):
if traverse_string:
obj = compat_str(obj)

View File

@@ -1,3 +1,3 @@
from __future__ import unicode_literals
__version__ = '2021.06.01'
__version__ = '2021.06.09'