[XVideos] Support playlists, searches and channels

[dirkf]

Please follow the guide below

---

Before submitting a pull request make sure you have:

• Searched the bugtracker for similar pull requests
• Read adding new extractor tutorial
• Read youtube-dl coding conventions and adjusted the code to meet them
• Covered the code with tests (note that PRs without tests will be REJECTED)
• Checked the code with flake8

In order to be accepted and merged into youtube-dl each piece of code must be in public domain or released under Unlicense. Check one of the following options:

• I am the original author of this code and I am willing to release it under Unlicense, except for code from yt-dlp for which either this or the below has been separately asserted
• I am not the original author of this code but it is in public domain or released under Unlicense (provide reliable evidence)

What is the purpose of your pull request?

• Bug fix
• Improvement
• New extractor
• New feature

---

Description of your pull request and other information

The XVideos extractor supported single videos presented via various URL formats. The site also offers various playlist-like pages:

• actual playlists: /favorite/ID/SLUG
• related videos/playlist: video URL + #_related-...
• channel activity: /...channels/ID
• channel videos: channel URL + #_tabVideos
• channel favourites (playlist): channel URL + #_tabFavorites
• search results: ...?k=(ID).

This PR tries to support extracting from those pages. It also pulls in small changes from the yt-dlp extractor.

To do:

• more tests
• simplify?

[afterdelight]

pls merge

[dirkf]

Before I merge it, please test it further if you can. Only a few URLs were used to develop the new extractor functions.

[afterdelight]

it works for channel, except the uploader output, -o "%(uploader)s" created a folder named NA and not the uploader's name

[dirkf]

These metadata fields are in the hydration JSON for, eg, https://www.xvideos.com/video64379435/verification_video:

    'uploader_id': 505962135,
    'uploader': 'tianmeichuanmei',

#30689 is supposed to

• extract uploader, uploader_id and uploader_url

I guess this uses those fields.

With just this PR, use %(playlist_id)s to get the uploader.

[afterdelight]

ok then will this PR merged with the other PR?

[xspish]

Hey guys! Here just to comment that it would be awesome to have channel dl support, as well as uploader/uploader_id as --output option for xvideos.

I've been doing all sorts of workarounds with other apps to get xvideos dls correctly, in yt-dlp above options work flawlessly on pornhub, but would be awesome to get them for xvideos too

thanks and I hope it gets implemented soon :)

[zapper]

There is some issue once the error "requested format not available" is thrown during download of "#_tabVideos".

[download] Downloading video 12 of 575
[XVideos] 59824163: Downloading webpage
[XVideos] 59824163: Downloading m3u8 information
[XVideos] 59824163: Checking hls-480p video format URL
[XVideos] 59824163: Checking hls-720p video format URL
[XVideos] 59824163: Checking hls-360p video format URL
[XVideos] 59824163: Checking hls-250p video format URL
ERROR: requested format not available
Traceback (most recent call last):
  File "/home/zap/git/youtube-dl_PR_30774/youtube_dl/YoutubeDL.py", line 816, in wrapper
    return func(self, *args, **kwargs)
  File "/home/zap/git/youtube-dl_PR_30774/youtube_dl/YoutubeDL.py", line 848, in __extract_info
    return self.process_ie_result(ie_result, download, extra_info)
  File "/home/zap/git/youtube-dl_PR_30774/youtube_dl/YoutubeDL.py", line 882, in process_ie_result
    return self.process_video_result(ie_result, download=download)
  File "/home/zap/git/youtube-dl_PR_30774/youtube_dl/YoutubeDL.py", line 1684, in process_video_result
    raise ExtractorError('requested format not available',
youtube_dl.utils.ExtractorError: requested format not available

Then the download is interrupted... Is this expected behaviour?

The used command line options are:

-f "bestvideo[height>=2160]+bestaudio/best[height>=2160]/bestvideo[height>=1440]+bestaudio/best[height>=1440]/bestvideo[height>=1080]+bestaudio/best[height>=1080]/bestvideo[height>=720]+bestaudio/best[height>=720]" --ignore-errors --no-progress --playlist-random --verbose --download-archive .download_archive

Also i can confirm the id's are being successfully stored for --download-archive

[dirkf]

Surely your format selection is equivalent to (bestvideo+bestaudio/best)[height >=? 720]

Post the full verbose log including the URL: perhaps use --playlist-items 12 to get the problem item directly, if only you weren't using --playlist-random.

[zapper]

I found the issue - the root cause is actually sitting in front of the keyboard.

So I can confirm that your PR works like a charm.

There was a bug in my wrapper script which took the command line arguments in a variable which was wrongly masked and therefore not correctly passed to the "-f" parameter:

[debug] Command-line args: ['-f bestvideo[height>=2160]+bestaudio/best[height>=2160]/bestvideo[height>=1440]+bestaudio/best[height>=1440]/bestvideo[height>=1080]+bestaudio/best[height>=1080]/bestvideo[height>=720]+bestaudio/best[height>=720] --ignore-errors --no-progress --playlist-random', '--verbose', '--download-archive', '/srv/x3/Hub/.download_archive_all', 'https://www.xvideos.com/amateur-channels/closeupfantasy#_tabVideos']

Changing my long argument string with your short version visually helped me a lot to find the problem - it's usually better to have everything more readable. Thanks for that. So my issue was absolutely unrelated to your PR.

Sorry for wasting your time - i am looking forward to see your changes soon in the master branch and later in the official binary.

[nicolaasjan]

Tested on channels and that worked, but not on profiles:

youtube-dl -v https://www.xvideos.com/profiles/lewdgamerx1990#
[debug] System config: []
[debug] User config: ['--rm-cache-dir', '-i', '-o', '/dev/shm/test-ytd/%(title)s.%(ext)s', '-f', 'bestvideo[height<=1080][ext=mp4][vcodec^=avc]+bestaudio[ext=m4a]/best[ext=mp4]/best', '--no-mtime', '--embed-thumbnail', '--force-ipv4']
[debug] Custom config: []
[debug] Command-line args: ['-v', 'https://www.xvideos.com/profiles/lewdgamerx1990#']
[debug] Encodings: locale UTF-8, fs utf-8, out utf-8, pref UTF-8
[debug] youtube-dl version 2023.10.03
[debug] Lazy loading extractors enabled
[debug] Single file build
[debug] Python 3.8.10 (CPython x86_64 64bit) - Linux-5.4.0-163-generic-x86_64-with-glibc2.29 - OpenSSL 1.1.1f  31 Mar 2020 - glibc 2.31
[debug] exe versions: ffmpeg N-112061-g654e4b00e2-Nico-20230914, ffprobe N-112061-g654e4b00e2-Nico-20230914, phantomjs 2.1.1, rtmpdump 2.4
[debug] Proxy map: {}
Removing cache dir /home/nico/.cache/youtube-dl ..
[XVideosPlaylist] lewdgamerx1990: Downloading webpage
[download] Downloading playlist: lewdgamerx1990
[XVideosPlaylist] playlist lewdgamerx1990: Collected 0 video ids (downloading 0 of them)
[download] Finished downloading playlist: lewdgamerx1990

[githubafterdark]

I tested this and for me it didn't seem to work with models pages, like for example:

 ~/Desktop/youtube-dl/bin git:(df-xvideos-playlist-patch) ✗ '/home/account/Desktop/youtube-dl/bin/youtube-dl'  https://www.xvideos.com/models/lizvicious1        
[generic] lizvicious1: Requesting header
WARNING: Falling back on generic information extractor.
[generic] lizvicious1: Downloading webpage
[generic] lizvicious1: Extracting information
ERROR: Unsupported URL: https://www.xvideos.com/models/lizvicious1

It also didn't work with #_tabVideos appended either, I don't think I'm doing anything incorrectly:

 ~/Desktop/youtube-dl/bin git:(df-xvideos-playlist-patch) ✗ '/home/account/Desktop/youtube-dl/bin/youtube-dl'  https://www.xvideos.com/models/lizvicious1#_tabVideos
[generic] lizvicious1#_tabVideos: Requesting header
WARNING: Falling back on generic information extractor.
[generic] lizvicious1#_tabVideos: Downloading webpage
[generic] lizvicious1#_tabVideos: Extracting information
ERROR: Unsupported URL: https://www.xvideos.com/models/lizvicious1#_tabVideos

[dirkf]

You may not have installed the working code, which gives this:

$ python -m youtube_dl --flat-playlist 'https://www.xvideos.com/models/lizvicious1#_tabVideos'
[XVideosChannel] lizvicious1: Downloading webpage
[download] Downloading playlist: Lizvicious - Pornstar page (videos,all)
[XVideosChannel] lizvicious1/videos: Downloading webpage
[XVideosChannel] lizvicious1/videos (+1): Downloading webpage
[XVideosChannel] lizvicious1/videos (+2): Downloading webpage
[XVideosChannel] lizvicious1/videos (+3): Downloading webpage
[XVideosChannel] playlist Lizvicious - Pornstar page (videos,all): Downloading 109 videos
[download] Downloading video 1 of 109
[download] Downloading video 2 of 109
[download] Downloading video 3 of 109
...
[download] Downloading video 107 of 109
[download] Downloading video 108 of 109
[download] Downloading video 109 of 109
[download] Finished downloading playlist: Lizvicious - Pornstar page (videos,all)
$

See, eg, https://stackoverflow.com/questions/13561618/pip-how-to-install-a-git-pull-request.

[githubafterdark]

Interesting, I didn't think I did anything incorrectly.

I cloned the repo and checked out the branch and ran make, ran the binary it gave me and it gave me that output, here are the commands I used:

git clone https://github.com/dirkf/youtube-dl/
cd youtube-dl
git checkout df-xvideos-playlist-patch
make

Then I ran it from the bin folder.

With that said, your pip install suggestion worked for me, thanks!

Here is the command I used: pip install git+https://github.com/ytdl-org/youtube-dl.git@refs/pull/30774/head

[dirkf]

After checking out the code, you can run it from the youtube-dl directory (that contains youtube_dl) using python -m youtube_dl ..., where you would normally have youtube-dl ..., so make isn't needed; here python is whatever you use to invoke the Python that wish to use, nowadays typically python3.

[master-leonardo]

Hey, could you please update this code for the new site changes?

[dirkf]

You mean like yt-dlp/yt-dlp#9502 ?

[master-leonardo]

You mean like yt-dlp/yt-dlp#9502 ?

Yes and no.
Yes as in it works for individual downloads.
And no as that change doesn't support profiles, channels, searches etc..

[dirkf]

If anyone would like to document the site changes with regard to the various types of pages that the site supports and the PR extractor is meant to support, that would make it a lot quicker to update.

[Lux-Hue]

If anyone would like to document the site changes with regard to the various types of pages that the site supports and the PR extractor is meant to support, that would make it a lot quicker to update.

The most important change is the formatting of the video links.

Was:
https://www.xvideos.com/video66313831/hottest_car_sex_ever_tesla_autopilot_driving

Is now:
https://www.xvideos.com/video.kfkdepm21e2/hottest_car_sex_ever_tesla_autopilot_driving

For some reason (haven't had much time to dive into the code) youtube-dl extracts the video style URL, which is invalid.

[Lux-Hue]

Way I see it, the issue is here:

extractor/xvideos.py around line 914

def _extract_videos(self, url, playlist_id, num, page):

This is the input of one video's data in the page variable.

{
  "id": 72799231,
  "u": "/prof-video-click/model/creamy-spot1/uopdkffea37/big_dildo_for_creamy_pussy_squirt_watch_full_uncensored_video_in_red_subscription_",
  "i": "https://cdn77-pic.xvideos-cdn.com/videos/thumbs169/28/6d/6f/286d6f37c99ae3ceee9d25ecbe688a59/286d6f37c99ae3ceee9d25ecbe688a59.27.jpg",
  "il": "https://cdn77-pic.xvideos-cdn.com/videos/thumbs169ll/28/6d/6f/286d6f37c99ae3ceee9d25ecbe688a59/286d6f37c99ae3ceee9d25ecbe688a59.27.jpg",
  "if": "https://cdn77-pic.xvideos-cdn.com/videos/thumbs169lll/28/6d/6f/286d6f37c99ae3ceee9d25ecbe688a59/286d6f37c99ae3ceee9d25ecbe688a59.27.jpg",
  "ip": "https://cdn77-pic.xvideos-cdn.com/videos/thumbs169poster/28/6d/6f/286d6f37c99ae3ceee9d25ecbe688a59/286d6f37c99ae3ceee9d25ecbe688a59.27.jpg",
  "c": 10,
  "tf": "Big Dildo for Creamy Pussy Squirt (Watch full uncensored video in RED subscription)",
  "t": "Big Dildo for Creamy Pussy Squirt (Watch full u...",
  "d": "2 min",
  "r": "99%",
  "n": "247.4k",
  "v": false,
  "vim": 0,
  "vv": 0,
  "hm": 1,
  "h": 1,
  "hp": 1,
  "td": 0,
  "fk": 0,
  "ve": 0,
  "ui": 611724437,
  "p": "creamyspot",
  "pn": "Creamyspot",
  "pu": "/creamyspot",
  "ch": true,
  "pm": false,
  "ut": null,
  "iu": false
}

It seems the extractor is using id which has the wrong value, instead it should use the u field which has the uopdkffea37 (My best guess)

Hope this makes things easier