imputnet / cobalt Goto Github PK
View Code? Open in Web Editor NEWsave what you love
Home Page: https://cobalt.tools
License: GNU Affero General Public License v3.0
save what you love
Home Page: https://cobalt.tools
License: GNU Affero General Public License v3.0
when i trying to downloading video from any tweets, it just says "an error occurred when i tried to get info about your link."
i guess time to check twitter parser lol
video proof:
https://files.catbox.moe/7zln7c.mp4
no idea if this is caused by cobalt itself or my nginx configuration, but after a few days way more ram is used by the server that cobalt is hosted on. i'll have a look into this issue in a few days (if ram usage ramps up again). if you host your own instance, you can help by checking if this issue applies to you too.
i downloaded a youtube video and tried to trim it using the windows video trimmer, it wouldnt let me trim it. also this happened
(i dowloaded it as a max-quality mp4)
if i tried to import it into any other video editor it shows an error
when i try to view the file's properties it shows a null length, and 0kbps audio??? im not exactly an ffmpeg master so i dont exactly know what causes this
Facebook URLs would be useful for downloading content without an account or having to open an app to view.
Example URL formats:
https://fb.watch/gsa6WvY0sF/
https://www.facebook.com/buzzfeedapop/videos/1061711167873198/?extid=NS-UNK-UNK-UNK-AN_GK0T-GK1C&ref=sharing
https://www.facebook.com/reel/606490164289064?extid=a&s=yWDuG2&fs=e
no idea what's the problem, whether it's cobalt or something on my end, but i've attached a video describing it
thank u in advance xx
Current and Crowdin (3.5.2). Test on YouTube and VK (audio, video):
Trying to download from YouTube resulted in a warning:
After trying to update: "npm install npm@latest -g". There is no result, no matter what the order. Problem with code or with npm 8.19.2/node 18.9.0, maybe this version does not work correctly with this code.
Your service on co.wukko.me is now work. And yes turn that shit off: https://github.com/wukko/cobalt/blob/current/src/modules/sub/currentCommit.js
For streamDefault just pass through the Content-Length header, for the rest parsing the ffmpeg output might be required. I can send a PR, do you have example URLs/settings for all the different types?
Hello,
I tried going to https://co.wukko.me/ and I pasted the link https://www.youtube.com/watch?v=OvhVLmD0DfQ
with auto mode on and I got the error saying that I have sent too many requests. However, the issue is that I haven't even used the site yet. I have visited the site before to check it out and see how it worked but never used it to download anything. And no, I didn't spam the button, I pressed it once.
Hope this issue is resolved soon, this site looks promising!
how to reproduce:
these tracks are able to be downloaded, as other sites i tested had no problems.
This will be useful because other downloaders are hard to find and it might just not work.
NicoNico video pages has authorization data for guest user in HTML attribute with id="js-initial-watch-data"
.
Front-end web sends to POST request to https://api.dmc.nico/api/sessions?_format=json
(The DMC service; Modern delivery system) with JSON body of some data from authorization data.
The DMC service will respond with authorized URL to the master M3U8. And yes, NicoNico uses HLS VOD...
This is working perfectly with Reddit videos. They're a real pain to download since they split the audio in a separate file.
What would make this really great is a window to paste some urls and let it go one by one with some custom wait time between them.
Could it be done?
An error occurs trying to download videos from the "Men Without Hats - Topic" channel. This was tested with "The Safety Dance" (https://youtu.be/nft1OVNLM7A), "All Into Stars" (https://youtu.be/sbSupcsaT80), and every other video I personally tried. There don't seem to be any issues with videos from other "Topic" channels, like https://youtu.be/w44au_CEfmI.
Just a question: can it be possible to use https://crowdin.com for translation for next time? Would be nice instead of edition json files directly
Support for Twitch Clips and Videos would be amazing.
Clip URL Format: https://www.twitch.tv/tpain/clip/TenderSuspiciousPotatoRaccAttack-L3MAL-BTve06zRqW
Video/Stream URL Format: https://www.twitch.tv/videos/1681039790
Thanks for an amazing tool!
I wanna download my spotify songs! (it would be awesome!!!)
https://youtu.be/JMkrrjKf5AE
this is the link for it. It's just this video. I've tried it multiple times.
On mobile, I have been constantly facing an issue when I try to download any youtube video that’s longer than 1 minute to an mp4. I find that usually after the first half, the video is suddenly missing its audio.
I have yet to check if I face the same issue on pc, so keep that in mind. If needed, I could go check when I have time.
Tested it locally on latest version, and also manually tested script part for tiktok.
Looks like it gives an error on step when you're trying to get html body of post page, before searching of file src.
Also I've found something strange in response
rejectUnauthorized: true,
session: undefined,
Maybe now we need to login before trying to find source link.
Any ideas?
I have my own instance hosted here: https://co.hyper.lol
YouTube videos/audios don't seem to work, and always result in a HTTP 400 error. Not sure what I can look into for this.
I am running commit 9b17300.
Any video I try results in this. Could it be something with Cloudflare?
Previously, Cobalt allowed you to download images, videos and mixed media from any public account. Then, downloading mixed media bugged out—only allowing the first image or video in a series of four. Then, now it just completely disallows download from age-restricted accounts even if they're public.
The error message is:
uh-oh...
an error occurred when i tried to get info about your link. are you sure it works? check if it does, and try again.
Here's an example link, from a person uploading "safe" samples of his paid content: https://twitter.com/leathersambam88/status/1589098906616745986 - Try loading his profile in incognito/private mode and it would show the age-restricted message. Cobalt used to be able to download these just fine.
I'm not sure if this is intentional or not, but trying to access an API endpoint at co.wukko.me via JavaScript (in this case, within a Firefox extension) will throw a CORS error, probably due to the header access-control-allow-origin: https://co.wukko.me/
. I'm not sure, but I'd assume this is not intended behavior for an open API.
Video in question: https://www.youtube.com/watch?v=MDsJJRNXjYI
suggested approach is to use the next best thing (vp9, h264)
Sometimes my friends send me a vxtwitter or fxtwitter URL, like this one
https://vxtwitter.com/RayaanTheAlien/status/1455029356032499714
It's basically only there to make the video embed better on Discord, and I'd assume cobalt could be able to handle these like it handles normal Twitter URLs.
P.S. I've been using cobalt for a while, and I absolutely love it! Thank you for your work!
When downloading ぼっちロック in video mode, its audio cuts out at around the 1-minute mark. It isn't consistent, sometimes it happens earlier, sometimes later (differs by only a couple of seconds) and it doesn't happen when downloading it in audio mode. This video seems to have some issues on youtube itself, since (at least for me) also around the 1-minute mark the video briefly stops and buffers before resuming.
https://cdn.syndication.twimg.com/tweet?id= Afaik this endpoint isn't rate limited & doesn't require a bearer token, may be better to use for this site
Example response from that url: https://upload.systems/p/1z9jHMKO8u7V
:)
seem reddit not working . anyone fix it ?
Test both on my own instance (running af1fc24) and on https://co.wukko.me.
Tested with this link: (and many other random videos) https://twitter.com/FreelanceMocha/status/1576304807794552833, returns 400.
https://co.wukko.me/api/json?audioFormat=best&quality=max&url=https%3A%2F%2Ftwitter.com%2FFreelanceMocha%2Fstatus%2F1576304807794552833
https://co.hyper.lol/api/json?audioFormat=best&quality=max&url=https%3A%2F%2Ftwitter.com%2FFreelanceMocha%2Fstatus%2F1576304807794552833
this will make easier to follow the code style for contributors.
https://editorconfig.org/
I can't seem to find an option to switch languages in the settings and the process of switching languages appears to be entirely bound to the system locale. An option to manually pick a language would be nice.
GNU General Public License v3.0 is great open-source license, of course!
But in your case better to use GNU Affero General Public License v3.0!
You can read explanation here:
https://www.gnu.org/licenses/why-affero-gpl.en.html
https://www.gnu.org/licenses/why-affero-gpl.ru.html
For some reason, any time i download a video from youtube, the video is registered as extremely long, for example, one video i had downloaded was listed as being 256204778 hours long (or 29247 years). While the video file does not take up much space, on average being about 1.5 mb for a 15 second video, it is impossible to share it to any other services without getting a file size warning.
MP4's that I download from Youtube won't play in QuickTime and can't be saved to the Photos app. VLC can play it so it's not completely broken, but there's something up with the encoding that's causing issues.
I tried with bloody mary sped up but it stopped at 260 kilobytyes and it never continued
would be nice if it can have an option in the settings to embed the link directly in a new tab instead of downloading (via setting the Content-Disposition to inline)
When making a POST request with the following data:
{"url": "https://www.youtube.com/watch?v=[url]"}
it returns 200 OK with the following data:
{
"status": "stream",
"url": "https://co.wukko.me/api/stream?t=[...]&e=[...]&h=[...]"
}
Opening that URL returns this:
{"status":"error","text":"Unauthorized"}
After downloading 5 songs from soundcloud with no problems it started to only download the first 30 seconds. I have tried multiple different songs in both smart and audio only mode as well as different audio formats and nothing seems to work.
I would like to contribute by translating cobalt into Croatian
only possible reason i can think of is download speed or duration. don't know whether it's possible for ffmpeg to pre-fetch one of inputs before merging them together, but this would be one of possible solutions.
cobalt could instead cache the audio file before live render, but that could lead to even more issues.
TikTok links can include the username in them, and usernames can include dots, seems like cobalt breaks trying to recognize them as tiktok links and throws an "it seems like this service is not supported yet or your link is invalid."
example link: https://www.tiktok.com/@random.video.on.tiktok/video/7090355705605016859
also, as a band-aid fix, seems like replacing the username with something else without the dots makes it work on cobalt, as tiktok redirects to the good one.
https://www.tiktok.com/@a/video/7090355705605016859
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.