Hacker Newsnew | past | comments | ask | show | jobs | submit | web0_cc's commentslogin

Another weekend hackathon project. Very basic prototype ATM.

If there is demand, i will add features like top rank timestamp, custom theme and widgets like live score.

Thanks


yt-dlp ${youtube_video_url} --get-url

ffmpeg -ss ${seconds} -i "${first_url_from_yt-dlp}" -t 1 -r 4/1 -q:v 2 -vf scale="0:-1" ${target}

-r 4/1: 4 frames per second

Again this app is for mobile devices which lack yt-dlp+ffmpeg


Nice, if you install termux (from fdroid) you can run it directly on your android.


Is it possible to get one screenshot from different periods of the same video like 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80% with just one command.

If yes, could you please share for me here, it'd be really helpful, thanks


Does this really not download the whole video? I havent checked but im just surprised that it's possible


No it doesn't. FFMPEG can seek video directly without loading all the content, even for Dash/HLS (but in this case yt-dlp typically returns a plain HTTP source, which I assume internally FFMPEG uses HTTP header range to seek?)

Keep in mind you need to make sure to use input seeking, though (-ss before -i).

Also I'm not sure why they need `-vf scale="0:-1"` part.


Thats true. its for mobile devices which lack desktop browser features and extentsions


I started working on this few days ago so havent had much time to think about paywalled content. Maybe this week i will try to find a solution.

There wont be much we could do If its behind captchas.

Thanks for giving it a try.


Formula i used is Ceil(CL/WL)/WPM)

CL= Content Length WP= Average Word Length (i used 5) WPM = Words per min, reading speed (i used 180)

not perfect but good enough.

Thanks breck for giving the app a try


Yea. due to lazy loading. Fix is simple but it comes at a cost

Thanks for finding the bug.


Thanks for the feedback & suggestions.

>> Nice. How about creating a bookmarklet to make it super easy to share a web0.cc link from the current page?

Sure, i will do it

>> Also maybe if the refresh can be reliably done after a few seconds then you could make the page refresh itself automatically (it's not difficult for the user but it's one less thing to handle). You could tweak the refresh time by how busy your site is.

Thats a good feature but it involves JS and for some reasons i decided not to put JS on the client side.


> Thats a good feature but it involves JS

Actually, you don’t need JS. The simplest mechanism is probably:

https://en.m.wikipedia.org/wiki/Meta_refresh

(Both HTML tag and HTTP header routes would probably work)


Thats good.

Honest question, Would you enjoy reading articles from archive.is on your mobile phone?

as far as i remember, its not mobile-friendly.


It is mobile friendly towards reader-mode of iPhone Safari which is more than enough for me since I have reader-mode on by default for articles.

But I can agree that it doesn't render mobile-friendly websites by default.


My bad.

First of all thank you all for the feedback & suggestions.

I think these all are good suggestions. I will pick one for sure


Thanks for checking the app, SamBam.

Do you mean like this https://web0.cc/a/zTHtZFEXci?url=https://www.nytimes.com/202...


That's good, although anyone can put anything there.

https://web0.cc/a/zTHtZFEXci?url=https://haxx0r.c0m

I was thinking more along the lines of

https://web0.cc/a/opinion/english-literature-study-zTHtZFEXc...

Of course it doesn't actually grant you security.

What if your original link with the `url=` refused to load if the target didn't match the query parameter? That would actually make it secure, no?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: