Is this new provider or existing #565
Labels
No labels
MediaFlow Proxy
autorelease: pending
autorelease: tagged
bug
documentation
duplicate
enhancement
help wanted
invalid
question
wontfix
🇩🇪 German
🇫🇷 French
🇮🇳 Indian
🇸🇦 Arabic
🌐 Multi
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference: Creepso/webstreamr-github#565
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
I found this somewhere
https://bysesayeveum.com/e/8c8tg87ak2awDomain:
byse.sxThe hls links kinda appear like filemoon but adding the domain on filemoon extractor doesnt seems to work...
Or this might be new?
This is definitly new
It was supported on ResolveURL just the domain is new...
Is the F16Px.py
I was able to port it to mediaflow
I will open a pr there
There is anyway to get the height from mediaflow and pass it to webstreamr?
no, atm it is not. which is why I have this height guesser method for hls streams here. for mp4 files it's not doable anyway in an efficient way
Your height guesser works if you get the master through addon but what if you get the master from mediaflow(proxied)? Because the link is encrypted with AES and is not visible on html...
For example
https://github.com/mhdzumair/mediaflow-proxy/pull/185
Also why not pass the height on headers for example? And detect it on addon !
it works fine via proxy too of course, that doesn't change anything. but I didn't check your extractor there, so can't say much. in general, since I want to avoid doing requests to the stream/mfp it would be nice if it can be parsed though.
a solution where it is passed from mfp sounds interesting. in general though, I don't want to do even a single MFP request when creating the results if I can to be scalable. but because of more stremio Android issues this has to be currently done unfortunately: https://github.com/Stremio/stremio-bugs/issues/1574
summary: my goal is to not do any stream request or mfp requests at all and only scrape the size from the source / embed if possible. if that's not possible it currently falls back to parsing the height from the stream and it caches it forever.
Yes got it so we have to do something like this when is not possible to get from source
const height = await guessHeightFromPlaylist(ctx, this.fetcher, proxiedUrl, url);Asking this dev on reddit about the issue you opened for android seems that addressed the issue to the right dev...lets hope they will find a solution
On that photo I am posting about another visual bug and earned a lifetime rpdb key
lol