[Bug]: Schemes not respected.

[jmercouris]

Preflight Checklist

• I have read the Contributing Guidelines for this project.
• I agree to follow the Code of Conduct that this project adheres to.
• I have searched the issue tracker for a bug report that matches the one I want to file, without success.

Electron Version

25.4.0

What operating system are you using?

Other Linux

Operating System Version

NixOS

What arch are you using?

x64

Last Known Working Electron version

No response

Expected Behavior

When a scheme is set to stream: true, it should allow streaming of a resource. When a scheme is set to stream: false, it should disable streaming of a resource.

Actual Behavior

The opposite behavior is observed.

Testcase Gist URL

https://gist.github.com/jmercouris/fca7cba9d8e6b30a2f861268a3180cff

Additional Information

No response

[fuwad121]

It seems the streaming behavior is the opposite of what's expected. I think flipping the stream property for each scheme could fix it : true for enable, false for disable in main.js
But, I could be mistaken.

[jmercouris]

Another thing; when I set standard to true and false the dispatch of schemes will change. For example a: will dispatch on a standard scheme, but a:asdf will NOT dispatch on a standard scheme. This is what we may in fact be observing here.

[codebytere]

@jmercouris see #41032 (which you were tagged in i think?)

tl;dr it's a misleading privilege that was implemented confusingly, and should probably be removed instead.

[jmercouris]

I agree, it should probably be removed. However, there is an important point above within this issue that has not been addressed.

When I set standard to true and false the dispatch of schemes will change. For example a: will dispatch on a standard scheme, but a:asdf will NOT dispatch on a standard scheme. This is what we may in fact be observing here.

Please don't close issues unless the problems have been addressed. There is still a very deep bug that dispatch won't happen for 'standard' schemes when the URL appears to be malformed.

[jmercouris]

Perhaps I should open up a new issue for this.

[codebytere]

@jmercouris i hear you but the initial issue is a duplicate of an open one - i folded it into that issue which remains open. the second comment you made:

#41548 (comment)

is a different issue. I can re-open this issue if you change the PR body to reflect that as the primary issue you'd like investigated. You can of course also open a new issue.

[jmercouris]

I've created a new issue #41601