Bugzilla

Adding matrix chat from [#necko:mozilla.org (2024-01-23)](https://matrix.to/#/!zrQfbihXkcbZeTYBVd:mozilla.org/$6LVawjkDfpu30FxoVUpQMb9pzUOpinZTERxarsxvUFI?via=mozilla.org&via=matrix.org&via=igalia.com):

erosman:

> Any chance of working on Bug 1794464?
>
> There might not be much (if any) work needed by the necko except agreeing to remove the restriction.
> WebExtension API can then be updated (remove one throw) which should be easy enough.
> 
> It is causing complication for example... https://github.com/foxyproxy/browser-extension/issues/76#issuecomment-1906398346

manuel:

> Do you feel comfortable preparing the patch yourself? The priority P3 indicates that we would accept patches enabling the behavior. It is currently not on our radar of bugs that we work on, though. I think this is the fastest way of getting it fixed would be to write the patch.
> It seems simple and a good bug candidate for a first bug to work on. I'd mentor you if you need help.

erosman:

> If I understand correctly, AFA `proxy.onRequest`, all that is needed is the removal of https://searchfox.org/mozilla-central/source/toolkit/components/extensions/ProxyChannelFilter.sys.mjs#177-181
>
> I have never done a patch for Firefox, and I am concerned about testing which is compiling and providing test cases. These are processes that I am not familiar with. Otherwise, providing a PR with those 5 lines removed is not difficult.

manuel:

> Test would make the patch harder indeed. Not sure right now which ones exists. Will check tomorrow. And I'm not sure if we still need to exclude SOCKS4(/5) proxies there. But will also take a look tomorrow.

I didn't take a look yet how the current test situation looks like and how hard creating tests would be. Adding the conversation from matrix, because it is hard to find for me and I am better at keeping track of stuff in bugzilla.

Adding matrix chat from [#necko:mozilla.org (2024-01-23)](https://matrix.to/#/!zrQfbihXkcbZeTYBVd:mozilla.org/$6LVawjkDfpu30FxoVUpQMb9pzUOpinZTERxarsxvUFI?via=mozilla.org&via=matrix.org&via=igalia.com):

erosman:

> Any chance of working on Bug 1794464?
>
> There might not be much (if any) work needed by the necko except agreeing to remove the restriction.
> WebExtension API can then be updated (remove one throw) which should be easy enough.
> 
> It is causing complication for example... https://github.com/foxyproxy/browser-extension/issues/76#issuecomment-1906398346

manuel:

> Do you feel comfortable preparing the patch yourself? The priority P3 indicates that we would accept patches enabling the behavior. It is currently not on our radar of bugs that we work on, though. I think this is the fastest way of getting it fixed would be to write the patch.
> It seems simple and a good bug candidate for a first bug to work on. I'd mentor you if you need help.

erosman:

> If I understand correctly, AFA `proxy.onRequest`, all that is needed is the removal of https://searchfox.org/mozilla-central/source/toolkit/components/extensions/ProxyChannelFilter.sys.mjs#177-181
>
> I have never done a patch for Firefox, and I am concerned about testing which is compiling and providing test cases. These are processes that I am not familiar with. Otherwise, providing a PR with those 5 lines removed is not difficult.

manuel:

> Test would make the patch harder indeed. Not sure right now which ones exists. Will check tomorrow. And I'm not sure if we still need to exclude SOCKS4(/5) proxies there. But will also take a look tomorrow.

erosman:

> `proxyAuthorizationHeader` is sent for the HTTP/S types as...
> ```
> `'Basic ' + btoa(username + ':' + password)
> ```
>
> I don't know how the user/pass is sent for SOCKS.


I didn't take a look yet how the current test situation looks like and how hard creating tests would be. Adding the conversation from matrix, because it is hard to find for me and I am better at keeping track of stuff in bugzilla.

Added comment to wrong bug. Moved it to [1794464 comment 6](https://bugzilla.mozilla.org/show_bug.cgi?id=1794464#c6)