857536 - Cleanup the CSS properties applied on <input type='file'> from forms.css

Status: RESOLVED FIXED

(Core :: Layout: Form Controls, defect)

Description

[Mounir Lamouri (:mounir)]

Attachment: Patch

We have some properties applying on |input|, some on |input[type='file']| and for some reasons, we don't revert all of them as expected. This should cleanup most things. It will make some bugs like border and background being ignored more visible because they will no longer apply to the label. They should apply to the element. This is another bug.

It should fixes the fact that the label had a background by default (inherited from a |input| property).

Comment 1

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 732767 [details] [diff] [review]
Patch

>+  background-color: none;

This isn't valid CSS.  You meant "transparent", right?

r=me with that.

Comment 2

[Mounir Lamouri (:mounir)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/2a6c83f2499e

Comment 3

[Mounir Lamouri (:mounir)]

(In reply to Boris Zbarsky (:bz) from comment #1)
> Comment on attachment 732767 [details] [diff] [review]
> Patch
> 
> >+  background-color: none;
> 
> This isn't valid CSS.  You meant "transparent", right?

Yes, and that was fixed by the next patch in my queue actually.

Comment 4

[Boris Zbarsky [:bzbarsky]]

> Yes, and that was fixed by the next patch in my queue actually.

Er... yes, but why did you check in the broken code, which will cause spammy warnings, in anyway?

Comment 5

[Mounir Lamouri (:mounir)]

... I *again* forgot to qref before importing in my m-i repository :( Will push a follow-up.

Comment 6

[Mounir Lamouri (:mounir)]

Oh no. I actually just forgot to update the test. That's less important I guess. My next patch should fix it anyway.

Comment 7

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/2a6c83f2499e

Comment 8

[Mounir Lamouri (:mounir)]

Comment on attachment 732767 [details] [diff] [review]
Patch

[Approval Request Comment]
This is required to push bug 52500. I thought this patch was landed earlier.
The user impact is similar to bug 52500 then and the risk is also similar. In the way that this patch isn't more risky than the one in bug 52500 and the user impact isn't going to be more than not able to psuh bug 52500 if this patch don't get uplifted.

(FWIW, it is in Aurora)

Comment 9

[Mounir Lamouri (:mounir)]

https://hg.mozilla.org/releases/mozilla-beta/rev/c70ddab527a3

Comment 10

[Mihaela Velimiroviciu (:mihaelav)]

Given the automation coverage for this bug, is there anything else QA can do to manually verify this? If so, please provide some guidelines.

Comment 11

[Mounir Lamouri (:mounir)]

(In reply to Mihaela Velimiroviciu [QA] (:mihaelav) from comment #10)
> Given the automation coverage for this bug, is there anything else QA can do
> to manually verify this? If so, please provide some guidelines.

I think we are good here. Thank you :)