1885536 - Border-collapsed table width not correctly adapting to zoom reset

Status: RESOLVED DUPLICATE of bug 1834487

(Core :: Layout, defect, P3)

Description

[stn]

Attachment: Example

User Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:123.0) Gecko/20100101 Firefox/123.0

Steps to reproduce:

I zoomed in on a page up to 500%, then I reset the zoom to 100%.

Actual results:

The border of a table correctly adapted its thickness while zooming in, but after resetting the border remained as wide as it was on the previous zoom level.

Expected results:

The border should have looked like before zooming.

I have found out that this bug only occurs when "border-collapse: collapse;" is set on the table. I have attached a minimal example which can be used to reproduce the bug by following the steps described above.

Regards,
stn

Comment 1

[BugBot [:suhaib / :marco/ :calixte]]

The Bugbug bot thinks this bug should belong to the 'Core::Layout' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.

Comment 2

[Alice0775 White]

I can reproduce this on Nightly125.0a1 Unbuntu22.04 [Wayland and XWayland] and LinxMint Cinnamon[X11].

Regression window:
https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=db4607bbd6d1046f430c2c0b7dbbefba88dc7f78&tochange=0f75cd1f132cae02560f8736469b3fc801fdccc1

Comment 3

[BugBot [:suhaib / :marco/ :calixte]]

:emilio, since you are the author of the regressor, bug 1834487, could you take a look? Also, could you set the severity field?

For more information, please visit BugBot documentation.

Comment 4

[Alice0775 White]

And I can also reproduce this on Nightly125.0a1 Windows11 if Window Device Pixel Ratios is 1.

Comment 5

[Emilio Cobos Álvarez (:emilio)]

So this is because before bug 1834487, a device pixel ratio change would've triggered CalcStyleDifference to return _Reflow, which caused this code to return true.

I can add a check for dpi change to BCRecalcNeeded but the real problem is that tables store borders as device pixels, which I thought we fixed, but turns out that it bounced (that is bug 1825384). That seems like the right fix. David, do you know what the state of bug 1825384 is, specially given your newer bc fixes? I'd rather help with that than paper it here.

Comment 6

[David Shin[:dshin]]

Confirming that after bug 1825384 fixes this (Checked via today's Nightly build)