1362499 - Support unit-less values for CSS -webkit-perspective (for interop)

Status: RESOLVED FIXED

(Core :: CSS Parsing and Computation, enhancement, P3)

Description

[Markus Stange [:mstange]]

One of the first 3d transform demos, https://webkit.org/blog-files/3d-transforms/morphing-cubes.html , does not work correctly in Firefox because the #container element has -webkit-perspective: 800; with a unit-less number.

Comment 1

[Daniel Holbert [:dholbert]]

Reclassifying as CSS Parsing & Computation -- the fix here would be in the CSS Parser.

Comment 2

[Daniel Holbert [:dholbert]]

Attachment: testcase 1

Comment 3

[Cameron McCormack (:heycam)]

Mike, is this something that should be in the Compat spec?

Comment 4

[Mike Taylor [:miketaylr]]

(In reply to Cameron McCormack (:heycam) from comment #3)
> Mike, is this something that should be in the Compat spec?

Yeah, if need to implement it, it should be. I'll file a bug (if one doesn't already exist).

Comment 5

[Xidorn Quan [:xidorn] UTC+10]

If we want to implement this, it can be a good first bug I think.

Comment 6

[Emilio Cobos Álvarez [:emilio]]

Hold on a second, do we really want to implement this just because of a webkit.org demo is broken? Or do we have any actual evidence of it mattering?

The page renders equally badly in Blink-based browsers.

Comment 7

[Emilio Cobos Álvarez [:emilio]]

Err, disregard my previous comment. Still I think it's not worth unless we find a real website which is broken that isn't webkit.org.

Eric, looks like Chromium still parses -webkit-perspective: <integer>. Does the Chrome team have any use counter for it by any chance?

Comment 8

[Eric Willigers]

Chrome does have a use counter:
unitless -webkit-perspective non-zero length
https://www.chromestatus.com/metrics/feature/timeline/popularity/1671

Comment 9

[Emilio Cobos Álvarez [:emilio]]

(In reply to Eric Willigers from comment #8)
> Chrome does have a use counter:
> unitless -webkit-perspective non-zero length
> https://www.chromestatus.com/metrics/feature/timeline/popularity/1671

Ouch, that use counter is quite high. I guess it doesn't account for fallback, i.e., something like

  -webkit-perspective: 800; perspective: 800px;

But still... :)

Comment 10

[Emilio Cobos Álvarez [:emilio]]

I'd prefer to at least get a single report of a broken page because of this before going ahead and implementing it, but I guess if it breaks websites we'd need to implement this quirk...

Comment 11

[Markus Stange [:mstange]]

I agree. I filed this bug only based on the Webkit demo and I haven't seen breakage caused by it elsewhere.

Comment 12

[Mike Taylor [:miketaylr]]

Karl, have you seen any real-world bustage related to this?

Comment 13

[Karl Dubost💡 :karlcow]

The two places I thought about were:
https://www.google.com/logos/2014/rubiks/iframe/
https://zacqary.github.io/rubix

but no issue. 


so… 0 bustage to my knowledge and/or memory

Comment 15

[Chris Mills [:cmills]]

Depending on how this goes, we might need to make sure this is documented properly.

Comment 16

[Firefox Bug Husbandry Bot]

Migrating Webcompat whiteboard priorities to project flags. See bug 1547409.

Comment 17

[Firefox Bug Husbandry Bot]

See bug 1547409. Migrating whiteboard priority tags to program flags.

Comment 18

[Mike Taylor [:miketaylr]]

It seems like we found a real site relying on this, but they updated it and it's working now (https://webcompat.com/issues/21898).

Comment 19

[Cheng-You Bai [:cybai] UTC+9]

hi there, I notice this bug while trying to find a good-first-issue and I'd like to work on it though, with reading through this bugzilla and the whatwg/compat issue, I'm not sure if Firefox would like to support unit-less values for -webkit-perspective. May I confirm it before trying to work on it? Thank you.

Comment 20

[Sick Leviathan]

https://www.chromestatus.com/metrics/feature/timeline/popularity/1671
use counter is down to less than 0.6% from the 10% reported in 2017-2019, should we still fix this? if so how do we make -webkit- variant have custom behavior different from the standard perspective? Is there an example in the servo folder for a similar CSS property?

Comment 21

[Emilio Cobos Álvarez [:emilio]]

We'd have to make -webkit-perspective a shorthand of perspective rather than a plain alias, and accept unitless units there. It's not hard but given we haven't seen compat bugs due to this in 5 years I think it's probably not worth it.

Comment 22

[Sick Leviathan]

Attachment: Bug 1362499 - Add compat support for unitless perspective for webkit transform and perspective properties. r?emilio

Comment 23

[Sick Leviathan]

Attachment: Bug 1362499 - Added tests for invalid tests for transform and perspective, and valid ones for their webkit counterparts. r?emilio

Comment 24

[Pulsebot]

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/b15cf7f3d998
https://hg.mozilla.org/integration/autoland/rev/02434507f7aa
Add compat support for unitless perspective for webkit transform and perspective properties. r=emilio,firefox-style-system-reviewers

Comment 25

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Created web-platform-tests PR https://github.com/web-platform-tests/wpt/pull/56336 for changes under testing/web-platform/tests

Comment 26

[Serban Stanca [:SerbanS]]

https://hg.mozilla.org/mozilla-central/rev/02434507f7aa

Comment 27

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Upstream PR merged by moz-wptsync-bot

Comment 28

[Chris Mills [:cmills]]

I've added a Firefox 147 rel note to cover this change on MDN, in https://github.com/mdn/content/pull/42275. I don't think the actual perspective documentation needs any changes.