1524428 - calc() function and tab-size

Status: RESOLVED INVALID

(Core :: CSS Parsing and Computation, defect)

Description

[Gérard Talbot]

Test

http://www.gtalbot.org/BrowserBugsSection/CSS3Values/calc-numbers-temp.html

Actual result

Firefox 60.5.0 ESR and Firefox 67.0a1 buildID=20190131093752 fail the 10 tab-size sub-tests.

Notes

• Chromium 71.0.3578.80 passes the 10 tab-size sub-tests.
• Epiphany passes 9 tab-size sub-tests out of 10.
• The original test is

https://chromium.googlesource.com/chromium/src/+/c825d655f6aaf73484f9d56e9012793f5b9668cc/third_party/WebKit/LayoutTests/css3/calc/calc-numbers.html

• I searched for a duplicate bug report and did not find any.
• I use Linux Debian 9.7 , 64bits.

Comment 1

[Gérard Talbot]

That's Epiphany version 3.22.7 (WebKit 2.18.6).

Comment 2

[Emilio Cobos Álvarez [:emilio]]

We don't support unprefixed tab-size, apparently. Using -moz-tab-size the tests should pass, so this is not related to calc() at all.

Jonathan, any reason to not unprefix -moz-tab-size?

Comment 3

[Jonathan Kew [:jfkthame]]

That's bug 737785, see discussion there for current status.

Comment 4

[Gérard Talbot]

We don't support unprefixed tab-size,

I totally miss the prefix thing.

apparently. Using -moz-tab-size the tests should pass,

The -moz-tab-size: calc(2 * -4) sub-test fails in

http://www.gtalbot.org/BrowserBugsSection/CSS3Values/calc-numbers-temp-with-moz.html

Comment 5

[Emilio Cobos Álvarez [:emilio]]

I think that test is wrong. Per https://drafts.csswg.org/css-values/#calc-range:

Parse-time range-checking of values is not performed within math functions, and therefore out-of-range values do not cause the declaration to become invalid. However, the value resulting from an expression must be clamped to the range allowed in the target context. Clamping is performed on computed values to the extent possible, and also on used values if computation was unable to sufficiently simplify the expression to allow range-checking. (Clamping is not performed on specified values.)

Which means that calc(2 * -4) should not be invalid, and should compute to 0.

Comment 6

[Gérard Talbot]

Good catch, Emilio! I missed that too!
Now, I need to edit the test to take into account this important piece of information.

Comment 7

[Emilio Cobos Álvarez [:emilio]]

Ok, great, thanks! :)

So then I think there's nothing style-system-related to fix here. Just unprefixing -moz-tab-size which is bug 737785.