556855 - Snow Leopard packaged unittests failing test_platform_colors.xul

Status: RESOLVED FIXED

(Core :: Widget: Cocoa, defect)

Description

[Lukas Blakk [:lsblakk] use ?needinfo]

Consistent failure:

10056 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/widget/tests/test_platform_colors.xul | platform color highlighttext is wrong: rgb(254, 255, 254)
testing color inactiveborder

10062 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/widget/tests/test_platform_colors.xul | platform color menu is wrong: rgb(254, 255, 254)
testing color menutext

10099 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/widget/tests/test_platform_colors.xul | platform color -moz-menuhovertext is wrong: rgb(254, 255, 254)
testing color -moz-menubartext

Comment 1

[David Dahl :ddahl]

http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1270240441.1270241786.26993.gz
Rev3 MacOSX Snow Leopard 10.6.2 mozilla-central debug test mochitest-other [testfailed] Started 12:51, finished 13:18

Comment 2

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1270264762.1270266132.24040.gz
Rev3 MacOSX Snow Leopard 10.6.2 mozilla-central debug test mochitest-other on 2010/04/02 20:19:22
s: talos-r3-snow-018
10076 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/widget/tests/test_platform_colors.xul | platform color highlighttext is wrong: rgb(254, 255, 254)
10082 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/widget/tests/test_platform_colors.xul | platform color menu is wrong: rgb(254, 255, 254)
10119 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/widget/tests/test_platform_colors.xul | platform color -moz-menuhovertext is wrong: rgb(254, 255, 254)

Comment 3

[Kyle Huey (Exited; not receiving bugmail, old account, do not use)]

http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1270281116.1270282731.23471.gz

Comment 4

[Jonathan Kew [:jfkthame]]

http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1270298796.1270300293.18641.gz

Comment 5

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1270407528.1270409079.27191.gz

Comment 6

[Daniel Holbert [:dholbert]]

http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1270530411.1270531167.9714.gz
Rev3 MacOSX Snow Leopard 10.6.2 mozilla-central opt test mochitest-other on 2010/04/05 22:06:51
s: talos-r3-snow-018

Comment 7

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

I marked the colors that were showing up as legal:
http://hg.mozilla.org/mozilla-central/rev/83491775950c

It would be good if somebody checks that that makes sense.

I'm a bit suspicious of the numbers behaving the way they are, though... could there be some sort of rounding bug in our interaction with color management or something?

Comment 8

[Markus Stange [:mstange]]

(In reply to comment #7)
> I marked the colors that were showing up as legal:
> http://hg.mozilla.org/mozilla-central/rev/83491775950c

Thanks!

> It would be good if somebody checks that that makes sense.

I think it does. If those colors are the ones that the system gives us, they're by definition correct.

> I'm a bit suspicious of the numbers behaving the way they are, though... could
> there be some sort of rounding bug in our interaction with color management or
> something?

Hmm. The code is:

>   NSColor* deviceColor = [aColor colorUsingColorSpaceName:NSDeviceRGBColorSpace];
>   return NS_RGB((unsigned int)([deviceColor redComponent] * 255.0),
>                 (unsigned int)([deviceColor greenComponent] * 255.0),
>                 (unsigned int)([deviceColor blueComponent] * 255.0));

We should check whether using rounding makes any difference.