Bug 120668 – [CSS Background] repeat: round should round the number of tiles to the nearest natural number

• Status: RESOLVED
• Resolution: FIXED
• Component: CSS

[Andrei Parvu]

According to the spec [1] repeat: round should round the number of tiles of an image to the nearest natural number greater than 0.

[1] - http://www.w3.org/TR/css3-background/#background-size

[Andrei Parvu]

Created attachment 210439 [details]
Patch

[Andrei Parvu]

Created attachment 211880 [details]
Patch

[Darin Adler]

Comment on attachment 211880 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=211880&action=review

> Source/WebCore/rendering/RenderBoxModelObject.cpp:1129
> +        int nrTiles = static_cast<int>(round((float)positioningAreaSize.width() / fillTileSize.width()));

I believe the round function takes, and returns, a double. So casting to float and then calling round doesn't make good sense. If we want to use float, then it should be roundf, or if we want to stay with double, we should cast to double, not float.

[Darin Adler]

Comment on attachment 211880 [details]
Patch

I’d like to see test cases right on the boundary, demonstrating the rounding behavior. So things just below and just above the place where we round in two different directions. Something that would fail if we used the wrong rounding function.

[Andrei Parvu]

(In reply to comment #4)
> (From update of attachment 211880 [details])
> I’d like to see test cases right on the boundary, demonstrating the rounding behavior. So things just below and just above the place where we round in two different directions. Something that would fail if we used the wrong rounding function.

Thanks for the comments, Darin!

Actually, the tests handle this: if we use ceil instead of round, mask-repeat-round-auto1.html, mask-repeat-round-auto2.html, mask-repeat-round-border.html and mask-repeat-round-content.html fail. If we use floor instead of round mask-repeat-round-padding.html fails.

Regarding the round function, we can use lroundf, which takes a float and also does the conversion to long - I'll upload a patch for this.

[Andrei Parvu]

Created attachment 211975 [details]
Patch

[Build Bot]

Comment on attachment 211975 [details]
Patch

Attachment 211975 [details] did not pass mac-ews (mac):
Output: http://webkit-queues.appspot.com/results/1886361

New failing tests:
svg/batik/text/smallFonts.svg

[Build Bot]

Created attachment 211980 [details]
Archive of layout-test-results from webkit-ews-08 for mac-mountainlion

The attached test failures were seen while running run-webkit-tests on the mac-ews.
Bot: webkit-ews-08  Port: mac-mountainlion  Platform: Mac OS X 10.8.5

[Andrei Parvu]

Created attachment 212430 [details]
Patch

[WebKit Commit Bot]

Comment on attachment 212430 [details]
Patch

Clearing flags on attachment: 212430

Committed r156322: <http://trac.webkit.org/changeset/156322>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Dean Jackson]

*** Bug 122857 has been marked as a duplicate of this bug. ***