Bug 125235

(In reply to comment #0) > Shapes specification no longer contains inset-rectangle. The new inset parsing support has been added in r159968, so we can remove the deprecated inset-rectangle support. I'll post the patch soon. We shouldn't remove that until we update the demos/blogs/etc. In which case, I'd argue that we should just have one bug to remove all of the deprecated shapes at that time. (Perhaps change this one into "remove deprecated shapes")

(In reply to comment #1) > (In reply to comment #0) > > Shapes specification no longer contains inset-rectangle. The new inset parsing support has been added in r159968, so we can remove the deprecated inset-rectangle support. I'll post the patch soon. > > We shouldn't remove that until we update the demos/blogs/etc. In which case, I'd argue that we should just have one bug to remove all of the deprecated shapes at that time. (Perhaps change this one into "remove deprecated shapes" It may not be practical to remove all of the deprecated stuff in one giant patch, but I agree that attempting to remove the bandage in as close to one quick gesture as possible is a good idea.

<rdar://problem/15958594>

What's the latest on this?

(In reply to comment #4) > What's the latest on this? I plan to remove inset-rectangle and the rest of the deprecated shapes in the next two weeks.

I'm changing this to cover removing all the deprecated shapes. A patch is forthcoming that removes the following: rectangle inset-rectangle It also removes the old syntax for ellipse and circle that used a comma separated argument list.

Created attachment 226053 [details] Patch

smfr, krit: I figure you guys will want to take a look at this, since it affects clip-path as well as shapes.

Comment on attachment 226053 [details] Patch After sleeping on this and letting my exuberance wear off, I will split this up into smaller patches; I'll do one for each shape type. However, the rectangle patch is probably still going to be pretty huge on account of the number of tests that still use it. This patch still shows what things will look like when the individual patches land.

(In reply to comment #9) > (From update of attachment 226053 [details]) > After sleeping on this and letting my exuberance wear off, I will split this up into smaller patches; I'll do one for each shape type. > > However, the rectangle patch is probably still going to be pretty huge on account of the number of tests that still use it. > > This patch still shows what things will look like when the individual patches land. IIRC you are just removing code. So I don't think that this needs to be split. I just want to hear the opinion of smfr first.

Comment on attachment 226053 [details] Patch Putting the r? back on Dirk's advice. I'll leave the dependent bugs for now waiting smfr's feedback.

Did anyone every ship these? Not sure what went out in Safari 7/iOS 7.

(In reply to comment #12) > Did anyone every ship these? Not sure what went out in Safari 7/iOS 7. We ship -webkit-clip-path and old syntax in iOS & Safari 7. Without inset-rect though.

(In reply to comment #13) > (In reply to comment #12) > > Did anyone every ship these? Not sure what went out in Safari 7/iOS 7. > > We ship -webkit-clip-path and old syntax in iOS & Safari 7. Without inset-rect though. So we're removing things that were shipping in iOS 7? Isn't this patch removing support for -webkit-clip-path: rectangle(0%, 0%, 100%, 100%);

Comment on attachment 226053 [details] Patch Discussed this with Dean on #webkit. Yes, we are removing the support for the old syntax of circle(), ellipse() and rectangle() which would break content using these functions. The syntax for polygon() stays. The patch looks good. Bem, please wait 24h to allow others to raise there concerns. If there are no objections, you can land the patch.

Created attachment 226501 [details] Patch for landing

Comment on attachment 226501 [details] Patch for landing No one yelled, so sending to the CQ!

Committed r165472: <http://trac.webkit.org/changeset/165472>