100999 2012-11-01 16:42:02 -0700 [CSS Exclusions] Shape-inside content area may be larger than default content area 2013-01-24 16:36:57 -0800 1 1 1 Unclassified WebKit CSS 528+ (Nightly build) Unspecified Unspecified RESOLVED INVALID P2 Normal --- 89256 1 betravis zoltan donggwan.kim eric webkit.review.bot oldest_to_newest 756726 0 betravis 2012-11-01 16:42:02 -0700 The shape-inside content area is currently clipped to the default content area. 761114 1 172882 zoltan 2012-11-07 15:37:18 -0800 Created attachment 172882 proposed patch 761337 2 172882 betravis 2012-11-07 22:08:28 -0800 Comment on attachment 172882 proposed patch View in context: https://bugs.webkit.org/attachment.cgi?id=172882&action=review The patch only tests a shape inside larger than the content area on the right/bottom sides. It also needs to deal with the upper/left sides. Check out the beginning of layoutRunsAndFloatsInRange and the setLogicalHeight call for how the top is set now. Also, the simple-rectangle.js script is designed to cover these basic test cases. I know that some of the changes here may break the simple float test case, but you might remove that test case to be added at a later date, since making floats work with shape-inside will require significant work. > Source/WebCore/rendering/RenderBlockLineLayout.cpp:165 > m_left = max<float>(m_segment->logicalLeft, m_left); > - m_right = min<float>(m_segment->logicalRight, m_right); > + > + if (!(m_block->m_floatingObjects && m_block->m_floatingObjects->hasRightObjects()) && m_segment->logicalRight > m_right) > + m_right = m_segment->logicalRight; > + else > + m_right = min<float>(m_segment->logicalRight, m_right); Do you know that if there are right floating objects, they necessarily affect this line? I think that the floating objects set contains all of this block's floats, so they may not. I think these lines should just be: m_left = m_segment->logicalLeft; m_right = m_segment->logicalRight; This leaves the issue of floats, but perhaps the better thing would be to pull out the float test and implement that in a subsequent patch. > LayoutTests/fast/exclusions/shape-inside/shape-inside-with-larger-content-area.html:16 > + -webkit-shape-inside: rectangle(0, 0, 200px, 200px); You need to test the rectangle going beyond the bounds of each of the boxes (content, padding, border, margin). I think to start, just going outside the margin box would be sufficient. eg, #shape-inside { position: absolute; left: 10px; top: 10px; width: 180px; height: 180px; -webkit-shape-inside: rectangle(-10px, -10px, 200px, 200px); } Although really, this should be wrapped into using resources/simple-rectangle.js, in which case you would include the custom styling: #shape-inside { position: absolute; left: 10px; top: 10px; } and the script: createRectangleTest('shape-inside', 'stylesheet', { width: 200, height: 200 }, { x: -10, y: -10, width: 200, height: 200 }, 'px', null); // use null for the last argument if you want to leave shape-inside's content in place You can also see shape-inside-subsequent-blocks, and shape-inside-multiple-blocks-dynamic. The scripts obviate the need for the 'helper' element. 761362 3 172882 zoltan 2012-11-07 22:54:49 -0800 Comment on attachment 172882 proposed patch I removed the patch from the review queue since it needs some improvement. Thanks for the comments, things are one step clearer know. 815968 4 zoltan 2013-01-24 16:36:57 -0800 The spec has been changed. This bug is not valid anymore. 172882 2012-11-07 15:37:18 -0800 2012-11-07 22:54:49 -0800 proposed patch shapeinsidecanbebigger.patch text/plain 5236 zoltan SW5kZXg6IFNvdXJjZS9XZWJDb3JlL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2Vi Q29yZS9DaGFuZ2VMb2cJKHJldmlzaW9uIDEzMzgxMSkKKysrIFNvdXJjZS9XZWJDb3JlL0NoYW5n ZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwzICsxLDE4IEBACisyMDEyLTExLTA3ICBab2x0YW4g SG9ydmF0aCAgPHpvbHRhbkB3ZWJraXQub3JnPgorCisgICAgICAgIFtDU1MgRXhjbHVzaW9uc10g U2hhcGUtaW5zaWRlIGNvbnRlbnQgYXJlYSBtYXkgYmUgbGFyZ2VyIHRoYW4gZGVmYXVsdCBjb250 ZW50IGFyZWEKKyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lk PTEwMDk5OQorCisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgorCisgICAgICAg IFRoZSBzaGFwZS1pbnNpZGUgY29udGVudCBhcmVhIGNhbiBiZSBiaWdnZXIgdGhhbiB0aGUgZGVm YXVsdCBjb250ZW50IGFyZWEsIHNvIHdlIG5lZWQKKyAgICAgICAgdG8gc3VwcG9ydCB0aGlzIGJl aGF2aW9yLgorCisgICAgICAgIFRlc3Q6IGZhc3QvZXhjbHVzaW9ucy9zaGFwZS1pbnNpZGUvc2hh cGUtaW5zaWRlLXdpdGgtbGFyZ2VyLWNvbnRlbnQtYXJlYS5odG1sCisKKyAgICAgICAgKiByZW5k ZXJpbmcvUmVuZGVyQmxvY2tMaW5lTGF5b3V0LmNwcDoKKyAgICAgICAgKFdlYkNvcmU6OkxpbmVX aWR0aDo6dXBkYXRlQXZhaWxhYmxlV2lkdGgpOgorCiAyMDEyLTExLTA3ICBBZGFtIEJhcnRoICA8 YWJhcnRoQHdlYmtpdC5vcmc+CiAKICAgICAgICAgW1Y4XSBJbnRydXNpdmVET01XcmFwcGVyTWFw IHNob3VsZCBiZSB1c2FibGUgZm9yIG1vcmUgdGhhbiBqdXN0IE5vZGVzCkluZGV4OiBTb3VyY2Uv V2ViQ29yZS9yZW5kZXJpbmcvUmVuZGVyQmxvY2tMaW5lTGF5b3V0LmNwcAo9PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0t LSBTb3VyY2UvV2ViQ29yZS9yZW5kZXJpbmcvUmVuZGVyQmxvY2tMaW5lTGF5b3V0LmNwcAkocmV2 aXNpb24gMTMzNTAzKQorKysgU291cmNlL1dlYkNvcmUvcmVuZGVyaW5nL1JlbmRlckJsb2NrTGlu ZUxheW91dC5jcHAJKHdvcmtpbmcgY29weSkKQEAgLTE1OCw3ICsxNTgsMTEgQEAKICNpZiBFTkFC TEUoQ1NTX0VYQ0xVU0lPTlMpCiAgICAgaWYgKG1fc2VnbWVudCkgewogICAgICAgICBtX2xlZnQg PSBtYXg8ZmxvYXQ+KG1fc2VnbWVudC0+bG9naWNhbExlZnQsIG1fbGVmdCk7Ci0gICAgICAgIG1f cmlnaHQgPSBtaW48ZmxvYXQ+KG1fc2VnbWVudC0+bG9naWNhbFJpZ2h0LCBtX3JpZ2h0KTsKKwor ICAgICAgICBpZiAoIShtX2Jsb2NrLT5tX2Zsb2F0aW5nT2JqZWN0cyAmJiBtX2Jsb2NrLT5tX2Zs b2F0aW5nT2JqZWN0cy0+aGFzUmlnaHRPYmplY3RzKCkpICYmIG1fc2VnbWVudC0+bG9naWNhbFJp Z2h0ID4gbV9yaWdodCkKKyAgICAgICAgICAgIG1fcmlnaHQgPSBtX3NlZ21lbnQtPmxvZ2ljYWxS aWdodDsKKyAgICAgICAgZWxzZQorICAgICAgICAgICAgbV9yaWdodCA9IG1pbjxmbG9hdD4obV9z ZWdtZW50LT5sb2dpY2FsUmlnaHQsIG1fcmlnaHQpOwogICAgIH0KICNlbmRpZgogCkluZGV4OiBM YXlvdXRUZXN0cy9mYXN0L2V4Y2x1c2lvbnMvc2hhcGUtaW5zaWRlL3NoYXBlLWluc2lkZS13aXRo LWxhcmdlci1jb250ZW50LWFyZWEtZXhwZWN0ZWQuaHRtbAo9PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBMYXlvdXRU ZXN0cy9mYXN0L2V4Y2x1c2lvbnMvc2hhcGUtaW5zaWRlL3NoYXBlLWluc2lkZS13aXRoLWxhcmdl ci1jb250ZW50LWFyZWEtZXhwZWN0ZWQuaHRtbAkocmV2aXNpb24gMCkKKysrIExheW91dFRlc3Rz L2Zhc3QvZXhjbHVzaW9ucy9zaGFwZS1pbnNpZGUvc2hhcGUtaW5zaWRlLXdpdGgtbGFyZ2VyLWNv bnRlbnQtYXJlYS1leHBlY3RlZC5odG1sCShyZXZpc2lvbiAwKQpAQCAtMCwwICsxLDMxIEBACis8 IURPQ1RZUEUgaHRtbD4KKzxodG1sPgorPGhlYWQ+Cis8c3R5bGU+CisgICAgZGl2IHsKKyAgICAg ICAgZm9udDogQWhlbSwgc2Fucy1zZXJpZjsKKyAgICB9CisgICAgI3NoYXBlLWluc2lkZSB7Cisg ICAgICAgIGJvcmRlcjogMnB4IHNvbGlkIGJsdWU7CisgICAgICAgIHBvc2l0aW9uOiBhYnNvbHV0 ZTsKKyAgICAgICAgbGVmdDogMTBweDsKKyAgICAgICAgdG9wOiAxMHB4OworICAgICAgICB3aWR0 aDogMTAwcHg7CisgICAgICAgIGhlaWdodDogMTAwcHg7CisgICAgICAgIHotaW5kZXg6IC0xOwor ICAgIH0KKyAgICAjc2hhcGUtaW5zaWRlLXJlY3RhbmdsZSB7CisgICAgICAgIGJvcmRlcjogMnB4 IHNvbGlkIGdyZWVuOworICAgICAgICB3aWR0aDogMTk4cHg7CisgICAgICAgIGhlaWdodDogMTk4 cHg7CisgICAgICAgIHBhZGRpbmc6IDJweCAwIDAgMnB4OworICAgIH0KKzwvc3R5bGU+Cis8L2hl YWQ+Cis8Ym9keT4KKyAgICA8ZGl2IGlkPSJzaGFwZS1pbnNpZGUtcmVjdGFuZ2xlIj4KKyAgICBU aGUgYmx1ZSByZWN0YW5nbGUgcmVwcmVzZW50cyB0aGUgc2l6ZSBvZiB0aGUgY29udGVudCBhcmVh LCB0aGUgZ3JlZW4gcmVjdGFuZ2xlIHJlcHJlc2VudHMgdGhlIHNpemUgb2YgdGhlIHNoYXBlIHdo YXQgaXMgYmlnZ2VyIHRoYW4gdGhlIGNvbnRlbnQgYXJlYS4gVGhlIHRleHQgc2hvdWxkIG5vdCBi ZSBjbGlwcGVkIHRvIHRoZSAoYmx1ZSkgZGVmYXVsdCBhcmVhLCBidXQgZmlsbCB0aGUgc3BhY2Ug aW5zaWRlIHRoZSBncmVlbiBzaGFwZS4gVGhlIHRleHQgc2hvdWxkIG92ZXJmbG93IHRoZSBibHVl IHJlY3RhbmdsZSBib3RoIGhvcml6b250YWxseSBhbmQgdmVydGljYWxseS4KKyAgICA8L2Rpdj4K KyAgICA8ZGl2IGlkPSJzaGFwZS1pbnNpZGUiPjwvZGl2PgorPC9ib2R5PgorPC9odG1sPgpJbmRl eDogTGF5b3V0VGVzdHMvZmFzdC9leGNsdXNpb25zL3NoYXBlLWluc2lkZS9zaGFwZS1pbnNpZGUt d2l0aC1sYXJnZXItY29udGVudC1hcmVhLmh0bWwKPT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PQotLS0gTGF5b3V0VGVzdHMv ZmFzdC9leGNsdXNpb25zL3NoYXBlLWluc2lkZS9zaGFwZS1pbnNpZGUtd2l0aC1sYXJnZXItY29u dGVudC1hcmVhLmh0bWwJKHJldmlzaW9uIDApCisrKyBMYXlvdXRUZXN0cy9mYXN0L2V4Y2x1c2lv bnMvc2hhcGUtaW5zaWRlL3NoYXBlLWluc2lkZS13aXRoLWxhcmdlci1jb250ZW50LWFyZWEuaHRt bAkocmV2aXNpb24gMCkKQEAgLTAsMCArMSwzMiBAQAorPCFET0NUWVBFIGh0bWw+Cis8aHRtbD4K KzxoZWFkPgorPHNjcmlwdD4KKyAgICBpZiAod2luZG93LmludGVybmFscykKKyAgICAgICAgd2lu ZG93LmludGVybmFscy5zZXR0aW5ncy5zZXRDU1NFeGNsdXNpb25zRW5hYmxlZCh0cnVlKTsKKzwv c2NyaXB0PgorPHN0eWxlPgorICAgIGRpdiB7CisgICAgICAgIGZvbnQ6IEFoZW0sIHNhbnMtc2Vy aWY7CisgICAgfQorICAgICNzaGFwZS1pbnNpZGUgeworICAgICAgICBib3JkZXI6IDJweCBzb2xp ZCBibHVlOworICAgICAgICB3aWR0aDogMTAwcHg7CisgICAgICAgIGhlaWdodDogMTAwcHg7Cisg ICAgICAgIC13ZWJraXQtc2hhcGUtaW5zaWRlOiByZWN0YW5nbGUoMCwgMCwgMjAwcHgsIDIwMHB4 KTsKKyAgICB9CisgICAgI3NoYXBlLWluc2lkZS1oZWxwZXItcmVjdGFuZ2xlIHsKKyAgICAgICAg Ym9yZGVyOiAycHggc29saWQgZ3JlZW47CisgICAgICAgIHdpZHRoOiAyMDBweDsKKyAgICAgICAg aGVpZ2h0OiAyMDBweDsKKyAgICB9Cis8L3N0eWxlPgorPC9oZWFkPgorPGJvZHk+CisgICAgPGRp diBpZD0ic2hhcGUtaW5zaWRlLWhlbHBlci1yZWN0YW5nbGUiPgorICAgICAgICA8ZGl2IGlkPSJz aGFwZS1pbnNpZGUiPiAgIAorICAgICAgICBUaGUgYmx1ZSByZWN0YW5nbGUgcmVwcmVzZW50cyB0 aGUgc2l6ZSBvZiB0aGUgY29udGVudCBhcmVhLCB0aGUgZ3JlZW4gcmVjdGFuZ2xlIHJlcHJlc2Vu dHMgdGhlIHNpemUgb2YgdGhlIHNoYXBlIHdoYXQgaXMgYmlnZ2VyIHRoYW4gdGhlIGNvbnRlbnQg YXJlYS4gVGhlIHRleHQgc2hvdWxkIG5vdCBiZSBjbGlwcGVkIHRvIHRoZSAoYmx1ZSkgZGVmYXVs dCBhcmVhLCBidXQgZmlsbCB0aGUgc3BhY2UgaW5zaWRlIHRoZSBncmVlbiBzaGFwZS4gVGhlIHRl eHQgc2hvdWxkIG92ZXJmbG93IHRoZSBibHVlIHJlY3RhbmdsZSBib3RoIGhvcml6b250YWxseSBh bmQgdmVydGljYWxseS4KKyAgICAgICAgPC9kaXY+CisgICAgPC9kaXY+Cis8L2JvZHk+Cis8L2h0 bWw+CkluZGV4OiBMYXlvdXRUZXN0cy9DaGFuZ2VMb2cKPT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PQotLS0gTGF5b3V0VGVz dHMvQ2hhbmdlTG9nCShyZXZpc2lvbiAxMzM4MTEpCisrKyBMYXlvdXRUZXN0cy9DaGFuZ2VMb2cJ KHdvcmtpbmcgY29weSkKQEAgLTEsMyArMSwxNiBAQAorMjAxMi0xMS0wNyAgWm9sdGFuIEhvcnZh dGggIDx6b2x0YW5Ad2Via2l0Lm9yZz4KKworICAgICAgICBbQ1NTIEV4Y2x1c2lvbnNdIFNoYXBl LWluc2lkZSBjb250ZW50IGFyZWEgbWF5IGJlIGxhcmdlciB0aGFuIGRlZmF1bHQgY29udGVudCBh cmVhCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3JnL3Nob3dfYnVnLmNnaT9pZD0xMDA5 OTkKKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAgICAgICBUaGUg c2hhcGUtaW5zaWRlIGNvbnRlbnQgYXJlYSBjYW4gYmUgYmlnZ2VyIHRoYW4gdGhlIGRlZmF1bHQg Y29udGVudCBhcmVhLCBzbyB3ZSBuZWVkCisgICAgICAgIHRvIHN1cHBvcnQgdGhpcyBiZWhhdmlv ci4KKworICAgICAgICAqIGZhc3QvZXhjbHVzaW9ucy9zaGFwZS1pbnNpZGUvc2hhcGUtaW5zaWRl LXdpdGgtbGFyZ2VyLWNvbnRlbnQtYXJlYS1leHBlY3RlZC5odG1sOiBBZGRlZC4KKyAgICAgICAg KiBmYXN0L2V4Y2x1c2lvbnMvc2hhcGUtaW5zaWRlL3NoYXBlLWluc2lkZS13aXRoLWxhcmdlci1j b250ZW50LWFyZWEuaHRtbDogQWRkZWQuCisKIDIwMTItMTEtMDcgIERpcmsgUHJhbmtlICA8ZHBy YW5rZUBjaHJvbWl1bS5vcmc+CiAKICAgICAgICAgVW5yZXZpZXdlZCwgdXBkYXRlIGNocm9taXVt LW1hYy1tb3VudGFpbmxpb24gYmFzZWxpbmVzIGZvciBjc3MyLjEvMjAxMS8wMzIzCg==