99194 2012-10-12 11:48:31 -0700 Regression r130057: incorrect block pref width for alternating InlineFlow and inline Replaced 2012-10-15 14:29:37 -0700 1 1 1 Unclassified WebKit Layout and Rendering 528+ (Nightly build) Unspecified Unspecified RESOLVED FIXED P2 Normal --- 1 leviw leviw arpitabahuguna eae eric hyatt jchaffraix mitz ojan rniwa robert tony webkit.review.bot oldest_to_newest 741079 0 leviw 2012-10-12 11:48:31 -0700 http://trac.webkit.org/changeset/130057 fixed block pref width calculation, but introduced a regression where alternating InlineFlows and inline Replaced elements will result in the prefs widths calculation never breaking the line. 741136 1 168466 leviw 2012-10-12 13:12:24 -0700 Created attachment 168466 Patch 741273 2 leviw 2012-10-12 15:42:51 -0700 Upstream chromium bug http://code.google.com/p/chromium/issues/detail?id=154797 741470 3 168552 arpitabahuguna 2012-10-13 07:52:57 -0700 Created attachment 168552 Testcase.html Apologize for not having verified all the scenarios and for causing this regression. I tried to come up with similar combinations which might regress the previous fix and in the process found one more issue. The case is when a RenderInline containing some text follows another RenderInline wrapping a RenderReplaced. For this case we are not breaking the line after the RenderInline containing the text. This causes it to display extra space after the line containing the RenderReplaced. Have uploaded the testcase containing two variations of the same issue. FF too does not seem to handle the second variation in the uploaded testcase. This issue occurs even after applying this patch. I think I may have a fix for the same. Should this case too be included as part of this regression or shall I raise another bug for it? 742044 4 arpitabahuguna 2012-10-15 08:00:16 -0700 Hey Levi, I was trying to fix the issue mentioned in my previous comment (#3) which resulted in handling of some other failing scenarios (arising out of the fix for #3 issue) such as : <div style="width: 50px"> <span style="float: left; border: 1px solid black;"> <span>FF doesn't handle this case.</span>Some text outside of span<span><img src="50x50.gif"/></span> </span> </div> Wherein a text followed by another text and then a RenderReplaced (enclosed within a RenderInline) causes the text to overflow. However, all these issues are independent of your patch. Am trying to fix all these scenarios and if it's okay with you, will try to submit a patch for them by tomorrow. 742128 5 leviw 2012-10-15 10:04:05 -0700 (In reply to comment #4) > Am trying to fix all these scenarios and if it's okay with you, will try to submit a patch for them by tomorrow. Of course I welcome your patch. In the meantime, I'd love to get this patch reviewed or the original reverted as this is causing painful regressions. 742412 6 leviw 2012-10-15 14:11:07 -0700 Anyone? This is a definite improvement from current behavior. We didn't cover all the cases prior to r130057, and won't with this fix, but at least we'll correct this regression. 742425 7 168466 eric 2012-10-15 14:16:33 -0700 Comment on attachment 168466 Patch OK. 742426 8 168466 leviw 2012-10-15 14:17:17 -0700 Comment on attachment 168466 Patch Thanks Eric. 742429 9 leviw 2012-10-15 14:18:21 -0700 Arpita, please open a new bug and add your test case (and patch when you have it) to it. 742456 10 168466 webkit.review.bot 2012-10-15 14:29:33 -0700 Comment on attachment 168466 Patch Clearing flags on attachment: 168466 Committed r131359: <http://trac.webkit.org/changeset/131359> 742457 11 webkit.review.bot 2012-10-15 14:29:37 -0700 All reviewed patches have been landed. Closing bug. 168466 2012-10-12 13:12:24 -0700 2012-10-15 14:29:33 -0700 Patch bug-99194-20121012131120.patch text/plain 4652 leviw SW5kZXg6IFNvdXJjZS9XZWJDb3JlL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2Vi Q29yZS9DaGFuZ2VMb2cJKHJldmlzaW9uIDEzMTIwOSkKKysrIFNvdXJjZS9XZWJDb3JlL0NoYW5n ZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwzICsxLDE4IEBACisyMDEyLTEwLTEyICBMZXZpIFdl aW50cmF1YiAgPGxldml3QGNocm9taXVtLm9yZz4KKworICAgICAgICBSZWdyZXNzaW9uIHIxMzAw NTc6IGluY29ycmVjdCBibG9jayBwcmVmIHdpZHRoIGZvciBhbHRlcm5hdGluZyBJbmxpbmVGbG93 IGFuZCBpbmxpbmUgUmVwbGFjZWQKKyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hv d19idWcuY2dpP2lkPTk5MTk0CisKKyAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISku CisKKyAgICAgICAgQ29ycmVjdGx5IGJyZWFraW5nIGxpbmVzIGFmdGVyIGlubGluZSByZXBsYWNl ZCBlbGVtZW50cyB0aGF0IGZvbGxvdyBpbmxpbmUgZmxvd3MuIFByZXZpb3VzbHksIGFsdGVybmF0 aW5nCisgICAgICAgIHRoZXNlIHR3byB0eXBlcyBvZiBlbGVtZW50cyB3b3VsZCBjYXVzZSB1cyB0 byBpbmNyZWFzZSBvdXIgcHJlZmVycmVkIHdpZHRoIGZvcmV2ZXIuCisKKyAgICAgICAgVGVzdDog ZmFzdC9ibG9jay9ibG9jay13aXRoLWlubGluZS1yZXBsYWNlZC1jaGlsZHJlbi1pbi1pbmxpbmUt Zmxvd3MuaHRtbAorCisgICAgICAgICogcmVuZGVyaW5nL1JlbmRlckJsb2NrLmNwcDoKKyAgICAg ICAgKFdlYkNvcmU6OlJlbmRlckJsb2NrOjpjb21wdXRlSW5saW5lUHJlZmVycmVkTG9naWNhbFdp ZHRocyk6CisKIDIwMTItMTAtMTIgIEFuZHJlYXMgS2xpbmcgIDxrbGluZ0B3ZWJraXQub3JnPgog CiAgICAgICAgIFJFR1JFU1NJT046IFJhcGlkIG1lbW9yeSBncm93dGggY2FsbGluZyBET00gQVBJ cyB3aXRoIGxhcmdlIHN0cmluZ3MuCkluZGV4OiBTb3VyY2UvV2ViQ29yZS9yZW5kZXJpbmcvUmVu ZGVyQmxvY2suY3BwCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT0KLS0tIFNvdXJjZS9XZWJDb3JlL3JlbmRlcmluZy9SZW5k ZXJCbG9jay5jcHAJKHJldmlzaW9uIDEzMTA5MCkKKysrIFNvdXJjZS9XZWJDb3JlL3JlbmRlcmlu Zy9SZW5kZXJCbG9jay5jcHAJKHdvcmtpbmcgY29weSkKQEAgLTU5MTQsNiArNTkxNCwxMSBAQCB2 b2lkIFJlbmRlckJsb2NrOjpjb21wdXRlSW5saW5lUHJlZmVycmVkCiAgICAgICAgICAgICAgICAg ICAgIGlubGluZU1pbiA9IDA7CiAgICAgICAgICAgICAgICAgfQogCisgICAgICAgICAgICAgICAg aWYgKGF1dG9XcmFwICYmIGNhbkJyZWFrUmVwbGFjZWRFbGVtZW50ICYmIGlzUHJldkNoaWxkSW5s aW5lRmxvdykgeworICAgICAgICAgICAgICAgICAgICB1cGRhdGVQcmVmZXJyZWRXaWR0aChtX21p blByZWZlcnJlZExvZ2ljYWxXaWR0aCwgaW5saW5lTWluKTsKKyAgICAgICAgICAgICAgICAgICAg aW5saW5lTWluID0gMDsKKyAgICAgICAgICAgICAgICB9CisKICAgICAgICAgICAgICAgICAvLyBX ZSBhcmUgbm8gbG9uZ2VyIHN0cmlwcGluZyB3aGl0ZXNwYWNlIGF0IHRoZSBzdGFydCBvZgogICAg ICAgICAgICAgICAgIC8vIGEgbGluZS4KICAgICAgICAgICAgICAgICBpZiAoIWNoaWxkLT5pc0Zs b2F0aW5nKCkpIHsKSW5kZXg6IExheW91dFRlc3RzL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBM YXlvdXRUZXN0cy9DaGFuZ2VMb2cJKHJldmlzaW9uIDEzMTIwOSkKKysrIExheW91dFRlc3RzL0No YW5nZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwzICsxLDEzIEBACisyMDEyLTEwLTEyICBMZXZp IFdlaW50cmF1YiAgPGxldml3QGNocm9taXVtLm9yZz4KKworICAgICAgICBSZWdyZXNzaW9uIHIx MzAwNTc6IGluY29ycmVjdCBibG9jayBwcmVmIHdpZHRoIGZvciBhbHRlcm5hdGluZyBJbmxpbmVG bG93IGFuZCBpbmxpbmUgUmVwbGFjZWQKKyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcv c2hvd19idWcuY2dpP2lkPTk5MTk0CisKKyAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BT ISkuCisKKyAgICAgICAgKiBmYXN0L2Jsb2NrL2Jsb2NrLXdpdGgtaW5saW5lLXJlcGxhY2VkLWNo aWxkcmVuLWluLWlubGluZS1mbG93cy1leHBlY3RlZC5odG1sOiBBZGRlZC4KKyAgICAgICAgKiBm YXN0L2Jsb2NrL2Jsb2NrLXdpdGgtaW5saW5lLXJlcGxhY2VkLWNoaWxkcmVuLWluLWlubGluZS1m bG93cy5odG1sOiBBZGRlZC4KKwogMjAxMi0xMC0xMiAgRW1pbCBBIEVrbHVuZCAgPGVhZUBjaHJv bWl1bS5vcmc+CiAKICAgICAgICAgVW5yZXZpZXdlZCBjaHJvbWl1bSBsdWNpZCByZWJhc2VsaW5l IGZvciByMTMxMjAyLgpJbmRleDogTGF5b3V0VGVzdHMvZmFzdC9ibG9jay9ibG9jay13aXRoLWlu bGluZS1yZXBsYWNlZC1jaGlsZHJlbi1pbi1pbmxpbmUtZmxvd3MtZXhwZWN0ZWQuaHRtbAo9PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09Ci0tLSBMYXlvdXRUZXN0cy9mYXN0L2Jsb2NrL2Jsb2NrLXdpdGgtaW5saW5lLXJlcGxh Y2VkLWNoaWxkcmVuLWluLWlubGluZS1mbG93cy1leHBlY3RlZC5odG1sCShyZXZpc2lvbiAwKQor KysgTGF5b3V0VGVzdHMvZmFzdC9ibG9jay9ibG9jay13aXRoLWlubGluZS1yZXBsYWNlZC1jaGls ZHJlbi1pbi1pbmxpbmUtZmxvd3MtZXhwZWN0ZWQuaHRtbAkocmV2aXNpb24gMCkKQEAgLTAsMCAr MSwxOSBAQAorPCFET0NUWVBFIGh0bWw+Cis8aHRtbD4KKzxzdHlsZT4KK3NwYW4geworCXBhZGRp bmctbGVmdDogMzBweDsKK30KK2ltZyB7CisJd2lkdGg6IDUwcHg7CisJaGVpZ2h0OiA1MHB4Owor fQorPC9zdHlsZT4KKzxib2R5PgorPGRpdj5UZXN0IGZvciBCdWd6aWxsYSBidWc6PGEgaHJlZj0i aHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTk5MTk0Ij4gOTkxOTQ8L2E+ IFJlZ3Jlc3Npb24gcjEzMDA1NzogaW5jb3JyZWN0IGJsb2NrIHByZWYgd2lkdGggZm9yIGFsdGVy bmF0aW5nIElubGluZUZsb3cgYW5kIGlubGluZSBSZXBsYWNlZC48L2Rpdj4KKzxkaXY+VGhpcyB0 ZXN0IHZlcmlmaWVzIHRoYXQgd2UgcHJvcGVybHkgY2FsY3VsYXRlIGJsb2NrIHByZWZlcnJlZCB3 aWR0aHMgd2hlbiB3ZSBoYXZlIHJlcGVhdGVkIGlubGluZSByZXBsYWNlZCBlbGVtZW50cyB3cmFw cGVkIGluIElubGluZUZsb3dzLjwvZGl2PgorPGRpdiBzdHlsZT0id2lkdGg6IDgwcHg7IGJvcmRl cjogMXB4IHNvbGlkIGJsYWNrOyI+CisgICAgPHNwYW4gc3R5bGU9IiI+PGltZyBzcmM9InJlc291 cmNlcy81MHg1MC5naWYiLz48L3NwYW4+PHNwYW4+PGltZyBzcmM9InJlc291cmNlcy81MHg1MC5n aWYiLz48L3NwYW4+Cis8L2Rpdj4KKzwvYm9keT4KKzwvaHRtbD4KXCBObyBuZXdsaW5lIGF0IGVu ZCBvZiBmaWxlCkluZGV4OiBMYXlvdXRUZXN0cy9mYXN0L2Jsb2NrL2Jsb2NrLXdpdGgtaW5saW5l LXJlcGxhY2VkLWNoaWxkcmVuLWluLWlubGluZS1mbG93cy5odG1sCj09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0tIExh eW91dFRlc3RzL2Zhc3QvYmxvY2svYmxvY2std2l0aC1pbmxpbmUtcmVwbGFjZWQtY2hpbGRyZW4t aW4taW5saW5lLWZsb3dzLmh0bWwJKHJldmlzaW9uIDApCisrKyBMYXlvdXRUZXN0cy9mYXN0L2Js b2NrL2Jsb2NrLXdpdGgtaW5saW5lLXJlcGxhY2VkLWNoaWxkcmVuLWluLWlubGluZS1mbG93cy5o dG1sCShyZXZpc2lvbiAwKQpAQCAtMCwwICsxLDIxIEBACis8IURPQ1RZUEUgaHRtbD4KKzxodG1s PgorPHN0eWxlPgorc3BhbiA+IHNwYW4geworCXBhZGRpbmctbGVmdDogMzBweDsKK30KK2ltZyB7 CisJd2lkdGg6IDUwcHg7CisJaGVpZ2h0OiA1MHB4OworfQorPC9zdHlsZT4KKzxib2R5PgorPGRp dj5UZXN0IGZvciBCdWd6aWxsYSBidWc6PGEgaHJlZj0iaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcv c2hvd19idWcuY2dpP2lkPTk5MTk0Ij4gOTkxOTQ8L2E+IFJlZ3Jlc3Npb24gcjEzMDA1NzogaW5j b3JyZWN0IGJsb2NrIHByZWYgd2lkdGggZm9yIGFsdGVybmF0aW5nIElubGluZUZsb3cgYW5kIGlu bGluZSBSZXBsYWNlZC48L2Rpdj4KKzxkaXY+VGhpcyB0ZXN0IHZlcmlmaWVzIHRoYXQgd2UgcHJv cGVybHkgY2FsY3VsYXRlIGJsb2NrIHByZWZlcnJlZCB3aWR0aHMgd2hlbiB3ZSBoYXZlIHJlcGVh dGVkIGlubGluZSByZXBsYWNlZCBlbGVtZW50cyB3cmFwcGVkIGluIElubGluZUZsb3dzLjwvZGl2 PgorPGRpdiBzdHlsZT0id2lkdGg6IDUwcHgiPgorICAgIDxzcGFuIHN0eWxlPSJmbG9hdDogbGVm dDsgYm9yZGVyOiAxcHggc29saWQgYmxhY2s7Ij4KKyAgICAgICAgPHNwYW4gc3R5bGU9IiI+PGlt ZyBzcmM9InJlc291cmNlcy81MHg1MC5naWYiLz48L3NwYW4+PHNwYW4+PGltZyBzcmM9InJlc291 cmNlcy81MHg1MC5naWYiLz48L3NwYW4+CisgICAgPC9zcGFuPgorPC9kaXY+Cis8L2JvZHk+Cis8 L2h0bWw+ClwgTm8gbmV3bGluZSBhdCBlbmQgb2YgZmlsZQo= 168552 2012-10-13 07:52:57 -0700 2012-10-13 07:52:57 -0700 Testcase.html testcase.html text/html 527 arpitabahuguna PCFET0NUWVBFIGh0bWw+DQo8aHRtbD4NCjxzdHlsZT4NCnNwYW4gPiBzcGFuIHsNCglwYWRkaW5n LWxlZnQ6IDMwcHg7DQp9DQppbWcgew0KCXdpZHRoOiA1MHB4Ow0KCWhlaWdodDogNTBweDsNCn0N Cjwvc3R5bGU+DQo8Ym9keT4NCjxkaXYgc3R5bGU9IndpZHRoOiA1MHB4Ij4NCjxzcGFuIHN0eWxl PSJmbG9hdDogbGVmdDsgYm9yZGVyOiAxcHggc29saWQgYmxhY2s7Ij4NCjxzcGFuPjxpbWcgc3Jj PSI1MHg1MC5naWYiLz48L3NwYW4+PHNwYW4+VGhpcyBpcyBzb21lIHRleHQ8L3NwYW4+PHNwYW4+ PGltZyBzcmM9IjUweDUwLmdpZiIvPjwvc3Bhbj4NCjwvc3Bhbj4NCjwvZGl2Pg0KDQo8ZGl2IHN0 eWxlPSJ3aWR0aDogNTBweCI+DQo8c3BhbiBzdHlsZT0iZmxvYXQ6IGxlZnQ7IGJvcmRlcjogMXB4 IHNvbGlkIGJsYWNrOyI+DQo8c3Bhbj5GRiBkb2Vzbid0IGhhbmRsZSB0aGlzIGNhc2UuPC9zcGFu PjxzcGFuPjxpbWcgc3JjPSI1MHg1MC5naWYiLz48L3NwYW4+DQo8L3NwYW4+DQo8L2Rpdj4NCjwv Ym9keT4NCjwvaHRtbD4=