79577 2012-02-25 12:08:39 -0800 [Gtk][WK1] Use mock scrollbars when testing 2012-04-20 16:47:15 -0700 1 1 1 Unclassified WebKit Tools / Tests 528+ (Nightly build) Unspecified Unspecified RESOLVED FIXED P2 Normal --- 0 zan webkit-unassigned mrobinson pnormand oldest_to_newest 565003 0 zan 2012-02-25 12:08:39 -0800 WebKitTestRunner utilizes mock scrollbars when testing. This creates small pixel differences in render trees compared to already generated results. There is no need to generate WebKit2-specific results, rather than that the mock scrollbars should be used when testing with DumpRenderTree as well. Such changes would need a lot of rebaselining, requiring careful approach. 565132 1 128909 zan 2012-02-26 01:02:20 -0800 Created attachment 128909 Patch 565133 2 zan 2012-02-26 01:04:53 -0800 (In reply to comment #1) > Created an attachment (id=128909) [details] > Patch This is just a preliminary patch, showing required changes to use mock scrollbars when running DumpRenderTree. I'll post some statistics about how many test results are affected, based upon r108931. 565141 3 zan 2012-02-26 02:03:03 -0800 Statistics for a release build of r108931, comparing to results at http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release/r108931%20(18789)/results.html css1/ - 66 new text diff mismatches css2.1/ - 31 new text diff mismatches editing/ - 22 new text diff mismatches fast/ - 365 new text diff mismatches fonts/ - 6 new text diff mismatches scrollbars/ - 1 new text diff mismatch svg/ - 32 new text diff mismatches tables/ - 125 new text diff mismatches transforms/ - 2 new text diff mismatches 565170 4 zan 2012-02-26 09:40:55 -0800 (In reply to comment #3) > Statistics for a release build of r108931, comparing to results at http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release/r108931%20(18789)/results.html > > css1/ > - 66 new text diff mismatches > css2.1/ > - 31 new text diff mismatches > editing/ > - 22 new text diff mismatches > fast/ > - 365 new text diff mismatches > fonts/ > - 6 new text diff mismatches > scrollbars/ > - 1 new text diff mismatch > svg/ > - 32 new text diff mismatches > tables/ > - 125 new text diff mismatches > transforms/ > - 2 new text diff mismatches Forgot to add results for http tests: http/ - 3 new text diff mismatches 573068 5 zan 2012-03-07 11:41:10 -0800 With r110053, rebaseline is required for 659 tests. If I can get a thumbs-up, I'd like to proceed with creating proper patches for enabling mock scrollbars when testing and for rebaselines. 573080 6 mrobinson 2012-03-07 11:52:01 -0800 (In reply to comment #5) > With r110053, rebaseline is required for 659 tests. If I can get a thumbs-up, I'd like to proceed with creating proper patches for enabling mock scrollbars when testing and for rebaselines. Let's do this! 573710 7 130795 zan 2012-03-08 01:37:36 -0800 Created attachment 130795 Patch Patch for review 573712 8 zan 2012-03-08 01:40:43 -0800 (In reply to comment #7) > Created an attachment (id=130795) [details] > Patch > > Patch for review Rebaselines after these changes work well with WKTR too, meaning future differences between WKTR and DRT output are originating from other issues. Land and rebaseline with care, the bots will go all volcano. 606973 9 130795 dbates 2012-04-20 15:41:04 -0700 Comment on attachment 130795 Patch LGTM 607050 10 130795 mrobinson 2012-04-20 16:47:11 -0700 Comment on attachment 130795 Patch Clearing flags on attachment: 130795 Committed r114799: <http://trac.webkit.org/changeset/114799> 607051 11 mrobinson 2012-04-20 16:47:15 -0700 All reviewed patches have been landed. Closing bug. 128909 2012-02-26 01:02:20 -0800 2012-03-08 01:37:27 -0800 Patch bug-79577-20120226100218.patch text/plain 1452 zan U3VidmVyc2lvbiBSZXZpc2lvbjogMTA4OTMxCmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViS2l0L2d0 ay9DaGFuZ2VMb2cgYi9Tb3VyY2UvV2ViS2l0L2d0ay9DaGFuZ2VMb2cKaW5kZXggNTliYjkwYWNi MDZmZTM3MjcwMTEyNWZmMmIwMjg1NzY5NzA4ZTgwMy4uODNmMTFmMjM3ZjIyZjI0NjQ3NDdhODU0 Njk1YjMxZTdlZTEyY2EyMyAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdC9ndGsvQ2hhbmdlTG9n CisrKyBiL1NvdXJjZS9XZWJLaXQvZ3RrL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE1IEBACisyMDEy LTAyLTI2ICBaYW4gRG9iZXJzZWsgIDx6YW5kb2JlcnNla0BnbWFpbC5jb20+CisKKyAgICAgICAg W0d0a11bV0sxXSBVc2UgbW9jayBzY3JvbGxiYXJzIHdoZW4gdGVzdGluZworICAgICAgICBodHRw czovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9Nzk1NzcKKworICAgICAgICBSZXZp ZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAgICAgICBFbmFibGUgbW9jayBzY3JvbGxiYXJz IGlmIHJ1bm5pbmcgRHVtcFJlbmRlclRyZWUuCisKKyAgICAgICAgKiB3ZWJraXQvd2Via2l0d2Vi dmlldy5jcHA6CisgICAgICAgICh3ZWJraXRfd2ViX3ZpZXdfdXBkYXRlX3NldHRpbmdzKToKKwog MjAxMi0wMi0yNCAgU2hpbnlhIEthd2FuYWthICA8c2hpbnlha0BjaHJvbWl1bS5vcmc+CiAKICAg ICAgICAgU3BlbGxDaGVja1JlcXVlc3QgbmVlZHMgdG8ga25vdyB0aGUgY29udGV4dCB3aGVyZSB0 aGUgc3BlbGxjaGVjayBoYXBwZW5lZC4KZGlmZiAtLWdpdCBhL1NvdXJjZS9XZWJLaXQvZ3RrL3dl YmtpdC93ZWJraXR3ZWJ2aWV3LmNwcCBiL1NvdXJjZS9XZWJLaXQvZ3RrL3dlYmtpdC93ZWJraXR3 ZWJ2aWV3LmNwcAppbmRleCA4OWI5YzIyOGZiOTNjYTZiMjExNjBhNmU5ZTc5NWU0ZTQ4ZWE5YTBi Li5lNmY5YTcxN2I0NmExM2JmZTAxMGFkODIyYmRiNmVmNTRjODBiZWE0IDEwMDY0NAotLS0gYS9T b3VyY2UvV2ViS2l0L2d0ay93ZWJraXQvd2Via2l0d2Vidmlldy5jcHAKKysrIGIvU291cmNlL1dl YktpdC9ndGsvd2Via2l0L3dlYmtpdHdlYnZpZXcuY3BwCkBAIC0zMzMwLDYgKzMzMzAsOCBAQCBz dGF0aWMgdm9pZCB3ZWJraXRfd2ViX3ZpZXdfdXBkYXRlX3NldHRpbmdzKFdlYktpdFdlYlZpZXcq IHdlYlZpZXcpCiAgICAgY29yZVNldHRpbmdzLT5zZXRVc2VIaXhpZTc2V2ViU29ja2V0UHJvdG9j b2woZmFsc2UpOwogI2VuZGlmCiAKKyAgICBjb3JlU2V0dGluZ3MtPnNldE1vY2tTY3JvbGxiYXJz RW5hYmxlZChEdW1wUmVuZGVyVHJlZVN1cHBvcnRHdGs6OmR1bXBSZW5kZXJUcmVlTW9kZUVuYWJs ZWQoKSk7CisKICAgICBpZiAoUGFnZSogcGFnZSA9IGNvcmUod2ViVmlldykpCiAgICAgICAgIHBh Z2UtPnNldFRhYktleUN5Y2xlc1Rocm91Z2hFbGVtZW50cyhzZXR0aW5nc1ByaXZhdGUtPnRhYktl eUN5Y2xlc1Rocm91Z2hFbGVtZW50cyk7CiAK 130795 2012-03-08 01:37:36 -0800 2012-04-20 16:47:11 -0700 Patch bug-79577-20120308103734.patch text/plain 1533 zan U3VidmVyc2lvbiBSZXZpc2lvbjogMTEwMTUyCmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViS2l0L2d0 ay9DaGFuZ2VMb2cgYi9Tb3VyY2UvV2ViS2l0L2d0ay9DaGFuZ2VMb2cKaW5kZXggNzk0MGNhM2Nk NDkwYTNmMGQ0YmRiZWU3Y2M1Yjc5YjU1Y2M2NDRhOS4uYTdiMDNjZWYzODZhZmUyYzRiOTRhMmNm ZGEwY2VkMTAzMjcwMTZlNyAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdC9ndGsvQ2hhbmdlTG9n CisrKyBiL1NvdXJjZS9XZWJLaXQvZ3RrL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE1IEBACisyMDEy LTAzLTA4ICBaYW4gRG9iZXJzZWsgIDx6YW5kb2JlcnNla0BnbWFpbC5jb20+CisKKyAgICAgICAg W0d0a11bV0sxXSBVc2UgbW9jayBzY3JvbGxiYXJzIHdoZW4gdGVzdGluZworICAgICAgICBodHRw czovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9Nzk1NzcKKworICAgICAgICBSZXZp ZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAgICAgICBFbmFibGUgbW9jayBzY3JvbGxiYXJz IGlmIHJ1bm5pbmcgRHVtcFJlbmRlclRyZWUuCisKKyAgICAgICAgKiB3ZWJraXQvd2Via2l0d2Vi dmlldy5jcHA6CisgICAgICAgICh3ZWJraXRfd2ViX3ZpZXdfdXBkYXRlX3NldHRpbmdzKToKKwog MjAxMi0wMy0wNyAgS2FuZ2lsIEhhbiAgPGthbmdpbC5oYW5Ac2Ftc3VuZy5jb20+CiAKICAgICAg ICAgW0RSVF0gUmVtb3ZlIFBsYWluVGV4dENvbnRyb2xsZXIgaW1wbGVtZW50YXRpb25zLgpkaWZm IC0tZ2l0IGEvU291cmNlL1dlYktpdC9ndGsvd2Via2l0L3dlYmtpdHdlYnZpZXcuY3BwIGIvU291 cmNlL1dlYktpdC9ndGsvd2Via2l0L3dlYmtpdHdlYnZpZXcuY3BwCmluZGV4IDg4MGYyNDQzODI5 MDZlMDBiNzYwNzk3OGZiMjQ1ODMzYzJlY2IyZjEuLjE3ODE3MjY2OWExNzIzZjdlYWU2NWQ1Yjky ZDY3ZTM4MDczOWE1NTEgMTAwNjQ0Ci0tLSBhL1NvdXJjZS9XZWJLaXQvZ3RrL3dlYmtpdC93ZWJr aXR3ZWJ2aWV3LmNwcAorKysgYi9Tb3VyY2UvV2ViS2l0L2d0ay93ZWJraXQvd2Via2l0d2Vidmll dy5jcHAKQEAgLTMzNDcsNiArMzM0Nyw5IEBAIHN0YXRpYyB2b2lkIHdlYmtpdF93ZWJfdmlld191 cGRhdGVfc2V0dGluZ3MoV2ViS2l0V2ViVmlldyogd2ViVmlldykKICAgICBjb3JlU2V0dGluZ3Mt PnNldEVuYWJsZVNjcm9sbEFuaW1hdG9yKHNldHRpbmdzUHJpdmF0ZS0+ZW5hYmxlU21vb3RoU2Ny b2xsaW5nKTsKICNlbmRpZgogCisgICAgLy8gVXNlIG1vY2sgc2Nyb2xsYmFycyBpZiBpbiBEdW1w UmVuZGVyVHJlZSBtb2RlIChpLmUuIHRlc3RpbmcgbGF5b3V0IHRlc3RzKS4KKyAgICBjb3JlU2V0 dGluZ3MtPnNldE1vY2tTY3JvbGxiYXJzRW5hYmxlZChEdW1wUmVuZGVyVHJlZVN1cHBvcnRHdGs6 OmR1bXBSZW5kZXJUcmVlTW9kZUVuYWJsZWQoKSk7CisKICAgICBpZiAoUGFnZSogcGFnZSA9IGNv cmUod2ViVmlldykpCiAgICAgICAgIHBhZ2UtPnNldFRhYktleUN5Y2xlc1Rocm91Z2hFbGVtZW50 cyhzZXR0aW5nc1ByaXZhdGUtPnRhYktleUN5Y2xlc1Rocm91Z2hFbGVtZW50cyk7CiAK