69590 2011-10-06 18:29:31 -0700 There should be a way to disable optimizer in webkit-patch rebaseline-expectations 2012-03-05 18:04:32 -0800 1 1 1 Unclassified WebKit Tools / Tests 528+ (Nightly build) Unspecified Unspecified RESOLVED FIXED P2 Normal --- 1 rniwa dpranke abarth dglazkov dpranke epoger eric tomhudson tony oldest_to_newest 479697 0 rniwa 2011-10-06 18:29:31 -0700 There must be a way to disable expectation optimizer in webkit-patch rebaseline-expectations. I often rebaseline then use pretty-diff to see test diffs because plain diff and wdiff hurt my eyes, but diffs are all messed when expected results are optimized. Also, it erroneously puts render tree dumps in cross-platform directory. 479698 1 rniwa 2011-10-06 18:31:14 -0700 I'd say that this is a requirement before we can remove rebaseline-chromium-webkit-tests. 479699 2 abarth 2011-10-06 18:32:00 -0700 Why is it wrong to put render tree dumps in the cross-platform directory when they're the same on most (all?) platforms? 479700 3 abarth 2011-10-06 18:32:38 -0700 I think the real issue is that there isn't a good way of viewing the effects of changing expectations, which is something we should fix. 479702 4 rniwa 2011-10-06 18:35:31 -0700 (In reply to comment #2) > Why is it wrong to put render tree dumps in the cross-platform directory when they're the same on most (all?) platforms? Because that'll make missing results to failures on some ports. (In reply to comment #3) > I think the real issue is that there isn't a good way of viewing the effects of changing expectations, which is something we should fix. Right but I don't want to learn how to use new tools. pretty-diff works as is, and I'd like to keep using that. 479704 5 abarth 2011-10-06 18:37:59 -0700 Right, pretty-diff should help you understand how a patch will change results on various platforms. 479706 6 rniwa 2011-10-06 18:39:42 -0700 Optimizing expected results is great but that can be done after I've verified the correctness of rebaselines. 571287 7 130251 dpranke 2012-03-05 17:50:42 -0800 Created attachment 130251 Patch 571291 8 130251 rniwa 2012-03-05 17:52:40 -0800 Comment on attachment 130251 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=130251&action=review > Tools/Scripts/webkitpy/tool/commands/rebaseline.py:206 > + if options.optimize: I think it's better to do an early return here. 571292 9 130251 abarth 2012-03-05 17:53:28 -0800 Comment on attachment 130251 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=130251&action=review > Tools/Scripts/webkitpy/tool/commands/rebaseline.py:160 > + optparse.make_option('--optimize', action='store_true', default=True, > + help=('Optimize/de-dup the expectations after rebaselining (default).')), > + optparse.make_option('--no-optimize', dest='optimize', action='store_false', > + help=('Do not optimize/de-dup the expectations after rebaselining ' > + 'You can use "webkit-patch optimize-baselines" to optimize separately.')), For webkit-patch, we usually just provide a flag for the non-default setting. 571295 10 dpranke 2012-03-05 17:54:23 -0800 *** Bug 72860 has been marked as a duplicate of this bug. *** 571312 11 dpranke 2012-03-05 18:04:32 -0800 Committed r109831: <http://trac.webkit.org/changeset/109831> 130251 2012-03-05 17:50:42 -0800 2012-03-05 17:53:28 -0800 Patch bug-69590-20120305175041.patch text/plain 6481 dpranke U3VidmVyc2lvbiBSZXZpc2lvbjogMTA5ODI0CmRpZmYgLS1naXQgYS9Ub29scy9DaGFuZ2VMb2cg Yi9Ub29scy9DaGFuZ2VMb2cKaW5kZXggMzgzNGE1ODJhOGZkY2U4ZTc4ZDJkOGE1NmQxYjVjOTcx MjE3MTFiMC4uYmE1MTA1NTc4ODMyNWMyMzY2ZjczMmVmMDI4YWIwNTU3M2ZmNjE5YyAxMDA2NDQK LS0tIGEvVG9vbHMvQ2hhbmdlTG9nCisrKyBiL1Rvb2xzL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE4 IEBACisyMDEyLTAzLTA1ICBEaXJrIFByYW5rZSAgPGRwcmFua2VAY2hyb21pdW0ub3JnPgorCisg ICAgICAgIFRoZXJlIHNob3VsZCBiZSBhIHdheSB0byBkaXNhYmxlIG9wdGltaXplciBpbiB3ZWJr aXQtcGF0Y2ggcmViYXNlbGluZS1leHBlY3RhdGlvbnMKKyAgICAgICAgaHR0cHM6Ly9idWdzLndl YmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTY5NTkwCisKKyAgICAgICAgUmV2aWV3ZWQgYnkgTk9C T0RZIChPT1BTISkuCisKKyAgICAgICAgQWRkcyBhIC0tbm8tb3B0aW1pemUgZmxhZyB0byB3ZWJr aXQtcGF0Y2ggcmViYXNlbGluZS1leHBlY3RhdGlvbnMKKyAgICAgICAgdG8gc2tpcCB0aGUgJ29w dGltaXplLWV4cGVjdGF0aW9ucycgc3RlcC4KKworICAgICAgICAqIFNjcmlwdHMvd2Via2l0cHkv dG9vbC9jb21tYW5kcy9yZWJhc2VsaW5lLnB5OgorICAgICAgICAoUmViYXNlbGluZUV4cGVjdGF0 aW9ucy5fX2luaXRfXyk6CisgICAgICAgIChSZWJhc2VsaW5lRXhwZWN0YXRpb25zLmV4ZWN1dGUp OgorICAgICAgICAqIFNjcmlwdHMvd2Via2l0cHkvdG9vbC9jb21tYW5kcy9yZWJhc2VsaW5lX3Vu aXR0ZXN0LnB5OgorCiAyMDEyLTAzLTA1ICBLZW50IFRhbXVyYSAgPHRrZW50QGNocm9taXVtLm9y Zz4KIAogICAgICAgICBbQ2hyb21pdW1dIEFkZCBuZXcgcG9wdXAgdHlwZTogUGFnZVBvcHVwCmRp ZmYgLS1naXQgYS9Ub29scy9TY3JpcHRzL3dlYmtpdHB5L3Rvb2wvY29tbWFuZHMvcmViYXNlbGlu ZS5weSBiL1Rvb2xzL1NjcmlwdHMvd2Via2l0cHkvdG9vbC9jb21tYW5kcy9yZWJhc2VsaW5lLnB5 CmluZGV4IDUxNWZmN2RmYTU4MWQxN2I0OTY5OTUzMWI3MGQ1ZTFhYjRmMTQ2ZTQuLmYyODE5Y2Yw OTM0ZWRiYTZiY2ViYTMwZGJkMmY2MDgyMmE5M2JjMDUgMTAwNjQ0Ci0tLSBhL1Rvb2xzL1Njcmlw dHMvd2Via2l0cHkvdG9vbC9jb21tYW5kcy9yZWJhc2VsaW5lLnB5CisrKyBiL1Rvb2xzL1Njcmlw dHMvd2Via2l0cHkvdG9vbC9jb21tYW5kcy9yZWJhc2VsaW5lLnB5CkBAIC0yNiw2ICsyNiw3IEBA CiAjIChJTkNMVURJTkcgTkVHTElHRU5DRSBPUiBPVEhFUldJU0UpIEFSSVNJTkcgSU4gQU5ZIFdB WSBPVVQgT0YgVEhFIFVTRQogIyBPRiBUSElTIFNPRlRXQVJFLCBFVkVOIElGIEFEVklTRUQgT0Yg VEhFIFBPU1NJQklMSVRZIE9GIFNVQ0ggREFNQUdFLgogCitpbXBvcnQgb3B0cGFyc2UKIGltcG9y dCBvcy5wYXRoCiBpbXBvcnQgcmUKIGltcG9ydCBzaHV0aWwKQEAgLTE1MCw2ICsxNTEsMTYgQEAg Y2xhc3MgUmViYXNlbGluZUV4cGVjdGF0aW9ucyhBYnN0cmFjdERlY2xhcmF0aXZlQ29tbWFuZCk6 CiAgICAgbmFtZSA9ICJyZWJhc2VsaW5lLWV4cGVjdGF0aW9ucyIKICAgICBoZWxwX3RleHQgPSAi UmViYXNlbGluZXMgdGhlIHRlc3RzIGluZGljYXRlZCBpbiB0ZXN0X2V4cGVjdGF0aW9ucy50eHQu IgogCisgICAgZGVmIF9faW5pdF9fKHNlbGYpOgorICAgICAgICBvcHRpb25zID0gWworICAgICAg ICAgICAgb3B0cGFyc2UubWFrZV9vcHRpb24oJy0tb3B0aW1pemUnLCBhY3Rpb249J3N0b3JlX3Ry dWUnLCBkZWZhdWx0PVRydWUsCisgICAgICAgICAgICAgICAgaGVscD0oJ09wdGltaXplL2RlLWR1 cCB0aGUgZXhwZWN0YXRpb25zIGFmdGVyIHJlYmFzZWxpbmluZyAoZGVmYXVsdCkuJykpLAorICAg ICAgICAgICAgb3B0cGFyc2UubWFrZV9vcHRpb24oJy0tbm8tb3B0aW1pemUnLCBkZXN0PSdvcHRp bWl6ZScsIGFjdGlvbj0nc3RvcmVfZmFsc2UnLAorICAgICAgICAgICAgICAgIGhlbHA9KCdEbyBu b3Qgb3B0aW1pemUvZGUtZHVwIHRoZSBleHBlY3RhdGlvbnMgYWZ0ZXIgcmViYXNlbGluaW5nICcK KyAgICAgICAgICAgICAgICAgICAgICAnWW91IGNhbiB1c2UgIndlYmtpdC1wYXRjaCBvcHRpbWl6 ZS1iYXNlbGluZXMiIHRvIG9wdGltaXplIHNlcGFyYXRlbHkuJykpLAorICAgICAgICBdCisgICAg ICAgIEFic3RyYWN0RGVjbGFyYXRpdmVDb21tYW5kLl9faW5pdF9fKHNlbGYsIG9wdGlvbnM9b3B0 aW9ucykKKwogICAgIGRlZiBfcnVuX3dlYmtpdF9wYXRjaChzZWxmLCBhcmdzKToKICAgICAgICAg dHJ5OgogICAgICAgICAgICAgc2VsZi5fdG9vbC5leGVjdXRpdmUucnVuX2NvbW1hbmQoW3NlbGYu X3Rvb2wucGF0aCgpXSArIGFyZ3MsIGN3ZD1zZWxmLl90b29sLnNjbSgpLmNoZWNrb3V0X3Jvb3Qp CkBAIC0xOTIsOSArMjAzLDEwIEBAIGNsYXNzIFJlYmFzZWxpbmVFeHBlY3RhdGlvbnMoQWJzdHJh Y3REZWNsYXJhdGl2ZUNvbW1hbmQpOgogICAgICAgICAgICAgc2VsZi5fcmViYXNlbGluZV9wb3J0 KHBvcnRfbmFtZSkKICAgICAgICAgZm9yIHBvcnRfbmFtZSBpbiB0b29sLnBvcnRfZmFjdG9yeS5h bGxfcG9ydF9uYW1lcygpOgogICAgICAgICAgICAgc2VsZi5fdXBkYXRlX2V4cGVjdGF0aW9uc19m aWxlKHBvcnRfbmFtZSkKLSAgICAgICAgZm9yIHRlc3RfbmFtZSBpbiBzZWxmLl90b3VjaGVkX3Rl c3RfbmFtZXM6Ci0gICAgICAgICAgICBwcmludCAiT3B0aW1pemluZyBiYXNlbGluZXMgZm9yICVz LiIgJSB0ZXN0X25hbWUKLSAgICAgICAgICAgIHNlbGYuX3J1bl93ZWJraXRfcGF0Y2goWydvcHRp bWl6ZS1iYXNlbGluZXMnLCB0ZXN0X25hbWVdKQorICAgICAgICBpZiBvcHRpb25zLm9wdGltaXpl OgorICAgICAgICAgICAgZm9yIHRlc3RfbmFtZSBpbiBzZWxmLl90b3VjaGVkX3Rlc3RfbmFtZXM6 CisgICAgICAgICAgICAgICAgcHJpbnQgIk9wdGltaXppbmcgYmFzZWxpbmVzIGZvciAlcy4iICUg dGVzdF9uYW1lCisgICAgICAgICAgICAgICAgc2VsZi5fcnVuX3dlYmtpdF9wYXRjaChbJ29wdGlt aXplLWJhc2VsaW5lcycsIHRlc3RfbmFtZV0pCiAKIAogY2xhc3MgUmViYXNlbGluZShBYnN0cmFj dERlY2xhcmF0aXZlQ29tbWFuZCk6CmRpZmYgLS1naXQgYS9Ub29scy9TY3JpcHRzL3dlYmtpdHB5 L3Rvb2wvY29tbWFuZHMvcmViYXNlbGluZV91bml0dGVzdC5weSBiL1Rvb2xzL1NjcmlwdHMvd2Vi a2l0cHkvdG9vbC9jb21tYW5kcy9yZWJhc2VsaW5lX3VuaXR0ZXN0LnB5CmluZGV4IDUyNzI0MzA4 ZGMwYzdhN2EzZWI0MzA1NmZjMDI2ODNmODZkNDIwZjYuLjU4ZDA5MjczMzRhNDAzNzk1NDA1MzVj NzNhNDUyYWVmZjU0ZDU4MDIgMTAwNjQ0Ci0tLSBhL1Rvb2xzL1NjcmlwdHMvd2Via2l0cHkvdG9v bC9jb21tYW5kcy9yZWJhc2VsaW5lX3VuaXR0ZXN0LnB5CisrKyBiL1Rvb2xzL1NjcmlwdHMvd2Vi a2l0cHkvdG9vbC9jb21tYW5kcy9yZWJhc2VsaW5lX3VuaXR0ZXN0LnB5CkBAIC0zMSw3ICszMSw3 IEBAIGltcG9ydCB1bml0dGVzdAogZnJvbSB3ZWJraXRweS5jb21tb24uc3lzdGVtLm91dHB1dGNh cHR1cmUgaW1wb3J0IE91dHB1dENhcHR1cmUKIGZyb20gd2Via2l0cHkudGhpcmRwYXJ0eS5tb2Nr IGltcG9ydCBNb2NrCiBmcm9tIHdlYmtpdHB5LnRvb2wuY29tbWFuZHMucmViYXNlbGluZSBpbXBv cnQgKgotZnJvbSB3ZWJraXRweS50b29sLm1vY2t0b29sIGltcG9ydCBNb2NrVG9vbAorZnJvbSB3 ZWJraXRweS50b29sLm1vY2t0b29sIGltcG9ydCBNb2NrVG9vbCwgTW9ja09wdGlvbnMKIGZyb20g d2Via2l0cHkuY29tbW9uLnN5c3RlbS5leGVjdXRpdmVfbW9jayBpbXBvcnQgTW9ja0V4ZWN1dGl2 ZQogCiAKQEAgLTg0LDkgKzg0LDggQEAgUmV0cmlldmluZyByZXN1bHRzIGZvciBjaHJvbWl1bS13 aW4td2luNyBmcm9tIFdlYmtpdCBXaW43LgogUmV0cmlldmluZyByZXN1bHRzIGZvciBjaHJvbWl1 bS13aW4teHAgZnJvbSBXZWJraXQgV2luLgogICAgIHVzZXJzY3JpcHRzL2Fub3RoZXItdGVzdC5o dG1sCiAgICAgdXNlcnNjcmlwdHMvaW1hZ2VzLnN2ZwotT3B0aW1pemluZyBiYXNlbGluZXMgZm9y IHVzZXJzY3JpcHRzL2Fub3RoZXItdGVzdC5odG1sLgotT3B0aW1pemluZyBiYXNlbGluZXMgZm9y IHVzZXJzY3JpcHRzL2ltYWdlcy5zdmcuCiAiIiIKKwogICAgICAgICBleHBlY3RlZF9zdGRlcnIg PSAiIiJNT0NLIHJ1bl9jb21tYW5kOiBbJ2VjaG8nLCAncmViYXNlbGluZS10ZXN0JywgJ1dlYmtp dCBMaW51eCAzMicsICd1c2Vyc2NyaXB0cy9hbm90aGVyLXRlc3QuaHRtbCddLCBjd2Q9L21vY2st Y2hlY2tvdXQKIE1PQ0sgcnVuX2NvbW1hbmQ6IFsnZWNobycsICdyZWJhc2VsaW5lLXRlc3QnLCAn V2Via2l0IExpbnV4IDMyJywgJ3VzZXJzY3JpcHRzL2ltYWdlcy5zdmcnXSwgY3dkPS9tb2NrLWNo ZWNrb3V0CiBNT0NLIHJ1bl9jb21tYW5kOiBbJ2VjaG8nLCAncmViYXNlbGluZS10ZXN0JywgJ1dl YmtpdCBMaW51eCcsICd1c2Vyc2NyaXB0cy9hbm90aGVyLXRlc3QuaHRtbCddLCBjd2Q9L21vY2st Y2hlY2tvdXQKQEAgLTEwMSw4ICsxMDAsMTcgQEAgTU9DSyBydW5fY29tbWFuZDogWydlY2hvJywg J3JlYmFzZWxpbmUtdGVzdCcsICdXZWJraXQgV2luNycsICd1c2Vyc2NyaXB0cy9hbm90aGUKIE1P Q0sgcnVuX2NvbW1hbmQ6IFsnZWNobycsICdyZWJhc2VsaW5lLXRlc3QnLCAnV2Via2l0IFdpbjcn LCAndXNlcnNjcmlwdHMvaW1hZ2VzLnN2ZyddLCBjd2Q9L21vY2stY2hlY2tvdXQKIE1PQ0sgcnVu X2NvbW1hbmQ6IFsnZWNobycsICdyZWJhc2VsaW5lLXRlc3QnLCAnV2Via2l0IFdpbicsICd1c2Vy c2NyaXB0cy9hbm90aGVyLXRlc3QuaHRtbCddLCBjd2Q9L21vY2stY2hlY2tvdXQKIE1PQ0sgcnVu X2NvbW1hbmQ6IFsnZWNobycsICdyZWJhc2VsaW5lLXRlc3QnLCAnV2Via2l0IFdpbicsICd1c2Vy c2NyaXB0cy9pbWFnZXMuc3ZnJ10sIGN3ZD0vbW9jay1jaGVja291dAotTU9DSyBydW5fY29tbWFu ZDogWydlY2hvJywgJ29wdGltaXplLWJhc2VsaW5lcycsICd1c2Vyc2NyaXB0cy9hbm90aGVyLXRl c3QuaHRtbCddLCBjd2Q9L21vY2stY2hlY2tvdXQKLU1PQ0sgcnVuX2NvbW1hbmQ6IFsnZWNobycs ICdvcHRpbWl6ZS1iYXNlbGluZXMnLCAndXNlcnNjcmlwdHMvaW1hZ2VzLnN2ZyddLCBjd2Q9L21v Y2stY2hlY2tvdXQKICIiIgorCisgICAgICAgIGNvbW1hbmQuX3Rlc3RzX3RvX3JlYmFzZWxpbmUg PSBsYW1iZGEgcG9ydDogW10gaWYgbm90IHBvcnQubmFtZSgpLmZpbmQoJy1ncHUtJykgPT0gLTEg ZWxzZSBbJ3VzZXJzY3JpcHRzL2Fub3RoZXItdGVzdC5odG1sJywgJ3VzZXJzY3JpcHRzL2ltYWdl cy5zdmcnXQorICAgICAgICBPdXRwdXRDYXB0dXJlKCkuYXNzZXJ0X291dHB1dHMoc2VsZiwgY29t bWFuZC5leGVjdXRlLCBbTW9ja09wdGlvbnMob3B0aW1pemU9RmFsc2UpLCBbXSwgdG9vbF0sIGV4 cGVjdGVkX3N0ZG91dD1leHBlY3RlZF9zdGRvdXQsIGV4cGVjdGVkX3N0ZGVycj1leHBlY3RlZF9z dGRlcnIpCisKKyAgICAgICAgZXhwZWN0ZWRfc3Rkb3V0X3dpdGhfb3B0aW1pemUgPSBleHBlY3Rl ZF9zdGRvdXQgKyAoCisgICAgICAgICAgICAiT3B0aW1pemluZyBiYXNlbGluZXMgZm9yIHVzZXJz Y3JpcHRzL2Fub3RoZXItdGVzdC5odG1sLlxuIgorICAgICAgICAgICAgIk9wdGltaXppbmcgYmFz ZWxpbmVzIGZvciB1c2Vyc2NyaXB0cy9pbWFnZXMuc3ZnLlxuIikKKyAgICAgICAgZXhwZWN0ZWRf c3RkZXJyX3dpdGhfb3B0aW1pemUgPSBleHBlY3RlZF9zdGRlcnIgKyAoCisgICAgICAgICAgICAi TU9DSyBydW5fY29tbWFuZDogWydlY2hvJywgJ29wdGltaXplLWJhc2VsaW5lcycsICd1c2Vyc2Ny aXB0cy9hbm90aGVyLXRlc3QuaHRtbCddLCBjd2Q9L21vY2stY2hlY2tvdXRcbiIKKyAgICAgICAg ICAgICJNT0NLIHJ1bl9jb21tYW5kOiBbJ2VjaG8nLCAnb3B0aW1pemUtYmFzZWxpbmVzJywgJ3Vz ZXJzY3JpcHRzL2ltYWdlcy5zdmcnXSwgY3dkPS9tb2NrLWNoZWNrb3V0XG4iKQorCiAgICAgICAg IGNvbW1hbmQuX3Rlc3RzX3RvX3JlYmFzZWxpbmUgPSBsYW1iZGEgcG9ydDogW10gaWYgbm90IHBv cnQubmFtZSgpLmZpbmQoJy1ncHUtJykgPT0gLTEgZWxzZSBbJ3VzZXJzY3JpcHRzL2Fub3RoZXIt dGVzdC5odG1sJywgJ3VzZXJzY3JpcHRzL2ltYWdlcy5zdmcnXQotICAgICAgICBPdXRwdXRDYXB0 dXJlKCkuYXNzZXJ0X291dHB1dHMoc2VsZiwgY29tbWFuZC5leGVjdXRlLCBbTm9uZSwgW10sIHRv b2xdLCBleHBlY3RlZF9zdGRvdXQ9ZXhwZWN0ZWRfc3Rkb3V0LCBleHBlY3RlZF9zdGRlcnI9ZXhw ZWN0ZWRfc3RkZXJyKQorICAgICAgICBPdXRwdXRDYXB0dXJlKCkuYXNzZXJ0X291dHB1dHMoc2Vs ZiwgY29tbWFuZC5leGVjdXRlLCBbTW9ja09wdGlvbnMob3B0aW1pemU9VHJ1ZSksIFtdLCB0b29s XSwgZXhwZWN0ZWRfc3Rkb3V0PWV4cGVjdGVkX3N0ZG91dF93aXRoX29wdGltaXplLCBleHBlY3Rl ZF9zdGRlcnI9ZXhwZWN0ZWRfc3RkZXJyX3dpdGhfb3B0aW1pemUpCg==