45424 2010-09-08 17:18:25 -0700 svn-apply tries to delete directories it shouldn't 2010-09-09 14:30:30 -0700 1 1 1 Unclassified WebKit Tools / Tests 528+ (Nightly build) Other All RESOLVED FIXED P2 Normal --- 1 mihaip mihaip dbates dglazkov eric oldest_to_newest 276197 0 mihaip 2010-09-08 17:18:25 -0700 svn-apply tries to delete directories it shouldn't 276205 1 mihaip 2010-09-08 17:29:43 -0700 Eric, can you review this? This is why the EWS failed for bug 45307 (see output at https://webkit-commit-queue.appspot.com/results/3961246). It called isDirectoryEmptyForRemoval on LayoutTests/platform/mac/http. That only has one item in it, the tests subdirectory. We ended up calling scmWillDeleteFile("LayoutTests/platform/mac/http/tests"), which returned true since there were files deleted in it (and for the git case we just look for a leading D). 276208 2 66971 mihaip 2010-09-08 17:32:46 -0700 Created attachment 66971 Patch 276652 3 eric 2010-09-09 12:11:56 -0700 Dan is the man. 276654 4 66971 eric 2010-09-09 12:12:42 -0700 Comment on attachment 66971 Patch We've been trying to unittest our perl code more. I'm not sure we have a way to test code which is still in svn-apply directly (instead of being in a seprate perl module), but Dan would know. What about svn-unapply? 276729 5 dbates 2010-09-09 13:55:06 -0700 (In reply to comment #4) > (From update of attachment 66971 [details]) > We've been trying to unittest our perl code more. I'm not sure we have a way to test code which is still in svn-apply directly (instead of being in a seprate perl module), but Dan would know. What about svn-unapply? Unfortunately, we don't have a way to test code in svn-apply and svn-unapply directly at this time. This function is specific to svn-apply. 276730 6 66971 dbates 2010-09-09 13:55:34 -0700 Comment on attachment 66971 Patch r=me. 276735 7 mihaip 2010-09-09 13:58:33 -0700 (In reply to comment #6) > (From update of attachment 66971 [details]) > r=me. Thanks. Mind landing this (I don't have committer access yet)? 276739 8 dbates 2010-09-09 14:00:27 -0700 (In reply to comment #7) > (In reply to comment #6) > > (From update of attachment 66971 [details] [details]) > > r=me. > > Thanks. Mind landing this (I don't have committer access yet)? Sure. 276744 9 66971 dbates 2010-09-09 14:08:00 -0700 Comment on attachment 66971 Patch Clearing flags on attachment: 66971 Committed r67115: <http://trac.webkit.org/changeset/67115> 276745 10 dbates 2010-09-09 14:08:09 -0700 All reviewed patches have been landed. Closing bug. 276766 11 eric 2010-09-09 14:28:40 -0700 You can also always set cq? :) Thanks dan for the info and landage! 276769 12 dbates 2010-09-09 14:30:30 -0700 (In reply to comment #11) > You can also always set cq? :) Thanks dan for the info and landage! Yes, I normally would cq+ such a patch, but the queue looked pretty full at the time. 66971 2010-09-08 17:32:46 -0700 2010-09-09 14:08:00 -0700 Patch bug-45424-20100908173245.patch text/plain 1473 mihaip ZGlmZiAtLWdpdCBhL1dlYktpdFRvb2xzL0NoYW5nZUxvZyBiL1dlYktpdFRvb2xzL0NoYW5nZUxv ZwppbmRleCBiOGRiZGFjMDFmY2Y5MGZmMTJiOGMzOWZmNmI5ZGQ1NzYwNjk1N2FiLi5mMzViZjNm YmYwMzYyZmIxYzc5ODdjZDI4ZTBmNzBmYWUyMzk0OGQyIDEwMDY0NAotLS0gYS9XZWJLaXRUb29s cy9DaGFuZ2VMb2cKKysrIGIvV2ViS2l0VG9vbHMvQ2hhbmdlTG9nCkBAIC0xLDMgKzEsMTYgQEAK KzIwMTAtMDktMDggIE1paGFpIFBhcnBhcml0YSAgPG1paGFpcEBjaHJvbWl1bS5vcmc+CisKKyAg ICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISkuCisKKyAgICAgICAgc3ZuLWFwcGx5IHRy aWVzIHRvIGRlbGV0ZSBkaXJlY3RvcmllcyBpdCBzaG91bGRuJ3QKKyAgICAgICAgaHR0cHM6Ly9i dWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTQ1NDI0CisKKyAgICAgICAgaXNEaXJlY3Rv cnlFbXB0eUZvclJlbW92YWwgaGFkIHRoZSB3cm9uZyBjaGVjay4gSWYgYW4gaXRlbSBpbiB0aGUK KyAgICAgICAgZGlyZWN0b3J5IGlzIGl0c2VsZiBhIGRpcmVjdG9yeSwgdGhlbiB0aGUgZGlyZWN0 b3J5IGlzIGRlZmluaXRlbHkKKyAgICAgICAgbm90IGVtcHR5LgorCisgICAgICAgICogU2NyaXB0 cy9zdm4tYXBwbHk6CisKIDIwMTAtMDktMDggIFZpY3RvciBXYW5nICA8dmljdG9yd0BjaHJvbWl1 bS5vcmc+CiAKICAgICAgICAgUmV2aWV3ZWQgYnkgT2phbiBWYWZhaS4KZGlmZiAtLWdpdCBhL1dl YktpdFRvb2xzL1NjcmlwdHMvc3ZuLWFwcGx5IGIvV2ViS2l0VG9vbHMvU2NyaXB0cy9zdm4tYXBw bHkKaW5kZXggMzNiMjI3OTdiOGM0ZTUwNjkzZjg1MGE2NmY3ZGNiNDRiMTFjZWU2ZC4uMWNmOWMw MWM2Nzk2N2I0ZWI4YmMxOGRjMWYyYzQ2MTQ3ZmUyYjU4NSAxMDA3NTUKLS0tIGEvV2ViS2l0VG9v bHMvU2NyaXB0cy9zdm4tYXBwbHkKKysrIGIvV2ViS2l0VG9vbHMvU2NyaXB0cy9zdm4tYXBwbHkK QEAgLTI4MCw3ICsyODAsNyBAQCBzdWIgaXNEaXJlY3RvcnlFbXB0eUZvclJlbW92YWwoJCkKICAg ICBvcGVuZGlyIERJUiwgJGRpciBvciBkaWUgIkNvdWxkIG5vdCBvcGVuICckZGlyJyB0byBsaXN0 IGZpbGVzOiAkPyI7CiAgICAgZm9yIChteSAkaXRlbSA9IHJlYWRkaXIgRElSOyAkaXRlbSAmJiAk ZGlyZWN0b3J5SXNFbXB0eTsgJGl0ZW0gPSByZWFkZGlyIERJUikgewogICAgICAgICBuZXh0IGlm IGV4aXN0cyAkcmVtb3ZlRGlyZWN0b3J5SWdub3JlTGlzdHskaXRlbX07Ci0gICAgICAgIGlmICgh IC1kIEZpbGU6OlNwZWMtPmNhdGRpcigkZGlyLCAkaXRlbSkpIHsKKyAgICAgICAgaWYgKC1kIEZp bGU6OlNwZWMtPmNhdGRpcigkZGlyLCAkaXRlbSkpIHsKICAgICAgICAgICAgICRkaXJlY3RvcnlJ c0VtcHR5ID0gMDsKICAgICAgICAgfSBlbHNlIHsKICAgICAgICAgICAgIG5leHQgaWYgKHNjbVdp bGxEZWxldGVGaWxlKEZpbGU6OlNwZWMtPmNhdGRpcigkZGlyLCAkaXRlbSkpKTsK