39985 2010-06-01 02:22:17 -0700 Several layout tests needlessly use invalid HTML 2010-06-02 14:44:17 -0700 1 1 1 Unclassified WebKit Tools / Tests 528+ (Nightly build) Other OS X 10.5 RESOLVED FIXED P2 Normal --- 39259 1 eric eric abarth ap commit-queue tonyg oldest_to_newest 232656 0 eric 2010-06-01 02:22:17 -0700 Several layout tests needlessly use invalid HTML 232657 1 57525 eric 2010-06-01 02:27:42 -0700 Created attachment 57525 Patch 232819 2 57525 abarth 2010-06-01 10:29:09 -0700 Comment on attachment 57525 Patch LayoutTests/ChangeLog:12 + for this test to depend on WebKit's old parser behavior. Can you cite a specific test here? I want to be sure we don't lose test coverage. LayoutTests/ChangeLog:18 + no need to depend on the old title behavior here. This is also tested in fast/tokenizer. I worry that some of these changes will cause web compat problems. We need to keep close tabs on this sort of thing. 232990 3 57525 ap 2010-06-01 16:37:22 -0700 Comment on attachment 57525 Patch Besides Web compatibility, there is compatibility with Mac applications using WebKit (such as Dashboard). I think that there are enough different changes to regression tests here to discuss them as part of review, not post-review. 233069 4 abarth 2010-06-01 22:03:41 -0700 @ap: I don't understand your comment. We're not changing any behavior. We're just improving the quality of our test suite by testing parsing issues in parser tests and editing issues in editing tests. 233076 5 ap 2010-06-01 22:20:31 -0700 Adam, I think I'm saying the same as you did in your review. The ChangeLog should mention which tests cover the aspects of behavior that are no longer covered by the changed tests, and a reviewer needs to verify those claims. I never said anything about changing behavior. 233081 6 abarth 2010-06-01 22:27:42 -0700 Ah, thanks. I misunderstood your comment. 233102 7 57625 eric 2010-06-01 23:44:34 -0700 Created attachment 57625 Patch 233118 8 ap 2010-06-02 00:37:09 -0700 I'm concerned that after this change, current behavior will mostly be tested as failure cases in html5lib test suite. It may be too easy to overlook that we'll be changing behavior. After all, HTML5 is nothing but a draft yet. Maybe such tests should be copied to fast/invalid? 233121 9 abarth 2010-06-02 00:55:13 -0700 > Maybe such tests should be copied to fast/invalid? The tests aren't invalid, as far as I can tell. They appear to be perfectly good tests with typos. > I'm concerned that after this change, current behavior will mostly be tested as failure cases in html5lib test suite. It may be too easy to overlook that we'll be changing behavior. Every time we change the results of a test in html5lib we're changing behavior. It's unclear to me why you think these particular parser behaviors ought to be tested in editing tests. The <title> changes are also tested in fast/tokenizer by two tests. The <foo<bar case is already tested in fast/invalid, which is the remedy you propose. As far as I can tell, there's nothing we can do to address your concerns. > After all, HTML5 is nothing but a draft yet. Well, perhaps we should implement the HTML4 parser then. :P 233126 10 abarth 2010-06-02 01:06:59 -0700 > > After all, HTML5 is nothing but a draft yet. > > Well, perhaps we should implement the HTML4 parser then. :P Ok. I thought of a more witty retort... Let's try that again. :) > After all, HTML5 is nothing but a draft yet. That's true, and, after all, evolution nothing but a theory. 233192 11 57625 commit-queue 2010-06-02 03:45:54 -0700 Comment on attachment 57625 Patch Clearing flags on attachment: 57625 Committed r60556: <http://trac.webkit.org/changeset/60556> 233194 12 commit-queue 2010-06-02 03:46:01 -0700 All reviewed patches have been landed. Closing bug. 233403 13 ap 2010-06-02 13:08:00 -0700 Mitz points out that we shouldn't modify tests in /css1. Can you roll out this patch? > Ok. I thought of a more witty retort... So suppose HTML5 changes in some way, and html5lib tests change consequently. Practically speaking, how will we ensure that upgrading our copy won't degrade coverage in areas that this patch delegated to html5lib tests? Delegating test coverage to a highly volatile external suite doesn't sound like a good idea to me, because upgrading responsibly will be a pain. Do we know if any policy governs maintenance of the suite? I'd expect that its goal is to test HTM5 requirements as well as possible, which is close to our goals, but doesn't align with them 100%. 233412 14 abarth 2010-06-02 13:21:29 -0700 (In reply to comment #13) > Mitz points out that we shouldn't modify tests in /css1. Can you roll out this patch? I think Eric agreed to undo the changes to /css1 via email. I'm not sure who was CCed on the thread, but I'd expect those changes shortly. > > Ok. I thought of a more witty retort... > > So suppose HTML5 changes in some way, and html5lib tests change consequently. Practically speaking, how will we ensure that upgrading our copy won't degrade coverage in areas that this patch delegated to html5lib tests? > > Delegating test coverage to a highly volatile external suite doesn't sound like a good idea to me, because upgrading responsibly will be a pain. Do we know if any policy governs maintenance of the suite? I'd expect that its goal is to test HTM5 requirements as well as possible, which is close to our goals, but doesn't align with them 100%. Alexey, I'm not looking to argue with you, but I'm not sure you have all the facts right. My understanding is that there are two parser behaviors at issue here: 1) Unterminated <title> tags. This is covered both by the html5lib test suite and by: http://trac.webkit.org/browser/trunk/LayoutTests/fast/tokenizer/missing-title-end-tag-1.html http://trac.webkit.org/browser/trunk/LayoutTests/fast/tokenizer/missing-title-end-tag-2.html as mentioned in Comment #9. 2) Nested start tags (e.g., "<a<b"). This is covered both by the html5lib test suite and by http://trac.webkit.org/browser/trunk/LayoutTests/fast/invalid/016.html as mentioned in the ChangeLog. Is there something else specific you're worried about? I read, understood, and acted on your feedback. You've already said it was the same as my review in comment #2 of the original patch. I'm not sure why we're still arguing about this. 233427 15 eric 2010-06-02 13:45:57 -0700 Committed r60576: <http://trac.webkit.org/changeset/60576> 233459 16 ap 2010-06-02 14:44:17 -0700 Looks like I did have my facts wrong. My feeling was that each unclosed tag's handling was (or could potentially be) different, and only some cases were covered. Looking at the ChangeLog again, I see that I misunderstood its structure - there are two lists in it, not one. 57525 2010-06-01 02:27:42 -0700 2010-06-01 23:44:29 -0700 Patch bug-39985-20100601022740.patch text/plain 5961 eric ZGlmZiAtLWdpdCBhL0xheW91dFRlc3RzL0NoYW5nZUxvZyBiL0xheW91dFRlc3RzL0NoYW5nZUxv ZwppbmRleCBlY2JiM2FmZmZhOTVjMDZkNmMzOTAyNzJkOGFhODFlNzc0ZWRkM2RiLi4zMmMwYzhh Yzk2ZDNmMDY2YjBhZWNjZmU4OWYxOTY3NjAzYjIyNmFhIDEwMDY0NAotLS0gYS9MYXlvdXRUZXN0 cy9DaGFuZ2VMb2cKKysrIGIvTGF5b3V0VGVzdHMvQ2hhbmdlTG9nCkBAIC0yLDYgKzIsMzEgQEAK IAogICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KIAorICAgICAgICBTZXZlcmFs IGxheW91dCB0ZXN0cyBuZWVkbGVzc2x5IHVzZSBpbnZhbGlkIEhUTUwKKyAgICAgICAgaHR0cHM6 Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTM5OTg1CisKKyAgICAgICAgKiBjc3Mx L2JveF9wcm9wZXJ0aWVzL2Zsb2F0X2VsZW1lbnRzX2luX3Nlcmllcy5odG1sOgorICAgICAgICAg LSBVbmNsb3NlZCA8dGQ+IHRhZ3Mgc3dhbGxvdyB0aGUgZm9sbG93aW5nIDxpbWc+IHRhZ3MuCisg ICAgICAgICAgIFRoaXMgcGFyc2VyIHF1aXJrIGlzIHRlc3RlZCBieSB0aGUgSFRNTDUgdGVzdHMg YW5kIGlzIGFuCisgICAgICAgICAgIGludGVudGlvbmFsIGJlaGF2aW9yIGNoYW5nZSBhcyBwYXJ0 IG9mIEhUTUw1LiAgTm8gbmVlZAorICAgICAgICAgICBmb3IgdGhpcyB0ZXN0IHRvIGRlcGVuZCBv biBXZWJLaXQncyBvbGQgcGFyc2VyIGJlaGF2aW9yLgorICAgICAgICAqIGVkaXRpbmcvc2VsZWN0 aW9uL2luYWN0aXZlLXNlbGVjdGlvbi5odG1sOgorICAgICAgICAgLSBTYW1lIGFzIGFib3ZlLiBI ZXJlIDxkaXYgc3dhbGxvd3MgdGhlIDxzY3JpcHQ+IHRhZy4KKyAgICAgICAgKiBlZGl0aW5nL3Nl bGVjdGlvbi9zZWxlY3QtdGV4dC1vdmVyZmxvdy1lbGxpcHNpcy5odG1sOgorICAgICAgICAgLSBB biB1bmNsb3NlZCA8dGl0bGU+IHdpbGwgc3dhbGxvdyB0aGUgd2hvbGUgZG9jdW1lbnQuCisgICAg ICAgICAgIEludGVudGlvbmFsIEhUTUw1IGNoYW5nZSwgdGVzdGVkIGJ5IEhUTUw1IHRlc3Qgc3Vp dGUgYW5kCisgICAgICAgICAgIG5vIG5lZWQgdG8gZGVwZW5kIG9uIHRoZSBvbGQgdGl0bGUgYmVo YXZpb3IgaGVyZS4KKyAgICAgICAgKiBmYXN0L2RvbS90aXRsZS1jb250ZW50LXdyaXRlLXNldC5o dG1sOgorICAgICAgICAgLSBBZ2Fpbiwgbm8gbmVlZCB0byBsZWF2ZSBhbiB1bmNsb3NlZCB0aXRs ZS4KKyAgICAgICAgKiBmYXN0L2Zvcm1zL3NlbGVjdC1zZXQtaW5uZXIuaHRtbDoKKyAgICAgICAg IC0gVW5jbG9zZWQgPGJvZHkgc3dhbGxvd3MgPHA+CisgICAgICAgICogZmFzdC9qcy9leGNlcHRp b24tbGluZW51bXMtaW4taHRtbC0yLmh0bWw6CisgICAgICAgICAtIFVuY2xvc2VkIDx0aXRsZT4g c3dhbGxvd3MgdGhlIGRvY3VtZW50LgorCisyMDEwLTA2LTAxICBFcmljIFNlaWRlbCAgPGVyaWNA d2Via2l0Lm9yZz4KKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKwogICAg ICAgICBIVE1MNSBwYXJzZXIgZG9lcyBub3QgdHJhY2sgbGluZSBudW1iZXJzCiAgICAgICAgIGh0 dHBzOi8vYnVncy53ZWJraXQub3JnL3Nob3dfYnVnLmNnaT9pZD0zOTk4NAogCmRpZmYgLS1naXQg YS9MYXlvdXRUZXN0cy9jc3MxL2JveF9wcm9wZXJ0aWVzL2Zsb2F0X2VsZW1lbnRzX2luX3Nlcmll cy5odG1sIGIvTGF5b3V0VGVzdHMvY3NzMS9ib3hfcHJvcGVydGllcy9mbG9hdF9lbGVtZW50c19p bl9zZXJpZXMuaHRtbAppbmRleCA1MDBmZjAyNDNiZjZiMjhiNzhlMmQ4MDZjNzk1OTQxMzU3N2Jk ZDQ5Li42NGYyYWRkMDY3OTE5NGMyMzM0N2M1ODViNjNlNDYxMDkyNmU2OWFkIDEwMDY0NAotLS0g YS9MYXlvdXRUZXN0cy9jc3MxL2JveF9wcm9wZXJ0aWVzL2Zsb2F0X2VsZW1lbnRzX2luX3Nlcmll cy5odG1sCisrKyBiL0xheW91dFRlc3RzL2NzczEvYm94X3Byb3BlcnRpZXMvZmxvYXRfZWxlbWVu dHNfaW5fc2VyaWVzLmh0bWwKQEAgLTU5LDcgKzU5LDcgQEAgVEFCTEUge21hcmdpbjogMjBweCAw cHg7fQogPHRyPgogPHRkIHdpZHRoPSIyMCIgc3R5bGU9ImJhY2tncm91bmQ6IGJsYWNrIj48SU1H IFNSQz0iYmxhbmsuZ2lmIiB3aWR0aD0iMjAiIGhlaWdodD0iMSI+PC90ZD4KIDx0ZCB3aWR0aD0x MDAgc3R5bGU9ImJhY2tncm91bmQ6IHllbGxvdyI+PElNRyBTUkM9ImJsYW5rLmdpZiIgd2lkdGg9 IjEwMCIgaGVpZ2h0PSIxIj48L3RkPgotPHRkIHdpZHRoPTIwIHN0eWxlPSJiYWNrZ3JvdW5kOiBi bGFjayI8SU1HIFNSQz0iYmxhbmsuZ2lmIiB3aWR0aD0iMjAiIGhlaWdodD0iMSI+PC90ZD4KKzx0 ZCB3aWR0aD0yMCBzdHlsZT0iYmFja2dyb3VuZDogYmxhY2siPjxJTUcgU1JDPSJibGFuay5naWYi IHdpZHRoPSIyMCIgaGVpZ2h0PSIxIj48L3RkPgogCiA8dGQgd2lkdGg9MjAgc3R5bGU9ImJhY2tn cm91bmQ6IHdoaXRlIj48SU1HIFNSQz0iYmxhbmsuZ2lmIiB3aWR0aD0iMjAiIGhlaWdodD0iMSI+ PC90ZD4KIApAQCAtMTEwLDcgKzExMCw3IEBAIFRBQkxFIHttYXJnaW46IDIwcHggMHB4O30KIDx0 cj4KIDx0ZCB3aWR0aD0iMjAiIHN0eWxlPSJiYWNrZ3JvdW5kOiBibGFjayI+PElNRyBTUkM9ImJs YW5rLmdpZiIgd2lkdGg9IjIwIiBoZWlnaHQ9IjEiPjwvdGQ+CiA8dGQgd2lkdGg9MTAwIHN0eWxl PSJiYWNrZ3JvdW5kOiB5ZWxsb3ciPjxJTUcgU1JDPSJibGFuay5naWYiIHdpZHRoPSIxMDAiIGhl aWdodD0iMSI+PC90ZD4KLTx0ZCB3aWR0aD0yMCBzdHlsZT0iYmFja2dyb3VuZDogYmxhY2siPElN RyBTUkM9ImJsYW5rLmdpZiIgd2lkdGg9IjIwIiBoZWlnaHQ9IjEiPjwvdGQ+Cis8dGQgd2lkdGg9 MjAgc3R5bGU9ImJhY2tncm91bmQ6IGJsYWNrIj48SU1HIFNSQz0iYmxhbmsuZ2lmIiB3aWR0aD0i MjAiIGhlaWdodD0iMSI+PC90ZD4KIAogPHRkIHdpZHRoPTIwIHN0eWxlPSJiYWNrZ3JvdW5kOiB3 aGl0ZSI+PElNRyBTUkM9ImJsYW5rLmdpZiIgd2lkdGg9IjIwIiBoZWlnaHQ9IjEiPjwvdGQ+CiAK ZGlmZiAtLWdpdCBhL0xheW91dFRlc3RzL2VkaXRpbmcvc2VsZWN0aW9uL2luYWN0aXZlLXNlbGVj dGlvbi5odG1sIGIvTGF5b3V0VGVzdHMvZWRpdGluZy9zZWxlY3Rpb24vaW5hY3RpdmUtc2VsZWN0 aW9uLmh0bWwKaW5kZXggOTk3MWMzY2E5MzBiYTJiZThjNjlmY2IxM2U3NTFlOTFhYTNmMWZkZi4u NmMxOTQzOTczYjdlYjhmMDYxNmQ0N2VhZTEwN2E0Yjg4YzU2MzNlNSAxMDA2NDQKLS0tIGEvTGF5 b3V0VGVzdHMvZWRpdGluZy9zZWxlY3Rpb24vaW5hY3RpdmUtc2VsZWN0aW9uLmh0bWwKKysrIGIv TGF5b3V0VGVzdHMvZWRpdGluZy9zZWxlY3Rpb24vaW5hY3RpdmUtc2VsZWN0aW9uLmh0bWwKQEAg LTYsNyArNiw3IEBAIGFmdGVyIHlvdSd2ZSBtYWRlIGEgc2VsZWN0aW9uIGluIGFub3RoZXIgZnJh bWUuPC9wPgogPGlmcmFtZSBzcmM9ImRhdGE6dGV4dC9odG1sLCA8ZGl2IGlkPSdpbnB1dCcgY29u dGVudGVkaXRhYmxlPSd0cnVlJz5zaG91bGQgYmUgYW4gaW5hY3RpdmUgc2VsZWN0aW9uPC9kaXY+ Ij48L2lmcmFtZT4KIDxicj4KIDxpbnB1dCBpZD0naW5wdXQnIHR5cGU9InRleHQiIHZhbHVlPSJy YW5kb20gdGV4dCI+Ci08ZGl2Cis8ZGl2PgogPHNjcmlwdD4KIHdpbmRvdy5vbmxvYWQgPSBmdW5j dGlvbigpIHsKICAgICBpZiAod2luZG93LmxheW91dFRlc3RDb250cm9sbGVyKQpkaWZmIC0tZ2l0 IGEvTGF5b3V0VGVzdHMvZWRpdGluZy9zZWxlY3Rpb24vc2VsZWN0LXRleHQtb3ZlcmZsb3ctZWxs aXBzaXMuaHRtbCBiL0xheW91dFRlc3RzL2VkaXRpbmcvc2VsZWN0aW9uL3NlbGVjdC10ZXh0LW92 ZXJmbG93LWVsbGlwc2lzLmh0bWwKaW5kZXggZjk0ZmRlMzdkNDVkNGUyMWNkOTExYTNlZDUzZDY2 MDU4M2E1Nzg1Ny4uNmNlMzE1NWQyMjRjZjFmMjRlODU0ZmViODlmOWMwOTU0MzBlY2E3OCAxMDA2 NDQKLS0tIGEvTGF5b3V0VGVzdHMvZWRpdGluZy9zZWxlY3Rpb24vc2VsZWN0LXRleHQtb3ZlcmZs b3ctZWxsaXBzaXMuaHRtbAorKysgYi9MYXlvdXRUZXN0cy9lZGl0aW5nL3NlbGVjdGlvbi9zZWxl Y3QtdGV4dC1vdmVyZmxvdy1lbGxpcHNpcy5odG1sCkBAIC0xLDYgKzEsNiBAQAogPGh0bWw+CiAg IDxoZWFkPgotICAgIDx0aXRsZT5UZXN0IGZvciBXZWJLaXQgYnVnIDI5OTY4OiBTZWxlY3Rpbmcg dGV4dCB3aXRoIHRleHQtb3ZlcmZsb3cgZWxsaXBzaXMgc2hvdWxkIG5vdCBzaG93IGN1dCBvZmYg dGV4dDwvaGVhZD4KKyAgICA8dGl0bGU+VGVzdCBmb3IgV2ViS2l0IGJ1ZyAyOTk2ODogU2VsZWN0 aW5nIHRleHQgd2l0aCB0ZXh0LW92ZXJmbG93IGVsbGlwc2lzIHNob3VsZCBub3Qgc2hvdyBjdXQg b2ZmIHRleHQ8L3RpdGxlPjwvaGVhZD4KICAgPHN0eWxlIHR5cGU9InRleHQvY3NzIj4KICAgICAu Y3V0b2ZmIHsKICAgICAgIHdpZHRoOiA5NXB4OwpkaWZmIC0tZ2l0IGEvTGF5b3V0VGVzdHMvZmFz dC9kb20vdGl0bGUtY29udGVudC13cml0ZS1zZXQuaHRtbCBiL0xheW91dFRlc3RzL2Zhc3QvZG9t L3RpdGxlLWNvbnRlbnQtd3JpdGUtc2V0Lmh0bWwKaW5kZXggYmY0ZDVmZTc0ZjFmZDA2NDY0Mzcw M2IyOGI0NTVhOTA5ODg3MzdmZi4uYWNkOGUxOWJiODViZWExZDMzNGQ1MTk5YzZmYWVhMjg4MTgz NTk4YyAxMDA2NDQKLS0tIGEvTGF5b3V0VGVzdHMvZmFzdC9kb20vdGl0bGUtY29udGVudC13cml0 ZS1zZXQuaHRtbAorKysgYi9MYXlvdXRUZXN0cy9mYXN0L2RvbS90aXRsZS1jb250ZW50LXdyaXRl LXNldC5odG1sCkBAIC01LDcgKzUsNyBAQCBpZiAod2luZG93LmxheW91dFRlc3RDb250cm9sbGVy KQogICAgIGxheW91dFRlc3RDb250cm9sbGVyLmR1bXBBc1RleHQoKTsKIGRvY3VtZW50LndyaXRl KCI8dGl0bGU+Iik7CiBkb2N1bWVudC50aXRsZSA9ICJQcm9wZXJ0eSI7Ci1kb2N1bWVudC53cml0 ZSgiV3JpdHRlbiIpOworZG9jdW1lbnQud3JpdGUoIldyaXR0ZW48L3RpdGxlPiIpOwogCiBmdW5j dGlvbiB0ZXN0KCkKIHsKZGlmZiAtLWdpdCBhL0xheW91dFRlc3RzL2Zhc3QvZm9ybXMvc2VsZWN0 LXNldC1pbm5lci5odG1sIGIvTGF5b3V0VGVzdHMvZmFzdC9mb3Jtcy9zZWxlY3Qtc2V0LWlubmVy Lmh0bWwKaW5kZXggODllMjIzODMwODE0NTIyM2I5NDQwYmViYjRmMjA0ZmEwYzRkODUyYS4uOTcy YTFkY2I1MDYyNzNkOWJkYmFhODdhNDU0ZDcwZjlkOTdhN2M3ZCAxMDA2NDQKLS0tIGEvTGF5b3V0 VGVzdHMvZmFzdC9mb3Jtcy9zZWxlY3Qtc2V0LWlubmVyLmh0bWwKKysrIGIvTGF5b3V0VGVzdHMv ZmFzdC9mb3Jtcy9zZWxlY3Qtc2V0LWlubmVyLmh0bWwKQEAgLTI0LDcgKzI0LDcgQEAgZnVuY3Rp b24gdGVzdCgpCiAKIDwvaHRtbD4KIAotPGJvZHkgb25sb2FkPSJ0ZXN0KCkiCis8Ym9keSBvbmxv YWQ9InRlc3QoKSI+CiAKIDxwPlRoaXMgdGVzdHMgYSBjYXNlIHdoZXJlIHJlbW92aW5nIGFuIG9w dGlvbiBlbGVtZW50IGZyb20gYSBzZWxlY3QgYnkgdXNpbmcgaW5uZXJIVE1MIHdvdWxkIGNhdXNl IGEgY3Jhc2guPC9wPgogCmRpZmYgLS1naXQgYS9MYXlvdXRUZXN0cy9mYXN0L2pzL2V4Y2VwdGlv bi1saW5lbnVtcy1pbi1odG1sLTIuaHRtbCBiL0xheW91dFRlc3RzL2Zhc3QvanMvZXhjZXB0aW9u LWxpbmVudW1zLWluLWh0bWwtMi5odG1sCmluZGV4IDc4ZGQ1OWY0OTkxZTY0YWUxZWZlN2RlYjNl ZjdkNGNiODkzZDIzZDIuLjZhMzlhYTA3NzM0NTNhYTE1M2UyYzM5MmVkZGVhZDEyN2U2MzE2M2Ig MTAwNjQ0Ci0tLSBhL0xheW91dFRlc3RzL2Zhc3QvanMvZXhjZXB0aW9uLWxpbmVudW1zLWluLWh0 bWwtMi5odG1sCisrKyBiL0xheW91dFRlc3RzL2Zhc3QvanMvZXhjZXB0aW9uLWxpbmVudW1zLWlu LWh0bWwtMi5odG1sCkBAIC0xLDUgKzEsNSBAQAogPGh0bWw+Ci08aGVhZD48dGl0bGU+dGl0bGUK KzxoZWFkPjx0aXRsZT50aXRsZTwvdGl0bGU+CiA8bGluayByZWw9InN0eWxlc2hlZXQiIGhyZWY9 InJlc291cmNlcy9qcy10ZXN0LXN0eWxlLmNzcyI+CiA8c2NyaXB0IHNyYz0icmVzb3VyY2VzL2pz LXRlc3QtcHJlLmpzIj48L3NjcmlwdD4KIDwvaGVhZD4K 57625 2010-06-01 23:44:34 -0700 2010-06-02 03:45:53 -0700 Patch bug-39985-20100601234433.patch text/plain 6892 eric ZGlmZiAtLWdpdCBhL0xheW91dFRlc3RzL0NoYW5nZUxvZyBiL0xheW91dFRlc3RzL0NoYW5nZUxv ZwppbmRleCAxNjI0MDk0MDhmYjQ1N2NmZWMyYzE2NjM5NmFlNjFkNTZhNzJiYzk5Li5lYzNkYTY0 MjQyMzY2ZGYwNWQxMzc1NmNhMTFkYTgzYWZmMGQ4MDhiIDEwMDY0NAotLS0gYS9MYXlvdXRUZXN0 cy9DaGFuZ2VMb2cKKysrIGIvTGF5b3V0VGVzdHMvQ2hhbmdlTG9nCkBAIC0xLDMgKzEsNDYgQEAK KzIwMTAtMDYtMDEgIEVyaWMgU2VpZGVsICA8ZXJpY0B3ZWJraXQub3JnPgorCisgICAgICAgIFJl dmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgorCisgICAgICAgIFNldmVyYWwgbGF5b3V0IHRlc3Rz IG5lZWRsZXNzbHkgdXNlIGludmFsaWQgSFRNTAorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0 Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9Mzk5ODUKKworICAgICAgICBUaGlzIGNoYW5nZSBkb2VzIG5v dCBjaGFuZ2UgYmVoYXZpb3Igb3IgcmVkdWNlIHRlc3QgY292ZXJhZ2UuCisgICAgICAgIEl0IG9u bHkgbWFrZXMgb3VyIHRlc3Qgc3VpdGUgbW9yZSBjbGVhciBieSBtYWtpbmcgb3VyIG5vbi1wYXJz ZXIKKyAgICAgICAgdGVzdHMgbm90IGJvdGhlciB0byB0ZXN0IHBhcnNlciBxdWlya3Mgd2hpY2gg YXJlIGFscmVhZHkgY292ZXJlZAorICAgICAgICBieSBvdXIgcGFyc2luZyB0ZXN0IHN1aXRlcy4K KworICAgICAgICBUd28gaW52YWxpZCBjYXNlcyB3aGljaCBhcmUgYWxyZWFkeSB0ZXN0ZWQgYnkK KyAgICAgICAgdGhlIGh0bWw1bGliIHRlc3RzIGFzIHdlbGwgYXMgb3RoZXJzLgorICAgICAgICAt IFVuY2xvc2VkIDx0aXRsZT4gc3dhbGxvd2luZyB0aGUgZG9jdW1lbnQgaXMgY292ZXJlZCBieQor ICAgICAgICAgIDwhZG9jdHlwZSBodG1sPjx0aXRsZT5mb28vdGl0bGU+PGxpbms+PC9oZWFkPjxi b2R5PlgKKyAgICAgICAgICBhbmQgb3RoZXIgc3VidGVzdHMgaW4gaHRtbDVsaWIvcmVzb3VyY2Vz L3Rlc3RzMTYuZGF0CisgICAgICAgICAgTWlzc2luZyA8L3RpdGxlPiBpcyBhbHNvIGNvdmVyZWQg YnkgdGhlIChpbnZhbGlkKSB0ZXN0OgorICAgICAgICAgIGZhc3QvanMvbWlzc2luZy10aXRsZS1l bmQtdGFnLWpzLmh0bWwKKyAgICAgICAgLSBVbmNsb3NlZCA8dGFnIHN3YWxsb3dpbmcgdGhlIG5l eHQgdGFnIGlzIGNvdmVyZWQgYnk6CisgICAgICAgICAgZmFzdC9pbnZhbGlkLzAxNi5odG1sLiAg SSBzdXNwZWN0IGl0J3MgYWxzbyBjb3ZlcmVkIGJ5CisgICAgICAgICAgdmFyaW91cyBodG1sNWxp YiB0ZXN0cyBidXQgcHJvdmVkIGRpZmZpY3VsdCB0byBncmVwIGZvci4KKworICAgICAgICAqIGNz czEvYm94X3Byb3BlcnRpZXMvZmxvYXRfZWxlbWVudHNfaW5fc2VyaWVzLmh0bWw6CisgICAgICAg ICAtIFVuY2xvc2VkIDx0ZD4gdGFncyBzd2FsbG93IHRoZSBmb2xsb3dpbmcgPGltZz4gdGFncy4K KyAgICAgICAgICAgVGhpcyBwYXJzZXIgcXVpcmsgaXMgdGVzdGVkIGJ5IHRoZSBIVE1MNSB0ZXN0 cyBhbmQgaXMgYW4KKyAgICAgICAgICAgaW50ZW50aW9uYWwgYmVoYXZpb3IgY2hhbmdlIGFzIHBh cnQgb2YgSFRNTDUuICBObyBuZWVkCisgICAgICAgICAgIGZvciB0aGlzIHRlc3QgdG8gZGVwZW5k IG9uIFdlYktpdCdzIG9sZCBwYXJzZXIgYmVoYXZpb3IuCisgICAgICAgICogZWRpdGluZy9zZWxl Y3Rpb24vaW5hY3RpdmUtc2VsZWN0aW9uLmh0bWw6CisgICAgICAgICAtIFNhbWUgYXMgYWJvdmUu IEhlcmUgPGRpdiBzd2FsbG93cyB0aGUgPHNjcmlwdD4gdGFnLgorICAgICAgICAqIGVkaXRpbmcv c2VsZWN0aW9uL3NlbGVjdC10ZXh0LW92ZXJmbG93LWVsbGlwc2lzLmh0bWw6CisgICAgICAgICAt IEFuIHVuY2xvc2VkIDx0aXRsZT4gd2lsbCBzd2FsbG93IHRoZSB3aG9sZSBkb2N1bWVudC4KKyAg ICAgICAgICAgSW50ZW50aW9uYWwgSFRNTDUgY2hhbmdlLCB0ZXN0ZWQgYnkgSFRNTDUgdGVzdCBz dWl0ZSBhbmQKKyAgICAgICAgICAgbm8gbmVlZCB0byBkZXBlbmQgb24gdGhlIG9sZCB0aXRsZSBi ZWhhdmlvciBoZXJlLgorICAgICAgICAqIGZhc3QvZG9tL3RpdGxlLWNvbnRlbnQtd3JpdGUtc2V0 Lmh0bWw6CisgICAgICAgICAtIEFnYWluLCBubyBuZWVkIHRvIGxlYXZlIGFuIHVuY2xvc2VkIHRp dGxlLgorICAgICAgICAqIGZhc3QvZm9ybXMvc2VsZWN0LXNldC1pbm5lci5odG1sOgorICAgICAg ICAgLSBVbmNsb3NlZCA8Ym9keSBzd2FsbG93cyA8cD4KKyAgICAgICAgKiBmYXN0L2pzL2V4Y2Vw dGlvbi1saW5lbnVtcy1pbi1odG1sLTIuaHRtbDoKKyAgICAgICAgIC0gVW5jbG9zZWQgPHRpdGxl PiBzd2FsbG93cyB0aGUgZG9jdW1lbnQuCisgICAgICAgICogaHRtbDVsaWIvcmVzb3VyY2VzL3dl YmtpdDAxLmRhdDoKKyAgICAgICAgIC0gQWRkZWQgYSB0ZXN0IGZvciA8aW1nPGltZz4gdG8gbWFr ZSBzdXJlIGl0J3MgY292ZXJlZC4KKwogMjAxMC0wNi0wMSAgU2hlcmlmZiBCb3QgIDx3ZWJraXQu cmV2aWV3LmJvdEBnbWFpbC5jb20+CiAKICAgICAgICAgVW5yZXZpZXdlZCwgcm9sbGluZyBvdXQg cjYwNTMwLgpkaWZmIC0tZ2l0IGEvTGF5b3V0VGVzdHMvY3NzMS9ib3hfcHJvcGVydGllcy9mbG9h dF9lbGVtZW50c19pbl9zZXJpZXMuaHRtbCBiL0xheW91dFRlc3RzL2NzczEvYm94X3Byb3BlcnRp ZXMvZmxvYXRfZWxlbWVudHNfaW5fc2VyaWVzLmh0bWwKaW5kZXggNTAwZmYwMjQzYmY2YjI4Yjc4 ZTJkODA2Yzc5NTk0MTM1NzdiZGQ0OS4uNjRmMmFkZDA2NzkxOTRjMjMzNDdjNTg1YjYzZTQ2MTA5 MjZlNjlhZCAxMDA2NDQKLS0tIGEvTGF5b3V0VGVzdHMvY3NzMS9ib3hfcHJvcGVydGllcy9mbG9h dF9lbGVtZW50c19pbl9zZXJpZXMuaHRtbAorKysgYi9MYXlvdXRUZXN0cy9jc3MxL2JveF9wcm9w ZXJ0aWVzL2Zsb2F0X2VsZW1lbnRzX2luX3Nlcmllcy5odG1sCkBAIC01OSw3ICs1OSw3IEBAIFRB QkxFIHttYXJnaW46IDIwcHggMHB4O30KIDx0cj4KIDx0ZCB3aWR0aD0iMjAiIHN0eWxlPSJiYWNr Z3JvdW5kOiBibGFjayI+PElNRyBTUkM9ImJsYW5rLmdpZiIgd2lkdGg9IjIwIiBoZWlnaHQ9IjEi PjwvdGQ+CiA8dGQgd2lkdGg9MTAwIHN0eWxlPSJiYWNrZ3JvdW5kOiB5ZWxsb3ciPjxJTUcgU1JD PSJibGFuay5naWYiIHdpZHRoPSIxMDAiIGhlaWdodD0iMSI+PC90ZD4KLTx0ZCB3aWR0aD0yMCBz dHlsZT0iYmFja2dyb3VuZDogYmxhY2siPElNRyBTUkM9ImJsYW5rLmdpZiIgd2lkdGg9IjIwIiBo ZWlnaHQ9IjEiPjwvdGQ+Cis8dGQgd2lkdGg9MjAgc3R5bGU9ImJhY2tncm91bmQ6IGJsYWNrIj48 SU1HIFNSQz0iYmxhbmsuZ2lmIiB3aWR0aD0iMjAiIGhlaWdodD0iMSI+PC90ZD4KIAogPHRkIHdp ZHRoPTIwIHN0eWxlPSJiYWNrZ3JvdW5kOiB3aGl0ZSI+PElNRyBTUkM9ImJsYW5rLmdpZiIgd2lk dGg9IjIwIiBoZWlnaHQ9IjEiPjwvdGQ+CiAKQEAgLTExMCw3ICsxMTAsNyBAQCBUQUJMRSB7bWFy Z2luOiAyMHB4IDBweDt9CiA8dHI+CiA8dGQgd2lkdGg9IjIwIiBzdHlsZT0iYmFja2dyb3VuZDog YmxhY2siPjxJTUcgU1JDPSJibGFuay5naWYiIHdpZHRoPSIyMCIgaGVpZ2h0PSIxIj48L3RkPgog PHRkIHdpZHRoPTEwMCBzdHlsZT0iYmFja2dyb3VuZDogeWVsbG93Ij48SU1HIFNSQz0iYmxhbmsu Z2lmIiB3aWR0aD0iMTAwIiBoZWlnaHQ9IjEiPjwvdGQ+Ci08dGQgd2lkdGg9MjAgc3R5bGU9ImJh Y2tncm91bmQ6IGJsYWNrIjxJTUcgU1JDPSJibGFuay5naWYiIHdpZHRoPSIyMCIgaGVpZ2h0PSIx Ij48L3RkPgorPHRkIHdpZHRoPTIwIHN0eWxlPSJiYWNrZ3JvdW5kOiBibGFjayI+PElNRyBTUkM9 ImJsYW5rLmdpZiIgd2lkdGg9IjIwIiBoZWlnaHQ9IjEiPjwvdGQ+CiAKIDx0ZCB3aWR0aD0yMCBz dHlsZT0iYmFja2dyb3VuZDogd2hpdGUiPjxJTUcgU1JDPSJibGFuay5naWYiIHdpZHRoPSIyMCIg aGVpZ2h0PSIxIj48L3RkPgogCmRpZmYgLS1naXQgYS9MYXlvdXRUZXN0cy9lZGl0aW5nL3NlbGVj dGlvbi9pbmFjdGl2ZS1zZWxlY3Rpb24uaHRtbCBiL0xheW91dFRlc3RzL2VkaXRpbmcvc2VsZWN0 aW9uL2luYWN0aXZlLXNlbGVjdGlvbi5odG1sCmluZGV4IDk5NzFjM2NhOTMwYmEyYmU4YzY5ZmNi MTNlNzUxZTkxYWEzZjFmZGYuLjZjMTk0Mzk3M2I3ZWI4ZjA2MTZkNDdlYWUxMDdhNGI4OGM1NjMz ZTUgMTAwNjQ0Ci0tLSBhL0xheW91dFRlc3RzL2VkaXRpbmcvc2VsZWN0aW9uL2luYWN0aXZlLXNl bGVjdGlvbi5odG1sCisrKyBiL0xheW91dFRlc3RzL2VkaXRpbmcvc2VsZWN0aW9uL2luYWN0aXZl LXNlbGVjdGlvbi5odG1sCkBAIC02LDcgKzYsNyBAQCBhZnRlciB5b3UndmUgbWFkZSBhIHNlbGVj dGlvbiBpbiBhbm90aGVyIGZyYW1lLjwvcD4KIDxpZnJhbWUgc3JjPSJkYXRhOnRleHQvaHRtbCwg PGRpdiBpZD0naW5wdXQnIGNvbnRlbnRlZGl0YWJsZT0ndHJ1ZSc+c2hvdWxkIGJlIGFuIGluYWN0 aXZlIHNlbGVjdGlvbjwvZGl2PiI+PC9pZnJhbWU+CiA8YnI+CiA8aW5wdXQgaWQ9J2lucHV0JyB0 eXBlPSJ0ZXh0IiB2YWx1ZT0icmFuZG9tIHRleHQiPgotPGRpdgorPGRpdj4KIDxzY3JpcHQ+CiB3 aW5kb3cub25sb2FkID0gZnVuY3Rpb24oKSB7CiAgICAgaWYgKHdpbmRvdy5sYXlvdXRUZXN0Q29u dHJvbGxlcikKZGlmZiAtLWdpdCBhL0xheW91dFRlc3RzL2VkaXRpbmcvc2VsZWN0aW9uL3NlbGVj dC10ZXh0LW92ZXJmbG93LWVsbGlwc2lzLmh0bWwgYi9MYXlvdXRUZXN0cy9lZGl0aW5nL3NlbGVj dGlvbi9zZWxlY3QtdGV4dC1vdmVyZmxvdy1lbGxpcHNpcy5odG1sCmluZGV4IGY5NGZkZTM3ZDQ1 ZDRlMjFjZDkxMWEzZWQ1M2Q2NjA1ODNhNTc4NTcuLjZjZTMxNTVkMjI0Y2YxZjI0ZTg1NGZlYjg5 ZjljMDk1NDMwZWNhNzggMTAwNjQ0Ci0tLSBhL0xheW91dFRlc3RzL2VkaXRpbmcvc2VsZWN0aW9u L3NlbGVjdC10ZXh0LW92ZXJmbG93LWVsbGlwc2lzLmh0bWwKKysrIGIvTGF5b3V0VGVzdHMvZWRp dGluZy9zZWxlY3Rpb24vc2VsZWN0LXRleHQtb3ZlcmZsb3ctZWxsaXBzaXMuaHRtbApAQCAtMSw2 ICsxLDYgQEAKIDxodG1sPgogICA8aGVhZD4KLSAgICA8dGl0bGU+VGVzdCBmb3IgV2ViS2l0IGJ1 ZyAyOTk2ODogU2VsZWN0aW5nIHRleHQgd2l0aCB0ZXh0LW92ZXJmbG93IGVsbGlwc2lzIHNob3Vs ZCBub3Qgc2hvdyBjdXQgb2ZmIHRleHQ8L2hlYWQ+CisgICAgPHRpdGxlPlRlc3QgZm9yIFdlYktp dCBidWcgMjk5Njg6IFNlbGVjdGluZyB0ZXh0IHdpdGggdGV4dC1vdmVyZmxvdyBlbGxpcHNpcyBz aG91bGQgbm90IHNob3cgY3V0IG9mZiB0ZXh0PC90aXRsZT48L2hlYWQ+CiAgIDxzdHlsZSB0eXBl PSJ0ZXh0L2NzcyI+CiAgICAgLmN1dG9mZiB7CiAgICAgICB3aWR0aDogOTVweDsKZGlmZiAtLWdp dCBhL0xheW91dFRlc3RzL2Zhc3QvZG9tL3RpdGxlLWNvbnRlbnQtd3JpdGUtc2V0Lmh0bWwgYi9M YXlvdXRUZXN0cy9mYXN0L2RvbS90aXRsZS1jb250ZW50LXdyaXRlLXNldC5odG1sCmluZGV4IGJm NGQ1ZmU3NGYxZmQwNjQ2NDM3MDNiMjhiNDU1YTkwOTg4NzM3ZmYuLmFjZDhlMTliYjg1YmVhMWQz MzRkNTE5OWM2ZmFlYTI4ODE4MzU5OGMgMTAwNjQ0Ci0tLSBhL0xheW91dFRlc3RzL2Zhc3QvZG9t L3RpdGxlLWNvbnRlbnQtd3JpdGUtc2V0Lmh0bWwKKysrIGIvTGF5b3V0VGVzdHMvZmFzdC9kb20v dGl0bGUtY29udGVudC13cml0ZS1zZXQuaHRtbApAQCAtNSw3ICs1LDcgQEAgaWYgKHdpbmRvdy5s YXlvdXRUZXN0Q29udHJvbGxlcikKICAgICBsYXlvdXRUZXN0Q29udHJvbGxlci5kdW1wQXNUZXh0 KCk7CiBkb2N1bWVudC53cml0ZSgiPHRpdGxlPiIpOwogZG9jdW1lbnQudGl0bGUgPSAiUHJvcGVy dHkiOwotZG9jdW1lbnQud3JpdGUoIldyaXR0ZW4iKTsKK2RvY3VtZW50LndyaXRlKCJXcml0dGVu PC90aXRsZT4iKTsKIAogZnVuY3Rpb24gdGVzdCgpCiB7CmRpZmYgLS1naXQgYS9MYXlvdXRUZXN0 cy9mYXN0L2Zvcm1zL3NlbGVjdC1zZXQtaW5uZXIuaHRtbCBiL0xheW91dFRlc3RzL2Zhc3QvZm9y bXMvc2VsZWN0LXNldC1pbm5lci5odG1sCmluZGV4IDg5ZTIyMzgzMDgxNDUyMjNiOTQ0MGJlYmI0 ZjIwNGZhMGM0ZDg1MmEuLjk3MmExZGNiNTA2MjczZDliZGJhYTg3YTQ1NGQ3MGY5ZDk3YTdjN2Qg MTAwNjQ0Ci0tLSBhL0xheW91dFRlc3RzL2Zhc3QvZm9ybXMvc2VsZWN0LXNldC1pbm5lci5odG1s CisrKyBiL0xheW91dFRlc3RzL2Zhc3QvZm9ybXMvc2VsZWN0LXNldC1pbm5lci5odG1sCkBAIC0y NCw3ICsyNCw3IEBAIGZ1bmN0aW9uIHRlc3QoKQogCiA8L2h0bWw+CiAKLTxib2R5IG9ubG9hZD0i dGVzdCgpIgorPGJvZHkgb25sb2FkPSJ0ZXN0KCkiPgogCiA8cD5UaGlzIHRlc3RzIGEgY2FzZSB3 aGVyZSByZW1vdmluZyBhbiBvcHRpb24gZWxlbWVudCBmcm9tIGEgc2VsZWN0IGJ5IHVzaW5nIGlu bmVySFRNTCB3b3VsZCBjYXVzZSBhIGNyYXNoLjwvcD4KIApkaWZmIC0tZ2l0IGEvTGF5b3V0VGVz dHMvZmFzdC9qcy9leGNlcHRpb24tbGluZW51bXMtaW4taHRtbC0yLmh0bWwgYi9MYXlvdXRUZXN0 cy9mYXN0L2pzL2V4Y2VwdGlvbi1saW5lbnVtcy1pbi1odG1sLTIuaHRtbAppbmRleCA3OGRkNTlm NDk5MWU2NGFlMWVmZTdkZWIzZWY3ZDRjYjg5M2QyM2QyLi42YTM5YWEwNzczNDUzYWExNTNlMmMz OTJlZGRlYWQxMjdlNjMxNjNiIDEwMDY0NAotLS0gYS9MYXlvdXRUZXN0cy9mYXN0L2pzL2V4Y2Vw dGlvbi1saW5lbnVtcy1pbi1odG1sLTIuaHRtbAorKysgYi9MYXlvdXRUZXN0cy9mYXN0L2pzL2V4 Y2VwdGlvbi1saW5lbnVtcy1pbi1odG1sLTIuaHRtbApAQCAtMSw1ICsxLDUgQEAKIDxodG1sPgot PGhlYWQ+PHRpdGxlPnRpdGxlCis8aGVhZD48dGl0bGU+dGl0bGU8L3RpdGxlPgogPGxpbmsgcmVs PSJzdHlsZXNoZWV0IiBocmVmPSJyZXNvdXJjZXMvanMtdGVzdC1zdHlsZS5jc3MiPgogPHNjcmlw dCBzcmM9InJlc291cmNlcy9qcy10ZXN0LXByZS5qcyI+PC9zY3JpcHQ+CiA8L2hlYWQ+Cg==