105407 2012-12-19 02:26:43 -0800 Current error reporting method used by check-layout.js should not affect subsequent sub-tests using checking data-offset-y. 2013-01-29 12:55:52 -0800 1 1 1 Unclassified WebKit Tools / Tests 528+ (Nightly build) Unspecified Unspecified RESOLVED FIXED P2 Normal --- 1 pravind.2k4 pravind.2k4 dglazkov jchaffraix ojan tony webkit.review.bot oldest_to_newest 794469 0 pravind.2k4 2012-12-19 02:26:43 -0800 Description: Consider a testcase which contains a set of sub-tests all checking if the element has correct data-offset-y(offsetTop). When one of the sub-tests fails, the failure msg is appended to the elements container(in our case, its the sub-test). This extra content changes the offsetTop of other subsequent sub-tests and all sub-tests following the failing test case also fail. This becomes confusing/time consuming when either creating a test-case or debugging test failure. 794671 1 180166 pravind.2k4 2012-12-19 07:50:51 -0800 Created attachment 180166 TestCase 794752 2 ojan 2012-12-19 09:40:44 -0800 Maybe we should save error messages + their containers into an array and only write them out at the end of the test? If we wanted to be paranoid, we could do it in a try/finally to ensure that we write out the errors even if the test throws an exception, although, it really shouldn't since all the relevant code is in check-layout.js. 794811 3 tony 2012-12-19 11:00:19 -0800 (In reply to comment #2) > Maybe we should save error messages + their containers into an array and only write them out at the end of the test? If we wanted to be paranoid, we could do it in a try/finally to ensure that we write out the errors even if the test throws an exception, although, it really shouldn't since all the relevant code is in check-layout.js. Some other options: We currently append PASS/FAIL next to the parent node, or if the parent node has the container class, we append next to the parent's parent. That's kind of hacky. We could add an optional param to checkLayout() for the node to append results to (and make the PASS text more clear like listing the id or class of the passing node). Alternately, we could add an attribute like data-log-results="some-div-id". If we see that, we write the results to some-div-id. Anyway, I think any of these approaches are fine. It's a little weird that for some of the existing writing mode tests the word "PASS" ends up being written vertically. 794879 4 ojan 2012-12-19 13:15:44 -0800 (In reply to comment #3) > We currently append PASS/FAIL next to the parent node, or if the parent node has the container class, we append next to the parent's parent. That's kind of hacky. We could add an optional param to checkLayout() for the node to append results to (and make the PASS text more clear like listing the id or class of the passing node). I agree this is less hacky, but it's also harder to use with tests that have lots of test-cases. It's hard to match up the failure to the test case. > Alternately, we could add an attribute like data-log-results="some-div-id". If we see that, we write the results to some-div-id. Same issue, right? Also, I don't like adding more work for people to do when writing these tests. > Anyway, I think any of these approaches are fine. It's a little weird that for some of the existing writing mode tests the word "PASS" ends up being written vertically. I agree that it's unfortunate that the logging output affects the tests and is affected by the tests. We could use shadow-dom with resetStyleInheritance. :) More seriously, with my original proposal, we could put them in a separate container div at the end and absolutely position the elements after the test-case to minimize the affects between tests and their output? I'd really like to keep the behavior of having the failure output be next to the test case if possible. 794891 5 tony 2012-12-19 13:33:10 -0800 All my suggestions were for an optional parameter. The default behavior would be to do what we currently do. I don't think there are that many tests that need to set a custom output location. We shouldn't need to edit or rebaseline any existing tests. I think it would be hard to know where to place the result using absolute positioning. The other option would be to do nothing. In the attached test case, you can easily avoid the problem by doing checkLayout('body'), although you only get a single PASS or FAIL. If you want a separate PASS/FAIL for each test case, you could wrap each test in a position:relative div. We do that for a lot of the flexbox tests. 794987 6 180234 pravind.2k4 2012-12-19 15:03:22 -0800 Created attachment 180234 Proped Patch 794990 7 180234 pravind.2k4 2012-12-19 15:05:58 -0800 Comment on attachment 180234 Proped Patch Reversing the order in which nodes are processed should ideally fix the issue. @Ojan, @Tony please let me know ur opinions. Also am not sure where to add a test case for this fix... Suggestion req on the same. 795011 8 180234 webkit.review.bot 2012-12-19 15:55:15 -0800 Comment on attachment 180234 Proped Patch Attachment 180234 did not pass chromium-ews (chromium-xvfb): Output: http://queues.webkit.org/results/15401804 New failing tests: css3/flexbox/flex-algorithm-with-margins.html css3/flexbox/box-sizing.html css3/flexbox/flex-flow-margins.html css3/flexbox/flex-align.html animations/suspend-resume-animation-events.html css3/flexbox/flex-align-percent-height.html css3/flexbox/flex-align-vertical-writing-mode.html css3/flexbox/align-absolute-child.html css3/flexbox/flex-item-min-size.html css3/flexbox/flex-no-flex.html css3/flexbox/flex-algorithm.html css3/flexbox/columns-height-set-via-top-bottom.html css3/flexbox/flex-flow.html css3/flexbox/flex-align-column.html css3/flexbox/flex-algorithm-min-max.html css3/flexbox/flex-item-child-overflow.html css3/flexbox/flex-flow-margins-auto-size.html css3/flexbox/flex-align-max.html css3/flexbox/auto-height-dynamic.html css3/flexbox/flex-flow-auto-margins.html css3/flexbox/flex-flow-padding.html css3/flexbox/columns-auto-size.html css3/flexbox/flex-align-end.html css3/flexbox/flex-flow-orientations.html css3/flexbox/flex-flow-overflow.html css3/flexbox/flex-align-stretch.html css3/flexbox/flex-flow-auto-margins-no-available-space.html css3/flexbox/flex-flow-border.html css3/flexbox/box-sizing-min-max-sizes.html css3/flexbox/flex-justify-content.html 795068 9 180234 buildbot 2012-12-19 17:38:47 -0800 Comment on attachment 180234 Proped Patch Attachment 180234 did not pass mac-ews (mac): Output: http://queues.webkit.org/results/15403871 New failing tests: css3/flexbox/flex-algorithm-with-margins.html css3/flexbox/box-sizing.html css3/flexbox/flex-flow-margins.html css3/flexbox/flex-align.html css3/flexbox/flex-align-percent-height.html css3/flexbox/flex-align-vertical-writing-mode.html css3/flexbox/align-absolute-child.html css3/flexbox/flex-item-min-size.html css3/flexbox/flex-no-flex.html css3/flexbox/flex-algorithm.html css3/flexbox/columns-height-set-via-top-bottom.html css3/flexbox/flex-flow.html css3/flexbox/flex-align-column.html css3/flexbox/flex-algorithm-min-max.html css3/flexbox/flex-item-child-overflow.html css3/flexbox/flex-flow-margins-auto-size.html css3/flexbox/flex-align-max.html css3/flexbox/auto-height-dynamic.html css3/flexbox/flex-flow-auto-margins.html css3/flexbox/flex-flow-padding.html css3/flexbox/columns-auto-size.html css3/flexbox/flex-align-end.html css3/flexbox/flex-flow-orientations.html css3/flexbox/flex-flow-overflow.html compositing/overflow/overflow-auto-with-touch-toggle.html css3/flexbox/flex-align-stretch.html css3/flexbox/flex-flow-auto-margins-no-available-space.html css3/flexbox/flex-flow-border.html css3/flexbox/box-sizing-min-max-sizes.html css3/flexbox/flex-justify-content.html 795551 10 180234 tony 2012-12-20 10:35:30 -0800 Comment on attachment 180234 Proped Patch CONSOLE MESSAGE: line 163: Uncaught TypeError: [object Array] is not a function is what the tests are spewing. 795697 11 180392 pravind.2k4 2012-12-20 12:56:49 -0800 Created attachment 180392 Patch 795698 12 pravind.2k4 2012-12-20 12:58:29 -0800 (In reply to comment #10) > (From update of attachment 180234 [details]) > CONSOLE MESSAGE: line 163: Uncaught TypeError: [object Array] is not a function is what the tests are spewing. > I extremely sorry for this. Have uploaded another patch, with which test case do not seem to fail. 795754 13 180392 tony 2012-12-20 13:42:31 -0800 Comment on attachment 180392 Patch Seems fine to me. Ojan might have an opinion, but he won't be back from vacation until January. 819195 14 180392 ojan 2013-01-29 12:25:40 -0800 Comment on attachment 180392 Patch I'm a bit torn on whether this is the best solution. I still somewhat prefer the approach of saving everything in an error array and then writing out all the errors at the end. But, I don't see any immediate problems with this simpler approach. 819238 15 180392 webkit.review.bot 2013-01-29 12:55:48 -0800 Comment on attachment 180392 Patch Clearing flags on attachment: 180392 Committed r141147: <http://trac.webkit.org/changeset/141147> 819239 16 webkit.review.bot 2013-01-29 12:55:52 -0800 All reviewed patches have been landed. Closing bug. 180166 2012-12-19 07:50:51 -0800 2012-12-19 07:50:51 -0800 TestCase test.html text/html 1979 pravind.2k4 PCFET0NUWVBFIGh0bWw+CjxodG1sPgo8aGVhZD4KPHNjcmlwdCBzcmM9ImNoZWNrLWxheW91dC5q cyI+PC9zY3JpcHQ+CjxzdHlsZT4KLmlubGluZSB7CiAgd2lkdGg6MjAwcHg7CiAgaGVpZ2h0OjIw cHg7CiAgYmFja2dyb3VuZDpncmVlbjsKICBkaXNwbGF5OmJsb2NrOwogIGxpbmUtaGVpZ2h0OjE7 CiAgZm9udC1zaXplOiAxNnB4Owp9Cjwvc3R5bGU+CjxzY3JpcHQ+CmZ1bmN0aW9uIHByZXBhcmVU ZXN0Q2FzZSgpIHsKICAgIHZhciBzYW1wbGVzID0gZG9jdW1lbnQuZ2V0RWxlbWVudHNCeUNsYXNz TmFtZSgnc2FtcGxlJyk7CiAgICB2YXIgb2Zmc2V0MSA9IHNhbXBsZXNbMF0ub2Zmc2V0VG9wOwog ICAgdmFyIG9mZnNldDIgPSBzYW1wbGVzWzFdLm9mZnNldFRvcDsKICAgIGNoZWNrTGF5b3V0KCcu c2FtcGxlJyk7CiAgICB2YXIgZGlmZiA9IHNhbXBsZXNbMV0ub2Zmc2V0VG9wIC0gb2Zmc2V0MjsK ICAgIHNhbXBsZXMgPSBkb2N1bWVudC5nZXRFbGVtZW50QnlJZCgnZHVtbXknKTsKICAgIGRvY3Vt ZW50LmJvZHkucmVtb3ZlQ2hpbGQoc2FtcGxlcyk7CiAgICByZXR1cm4gZGlmZjsKfQpmdW5jdGlv biBydW5UZXN0KCkgewogICAgdmFyIGRpZmYgPSBwcmVwYXJlVGVzdENhc2UoKTsKICAgIHZhciBl bGVtZW50cyA9IGRvY3VtZW50LmdldEVsZW1lbnRzQnlDbGFzc05hbWUoJ2lubGluZScpOwogICAg QXJyYXkucHJvdG90eXBlLmZvckVhY2guY2FsbChlbGVtZW50cywgZnVuY3Rpb24oZWxlbWVudCkg ewogICAgICAgIGVsZW1lbnQuc2V0QXR0cmlidXRlKCJkYXRhLW9mZnNldC15IiwgZWxlbWVudC5v ZmZzZXRUb3AgKyBkaWZmKTsKICAgIH0pOwogICAgZWxlbWVudHNbMF0uc2V0QXR0cmlidXRlKCJk YXRhLW9mZnNldC15IiwgZWxlbWVudHNbMF0ub2Zmc2V0VG9wKTsKICAgIHZhciBlbGVtZW50ID0g ZG9jdW1lbnQuZ2V0RWxlbWVudEJ5SWQoJ3Rlc3QnKTsKICAgIGVsZW1lbnQuc2V0QXR0cmlidXRl KCJkYXRhLW9mZnNldC15IiwgIi0xIik7CgogICAgY2hlY2tMYXlvdXQoJy5pbmxpbmUnKTsKfQo8 L3NjcmlwdD4KPC9oZWFkPgo8Ym9keSBvbmxvYWQ9InJ1blRlc3QoKTsiPgo8ZGl2IGlkPSJkdW1t eSI+CjxkaXYgY2xhc3M9ImlubGluZSBzYW1wbGUiIGRhdGEtb2Zmc2V0LXk+IEZpbGxlciB0ZXN0 Y2FzZTwvZGl2Pgo8ZGl2IGNsYXNzPSJpbmxpbmUgc2FtcGxlIiBkYXRhLW9mZnNldC15PiBGaWxs ZXIgdGVzdGNhc2U8L2Rpdj4KPC9kaXY+CjxkaXYgY2xhc3M9ImlubGluZSIgZGF0YS1vZmZzZXQt eT4gRmlsbGVyIHRlc3RjYXNlPC9kaXY+CjxkaXYgY2xhc3M9ImlubGluZSIgZGF0YS1vZmZzZXQt eT5GaWxsZXIgdGVzdGNhc2U8L2Rpdj4KPGRpdiBpZD0idGVzdCIgY2xhc3M9ImlubGluZSIgZGF0 YS1vZmZzZXQteT5UaGlzIHRlc3RjYXNlIHNob3VsZCBmYWlsPC9kaXY+CjxkaXYgY2xhc3M9Imlu bGluZSIgZGF0YS1vZmZzZXQteT5UaGlzIHRlc3RjYXNlIHNob3VsZCBub3QgZmFpbDwvZGl2Pgo8 ZGl2IGNsYXNzPSJpbmxpbmUiIGRhdGEtb2Zmc2V0LXk+VGhpcyB0ZXN0Y2FzZSBzaG91bGQgbm90 IGZhaWw8L2Rpdj4KPGRpdiBjbGFzcz0iaW5saW5lIiBkYXRhLW9mZnNldC15PlRoaXMgdGVzdGNh c2Ugc2hvdWxkIG5vdCBmYWlsPC9kaXY+CjxkaXY+CjxiPkFib3V0IHRoZSBUZXN0Q2FzZTogPC9i PgpUaGUgc3ViLXRlc3RjYXNlcyBhZnRlciAiVGhpcyB0ZXN0Y2FzZSBzaG91bGQgZmFpbCIgc2hv dWxkIG5vdCBiZSBhZmZlY3RlZCBieSB0aGUgZmFpbHVyZSBtZXNzYWdlLiBUaHVzIHN1YnNlcXVl bnQgdGVzdGNhc2VzIHNob3VsZCBub3QgZmFpbC4KQWxzbyBwcmVwYXJlVGVzdENhc2UoKSB3YXMg dXNlZCBhcyB0aGUgc3VjY2VzcyBtZXNzYWdlKCJQQVNTIikgb2YgdGhlIHN1Yi10ZXN0Y2FzZXMg IkZpbGxlciB0ZXN0Y2FzZSIgYWxzbyBhZmZlY3RlZCB0aGUgb2Zmc2V0VG9wIG9mIHN1YnNlcXVl bnQgc3ViLXRlc3RjYXNlcy4KPC9kaXY+CjwvYm9keT4KPC9odG1sPgo= 180234 2012-12-19 15:03:22 -0800 2012-12-20 12:56:44 -0800 Proped Patch bug-105407-20121220043557.patch text/plain 1758 pravind.2k4 SW5kZXg6IExheW91dFRlc3RzL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBMYXlvdXRUZXN0cy9D aGFuZ2VMb2cJKHJldmlzaW9uIDEzODE5NSkKKysrIExheW91dFRlc3RzL0NoYW5nZUxvZwkod29y a2luZyBjb3B5KQpAQCAtMSwzICsxLDE4IEBACisyMDEyLTEyLTE5ICBQcmF2aW4gRCAgPHByYXZp bmQuMms0QGdtYWlsLmNvbT4KKworICAgICAgICBDdXJyZW50IGVycm9yIHJlcG9ydGluZyBtZXRo b2QgdXNlZCBieSBjaGVjay1sYXlvdXQuanMgc2hvdWxkIG5vdCBhZmZlY3Qgc3Vic2VxdWVudCBz dWItdGVzdHMgdXNpbmcgY2hlY2tpbmcgZGF0YS1vZmZzZXQteS4KKyAgICAgICAgaHR0cHM6Ly9i dWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTEwNTQwNworCisgICAgICAgIFJldmlld2Vk IGJ5IE5PQk9EWSAoT09QUyEpLgorCisgICAgICAgIFdoZW4gYSB0ZXN0Y2FzZSBpcyBwcm9jZXNz ZWQgYnkgY2hlY2stbGF5b3V0LmpzLCB0aGUgcmVzdWx0IGlzIHdyaXR0ZW4ganVzdCBhZnRlciB0 aGUgbm9kZSBiZWluZyBwcm9jZXNzZWQuIFRoaXMgY2F1c2VzCisgICAgICAgIG9mZnNldFRvcCBv ZiBzdWJzZXF1ZW50IHN1Yi10ZXN0Y2FzZXMgdG8gYWx0ZXJlZC4gSWYgaG93ZXZlciBpZiB3ZSBw cm9jZXNzIHRoZSBub2RlcyBpbiB0aGUgcmV2ZXJzZSBvcmRlciBpbiB3aGljaCB0aGV5IAorICAg ICAgICBhcHBlYXIgaW4gdGhlIERPTSB0cmVlLCB0aGUgcmVzdWx0IG9mIG5vZGUgYmVpbmcgcHJv Y2Vzc2VkIHdpbGwgbm90IGFmZmVjdCBvdGhlciBub2RlcyBvZmZzZXRUb3AuCisgICAgICAgIFRo aXMgcGF0Y2ggcHJvcG9zZXMgdG8gZG8gdGhlIHNhbWUuCisKKyAgICAgICAgKiByZXNvdXJjZXMv Y2hlY2stbGF5b3V0LmpzOgorICAgICAgICAoLik6CisKIDIwMTItMTItMTkgIExldmkgV2VpbnRy YXViICA8bGV2aXdAY2hyb21pdW0ub3JnPgogCiAgICAgICAgIFJlLXNraXBwaW5nIHRvdWNoLWlu cHV0LWVsZW1lbnQtY2hhbmctZG9jdW1lbnRzLmh0bWwgc2luY2UgaXQncyBzdGlsbApJbmRleDog TGF5b3V0VGVzdHMvcmVzb3VyY2VzL2NoZWNrLWxheW91dC5qcwo9PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBMYXlv dXRUZXN0cy9yZXNvdXJjZXMvY2hlY2stbGF5b3V0LmpzCShyZXZpc2lvbiAxMzgxMjgpCisrKyBM YXlvdXRUZXN0cy9yZXNvdXJjZXMvY2hlY2stbGF5b3V0LmpzCSh3b3JraW5nIGNvcHkpCkBAIC0x NTgsNyArMTU4LDkgQEAgd2luZG93LmNoZWNrTGF5b3V0ID0gZnVuY3Rpb24oc2VsZWN0b3JMaQog ICAgICAgICByZXR1cm47CiAgICAgfQogICAgIHZhciBub2RlcyA9IGRvY3VtZW50LnF1ZXJ5U2Vs ZWN0b3JBbGwoc2VsZWN0b3JMaXN0KTsKLSAgICBBcnJheS5wcm90b3R5cGUuZm9yRWFjaC5jYWxs KG5vZGVzLCBmdW5jdGlvbihub2RlKSB7CisgICAgbm9kZXMgPSBBcnJheS5wcm90b3R5cGUuc2xp Y2UuY2FsbChub2Rlcyk7CisgICAgbm9kZXMucmV2ZXJzZSgpOworICAgIG5vZGVzLmZvckVhY2go bm9kZXMsIGZ1bmN0aW9uKG5vZGUpIHsKICAgICAgICAgdmFyIGZhaWx1cmVzID0gW107CiAgICAg ICAgIGNoZWNrRXhwZWN0ZWRWYWx1ZXMobm9kZS5wYXJlbnROb2RlLCBmYWlsdXJlcyk7CiAgICAg ICAgIGNoZWNrU3VidHJlZUV4cGVjdGVkVmFsdWVzKG5vZGUsIGZhaWx1cmVzKTsK 180392 2012-12-20 12:56:49 -0800 2013-01-29 12:55:48 -0800 Patch bug-105407-20121221022927.patch text/plain 1583 pravind.2k4 SW5kZXg6IExheW91dFRlc3RzL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBMYXlvdXRUZXN0cy9D aGFuZ2VMb2cJKHJldmlzaW9uIDEzODI4NykKKysrIExheW91dFRlc3RzL0NoYW5nZUxvZwkod29y a2luZyBjb3B5KQpAQCAtMSwzICsxLDE2IEBACisyMDEyLTEyLTIwICBQcmF2aW4gRCAgPHByYXZp bmQuMms0QGdtYWlsLmNvbT4KKworICAgICAgICBDdXJyZW50IGVycm9yIHJlcG9ydGluZyBtZXRo b2QgdXNlZCBieSBjaGVjay1sYXlvdXQuanMgc2hvdWxkIG5vdCBhZmZlY3Qgc3Vic2VxdWVudCBz dWItdGVzdHMgdXNpbmcgY2hlY2tpbmcgZGF0YS1vZmZzZXQteS4KKyAgICAgICAgaHR0cHM6Ly9i dWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTEwNTQwNworCisgICAgICAgIFJldmlld2Vk IGJ5IE5PQk9EWSAoT09QUyEpLgorCisgICAgICAgIFdoZW4gYSB0ZXN0Y2FzZSBpcyBwcm9jZXNz ZWQgYnkgY2hlY2stbGF5b3V0LmpzLCB0aGUgcmVzdWx0IGlzIHdyaXR0ZW4ganVzdCBhZnRlciB0 aGUgbm9kZSBiZWluZyBwcm9jZXNzZWQuIFRoaXMgY2F1c2VzCisgICAgICAgIG9mZnNldFRvcCBv ZiBzdWJzZXF1ZW50IHN1Yi10ZXN0Y2FzZXMgdG8gYWx0ZXJlZC4gSWYgaG93ZXZlciBpZiB3ZSBw cm9jZXNzIHRoZSBub2RlcyBpbiB0aGUgcmV2ZXJzZSBvcmRlciBpbiB3aGljaCB0aGV5IAorICAg ICAgICBhcHBlYXIgaW4gdGhlIERPTSB0cmVlLCB0aGUgcmVzdWx0IG9mIG5vZGUgYmVpbmcgcHJv Y2Vzc2VkIHdpbGwgbm90IGFmZmVjdCB0aGUgb2Zmc2V0VG9wIG9mIHN1c2VxdWVudCBub2Rlcy4K KworICAgICAgICAqIHJlc291cmNlcy9jaGVjay1sYXlvdXQuanM6CisKIDIwMTItMTItMjAgIFJ5 b3N1a2UgTml3YSAgPHJuaXdhQHdlYmtpdC5vcmc+CiAKICAgICAgICAgQWRkIGEgTWFjIHRlc3Qg ZXhwZWN0YXRpb24gcGVyIGJ1ZyAxMDQ4NDguCkluZGV4OiBMYXlvdXRUZXN0cy9yZXNvdXJjZXMv Y2hlY2stbGF5b3V0LmpzCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0tIExheW91dFRlc3RzL3Jlc291cmNlcy9jaGVj ay1sYXlvdXQuanMJKHJldmlzaW9uIDEzNzczNikKKysrIExheW91dFRlc3RzL3Jlc291cmNlcy9j aGVjay1sYXlvdXQuanMJKHdvcmtpbmcgY29weSkKQEAgLTE1OCw2ICsxNTgsOCBAQCB3aW5kb3cu Y2hlY2tMYXlvdXQgPSBmdW5jdGlvbihzZWxlY3RvckxpCiAgICAgICAgIHJldHVybjsKICAgICB9 CiAgICAgdmFyIG5vZGVzID0gZG9jdW1lbnQucXVlcnlTZWxlY3RvckFsbChzZWxlY3Rvckxpc3Qp OworICAgIG5vZGVzID0gQXJyYXkucHJvdG90eXBlLnNsaWNlLmNhbGwobm9kZXMpOworICAgIG5v ZGVzLnJldmVyc2UoKTsKICAgICBBcnJheS5wcm90b3R5cGUuZm9yRWFjaC5jYWxsKG5vZGVzLCBm dW5jdGlvbihub2RlKSB7CiAgICAgICAgIHZhciBmYWlsdXJlcyA9IFtdOwogICAgICAgICBjaGVj a0V4cGVjdGVkVmFsdWVzKG5vZGUucGFyZW50Tm9kZSwgZmFpbHVyZXMpOwo=