75417 2012-01-01 11:57:55 -0800 NRWT should use test_expectation.txt on wk2 platforms 2012-01-06 00:14:22 -0800 1 1 1 Unclassified WebKit WebKit2 528+ (Nightly build) Unspecified Unspecified RESOLVED FIXED P2 Normal --- 1 kbalazs ossy abarth dpranke eric jchaffraix ojan ossy robert webkit.review.bot oldest_to_newest 528362 0 kbalazs 2012-01-01 11:57:55 -0800 Tests added in http://trac.webkit.org/changeset/103875 has metric differences between Qt-WK1/2 which is weird. Need to investigate wetter it can be fixed without diverging the expectations. An example diff: --- /ramdisk/qt-linux-32-release-webkit2/build/layout-test-results/fast/css/caption-width-absolute-position-offset-top-expected.txt +++ /ramdisk/qt-linux-32-release-webkit2/build/layout-test-results/fast/css/caption-width-absolute-position-offset-top-actual.txt @@ -10,7 +10,7 @@ RenderTableSection {TBODY} at (0,2) size 106x36 RenderTableRow {TR} at (0,0) size 106x36 RenderTableCell {TD} at (0,15) size 106x6 [border: (2px solid #FFA500)] [r=0 c=0 rs=1 cs=1] -layer at (8,100) size 56x20 - RenderBlock (positioned) {CAPTION} at (8,100) size 56x20 [border: (4px solid #008000)] - RenderText {#text} at (4,4) size 48x12 +layer at (8,100) size 56x21 + RenderBlock (positioned) {CAPTION} at (8,100) size 56x21 [border: (4px solid #008000)] + RenderText {#text} at (4,4) size 48x13 text run at (4,4) width 48: "xxxx" 528363 1 kbalazs 2012-01-01 12:14:18 -0800 As a first I skipped them in http://trac.webkit.org/changeset/103881. 528455 2 ossy 2012-01-02 03:25:12 -0800 See http://build.webkit.sed.hu/results/x86-32%20Linux%20Qt%20Release%20WebKit2/r103875%20%2817825%29/results.html for details. 528469 3 kbalazs 2012-01-02 05:16:29 -0800 Find out that these needs platform results. The strange thing is that these tests was supposed to be skipped on Qt by test_expectations, but they still failed on wk2. This is likely an NRWT bug. *** This bug has been marked as a duplicate of bug 74888 *** 528489 4 ossy 2012-01-02 06:09:58 -0800 path_to_test_expectations_file() in webkit.py: def path_to_test_expectations_file(self): # test_expectations are always in mac/ not mac-leopard/ by convention, hence we use port_name instead of name(). expectations_directory = self._wk2_port_name() if self.get_option('webkit_test_runner') else self.port_name return self._filesystem.join(self._webkit_baseline_path(expectations_directory), 'test_expectations.txt') It returns qt-wk2 incorrectly. As far as I know nobody uses platform-wk2/test_expectations.txt, because there isn't any checked in test_expectations.txt in these directories. 528492 5 ossy 2012-01-02 06:27:03 -0800 This change introduced in http://trac.webkit.org/changeset/90077 . See https://bugs.webkit.org/show_bug.cgi?id=63501#c6 for details. This change was implemented to avoid duplicated expectations (mac, mac-wk2, test_expectations.txt vs Skipped files) But now there is separated wk2/Skipped file instead of including mac-wk2/Skipped for all wk2 port and now NRWT handles duplicated expectations correctly - Skipped list entry is stronger than a test_expectation.txt entry. So I propose reverting this part of the original change to enable test_expecatitons.txt on wk2 platforms too. 528494 6 120878 ossy 2012-01-02 06:32:42 -0800 Created attachment 120878 Patch 528548 7 120878 eric 2012-01-02 09:55:17 -0800 Comment on attachment 120878 Patch Please test this. 528718 8 120930 ossy 2012-01-03 03:54:05 -0800 Created attachment 120930 Patch 529692 9 120930 dpranke 2012-01-04 13:13:51 -0800 Comment on attachment 120930 Patch So is the intent that we have a single test_expectations.txt file for both wk1 and wk2? Do we need to distinguish different results for each, and, if so, how will you do that? 530106 10 kbalazs 2012-01-05 01:49:42 -0800 (In reply to comment #9) > (From update of attachment 120930 [details]) > So is the intent that we have a single test_expectations.txt file for both wk1 and wk2? Do we need to distinguish different results for each, and, if so, how will you do that? This is how NRWT works. It is a question wetter it should be improved - for example kkristof has a patch for cascading expectations - but it does not belong to this bug. The patch is an easy fix for a real bug, so I hope we can short-circuit this story. 530281 11 ossy 2012-01-05 07:49:15 -0800 (In reply to comment #9) > (From update of attachment 120930 [details]) > So is the intent that we have a single test_expectations.txt file for both wk1 and wk2? Do we need to distinguish different results for each, and, if so, how will you do that? There are only 4 test_expectations.txt: qt, gtk, chromium, mac and win. Now nobody uses actively test_expectations.txt except chromuium port, everybody prefers Skipped lists (because it is simpler and you can use cascaded Skipped lists.) And we can add different results into qt,qt-wk2 ; mac,mac-wk2 and gtk,qtk-wk2 dirs irrespectively of test_expectations.txt But now NRWT ignores test_expectations.txt files for wk2 platforms, which isn't correct. Chromium guys usually add tests into test_expectations.txt not to break bots and let the port maintainters do rebaselines themselves. 530378 12 120930 eric 2012-01-05 10:14:25 -0800 Comment on attachment 120930 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=120930&action=review > Tools/Scripts/webkitpy/layout_tests/port/webkit_unittest.py:97 > + def test_path_to_test_expectations_file(self): > + port = TestWebKitPort() > + self.assertEqual(port.path_to_test_expectations_file(), '/mock-checkout/LayoutTests/platform/testwebkitport/test_expectations.txt') I don't see how this tests your change. You want to test wha thappens in the wk2 case, no? 530425 13 kbalazs 2012-01-05 10:49:00 -0800 (In reply to comment #12) > (From update of attachment 120930 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=120930&action=review > > > Tools/Scripts/webkitpy/layout_tests/port/webkit_unittest.py:97 > > + def test_path_to_test_expectations_file(self): > > + port = TestWebKitPort() > > + self.assertEqual(port.path_to_test_expectations_file(), '/mock-checkout/LayoutTests/platform/testwebkitport/test_expectations.txt') > > I don't see how this tests your change. You want to test wha thappens in the wk2 case, no? Ossy, I recommend to test wetter the file exist, in the wk2 case. At least that way it can not regress again. 530493 14 120930 dpranke 2012-01-05 11:47:14 -0800 Comment on attachment 120930 Patch The change in webkit.py looks fine. Please add tests for both with and without the --webkit-test-runner option just to get coverage of both options. 530595 15 121324 ossy 2012-01-05 13:40:58 -0800 Created attachment 121324 Patch 530619 16 121324 dpranke 2012-01-05 14:11:22 -0800 Comment on attachment 121324 Patch looks good. thank you! 530967 17 121324 ossy 2012-01-06 00:14:13 -0800 Comment on attachment 121324 Patch Clearing flags on attachment: 121324 Committed r104273: <http://trac.webkit.org/changeset/104273> 530968 18 ossy 2012-01-06 00:14:22 -0800 All reviewed patches have been landed. Closing bug. 120878 2012-01-02 06:32:42 -0800 2012-01-03 03:53:58 -0800 Patch bug-75417-20120102063239.patch text/plain 1684 ossy U3VidmVyc2lvbiBSZXZpc2lvbjogMTAzODk5CmRpZmYgLS1naXQgYS9Ub29scy9DaGFuZ2VMb2cg Yi9Ub29scy9DaGFuZ2VMb2cKaW5kZXggNTg3ZDI5NGFjY2NkMzQ0M2E0NGE2Y2M1MjUxMjFiZGM1 Nzc4MTQ0OC4uZDkwMGIxMDU1NmVmM2UzYTkzOTQ5ZTY3N2E4ZmJiZDU1NTUzYjg1MyAxMDA2NDQK LS0tIGEvVG9vbHMvQ2hhbmdlTG9nCisrKyBiL1Rvb2xzL0NoYW5nZUxvZwpAQCAtMSwzICsxLDEz IEBACisyMDEyLTAxLTAyICBDc2FiYSBPc3p0cm9nb27DoWMgIDxvc3N5QHdlYmtpdC5vcmc+CisK KyAgICAgICAgTlJXVCBzaG91bGQgdXNlIHRlc3RfZXhwZWN0YXRpb24udHh0IG9uIHdrMiBwbGF0 Zm9ybXMKKyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTc1 NDE3CisKKyAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISkuCisKKyAgICAgICAgKiBT Y3JpcHRzL3dlYmtpdHB5L2xheW91dF90ZXN0cy9wb3J0L3dlYmtpdC5weToKKyAgICAgICAgKHBh dGhfdG9fdGVzdF9leHBlY3RhdGlvbnNfZmlsZSk6CisKIDIwMTEtMTItMzEgIERhbiBCZXJuc3Rl aW4gIDxtaXR6QGFwcGxlLmNvbT4KIAogICAgICAgICBNaW5pQnJvd3NlciBzZXRzIHRoZSBXS1Zp ZXfigJlzIGZyYW1lIGluY29ycmVjdGx5CmRpZmYgLS1naXQgYS9Ub29scy9TY3JpcHRzL3dlYmtp dHB5L2xheW91dF90ZXN0cy9wb3J0L3dlYmtpdC5weSBiL1Rvb2xzL1NjcmlwdHMvd2Via2l0cHkv bGF5b3V0X3Rlc3RzL3BvcnQvd2Via2l0LnB5CmluZGV4IDE5ODVhNGJkZWE3NTdhMTAwMTNkMmIy MTBmZTM4Njg5NjZiY2RjYWQuLjYxMWVjMGYwMWJlYjUxNjk2MTU2ZDg1ZDU3ODFhMzZhMzIyM2E5 YTcgMTAwNjQ0Ci0tLSBhL1Rvb2xzL1NjcmlwdHMvd2Via2l0cHkvbGF5b3V0X3Rlc3RzL3BvcnQv d2Via2l0LnB5CisrKyBiL1Rvb2xzL1NjcmlwdHMvd2Via2l0cHkvbGF5b3V0X3Rlc3RzL3BvcnQv d2Via2l0LnB5CkBAIC03OCw4ICs3OCw3IEBAIGNsYXNzIFdlYktpdFBvcnQoUG9ydCk6CiAKICAg ICBkZWYgcGF0aF90b190ZXN0X2V4cGVjdGF0aW9uc19maWxlKHNlbGYpOgogICAgICAgICAjIHRl c3RfZXhwZWN0YXRpb25zIGFyZSBhbHdheXMgaW4gbWFjLyBub3QgbWFjLWxlb3BhcmQvIGJ5IGNv bnZlbnRpb24sIGhlbmNlIHdlIHVzZSBwb3J0X25hbWUgaW5zdGVhZCBvZiBuYW1lKCkuCi0gICAg ICAgIGV4cGVjdGF0aW9uc19kaXJlY3RvcnkgPSBzZWxmLl93azJfcG9ydF9uYW1lKCkgaWYgc2Vs Zi5nZXRfb3B0aW9uKCd3ZWJraXRfdGVzdF9ydW5uZXInKSBlbHNlIHNlbGYucG9ydF9uYW1lCi0g ICAgICAgIHJldHVybiBzZWxmLl9maWxlc3lzdGVtLmpvaW4oc2VsZi5fd2Via2l0X2Jhc2VsaW5l X3BhdGgoZXhwZWN0YXRpb25zX2RpcmVjdG9yeSksICd0ZXN0X2V4cGVjdGF0aW9ucy50eHQnKQor ICAgICAgICByZXR1cm4gc2VsZi5fZmlsZXN5c3RlbS5qb2luKHNlbGYuX3dlYmtpdF9iYXNlbGlu ZV9wYXRoKHNlbGYucG9ydF9uYW1lKSwgJ3Rlc3RfZXhwZWN0YXRpb25zLnR4dCcpCiAKICAgICBk ZWYgX3BvcnRfZmxhZ19mb3Jfc2NyaXB0cyhzZWxmKToKICAgICAgICAgIyBUaGlzIGlzIG92ZXJy cmlkZW4gYnkgcG9ydHMgd2hpY2ggbmVlZCBhIGZsYWcgcGFzc2VkIHRvIHNjcmlwdHMgdG8gZGlz dGluZ3Vpc2ggdGhlIHVzZSBvZiB0aGF0IHBvcnQuCg== 120930 2012-01-03 03:54:05 -0800 2012-01-05 13:40:49 -0800 Patch bug-75417-20120103035404.patch text/plain 2824 ossy U3VidmVyc2lvbiBSZXZpc2lvbjogMTAzOTMyCmRpZmYgLS1naXQgYS9Ub29scy9DaGFuZ2VMb2cg Yi9Ub29scy9DaGFuZ2VMb2cKaW5kZXggZTJlMjFiYjQ5MTY5MzhkNzY5ZjBjMThhZGVmNjI3ODE1 NmE1ZTYxYS4uMzQwYWE4MWM0MzUzOWYyMzU1M2Y3YTIzOWNlOWJhMGRhMGI0Nzk0NCAxMDA2NDQK LS0tIGEvVG9vbHMvQ2hhbmdlTG9nCisrKyBiL1Rvb2xzL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE1 IEBACisyMDEyLTAxLTAzICBDc2FiYSBPc3p0cm9nb27DoWMgIDxvc3N5QHdlYmtpdC5vcmc+CisK KyAgICAgICAgTlJXVCBzaG91bGQgdXNlIHRlc3RfZXhwZWN0YXRpb24udHh0IG9uIHdrMiBwbGF0 Zm9ybXMKKyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTc1 NDE3CisKKyAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISkuCisKKyAgICAgICAgKiBT Y3JpcHRzL3dlYmtpdHB5L2xheW91dF90ZXN0cy9wb3J0L3dlYmtpdC5weToKKyAgICAgICAgKHBh dGhfdG9fdGVzdF9leHBlY3RhdGlvbnNfZmlsZSk6CisgICAgICAgICogU2NyaXB0cy93ZWJraXRw eS9sYXlvdXRfdGVzdHMvcG9ydC93ZWJraXRfdW5pdHRlc3QucHk6CisgICAgICAgICh0ZXN0X3Bh dGhfdG9fdGVzdF9leHBlY3RhdGlvbnNfZmlsZSk6CisKIDIwMTItMDEtMDMgIFRvciBBcm5lIFZl c3Riw7ggIDx0b3IuYXJuZS52ZXN0Ym9Abm9raWEuY29tPgogCiAgICAgICAgIFtRdF0gSW50cm9k dWNlIHFtYWtlIHJlcGxhY2UgZnVuY3Rpb24gZm9yIHRoZSBhY3RpdmUgYnVpbGQgY29uZmlnCmRp ZmYgLS1naXQgYS9Ub29scy9TY3JpcHRzL3dlYmtpdHB5L2xheW91dF90ZXN0cy9wb3J0L3dlYmtp dC5weSBiL1Rvb2xzL1NjcmlwdHMvd2Via2l0cHkvbGF5b3V0X3Rlc3RzL3BvcnQvd2Via2l0LnB5 CmluZGV4IDE5ODVhNGJkZWE3NTdhMTAwMTNkMmIyMTBmZTM4Njg5NjZiY2RjYWQuLjYxMWVjMGYw MWJlYjUxNjk2MTU2ZDg1ZDU3ODFhMzZhMzIyM2E5YTcgMTAwNjQ0Ci0tLSBhL1Rvb2xzL1Njcmlw dHMvd2Via2l0cHkvbGF5b3V0X3Rlc3RzL3BvcnQvd2Via2l0LnB5CisrKyBiL1Rvb2xzL1Njcmlw dHMvd2Via2l0cHkvbGF5b3V0X3Rlc3RzL3BvcnQvd2Via2l0LnB5CkBAIC03OCw4ICs3OCw3IEBA IGNsYXNzIFdlYktpdFBvcnQoUG9ydCk6CiAKICAgICBkZWYgcGF0aF90b190ZXN0X2V4cGVjdGF0 aW9uc19maWxlKHNlbGYpOgogICAgICAgICAjIHRlc3RfZXhwZWN0YXRpb25zIGFyZSBhbHdheXMg aW4gbWFjLyBub3QgbWFjLWxlb3BhcmQvIGJ5IGNvbnZlbnRpb24sIGhlbmNlIHdlIHVzZSBwb3J0 X25hbWUgaW5zdGVhZCBvZiBuYW1lKCkuCi0gICAgICAgIGV4cGVjdGF0aW9uc19kaXJlY3Rvcnkg PSBzZWxmLl93azJfcG9ydF9uYW1lKCkgaWYgc2VsZi5nZXRfb3B0aW9uKCd3ZWJraXRfdGVzdF9y dW5uZXInKSBlbHNlIHNlbGYucG9ydF9uYW1lCi0gICAgICAgIHJldHVybiBzZWxmLl9maWxlc3lz dGVtLmpvaW4oc2VsZi5fd2Via2l0X2Jhc2VsaW5lX3BhdGgoZXhwZWN0YXRpb25zX2RpcmVjdG9y eSksICd0ZXN0X2V4cGVjdGF0aW9ucy50eHQnKQorICAgICAgICByZXR1cm4gc2VsZi5fZmlsZXN5 c3RlbS5qb2luKHNlbGYuX3dlYmtpdF9iYXNlbGluZV9wYXRoKHNlbGYucG9ydF9uYW1lKSwgJ3Rl c3RfZXhwZWN0YXRpb25zLnR4dCcpCiAKICAgICBkZWYgX3BvcnRfZmxhZ19mb3Jfc2NyaXB0cyhz ZWxmKToKICAgICAgICAgIyBUaGlzIGlzIG92ZXJycmlkZW4gYnkgcG9ydHMgd2hpY2ggbmVlZCBh IGZsYWcgcGFzc2VkIHRvIHNjcmlwdHMgdG8gZGlzdGluZ3Vpc2ggdGhlIHVzZSBvZiB0aGF0IHBv cnQuCmRpZmYgLS1naXQgYS9Ub29scy9TY3JpcHRzL3dlYmtpdHB5L2xheW91dF90ZXN0cy9wb3J0 L3dlYmtpdF91bml0dGVzdC5weSBiL1Rvb2xzL1NjcmlwdHMvd2Via2l0cHkvbGF5b3V0X3Rlc3Rz L3BvcnQvd2Via2l0X3VuaXR0ZXN0LnB5CmluZGV4IDM1MmVhNjcxMzMxN2M1YTVlMDgwNTg2NWZh MGM5OGJjNzEyNTllYzcuLjVjZTBkMjU3ZDUwNDA5Y2UxNmY2OTgxZTA1MTBkOGRhYTQ5ZmMwY2Eg MTAwNzU1Ci0tLSBhL1Rvb2xzL1NjcmlwdHMvd2Via2l0cHkvbGF5b3V0X3Rlc3RzL3BvcnQvd2Vi a2l0X3VuaXR0ZXN0LnB5CisrKyBiL1Rvb2xzL1NjcmlwdHMvd2Via2l0cHkvbGF5b3V0X3Rlc3Rz L3BvcnQvd2Via2l0X3VuaXR0ZXN0LnB5CkBAIC05Miw2ICs5MiwxMCBAQCBjbGFzcyBXZWJLaXRQ b3J0VGVzdChwb3J0X3Rlc3RjYXNlLlBvcnRUZXN0Q2FzZSk6CiAgICAgZGVmIHRlc3RfYmFzZWxp bmVfc2VhcmNoX3BhdGgoc2VsZik6CiAgICAgICAgIHBhc3MKIAorICAgIGRlZiB0ZXN0X3BhdGhf dG9fdGVzdF9leHBlY3RhdGlvbnNfZmlsZShzZWxmKToKKyAgICAgICAgcG9ydCA9IFRlc3RXZWJL aXRQb3J0KCkKKyAgICAgICAgc2VsZi5hc3NlcnRFcXVhbChwb3J0LnBhdGhfdG9fdGVzdF9leHBl Y3RhdGlvbnNfZmlsZSgpLCAnL21vY2stY2hlY2tvdXQvTGF5b3V0VGVzdHMvcGxhdGZvcm0vdGVz dHdlYmtpdHBvcnQvdGVzdF9leHBlY3RhdGlvbnMudHh0JykKKwogICAgIGRlZiB0ZXN0X3NraXBw ZWRfZGlyZWN0b3JpZXNfZm9yX3N5bWJvbHMoc2VsZik6CiAgICAgICAgICMgVGhpcyBmaXJzdCB0 ZXN0IGNvbmZpcm1zIHRoYXQgdGhlIGNvbW1vbmx5IGZvdW5kIHN5bWJvbHMgcmVzdWx0IGluIHRo ZSBleHBlY3RlZCBza2lwcGVkIGRpcmVjdG9yaWVzLgogICAgICAgICBzeW1ib2xzX3N0cmluZyA9 ICIgIi5qb2luKFsiR3JhcGhpY3NMYXllciIsICJXZWJDb3JlSGFzM0RSZW5kZXJpbmciLCAiaXNY SFRNTE1QRG9jdW1lbnQiLCAiZm9vU3ltYm9sIl0pCg== 121324 2012-01-05 13:40:58 -0800 2012-01-06 00:14:13 -0800 Patch bug-75417-20120105224056.patch text/plain 3124 ossy U3VidmVyc2lvbiBSZXZpc2lvbjogMTA0MjA2CmRpZmYgLS1naXQgYS9Ub29scy9DaGFuZ2VMb2cg Yi9Ub29scy9DaGFuZ2VMb2cKaW5kZXggZDQ0ZmFmZjUzM2JjNDYzZTNhOTQ4OTJjMTMwZjZhMjE2 ZmExNDBkZC4uZjA3ZDkzMWMyZWZhODJhMTM3OTgxZGZmY2ZhNTc0OTQwZDAxZjk5MCAxMDA2NDQK LS0tIGEvVG9vbHMvQ2hhbmdlTG9nCisrKyBiL1Rvb2xzL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE1 IEBACisyMDEyLTAxLTA1ICBDc2FiYSBPc3p0cm9nb27DoWMgIDxvc3N5QHdlYmtpdC5vcmc+CisK KyAgICAgICAgTlJXVCBzaG91bGQgdXNlIHRlc3RfZXhwZWN0YXRpb24udHh0IG9uIHdrMiBwbGF0 Zm9ybXMKKyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTc1 NDE3CisKKyAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISkuCisKKyAgICAgICAgKiBT Y3JpcHRzL3dlYmtpdHB5L2xheW91dF90ZXN0cy9wb3J0L3dlYmtpdC5weToKKyAgICAgICAgKHBh dGhfdG9fdGVzdF9leHBlY3RhdGlvbnNfZmlsZSk6CisgICAgICAgICogU2NyaXB0cy93ZWJraXRw eS9sYXlvdXRfdGVzdHMvcG9ydC93ZWJraXRfdW5pdHRlc3QucHk6CisgICAgICAgICh0ZXN0X3Bh dGhfdG9fdGVzdF9leHBlY3RhdGlvbnNfZmlsZSk6CisKIDIwMTItMDEtMDUgIEpvY2hlbiBFaXNp bmdlciAgPGpvY2hlbkBjaHJvbWl1bS5vcmc+CiAKICAgICAgICAgUmVwbGFjZSB3ZWJraXRweS5j b21tb24uc3lzdGVtLmZpbGVzeXN0ZW0uZmlsZV9wYXRoX2FzX3VybCB3aXRoIHdlYmtpdHB5LmNv bW1vbi5zeXN0ZW0ucGF0aC5hYnNwYXRoX3RvX3VyaQpkaWZmIC0tZ2l0IGEvVG9vbHMvU2NyaXB0 cy93ZWJraXRweS9sYXlvdXRfdGVzdHMvcG9ydC93ZWJraXQucHkgYi9Ub29scy9TY3JpcHRzL3dl YmtpdHB5L2xheW91dF90ZXN0cy9wb3J0L3dlYmtpdC5weQppbmRleCAxOTg1YTRiZGVhNzU3YTEw MDEzZDJiMjEwZmUzODY4OTY2YmNkY2FkLi42MTFlYzBmMDFiZWI1MTY5NjE1NmQ4NWQ1NzgxYTM2 YTMyMjNhOWE3IDEwMDY0NAotLS0gYS9Ub29scy9TY3JpcHRzL3dlYmtpdHB5L2xheW91dF90ZXN0 cy9wb3J0L3dlYmtpdC5weQorKysgYi9Ub29scy9TY3JpcHRzL3dlYmtpdHB5L2xheW91dF90ZXN0 cy9wb3J0L3dlYmtpdC5weQpAQCAtNzgsOCArNzgsNyBAQCBjbGFzcyBXZWJLaXRQb3J0KFBvcnQp OgogCiAgICAgZGVmIHBhdGhfdG9fdGVzdF9leHBlY3RhdGlvbnNfZmlsZShzZWxmKToKICAgICAg ICAgIyB0ZXN0X2V4cGVjdGF0aW9ucyBhcmUgYWx3YXlzIGluIG1hYy8gbm90IG1hYy1sZW9wYXJk LyBieSBjb252ZW50aW9uLCBoZW5jZSB3ZSB1c2UgcG9ydF9uYW1lIGluc3RlYWQgb2YgbmFtZSgp LgotICAgICAgICBleHBlY3RhdGlvbnNfZGlyZWN0b3J5ID0gc2VsZi5fd2syX3BvcnRfbmFtZSgp IGlmIHNlbGYuZ2V0X29wdGlvbignd2Via2l0X3Rlc3RfcnVubmVyJykgZWxzZSBzZWxmLnBvcnRf bmFtZQotICAgICAgICByZXR1cm4gc2VsZi5fZmlsZXN5c3RlbS5qb2luKHNlbGYuX3dlYmtpdF9i YXNlbGluZV9wYXRoKGV4cGVjdGF0aW9uc19kaXJlY3RvcnkpLCAndGVzdF9leHBlY3RhdGlvbnMu dHh0JykKKyAgICAgICAgcmV0dXJuIHNlbGYuX2ZpbGVzeXN0ZW0uam9pbihzZWxmLl93ZWJraXRf YmFzZWxpbmVfcGF0aChzZWxmLnBvcnRfbmFtZSksICd0ZXN0X2V4cGVjdGF0aW9ucy50eHQnKQog CiAgICAgZGVmIF9wb3J0X2ZsYWdfZm9yX3NjcmlwdHMoc2VsZik6CiAgICAgICAgICMgVGhpcyBp cyBvdmVycnJpZGVuIGJ5IHBvcnRzIHdoaWNoIG5lZWQgYSBmbGFnIHBhc3NlZCB0byBzY3JpcHRz IHRvIGRpc3Rpbmd1aXNoIHRoZSB1c2Ugb2YgdGhhdCBwb3J0LgpkaWZmIC0tZ2l0IGEvVG9vbHMv U2NyaXB0cy93ZWJraXRweS9sYXlvdXRfdGVzdHMvcG9ydC93ZWJraXRfdW5pdHRlc3QucHkgYi9U b29scy9TY3JpcHRzL3dlYmtpdHB5L2xheW91dF90ZXN0cy9wb3J0L3dlYmtpdF91bml0dGVzdC5w eQppbmRleCAzNTJlYTY3MTMzMTdjNWE1ZTA4MDU4NjVmYTBjOThiYzcxMjU5ZWM3Li43OGI4ZTc1 MWU4ODg2YTQ4MjUwN2YxNThlYWJiYjQ3MmU0ZjVkMmU5IDEwMDc1NQotLS0gYS9Ub29scy9TY3Jp cHRzL3dlYmtpdHB5L2xheW91dF90ZXN0cy9wb3J0L3dlYmtpdF91bml0dGVzdC5weQorKysgYi9U b29scy9TY3JpcHRzL3dlYmtpdHB5L2xheW91dF90ZXN0cy9wb3J0L3dlYmtpdF91bml0dGVzdC5w eQpAQCAtOTIsNiArOTIsMTMgQEAgY2xhc3MgV2ViS2l0UG9ydFRlc3QocG9ydF90ZXN0Y2FzZS5Q b3J0VGVzdENhc2UpOgogICAgIGRlZiB0ZXN0X2Jhc2VsaW5lX3NlYXJjaF9wYXRoKHNlbGYpOgog ICAgICAgICBwYXNzCiAKKyAgICBkZWYgdGVzdF9wYXRoX3RvX3Rlc3RfZXhwZWN0YXRpb25zX2Zp bGUoc2VsZik6CisgICAgICAgIHBvcnQgPSBUZXN0V2ViS2l0UG9ydCgpCisgICAgICAgIHBvcnQu X29wdGlvbnMgPSBNb2NrT3B0aW9ucyh3ZWJraXRfdGVzdF9ydW5uZXI9RmFsc2UpCisgICAgICAg IHNlbGYuYXNzZXJ0RXF1YWwocG9ydC5wYXRoX3RvX3Rlc3RfZXhwZWN0YXRpb25zX2ZpbGUoKSwg Jy9tb2NrLWNoZWNrb3V0L0xheW91dFRlc3RzL3BsYXRmb3JtL3Rlc3R3ZWJraXRwb3J0L3Rlc3Rf ZXhwZWN0YXRpb25zLnR4dCcpCisgICAgICAgIHBvcnQuX29wdGlvbnMgPSBNb2NrT3B0aW9ucyh3 ZWJraXRfdGVzdF9ydW5uZXI9VHJ1ZSkKKyAgICAgICAgc2VsZi5hc3NlcnRFcXVhbChwb3J0LnBh dGhfdG9fdGVzdF9leHBlY3RhdGlvbnNfZmlsZSgpLCAnL21vY2stY2hlY2tvdXQvTGF5b3V0VGVz dHMvcGxhdGZvcm0vdGVzdHdlYmtpdHBvcnQvdGVzdF9leHBlY3RhdGlvbnMudHh0JykKKwogICAg IGRlZiB0ZXN0X3NraXBwZWRfZGlyZWN0b3JpZXNfZm9yX3N5bWJvbHMoc2VsZik6CiAgICAgICAg ICMgVGhpcyBmaXJzdCB0ZXN0IGNvbmZpcm1zIHRoYXQgdGhlIGNvbW1vbmx5IGZvdW5kIHN5bWJv bHMgcmVzdWx0IGluIHRoZSBleHBlY3RlZCBza2lwcGVkIGRpcmVjdG9yaWVzLgogICAgICAgICBz eW1ib2xzX3N0cmluZyA9ICIgIi5qb2luKFsiR3JhcGhpY3NMYXllciIsICJXZWJDb3JlSGFzM0RS ZW5kZXJpbmciLCAiaXNYSFRNTE1QRG9jdW1lbnQiLCAiZm9vU3ltYm9sIl0pCg==