14941 2007-08-11 05:03:32 -0700 REGRESSION: textarea value from JavaScript includes extra newline 2008-03-16 13:18:11 -0700 1 1 1 Unclassified WebKit Forms 523.x (Safari 3) All All RESOLVED FIXED http://shadow2531.com/js/jsuri.html HasReduction, InRadar, Regression P1 Normal --- 1 shadow2531 darin ap ddkilzer justin.garcia mitz zwarich oldest_to_newest 2541 0 shadow2531 2007-08-11 05:03:32 -0700 In WebKit r24872, pressing enter on a non-empty line in a textarea adds two newlines to the DOM value (not visually) instead of one. No other browser does this. This causes double %0D%0A in the encoded value after pressing enter. Also, pressing the insert key in a textarea inserts null bytes into the DOM value. 2542 1 15929 shadow2531 2007-08-11 05:04:08 -0700 Created attachment 15929 Demo 2543 2 shadow2531 2007-08-11 05:09:48 -0700 (In reply to comment #0) >This causes double %0D%0A in the encoded value after pressing enter. At the URI listed above I mean. 2519 3 ddkilzer 2007-08-11 08:53:31 -0700 Confirmed with a local debug build of WebKit r25008 with Safari 3 Public Beta v. 3.0.3 (522.12.1) on Mac OS X 10.4.10 (8R218). 2518 4 ddkilzer 2007-08-11 08:53:49 -0700 <rdar://problem/5404093> 2516 5 ddkilzer 2007-08-11 08:54:38 -0700 BTW, excellent test case, Michael! 2414 6 zwarich 2007-08-13 00:49:35 -0700 I did some hunting with nightlies. This bug wasn't introduced all at once. Here is an exciting timeline of these problems. I can't test the insert bug, so it is excluded. Before the r14941 nightly (the last nightly prior to that is r14926), all is well. With the r14941 nightly, the following problems are introduced: 1) The phantom newline after deleting text that contains a newline. 2) Typing "line1" and then enter does not actually display a newline, it only adds an invisible one to the value of the textarea. When you hit enter again, there are two newlines. There are two revisions between these nightlies that affect the textarea element, r14931 <http://trac.webkit.org/projects/webkit/changeset/14931> and r14935 <http://trac.webkit.org/projects/webkit/changeset/14935>. My money is on r14931, but I don't have the bandwidth right now to download a tree to test. I noticed that at some point in time, both of these bugs were fixed. I tracked it down to the period between the nightlies r15614 and r15618, surely by r15617 <http://trac.webkit.org/projects/webkit/changeset/15617>. The phantom newline after delete was reintroduced between the nightlies r17205 and r17233. The most likely candidate for introduction is r17223 <http://trac.webkit.org/projects/webkit/changeset/17223>. Finally, the phantom newline after hitting enter was introduced by r17386 <http://trac.webkit.org/projects/webkit/changeset/17386> (I had a tree from that day sitting around from another bug fix). I am not sure I will have the time to put together a patch to fix all of these problems in the next week, because I'm at a conference, but hopefully this will help anyone who is trying to fix it. 2382 7 sullivan 2007-08-13 13:53:37 -0700 What user symptoms are caused by these problems? 2377 8 shadow2531 2007-08-13 23:07:09 -0700 With textarea counter scripts that limit the length of input, the stray newline and extra newlines decrease the limit of real text and lines you can input. If you press insert and it enters null bytes, those are counted by textarea counters also. In fact, if you hold down insert, you can exhaust your limit even though the textarea still looks blank. A user could unknowingly decrease their limit. With the stray newline (once triggered), it makes it difficult (if you don't know about the bug) to submit a textarea with an empty value to the server. The extra newlines are also submitted to the server causing an incorrect value to be submitted. If the server script takes each newline pair and converts it to <p></p> for example, you'll get an extra <p></p> with this problem, which would produce the wrong output in whatever was using the them. With my jsuri composer, it causes the js uri to have 6 extra bytes for every stray newlinee and 3 for every null byte. This causes the js URI to be longer than other browsers and when the textarea produces null bytes and causes %00 to be in the js uri, the js uri will be broken. (I can filter out the null bytes, but the extra newlines produced I cannot because I can't differentiate them from non-extra ones.) In SquirrelMail, if you hit insert, this can cause the outgoing body to get cut off at the null byte. This breaks Gmail in the same way. It probably breaks other webmails in the same way. (In reply to comment #7) > What user symptoms are caused by these problems? > In short, they break form submission. 2047 9 mitz 2007-08-18 02:24:15 -0700 See also bug 14562. 2013 10 ddkilzer 2007-08-18 18:42:01 -0700 Using autospade (Bug 15002): Works: r17205 Fails: r17233 Note that the behavior in r17233 is different from ToT when using the test case (Attachment #15929). Hitting Enter adds a %0A character ("line1%0A"), but then typing "a" puts it BEFORE the return character ("line1a%0A") instead of after it! In later revisions, you end up with "line1%0Aa". In both cases, the "%0A" character is undeletable. 2012 11 ddkilzer 2007-08-18 19:03:07 -0700 Behavior changed again between r17354 ("line1a%0A") and r17388 ("line1%0Aa"). 2010 12 ddkilzer 2007-08-18 19:10:07 -0700 (In reply to comment #11) > Behavior changed again between r17354 ("line1a%0A") and r17388 ("line1%0Aa"). Likely r17386: http://trac.webkit.org/projects/webkit/changeset/17386 (In reply to comment #10) > Works: r17205 Fails: r17233 Likely r17223: http://trac.webkit.org/projects/webkit/changeset/17223 73851 13 darin 2008-03-15 18:59:08 -0700 It's a mistake to have two bugs both reported in the same bug report. I have a fix for the multiple newline one, but since my keyboard doesn't even have an insert key in it I'm not going to be able to fix the other. 73852 14 mitz 2008-03-15 19:38:44 -0700 (In reply to comment #13) > Since my keyboard doesn't even have an > insert key in it I'm not going to be able to fix the other. I am pretty sure the insert key bug is bug 14562 has been fixed long ago in <http://trac.webkit.org/projects/webkit/changeset/25251>. 73853 15 shadow2531 2008-03-15 20:12:02 -0700 (In reply to comment #14) > (In reply to comment #13) > > Since my keyboard doesn't even have an > > insert key in it I'm not going to be able to fix the other. > > I am pretty sure the insert key bug is bug 14562 has been fixed long ago in > <http://trac.webkit.org/projects/webkit/changeset/25251>. > Yep, the insert key no longer inserts null bytes. Just need the newline fix now. 73950 16 19805 darin 2008-03-16 12:58:19 -0700 Created attachment 19805 patch 73956 17 19805 mitz 2008-03-16 13:13:53 -0700 Comment on attachment 19805 patch r=me 73959 18 darin 2008-03-16 13:18:11 -0700 Committed revision 31081. 15929 2007-08-11 05:04:08 -0700 2007-08-11 05:04:08 -0700 Demo zinput.html text/html 1587 shadow2531 PCFET0NUWVBFIGh0bWw+CjxodG1sPgogICAgPGhlYWQ+CiAgICAgICAgPHRpdGxlPlNhZmFyaSBp bnZlbnRzIGV4dHJhIG5ld2xpbmUgaW4gdGV4dGFyZWEgdmFsdWU8L3RpdGxlPgogICAgICAgIDxz Y3JpcHQ+CiAgICAgICAgICAgIHdpbmRvdy5vbmxvYWQgPSBmdW5jdGlvbigpIHsKICAgICAgICAg ICAgICAgIHZhciBwcmUgPSBkb2N1bWVudC5nZXRFbGVtZW50c0J5VGFnTmFtZSgicHJlIilbMF07 CiAgICAgICAgICAgICAgICBwcmUuaW5uZXJIVE1MID0gIiI7CiAgICAgICAgICAgICAgICB2YXIg dGEgPSBkb2N1bWVudC5nZXRFbGVtZW50c0J5VGFnTmFtZSgidGV4dGFyZWEiKVswXQogICAgICAg ICAgICAgICAgdGEub25rZXl1cCA9IGZ1bmN0aW9uKCkgewogICAgICAgICAgICAgICAgICAgIHBy ZS5pbm5lckhUTUwgPSBlbmNvZGVVUklDb21wb25lbnQodGEudmFsdWUpOwogICAgICAgICAgICAg ICAgfTsKICAgICAgICAgICAgfTsKICAgICAgICA8L3NjcmlwdD4KICAgICAgICA8c3R5bGU+CiAg ICAgICAgICAgIHRleHRhcmVhLCBwcmUgewogICAgICAgICAgICAgICAgd2lkdGg6IDUwMHB4Owog ICAgICAgICAgICAgICAgaGVpZ2h0OiAyMDBweDsKICAgICAgICAgICAgfQogICAgICAgICAgICBw cmUgewogICAgICAgICAgICAgICAgYm9yZGVyOiAxcHggc29saWQgYmxhY2s7CiAgICAgICAgICAg IH0KICAgICAgICA8L3N0eWxlPgogICAgPC9oZWFkPgogICAgPGJvZHk+CiAgICAgICAgPHVsPgog ICAgICAgICAgICA8bGk+SW4gdGhlIHRleHRhcmVhIGJlbG93LCB0eXBlICJsaW5lMSIgYW5kIHBy ZXNzIGVudGVyLiBZb3Ugd2lsbCBzZWUgdGhhdCB0aGVyZSBpcyBhbiBleHRyYSBuZXdsaW5lIChu ZXdsaW5lcyBhcmUgZW5jb2RlZCBzbyB5b3UgY2FuIHNlZSB0aGVtKSB0aGF0IHNob3VsZG4ndCBi ZSB0aGVyZS48L2xpPgogICAgICAgICAgICA8bGk+Tm93LCB3aXRob3V0IGNoYW5naW5nIGFueXRo aW5nIGVsc2UsIHR5cGUgImEiIGZvciBleGFtcGxlLiBZb3Ugd2lsbCBzZWUgdGhhdCB0aGUgZXh0 cmEgbmV3bGluZSBpcyBub3cgZ29uZS48L2xpPgogICAgICAgICAgICA8bGk+Tm93LCBjbGVhciB0 aGUgdGV4dGFyZWEgYW5kIHR5cGUgImxpbmUxIiBhbmQgcHJlc3MgZW50ZXIgYWdhaW4uIFRoaXMg dGltZSwgaG9sZCBkb3duIGJhY2tzcGFjZSB0aWxsIHRoZSB0ZXh0YXJlYSBpcyBjbGVhci4gWW91 IHdpbGwgc2VlIHRoYXQgdGhlcmUncyBhIG5ld2xpbmUgdGhhdCB5b3UgY2FuJ3QgZ2V0IHJpZCBv ZiAoZXZlbiB3aXRoIGRlbGV0ZSkuIFRvIGdldCByaWQgb2YgdGhlIG5ld2xpbmUgeW91IGhhdmUg dG8gdHlwZSBzb21ldGhpbmcgZWxzZS48L2xpPgogICAgICAgICAgICA8bGk+QWxzbywgcHJlc3Mg dGhlIGluc2VydCBrZXkgd2hpbGUgdGhlIHRleHRhcmVhIGlzIGZvY3VzZWQuIFNhZmFyaSB3aWxs IGluc2VydCBudWxsIGJ5dGVzIGludG8gdGhlIHRleHRhcmVhIHZhbHVlLjwvbGk+CiAgICAgICAg PC91bD4KICAgICAgICA8cD48dGV4dGFyZWE+PC90ZXh0YXJlYT48L3A+CiAgICAgICAgPHByZT5U aGlzIHRlc3QgcmVxdWlyZXMgSlM8L3ByZT4KICAgIDwvYm9keT4KPC9odG1sPgoK 19805 2008-03-16 12:58:19 -0700 2008-03-16 13:13:53 -0700 patch TextAreaNewlinePatch.txt text/plain 5329 darin SW5kZXg6IFdlYkNvcmUvQ2hhbmdlTG9nCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0tIFdlYkNvcmUvQ2hhbmdlTG9n CShyZXZpc2lvbiAzMTA3OSkKKysrIFdlYkNvcmUvQ2hhbmdlTG9nCSh3b3JraW5nIGNvcHkpCkBA IC0xLDMgKzEsMTggQEAKKzIwMDgtMDMtMTYgIERhcmluIEFkbGVyICA8ZGFyaW5AYXBwbGUuY29t PgorCisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgorCisgICAgICAgIC0gZml4 IGh0dHA6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTE0OTQxCisgICAgICAgICAg dGV4dGFyZWEgdmFsdWUgZnJvbSBKYXZhU2NyaXB0IGluY2x1ZGVzIGV4dHJhIG5ld2xpbmUKKwor ICAgICAgICBUZXN0OiBmYXN0L2Zvcm1zL3RleHRhcmVhLXRyYWlsaW5nLW5ld2xpbmUuaHRtbAor CisgICAgICAgICogcmVuZGVyaW5nL1JlbmRlclRleHRDb250cm9sLmNwcDoKKyAgICAgICAgKFdl YkNvcmU6OlJlbmRlclRleHRDb250cm9sOjpmaW5pc2hUZXh0KTogQWRkZWQgY29kZSB0byBzdHJp cCB0aGUgdHJhaWxpbmcKKyAgICAgICAgbmV3bGluZS4gSXQncyBwb3NzaWJsZSB0aGVyZSBhcmUg c29tZSBvYnNjdXJlIGNhc2VzIHdoZXJlIHRoaXMgaXMgbm90IHdhbnRlZCwKKyAgICAgICAgYnV0 IEkgY291bGRuJ3QgZmluZCBhbnkuIElmIHNvbWVvbmUgZmluZHMgYSBjYXNlIHdoZXJlIHRoaXMg aXMgYmFkLCB3ZSBjYW4KKyAgICAgICAgbWFrZSB0aGUgY29kZSBjb25kaXRpb25hbC4KKwogMjAw OC0wMy0xNSAgVGltb3RoeSBIYXRjaGVyICA8dGltb3RoeUBhcHBsZS5jb20+CiAKICAgICAgICAg UmV2aWV3ZWQgYnkgQWRhbSBSb2Jlbi4KSW5kZXg6IFdlYkNvcmUvcmVuZGVyaW5nL1JlbmRlclRl eHRDb250cm9sLmNwcAo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBXZWJDb3JlL3JlbmRlcmluZy9SZW5kZXJUZXh0 Q29udHJvbC5jcHAJKHJldmlzaW9uIDMxMDc4KQorKysgV2ViQ29yZS9yZW5kZXJpbmcvUmVuZGVy VGV4dENvbnRyb2wuY3BwCSh3b3JraW5nIGNvcHkpCkBAIC01NTUsMTIgKzU1NSwxOCBAQCB2b2lk IFJlbmRlclRleHRDb250cm9sOjpzdWJ0cmVlSGFzQ2hhbmdlCiAKIFN0cmluZyBSZW5kZXJUZXh0 Q29udHJvbDo6ZmluaXNoVGV4dChWZWN0b3I8VUNoYXI+JiByZXN1bHQpIGNvbnN0CiB7CisgICAg Ly8gUmVtb3ZlIG9uZSB0cmFpbGluZyBuZXdsaW5lOyB0aGVyZSdzIGFsd2F5cyBvbmUgdGhhdCdz IGNvbGxhcHNlZCBvdXQgYnkgcmVuZGVyaW5nLgorICAgIHNpemVfdCBzaXplID0gcmVzdWx0LnNp emUoKTsKKyAgICBpZiAoc2l6ZSAmJiByZXN1bHRbc2l6ZSAtIDFdID09ICdcbicpCisgICAgICAg IHJlc3VsdC5zaHJpbmsoLS1zaXplKTsKKworICAgIC8vIENvbnZlcnQgYmFja3NsYXNoIHRvIGN1 cnJlbmN5IHN5bWJvbC4KICAgICBVQ2hhciBzeW1ib2wgPSBiYWNrc2xhc2hBc0N1cnJlbmN5U3lt Ym9sKCk7CiAgICAgaWYgKHN5bWJvbCAhPSAnXFwnKSB7Ci0gICAgICAgIHNpemVfdCBzaXplID0g cmVzdWx0LnNpemUoKTsKLSAgICAgICAgZm9yIChzaXplX3QgaSA9IDA7IGkgPCBzaXplOyArK2kp CisgICAgICAgIGZvciAoc2l6ZV90IGkgPSAwOyBpIDwgc2l6ZTsgKytpKSB7CiAgICAgICAgICAg ICBpZiAocmVzdWx0W2ldID09ICdcXCcpCiAgICAgICAgICAgICAgICAgcmVzdWx0W2ldID0gc3lt Ym9sOworICAgICAgICB9CiAgICAgfQogCiAgICAgcmV0dXJuIFN0cmluZzo6YWRvcHQocmVzdWx0 KTsKSW5kZXg6IExheW91dFRlc3RzL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBMYXlvdXRUZXN0 cy9DaGFuZ2VMb2cJKHJldmlzaW9uIDMxMDc5KQorKysgTGF5b3V0VGVzdHMvQ2hhbmdlTG9nCSh3 b3JraW5nIGNvcHkpCkBAIC0xLDMgKzEsMTYgQEAKKzIwMDgtMDMtMTYgIERhcmluIEFkbGVyICA8 ZGFyaW5AYXBwbGUuY29tPgorCisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgor CisgICAgICAgIC0gdGVzdCBmb3IgaHR0cDovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/ aWQ9MTQ5NDEKKyAgICAgICAgICB0ZXh0YXJlYSB2YWx1ZSBmcm9tIEphdmFTY3JpcHQgaW5jbHVk ZXMgZXh0cmEgbmV3bGluZQorCisgICAgICAgICogZmFzdC9mb3Jtcy90ZXh0YXJlYS1wYXN0ZS1u ZXdsaW5lLmh0bWw6IFVwZGF0ZWQgdGVzdCB0byBleHBlY3QgY29ycmVjdCBiZWhhdmlvciBpbnN0 ZWFkCisgICAgICAgIG9mIGV4cGVjdGluZyB0aGUgYnVnLgorCisgICAgICAgICogZmFzdC9mb3Jt cy90ZXh0YXJlYS10cmFpbGluZy1uZXdsaW5lLWV4cGVjdGVkLnR4dDogQWRkZWQuCisgICAgICAg ICogZmFzdC9mb3Jtcy90ZXh0YXJlYS10cmFpbGluZy1uZXdsaW5lLmh0bWw6IEFkZGVkLgorCiAy MDA4LTAzLTE1ICBEYXJpbiBBZGxlciAgPGRhcmluQGFwcGxlLmNvbT4KIAogICAgICAgICBSZXZp ZXdlZCBieSBTYW0uCkluZGV4OiBMYXlvdXRUZXN0cy9mYXN0L2Zvcm1zL3RleHRhcmVhLXBhc3Rl LW5ld2xpbmUuaHRtbAo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBMYXlvdXRUZXN0cy9mYXN0L2Zvcm1zL3RleHRh cmVhLXBhc3RlLW5ld2xpbmUuaHRtbAkocmV2aXNpb24gMzEwNzgpCisrKyBMYXlvdXRUZXN0cy9m YXN0L2Zvcm1zL3RleHRhcmVhLXBhc3RlLW5ld2xpbmUuaHRtbAkod29ya2luZyBjb3B5KQpAQCAt MTQsMTIgKzE0LDEyIEBAIGZ1bmN0aW9uIHRlc3QoKQogICAgIGRvY3VtZW50LmV4ZWNDb21tYW5k KCJwYXN0ZSIpOwogICAgIHZhciByZXN1bHQyID0gdGEudmFsdWU7CiAKLSAgICBpZiAocmVzdWx0 MSA9PSAiYWJjXG5cbiIgJiYgcmVzdWx0MiA9PSAiYWJjXG5hYmNcblxuIikKKyAgICBpZiAocmVz dWx0MSA9PSAiYWJjXG4iICYmIHJlc3VsdDIgPT0gImFiY1xuYWJjXG4iKQogICAgICAgICBkb2N1 bWVudC53cml0ZSgiPHA+SG9vcmF5LCB0aGUgdGVzdCB3YXMgc3VjY2Vzc2Z1bCE8L3A+Iik7CiAg ICAgZWxzZSBpZiAocmVzdWx0MSA9PSAiIikKICAgICAgICAgZG9jdW1lbnQud3JpdGUoIjxwPlRo ZSB0ZXN0IGZhaWxlZDsgZG9lc24ndCB3b3JrIGluIHJlbGVhc2UgYnVpbGRzIG9mIFNhZmFyaSBi ZWNhdXNlIHBhc3RlIGlzIG5vdCBhbGxvd2VkLjwvcD4iKTsKICAgICBlbHNlCi0gICAgICAgIGRv Y3VtZW50LndyaXRlKCI8cD5UaGUgdGVzdCBmYWlsZWQsIHJlc3VsdCAxIHdhcyAnIiArIHJlc3Vs dDEucmVwbGFjZSgiXG4iLCAiXFxuIikgKyAiJyBhbmQgcmVzdWx0IDIgd2FzICciICsgcmVzdWx0 Mi5yZXBsYWNlKCJcbiIsICJcXG4iKSArICInLiBUaGlzIHRlc3QgZGVtb25zdHJhdGVzIGEgYnVn ICg1Mjk5NDI1KTogdGhlIGZpbmFsICdcXG4nIGluIGEgdGV4dGFyZWEgaXMgcmVwb3J0ZWQgYnkg LnZhbHVlIGV2ZW4gaWYgaXQgaXMgY29sbGFwc2VkLjwvcD4iKTsKKyAgICAgICAgZG9jdW1lbnQu d3JpdGUoIjxwPlRoZSB0ZXN0IGZhaWxlZCwgcmVzdWx0IDEgd2FzICciICsgcmVzdWx0MS5yZXBs YWNlKCJcbiIsICJcXG4iKSArICInIGFuZCByZXN1bHQgMiB3YXMgJyIgKyByZXN1bHQyLnJlcGxh Y2UoIlxuIiwgIlxcbiIpICsgIicuPC9wPiIpOwogfQogPC9zY3JpcHQ+CiA8Ym9keSBvbmxvYWQ9 InRlc3QoKSI+CkluZGV4OiBMYXlvdXRUZXN0cy9mYXN0L2Zvcm1zL3RleHRhcmVhLXRyYWlsaW5n LW5ld2xpbmUtZXhwZWN0ZWQudHh0Cj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0tIExheW91dFRlc3RzL2Zhc3QvZm9y bXMvdGV4dGFyZWEtdHJhaWxpbmctbmV3bGluZS1leHBlY3RlZC50eHQJKHJldmlzaW9uIDApCisr KyBMYXlvdXRUZXN0cy9mYXN0L2Zvcm1zL3RleHRhcmVhLXRyYWlsaW5nLW5ld2xpbmUtZXhwZWN0 ZWQudHh0CShyZXZpc2lvbiAwKQpAQCAtMCwwICsxIEBACitIb29yYXksIHRoZSB0ZXN0IHdhcyBz dWNjZXNzZnVsIQoKUHJvcGVydHkgY2hhbmdlcyBvbjogTGF5b3V0VGVzdHMvZmFzdC9mb3Jtcy90 ZXh0YXJlYS10cmFpbGluZy1uZXdsaW5lLWV4cGVjdGVkLnR4dApfX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fCk5hbWU6IHN2 bjplb2wtc3R5bGUKICAgKyBuYXRpdmUKCkluZGV4OiBMYXlvdXRUZXN0cy9mYXN0L2Zvcm1zL3Rl eHRhcmVhLXRyYWlsaW5nLW5ld2xpbmUuaHRtbAo9PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBMYXlvdXRUZXN0cy9m YXN0L2Zvcm1zL3RleHRhcmVhLXRyYWlsaW5nLW5ld2xpbmUuaHRtbAkocmV2aXNpb24gMCkKKysr IExheW91dFRlc3RzL2Zhc3QvZm9ybXMvdGV4dGFyZWEtdHJhaWxpbmctbmV3bGluZS5odG1sCShy ZXZpc2lvbiAwKQpAQCAtMCwwICsxLDE3IEBACis8c2NyaXB0PgorZnVuY3Rpb24gdGVzdCgpCit7 CisgICAgaWYgKHdpbmRvdy5sYXlvdXRUZXN0Q29udHJvbGxlcikKKyAgICAgICAgbGF5b3V0VGVz dENvbnRyb2xsZXIuZHVtcEFzVGV4dCgpOworICAgIHZhciB0YSA9IGRvY3VtZW50LmdldEVsZW1l bnRCeUlkKCJ0YSIpOworICAgIHRhLmZvY3VzKCk7CisgICAgZG9jdW1lbnQuZXhlY0NvbW1hbmQo Ikluc2VydExpbmVCcmVhayIpOworICAgIHZhciByZXN1bHQgPSB0YS52YWx1ZTsKKyAgICBpZiAo cmVzdWx0ID09ICJcbiIpCisgICAgICAgIGRvY3VtZW50LndyaXRlKCI8cD5Ib29yYXksIHRoZSB0 ZXN0IHdhcyBzdWNjZXNzZnVsITwvcD4iKTsKKyAgICBlbHNlCisgICAgICAgIGRvY3VtZW50Lndy aXRlKCI8cD5UaGUgdGVzdCBmYWlsZWQsIHJlc3VsdCB3YXMgJyIgKyByZXN1bHQucmVwbGFjZSgi XG4iLCAiXFxuIikgKyAiJy48L3A+Iik7Cit9Cis8L3NjcmlwdD4KKzxib2R5IG9ubG9hZD0idGVz dCgpIj4KKzxwPjx0ZXh0YXJlYSBpZD0idGEiPjwvdGV4dGFyZWE+PC9wPgoKUHJvcGVydHkgY2hh bmdlcyBvbjogTGF5b3V0VGVzdHMvZmFzdC9mb3Jtcy90ZXh0YXJlYS10cmFpbGluZy1uZXdsaW5l Lmh0bWwKX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fXwpOYW1lOiBzdm46bWltZS10eXBlCiAgICsgdGV4dC9odG1sCk5hbWU6 IHN2bjplb2wtc3R5bGUKICAgKyBuYXRpdmUKCg==