15266 2007-09-23 15:26:18 -0700 Canvas shadows on gradients and patterns don't work sensibly 2011-06-02 20:55:11 -0700 1 1 1 Unclassified WebKit Layout and Rendering 523.x (Safari 3) PC Windows Server 2003 RESOLVED FIXED P2 Normal --- 1 excors webkit-unassigned andreip helder john.engelhart john.krueger mihaip mlamb1976 senorblanco oldest_to_newest 80 0 excors 2007-09-23 15:26:18 -0700 Shapes with gradients and patterns should get shadowed like any other shape or image. The bottom three examples of http://philip.html5.org/demos/canvas/shadows/various.html demonstrates the cases, and http://philip.html5.org/demos/canvas/shadows/expected.png is how I think it should be rendered. WebKit appears to treat patterns as if they were totally solid, and does something weirder with gradients. (There's no specification of how shadows should behave, but http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-June/011799.html is my suggestion.) 78 1 16364 excors 2007-09-23 15:27:41 -0700 Created attachment 16364 test case This does some minimal testing of the behaviour. 79 2 aroben 2007-09-23 15:32:05 -0700 The pattern part of your testcase and <http://philip.html5.org/demos/canvas/shadows/various.html> both work fine for me, though I do see problems with the gradients. 77 3 aroben 2007-09-23 15:33:10 -0700 (In reply to comment #2) > The pattern part of your testcase and > <http://philip.html5.org/demos/canvas/shadows/various.html> both work fine for > me, though I do see problems with the gradients. Sorry, that was a bit unclear. I don't see any problems with patterns+shadows in either your testcase or the page at philip.html5.org. I do see problems with gradients+shadows in both of those. I'm testing using ToT WebKit on Windows XP. 75 4 16366 excors 2007-09-23 16:51:19 -0700 Created attachment 16366 output from http://philip.html5.org/demos/canvas/shadows/various.html From the bottom of various.html, I see this with r25699 (from nightly.webkit.org) on Win2003, and also with r25567 on Win2000, both in Safari 3.0.3. 74 5 16367 excors 2007-09-23 16:54:32 -0700 Created attachment 16367 output from earlier test case I see this, with the same WebKit/Safari versions. 266900 6 64888 mlamb1976 2010-08-19 12:44:12 -0700 Created attachment 64888 Test repro case HTML page showing shadow rendering bug. Cases Rendered: 1. Rectangle with gradient fill and shadow 2. Stroked circle with gradient fill and shadow 3. Rectagle with solid fill and shadow 4. Circle with solid fill and shadow 5. Stroked rectagle with shadow 6. Stroked circle with shadow 7. Text with gradient fill, and shadow. 8. Text with gradient stroke, and shadow 9. Text with solid fill and shadow 10. Text with solid stroke, and shadow. 266901 7 64889 mlamb1976 2010-08-19 12:45:10 -0700 Created attachment 64889 PNG of expected rendering of Test Repro Case This is a PNG of the expected rendering of the Test Repro Case 266905 8 mlamb1976 2010-08-19 12:47:13 -0700 I am able to reproduce this with: Mac OS X Version 10.6.4 Safari Version 5.0.1 (6533.17.8) Hardward details: Model Name: MacBook Pro Model Identifier: MacBookPro4,1 Processor Name: Intel Core 2 Duo Processor Speed: 2.6 GHz Number Of Processors: 1 Total Number Of Cores: 2 L2 Cache: 6 MB Memory: 4 GB Bus Speed: 800 MHz Boot ROM Version: MBP41.00C1.B03 SMC Version (system): 1.28f2 320519 9 mihaip 2010-12-10 17:53:34 -0800 I briefly investigated this. It looks like the problem is that the GraphicsContext::fillRect implementation in GraphicsContextCG.cpp ignores shadows altogether for the case where there's a gradient. It ends up clipping to the rect to the filled and calling Gradient::Paint (as implemented in GradientCG.cpp). That just uses CGContextDrawLinearGradient, which doesn't use the context shadow. Fixing this probably involves drawing the gradient into a temporary CG context and then drawing it into the canvas one using the codepath that drawImage uses, which uses drop shadows (and does the right thing for alpha). 329985 10 helder 2011-01-05 19:40:08 -0800 (In reply to comment #9) I just found this bug accidentally. Your analysis is totally correct. I recently noticed the issue with gradients+shadow and have fixed it in bug 51869. 330198 11 andreip 2011-01-06 10:06:33 -0800 It looks like these tests were fixed on non-Skia platforms (i.e. Mac) by http://trac.webkit.org/changeset/75139 330250 12 mihaip 2011-01-06 11:22:56 -0800 Marking this as fixed, since bug 51869/http://trac.webkit.org/changeset/75139 fixed this for Mac. Bug 51989 is open about the Chromium/Skia failures, other ports should open bugs as appropriate. 392021 13 john.krueger 2011-04-25 16:15:21 -0700 using WebKit r84622, and the code here: http://jsfiddle.net/simevidas/MDgR3/ I am unable to see a shadow on the text that has a gradient. 414635 14 simon.fraser 2011-06-02 20:55:11 -0700 *** Bug 20933 has been marked as a duplicate of this bug. *** 16364 2007-09-23 15:27:41 -0700 2007-09-23 15:27:41 -0700 test case gradient-pattern-shadow.html text/html 1055 excors PCFET0NUWVBFIEhUTUw+Cjx0aXRsZT5jYW52YXMgZ3JhZGllbnQvcGF0dGVybiBzaGFkb3cgcmVu ZGVyaW5nPC90aXRsZT4KPHA+VGhpcyBzaG91bGQgYmUgc29saWQgZ3JlZW46CjxwPjxjYW52YXMg d2lkdGg9IjEwMCIgaGVpZ2h0PSIxMDAiIGlkPSJjMSIgc3R5bGU9ImJhY2tncm91bmQ6IGxpbWUi PjwvY2FudmFzPgo8cD5UaGlzIHNob3VsZCBiZSBzb2xpZCBncmVlbiB0b286CjxwPjxjYW52YXMg d2lkdGg9IjEwMCIgaGVpZ2h0PSIxMDAiIGlkPSJjMiIgc3R5bGU9ImJhY2tncm91bmQ6IGxpbWUi PjwvY2FudmFzPgo8c2NyaXB0Pgp3aW5kb3cub25sb2FkID0gZnVuY3Rpb24gKCkgewogICAgdmFy IGN0eCA9IGRvY3VtZW50LmdldEVsZW1lbnRCeUlkKCdjMScpLmdldENvbnRleHQoJzJkJyk7Cgog ICAgY3R4LnNoYWRvd0NvbG9yID0gJyNmMDAnOwogICAgY3R4LnNoYWRvd09mZnNldFggPSA1MDsK ICAgIGN0eC5zaGFkb3dPZmZzZXRZID0gMTAwOwogICAgY3R4LmZpbGxTdHlsZSA9IGN0eC5jcmVh dGVMaW5lYXJHcmFkaWVudCgwLCAwLCAxMDAsIDApOwogICAgY3R4LmZpbGxTdHlsZS5hZGRDb2xv clN0b3AoMCwgJ3JnYmEoMCwgMjU1LCAwLCAxKScpOwogICAgY3R4LmZpbGxTdHlsZS5hZGRDb2xv clN0b3AoMC40OSwgJ3JnYmEoMCwgMjU1LCAwLCAxKScpOwogICAgY3R4LmZpbGxTdHlsZS5hZGRD b2xvclN0b3AoMC41LCAncmdiYSgwLCAyNTUsIDAsIDApJyk7CiAgICBjdHguZmlsbFN0eWxlLmFk ZENvbG9yU3RvcCgxLCAncmdiYSgwLCAyNTUsIDAsIDApJyk7CiAgICBjdHguZmlsbFJlY3QoMCwg MCwgMTAwLCAxMDApOwoKICAgIGN0eCA9IGRvY3VtZW50LmdldEVsZW1lbnRCeUlkKCdjMicpLmdl dENvbnRleHQoJzJkJyk7CgogICAgY3R4LnNoYWRvd0NvbG9yID0gJyNmMDAnOwogICAgY3R4LnNo YWRvd09mZnNldFkgPSA1MDsKICAgIGN0eC5maWxsU3R5bGUgPSBjdHguY3JlYXRlUGF0dGVybihk b2N1bWVudC5nZXRFbGVtZW50QnlJZCgnYzInKSwgJ3JlcGVhdCcpOwogICAgY3R4LmZpbGxSZWN0 KDAsIDAsIDEwMCwgNTApOwp9Cjwvc2NyaXB0Pgo= 16366 2007-09-23 16:51:19 -0700 2007-09-23 16:51:19 -0700 output from http://philip.html5.org/demos/canvas/shadows/various.html webkit-pattern.png image/png 6032 excors iVBORw0KGgoAAAANSUhEUgAAAMkAAADhCAYAAABiOZFeAAAXV0lEQVR42u2df2gc553GH7vj3jvc ms4eNuyCXDwl4bJBPrK6BipRH1ghBa/RQWR8EBn3jyop7TkXLpZbLie3pDnlCjmlR43swtXOwRW5 kOAYYiRDQ9YcLavQFG1AZdeHjMdU5mY5iZsBLzcv9ZC5P2Z/zK5W0spa7a5GzwcWyTOr3efdmWfe 9/u+s4/3eJ7nGZ/MIrsMiIgApEQF4f9bulEMDvVDAyG7DwUApGUjf3saF/51dvUzhsYw9cIg4Jaf TbaTwm9mYfWmkNAebz/ZBrwAmYspD0D10TvpWR5pG1bGSwDe5Jz1ePvJtrA3aBhNqbs86QKC15E2 YePSNweQL41yN7+fbBd7191b5AfUHiQ+ODeAV26WB8Fik/vJttckTbGSx7WbGUCo5eMG/fgwksUM 3rl8FZnPCpCRGIbPnMfZU0nAtTH7i0uYvpFBoSgR+9owLpw7i8SB1SdI/vYsZj+aRfoTAxKA6NGR OnUWo0PJ1T2ZLGD2l9OYvpFGwTAAPYnB5wfRn3waaqk1Qksg+VS1V7TvpHHp364i/VEWOKij/4Wz eOW7KcQ2aH3+1jVkbECt9rXof74Pxs2ruHojjUJRQD82jLG/G21QIzTZLreAS9+K45VfVP9y9tpV xJNRSOhIvaBjer39p6oTKhu2cyWLazfna4/h0WH0PwGkf/YO3nk/DRnpx8S7E+hHFldv5qEK1Bzv xEoakxenMXe3ANGTxNnvX0CqN+QFUnDslbs8UluTHJvynPLO5Yw3fgy1+9d4pN6c8sYaPnfcMx8F x9jz3nhv+b3GvMk3R2uf//L16vt7nucszniDgf0jb05640ON3qdaS82/O1bZPvr2lDfaW9WSc9Yf i+Y+nKp5v7Ufg96M+Tjtcrwr636mCW/86+vvzzzcRDuX51cdw9TFGe/Ka4nabZdznrecWeMYrn7M LDmhrkmaN4nned7ilVUf0ODr173c3HSDkynlXZ+bX3UAprLl09fxrp+pbh9P+9vvXas9oaYWygpM b7zuRCsXszXv/eZ1L7Pgn7HWrydXPd9ZqLYh8frHG35AucupVW2eXjA989dTtduHrlRO/E21y3G8 e7MTNfsmZnOeY1meZTkb799sO52Ml9roInc55z83v/p4j1yc8WYChvSNNk+TVHhYd0K+PF29Ig7V XuGuL5aOycLUmiaZCRqod8q/+tcdmPLz61+nemAsb2qo8fbJ3ur2yozQo5w3UnmdkQ17k8xrtSfJ 6Hv3Kvuuv1x7AbiSdzbdrsafUa2o9fdvsp3OfGC7/xj7MOeZ2Wkv1ZvwEsfGvMxy4/dNvDbT4LXD b5K9m6tgBGKBf6aSyTVmv5LQe5qYz1nJV35P/E0CGiTm5vKNn+zW/jN+oPzOtTNwszcysAHAzmP2 99Xt1rIJe8VG4W4e2crWa8jcletq1I6M1LYsEa9qTiZr22PLzbdrjUK+6f1bbWfvBM4PJRB7ZgQz Cznk0pPoP9D4qfqTeulY1L6WtGWoS5KtLQ+6cs3pzPqTejUCI//hIPn9PKQrYeYzeG7Pc0iv9ewD OhIAyqeaaVdPGBmYhRs83gcNgLw/X/Nab/3103jrcT4B2ajNvi31/gEA1yq70p8aGPtaclPt2vK8 2Fbb2Z9o/k6KNY53+vY87B+E946MvZ19ewFdk7h+bgAnRs4jjTFMvzve+Kk9fRjtDfQY2ZJdVjK4 dDvQu/UnGp4UY+/l4D1y4DysfYw+JVp2iZGP064WX+Y2285UMrH1tbAIQr2e1lGTFH7zDtRDA3jr NgCkkLEmMdKvr/HsGMY+vYepl0sm+PlJPHfuAk4fPFHpXSY+zGHs61rD4dls1gQUARGpfWzpKl73 Hv2H44/Rri2y1Xa6LRgqhXw9rf0mKS+EyTzGjp6vDpMuTqBfA+SqgxY4wEJH3xH/ZEu9NoFUPIq+ tycx9d7HMC0P40OJ6lMP92EwuObx47eQXqm97l/79mlcuyMfTz8Acy5TWycd0h6vXQ162I164G1v J1nDJPXHpv4KVLRQWOOEqSt3q8OAuucYebNyBbOD49pXJ3H15jVcePGV2upmKQ/jgf/MuX95DgOv zgIYxdWfjGPse2MY+94Yzp4aRKx+QKwlMTJUM3LGc8fOY+6BBKSN2X86gdM/L0CPba43+eBGpnLy ZW5dC+wZwcAT4rHaVd8bmJYN2HN4bs9LyMsN9otNtlPKJo/hehcIsWq4tTsWEx/eq5lKLa91zCxW pyrvfTheu//MlL84+DDnjdUtck0v+H+XuVy7PpB4fcafNnbmN5yvr05FX/Ecz/QmAlOdiWMpb+TM iDcylPJSQylv5MyoN3Fx2ssE9DqL19d93fFZs4l1kpHV6wivT3nTF0cbTw1vul2e5y1/7CUa7S9P uW6wfzPtNGcbHENnrcXU8YbHzlqYrtXTu/HC7A5fJ6mb10fjlfLcwpXG+16e9CbPrLEoNZRYd7Eq U7MoNehNvJvxnEdmrVmHJr3cQ38xMWiS9R5TgbtknfxMg5M24U1+mGvqA1ptksFVJ+zo5UzNetLm 2lU6edNTNa87+NqVmrsTNtrfTDud/BrHsMF6Uf0aSfUuh6lV6ywN19RCxB7P87yOdmVS+vc11d3a Km0bUhHQAkM+4/2X8JW/ubrxaw5dgfPhaGBQIGHcMfxCW0ShH45BNDn5nf/ZaTz9t9Vh1ZW8g9HD EsZ9ExIC8R4dWmRr7QoW0eXvvInH2b+FdpLtWidpySxw49vxhabVbncLmP2oZJBjIxg/PoCnewSs FQlp2zAX03jnF6UVg2LpBA1OyT6VaJGnJSA06E9prWlX3ZhfRNavCdbd38J2km4ySbMn5++v45Wf +79PXZzG2d76Z4wjYe/BSzeBxLG+7VvY4m3qNEm3Ig5U1xleefU8oj84jcEjOgQkrIKBuVtX8dJN f13i6sv9LXtfu2jXziwZNtAb45mzm9hR327Nz3jjZ1Jr3q4+9va0l2vZN1sdb3qNCQmcueKF++Zw 0l2F++MOv4p2ZcVbKFtfPV9zUsFt3P8Kfod217BjTUJIu9jLj4AQmoQQmoQQmoQQmoQQmoQQmoQQ moQQmoQQQpMQQpMQQpMQQpMQQpMQQpMQQpMQQpMQQmgSQmgSQmgSQmgSQmgSQmgSQmgSQmgSQghN QghNQghNQghNQghNQghNQghNQghNQghNQgihSQihSQihSQihSQihSQihSQihSQihSQghNAkhNAkh NAkhNAkhNAkhNAkhNAkhNAkhBEo3iEjfSWPwqcEtvYYsAtKVEAoARUCI5v+2YBdw9t/PAhHAcA3A BaAB+pM69CUdqSMp6uuwvk6yx/M8r9Mi4j+O4+RXT2Li6AQ0oW3qb/N3JKavZZG/Y/oHRwGiEQHt gMDAs3GkjusQkbWPuHQlzr9/HvP5eQwcG8DsJ7OQjoR2UAM0wF604agO9XVIXzfwhTfeeOONTouI qiry9iKy97L4xp9/o+m/m/ukgAv/mMHt/zQgpUTfX8Zx/75E/r+KWHrg4lcff4qis4K/6k+sObD8 6cc/Rf6/8zh48CD2H9yP5f9ZhrJPgZQS7ucuIl+K4B+O/j31dUAfa5JgTyJ09Ol9SK+kYRftJq+A BVz44TTyi3kABuAayP8+C9PMQ0oDUpqAK5H+aB6FQuPXtKWNzJ0M4AKqVP2NLiAgoCkaNKlBuIL6 OqCPNUkdE7cnYEdswAasogUtsnGXPH1tBoZhIHZAh6pGS1s1JJMqVEXD/EIW5oN5xA/pMFcsxHpW v2b+7jyEKaEqGkRCAD2AGlGRPJBE3s7Dsi2gSH2d0EeTNBjXSltCoMlq0QWMuwYcKVFYMSAEIIRA XAhEtT5Ytgm4DqIaYJo2LEtW36tQgJnPoXAnj/mPZvH0ov869otJyIQOTWiYeGECH6x8gKmbU1AL KizXor526qNJGiOEQLwnDlWoTalWRRzxGDDQH0f6dgZmAbBtCdOU/lABEtqBOOwVAfu+gcInNpwl E4U7eRiGAWvFhJASuiJgAZDvZ2FGJeRR4MLvLsBesoEi4MChvnbrY03SoCeBBJTqWLYZEgkdQlGR eDKOeDwOVcQhRBy2Pe8fFNcCHsxjUGQRN7LQXKCwMI9cdh7Wij+TYwMQz/YhMZRCVAK4ZUCaEpl8 Brn7OaiuClWo1Lfd+tw2PXZ0T6IBsiih3dUghprrkhO9cRgXZzHxtgVVsQE4kNKG4wJaMQvdzUEA iCOOmCqQvZ3G/O+y/joAgL5Tw4AL6P0DiPUmkJ3LIG4UYNyygAMC2Ac4iuNf+ahvW/VJt03nWeB9 hNK8E7rCJCIrIFckTrx4oum/6Tui4+kndeQWJaBE/VkV4UA/BGBhDpA2ZEQAUiJ9Ow0IIBqLY/DV szAXDSSeH4RQBOben4aZzUJTAKkIxAsChisrawa2sCFAfdupT8o2n29KwJglB4h1HNEVJjFsA6eT pzH8/HDTfxPr0TB8KoncP6fhSH92Rj+oYeRMAumfpBEtOpBCheFKCEdCEwJC0xB7JgkIgfTPLgGK gHE7Dblk+KvMigDMAsweB7rQMXBoAOmlNIwC9W2nPllsbxUuyz2JUpqyLm1byyxdsZi4T9mHCy9e gBCbm/34iyM6vhxX4f7xIbBXwv2jgju5FRT/bwX7/8yFiMaxtBdIRKNQvijwFezBof9dxm9v/QoL v53Dyh+WoAiByBf2wSwUsGcvoEcikIkvo7C/iMXP84j8SQTfOfYd6ttGfUt/8HuTbX+4pV7L9X/f 4wLu5wA+93+WPANlb2213hW3pWy58JcS9oqEVSzN0OQzyFycgJQSeVcieUhH4oCGuBDoEyrSKxYs V/pXDxfQNA19L5zE/PvTkEsm7G8OwIoB+iEdJ586iVgkRn3bqC99uz3nibrP7yHKvYgQ/u+q8H+v 2adUexMFIUAIgViPQOVQfG0Yxo2rkIaB/AMTZkSDrmkwihJmUQKuhFCEP8b+9ijMRQPJU8OQ9w1k Pr2EkWdPQ/9qkvrapK9QaE8dYtcZRFX9n9IFoij93qCoD4VJGjE4NIzJH16AKF8pbRsxOICiQj+e Qt+p0zCz84g9k4S1kEf63Hngsyz0WAzRnjj1tVGf2SaTlHuIcs8hI0B0v98bWqgaBW7JLCWHhNYk 8WODwGEdeGBAHNDgWCaiCgDXRlzToPXEkP9lHoVPM5CfZRFdMmEBwPBJaLEY9bVRn7m0/cU6SgZQ 9wEi4v8eLa2fiAgQVQCnrpgPfU8Se0LH4LN9yK7YQDSKqAsISEAA9o0PMPtpFvbCPCABBxJ2REAM n0Tqu2epr8362tGTINCDiIfVHgQu/OlrCUgBqMHFRyXkJpHSRvyIhPxIQi7koWkaMhGB+H4VUSlh 3k6j4ALzAogfG8Dpc+OIHdaprwP6tr0mUao1iRYB1EjABKX6xBGBn924TrId2LYB8YSJ6Lc0GO9J wLZhIAbbdWADsOICOCygH9GReLa/rScg9dX1JMttqkf8tVFosmoSofi1SXlquDw9LFD6GeaaxFox UHhgQhwWSHxfh7QlBgoS9jJguQA0iWjUhtCA+IE49XVQn7XNXzFRAyZBYKVdlMzhSEArmUO6gBow CNwQmwQQcFx/pRgAtCf8ac78XRuwSx2qK/1CNJagvg7q2+4Vd6lUT3gbtb2K86hqjoqB3F0y3NIP D8BcPg3bTgOwYRcEHBuQtoAeS8IsWhCKQDyWhKbp1NdBfdt+71ZgYRBKdWjlyNphlvPI71GwW2oS EdEweHQcxoMU8vlZGPfnYS37d7vCTsM0DQgRRer4CerrsL62mKQ01FIDdUfl/d3170QOxW0pTc2g PDCQW8yi8CAPy85BShvJr57G4NGRrrhU7GZ9e/a0wfTl6d8IENWA+EEgFgPiMSB+yP8Zi/nbo1p1 LSXUhXs9sR4dsR69dOUq3VrRRd+1pr7uZdeYpPaqIoAu/r419XUXjDklhCYhhCYhhDVJO5BSIv1Z GnkrD7FfQNd09B3u2/IXmqgvRpO0gq5NlS92eWr7LtLXSZgqz1T5juvr9nUSpsozVb7j+n70ozYM mZTS44v+90r2/ykQiQD7I8D+L/k/IxF/uyr85ymKHwrBVHmmyoOp8jugJmFqO/XRJBsVjUyVpz6a pJnCiqnt1NedMFWeqfJdqY89ST1Mbac+mmSDoRZT5amPJlkfpspTXzfDVHmmyndcX7cvJjJVnqny HdfHr++2afqYqe3h1ccp4G1icGgYuRWrJhVd2CZQtKEfG8TJi5eQPJ5C7JkkhJRInzsP59ZsW1Pb qY+Fe0dhanu49dEkLYCp7eHWR5O0pJhnanuY9dEkLYCp7eHWR5O0AKa2h1sfTdISmNoebn3tI7RT wPrhAeiHT0O6KqRro1CQMO7LSiq6iAhokc6mtlMfTdLZfqSUip5MTkKIQZgFAWNJwnIsmHYa5pIB 0wQSiRPUtwP1tROmyjNVvuP6eFtKl8DU9nDrY+He8qsKU9vDrI81CSE0CSE0CSGsSXYiTG0Pt74d bxKmylNfN8NUeabKM1WeqfJMbe92fUyVb6YnYWo79bEmWR+mtlMfTbJR0chUeeqjSZoprJjaTn3d CVPlmSrPVPkd0ZMwtZ36aJINhlpMlac+mmR9mCpPfd0MU+WZKs9UeabKM7W92/Xx67ttmj5mant4 9XEKeJtganu49e2+KeBtgKnt4dZHk7QApraHWx9N0pJinqntYdZHk7QApraHWx9N0gKY2h5ufTRJ S2Bqe7j1tQ+myjO1fUfqo0lacR1kanuo9bUTpsozVb7j+nhbSpfA1PZw62Ph3vKrClPbw6yPNQkh NAkhNAkhrEl2IkxtD7e+HW8SpspTXzfDVHmmyjNVnqnyTG3vdn1MlW+mJ2FqO/WxJlkfprZTH02y UdHIVHnqo0maKayY2k593QlT5Zkqz1T5HdGTMLWd+miSDYZaTJWnPppkfZgqT33dDFPlmSrPVHmm yjO1vdv18eu7bZo+Zmp7ePVxCnibYGp7uPXtvingbYCp7eHWR5O0AKa2h1sfTdKSYp6p7WHWR5O0 AKa2h1sfTdICmNoebn00SUtganu49bUPpsoztX1H6qNJWnEdZGp7qPW1E6bKM1W+4/p4W0qXwNT2 cOtj4d7yqwpT28OsjzUJITQJITQJIaxJdiJMbQ+3vh1vEqbKU183w1R5psozVZ6p8kxt73Z9TJVv pidhajv1sSZZH6a2Ux9NslHRyFR56qNJmimsmNpOfd0JU+WZKs9U+Z0wBbztuIB0ASn99QDLBsxl oFAAzAJgLvk/CwV/u2WX1g2k/yA7n61MAfO2FEJ2wnCLEJqEEJqEEJqEEJqEEEKTrI1SfQgFUJXa bSQ8x1ctHePNHN9dfQqI8gcW+HfQLFDg5x24PM92ukmE0uD4rnEe0CSlVqv7ALv8gQn/9mhbqS46 VYzh0iRhMYkqAsc38O9KL7OvsSOU3fiB1V9d1H3VD09VUbnlAgqguoBDk+xo1MDFUIv4x7hyvPfV 9TINXKHsGmO4/gchUfuhBK8u0Uig11BKt6S4gKBJdvywujJiUH2jRCMBo4i6IRkCw6/dlOAoFP/+ LaEATuBDq9zPUx5WlbbbRVTu+eJwKxzDLSilniRSMko0cI+WWF2vlH/uyppEVQBZuqrICPw7WOtM okYAR9IgoapDg6OG/YEeJTAEU3d9TVI3u1H+wDQtsD3i3wHsPArcAUyjhKIOrdQgpd4jqvlGqYwo 1pga3nU9SXBqV7r+FUXuqxpHSsCJVIdZkgYJV10S7FEi/vEP1iaNpoOV3XQ1EfAL9/KQKgr4/62y AkQVwBG+STSXJgmzSYJFfHDSZlVdsptrksosV8koTqlGKa+PlA3iPOLJFapadN9qo4i69ZJGi4q7 5j/xqVDfSwRNUd97sBcJZ20SuEWlxjQNepFdaxKg1ghr/U7Cb5SGv2O3m2Qds5RhHbIL6pNGplmj ANm9JkHjHoMG2YVG2aA6p0k2YSASjqHWZvl/TNRbiZBsEr8AAAAASUVORK5CYII= 16367 2007-09-23 16:54:32 -0700 2007-09-23 16:54:32 -0700 output from earlier test case webkit-output.png image/png 5100 excors iVBORw0KGgoAAAANSUhEUgAAAMIAAAEeCAYAAADYasrRAAATs0lEQVR42u2dQWgbZ9rH/1N8mEAO E3BBhhTqZQ+VcQ4Oe1ib3YPH7GWMC19DFyqTXOwUts5XSO0NNO7uwXX24ChbcKT0YicnKZBiHxKk wBYph35IhV3kgItUyCIfEhhBAxqwYQZseL7DjC3Z0oykOE5S5/8rKcTzat73ed73p3mfd5oU0gHV JzlJ/VD2vW6bBVm5l5OqvBxa9XdoNk3Jpf3Ha1fLkrrvf91cz0hmrSqvlQMxtJqDVjG9raB4OyIA An/F/l2U+X7v9/0x3yTmFtw20TX7cKPaLrfV32Eo35v24tMlt9mwXCT1pe5/3S7KtJeb8ELutU1e sxj856BFTG+7CIVFQ4yFjFQ3bbHNjOiAYDgqpm2LXTUlcRESXiiI2N7iHI6J3zK3n+YkkcxIdfsl jKyN/g7ZgeQWDQEMKTRbFNu2pOZ03+v2LzkxADEWC69x+hpjCJyDFjG9zXSp3Qai/6NDUwF0nUII AE6q0FQVqhpCZCGH6vcqoPYg1AtgC76opwcR+QQvhzb6O2QH0LoCLnepCP82BMBq/umTKgDAgYrX R2MMgXPQIqa3mXfCn0whHDSX2iCmPg7Xfn8SWHsYR5+iQDkziuRjL6nWBtJ3rmHkzAxKjrtE8ndm 0Kf0oe/MCEb0USR/dpp00KKdX38A4FSQ/HociqJAURSMXl1GxQGws4EbF8YxfmEc174rAQDS1ycx /uk4Jr9Ow/EN1kL6une/M334TSTZxlLcQPrbSSiKgj59Eumf68e3gfinI974+vDVnbxP3wE58Iux 6fAPzsELxORUsHx1dG/MM9eXkX6YRvrRGio/5RG/Oo6RL+JYvTkDRVFw7WElOE6fa+69RjHyRRL5 R3GMePO7XDe/a3cmoSijyFZegQn7n7QFiQCCsWbbEVuWznv79uFZWbkX87ZRbttyOiZ6PwQw3P3n k4QAhhRtEdl099TRZoWlb7vg/kSqEht2x2pui1TXExIGBP1RMUVEqjmJAKLvbl22qxIbcz+/O4ri rfpthS0rFyHArBQ3RewnKbc/v23Ebq4AwdiszF82avvvqoiIKbOATCQL3hbGrcUm7pU7yEGLGA/E 0DAHncYkVVkaq93fLiX26kTjH0sS3YsRgn5DwoBEbmcC4vTJQTIjsX33mpWV5MH5FUlddq8vleyj rxE6EcFdSNG9hVRY1AWYlbK3H3ULbzfJdmnJLdruF90UZ6MSyzeK4N8uuD97LeYW8uu1kRZvuYme TZuuSMOQcN0evnjLEPTXYqtfRPa6O45YXZGZW9ADRTAACX+ZqrVfdCc3crsoVW988/dSkrqfktTt ad/i3y8HrWM8KPOBOeg0pk0vpoXCnhjRfgjGlryc2Z4o866Ith0YZzkwB9696ubXHVvE/UJwj7jE fPpqzrfe6WxH6v57dyeldqkA8rVHtVN7rKlaD8IAZj7sg6JPIntiHFO/1xrvGdAuqL+NfBpAGOFQ bV/X+8chAED+v6a34QB6DmxB/Cg9WgVgYOiD2v20k633/73v1XoYPP8VdACVLccbn4He3Vt060il U8jdPQetzRy0E2Pj1sZ68Zi8eqP0xKzVIL1N2r2ruTGoamCcVls5qM2vOzYL1s5eYhA6rb2SGqHr ZRRsTVMbMlB4msP8Z0O49mAZ5waXMX2/jOhYb5vtelr0pwEoofTMgd7tKfP+EIzA5R4QRZcriuPs GdgWzk5QNjUMjBkI71unTtu5mjhkjB3HpA5gek5H+u+jiF8oQ99cxeQDYOKe3uLjzeMsdZKDhi/c V1wsH9WNncdxLD8bwPx9QTkbgw7gxjfphvOKdts15v4EAMA0rf3FIgC9vyZbdsPcv2jf9Rmv2xrL 32/Uf1XXLSi/xVY3bVtVZAEM9vdA6w4DSGLmZr5ufFmMfrTcsIj9cuC0GaPvHLxATGf/MAAgAqyv YvU/KnKlKpY+9u8rKE61gxy89lOjxnMT/yNL58DPrR2nyTehsxfkpYtxVAD0Dk9hYgzA6Z6mtvu1 C+ovPDYFA8A1Yxr55+6P83fjACbw0WDd4/SbLPLPKsjfmcHZz7PAozQSD9dqj19vvANjEwCA5T9P YvnHCpznJST+mQSwgez3+br2Hjvu59L5HCo7buaSV0YBzOPScAi9fzoHHUD68yGMfHEDq3fjGDk1 goG/nms7B23HWJfzw8W0gRn9BsLn+9DzfhgDv+uFbRaQf7xRu+/W/k1mUJzhFjlotZ7yN8ehKOOv 9tSoml8So7/2Njk8PC0Z027yVhISWViR1OJEre35mBT+Fa07BZiXolcoAfBOMiKSMZvUnOvN2rXu z9x2T0GM3Z95n009rRWGZnq+NqbLS5KY0wUIy8TCihSzsbrxTkjuF5FyOtrkzXpYInMJMbcbDw8y t6b3tx2LSrGuCK2uJbxTGu/kZSHT9OVg8xx41wJiNH/YH8PK6vy+OShvdx5T4nzz/7rAuJWRpYu1 OQlfXNorcoPibH7NlpXLdfO7mNo3v+4pWW0siVdwaqSIiByZZTuAs2XBcgAtpPnv/dpt57MBsCoW nC4VoW6tSe1owdpREepWgS0LjqoFbnXgWKhYDrTuENQdCxY092Wj79gdWM8toEuD1q36X1c1aEE3 CsxBcIytU9RmTJU0RnrimDdXMKgBluN+Q5v5OPquaLDXpwLmMCDOdnPQ5J4VC+7cHTFHKwL5VbF2 vQ9nrwwhU4pC/0Dbkyg9N4T4eytI/SV8bGOnCKS2R/8xjlODlxp+rs+lsPI3A9oxjp0ikIZtVOlx CaZtAziB3jMD6O1Wj33YFIEQHOF7BEIoAiEUgRCKQAhFIIQiEEIRCKEIhFAEQigCIRSBEIpACEUg hCIQQhEIOaYiKPj1/3PjigIov+5fM9fdWAifCC/OzjGIoYuLkSJwER0PmSkCIRSBEIpAuL2jCIQ1 AkUghCIQQhEIoQiEUARCKAIhFOGNhkePhCKAL6MIRSCEIhBCEY5ZjcDtHUXgImLBTxEIoQiEUATC GoEiENYIFIEQikAIRSCEIhBCEQihCIRQhKOHR4+EIoAvowhFIIQiEEIRjlmNwO0dReAiYsFPEQih CIRQBMIagSIQ1ggUgRCKQAhFIIQiEEIRCKEIhFCEo4dHj4QigC+jCEUghCIQQhGOWY3A7R1F4CJi wU8RCKEIhFAEwhqBIhDWCBSBEIpACEUghCIQQhEIoQiEUISjh0ePhCKAL6MIRSCEIhBCEY5ZjcDt HUXgImLBTxEIoQiEUATCGoEiENYIFIEQikAIRSCEIhBCEQihCIRQhKOHR4+EIoAvowhFIIQiEEIR jlmNwO0dReAiYsFPEQihCIRQBMIagSIQ1ggUgRCKQAhFIIQiEEIRCKEIhFCEo4dHj4QigC+jCEUg hCIQQhGOWY3A7R1F4CJiwU8RCHkDUEREmAbCJwIhhCIQQhEIoQiEUARCKAIhFIEQikAIRSCEIhBC EQihCIRQBEIoAiEUgRCKQAhFIIQiEEIRCKEIhFAEQl6dCNZ/80j/34bvdaeyhtXv8rBe0uBa9Xdo tirIP/Qfr2NtIP3A/3rlpyyyj63XO4MHYmg1B61iemsp3o4IgMBfsX8XZb7f+31/TKrSnNyC2ya6 Zsuh2C631d9hKN+b9uLTJbd58KotqS91/+t2Uaa93IQXcvK6aBaD/xy0iOktB4VFQ4yFjFQ3bbHN jOiAYDgqpm2LXTUlcRESXiiI2N7iHI6J3zK3n+YkkcxIdfsljKyN/g7ZgeQWDQEMKTRbFNu2pOZ0 3+v2LzkxADEWC69x+hpjCJyDFjG9+ZQl+mXiSNbDO2q3gej/6tBOqlC1UwgBwEkVmqpC1UKILOQw 9b4KqD0I9QY/XdTTg4h8okN7GX8FYxv9HbKD4HF2qQj/NuT/6ZOqu9WA+hqf540xBM5Bi5jedPLf XsLMg+rR1AjhT6YQDppLbRBTH4drvz8JrD2Mo09RoJwZRXJ3j2xtIH3nGkbOzKDkuEskf2cGfUof +s6MYEQfRfJnp9muNbidX3/uhhjJr8ehKAoURcHo1WVUHAA7G7hxYRzjF8Zx7bsSACB9fRLjn45j 8us0HP+qBOnr3v3O9OE3kWQbS3ED6W8noSgK+vRJpH+uH98G4p+OeOPrw1d38j59B+TAL8amwz84 By8Qk1PB8tXRvTHPXF9G+mEa6UdrqPyUR/zqOEa+iGP15gwURcG1h5XgOH2uufcaxcgXSeQfxTHi ze9y05qrgviHCoY+SwM/XcJZfRTJn5zWuWlxfe3OJBRlFNkKgP1P2oJEAMFYs+2ILUvnvX378Kys 3It52yi3bTkdE70fAhju/vNJQgBDiraIbLp76uhak92+b7vg/kSqEht2x2pui1TXExIGBP1RMUVE qjmJAKLvbl22qxIbcz+/O4rirfpthS0rFyHArBQ3RewnKbc/v23Ebq4AwdiszF82avvvqoiIKbOA TCQL3hbGrcUm7pU7yEGLGA/E0DAHncYkVVkaq93fLiX26kTjH0sS3YsRgn5DwoBEbmcC4vTJQTIj sX33mpWV5MH5PTAysyjzw24shaemVO1WuWmdu9Rlt/+lki0dieAupOjeQios6gLMStnbj7qFt5tk u7TkFm33i+6wslGJ5RtF8G8X3J+9FnML+fXaSIu33ETPpk1XpGFIuG4PX7xlCPprsdUvInvdHUes rsjMLeiBIhiAhL9M1dovupMbuV2Uqje++XspSd1PSer2tG/x75eD1jEelPnAHHQa06YX00JhT4xo PwRjS17ObE+UeXcx2XZgnOXAHHj3qptfd2wR9wuh2RfxWE2UVrlpJ3diV8V86vb+Tmc7Uvffuzsp tUsFkK89bpzaY03VehAGMPNhHxR9EtkT45j6vdZ4z4B2Qf1t5NMAwgiHavu63j8OuXvJ/5rehgPo ObAF8aP0aBWAgaEPavfTTrbe//e+V+th8PxX0AFUthxvfAZ6d2/RrSOVTiF39xy0NnPQToxNzkdf PCavrig9MWs1SLM67V3NjUFVA+O02spBbX7dsVmw2vgr8lvlpq3cqRpCp7X60A9XsDVNbchA4WkO 858N4dqDZZwbXMb0/TKiY71ttutp0Z8GoITSMwd6t6fM+0MwApd7QBRdriiOs2dgWzg7jQuprsDC wJiB8L516rSdq4lDxthxTOoApud0pP8+iviFMvTNVUw+ACbu6S0+3jzOUic5aPjCbUWr3HSWuyN7 s+w8jmP52QDm7wvK2Rh0ADe+STe8yGm3XWMeTgAATNPaXywC0PtrsmU3zP2L9l2f8bqtsfz9Rv1X dd2C8ltsddO2VUUWwGB/D7TuMIAkZm7m68aXxehHyw0T4ZcDp80YfefgBWI6+4cBABFgfRWr/1GR K1Wx9LF/X0Fxqh3koGNa5abD3L3TeG4CYMsnsQd+bu04Tb4Jnb0gL12MowKgd3gKE2MATvc0td2v XVB/4bEpGACuGdPIP/eO1+7GAUzgo8G6zcc3WeSfVZC/M4Ozn2eBR2kkHq7VPX7d8Q6MTQAAlv88 ieUfK3Cel5D4ZxLABrLf5xsf1zvu59L5HCo7buaSV0YBzOPScAi9fzoHHUD68yGMfHEDq3fjGDk1 goG/nms7B23HWJfzw8W0gRn9BsLn+9DzfhgDv+uFbRaQf7xRu+/W/k1mUJzhFjlobz1hf9+/mMg+ iGPZMgJz007u8jfHoSjj+0+NqvklMfprb5PDw9OSMe0mbyUhkYUVSS1O1Nqej0nhX9G6U4B5KXrF CgDvJCMiGbNJCbTerF3r/sxt9xTE2P2Z99nU01pxZKbna2O6vCSJOV2AsEwsrEgxG6sb74TkfhEp p6NN3qyHJTKXEHO7sXjL3Jre33YsKsW6IrS6lvBOabyTl4VM0xOR5jnwrgXEaP6wP4aV1fl9c1De 7jymxPnm/3WBcSsjSxdrcxK+uLRX5AbF2fyaLSuX6+Z3MbVvfndPeg5SuLXbJiK5auv5D75eizVR suVo/485O4CzZcFyAC2k+e/92m3nswGwKhacLhWhbq1J7WjB2lER6laBLQuOqgVudeBYqFgOtO4Q 1B0LFjRoQQPacWA9t4AuDVq36n9d1aAF3SgwB8Extk5RmzFV0hjpiWPeXMGgBliO+w1t5uPou6LB Xp8KmMOAONvNQVuxOICqdpCbgOs7DioWEOpW+b+OInUvmK734eyVIWRKUegfaHsSpeeGEH9vBam/ hI9t7BSB1PboP8ZxavBSw8/1uRRW/mZAO8axUwTSsI0qPS7BtG0AJ9B7ZgC93eqxD5siEAL+CTVC KAIhFIEQikAIRSCEIhBCEQihCIRQBEIoAiEUgRCKQAhFIIQiEEIRCOkQBQL+eYQ3CE4HnwiEUARC KAIhFIEQikAIRSCEIhBCEQihCIRQBEIoAiEUgRCKQAhFIIQiEEIRCKEIhFAEQigCIRSBEIpACEUg hCIQQhEIoQiEUARCKAIhFIEQikAIRSCEIhBCEQihCIRQBEIoAiEUgRCKQAhFIIQiEEIRCKEIhFAE QigCIRSBEIpACEUghCIQQigCIRSBEI8uUZiENwphCvhEIIQiEEIRCKEIhFAEQigCIRSBEIpACEUg hCIQQhEIoQiEUARCKAIhFIEQikAIRSCEIhBCEQihCIRQBEIoAiEUgRCKQAhFIIQiEEIRCKEIhFAE QigCIRSBEIpACEUghCIQQhEIoQiEUARCKAIhFIEQikAIRSCEIhBCEQihCIRQBEIoAiEUgRCKwBQQ QhEIAQD8P18mYCEl/npBAAAAAElFTkSuQmCC 64888 2010-08-19 12:44:12 -0700 2010-08-19 12:44:12 -0700 Test repro case BugID15266Repo.html text/html 3060 mlamb1976 PCFET0NUWVBFIGh0bWw+IAo8aHRtbCBsYW5nPSJlbiI+IAogIDxoZWFkPiAKICAgIDxtZXRhIGNo YXJzZXQ9InV0Zi04Ij4gCiAgICA8dGl0bGU+QnVnICMxNTI2NiAtIENhbnZhcyBzaGFkb3dzIG9u IGdyYWRpZW50cyBhbmQgcGF0dGVybnMgZG9uJ3Qgd29yayBzZW5zaWJseTwvdGl0bGU+IAogICAg PHNjcmlwdCB0eXBlPSJ0ZXh0L2phdmFzY3JpcHQiPjwhLS0Kd2luZG93LmFkZEV2ZW50TGlzdGVu ZXIoJ2xvYWQnLCBmdW5jdGlvbiAoKSB7CiAgLy8gR2V0IHRoZSBjYW52YXMgZWxlbWVudC4KICB2 YXIgZWxlbSA9IGRvY3VtZW50LmdldEVsZW1lbnRCeUlkKCdteUNhbnZhcycpOwogIGlmICghZWxl bSB8fCAhZWxlbS5nZXRDb250ZXh0KSB7CiAgICByZXR1cm47CiAgfQogCiAgLy8gR2V0IHRoZSBj YW52YXMgMmQgY29udGV4dC4KICB2YXIgY29udGV4dCA9IGVsZW0uZ2V0Q29udGV4dCgnMmQnKTsK ICBpZiAoIWNvbnRleHQpIHsKICAgIHJldHVybjsKICB9CgogIGNvbnRleHQuZ2xvYmFsQ29tcG9z aXRlT3BlcmF0aW9uID0gJ3NvdXJjZS1vdmVyJzsKICAvLyBzZXQgdXAgdGhlIHNoYWRvdyAKICBj b250ZXh0LnNoYWRvd09mZnNldFggPSA1OwogIGNvbnRleHQuc2hhZG93T2Zmc2V0WSA9IDU7CiAg Y29udGV4dC5zaGFkb3dCbHVyICAgID0gNDsKICBjb250ZXh0LnNoYWRvd0NvbG9yICAgPSAncmdi YSgwLCAwLCAwLCAwLjUpJzsKICAKICAvLyBkcmF3IGEgY2lyY2xlIHdpdGggc29saWQgZmlsbAog IGNvbnRleHQuYmVnaW5QYXRoKCk7CiAgY29udGV4dC5hcmMoMjUwLCA1MCwgMjUsIDAsIE1hdGgu UEkqMiwgdHJ1ZSk7IAogIGNvbnRleHQuY2xvc2VQYXRoKCk7CiAgY29udGV4dC5maWxsU3R5bGUg PSAnI2ZmMDAwMCc7CiAgY29udGV4dC5maWxsKCk7CgogIC8vIGRyYXcgYSBzcXVhcmUgd2l0aCBz b2xpZCBmaWxsIAogIGNvbnRleHQuZmlsbFJlY3QoMTUwLCA1LCA1MCwgMTAwKTsKCiAgLy8gZHJh dyBhIHN0cm9rZWQgc3F1YXJlCiAgY29udGV4dC5zdHJva2VSZWN0KDI5MCw1LDUwLDEwMCk7Cgog IC8vIGRyYXcgYSBzdHJva2VkIGNpcmNsZQogIGNvbnRleHQuYmVnaW5QYXRoKCk7CiAgY29udGV4 dC5hcmMoMzgwLCA1MCwgMjUsIDAsIE1hdGguUEkqMiwgdHJ1ZSk7IAogIGNvbnRleHQuY2xvc2VQ YXRoKCk7CiAgY29udGV4dC5zdHJva2VTdHlsZSA9ICcjMDAwMDAwJzsKICBjb250ZXh0LnN0cm9r ZSgpOwoKICAvLyBUaGUgaHVlIHNwZWN0cnVtIHVzZWQgYnkgSFNWIGNvbG9yIHBpY2tlciBjaGFy dHMuCiAgdmFyIGNvbG9yLCBodWUgPSBbCiAgICBbMjU1LCAgIDAsICAgMCBdLCAvLyAwLCBSZWQs ICAgICAgIDDCsAogICAgWzI1NSwgMjU1LCAgIDAgXSwgLy8gMSwgWWVsbG93LCAgIDYwwrAKICAg IFsgIDAsIDI1NSwgICAwIF0sIC8vIDIsIEdyZWVuLCAgIDEyMMKwCiAgICBbICAwLCAyNTUsIDI1 NSBdLCAvLyAzLCBDeWFuLCAgICAxODDCsAogICAgWyAgMCwgICAwLCAyNTUgXSwgLy8gNCwgQmx1 ZSwgICAgMjQwwrAKICAgIFsyNTUsICAgMCwgMjU1IF0sIC8vIDUsIE1hZ2VudGEsIDMwMMKwCiAg ICBbMjU1LCAgIDAsICAgMF1dLCAvLyA2LCBSZWQsICAgICAzNjDCsAogCiAgICAvLyBDcmVhdGUg dGhlIGxpbmVhciBncmFkaWVudDogc3gsIHN5LCBkeCwgZHkuCiAgICAvLyBUaGF0J3MgdGhlIHN0 YXJ0ICh4LHkpIGNvb3JkaW5hdGVzLCBmb2xsb3dlZCBieSB0aGUgZGVzdGluYXRpb24gKHgseSku CiAgICBncmFkaWVudCA9IGNvbnRleHQuY3JlYXRlTGluZWFyR3JhZGllbnQoMCwgMCwgZWxlbS53 aWR0aCwgMCk7IAogIAogIC8vIEFkZCB0aGUgY29sb3Igc3RvcHMuCiAgZm9yICh2YXIgaSA9IDA7 IGkgPD0gNjsgaSsrKSB7CiAgICBjb2xvciA9ICdyZ2IoJyArIGh1ZVtpXVswXSArICcsICcgKyBo dWVbaV1bMV0gKyAnLCAnICsgaHVlW2ldWzJdICsgJyknOwogICAgZ3JhZGllbnQuYWRkQ29sb3JT dG9wKGkgKiAxLzYsIGNvbG9yKTsKICB9CiAKICAvLyBVc2UgdGhlIGdyYWRpZW50IGZvciB0aGUg ZmlsbFN0eWxlLgogIGNvbnRleHQuZmlsbFN0eWxlID0gZ3JhZGllbnQ7CiAKICAvLyBkcmF3IGEg cmVjdGFuZ2xlIHdpdGggZ3JhZGllbnQgZmlsbAogIGNvbnRleHQuZmlsbFJlY3QoNSwgNSwgNTAs IDEwMCk7CgogIC8vIGRyYXcgYSBjaXJjbGUgd2l0aCBncmFkaWVudCBmaWxsCiAgY29udGV4dC5i ZWdpblBhdGgoKTsKICBjb250ZXh0LmFyYygxMTAsIDUwLCAyNSwgMCwgTWF0aC5QSSoyLCB0cnVl KTsgCiAgY29udGV4dC5jbG9zZVBhdGgoKTsKICBjb250ZXh0LmZpbGwoKTsgCgogIC8vIGRyYXcg c29tZSB0ZXh0IHdpdGggZ3JhZGllbnQgZmlsbCAKICBjb250ZXh0LmZvbnQgPSAnYm9sZCAzNnB4 IHNhbnMtc2VyaWYnOwogIGNvbnRleHQudGV4dEJhc2VsaW5lID0gJ3RvcCc7CgogIGNvbnRleHQu ZmlsbFRleHQoJ0hlbGxvIHdvcmxkIScsIDUsIDEyMCwgMjAwKTsKICAKICAvLyBkcmF3IHNvbWUg c3Ryb2tlZCB0ZXh0IHdpdGggZ3JhZGllbnQgc3Ryb2tlCiAgY29udGV4dC5zdHJva2VTdHlsZSA9 IGdyYWRpZW50OwogIGNvbnRleHQuc3Ryb2tlVGV4dCgnSGVsbG8gd29ybGQhJywgNSwgMTYwLCAy MDApOwoKICBjb250ZXh0LmZpbGxTdHlsZSA9ICcjZmYwMDAwJzsKICBjb250ZXh0LmZpbGxUZXh0 KCdIZWxsbyB3b3JsZCEnLDIyMCwxMjAsMjAwKTsKCiAgLy8gZHJhdyBzb21lIHN0cm9rZWQgdGV4 dCB3aXRoIHNvbGlkIHN0cm9rZQogIGNvbnRleHQuc3Ryb2tlU3R5bGUgPSAnIzAwMDAwMCc7CiAg Y29udGV4dC5zdHJva2VUZXh0KCdIZWxsbyB3b3JsZCEnLDIyMCwxNjAsMjAwKTsKCn0sIGZhbHNl KTsKICAgIC8vIC0tPjwvc2NyaXB0PiAKICA8L2hlYWQ+IAogIDxib2R5PiAKICAgIDxjYW52YXMg aWQ9Im15Q2FudmFzIiB3aWR0aD0iNDUwIiBoZWlnaHQ9IjIzMCIgc3R5bGU9ImJvcmRlcjoxcHgg c29saWQgYmxhY2siPjwvY2FudmFzPiAgCiAgICA8cD5UaGlzIHNob3VsZCByZW5kZXIgYXMgZm9s bG93czogPGJyPgogICAgICAgPGltZyBzcmM9ImV4cGVjdGVkLnBuZyIgYWx0PSJFeHBlY3RlZCBy ZW5kZXIuIj4KICAgIDwvcD4gCiAgPC9ib2R5PiAKPC9odG1sPiAK 64889 2010-08-19 12:45:10 -0700 2010-08-19 12:45:10 -0700 PNG of expected rendering of Test Repro Case expected.png image/png 0 mlamb1976