123012 2013-10-18 01:30:01 -0700 REM unit down scaling is incorrect 2015-02-06 05:42:06 -0800 1 1 1 Unclassified WebKit Layout and Rendering 528+ (Nightly build) All All UNCONFIRMED HasReduction P2 Normal --- 0 jon.ronnenberg webkit-unassigned jon.ronnenberg oldest_to_newest 941113 0 214548 jon.ronnenberg 2013-10-18 01:30:01 -0700 Created attachment 214548 Shows 2 HTML elements with their pre calculated height, run time calculated height and actual height. Steps to reproduce the problem: 1. Go to http://dotnetcarpenter.bitbucket.org/bugz/rem-bug/ 2. Resize the window to a smaller size. 3. Watch the height text inside the blue and green squares - at some point, all rem scales are askew. Or on a small screen device, e.i. iPhone/Android phone: 1. Go to http://dotnetcarpenter.bitbucket.org/bugz/real-example/ 2. Notice that the right pointing arrow will drop down below the other controls. 3. Touch the histogram icon 4. The icon legend to the right of the screen is rendered on top of the histogram bars. What is the expected behavior? Go to the aforementioned URI's with a gecko based browser. Scaling should always be correct. What went wrong? As demonstrated in the attached test case and on the live links, webkit browsers stop down-scaling correctly at a certain point. Did this work before? I don't know. Any other comments? This is observed on iPhone Safari iOS6, Android stock browser 4.0.4 and 4.2.2, Chrome 30 for Android, and Opera 16. webkit doesn't seem to having any issues with up-scaling content. As seen in the second attachment: rem-height-workaround.html I've also commented on a similar issue in the webkit bug tracker: https://bugs.webkit.org/show_bug.cgi?id=117680 941114 1 214549 jon.ronnenberg 2013-10-18 01:31:04 -0700 Created attachment 214549 Test-case shows that up scaling is not an issue 941116 2 214550 jon.ronnenberg 2013-10-18 01:32:38 -0700 Created attachment 214550 A real example of the issue Can be used as a test case when the reduced test case passes. 1067345 3 jon.ronnenberg 2015-02-06 05:42:06 -0800 -webkit-text-size-adjust:none; might solve the issue. But haven't checked since this means going through hundreds of lines of SASS to change the factor. 214548 2013-10-18 01:30:01 -0700 2013-10-18 01:30:01 -0700 Shows 2 HTML elements with their pre calculated height, run time calculated height and actual height. rem-height.html text/html 3040 jon.ronnenberg PCFET0NUWVBFIGh0bWw+DQo8aHRtbD4NCjxoZWFkPg0KICA8bWV0YSBjaGFyc2V0PSJ1dGYtOCI+ ICANCiAgPG1ldGEgbmFtZT0idmlld3BvcnQiIGNvbnRlbnQ9IndpZHRoPWRldmljZS13aWR0aCI+ DQogIDx0aXRsZT5yZW0gaGVpZ2h0IGJ1ZzwvdGl0bGU+DQogICAgPHNjcmlwdD4NCiAgICAgIC8q IGZvbnQgc2l6ZSAqLw0KICAgICAgZnVuY3Rpb24gc2V0Rm9udFNpemUoKSB7DQogICAgICAgIGh0 bWwuc3R5bGUuZm9udFNpemUgPSBjYWxjRm9udFNpemUoKSArICclJzsNCiAgICAgIH0NCiAgICAg IGZ1bmN0aW9uIGNhbGNGb250U2l6ZSgpIHsNCiAgICAgICAgcmV0dXJuIDAuNjI1ICogaHRtbC5j bGllbnRXaWR0aC8xMDI0ICogMTAwOw0KICAgICAgfQ0KICAgICAgdmFyIGh0bWwgPSBkb2N1bWVu dC5kb2N1bWVudEVsZW1lbnQ7DQogICAgICB3aW5kb3cuYWRkRXZlbnRMaXN0ZW5lcigicmVzaXpl Iiwgc2V0Rm9udFNpemUpOw0KICAgICAgc2V0Rm9udFNpemUoKTsNCiAgICA8L3NjcmlwdD4NCiAg PHN0eWxlPg0KICAgIGJvZHkgew0KICAgICAgbWFyZ2luOiAwOw0KICAgICAgcGFkZGluZzogMDsN CiAgICB9DQogICAgYXJ0aWNsZSB7DQogICAgICBiYWNrZ3JvdW5kLWNvbG9yOiBsaWdodGJsdWU7 DQogICAgICBvcGFjaXR5OiAuNTsNCiAgICAgIGhlaWdodDo1MHJlbTsNCiAgICAgIHdpZHRoOjEw MHJlbTsNCiAgICB9DQogICAgc2VjdGlvbiB7DQogICAgICBiYWNrZ3JvdW5kLWNvbG9yOiBkYXJr Z3JlZW47DQogICAgICBvcGFjaXR5OiAuNzsNCiAgICAgIGhlaWdodDogNDByZW07DQogICAgICB3 aWR0aDogNzRyZW07ICAgICAgDQogICAgfQ0KICAgIGgxLHAgew0KICAgICAgZm9udC1zaXplOiAz LjVyZW07DQogICAgfQ0KICA8L3N0eWxlPg0KPC9oZWFkPg0KPGJvZHk+DQogIDxoMT53ZWJraXQg ZmFpbHMgd2l0aCBmb250LXNpemUgJmx0Oz0gNTYlIG9uIGh0bWwgZWxlbWVudDwvaDE+DQogIDxw Pg0KICAgIElmIHRoaXMgaXMgdGhlIGluaXRpYWwgY2FsY3VsYXRpb24sIHRoZSBwcmVzZW5jZSBv ZiB0aGUgdmVydGljYWwgc2Nyb2xsIGJhciBpcyBub3QgdGFrZW4gaW50byBhY2NvdW50Lg0KICAg IFRyaWdnZXIgYSByZXNpemUgZXZlbnQgKGUuaS4gcmVzaXplIHRoZSB3aW5kb3cgb3IgY2hhbmdl IG9yaWVudGF0aW9uKSwgdG8gdGhlIGdldCB0aGUgY29ycmVjdCB2YWx1ZXMuDQogIDwvcD4NCiAg PHA+U2VlIDxhIGhyZWY9InJlbS1oZWlnaHQtd29ya2Fyb3VuZC5odG1sIj5yZW0taGVpZ2h0LXdv cmthcm91bmQuaHRtbDwvYT4gZm9yIGEgd29yay1hcm91bmQuPC9wPg0KICA8YXJ0aWNsZT4NCiAg ICA8cD5JbiBhIDEwMjRweCB3aWRlIHZpZXcgcG9ydCB0aGlzIHNob3VsZCBiZSBleGFjdGx5IDxz dHJvbmc+NTAwcHg8L3N0cm9uZz4gaGlnaC48L3A+DQogICAgPHA+VGhlIHZpZXcgcG9ydCBpcyA8 c3Ryb25nIGNsYXNzPSJ2aWV3cG9ydCI+PC9zdHJvbmc+cHggd2lkZS4gQW5kIHRoaXMgZWxlbWVu dCBzaG91bGQgYmUgPHN0cm9uZyBpZD0iYXJ0Q2FsY0hlaWdodCI+PC9zdHJvbmc+IGhpZ2gNCiAg ICBidXQgaXMgPHN0cm9uZyBpZD0iYXJ0UmVhbEhlaWdodCI+PC9zdHJvbmc+IGhpZ2guPC9wPg0K ICAgIDxzZWN0aW9uPg0KICAgICAgPHA+SW4gYSAxMDI0cHggd2lkZSB2aWV3IHBvcnQgdGhpcyBz aG91bGQgYmUgZXhhY3RseSA8c3Ryb25nPjQwMHB4PC9zdHJvbmc+IGhpZ2guPC9wPg0KICAgICAg PHA+VGhlIHZpZXcgcG9ydCBpcyA8c3Ryb25nIGNsYXNzPSJ2aWV3cG9ydCI+PC9zdHJvbmc+cHgg d2lkZS4gQW5kIHRoaXMgZWxlbWVudCBzaG91bGQgYmUgPHN0cm9uZyBpZD0ic2VjQ2FsY0hlaWdo dCI+PC9zdHJvbmc+IGhpZ2gNCiAgICAgIGJ1dCBpcyA8c3Ryb25nIGlkPSJzZWNSZWFsSGVpZ2h0 Ij48L3N0cm9uZz4gaGlnaC48L3A+DQogICAgPC9zZWN0aW9uPg0KICA8L2FydGljbGU+DQogIDxz Y3JpcHQ+DQogICAgdmFyIHZwID0gZG9jdW1lbnQuZ2V0RWxlbWVudHNCeUNsYXNzTmFtZSgidmll d3BvcnQiKSwNCiAgICAgICAgYXJ0Q0ggPSBkb2N1bWVudC5nZXRFbGVtZW50QnlJZCgiYXJ0Q2Fs Y0hlaWdodCIpLA0KICAgICAgICBhcnRSSCA9IGRvY3VtZW50LmdldEVsZW1lbnRCeUlkKCJhcnRS ZWFsSGVpZ2h0IiksDQogICAgICAgIGFydCA9IGRvY3VtZW50LnF1ZXJ5U2VsZWN0b3IoImFydGlj bGUiKSwNCiAgICAgICAgc2VjQ0ggPSBkb2N1bWVudC5nZXRFbGVtZW50QnlJZCgic2VjQ2FsY0hl aWdodCIpLA0KICAgICAgICBzZWNSSCA9IGRvY3VtZW50LmdldEVsZW1lbnRCeUlkKCJzZWNSZWFs SGVpZ2h0IiksDQogICAgICAgIHNlYyA9IGRvY3VtZW50LnF1ZXJ5U2VsZWN0b3IoInNlY3Rpb24i KTsNCg0KICAgIHdpbmRvdy5hZGRFdmVudExpc3RlbmVyKCJyZXNpemUiLCB1cGRhdGVUZXh0KTsN CiAgICB1cGRhdGVUZXh0KCk7DQogICAgZnVuY3Rpb24gdXBkYXRlVGV4dCgpIHsNCiAgICAgIHZh ciBmYWN0b3IgPSAoY2FsY0ZvbnRTaXplKCkvMTAwKS8wLjYyNTsNCiAgICAgIFtdLmZvckVhY2gu Y2FsbCh2cCwgd3JpdGUoZG9jdW1lbnQuZG9jdW1lbnRFbGVtZW50LmNsaWVudFdpZHRoKSk7DQog ICAgICAvLyBhcnRpY2xlIG9yaWdpbmFsZSAxMDAwcHgqNTAwcHgNCiAgICAgIGFydENILmlubmVy SFRNTCA9IE1hdGgucm91bmQoZmFjdG9yICogNTAwKSArICJweCI7DQogICAgICBhcnRSSC5pbm5l ckhUTUwgPSBhcnQuY2xpZW50SGVpZ2h0ICsgInB4IjsNCiAgICAgIC8vIHNlY3Rpb24gb3JpZ2lu YWxlIDc0MHB4KjQwMHB4DQogICAgICBzZWNDSC5pbm5lckhUTUwgPSBNYXRoLnJvdW5kKGZhY3Rv ciAqIDQwMCkgKyAicHgiOw0KICAgICAgc2VjUkguaW5uZXJIVE1MID0gc2VjLmNsaWVudEhlaWdo dCArICJweCI7DQogICAgICAvL2NvbnNvbGUubG9nKCJmYWN0b3I6IiArIGZhY3Rvcik7DQogICAg fQ0KICAgIGZ1bmN0aW9uIHdyaXRlKG1zZykgew0KICAgICAgcmV0dXJuIGZ1bmN0aW9uKGVsKSB7 DQogICAgICAgIGVsLmlubmVySFRNTCA9IG1zZzsNCiAgICAgIH0NCiAgICB9DQogIDwvc2NyaXB0 Pg0KPC9ib2R5Pg0KPC9odG1sPg== 214549 2013-10-18 01:31:04 -0700 2013-10-18 01:31:04 -0700 Test-case shows that up scaling is not an issue rem-height-workaround.html text/html 3331 jon.ronnenberg PCFET0NUWVBFIGh0bWw+DQo8aHRtbD4NCjxoZWFkPg0KICA8bWV0YSBjaGFyc2V0PSJ1dGYtOCI+ DQogIDxtZXRhIG5hbWU9InZpZXdwb3J0IiBjb250ZW50PSJ3aWR0aD1kZXZpY2Utd2lkdGgiPg0K ICA8dGl0bGU+cmVtIGhlaWdodCBidWc8L3RpdGxlPg0KICAgIDxzY3JpcHQ+DQogICAgICAvKiBm b250IHNpemUgKi8NCiAgICAgIGZ1bmN0aW9uIHNldEZvbnRTaXplKCkgew0KICAgICAgICBodG1s LnN0eWxlLmZvbnRTaXplID0gY2FsY0ZvbnRTaXplKCkgKyAnJSc7DQogICAgICB9DQogICAgICBm dW5jdGlvbiBjYWxjRm9udFNpemUoY29udGV4dCkgew0KICAgICAgICAvL3JldHVybiBNYXRoLnJv dW5kKDAuNjI1ICogaHRtbC5jbGllbnRXaWR0aC8xMDI0ICogMTAwKTsNCiAgICAgICAgcmV0dXJu IE1hdGgucm91bmQoMC42MjUgKiBodG1sLmNsaWVudFdpZHRoLzEwMi40ICogMTAwKTsNCiAgICAg IH0NCiAgICAgIHZhciBodG1sID0gZG9jdW1lbnQuZG9jdW1lbnRFbGVtZW50ICAgICAgDQogICAg ICB3aW5kb3cuYWRkRXZlbnRMaXN0ZW5lcigicmVzaXplIiwgc2V0Rm9udFNpemUpOw0KICAgICAg c2V0Rm9udFNpemUoKTsNCiAgICA8L3NjcmlwdD4NCiAgPHN0eWxlPg0KICAgIGh0bWwgew0KICAg ICAgLyogbW9zdCBjb21tb24gZm9udCBub3JtYWxpemF0aW9uIG9uIHRoZSB3ZWIgdG9kYXkNCiAg ICAgICAqIHRoaXMgaGFzIG5vIGVmZmVjdCBvbiB0aGUgdGVzdCBjYXNlDQogICAgICAgKi8NCiAg ICAgIGZvbnQtc2l6ZTogNjIuNSU7DQogICAgfQ0KICAgIGJvZHkgew0KICAgICAgbWFyZ2luOiAw Ow0KICAgICAgcGFkZGluZzogMDsNCiAgICB9DQogICAgYXJ0aWNsZSB7DQogICAgICBiYWNrZ3Jv dW5kLWNvbG9yOiBsaWdodGJsdWU7DQogICAgICBvcGFjaXR5OiAuNTsNCiAgICAgIGhlaWdodDo1 cmVtOw0KICAgICAgd2lkdGg6MTByZW07DQogICAgfQ0KICAgIHNlY3Rpb24gew0KICAgICAgYmFj a2dyb3VuZC1jb2xvcjogZGFya2dyZWVuOw0KICAgICAgb3BhY2l0eTogLjc7DQogICAgICBoZWln aHQ6IDRyZW07DQogICAgICB3aWR0aDogNy40cmVtOyAgICAgIA0KICAgIH0NCiAgICBoMSwgcCB7 DQogICAgICBmb250LXNpemU6IC4zNXJlbTsNCiAgICB9DQogIDwvc3R5bGU+DQo8L2hlYWQ+DQo8 Ym9keT4NCiAgPGgxPndlYmtpdCBmYWlscyB3aXRoIGZvbnQtc2l6ZSAmbHQ7PSA1NiU8L2gxPg0K ICA8cD4NCiAgICBJZiB0aGlzIGlzIHRoZSBpbml0aWFsIGNhbGN1bGF0aW9uLCB0aGUgcHJlc2Vu Y2Ugb2YgdGhlIHZlcnRpY2FsIHNjcm9sbCBiYXIgaXMgbm90IHRha2VuIGludG8gYWNjb3VudC4N CiAgICBUcmlnZ2VyIGEgcmVzaXplIGV2ZW50IChlLmkuIHJlc2l6ZSB0aGUgd2luZG93IG9yIGNo YW5nZSBvcmllbnRhdGlvbiksIHRvIHRoZSBnZXQgdGhlIGNvcnJlY3QgdmFsdWVzLg0KICA8L3A+ DQogIDxwPlRoZSB3b3JrIGFyb3VuZCBpcyB0byBzY2FsZSB1cCBpbnN0ZWFkLCBsaWtlIGhlcmUu IEkgaGF2ZW4ndCBmb3VuZCBhbnkgdXBwZXIgbGltaXQgYnV0IEkgYXNzdW1lIHRoZSB1c3VhbCB1 cHBlciBib3VuZGFyeSBmb3IgZWNtYXNjcmlwdCBudW1iZXIgYXBwbGllcy48L3A+DQogIDxhcnRp Y2xlPg0KICAgIDxwPkluIGEgMTAyNHB4IHdpZGUgdmlldyBwb3J0IHRoaXMgc2hvdWxkIGJlIGV4 YWN0bHkgPHN0cm9uZz41MDBweDwvc3Ryb25nPiBoaWdoLjwvcD4NCiAgICA8cD5UaGUgdmlldyBw b3J0IGlzIDxzdHJvbmcgY2xhc3M9InZpZXdwb3J0Ij48L3N0cm9uZz5weCB3aWRlLiBBbmQgdGhp cyBlbGVtZW50IHNob3VsZCBiZSA8c3Ryb25nIGlkPSJhcnRDYWxjSGVpZ2h0Ij48L3N0cm9uZz4g aGlnaA0KICAgIGJ1dCBpcyA8c3Ryb25nIGlkPSJhcnRSZWFsSGVpZ2h0Ij48L3N0cm9uZz4gaGln aC48L3A+DQogICAgPHNlY3Rpb24+DQogICAgICA8cD5JbiBhIDEwMjRweCB3aWRlIHZpZXcgcG9y dCB0aGlzIHNob3VsZCBiZSBleGFjdGx5IDxzdHJvbmc+NDAwcHg8L3N0cm9uZz4gaGlnaC48L3A+ DQogICAgICA8cD5UaGUgdmlldyBwb3J0IGlzIDxzdHJvbmcgY2xhc3M9InZpZXdwb3J0Ij48L3N0 cm9uZz5weCB3aWRlLiBBbmQgdGhpcyBlbGVtZW50IHNob3VsZCBiZSA8c3Ryb25nIGlkPSJzZWND YWxjSGVpZ2h0Ij48L3N0cm9uZz4gaGlnaA0KICAgICAgYnV0IGlzIDxzdHJvbmcgaWQ9InNlY1Jl YWxIZWlnaHQiPjwvc3Ryb25nPiBoaWdoLjwvcD4NCiAgICA8L3NlY3Rpb24+DQogIDwvYXJ0aWNs ZT4NCiAgPHNjcmlwdD4NCiAgICB2YXIgdnAgPSBkb2N1bWVudC5nZXRFbGVtZW50c0J5Q2xhc3NO YW1lKCJ2aWV3cG9ydCIpLA0KICAgICAgICBhcnRDSCA9IGRvY3VtZW50LmdldEVsZW1lbnRCeUlk KCJhcnRDYWxjSGVpZ2h0IiksDQogICAgICAgIGFydFJIID0gZG9jdW1lbnQuZ2V0RWxlbWVudEJ5 SWQoImFydFJlYWxIZWlnaHQiKSwNCiAgICAgICAgYXJ0ID0gZG9jdW1lbnQucXVlcnlTZWxlY3Rv cigiYXJ0aWNsZSIpLA0KICAgICAgICBzZWNDSCA9IGRvY3VtZW50LmdldEVsZW1lbnRCeUlkKCJz ZWNDYWxjSGVpZ2h0IiksDQogICAgICAgIHNlY1JIID0gZG9jdW1lbnQuZ2V0RWxlbWVudEJ5SWQo InNlY1JlYWxIZWlnaHQiKSwNCiAgICAgICAgc2VjID0gZG9jdW1lbnQucXVlcnlTZWxlY3Rvcigi c2VjdGlvbiIpOw0KDQogICAgd2luZG93LmFkZEV2ZW50TGlzdGVuZXIoInJlc2l6ZSIsIHVwZGF0 ZVRleHQpOw0KICAgIHVwZGF0ZVRleHQoKTsNCiAgICBmdW5jdGlvbiB1cGRhdGVUZXh0KCkgew0K ICAgICAgdmFyIGZhY3RvciA9IChjYWxjRm9udFNpemUoKS8xMDApLzAuNjI1Ow0KICAgICAgW10u Zm9yRWFjaC5jYWxsKHZwLCB3cml0ZShkb2N1bWVudC5kb2N1bWVudEVsZW1lbnQuY2xpZW50V2lk dGgpKTsNCiAgICAgIC8vIGFydGljbGUgb3JpZ2luYWxlIDEwMDBweCo1MDBweA0KICAgICAgYXJ0 Q0guaW5uZXJIVE1MID0gTWF0aC5yb3VuZChmYWN0b3IgKiA1MCkgKyAicHgiOw0KICAgICAgYXJ0 UkguaW5uZXJIVE1MID0gYXJ0LmNsaWVudEhlaWdodCArICJweCI7DQogICAgICAvLyBzZWN0aW9u IG9yaWdpbmFsZSA3NDBweCo0MDBweA0KICAgICAgc2VjQ0guaW5uZXJIVE1MID0gTWF0aC5yb3Vu ZChmYWN0b3IgKiA0MCkgKyAicHgiOw0KICAgICAgc2VjUkguaW5uZXJIVE1MID0gc2VjLmNsaWVu dEhlaWdodCArICJweCI7DQogICAgICAvL2NvbnNvbGUubG9nKCJmYWN0b3I6IiArIGZhY3Rvcik7 DQogICAgfQ0KICAgIGZ1bmN0aW9uIHdyaXRlKG1zZykgew0KICAgICAgcmV0dXJuIGZ1bmN0aW9u KGVsKSB7DQogICAgICAgIGVsLmlubmVySFRNTCA9IG1zZzsNCiAgICAgIH0NCiAgICB9DQogIDwv c2NyaXB0Pg0KPC9ib2R5Pg0KPC9odG1sPg== 214550 2013-10-18 01:32:38 -0700 2013-10-18 01:32:38 -0700 A real example of the issue real-example-offline.zip application/x-zip-compressed 2438 jon.ronnenberg UEsDBBQAAAAIALegT0O/n5OwOQQAANMLAAAaAAAAcmVtLWhlaWdodC13b3JrYXJvdW5kLmh0bWy9 VtuO0zAQfUfiH4ZIQLd0kxZaQG1aicsikLgJVkII8eAmbmOtEwfH6QXEvzN2nDjZLmUREquVUo/n eM5c7Jnw1vN3z84/vz+DRKV8cfNG2HwpifELEKZUEYgSIguq5l6pVqePvdZORlI69zaMbnMhlQeR yBTNUHPLYpXMY7phET01CwtTTHG6kDSFhLJ1omBZrsPASI0CqhSRZLnSK/MX9GGFx0LBvlPoB7V4 VWaRYiIDZPYC9z/idu8EftT7YLzyC7Xn1F9ZBZhDRHjU0r8Hd2/fndWgn5dP76ob93aqYyUIJFWl zOANUYkvRZnFvaH/8P4E+hWDiDMMyScdg2A0vD9G+Wg4PJm5I/7uAL8+4ZD1hkgDQTdjEZUpwvz6 xxmn+lMp1oAty2Kx9Ukcn21w8zUrMH1U9jxJdbi9QTu6jcFOyGtpGLTyFmLY64RWjH600pmKQmGp pKnIqtRmQqaEs+/EhBz/VUJhS5egREz2mHQH1vqnmtsUHt73J7dnZtMFYSnivTOWErlm2RSGDfUc fWXZ2oocjkjFIk4ddEmii7XJxmkkuJBT4Lpel7ykzWEiJxFT+yn4k1pmy3o6wRJvZOYCTEdDK3NW C2qq7JjVmEiUUJpdYfXRZaswPjQLj3wjrWQd88loALkz3oqt/2ByiWwY1DkNA/s+hDrYVb6T0QLz dcEUrAjjBZpWiTsP7nA1m8Pk4W3EjipEvjjXSRbyAohxGFiB+YYiIpiGMgeWFQrtDDDuFxQdlNSH V5AQLNS7aMZASLZH1ZxKVEqZfkwU6pCiwJI3RVQWJeFWZakhRO4RK4FGKakKFrIyXeI2yXO8aIUf BnnF0JaEWRjCrzIgoK9wvtOxpaAfPtAvH9pC9kUiSh7DkgLdkUjxvb4HUmTrxWQ4zHdhYFeQYLIa O00s3GmsQGRO8PXh6EvrgV3gGShfWAI+PMniyjatrrfjYE9g8dxDR57hM/ZS10hzRsXCXpqytukQ HyjhXcQhb1u+ZvnPMRofidH/iBI6cyxKLk5dzHXiFLhA6ZUrrHa3qx/wTd5+vte0frmLp/tn2tG3 2HN7ztmTgWskOtEvr0Y/3b+Ke51KOIB++BO0dvUQ2gZ+K6ncf6QcXRbYSayzHYQO9TFj3Vxchh7l WafkaugRnjZFnu5nN29crzmWeUwUPceRoO6CTmIboxslOnvwo92zV0RzQHK97nQS6DYf6GHA9Zcv X318v85IlPiozHubfABbyRTt/a7ltweIE9fEg6BpekIybJP63UWDeAv75r1CRVdVPsvQ95fnb17D vD2pWOp9mAz1MOXlO2/WAn7oAFFSs6ny0yAcJ5uHFqdHY01p3KKkla5BaewotYBdSlryB0o49RUC h0gu1j2vOnzqoVr1sw7oz262bU7SYm1z7Qa9RqdHeWeUpLzDDLHtEc9+OpNWGOgWbH6Y2f0XUEsD BBQAAAAIADRPUUP45rDwBQQAAGILAAAPAAAAcmVtLWhlaWdodC5odG1svVbpjtMwEP6PxDsMkYBs 2SYttIC6aSSORSBxCVZCCPHDTdzEWicOjtODFe/O2Dmc7C5lJSSqSonH8818c8Tj4M7LDy/Ovn48 hVRlPLx9K+ielMT4BAgyqghEKZElVUunUuvxUycEsHs5yejS2TC6LYRUDkQiVzRH3S2LVbqM6YZF dGwWTm1SMcVpKGkGKWVJqmBVJYFvpEYBVcpIskLplfn5I1ijWSjZTwojvxWvqzxSTOSA3F7h/mfc do/got0HE5dXqj2n3rpRgCVEhEc9/Qdw/+79kxb067L1oboJb6cGXiRVlczhHVGpJ0WVx+7Ee/xw DqPaf8QZJuSLzoA/nTycoXw6mRxd9bgh0iCQYiyiKkOU176ccqofHWjL8lhsPRLHpxuUv2UlZp1K 15FUZ8k57iel8zXIVCsN/F66A8xWW4eazEWvCpkoFVY4y0ReVyQXMiOc/SQmU/hXKYUtXYESMdlj rSxY6481twU8fujN756YTRv/SsR76ywjMmH5AiYd9QJjZXnSiCyOSMUiTi10RaLzxNRhHAku5AK4 brMVr2hnTBQkYmq/AG/eyppuXMwn2Jqd0DTuYjpphdZvSU17HPIbE4kSSvNr/D657Bdm1ziGJzMt rEUD9+n0uLC+e8l95M0vcQ38tqiB33zZgc52XfB0GmLBzpmCNWG8RMcqtfbgHlcnS5g/vovVrRuC 1q2Itqa1hSL8TCkEBFJJ10sHvY/rmMZbIc+JSYmnoU74573AJyG6lUBA74ybrcAvaidNmc3C+HyT o6r+noqdThYFfQaBPoSwCVkJZSoqHsOKAt2RSPG97m0p8iScTybFLvCbFaRIR/uxps/SvjVWIrIg eBBwUpa9sy5EGygPGwIePMtj47tNkeXQWGDx0sFAXuCJ8lpnobNRs2g+hKr1aRGfKOFDxFXeTUOa 5T/naHYgR/8jSxjMoSzZPA0xN8mTbxOlV7ax+oOnPY83Rf80Tmh7EJfP9y90oO9x/Lk22KNjOxZ0 oV9fj36+fxO7g064Av30N2gb6lVoH/ijonL/mXIMWeB0aIIdIHSqDzkb1uIy9CDPtiTXQw/wbErk 6Bl1+9bNBl5VxETRM5zO7WSzkmbY2ak+2IOL/gheE80BybnDi4Kvp7avR7udGd++e3hmnZIo9VCZ u5viGLaSKer+aYL3rwNHdjD7fjfIhGQ4+gi+oUP8CkfmvEJF21UeyzH212fv3sKyf+9oqI8AIfpi 4xQ756SH/DRAoqSlUxeoQ1hSTSF6pJ7MNKdZj5NWugGnWZ+TRQ45aclfOOEVrBR4o+MicZ3a+sJB tfq1TemvYb2bqmRl0lTbXtw6HZfywb2O8gEzxPbvbM1jcH8KfD1XzYu5Sv8GUEsBAj8AFAAAAAgA t6BPQ7+fk7A5BAAA0wsAABoAJAAAAAAAAAAgAAAAAAAAAHJlbS1oZWlnaHQtd29ya2Fyb3VuZC5o dG1sCgAgAAAAAAABABgAW/nyKtHJzgHn5/Eq0cnOAefn8SrRyc4BUEsBAj8AFAAAAAgANE9RQ/jm sPAFBAAAYgsAAA8AJAAAAAAAAAAgAAAAcQQAAHJlbS1oZWlnaHQuaHRtbAoAIAAAAAAAAQAYAKyL a4wOy84BnGRrjA7LzgFb+fIq0cnOAVBLBQYAAAAAAgACAM0AAACjCAAAAAA=