242740 2022-07-14 02:49:31 -0700 [JSC] ReferenceError when multiple modules are simultaneously importing a module containing a top-level await 2026-04-16 08:53:46 -0700 1 1 1 Unclassified WebKit JavaScriptCore WebKit Nightly Build Unspecified Unspecified RESOLVED FIXED https://bugs.webkit.org/show_bug.cgi?id=270357 BrowserCompat, HasReduction, InRadar P2 Normal --- 1 ardtimmerman+webkit webkit-unassigned ashvayka brian dpaddock eixh3poxg hcampos hello herr.ernst jcoppeard jimmy karlcow karl.horky keith_miller lea marcustyphoon ncpncebij nico.m.hoerter pierre saam sambeolatki strehle.benedikt sviatsv syg vrugtehagel webkit-bug-importer webkit-bugzilla ysuzuki oldest_to_newest 1883582 0 460889 ardtimmerman+webkit 2022-07-14 02:49:31 -0700 Created attachment 460889 Test case reduction -- also includes a README explaining the steps to reproduce the issue When modules are simultaneously fetching an import which requires some setup time due to top-level await, one of the imports will be rejected. This doesn't happen when the modules are fetched sequentially. Note, I'm not aware of what the spec has to say about how this should be implemented, but I did note that this behavior is not consistent with other browsers. When running the attached test case reduction on either chrome or firefox, none of the promises will be rejected. (reproduced this using WebKit 16612(.3.6.1.8)) 1885668 1 webkit-bug-importer 2022-07-21 02:50:20 -0700 <rdar://problem/97370038> 1934879 2 jimmy 2023-02-18 03:38:14 -0800 able to reproduce, i stubble upon this issue today myself... FF/Blink works OK it's hard to really do any workaround without doing a major refactor. It's not like you can polyfill this in anyway. so i wish this could be fixed/prioritized. ...non essential issues that can be solved with polyfills can get a lower priority I would consider this to be more of a bug than a feature request. 1973240 3 agafvv 2023-08-24 02:36:23 -0700 I've encountered into this problem myself and spent a lot of time debugging due to unformatted error messages. https://github.com/sveltejs/kit/issues/7805 - message in the SvelteKit repository, where other developers have encountered this bug. Please prioritize fixing it! 1991621 4 hello 2023-11-10 10:37:18 -0800 Noting that we've encountered a similar issue, in our case using Web Components from Stencil - https://github.com/ionic-team/stencil. Issue only occurs on Safari (as per the test case reduction), and is resolved if we refactor to remove our top-level await. 2004665 5 pierre 2024-01-12 04:57:49 -0800 I've encountered "Unhandled Promise Rejection: ReferenceError: Cannot access uninitialized variable" while developing with Astro and can confirm that WebKit-based browsers handle things differently than other browsers. 2012064 6 ashvayka 2024-02-08 19:26:54 -0800 Pull request: https://github.com/WebKit/WebKit/pull/24122 2035714 7 sambeolatki 2024-05-15 23:49:36 -0700 (In reply to Jack Koppa from comment #4) > Noting that we've encountered a similar issue, in our case using Web > Components from Stencil - https://github.com/ionic-team/stencil https://dinosaurgame3d.com. Issue only > occurs on Safari (as per the test case reduction), and is resolved if we > refactor to remove our top-level await. How can the issue that occurs specifically on Safari when using Web Components from Stencil be resolved by refactoring the code to remove the top-level await? 2039302 8 vrugtehagel 2024-06-02 00:50:53 -0700 This bit me today - it actually happens in a much simpler use-case than the one presented in this issue as well. I have a reusable async function that imports a module, and therefore is subject to importing that module multiple times. When it is called twice (or more) before the module has loaded, some of the imports resolve immediately. In a nutshell: ```js // index.js import('./mod.js').then(() => console.log('one')); import('./mod.js').then(() => console.log('two')); // mod.js await new Promise(resolve => setTimeout(resolve, 1000)); ``` Here, I expect to see the "one" and "two" logged after a second (not necessarily in that order), but instead, Safari instantly logs "two" and then logs "one" after a second. Adding more `import()` statements to this causes all but the first import to resolve immediately. 2047324 9 mark.lam 2024-07-19 21:42:43 -0700 *** Bug 270687 has been marked as a duplicate of this bug. *** 2098180 10 hcampos 2025-02-26 01:06:13 -0800 At Esri, we stumbled upon this issue within our ArcGIS Maps SDK. We use top-level await to dynamically load different modules based on the environment. Somewhat similar to what Sveltekit stumbled upon in https://github.com/sveltejs/kit/pull/11601 Another repro we put together, similar to vrugtehagel@pm.me above: ```js // utils.js await new Promise((resolve) => setTimeout(resolve, 100)); export var utils = "utils"; // common.js const { utils } = await import("./utils.js"); export var common = utils; console.log("common", { utils }); // entry.js const [{ utils }, { common }] = await Promise.all([import("./utils.js"), import("./common.js")]); console.log("a", { common, utils }); ``` Works as expected in Chrome, Edge and Firefox. Fails in Safari, with one of the imports resulting in undefined. 2132007 11 kirk 2025-07-22 21:16:22 -0700 Pull request: https://github.com/WebKit/WebKit/pull/48409 2154439 12 karl.horky 2025-10-27 06:56:05 -0700 Since Jake Archibald talked about this on Twitter (https://x.com/jaffathecake/status/1982542211712712878) I wanted to also add the quick research that I did into how top-level await code which is bundled with bundlers webpack and Vite breaks with with Safari (webpack works out of the box, Vite breaks out of the box, Vite can be configured to work with `rollupOptions.output.inlineDynamicImports= true`) https://github.com/karlhorky/repro-webpack-and-vite-top-level-await-safari-bundling-bug 2155363 13 syg 2025-10-30 15:15:00 -0700 *** Bug 301711 has been marked as a duplicate of this bug. *** 2155441 14 ncpncebij 2025-10-30 22:41:50 -0700 should rename this browser to safarIE 2177779 15 k_tamkun 2026-02-03 17:58:01 -0800 Pull request: https://github.com/WebKit/WebKit/pull/57827 2200367 16 ews-feeder 2026-04-14 14:41:33 -0700 Committed 311236@main (4a638109b905): <https://commits.webkit.org/311236@main> Reviewed commits have been landed. Closing PR #57827 and removing active labels. 460889 2022-07-14 02:49:31 -0700 2022-07-14 02:49:31 -0700 Test case reduction -- also includes a README explaining the steps to reproduce the issue safari-top-level-await-bug.zip application/zip 13234 ardtimmerman+webkit UEsDBAoAAAAAAFFd7lQAAAAAAAAAAAAAAAAbABwAc2FmYXJpLXRvcC1sZXZlbC1hd2FpdC1idWcv VVQJAAMK5c9i3uTPYnV4CwABBOgDAAAEZAAAAFBLAwQUAAAACABGXe5UQlijkIsAAADJAAAAJQAc AHNhZmFyaS10b3AtbGV2ZWwtYXdhaXQtYnVnL2luZGV4Lmh0bWxVVAkAA/Tkz2Li5M9idXgLAAEE 6AMAAARkAAAANY67DsIwDEV3vsJkb7MyJFl4rDCUgTFtLFKUKFHigvr3lLpM99jHlq7an67H7nE7 g6cYzE5xACiP1v1gwYhkYfC2VCQt7t2lOQiQm6SRApou5SbgGwPYjx0J+umpJCs+q0MZMwHNGbWI yU0BBdQyaNFKHuuWjW1fVRgl+WUtI/9tVJ/czCumRa2Nv1BLAwQUAAAACAA7Xe5UGeyuE6AmAAAX 2wAALAAcAHNhZmFyaS10b3AtbGV2ZWwtYXdhaXQtYnVnL3BhY2thZ2UtbG9jay5qc29uVVQJAAPi 5M9i4uTPYnV4CwABBOgDAAAEZAAAANRaWZOrSnJ+96840a8amk0CNDHXNloQWhBCCBBy+DrYF7Gv QjN3frsB9SL16YXWPbbDD92iiqqPqi+zMrOy6u//8uPHQ6D4xsNffzw4fhQmGeCHeu4ZQOlkNpCF EeAZheEBSqk42cNfmvaFkaROGDRd4EfoEbrUeqF2NB3PEF/eIm19YsS5kxhpXZEludHWRYp2VKy2 7u91ua55efr+eNo+746pfeM5mhGkLeCcH7/W60ZkBLoRaM7VONo3qZEUbfvf4X6L9fTqj/b3jwvC QxDqxn9dxpaC/342nAxMNdvwlWu063Ehjwh8PbCak9ArDL15Z2dZlP4VBBPDctIsqR6DyHfTxzCx bqFB4PkJuMA9Ztb5FdIJMsNKnKxqMFNbGcAIAC760mgXSRnXd7I1nKioEhIrc+z5ouafkYjFQlnF Ob1v4yIfC/ZcXcyLpDqeh9MSlhgLMygpXeopQ2uDfKNSp9gbR9Zvvz18zIaiaUaUfcQD/Ig+Et+m 4QmzJuDpCWhxvp7/RiYze6cgPiIuT3nBT9DJZgzOqp7QS1kSXpH5jvCP/jYvyt6aHPOKsZ9yvTWt iRSesGsPizXlJKJCbvilUh57eI9frMp6/h00yXd8A8iqqK1/+CfyWI+5/9Kxoc2wwsxRsjBpGkCP 2CP6om0v+EZgOcFb6Ibwps+//vaj7tZBRRW3+EAgxCN8j2LWgI003AK4AHwtinJmRYk+LhjFG0Oj c7YvnGyC2GOoL5smWWSpEAzRhTrbyysyGwts/6TDuD8eH7iTwM4zZ1QeigWvSSGxnhccw+Fworeq 2EEUppJmgG4YEVDbI8VrlzdaywO+FoebhgFwWWBAligNR0+G4MamvFo1wExCH6iJcQKrbYjUDZHr hnniAG4r/t/7j/Wa/Vm8Zh7ol+5Xw22Upulk1VYvV28BvWvBXBo8aqEPplEYpGGSgkYUqkZinI9O 0EUxgtQBFM+xgg/0A63nBH9fPV5gGy15KQAt2te6MmfNshxRA0c7s24ESobcl0+VG5MUw4WH3NQ5 aaLDQ346Paniar2Pi+jML0AGPyh9MB1vfA7qzbZxHpK4OtPma+1klFjHZXuRZ+1x9Mx+khzcyQ1c zfnn+poj4/QBxYP7KW5hnyluC8CgG8VxvuD23o4XZitkRQ95YR8SYpnK/TRUAtsKsdWa94n+FF6N tCmRoixURrqzPVfb4Rk8HBaMx0gFiscsqw4jfeExwnRJcdcUf2G5iD9BqeGH7uecErdxQEdOr3Br Uq9KQIv3NasM78pnTVqzJGSUJB2dIeYUyVRRWljlVvB6hp4yI4zPWL8vjkNwu7GCBAZns6znaKiz 2J05YpqjO52B+7xSBqLgz6YY+bm//YCjN3r8HkmNTUK/TdI1cBOSXBWBFrGDL1hW3JaL3AWUziMM mSb8QZ+lLrOQUiUahNu1Y9N2js3whciCQ3c+woTKKFa9bJ+zcRJYhAtqp+VOqiShLPhd4jiHM9jR F9zqzsPvxFvr7qSAmXveZUJaTSkQhfUknpzGm8bN7COgkUH7HmuX8ned+J9ZCjcDeE/I2F3W5RX2 ScSXAoB1sy4ySogbntbiI5WMIyrghkXO7/04z4tBuGeX0czYYcosSdAQnWlD0hMVbDEyBZ4dB+pp NjtQPRAKwzm+TDa5wJ8zQRcGZDcB39jbh98H/7Mi+cwI3Uf9R4a9I/XBgCEGA3OpInxqMnOHokIx IUf0YkxwGV2Kp55PSxuU06bxiJYdI3AHOs1brnrKxpA0PezlrF9qE+yccFNvsrGp/GCZ5DcMO9w1 zPk4mNFsxTtezf3fnqKb3+CuYkmzyjM+2oRgj3eFvK+4z4K5lIAW72vJiGq8TsPzYMDLM+2s5Hke QQwlVPs1Vzh8RO1ceMtl1NnnVkeYiFWRj3YUNWN2yBBN/InC5cvxDpsHvppPKyEw+v4uP33H5f5a yVwm/z3RJNpHvqiOju+RSQ3YCKP+AVqEr6XAmuAWKuOIQDWLDs/mXF6pIyYwXHwogLw4Y9kqH6Gi SBXwesuAG55iljByGi7FzJHPi9gPhIG0ReIIOgwSWsVTaRbcBj6vlP7HC6Wv5H4W33eL8E0jCdP0 4aXXH3/5/DORkiVGzfJn3ynL8vGpXfux735Dq8eWe1kz7c8+c4G9eJg8ajI8V594evrPz/TH+kB9 Bjdbr+7qY7XaY7VRM9Iln+BWFLtacsNqgC8KDuNWIaGT1jpOS5tg4C2zkHVOz3fYSZV4a4liG/AA Fr2IIIiIwYYuw5g8sdcJCpvDRw6cD2lh9nk+RVU8JdAMHfCV7MPVA981/VvomonbCgDuRgoa8obA QjtGxHB7DZO+uk+X/UqJBTxzHdpTT9vJgY1oaCkNe2Nj31e3pLKHyOCUQdkJYZJoG0klp2zKuTfn pwtbFgab8nNSwpPx0Y4Vvyv0bxEbCppfAO8W7rsgqI/EvFK0hVpIvaw4DOmchlDz0Mc4TVkWhX1w tOMhzkYIDk9OC5xfb/n1KRFB/bCHIGLg4/ac3x4k9Sx5i+IEJrUr/kaY87xzv0SmN+kMTfENT1Oe chj428C1NeRXAdL1K88BGhrSj0Pemw3yoHZ/N2mPxigAppFmT1kRGL1FqLvWLwGv9lVPO+w3IygT 5SqoJu6K4OB63479aX/3YnRTJ9AbVUrsPO3g6dRE0QzAOEXNLMKPVBVuclDfjxDfgDdae1sDXIC/ VmBnnG/oCWUl9B6nkWI6BwenetOKn3l6BEG7KLdi25cz0WdYbAAT/jjfMiErU54+HXmQBRN4boZn XaEtaanJlIrB8czpGKf/ZNjeybTVbkVTsrpFdEmRXivCJ+RX2YexH3qfdWgQG56b3zaL1cE6RIzN mtRk4xTDbOC68/2RDlPJP/IVpxapNRXXxzM03dJqvrKRHYi7fSaOo/MAQSmDYESZGC7HGG+DnMkj KdPjXYuaaOU34r0f0GOXfcy1pfh1BvUFtabt5bmrYV3swnnB+mcssGepYJMyr6khomObalCGrrs2 w1OoIf5UXAn6AoesxRKxAr1HjmR4p6GpyFRLnJtssjW0A/e6N+ak+a76Vqj8f2o9no3zr8sWXuJ2 4PLbNUdIhRAusfIsZsyxRLNnfs9myPg0UcdYeuRBlx1WwVTzI3LNhJtk0pPgBF5i4gl31n6P5Pzp ToD3CZL1AiI4H8Vhhp1stLsC/w7XrqNWmB//+MeP5jQORpunWjbYzcHcn9vKtP+/tYm5do+/zsK8 oDZCen7uamlA7zyLptIKJDbmHBr5I3azLSNIA0tqvYe3usCsUHcJbjV+RA53hH+Y6FwUy6NlNqZ2 rMmp5SaWpmzkxlKPCpRKg4bY0PrOcvnz4rh7rXhOpIZKole/Wh5PsBeBPBW6SoTPe7kwSBPYPQoV nM620WrpLpNiyop7i7ccfh6UChgb8wmmLCC8snuDtaShNsKZLu2OAnPfX9kZpeWktBTSBZr0TFft GhkmT+70dk/dSPFkaMpLzHYTcjkpUKbeVbfvxVvNWkWg67X6CP/S1Xq3eoR+VDdPHdX7yL81aQP4 +4fS18iNjlwVgQvk12pCUmiCbwa6dFoRzKma77aMx265tcKSyPxIUsou6R/xkZD77ibbRrPJ7hB5 0EaF92MWWxiT8UiPYAQbW+kwZU5xHSLI1bajmrTn07r6FCrAj/1atX/87cc7OaJfcwb9QtAn4TD+ 2L9fDm0kfFUCWrwOuziFn0/Om2F0EPn+weR6u3OROZKtR5Q9RbYT+kxwi2O0H/LzaBUThJgb1MBd lFjWw8YCp5EhWT+eiCGzSiQyjszJlHC4jmv15bbEwz+bOw2D6zX5HMc+/LQHe6PVze0CqAlYrpro hppbF/XGHofXb2pmbEPRa96fPvvm1DpVTANQc9M0kkvA8WZ7VyhJ9dQTfu9Mu0NU2ulA9Sb4f09f oPtCoRfYVl2eCwDULSgC+aTo62lZceuRGdqz6DwMMVlFZ2eUT4RJPKIHyS63Z1tcsTNPlZeBeCoR dUwxW89VSHy6p5X9gShTWg91qlCP9tKPP08B1YPMjCADdKe2gqmTfbyCoMfBHXmgd/Av1LytBVr8 rzk6bmfbQ4mqK4+a85ORVe54Et4xI2qNLbCZNDFy1Z9MUZLqob1qvyLsnVQQySA5rtRY3sv9rbs5 gtMzHYy9OXrmTFijfPo7ZxBdTVKTjQTSSCk/SyZ9/4j0Crch8rXU7n86HJA628lmsRQ24nQ9wXV6 wxLJUWWDTVRNKkubUJnkRoa0HlsVIW2QZK+dahZ36XFrcB7GJiNkhveERaJYKRzDAyrXNXzW9QJE pGQ2cDSqh+dbMrfZH9tQlcACagvkK4H+8Hzv5TbBYzsvIcg9uZsfXbatzwbufZd+bfO6iq1FrAXW /gItxteiUsf41Ev0hbIINiq56cHTMJOL+JCqqKF5g9LBRmb9t9gJ2josI0wr0siC3P052O6WE9cH pzipW9RINMiNyKy15SylmY4pFD99jl+62NPLBahTvaj1D80HdkeofIXbkvdSAlq8rylcsfRpziob Wae9xTYbc+ZkLh4yswRzkyF6iehi/eF2Tp81HyxmJbfdUxnGbuI5JaU+sp8yyXo+zaQt1k9leDVd TpCCZL9jLrre9DSUNFNSRwk+u1UB3XWSdQvd3D65qQCgbqdbc4LYyQdJG+4dmd5yfRDVBonrmtbR tVeVhHj5bG4bs+lWXXMYK+PVrqqXMx9N9geFqHBR2Axlf6LJSs+o4r6m4M4qlj6/gPL1ZZxhe/ft 113GafE6qBVMTJRFuueFI9LrI1FB+Ku0Dtx2iB1TR2radxG4ZH1LFeL0gNgrHFmngoATA2u43Vsh mkKHtSV6fWTn0BF6yNgCFKvPPfXzBuv9xA18R7TSIjbzb36BFuPrmRM5Hx0OoUYetiNyM5/O1wsX XYVDupo5medpZw3Bp/ZAlkPXZSgmICqamTCKQE+RRWzG0GAClrlWzvuLmbCt9hlcaHbVcVtx61Mv Jw/otXOwjKy5NGko/sPzlZkb52HntWMBUscKFC99PkH46YrOFcJP7sc3Esv4tEXNL5DkAdD4uveP IMLAyJzLVfSfDzgug6utnPN6LehmjpcbM6bTtAqM8uWkA3l/L/y/kF252jVfdOlbOa+fL9C+n2mB 7wiW3mDXuv6mBmhxv9Z6E425YchVQxch7ZEB0kNxPITLlb8cjccCaE3YYLkm5dnATksWH62WLjSk tQEtr89DbTbv9U5eNBnPrYniZGIKoeRuS/S5T9d7O9Ka9VqRkrTdLn10zHMnL6/Yz7y81rSnPB14 GbjsWNyTE3mdb49L6rxeQPp4bJTpwZFDSK93t2vZc49sRI2xBBH67MhfZoUP7tLc7pewLOuTeKKi sELsRHu0QNjFUo877m+jPKi0lxxRPVqkg7bdWIhfd7HqFbbm8rXQ9WJVlmKS00Nc1OXc2MGhckB6 a2JCBUe+HPeY2D9Nz/qUXI2QeH+QRXQPqvB4Z1qb1WyxZgxSOukbUyAoFvZTdC2YCj3eyP9Pkqxv bfL7Mfc9t6pukGu53JQBpNvNqlGfog5YDO4Q17aPqXVUpxJKjwppbk4kYpAGR85SygzCeXwx4Pa7 IzY6rsUeVJHGfzN3Jc2JK132r7x4y1b4A6GRJRqQhAChCSEWHaF5nmcW/dsb8FDGZbsE7RevN2VJ rrrYJ1OZee8599Sx3CBivlnYepxZTBaH6bRaTTQxWt7DtIHTK1MxAskg/Uqo+VhrzDneGbXzn2Nb Ylb7ydB4+1lz0pBgI8nBcTFQtWgDaV+raeis66lhbe1ujkf8umwbaT8Q+LSuJFqgsaFl9p4R+0dU 2h7IhGYjlSf43vlerXHeru3Mir5cHy+nuftf57eol9//9fqqAhtDbQHBjOCrZcmiaGCoS2vaZRRe rQf8vDBEndLVAhG2FYKSPUx6K8rhRadwT3i30APZxFdCXmo+yyI5hu3MRCfbVm0ON2uiGaS3s+U9 CheNxX/C6u6TAP6vrQBfq6J/jnP54jOex/ez74xlY05DkiBAA1RkJWsDNS+2hhPuIWrZ+iRvxbwd nHRjpTbsAoekdWhiqdwwqW2TLNNO/QjgdYzYLOJKc3hYdTyU3qMH8Z6le0w94/w7Xlsez3uS5Rtf 0xXwo9jeRn9G9fbZEzwOz7maBXnW64kcoYcBE5ee1M5YajCGKvFkBkOUpVihMgDl1AY7+ryynJhV dDyvuGq8INgTH273/i7UybXrHFXaA7tDH4zH8/8t2fQ+L/mSaXpk6N6OLW/XzxTTn4fKX2bBKlBi FEpTIKpqliFqp9NkPhXX+WCCCMKzvrMAY2jetLUwTzUuy8EcauxiZ6n2pDeX4H69EeAUEzg0xyQz 9ajHpv6/MBrPlObP6Y2fQz6Pw/lirObY5U+L0piWzEpViY4hvG3EHhslO1opoNQHNndIL4k2q81G 78FKpHSDlIdwRcikMUtZkGISrIJSMCPSzM/2CmFssXU3spx7s+08mICOW77OaeU38/4RsM8Rr1if v17n+wioJbbfrPN5nPIsw0qpNxeXvkRuTH1h7fMUnVeJzFgF5EJoH3r7vab4IVwUa1NZxwUmVyFO AC4ngnRD0gw3i/lqEB3u++PNF42jX+l07wfisw844/LZ46tmdwRM2w0+2c1T6BAefO6YgvHaj5bG glQEVRIPoebvYWIxkVKhxfvWK+oiXxzmnAAnktjLPSjFGVwXJ6cwp5UQl40hDJDdfF8b+1CV+bl5 8j7wGZb3t2NnjWG2k0Kwmp0bQaq0c6lTclTBNZ80HR4pwDTl2JYveW9ZdoGbxRM70yx7AR+LShNx elsu+chGsY2bZ5N11pqcWpk19P2s+UW/fz5RkEfWp5egFxSer56eA/0ZgWqnws0eO6dBba2t+q7r WWq6awJ/vw1oTBkWpojIWk+RxDxR9VbwykwUOl0UBX63wknOpAGQVvaxkPGgSs5SFfL09p59YiRl d9NV/1UaCiGPIXeN+wre9ebpOdyf8TtSOmDuEuSkKEtAz8lyoc5Tfqp6HKnvlKlYK+D2qC1buAIA ZRtZ3t5Qph4KwNKs4eBNKIanoWIJ8NQ32hqZ+ifDOQT3OA08Kzle5s29C/x48APryf2KLH2sAvAa 9Bn36+XYvF8oTCFCMlp0FkdcO2iDHTdzdhWe5treprnWSDakdUiaQtVDJUrZSXV+YVtfhFqv45eS BLJ52G4JXoSw1Fx5+okpGOueSTsOtDT4rnfjksPcr3p5i3qF7eX6mg+NULwMK3Yv0v5Q7Naq4tVz Aod2hwbVoLKHN23LtrLaKlnm5V3GUIIqAgNVSlNWympgEEibFLAG9g9Qrrk1GmkEYRX+VB0r+/5N GX8pDIL/we6ctP81Fvqg+ipLBc+nN/Qx5M8PX4E/Xz5dI42ovVRhGsFmB+mrQj+Z9kAE22qnsTtX AACVYTB4TvYVej4ONsjBg9PSSQ1U2PeC3peiUnTddJV0NSA5Ll56XpNEyvo0/75CnXy9Tj60116X x2rsvqrkOTrdoZy6opRGKCbIEbfsKjqtmtDtCldQorLjuhDjJE7O02iVKqg8rOAduUuZTbYMyblt +0aM8HtOrxf1HAv6KfI9HXnjqvIVyX1/Nf5X2DMAv26uFPeIGjxAq1XB7NbVBggnJ5v0j2GVHhDQ gw8luxS4Q+em5H7HxGI0iRAvMCKLr5qeFExCalBKTuZZ3k0EtIo1c2V7vriBTeIf2F4/EGNf1R/u T2LfB74g+O72WnUYkcrKMK6djhqGYSd/y5Up5vVCzMnbRRHMj+SkEZepZJsOx/pHn1ygapFHgoIr IIMRrWt7O5gucW8JV6ppyNMAE1vOJbXHRDH/VBJ1I837uUa/X2HP0P+6Gdvglx8XNLBcrbOBU4ba Ls7HGqQCllwebqerHSSclsGkwaUeoPfIRANWisYcGLxfnpyWxhYh4VIF207wdo0XYsX5bFpIUXmn nGsceq+c7Vf8+yPQXWNecbteXTn4MRo4MzdkmTmDYS90hN/tnY3AH+RyV5c4iZNYl9kr06psBAyl NKEA5tjg+nSfqSiVIEe41cop503ALd9oEu9iIWIY+tjewHfntFf2/N7Z+u81v1xftaB6Cs7nA/u7 RPr+wbwNfR7T2wdj3wdK1VYlBKjrHD4dknM6KB+jpQutmEqeT2SuBXRIqufQRA13OZVHMYG4oFPC ArQTOFNt1BW0PJxzamcmYnKpoqdF0UX499niu/XnK+b9/jX5NegrEOfLK9c+Yi3OwsTZTgsbGKZh daLrTIfxcrpzCsTuNnTExaTQoCLiwmDsRlV3aPhaL30vU/J1OnAWm8xg1S96AFGKZMalcrf2afSn 7YOuv1SdPauGvhIRP0Y+3YZ+BfDtwVgaymvmJoUqBmJ2TFmqeNKwp71ALJcSneezQIr2rk+sDL9b oTl66GZTOVSmm36ddlEoBKaYMLLuB2rqzpZrB4uA0FomzPcHwneM/OcvFvwIHC9BL0C8XD5dI41Q 3yf6lo41bGgFTNlDEBnKSSsH9OwcBTL3M+vUTIQTxdub7f4oam5PmrEqrX0XgLFqECUirUN3bdsT dnqOfeIYIDOW30NQGqnn/Emy8Uj55X3gMxTvb68pwogTc7RANDHb7jO4nsfbfhaJ2yXZ8w4RI5xp ylsiBjeTeh11c41MAb/fEgsNEX3cs/0KhchV6oer2UnUllmxW+1mFQ8u9sg/IJ0urW+Qu5/GLq0L XtYVpREk9gCZjFfwISQSdr8211GWxJkftdXiDMQiUMVYJlYp4bspQOx6DzYtdGbjikVoINIZLeWe X1LSsywr8q1jO3SDqO+EkUfBWxnt3/99K5q9/sDXvvVrP8etLu1XOnrJeT80Zz1rxq5F3Yu42knr V+PITyXUv/HL1xH5SCx/M4Avg/BkNNeVK/rSUAG6ynbuHtHf41+G+PenT9f4I5LnNTS30OkhPpF7 b1sKQFGg+DLbmgjaTRJspbeMNAsVJJDAnijVBbRxe0TpIqQsFMXfJSK/0QabSQo8W9ZU3Ltl0qcj ZUzP2F7rFDdNNh9aZT6YTY0A//n49NWW/sAK9C7we7gvqjFoXFHtaHoSYhxp24VV/rgndjrnGV5M JPJyxgZzuJrtrHOOJTHLkNcbwDovOLbusilc1pBPoA7Jr1e4F+w5zvU2Buj4lHvkRxaH3nB+pGvg /G/Gybg/9+/8qkjyyNT/Lf51LH57ei2jjJj6B3eaag4qVc4pWG4aYJHPzSBQzSSPC/R85swBrMP0 w26yIli/VClK9TsDaBfboTGbwj1qSuuokwml75mKYjCJX09a+h5902hob1+In0vD3sW9GNL9uhub jjH2TD0SKyo6xDrGmKverWh8krZ8pNIqyEC4BVbBFhV3qDHfKfNNwhI4k+aVJSeblZwt8bXAlTXe 2JNuV2fhUMEMCnxPeL1aSX+6Sz6m4riGvABw+foEjpRriLP5ovJmJ8K3ZSdvQJwMUGjQ1KNJpgXd ytvtABLnbL4BsJwQRLAQkYKlcikC6hCVjkBXEC1dDCtte36RxUV9iBeIOzIV/WiO/dEJ+/xXns2J PzoRX75xtXv64O102fpe3H/+/trV5jNTm19t4t80VH7WifrXlUX/TZzzQYnz14t5+JNvpHb8vCGg N8rfS7qc20btPJ3hsKLnD0LOH/Tnzf3NltxsgtieJEaQPqIf+wuEx7zEH36Nrzz0HvAOfR/5bR6/ 3F999EbUUpdZtSmWxLZcD07AUe0OPHkCqmfKaamz/ZqiOaP140QEopSYHbEYrIGIWbPRkaO2mLrX CnGR8BxEzaYLdImGGZRSNjzy8PddS+5nTZofOjL/evGivpGMf9CH//XROvySSb7rT389SL665tzy R6+1yvcFlL9/a/H9LTP+kAZfRvQ2SXrJiO6cPJOP9OG3lDcE/RDlfQk0QjXMriY0XzZbrli5E9Hw 274Tjwd+8ARxhqDJ4GwndAcorLUwkALYhYOibDbOViRh6tghi85tnbhSywVKoNiW3xQHyjzdU8MY mWr9CdE/seEPeBt8x4aP8TWIrYlh+ClQwxNZ208KK2kSPRySdaW5W4he+3myb9Rl6FWCVDbNcbeX u4VYDgXAskwrTGRuxu8PM9TsAToFTv4G93B/5Fn9HRv+Py9z6u5lcuTA/NYL+3NE24fYl7Xy9slY Ci5i+3J2OuZ6bSflFiyosCwPkyNYSmDE4FQPeFHOgwzsHJL2UHWJRYO+ohGarp7i0tA7cDIcc1Rv KB3D9Lo1t1M7oUYe519/5F9uvP8cdfLxs35OlnwT+d1APDcjjpQgYwBgL/3a6iEIgRrCKHCKkuBt 0xNOTZxI7CgKfqKIXMrSNmrK5cEO6OFEtswUO8Kwqttr2dfU4VAjm8ngn3AnskzwnrrOKAxv+tq+ QhC7H8FfcS/4/bq7ooeNMAxPqdmRBmJA3pGTJpOn7cGh52sQmHaNsSn42D3NFxs1g/BVNaxlQpOX FiuBUhmgs/XRAmftugQ9EwpjLOhE3+vyhfmHOukI//QHU5lv/NNHJjNsuhZICWpDS8dNJ9uuays8 IpO0T2eGm2xoNPZLqnTMLCo2Rg4cTEfD07lyyHeUUNgMjxQWB2XKlOEVDe1IjN4E+2LkO/1bD/Sl AvbRZ+iDx/ptA/BbrevN4RF7zOHx/25n/ChH9UfXdexHXdexcQS5ddj2rYLby620d/39hoZWC2eO J9GB0rczAYyxMLG6TSpUMeXQsgeWy4lw3NT8NJUWfmCUAdikDGJRcMhApZGri2htjjwYf3Bdf8wI /8cMqn9hd1eb6+dtuz+4qf8e/22gb56O3dyJMm/d7YKGbKstsDiG+n2TnEI+PIpIzYEVrXJ8CalZ VE0Pagwjcg0tY6PeF447z8sIDiXq5GuodvQAyIzm0IDm6/CeXWXUMenTwvbP9Ud8Ev8N1ZunY3sm YI/AU2OKuY6130cCh0oV7GAKuhWUKEDoZI0pqt0oFWo2EG2n8mCl9X4Flw4eeccF0PVzrnG4mQ6C pmt6XX1q6Cl/T2Iwutz23mn3i+P/I44eb2HPOL5dPz0HG/Ffy+SG4iszVoyWJg55G6G0+X3lklts zfe7GaoXQJXsTjq4LDJWCoskLNjDAksRJuoD3HGDsjYdWu1PLm07u9oXNbJh7+rGvLQH/Wt7w4cS zxfq7gckn+8Dnwfm/e3TNeKIVnCdrdz4xO7gGmQAXGN2ccubK9pcrBjS6wA6AWxGAvn/be9OmhNl wjiA3/0UqblaU7LE7ajRCGgUd/SGAirR4BIxWJXvPmrUsZtubSytd14fDpOYzBRO/x+2ptv+TbLN 3pxXxERdyc7GUsmuJR2nl1z0nfiwLRWqrf48nRWmjtzK5RqsYyjEASdsdOn03+1LIzL6S0dsizxp 7Zph7Z9NbrPevfj9zDaknZwXm54nilXe66dT0yqfy0xKnyuN59KvyYZstcdtvmM4qfaYH434uimN THPISZKlmiXBetOKiTW3SJn94iJtzV+WJcX9MLKMz1uRD9wLgaJDHy1cmCcgXDfjhDBPQGCbcaLV FjXbsKJ2fDKOerFEoSip3d7mDiY1FToDnVPUT2PWkDuCZsadoaLm0gs9pygl1Z2mo+V6djpvuaZr aF2v92ELfTdVyRvcrS91+6XxaItABO8sbDe4CWv7bbfcA0PnIFsu9Noly+DMpqr3hu9aszXh7NRy 6dbyXmMYd9yaHjM8ZxXLzjK9ruK9RONWMyqvq5X1JFNcd2adWkauO7WhUZ2LptyNqdos4FRTls7m YZGw24277ba4SWv3nXVsLVuSxcaYb7RF1R0nubHozcRO4nkU1VfTlTZbFLKd9vumH/X50dMG87dc 1LOT82Gbe1/mc19rRe9oBTvfjjsDfVosv+U3f+xRj7EvdfjMHPVzeb5xgG2Cv4/Lq23+Onbym8CP uZjKhCzWf7sJwSfb3Zbs70+s04Edru/lC5yZKqjrxvOoIE2HcmXI9ZVKK2WN3IX2ZRhVfSGpkuV0 Lb5rfs0tMyPwK6Gs58t8TcpHLW1syfVJJZ+qpcviUqyOGM+yBAPhf9aDPSUWSEVNXVfUw1a3JT28 3uF5DAWtvr5Gl1ZmNnQ6kutIhm43YqqTLDpaK6tqdaE3sGV9Eh8trLaiWnLlo9qNc8pnc9RyX9zY wh7ofUv0avNPoVtsdmOtLXn7HETTOKJR2/4rvu4hueL/4DONnx7wMX1qBziyfw9yKr6BW8rcycdV jYFBxidUNz6Yci1ijMYJgSAmh/jP8cNoYYASwORa+c7zKP8b9Eno398/YUMH6NUjYK5M7i8127vZ v8c3PuHifpHXB/WfBNC9nD2QIGgvNZL7wr14JJSxJTwT1OcNksm5MSa/0UtN5e5OL/1AJOw6fqv3 5z8ewOs9pk8ze3eVotbMd8eMVyxx9WFNG/rxL05Irdbd0d0L1ToH72K50i9EcDxdf9tBobVo8x8Z hsVaCoIwDdXSfQ5QoFLy3TRUpBTdB0IL9Nw+clMH9NHpT6SxkOBOtOEP7mOijQXkTeINByk7Ui6l N1Ud0aCBG4lnHn2f8xHJGcLTDSn76+PIhlih4aqEIUR4gAjRWOAyguRj/46EIBI8GAKQcoW6xP/B E/+Q5odc3z4H2EIfGJSPfJYIQb4nVpAP2W1C8w6NImTuKDc7jMQdMNUOaTtUPw4JAQj9Bkl7wxse Emk+Ig2PCLxxhgdyUcR5FDEMbzgYnIt833AO5sKiAiRrhZgWDdMK/SwoZJa/0SDBq4vDchh2hccG g6rCWg3QmqJMK7roTPmTA0dFgdChYIFQaNtBc05Mg5PU225gFhPWdlCSEvVOg6YoIWGFYNFJDhCM In+rQ2ToF0xXCCQlhCbw4BAQ+drw3yJAaAFCyAeHfMg1uwHiA9vtuRDrmdRCb+fH20FiAW7lAOJx yAdOSOMEoHH8uw5koIbyyDPEaU72kic2JIS2lAfG0QTZiZhIGureczeW5vjG37Rq0rLwFTlgFmyS DDWRu2sy+wbRFjahizK7NCP4K+xkFSox1HgvCjHEJKERL2gIsI0WytJZQFgW8kF0f5LFVwKI+MmZ JU5o8Akht1AXIRzRIQ/iO9HBoj2QpsMWNChP2W6vZ0ACMxg+uoOO8DL0IQkjXoSD9Mr1INmkDGqi d9MyLnR2HhrAQFoKhq+gzV4n0xVoRnDtCMbVpc+MNYEUGs7cY99RZ4gcvn5HviN/AFBLAwQUAAAA CAA7Xe5UVvNBUM0AAABOAQAAJwAcAHNhZmFyaS10b3AtbGV2ZWwtYXdhaXQtYnVnL3BhY2thZ2Uu anNvblVUCQAD4uTPYuLkz2J1eAsAAQToAwAABGQAAABtULtuwzAM3P0VhKcUqIwE6NS1U+dmLiDQ RMXCpgSKtgsE+ffo0W7dpLvj3ZG3AWAUv9L4CiOvKaq5Nc7bQu5gC85icgvttDh/eLbxuep30sxR 6shlOk/njs6UUTnZL3ONoJS0eCHBQawzoM+UwYI3iIibwhFIoKeyfIGHHl1wxgAYxTwLlA7QOkDr cMpPU09cC9t6y0w/03fuaG+RC3Er3wqQ7m2//ijgvQn9ZiFqJfrgwkiSm/L94+1vqUTFXZDpP8fP y0s7QPUc7sMDUEsDBBQAAAAIADtd7lSJfXwCSgEAAAcDAAAkABwAc2FmYXJpLXRvcC1sZXZlbC1h d2FpdC1idWcvUkVBRE1FLm1kVVQJAAPi5M9i4uTPYnV4CwABBOgDAAAEZAAAAJVSzU7DMAy+5yks 7cr2BFzgxn0SHJBotnptpDQuibMKqQeeYrwHZ55mT4LTrNVWBhJWVdf25/rvW8Ca2qXFPVrQnTYM m1gp9Vijg4bKaDGA9gjBNNGydkgx2DfYIW9r4yrQDkzTkmfgWjN4fI3GS06gRpKQYwts5LOMCEzy XBS7kSwcf9AZa2GnjQVyEPROe7NSak0g/hjYa0boUhEToNZti04aUKooCgUvWWAuJ7/qT/agc70w 2UkPiLvsOX68LyfJmKccuYUtOdbGDaPPp1H9qVw/62LyK5XM4+HzR6NjG+dN/TM6zvUn6nj4+j0+ busCcHVb6f28nEtG3E+Iq21Ou0iHU4uFEBAYQyKQHJYi55O6toGtGZSPwgf0e8wpa6EMek8+M2aD YKmqsBz4JbGNp07g6VSBLK7ggYVRQtqA2ISEyuxRZzT7BlBLAwQKAAAAAAA7Xe5UAAAAAAAAAAAA AAAAIwAcAHNhZmFyaS10b3AtbGV2ZWwtYXdhaXQtYnVnL21vZHVsZXMvVVQJAAPi5M9i4uTPYnV4 CwABBOgDAAAEZAAAAFBLAwQUAAAACAA7Xe5UA0BzvZkAAABDAQAALgAcAHNhZmFyaS10b3AtbGV2 ZWwtYXdhaXQtYnVnL21vZHVsZXMvbW9kdWxlLWIuanNVVAkAA+Lkz2Li5M9idXgLAAEE6AMAAARk AAAAhY/LCsJADEX3fkXoqgXTosuKK8H/SKexjLSJmCkq4r/b8UVBi7sDCefeWxR0Ih/Adwc9hjTJ i07rvmV0+d6SbDWbORULcDXteBNxDhG3vbjgVeAGa5hQ9MY1VhckJBlgLNSW81abNHm+QgUIjoxh UQI5x2ZeGiABPkcp1/CoQRLK6BgrPsW+3Ij/85ZTebvXwJ957/VpNhzvUEsDBBQAAAAIADtd7lQy /KKmagEAAGUEAAAuABwAc2FmYXJpLXRvcC1sZXZlbC1hd2FpdC1idWcvbW9kdWxlcy9tb2R1bGUt YS5qc1VUCQAD4uTPYuLkz2J1eAsAAQToAwAABGQAAAClUktvwjAMvvdXWJxaaU21HZnYZRLXTYLb tIObGpopTVCdDhDivy99IQpU2tgtcvw9bH/SGnawKW2hmBhm8PH5HATyvDpHpZVZv5kFrrBUvkcV G1u6cCKSwmaVpjgVXzyJPLAnEpuK8/A2ge8btl3RVUxZnO5jjNFkHblwOZkwPADbgl5rfw/Nc14Z 6ZQ1xwhmL3AIAGrzVpPQdh1OWkZAiEEiEzxOAaUkZm8I0ADtam3KGpRD46b1HEOSk2I94QV/HE9u VC9Vn8ZUV537EdV+uDDy38coCgLconLw3u5PoNYLck5T1u+ao+58Jy/LXDFYQ8BEBYOzsLawLa13 UjV+uLlK4+AcOHq84PaPkOhkHlJ/hszKqiDjRGqzvVDGULmknfPxoXoWzxMkCaB2VBp0BN9Ush8V UtJ267/uzWCPPIwFxYPbLf4qdx3fPwM1JDkPVJJcKrSRuq7fF6pr5UGormX+kpchdDQxP1BLAwQU AAAACAA7Xe5Ujeu35/MAAAC2AQAAPAAcAHNhZmFyaS10b3AtbGV2ZWwtYXdhaXQtYnVnL21vZHVs ZXMvbW9kdWxlLXVzZWQtYnktYS1hbmQtYi5qc1VUCQAD4uTPYuLkz2J1eAsAAQToAwAABGQAAAB9 ULFOxDAM3fMV3pJKqFexVmVBYj2E2BBDaA0Xzo2rJKWcqv47SdvATXhK3nt+9nPL1gcYHPfGo4cG Xl5r0a5gh6QvEbE4weMmUA490xdCcwezgFhJyoQl8YeSnnh6GG0bDFtZ1KvAY3g2PfIYlCr++v7t hX1MV2aXVDuodmi5gduqqgqxREDkBOUw+pNaV0/w4QBP2HPcOJwQ9KRNAG07MEESQc8xJpkz0gUm dudSbIo9bamJVDZObttdPPd4v74akL8fWQvCjTy+fWKb2HkR+D2wC/CekyU+x4z3SMdwGEZnQWoJ xl4Z1GIRuX++mrrU4gdQSwECHgMKAAAAAABRXe5UAAAAAAAAAAAAAAAAGwAYAAAAAAAAABAA7UEA AAAAc2FmYXJpLXRvcC1sZXZlbC1hd2FpdC1idWcvVVQFAAMK5c9idXgLAAEE6AMAAARkAAAAUEsB Ah4DFAAAAAgARl3uVEJYo5CLAAAAyQAAACUAGAAAAAAAAQAAAKSBVQAAAHNhZmFyaS10b3AtbGV2 ZWwtYXdhaXQtYnVnL2luZGV4Lmh0bWxVVAUAA/Tkz2J1eAsAAQToAwAABGQAAABQSwECHgMUAAAA CAA7Xe5UGeyuE6AmAAAX2wAALAAYAAAAAAABAAAApIE/AQAAc2FmYXJpLXRvcC1sZXZlbC1hd2Fp dC1idWcvcGFja2FnZS1sb2NrLmpzb25VVAUAA+Lkz2J1eAsAAQToAwAABGQAAABQSwECHgMUAAAA CAA7Xe5UVvNBUM0AAABOAQAAJwAYAAAAAAABAAAApIFFKAAAc2FmYXJpLXRvcC1sZXZlbC1hd2Fp dC1idWcvcGFja2FnZS5qc29uVVQFAAPi5M9idXgLAAEE6AMAAARkAAAAUEsBAh4DFAAAAAgAO13u VIl9fAJKAQAABwMAACQAGAAAAAAAAQAAAKSBcykAAHNhZmFyaS10b3AtbGV2ZWwtYXdhaXQtYnVn L1JFQURNRS5tZFVUBQAD4uTPYnV4CwABBOgDAAAEZAAAAFBLAQIeAwoAAAAAADtd7lQAAAAAAAAA AAAAAAAjABgAAAAAAAAAEADtQRsrAABzYWZhcmktdG9wLWxldmVsLWF3YWl0LWJ1Zy9tb2R1bGVz L1VUBQAD4uTPYnV4CwABBOgDAAAEZAAAAFBLAQIeAxQAAAAIADtd7lQDQHO9mQAAAEMBAAAuABgA AAAAAAEAAACkgXgrAABzYWZhcmktdG9wLWxldmVsLWF3YWl0LWJ1Zy9tb2R1bGVzL21vZHVsZS1i LmpzVVQFAAPi5M9idXgLAAEE6AMAAARkAAAAUEsBAh4DFAAAAAgAO13uVDL8oqZqAQAAZQQAAC4A GAAAAAAAAQAAAKSBeSwAAHNhZmFyaS10b3AtbGV2ZWwtYXdhaXQtYnVnL21vZHVsZXMvbW9kdWxl LWEuanNVVAUAA+Lkz2J1eAsAAQToAwAABGQAAABQSwECHgMUAAAACAA7Xe5Ujeu35/MAAAC2AQAA PAAYAAAAAAABAAAApIFLLgAAc2FmYXJpLXRvcC1sZXZlbC1hd2FpdC1idWcvbW9kdWxlcy9tb2R1 bGUtdXNlZC1ieS1hLWFuZC1iLmpzVVQFAAPi5M9idXgLAAEE6AMAAARkAAAAUEsFBgAAAAAJAAkA 6AMAALQvAAAAAA==