16438 2026-09-26 09:26:21 +0000 spdx 3.0.1: Missing internal dependencies: example Go blindspot for CVE scanner 2026-10-08 16:41:46 +0000 1 1 7 Build System, Metadata & Runtime OE-Core oe-core other 6.1 x86 Multiple NEW Medium+ normal 6.2 M2 1 pokylinux JPEWhacker ccasciato randy.macleod romain.naour --- --- Don't know oldest_to_newest 106938 0 pokylinux 2026-09-26 09:26:21 +0000 [Bug Summary] spdx 3.0.1: Missing internal Go module dependencies and stdlib tracking in create-spdx [Component] Bitbake / Core Metadata (specifically classes/create-spdx and classes-recipe/go-mod) [Version] Yocto 6.0 (Wrynose) / Yocto 6.1 (Blacksail) [Description] Starting with Yocto 6.0/6.1, the `create-spdx` architecture introduces native SPDX 3.0.1 JSON-LD output. While it perfectly maps the system-level build topology (linking the recipe to go-cross and go-runtime), there is an architectural visibility gap regarding language-specific inner application dependencies. When compiling a Go application via `go-mod.bbclass`, the external Go modules defined in `go.mod` (fetched and compiled statically into the binary) are not declared as discrete `software_Package` or `SoftwareMaterial` elements inside the generated package SPDX file (`package-<name>.spdx.json`). The entire source repository context is summarized as a single package entity. Consequently, downstream compliance tools and vulnerability scanners parsing the SPDX 3.0.1 graph cannot detect internal application vulnerabilities. The generated Yocto SBOM leaves the application layer completely blank. [Steps to Reproduce] 1. Create a Go application recipe that fetches external dependencies using `go-mod` (e.g., github.com-robertberger-go-hello). 2. Enable `INHERIT += "create-spdx"` in local.conf. 3. Build the recipe: `bitbake github.com-robertberger-go-hello`. 4. Inspect the generated package SPDX file using jq: jq -r '.. | .externalIdentifier?[]? | select(.externalIdentifierType == "purl") | .id' tmp/deploy/spdx/3.0.1/<arch>/packages/package-github.com-robertberger-go-hello.spdx.json [Actual Results] The output is completely blank. No `pkg:golang/...` entries exist for the sub-modules or the implicit standard library dependencies, even though they are compiled into the binary. [Expected Results] The inner Go module graphs fetched during the build lifecycle should be appended to the package-level SPDX 3.0.1 data stream (similar to how other ecosystems look to leverage `SPDX_PACKAGE_URLS` or macro definitions), providing visibility into statically linked dependencies. [Example Context] If you build this[1] go app you see that go.mod pulls in go 1.25.4, which has known vulnerabilities. [1] https://gitlab.com/meta-layers/meta-yocto-training/-/blob/2026-08-18-training-wrynose-blacksail/recipes-training/golang-helloworld/github.com-robertberger-go-hello_git.bb?ref_type=heads It looks like those "internal" dependencies do not show up in the SPDX, so the CVE scanner can not detect the CVEs. With a third party scanner e.g. trivy and some cooking I get this: ## ❤️ Unpatched Vulnerabilities (Combined Audit Scope) | Audit Scope | Severity | Package | CVE ID | Score | Dep. Type | Linkage | Remediation Action | | :--- | :--- | :--- | :--- | :--- | :--- | :--- | :--- | | 🎯 TARGET | 🔴 CRITICAL | `stdlib` | **[CVE-2025-68121](https://avd.aquasec.com/nvd/cve-2025-68121)** | 9.5 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.24.13, 1.25.7, 1.26.0-rc.3_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2025-61726](https://avd.aquasec.com/nvd/cve-2025-61726)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.24.12, 1.25.6_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2025-61729](https://avd.aquasec.com/nvd/cve-2025-61729)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.24.11, 1.25.5_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-25679](https://avd.aquasec.com/nvd/cve-2026-25679)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.8, 1.26.1_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-27145](https://avd.aquasec.com/nvd/cve-2026-27145)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.11, 1.26.4_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-32280](https://avd.aquasec.com/nvd/cve-2026-32280)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.9, 1.26.2_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-32281](https://avd.aquasec.com/nvd/cve-2026-32281)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.9, 1.26.2_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-32283](https://avd.aquasec.com/nvd/cve-2026-32283)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.9, 1.26.2_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-33811](https://avd.aquasec.com/nvd/cve-2026-33811)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.10, 1.26.3_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-33814](https://avd.aquasec.com/nvd/cve-2026-33814)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.10, 1.26.3_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-33818](https://avd.aquasec.com/nvd/cve-2026-33818)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.13, 1.26.6, 1.27.0-rc.3_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-39820](https://avd.aquasec.com/nvd/cve-2026-39820)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.10, 1.26.3_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-39821](https://avd.aquasec.com/nvd/cve-2026-39821)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.13, 1.26.6, 1.27.0-rc.3_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-39822](https://avd.aquasec.com/nvd/cve-2026-39822)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.12, 1.26.5, 1.27.0-rc.2_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-39836](https://avd.aquasec.com/nvd/cve-2026-39836)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.10, 1.26.3_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-42499](https://avd.aquasec.com/nvd/cve-2026-42499)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.10, 1.26.3_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-42504](https://avd.aquasec.com/nvd/cve-2026-42504)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.11, 1.26.4_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-56853](https://avd.aquasec.com/nvd/cve-2026-56853)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.13, 1.26.6, 1.27.0-rc.3_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-56858](https://avd.aquasec.com/nvd/cve-2026-56858)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.13, 1.26.6, 1.27.0-rc.3_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-56859](https://avd.aquasec.com/nvd/cve-2026-56859)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.13, 1.26.6, 1.27.0-rc.3_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-56860](https://avd.aquasec.com/nvd/cve-2026-56860)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.13, 1.26.6, 1.27.0-rc.3_ | | 🎯 TARGET | 😠 HIGH | `stdlib` | **[CVE-2026-56862](https://avd.aquasec.com/nvd/cve-2026-56862)** | 8.0 | GO MOD | **GO (go-hello)** | _UPDATE TO v1.25.13, 1.26.6, 1.27.0-rc.3_ | The fix is trivial. Just replace go 1.25.4 -> go 1.26.6. But first of all some tool needs to shout that there is an issue to be fixed. This seems to be an architectural limitation and I'm not sure how this can be fixed system wide. It also applies to many more use cases like go, rust, npm, python where dependencies are pulled in by the recipe. Please advise. 106977 1 JPEWhacker 2026-10-01 14:42:52 +0000 The dependent go modules should be present in the SPDX output as input packages to the build. I believe the thing that is missing is: 1) I'm not sure if these go modules have PURLs assigned to them. I don't know if this is possible to do, but we can investigate that further 2) The PURLs from things in SRC_URI do not automatically propagate down the the packages that are built using those SRC_URIs. We might be able to improve that also 106997 2 pokylinux 2026-10-02 10:06:34 +0000 bitbake virtual/robertberger-go-hello -c sbom_cve_check_recipe returns this: WARNING: github.com-robertberger-go-hello-vuln-git-r0 do_sbom_cve_check_recipe: binutils-cross-aarch64-2.47: Found unpatched CVEs: CVE-2026-3441, CVE-2026-3442, CVE-2026-6844, CVE-2026-6845, CVE-2026-90801, CVE-2026-90802, CVE-2026-90803, CVE-2026-90804, CVE-2026-90828, CVE-2026-90829, CVE-2026-90830, CVE-2026-90831, CVE-2026-91779, CVE-2026-91780, CVE-2026-91781, CVE-2026-91782 WARNING: github.com-robertberger-go-hello-vuln-git-r0 do_sbom_cve_check_recipe: bison-native-3.8.2: Found unpatched CVEs: CVE-2026-56390 WARNING: github.com-robertberger-go-hello-vuln-git-r0 do_sbom_cve_check_recipe: glibc-2.44+git: Found unpatched CVEs: CVE-2026-18374, CVE-2026-19499, CVE-2026-19542, CVE-2026-77117, CVE-2026-80489, CVE-2026-8674, CVE-2026-86805, CVE-2026-89092, CVE-2026-95818, CVE-2026-97399 WARNING: github.com-robertberger-go-hello-vuln-git-r0 do_sbom_cve_check_recipe: popt-native-1.19: Found unpatched CVEs: CVE-2026-18739, CVE-2026-18743, CVE-2026-18839 WARNING: github.com-robertberger-go-hello-vuln-git-r0 do_sbom_cve_check_recipe: zlib-native-1.3.2: Found unpatched CVEs: CVE-2026-76844, CVE-2026-85091 which means it does not care about the go version used. trivy scan returns a lot of CVEs due to the go version and the go stdlib. tree github.com-robertberger-go-hello-vuln github.com-robertberger-go-hello-vuln ├── aarch64 │   ├── builds │   │   └── build-github.com-robertberger-go-hello-vuln.spdx.json │   ├── common-package │   │   └── github.com-robertberger-go-hello-vuln-package-common.spdx.json │   ├── packages │   │   ├── package-github.com-robertberger-go-hello-vuln-dbg.spdx.json │   │   ├── package-github.com-robertberger-go-hello-vuln-dev.spdx.json │   │   └── package-github.com-robertberger-go-hello-vuln.spdx.json │   ├── packages-staging │   │   ├── package-github.com-robertberger-go-hello-vuln-dbg.spdx.json │   │   ├── package-github.com-robertberger-go-hello-vuln-dev.spdx.json │   │   └── package-github.com-robertberger-go-hello-vuln.spdx.json │   └── static │   └── static-github.com-robertberger-go-hello-vuln.spdx.json ├── src │   ├── go.mod │   ├── main.go │   └── readme.txt └── trivy-scan └── github.com-robertberger-go-hello-vuln-src-vuln.json under aarch64: our generated spdx under src: what trivy uses for scanning under trivy-scan: the output of the scan I'll attach a tarball with this info. 106998 3 5251 pokylinux 2026-10-02 10:08:12 +0000 Created attachment 5251 spdx, try vuln report, sources Maybe this helps? 107031 4 romain.naour 2026-10-08 14:38:06 +0000 Hello, Thanks for the detailed report! I noticed a similar issue for a custom npm package using a npm-shrinkwrap.json (npmsw fetcher). The generated SPDX 3.0.1 sbom doesn't list any dependencies listed in the npm-shrinkwrap.json while external sbom tool (cdxgen) list several vulnerabilities. The sbom only contains the "software_packageUrl": "pkg:npm/<custom_pkg>" for the main custom package. But there is no `pkg:npm/...` entries related to the "sub-module" list of npm packages from npm-shrinkwrap.json. I'm agree with the analysis: "It also applies to many more use cases like go, rust, npm, python where dependencies are pulled in by the recipe." Also, when a vulnerability for such "sub-module" is added to the SBOM, the sbom-cve-check should be able to know if it has been fixed or not. How to distinguish a patch fixing a CVE for a "sub-module" or the main package? Best regards, Romain 107032 5 pokylinux 2026-10-08 16:41:46 +0000 @Romain Naour: I think npm shrinkwrap is disabled in recent Yocto versions due to security concerns. Nevertheless you can still use the standard fetcher to use npm packages with "internal dependencies" which are not seen by the current design of OE/YP. One solution I am playing with is to use use a 3rd party scanner like trivy and let it run over parts of the sources (lock files,...). It looks like this can detect such issues. If you want to play with my bbclass I can send it over. It's not yet in my public layers. It will eventually be here[1] [1] https://gitlab.com/meta-layers/meta-resy/-/tree/master/classes?ref_type=heads 5251 2026-10-02 10:08:12 +0000 2026-10-02 10:08:12 +0000 spdx, try vuln report, sources github.com-robertberger-go-hello-vuln.tar.gz application/gzip 26161 pokylinux H4sIAAAAAAAAA+xdaXMbOZL1T2Ho464pJm5An3p72rPtCO/2RLd7NmI3OhQ4EhK3KZaCRXnaO9H/ fROkKFOHzUMUxQPwQYmFqgLyeJVAPSQu+uPLm3Aam6vuqAk4GtO/Cxx1L5ruJQ4GTffTzWDYe/Os AlSMUeWTGQXzn7PyhimujDYGpHgDHLhUbzrqebddrty0Yz/qdN6045uEw/FX6y06vqflYin9ez+K l1quaQcr6F9D+Z4zaXjV/zbKavq/9vF3f4Htaoawgv6VhuL/gitR9b+Ncto77X33oRlefOgPf3+h exQFayln+n74CUyLe7YAjBUz6Hx4ofbcKzP9j5rmm8pddHxPy5r+f/tDd6nTuylcnLbX6Y/T/22b 4Wv3uJb5sg39z/n/E/jPmRLsDv+V0m9ASACK/2AbAjhy/P/nyXexGY7xj/HJ2cnleHzdnvV6E19t Rhe9Uco9cQqnbPJV97bmxI0H6eTtyXcXI399eXL2P/88GX++RrrEX0box/1m+H6Ym1Khn+jL87P5 r4G+j+V3LMc4oX0XZBfURy7OAOjvf3+p8P1nuvikXfPNKj+kJra90B8H/zt2fUrBK2a60tKllEfR tSbyLkThmVeW8eBnlXtkt8Nx76drHL67CpgSNeO3uxv+2vaHF5u/57hpBr0Gu1Mxlls1o3N2DuXO 7TXGv+OoJfGQPCbiPvnz7SoSZQ8kynUXqCHuI+dnXJ4xUyU6L9GfRhd+2P+/ifhOSu30x/t0a/4v KpRbmcyU+ZRjkHvhcNrwL04180tyx9OGLoi3F7xrZXHSXj91/aDv25M/SQRDfzXp6vzt50TwkaT3 cl3/qm62L4AHreiwU5iXwy907Icm3lyRrpaSx3IBT7SGaZS+G7PSXaVIWl772M3KZICQwTno5Rgt CGfQMp3p0ReVZ0lr6awAy6OSTroMHGQEF2gEJRSLKK2yTgvEbHvptuEPnsbni6Mx6ur1qMn9Af6l IbmPrvwwYvHRcNOfAHtsRljkgfFm1B9/Lj/2r64H+KEfi3aG5Qptk8f/8FSPpF2i43cDnIjxW66+ S+K7lVpvean9tsB+2Ybs9+2SJ0xr37PWIuVJpfmzppct9rKcAlYP75e87q//+f7Hf/vlx+dZ7nKy GdB4tnsLAimROQWPXIqkJQOhZHI+chcFkyZEjBm08AKzIcOJIFBLsizDuUkYQpoHlJVFMw83M5c5 /9v0KnsHOav7zBecOL89+dfRoMjx94uzz00cN70rHPvu5MfueOT7Q0KX5fr8HdV64vpfnvwPj4/6 V370+W83o+umLeroDyn4HwwWPZaKWyds46h/PZ5e+N+bzo+lGZ3/akaDdNrpfLzst53b+3dKmEz9 aDvvPvy1036+Cs2g7fhh6oxwUCKiTtvcjCK2HULeDhnYzcUF9blzPW1Xe7rbKBI+d+lw99K3l73n OtaKsLG69RXHLbIcDf3gfRm/9XMfR/NCfff46FOnfJzWnjPhtyf9uQtuzJ7n0GK1psVr5OJhq+jL M34qzv6F/izVhlJrUvvuzzz4LQ965HY3V8XbHnlLp9v54c7mSwjSPomQf6Ujhw+PvSKAHlsGgSok bAQSZhKfs+qbdkThQuhNoLhXfgv9YW925unkaxLdJxwVt7ob0N4p4UfqNx33g4uGguXLq2LGl54r XeRNx/7uBzeTMMRHwMjBxySzCAoS1ypxFsmUgpUheZJENM6TOUZgkKMMMWljPXASqZz3k6fvqRh/ cE/LAtiomQnGMekhQNBRK61MYNEUVbiQhdaCGUDvweuocnYgUlLR68SUtIF8R2aQCXI2CulKBjPp SofMJDihNPWGyehsTsomxQVnNmjSry+h01yrP/Qzxs9xgL9EMqP0c3kekkW3l/3rvfP10Vzje0II oy1nijvPXNErlzLTwNihDYr70rvSZercbIiVR83Vei6WfYhKkzaVE75ohWvrLHXYeu9U1AElSK6y huyYdYZj9l4rEW12NmFe0cUmDabuxv51QeT5ft8+eqhVP92Mr29KtDVu6gCw4vQjI1rNW44DNLSM LCsShkTuAklWM7ocyw4ZKMLijYIGyTY4o7yyhPA+gM4hB+uttZGwHSFZcMwBC1IHy0QM1ilpIj0N nCY08Svru8DFcq71NKhcUHdGNFxqK6hUUFkCVBZ60z1QORwU8YAcgCKvkKRBZ3mQMoAjt9U+qxye RI3d7ds6M0xPoMdsDubAwWM2mqkYshEMWehLB4ohdDfOkJ74ystoUE4ImCRo7bgpg4njxBAyxx8w DvwI0/S9Fz6FJjs6ahtMW9wT59yeQ+8/3n+sILEZkFjoLAQSv/352sSeJUvlf1b+58vzP/FT5X/u Znl9/idTiqsv/E8jJ/xPJir/cxul8j93na1Y+Z+V/1n5n5X/uTn+J36q/M/VZwlIans+gH4x/id+ 2hL/c2K5q/M/RZYqCG5tMuAFJuslhJQo5szRS0GRp2CoGRmN8kLaKMimkhY82MSTMfYZ/E9q8RHz Pyf6Ol7+55T72Y+dYozt284l+oSjKent7T1GaH+MV21niBHblpo4YYbO2t0hMeKguS4Os0fc0Oc6 3csSwYpfVm7oK3BDp5DwLW7onbE/ZoceLBXEeO8owKZaTlnpI2hkQUF0HC23xlT+2A5q8OgCyFeE 99W85UBf0wZB/RZZJ8MTEyplq4JFxbQRhroPh/2advLkqK9pK0gsBxILneU4SKYxMxQpBs+ENqAA BdlvYtZoAJlBVJLp4ZNMa+SxKVBZ6E37xP0oZSf4H8o85n/oyv/YRtnG+//K/djd8vr8D864epD/ UQhjZOV/bKNU/seusxUq/6PyPyr/o/I/NsX/qNyPVQfQ+z54fiHex1Y4H2vxPVhQLENUGI2TNOoX QmYyJ/RkLiIZmwWZFLOc62yU8D5zLlCAsQ45/RbW5nscL9fjKHgeu+3m83Nkz/WAlyRfVOLFKxAv vkG6ONL0W4eXemuXnf4rabfmE21tIsWWMcgwCMssOCmQ8wxMc1QuZx5S4JHHpOkjSI3Ol3nmnLRl CmhUI1jm66TYYsybRDdKPmtnwVrIKHLK6BNpIHjODAuO5B7IUplyLMTgHFUlixYBPDdRGU/jJ1s+ mHGCVOfRB59BIwCpM0XPE3BuZJTSUre0iM5KUm+U/FDZDgxdSYOmPDAmPPWcBAsJUclkHMODXpRe mQ4Ve5d+KbnQUY6D6SAgy4hIlYGOck5GarnUKgptNEZdOZQ7qMGjmojZFcBY6CkHGlE4CtGSjTAh hWEwXhoGgmIu5TwYpY4vojiCNFkHmCLrNbFjoQ8dR7CRaPTFdbAiCRdFdtKDBMGcFzkgM1hplYdO q6yAshlAWehJ+0ap3KuyGv+Lql01w+7MT5a8x0r7/2r+BrgUuu7/u5WyE/xfIx/zf1Xl/26jPMv/ l3t2z96gT0+uZOCdKtvQ/7f5vyCZMPfxXwgaSFT+7zZK5f9Wtmplq1a26sLBteHS6MgzjZ0dE9aU gVrw3kitrGdR8+AI47V3wMA4pHqRKxpau6zou+C+8P+WW7R4P2rYOGn1vma/Bl+VVL8IpnZ5hmN5 3us6gey69Nc1zX91FqyzgYUA0jqugjdgPPkx+KwRkpMy+pBZcIkJF7VFG3SUDHnOKjlAwezqyX0e NLpO2+x+WS3+n6Btu+y8z6ysMv9DT6I3heikZZ3/2UbZifkf9UT+f17nf7ZR1vL/yUdd+30IZRv6 X7D+G4TW9/FfSIA6/7OVUud/dnRgVdd/1xm1OqP20jNqkyfZ+fe3k0j7xUd6wAmedmXy/8cvUqNe PBLYTE2pOZ/66PlEhHdXOhRWy2tytFedOzm7r4yz2WV2Y/J3c6kK5kPHmqhgRSffc8bZ8vOxKwww 1p2GXc0QV599fS74zCPICuK4t5h34g2DmTec364Ze/fH9QjbdtngZ5e84Yk1ZG8fdXPwqJtnJ9+s +aHfjufDRW5PF4YqO+5qu7XO7zgo4SyAKBlGZXCaxmDlNaihyugxBx24XYoSvrt9XbwELSEZYmp/ Gj5nVSp4Sc1WmgfjGQihJJlsUM5TC0XkdBLj1rkgJQiPoITKTGQtE3dJp8h7F4N+iF9bOLdWg3QE o41kHg3DZFIkGZGmJSbuWSQlJ9TSOx9BSmF8jMHqTI6luUjalPXto5vhuH+FG2yVUSFnn6One1mh o/fcRioJMHIbmNOW5xBtZI5JHUlI2nnwmRvHWOaqtCqOmrbtzub2Ntg2b2wui8mYy1YoktVkZ08V k2QCRSQtBZGNUSwlDoJJIamxHMlHMsmUJBbjizUOLWQyfJalkQ41Gbclk0s5cIUhMMZNUFmBBkgC tEpoMDkyRHCStF3QkBr3TX3ueZT2io+O1eD0QBcpRjLOGJjnJFGfgrSZ0I1sT6KnH8K+PTba5mYU p3lWakaD6vtLpllf5APHEVGioLhG6rL8jx7hLAUBHkusoQRhoF5u74bd7etSSQ3eD/cup8Ed5FU0 2AgaLHSDe2iw93kX556Yd31JzT+Gg8anD03047ukh/86U/uX5vd+njT/+0nz71JrnVKF7wyHgBq8 IRUh3dajcjQ+4JCVcFY6FaI0YPg3UzlO7/T0XZa+w3Oa8mWiZq2LPEwNORV2ne1Z11HvjHU+qePP mJ/K5li+vlfpFuU/xSKZQTHtZjRD+WXN+uS3J1+1nM6fcHrvhEpLnSvr7f/Qbcf+oqTQXOoeK/A/ yzY6Zf2vAVb5n9soO8H/NE/wP2Xlf26jPNP/V9wNPVxUPuhulW3ofwH/kwk6dsf/VGq6/pdV/uc2 SuV/7jpbsfI/K/+z8j8PcEX1K+3/U6KwSq1bfRNdktqez+e+0D5ARTRb2QtoooN1t+aZnnxA08Sr m+8xbNHT6Xy87Led2/t3ZvltO+8+/LXTfr4KzaDt+GHqTF4wYOpM5y/bDoFgJ2G4uSjDms71tF3t Hu34kxLZXPDIpUhaMhBKJucjd1EwaULEmEELLzAbsq4IArUk8zOcG+p3SC+5+ced49Zdf7a768/M 7b+y80+n2/nhzuZLNNAe6V5Ak0fYwe0HtOuQ8JU9gQb90JtAce/hDkGnk69PNrBPUPKxECXBxySz CAoS1ypxFsmUgpUheZJENM6TOUZgkKMMMWljPXASqVxnnyDLAtiomQnGMekhQNBRK61MYNEUVbiQ hdaCGUDvC+9U5exApKSi14kpaQP5jsxQ3utlo5CuZDCTrnTITIITSlNvmIzO5qRsUlxwZoMm/fpj Ic0IIYy2nCnuPHNFr1zKTGNUhzYo7pcizezo29a6DcihjMVeEadX85bjAA0tI8uKhCGRu0CS1Ywu x7JDBoqweKOgUdP5V1A5aFBZ6E0HSt33gByAIq+QpEFneZAygCO31T6rHPaMn7vODNNx7jE0P5qp GLIRDFnoSweKIXQ3zpCe+MrLaFBykjQnQWvHTRlMHCeG1HVCFSSeWhmwyFnq3kELSuV/Vv7n9vif +KnyP3ervD7/kykh5/I/Gzbhf0pd+Z/bKJX/uetsxcr/rPzPyv+s/M/N8T/xU+V/rj41QVLb81H7 i/E/8dOW+J+kg/X5n+Xk4+V/Tpz+ePmfU+5nP3ZKjs72becSfcLRlPT29h4jtD/Gq7YzxIhtS02c MENn7e6QGHHQXBfb3SNuqMhSBcGtTQa8wGS9hJASjbly9FLQyEsw1IysTnkhbRRklEkLHmxJwGXs yxLBbp26ckO3zQ2dQsK3uKF3xv6YHXqwVBDjvaNYl2o5ZaWPoJEFBdFxtNwaU/ljO6jBo4vlXhHe V/OWA31NGwT1W2SdDE9MqJStChYV00YY6j4c9mvayZOjvqatILEcSCx0luMgmcbMUKQYPBPagAIU ZL+J2UnWowzLpXOsJNPd0nCNPF4JVBZ6U+V+LFF2gv+hzWP+h6n8j22UbfI/Kvdj98rr8z84Ay3u 538UwghR+R/bKJX/setshcr/qPyPyv+o/I9N8T8q92PVUfu+j9hfiPexFc7H2nyP4+V6HAXPY7c9 bn6OjAXFMkSF0ThpExdCZjIW9GQMIhmbBRkMs5zrbJTwPnMuUICxDjn9Fl6SfFGJF69AvPgG6eJI 028dXuqtXXb6r6Tdmk+0tYkUW8YgwyAss+CkQM4zMM1RuZx5SIFHHpOmjyA1Ol+mfHPSlimgAYZg ma+TYosxbxLdKPmsnQVrIaPIKaNPpIHgOTMsOJJ7IEtlyrEQg3NUlSxaBPDcRGU8DWVs+WDGCVKd Rx98Bo0ApM4UPU/AuZFRSkvd0iI6K0m9UfJDZTswdCUNmvLAmPDUcxIsJEQlk3EMD3pRemU6VOxd +qXkQkc5DqaDgCwjIlUGOso5GanlUpc9oY3GqCuHcgc1eFRzIrsCGAs95UAjCkchWrIRJqQwDMZL w0BQzKWcB6PU8UUUR5Am6wBTZL0mdiz0oeMINhKNvrgOVqT/Z+9am5tIsuz3+RUV/tS9gaR8Pxwx sesx0DBLMwym6ZhZNibqkYW1SCp3STK4J/q/780qyZZkSZY4gKFbisDYqjz5uJl5897MrHOlz2Xp VcoUk9ynsswCt+FwrfL3fq3yoFA+jUK5cyYdrlT+3j773f8aT2iw5LtF/b357BX/N94FFSLSAB3i /36Bz1dx/1er2/d/5eH+75f4fNT8b/87XPr9HXy+RP/fwf8mlOLL+l9Gq+Nw//dLfA73f7/226of cf/3INHD/d/D/d+v/f4v9ybELSFvWc5SyUyMCMZpQZSeXHounStLcusLLrwrs5iMsYLbQtk8pEbK mxuVS+vxH+D6Lyy5vbbVvvHNpd1v/+5j1n3s5d89h+puEomcXp3ZhEc3aBeVxz4C+ZruG3/Z+fEV XTduo/nOZum/3xw9fvrs0fOTHx+9OTp+s9vlzH9Rqm6WvTl60KL/9ezkH49eXuexpiVvjn7beuM5 GiP9sp/PzY0/yr3nr/mk5HDt+au99vwVHVN+AjW6GNDIGi+5L0tTSFNk8dIomdJFVuQmLQuhdjqm /Jrbip5UFoEUSjH+2wi5Scmky6XNfPDx9CqzpeQh09yYGAzXk/RDqkwIWXBG8TRXNuSiLErpC/ou L/J4BTqvq/G4M9/zWlUwK89nqnTPapKPnskgrAnclrbgLCO9WNCQyAvNMsmyNMvTPJVCZTqN0RAc dVKelWnqCia9j9WopzShh2FNBWdPPq5qKc8sCSTLnMyEyzRJKKWCbUiFSkvmckejxucF/U5ZpKFI udYxWFfmgk3Tovc2z+8Q4WqCj6torl3OjcuFLHkpginz3NOgTUNZlkLwoIPxyuhSaRUJaVIpc6Zp CvhQyoJZWmAG/Sy/Vbn45cdVKGhjgqcOtEYZFbmtrCQZOaGdYHHUFSrnTus8WOpiKzNNMzCj2vCc OtvpRjCbevXm0bfukNyrkbCPQv4jnHbvJr9J3b+86ozzdLTv2W/87HP+Syoxnv9KLQ/nv1/is3f/ 72ZzjOu8Pf2JRwVbz38so8X5hv9FCBPPf4QUh/OfL/H595+S5OgsPw/D9NqrTsSD+O2r2On0V0xC f948PoosdvTlb02yl+GiqidPH8YHjKeMNLjuOFUUHctF2vE85R0yF0WQzClh0qMGddoeDpxMIqxh LOGsw8Qr5o+lOFama5yOSkNG7pIIOKnJe0nzyfPG5UiOeu+r+l3Rr2f38VvHKh1eDPoX51edYTom f6jTUqG4DncdxjsX/dEoFNdb2g1wvIqcJ2pynQwvmnLmB6CdOoyvOoP+aPphx5WHUs1mT39cNQTz u27G1flyw1srOTlqmLivqHnDo5n8x9PBhNbf5H+ajmq7K/ZfStk28n1bdYdV0SRvnpwO0nEEHA3S 0dsOuajjm2fzYt5WS5DZTtZNMYtFNSnaEXDTuN7LpnF/aRq38GbhImbemfuhFj32pTrEev708lnM Mfrdb6vYvgV59xblvT5zyuKntiWeLEZLXgRZ5TLl9mgh0W9L1VlyUQkYN8SXK/zw2rNZlF7zaDwp yOD8r0veFbqrFgv536UsTsi3v/q1Pb6bd87184X6rOmSlSLWdkCbBhLzlmJuhEpmLyNz3Njcxlce 2GapLqibtTVflTqpgpCvyH270P60Iuej1zTvQp1m/UF/0t821BcTXrUNO339KGob3TGOC3600pRR UdVPH45vd/8Pf2t1lJLSbu78F+/e3t2VlGhbb8Y87rVDn7bMD6G4q2cf9z8sJaI0qsvlg6RJbJv/ TZd16rwrl5Fnk3QyHbdK8kMoVh6G+Erz5Oqs2SOOiUaXK0letHu3M1HMnZ/0suimv0zTccgbJUKw Xn4ZNnb3w3SSXheyIuaZkqyiii+yVZnO++/VeUh+qJKlYZbEbLM0vme60hHLtaUO7JK+jzF5mnVk c4c87o9IDV7U/VGzSLQvcR97ngmjc0d+qeLBB0feUR7IxVfWM+aCIBMt6LL05FGHItdZljsZb3Hn 1hcqzZdl8ao/GTRNyuuri0nVmwzGx8ni709HeVXHmZvkIS50cZ88JJfkKRbN9E6KKVXwbfLq2Vky DuM4JhJahKfDdn98Rc0ubJzTny3yNirpjxbq8CDpl8mEJD4dFaEeXEVMPBrtv03IRU36k3FyOujT zDk9GSdVnbwk9R5/pXlE5kMynE6asDFZmLwPYdTk1B+RAkkHhCetf062RhNgJj5pKkGpr588WP91 MkyvkvE0zwN9+/6c8u1PkvF5NR3ERJchKVMyA4puG+0mJj5PL2iRadOmyXQc6iSn6TaetaV7OqhG bT3aGo9nBU+mNdk7s1QPYgMJe436IUza3x5XdSuFWZHkCtA/SkqF5c2DZFLN2kFfzYX+ntbe5s+a Jh+VmE5iQ9437RhVk7Yt89Y3iWf9vVaMTfXmpc7z3FzqvF5gqd31eqTR+i+fvnp6evJsOcXp+7BW 2Z/+TGuE15v1fLtQLBSwoj3SwTClb+WKBkiH6a83dvvig1+ndVjzPVnAo3TYpydq5UmeDch2Jlk0 a8QqLirM25iqTvNB6FQ0addU7uK8mjSVW4XVoaCOWVNKXeXvrtbkNM2mo8k0AjaqtNPXZ2e3xXbT 3OUHUeLyNemeqm6X8LOzY9nlvZPXx897J6fHz3ovXtJvPz2lH2fHP/VOj5/0ntK/k+PnK0q4yems uTlxnPguX3r421opIlU5XajKk21V4WxrTa67YL/KPPk4udiuWq7NFou6DHUY5WHNJLpelfMYoKvb NqFZl0Nd0xLZe/nk7KQxqo4Vj9F6doXPL7zEtb23xZxbyCSbvv21Pxiki9kIJZWSYk+I5XyvUmgp eP8vetDN3/b/s1/8+SNKXZPFtlq0O86XgaBDsrqbzd0Zj0Tc0u2pLuttxF6G7rA/qUODovI6kWyG vo3nh0358Zc/3widW2E+WWbberAdM92oWBuvvsnT906ezcZQJ46h7vlkOLgjh0Zp3WTx8tkug3DB M537qlWvPx5PSaSWOkNuBLYGXj7oWWktWd3b0zVZUo6Cb65KXU0vxmQ5Vm8Hoa3RrE6ddDSqpjQZ e3nvv5mfZn3/+O+P32EZvS7E5B+vf3Thp3xTRnNhxsWlyef2vNzaMbfwM/3w6LpvNVdmaxakqbuj /nhCrZlZ00WYkOG1o3pYNcWXnL0NmPfv33fjCI+DiEp5SfOtLuL0XClys684zQb98XkoyF8I15tr kQ5Yv+LumJtjzrpOz7bVrmHP0vHkx6poKJ+WkabDbdyW0+5Yyq6T8p93bzxsdZBvTe7dHGS1rdEH B5mLmYO8ItutPvHHOLy3u+/bdnhNyKTKuHCFUsqKXFuXyUx4oRmTnFnDhMpTU2SlMzaItNCcM88K WcpSlGalb64d3naMHCejMOlN68Fx8mMYVvVVEj6ckwMzd0N/mQb67iKtSRAT8mbot3jFNj6aAbe6 uFFus3TXIU+Lity36OWMw4ScoEGfFsukah3T0XSYUSlVuVrwOJaYtt92k5/PyblsAMP0Q384HSbj /q9hE4z6g/4fJ+2b5oPBVVtm9ImvljKpA8HHk3ns1Zhn6/7GJsTu7L5sU3RfkBgCuZzDhIo5r4rG 3YyyCckg7mrHMjtkIFYFlRIvMY+7yYuZ6NJZkvj1PP5r/H6Yjq7Ix+9TCbebcePOhg/RMCRDh4pu OoyymG8cbPEEnzz9YdkS3+oF0pr9+bzAWw/u9AJXEdu9wB39vdUH187GrQdzf+/e3Lrnvfj7Hb6U 7erP4Ux9qtp8bmeKM+Y32uQ74Tm7tUbuh3cYXgiN4fVGO3sXPFfMYXgOtZ9btYc3vBbvETz9hNpP P8HyJdR/hN9o6e+I3+if74jf6A/uhBcWk59wUP2lVigeGj9SW6j/FN/sZ++Etxaav8oZqP3KYf2v mYfkpz2mvw1pYAhvsPXLMgX1v2VY+61SWP2VweqvDKR/4iVmCK89tH4THirf+c37cbvgPQPxHFs/ vLSQ/LyR0Pj1liP1F4wpRP8IMiAR+0dwo0E8pD+FUJD9KYQRUP2FhexPIfzmXc9d8FJA9qOQ0kDy V36f06TbeM0csv4KzbH+0wKynwiPjT/CQ/InPKQ/CI/oX2GUgNpvNNZ/RmP6zxiJ2A/COKh4y6DS LeQ8Exyae1ZDqs8xyPUWTmHFK2zkOOegvnMe07weW3g8NG8lk4jZJDGvQWKLhmQe0ZkS23IjODJu CI7MGoIjKkNyUHQeWW2kgIxFgkOSFx6SvIA8LSmhhUpuOV3fCY5NWAnpWoJDw0Ziw0ZxSPIKsnCk 0pCuw/YXCI6VbqFho6DNSYJDykpZaNQpbJHSkGkotQbhUOUNNuoMNuqMw3rOOGhTXxoPVd9CW2LS CmhLS1rIoyQ4VntM9A7aTiM4VHkH2eQEhzrOYeuUg3YCY0hDCI75I27zhcPd4NiwgQ5BCY6NOswX c5hh6STW7xLSlQ6zSx1mlzrMLnXQBgrBsX7HLARnsMpjS7yDjo0JjokOOjSWzmGDFtt/cJgf6hkk Oo+ZtR467iA41HEeuixDcEjbeOioU3qNiQ7zCDx0zi495kh6zJH0mCPpMXfEY7oOO6GT3nFM9A7R F4p5gbRecWiZIjiirhSHhi3BscpDPa84pmwJD1lXSmx+724nOORRKAGZR0pg404YqOsENu6Ehdqu oDsxBIc0Bnaji+BY27GOU5BxpjR0G1NpSNUTHJqwBlpllfEMWWYVtnGlHHQypjyDZOcZNOw85IUr D11iJjgkeQ9tWCrvoYsEmkF7Rxq7gK+x+/eRlRWDI8OO4Ii2IzgyaknZQftu8Y1HCC6go3jCY63H xh3Hxh12bVdzhZgI8RV3CA6dbmkJHQhr6SF1pSBVrxWmLBV050srSNVrBV250thdf00WBtR4jakr jfWchl5009g9ywjH8ArTVwZTOAYUnoFKt9hLFtpBxqV20BVngkNz3kFmvfbQ/XDtoevhGjNttYcW KoNddDTYXcFIeg3BoVFnBGTZEhySvIJ23giOqBtSVlDbNXSsaTS0c0ZwrPLYlDEK8uMtgya8FdA6 Y7H7tVZCW/2R6B6BY6fpFtvDcBxSN45Do9YJ6PKNE5C+cNjdYoJDlccsO4Ij9oFTDoF7BqlagiOi 8xx6f4jgyLCJkWV2h28gotzGiXggojwQUc5y+DgiygU+SbPlPe0VPkm+2QL4UjSQsYO+AhrIbeNk Mw3k5kPjuQ7o3lYPRax9Pk7L3qXoXYYPvViDFX6+JuzOprx3oJhsmrOA3olikneEeyVYpJhkvmu5 3Jli0keKSS6OuT3mvEv64xNQTC7r67spJjXNML2Fm/tAMdnlfEYxuWz7fhaKyZXu+7YpJr0I2iqW FcxZUdqcCZWmRSiL0nhWMJ9JF3jmhVIxGJ3TMouxKgPLhC1ssXI741ZMhQ+a+eNkzjf5MIwie31V JmehvqR1MSmmIbLk3xAb1qGNk7xIbZhc9tMkr9My8jYuhGJY6YZl+smf+5Pz/ih5Uo0jE194VNdV 3W1+fvf9gzYUQSxiUk/zSUPNOEpCfJrQV/R3Q/5Yh8gdOapmRJVU0WWiynPKfNxy+L8nyyHJQtKI l6pZTSfd5PG0jrkMq3pGJrlYQsz6ugYtIWUdLprIU/MU9Di2c9REm3iQDEJaNJEAqoSGZ1HHAODJ LPBiDH1A9S2botKlgBUXdXXZL9oS0iSu4Xm/mo6TNPL+NeySdROmKfJlbu+IT8Yx+VkjDWzimFz9 fleOSb7y9MAxebuKye+eY3K/O8+b/aSNvtp6p4c7Zfb1k+7Ftdq7ogfvbK/MNo6bL+vqxWkAuXrb 5tGCqyf0FmqMZVfPqM2ngru5eu7xX6u/vHtRFMONe5d3u3r+63D1No6T9a7ezLWA3DG/nzumO1zE cJrcH3N9rKNT9RGM/+xYua5mFnLHTIdMYLuvO0busWEHxv9ZwjXumO66WSS8lfBzn9YbW9t737Y3 xnieFllZutQrY7gwpN0yUQqZk1vmQ5CF5rlPtbSl5VqIMkRNmTpndYxLviLRa2/smuj/JqTdnM2f 3JinLy5N48uQnzOJNPnjXTn+6XnLh0+A8bQkhyNGNRtczYPkhTayXMy7l04n51W0icijGF5Uoxj+ LIZ8i8bTReP0VMOYTwNtiPs/mbuhbrNrx6+5cFt2kj5XwLPd3ZCDo3G7ir9/R4MzBt0z4oyDZPIc I/NmHCMjJjxE5sw49Iol4aGXoiOXKyj/bzuYAJkYWP9hZLycYa8Xc74lFNlOeJDMmIY/1H7CQ/OX G4gaL+KxYBL3HAyCW4wMnDuMjJ9jlBCce4iBhvAYGb04BNO452AaaDAMLBiM1Jj+khq65c2lAYN5 GOhFOB5vKUB4iwWzkA6bP4pBr2wfgpGAwUjuO5iIwVh9Ih4LhnHfwUwExO0LB/PwWDAIwkOvwkY8 NH4ID9YfW389uP56htmPnkGMYGS+gfLnmP0FB3MB9w8ID8oPigYQ8VgwGQ5dwOdeQC96cS8x/53w WP+hwXzUvQcDwua/xewnwmPj12LBUNBgRoxB/r9gGovmQXgsmJOG5r+g5kMBPT5BMChk/tx7MCmO 2a+CW2j/VgiJRUQRSkDBfL75YFoWWv+FcJsvEu2Ev/dgXtD+hVASC2mjsHethWbQ/gceDIxDxKyH YGJwMDFs/N57MDHs5WHCQ/Y7HMzMYmxQnyKi1/2G5HJQeCI8pBcWVUt4DgWDFV5h9pu30P5hxEPy 9yChurRYGACFvc0rlYTuX0gjsRAOhAejvkDrjzQWC9AFUhUrxqHX6Kn7ofVLkf8F0TCihL2CQ/ZH xEMklo5B6xfM5HjPhH5CQu474T1Uf4HFSSPjCcSj/GggyZaFrC+y/TCyJuOx4WcVtPukMbZujfGn aIwDVoMz3zvo5oZhDFo5DBYfkeAQZZKEBg7Bka4jOCQ6lC4Ki1+EEj4ZbMveOMjiNh7ji2JQLBiC Y6VDdOPx9QUIDnJlQbHerIQsPZipC/JTLBbe0RrolojFoq0RHCsdJDmDaAkJDg0b7H6SddBCYT3E J3nPDG0SOtpwEtqZIDhW+j0TvGEMbdCEdQq6Ew3ywxEcqzzkERIco9aDtlMdFtvRYbEdCQ7NOAdt BREcmnF7Ueutg0N8kBjhs8P8aIcdQjosqqbDXqFwDgpe5DDzwGHmgcPMA+eg2zcOC9rtsNiODovt 6DBHzGMH5wRHVhmPkRZ7Bi1SBAcZWLG2/4EJXDEv0mOGmcfCxBEcUdQeC6PiFWTbEBzqdwUtkR4L o+INdErtLWSYebdPlLl1bFrryD8WMlnPc0US32wY7MYwtTWLL8cwdUfzW4Fv4XGKiw3E47RttVrg cdJiy22uFcpevdnu2Y3H6VFx/s9f6r97d/5uU0bbeJxmMgV5nO6U7F08Tnd07UbKXrN5C3xPyl6z wrADUfbOm7OA3omyV3aYeSVEQ9nLulbsw9jrXnEZGXsl6zIDU0TZ1Thnu1BEaSbtgSKqTbiWImrG 2GtWE39yjqjb3fdtc0QVJjdMeFnywHNmeZ4VhdWGSy6lL+hH6n28hKeEZywNvEjT0rOycDxlhQzL OvYOxt72/2Tp2W0W38jPe8Mce0FrRvy1JZd6+PwsOTt5ntBArfthvNIRy3RS3/1HLKF7ekNe+32X hk2/vJrT+FLm4TKy1w6ukrwZe8kwneTn8+cNc1WaDKrqIqkuQ51QmrYK0+z/Iu3VyWAS6lGzSCex M5PvqHLfz2sXWXQjLW86HV+z8I67ZxfUvd9FEqsHyZuj7puj71vOYlKqk3BN2vv/7F1bc+M2mv0r rLzsbq3txh1E3np6emZck3S60s5mZipVKV5tbcuiRpTa8f76/UD5IskCTekwltVhVWaSVvOAIECC BHAuVKNq0thd1UljZDVdzBvnq+b9HXk/Yl+zZHZZPDQKjXtUpneYai7Xex1njadwTRdHNXiw+aXL uI38vZXfn+TRhXildaPrxXg+ouou3X7vvbd8Df6jpnOnxZhOsJjMz6I/Fc1VRk2LLxv5wbWrfkDV Udp4CkfpYjRemg9TmavuwtlVMpp4d2NqN9/iV0VCdczofUMvmaigz5eICojonLNFvWHg3J/Dl2EH cPLa/H1w8roraXDyaj6RhRSYkkVi8fXCOUxJQXiISSo1tEcslYSoaB4PMWl1DCmhcSawgRYYPJMX On/soDUC6TBminTaIisk0mFL4YpjTkIeDzFpObYargS2IO3xUJ69wJwUCA8yoRnGhBZYtDnhofHX 45HxRwkOLdIdnIkuBLQ+q7yUDcI7KKFcKQPxRZTBQsY9HlJC0MwZuv+thTYHPB4af0AllAKVUB4P Xb/DnDAJDzmBER57fpyFGIqEtxAXnwmo/TXDlFiaM0xKwTnkRKXB8UtLCX2/a4k5mXo8JEcgPNT/ hMfaT0Pfb34VFpIyKYwVT3hIya3b8lw64qH+U5iSWSsNORkSHnJiJjxEZ9HKQO8vXApGE0gEbzn0 /QNrsQgPPX/WYFo4pyEam8dD/Ud46PoJD14/NP8iPERKMoxBzy89fdD7h/CQkp7wkBODxyP3j8dD qi6GSWk9HlKFccxJxW/JYYpABn1/GIM5gZpYQtwwj9/h/g+wfJ5s364U0hPLR6hYtZgu9kAM6lZE ay1eDbdo58LaevBZolKziI4lzrUtw69QlWJpwsaL61QlZ8IfJt2oSvO/13J2nf0p+/vNnlQl36po 5Jyf4KBcpbbeDXGVGnJMAANwlZomAblKzeWsoDtxlYzPsxPSc5WkPtPd4+xWuEpKnokYSxenp0WI jSXnTnF2Tg1cpbsDt3KV3B1Vaf1t2jtVaUvvHTlVKdXMpalPFndO6DTPsjjJraQn27pYi1zEdO8p l8axyTJlXVIoa2RW8lTzRCZdqEp3f5iP60fa0l+rrTSltSYYTZ7yWB5oLq00pT8vlunca4hlwHdy 7dk1/qQ31ezzMhp8VEd5NbmLEp9H2TKAb3y7TBUv8mUY+T016JHX44PEZ9dFPmqqt8KaiZJZEU2T 2rOBqAZL8s4PTfXqs/MVVO2DzkfZVRPy7dPDPW8pifKHlqmXLXNHe0rKkipWR2k1v4ryURMTuKBD an/oSps3sX2bf0Ht3xubxy/7BoejvuPBD8Da2fh9kdI9s/CEouDjNdB5nq9NsPF6C+YDg8EIjwXL WQYZm3ONGctzDU1n4WAwOBgID9aBghkOH8wD0dE4zU6x82PLMVxjdDZuGGasT3gsGIdjwYhGQa4l Hg89/0ZD26keDz3/RmPBJITH+k9jwSSEB9sfCzY7fLDTYYOZYjCYKEaDhQ4ejAPROb6CYBmw/hrE 48E0KB5rv6MPxgGD0bBgHe5iSK0uGLYdTHgsGIVpiI7l8Viwj4HmLx6PBRMZyKjh8MFAmAMYHiwE BvNwu4tpwFM8NQAWrIMF0wshIDoF4SFbIo/Hzi+x+x8PVoLeXz6YCbt/FERnFT6ZB8JbsP0tRKcU ArOC6yHYCXx+HERHIjz4/IHBJITH5IRYsLDHQ/ePFND8Aw7WkhJrf4UZcApQjkith31/ER56fpSF 6EiEB9vfYvc/Gozm9QgIXjPs+28IVjv6YDUwGA0NNsPmX8cfjIbNf+BgNQatv4qYY8GohMeu/9DB cEOw2xDshuA5ZF3v8VAwGuEPHUwHBstBcohjD6YjPCbH/sMH2w12FJAdggXl4BbbP/J4yA7BYvtH hIf2fzweaz8sGo7w0PwDDmaMLSSnUk5D64+Eh/hTyhmIf6QcFjKGB1syaP8fTraEkyEPLCeGoyEt luwJR0tioRZwNCXh0WhMqP+tgsaPg8u5wXhKj8fSNbHQMzgek2ksJZEb6Pk13DFIDkt4SM6sMf4g HHKpsfUTPCQTmz9S70P89d2CLrariZ8qrFYK2S4N1kaGH9swZKdkgqdS4J3Pur0IsBb0xgwP2Z2L CI8a3YsIDvwdldXM9iHODt6+nYrQXMdhBkPnIsDmbIrAtO6+iPBQ1r0I7NbyRYQJCS8pl28dVfYp LDhM71yYjMPxvPsUFuz1nQtzLTfyPoX1dplKqLBt0e6Ftbz0di3MzyCCL4V9CuutN31hvbUZFRYm 2uxTWPDlsfzGaHGfaNYMIPeJ1q+21aCcWIafh3X3iViE50Hd3CfY4p/pl/Tjv378+adQQe3uE82I B7pPNC1z2l7K8/4TrWNvMCvHhVeS9/efWDYK5j+xvJwVdCf/CXXK4gu2zMqRZ0LtZ0Chz5zowX9i fSDu6D/REt8w+E/s5T9xr5B/cI/4hh6mjVbZz6Jio4OP26Ii1bxgTjkWu0QZLvNUqtIVqkwVszY1 Kk8kS7l0LJE60aXRhRa2NHSES40p19uiPU2HrrFDks5oUpTlKBvR7brFoeIuo2W0MchselT8z91h 3qdis4z6zhXCx79Mq7E/Ve3dKRaT4rdpkS2zbFZr0RhUrDlQjJYhNMsqZdVknjS5O3TH3T7xsWhO cRtdJ9Opj9M5oR/qepQ2CT6LJiYoZEhRTeige1eKxwv3B1Z0+tnNiK5gS67M3WWeRM0Ls4nCSaJZ Vc2jd2/vq77umvEj/WUd+eChj1U1Pomq2f1x9S2Vfr1W+pQO6c3rQrgWu6GX87rwY0P/DhibCTlf SW7NsrFeRVUemnq32vxtv9rQm2i9NsF+7MlzA92z23HNNLxoGJxdDouGw6LhsGg4LBoOi4bDouGw aDgsGr7woiELU+KfLBqGj3y5RUOOLhparuJeFg2DrdGyaBjsk24Lexxc2JO687qePRX6grNv6R/G zjgcgu2rvz6KdFrYi9ucHIeFvVe2sLfRwce9sMeSWOX0BZ5bJVLJXVkanUpT6jgV0uVWa5HFxmrL GU8yl7OkdHkuyyKzPNcb+wNPFvY27GbXXGi3L+vdRTMX0cV3n6jTZfS5uI0WU5+3HF3TPDS5fCYd +7yMvI1sPcoLX7ZfnPJFZdVkQhN9vy5W0xNZP55oywn82tv89Cqh466Sz8UyLduvwY19kLUftJY+ tiuFeu/YvEjyMb2sTh6W7OidUE3ms6qJ4Kb/rBfX0/u1uVlRN3fNY5x2F/fZtcW+uzbq1VH2ZMvP Rn29i29faWj08S++DYa3XfHHbniL1t+G5+Ed8ZDhGY8NZNjHYyg/fDCsPXbDWsxwpgfD0cGwFME7 jhleO35ww1TMMBM2XIUEex4P9T/hMcNO1DCWQ4aVfRjOYoahEhJ89WB4CxkmDoa3g+HtYHgL4QfD WwSPGr7ihqECM7zExs/jN/zExl8hHGaYeHjDSsjwBDashA0nUcPIAxs2/tENB3HDQNDwbzDsg8a/ r8BwDhk/ccO3P7ph22ENzw5vGGag9YfBMOvIDbOO3vAKNaxCDacObRiloPcfbBilY8jwVmsH9Z/H Q+1HeMjwiPCQ4RBqOAUbPsVY+xEeaT/Y8Ehj+yc+rxDpP6Ox/WtjGDT+Ex7aP4MNi4yG1n8ObXhk BTT82p78knbUMSktd1PN7K2W2klM0ou0aedr+900WpiQpr0tXlIFIzRNcvoq7BVLatqeo+WT2yIo aCbikKCgdSq/Iigw0prgJ9eGoECGN/deUFDQ3i7PCwqaSUIPcoJg6wblBHF4cRbzIJE9eJBIUKqg 470sSOIzJWGpwhP5WjcPklZ5xiBV4OxOq7ChEjiEVuFpDx+3VsG6tBRZmiZMS5OUmRLC+Eefp2Vc yqKwSa5VrDLBXVZkZSl0kWuhWMrKNMvy9U+SR63CpJiviRToz9E0yT4nl0VIpTCuJpfRuw9vv3/v afxTev03IoHvqurzYtr8vtHu6xqFn71TyFIhsAKJbkbzq6Wu4LKK/vzhUyMRGNOtcfJgGrLlvF43 MJ+NLqnBvG6gonGnOC1nRSN/eBddF9cVIZNJTn+bzZL6qje5wJagSf8zl6pljBj0Apv1fhUk/UEv cF+bYD/2pBcQgkPr7YSH1ls8HtrvlAIM2BQGC0jCrt/jsf06rqD9Nsmh/XYpNbTfKpWE1oslGNDp 9/ug+nsPSAjvHNT/hMfq76D9amkExHfrIeAK0jv1EJAF8V16CNiC+BrwfjXhofYH11ulEwbbbxdQ QL104P3rtIWef5QvAAeccWy/++ABaQwKCCE8xNf5GgLa0PofdcAbvf7A6w+vpXfDQwE7SmB6L8KD 7c/B9hdha7CX4CsJLGBKiRaHmi54ifEt6fMX4uspZTC+kYkhvr2yHBs/rIX4LnhAIHj/gwHDCuSL Eh4LKGM0AEN4ifF9GPb9RniI76ZZDL2/8YBChvG16PMRxEPzV92s7wJ4gkP3L+Gh9pdb1q13wmss 4E5xc1i+oYDWjwgf3sbtiEfPj/ElBRQQ5/HQ8wfzPRX2/CmNjd9KYwGfCtOrEh68fwzGFz14wKuE vr8PHxCL8c2HgFk0YBZ8f1jw++HQAbXWQPMXTdNH6Pl1Gto/wQNqGbT+TqPfYfniXEB+IYZL6PnB +epwQC4YcAvy1cH5vwH9nnbkmwcY223m7duJ0T1EJQzJE0PyxJA88WwRQ/LEkDwxJE/sXdiQPNGo PMJCkYZB8CLJE/SSC1/0hlAkbpGedBKK/Dt7d/7+H3Wl7Pehgp4Rivhx5XXE1baNcOHkifAQC0hF mkYBpSLN5aygO0lF9CmzF4J5qYgSZ3avtFptzzTvQSmyIX3opBSh10r4kgelyCtTimxU+biVInns kjJRmbWujHPHFcuSVAthdZJJlhoTxy51sZEFT02exbJMHFOGKRtnKtbJelusKkXe+Cq+obMVs0ky bv4kHtUjy383I8FvbxpdCbXA3y4uPr4RwcCLZFxWs+sijz69v7g4//DXT79+//Yfv/7lx7ffv//1 0/m/3kflzLfoeu9s0ZNM6W1FQ5sXlSzP+FDgsoT6JJrPkkk9rWbz6IY+xKPCX0WUTKLRpBxNRvMi GlfV1Ffwhu4xX9C7Hz5QCT+9vTj/4cNdKdGojEZzH4NR0Id4HSXBejdylsQn0i4aLQrrTXgSP91A 8T8b1pKP3bfuJPm/RaMf2AxvfTV6lI3fF+liMl/4CgefpUGosltVHvpgEKoE8INQBcQftVBFSez8 VkILxTDRECXq4USxwxK9BqLIYYkiJtbQ9aMbzQffKMaMIT0ean+rIGNwbTUktIA3qm2Mnd/FENHR bzSjeGyjGiNKeTy20a0gornHQxulTEHjh58Og+eHhBowUYAZiGgJEw2EgohCRmDBIh4P3f+SYfev xIROHg/1n8SESoSH3n8eD7YfJPQ0oLG/x0Pjh8SM0T0eu34DBavARB+zUzBTmKiyE0GhF2ZFPxSR 8OjXvYidSDrbiwAZN74IjN/xSugyvgi8LcKfZN2LAEk7vojgwPCiFJG+iRi9kQq8SD3cRHsU1p/7 ZN8Ukd64E76w4M29T2HB23yPwsJv0oMzYYJPwFYiRrzq2OlXmDHHzrY14svR/GqRLvkSyw3gy2pJ t6jf0Nw47OJxWZ3Oii+j4uaOcrEMal8SJnwZ//3GGk/72wvvNxGbAkyY57ZKIWk/0cpxz5d3TzVp u/SXo5ookGrS9H0PRJPgoBsmmoQZcxjRRLUSTZZbeOsfi5NqTi+3+s1Pnz6cxkqy0+C98jzacggt ILQMo7sRbBRIsIn38mLV7kw42QPDZv2O6sKw0c6KgWGzPHA7w0beMWxM8297xk5n2U50mz25NBt9 edxcmiTzbxaTlLnSiXCpKLMkLmSZc86ENULIQsRaikyYOGOF4tJK5VjqeJqXLF//1Hjk0hSTjJ4t umWSesLX/Fc3/mY7bab4rSG7fCk8E2XR9Lt3YP1pQv1FFR9v9MA6X+b9pKyoN6JkBTwe0XfPWhHR vIqm9B6nByCi2zD7TCe9Sha1LyW6aSg3dKCnyeRFMR3fnk6Kel7kJ/eFUt3q+WyRzRezou7PdHX9 eTmUuepgo/q0itFXTwmxzOyypRJe0totlGJ3jc0empo9NDR7aGb60MgoJlhYQjiolvYp4hXEfOhT zmXYGvSgwp5XLVJ5LbqSZnSEljNax9fHyT5NdlxYm7Y22Y+ZCpOAuk32nZq+r2aT6Y/j8/0n+zE4 2X++bbtN9oP3XWiy38yyoIlpmwxi+8Q0PuXyQgg/v+TuLFY7TEyZvOCeOvetFmcczgh5slbbMSOk hYc9zEtflfLjaQ8f92w1L0RBw6ONYyetKVVS8iyRiltW6qxMXMmdFnGRSyUKXeRFyVKTehW4cTq1 ciMvZU35cU2DyWZQyN2P2+eo2SwpaUIYFf4gmlpOF/O6dV76czEen96jlsfTZDLJrvwk8yNNNou3 eU6Tyfpk7U/f0ah30iR8NL/6ISG6oc/9KEnHhZ/F3seDPE6a3338aXU267H3QSHjcZUl/tf+5qz2 aYbTi+s1Xo0u47gnuIP84r42v/dce5BfDPKLA+eEYO2P50yAOREWor+iOQuKYfe/98mHfIo5g3Ia cJ98jvlswz7hWM4G7vPNMJ9qVH6E+mQfu3xp8KkGfaqx749BPnZon2ENvT/68PnF2k+D/Y/6/MbY +Xvw6cV8dnGfXEy+B/vEhqmF3fCwzywmHzywfM+6MLWtk/wP9bmNMZ98XH4I++Ri5z+wz60Q4Y2x bvI1BslfCA/Jd4QFrx+b/xnR4sjWTb6F+ezSBxjmM8yhnEbCY+2vJZRz4PEH9lmG5n+wT7OOoZxQ Yxh2/6A+0UaGuR1d8FZj96910PzD46H2p9cXdH4wp9N//u3QfgGuVptaJyg/3E07uIc+bg/v5z28 nv/gvLNnvJgHy/N9ihgsz/9I5MHB8ryvwgbL88MLPQfL86OlpjaTuRexPNfOddah6vAG5UvpUP3T i1ueO41bnreNIy2W58EWBJSoTaOAlufN5aygd1ZkyjMe7+F5Hn/L7JmzfVBfN8iAXSSZ9H8D9XV5 4CuSZD7tyyMnuZqS2azMEycTp5xhiZClyjLjciUzY5yyOhdZrnlaZC7jTFkZO1lKEVtj0o0WfSC5 brqXvxnlk+SR8Xrvfv7U5vzuQOqaL6NxcVlERZ0l44ZE2hBhRxMaH2ZFNo8+Lia3WZUX0ThJi/GK Z/lG96zzYn1bX1RvP707P2+oqhfVT5NRU0xJL4KGq/p4jvFt5Ee86ePJThtFaZEvT1pHcxoHo7xo CphX3gC9Kfq0mhC2OeYs+kvl6bLJ9XRcnDye7z9/+ea3yenp3V+c+lvtl2/+a+3cs2K+mE38SYrI v8GjX765O3x5dERvj6ti5ivhebcRvU6q2Vl0cTWqo7S4Sr6M6MwZ/cW4SPKlBHVLq44av/fLWXJd R4tGg+rP57shmibZ5+Sy2LiE5P746OZqlF1F02LmbefrleKzqyL7XEdUui9r2dpXVT1vruI68Zf2 v74PNy8oXcx9cV4+u615zqLzctnk9zWoF2ld/HtBoxi1V1ZNvhSzeb3tpHT1dw1/4j3nG9P60STJ PWCJXp63jpbvuIbrTMXc3x3b2r83UjMXsXt5u3m18Xs7Tfl5NvLG7w8u8bvSlPck4/7Nk3G/uyfj vnvjfzmn/7399sPGGBqtknHjM7H2l78/GffAZFZhNESm83hkMV0yg5FRCY9s5kvWMmPsiIfIoAwL DfZkXhAPbeZ6PEhmtiAeItNIqQVGBsfIULCXvZJQ6LfHQ2RqlMytMDKS1C12aB3xUPsRHtnM9Xjw +iEyjMdD9692Dnr+/LIOgjccIvMSHiIjSSMgMoo0BvIiJjz2/jMYmR0VM0gTQ174hMfEOAYjY6JZ ItIqiMzi8VD/xQ4ig0jHsOfPaez972KLjJ+KcYjMB4txmIPImIozSEzgxUDI89ODmAi6fwkPff8e XMzEHdb+gkFiDo/HspAwL32Ph+5fgXnxezx0/9LrH7x+bPwSLLyz3hEPjV+CSxAPzX9wMSAmhvV4 7P4RkJjj4GJGgYlBjl4MqQxEZvV4qP9Mi7NXF7zl2PhrLfT96vHQ++foxaiomPTIxZz0+Q2JgTgo BuMSyvLSAlt/0lJiYlaqPtT+hIf6n/CQmExqTMylwCzBQUx8YDEx9v7SChRzKmz9hfDY+IGLqaHv f60stH+gDcfeP97nG8KjYmiFtT8sRtbY/XNoMbDD9o89Hqq/09D6m2ESen8RHszCxNaPexADQ/2P i4mx/jeSg2JMMIsPF7NC+689iFGxLE9UjHp4MSmW5UnvD+j84Pyb8Nj9YzH+h7HY95+xmBkSLObd TQwbFrPulCc5iFkHMesgZg0XMYhZBzHrIGYdxKyDmLXPwgYxq//4O6CYtTU21oZbYTWN1Yrwg7Sh gm0rsa+Alm4Fjc5Pr//56ZyNF8F78Hk5bXtKSwc5rZ+q9SKnDQ4+wawXFl5kwOS0HAx25QaKR20a YwXdUYwrhA+ToX8EO1N8Fy2uveCi0eLqM3oWetDiro8W3WJobJsAedDicnGnxdWrWtyNhv49tLgb pzhuLa5NdFmIjNs0KxjPS65oVpOnhtHJ8yRROjEypQ7NbZmyWIo4EYUoucmznKlyY1NmU4vbxMxU 9dmPVTX/Nvp0e03D5+eorMbj6saLPr+sXbGPbrmpo3w0awR3t9F8lnyhTn8mDfUHn3k6+i2qb+t5 cV2fRNW0mPjSk6gcjQuvNU0m97WIRtfTGR0wG9/e1aMmYFOvRn75EB4TVYt5PcoLH4rjNZkNuElN 9X8qR5NkHE3/n71rbW4T2bZ/pStfMnPLlps3nVvnQybJTOaM8zixc+ZRqZpq6MbiWhI6gOz4/Pq7 dwMylgQCtWzLCalUJSXoNzR7rd57bZ6PCTwi82SGSVbLO3nxe5zB/6DqIAGEQNTAMfYX71DXYQfA qF9yshLp+jyFlkYfYAw/fHlWdu3ky7Mfnxfxozg4srzw5RnBBK9Fx+q/bmh9nsDK47ysjEy1t/LW 7BxV6rbs0/cVU/pdZH09XeZ+Oa3nfqnCTbfkfvHvXNTPRLPX3jRO3p6CXy3f7+M820y+9uKLAPD5 RrPTWDd+qLWKB2Uj2oa/FfGoFdBCPK1rWAMuzPM7Zpb02kJKuuGND+n0X2/+/nPunIW74w1TE2+o 0Jo9oI3G1W0W71lPi9bPpjd72/TeMfXPMa2k9wIwidXXpoeCFMEAtfdg068ElHS06dtwzGDTH1Zq ydUVftqWvhEaLvMsi8uISpfZthWywAA7X/pOIF0r4q4dGoHhBpwz3zeMyA5kwC1fiJCG0cqDVk8t uZpVstJwaUos+f7zKQlucmWav8braBifJsnlYv4xSXOUcPk9hpfpuj3dJE7mpuK3ojpzPgPr97a6 wk4GAHAVo57OdQx2OJ8VqSpRyQV2xRI5YBd/oD/uzS62vf3ninzolJDm41m/Q0rIISVk75SOjYZ8 cyhos0uEpiG/n0NvQ//Qu/nkrnsVmj4EWIXeifWBOABgFfpz0ZwlsnsVmm4IWEXj6f2Dwsx9Hy3v 7ZgUto1mD7BdKtvbyebeD733dhqMlTU+3LtU1viY71BZ80fg0c/2G9+AjUSLXyNa1BdWi2hp/cbW iBbPsbqeEDPanIjp4XSSLVeTaFEzo0+0ND93LUSL3uGp1Sbq20XJ2GFWZ6bFVUwLfWGbL2w6smxN omX95epItHgD0VLcuD+iZQcWZcPyPW0WRdoRk9LwqB0wJ7SMMPJNO7CF5fkSUw5Sz/Qo5b7BQysy XBlGUsiQRoG0A4+uwJ87LMrJFLaKu3LFxS+bSRQ8REwmyUUcwjWJdxIuRAoYCC6lWzWJP9aLKwIk I2GymAgS8kUmyevkjORj2J0v8Ihzli2m8uM4hfktOZSiCaUAfKdtgGGwAylF34R8+vkVge+EuT/R Wmrvn0dpPGD8Jg8SB/5iZ/7B8DytKHYsr6Wiuwf+RCcKpyf/sn4Qa+m1j+U1VVz1VFAtPRUfy3K0 VAC0VVBdvSgoy3Wb8Vyn8nopTXVVILVV9PRV6B5XBU5bxeqRVaCeuorTd68ipLf/QHktFccnr2I0 qMg8soqMtoqK1vuP5fXmz9Fcf1dr/9FXUfH0nh9XTwUeA5L0xq+XxcFxWfMJYMfyWv3XVqGxtVRA sbzW+/vYKjYe08J/yC9pjZ/pqfhjeS0VEkq1VLChvBb+24OKjaYKjamngmHazScpHctrqViYnub4 9fCfa/pUq31LU8VFMwsNyi/ote9q2b+u3RJr3U2FSE9FB8prPX/6KkhaKnxYXm/82ipKWvhXW0UJ M8XqlNdXUdLKooMqSFrzp6sC5DEt/IblteYfPt9a7fdTMdpUvs/6bfbfa/W2GFSQBhWkQQWpXxWD CtKggjSoID2+q+qggjSoIA0qSOtV7KiCVHNVNVponxUxI9ZMMD6Qq6p6RzRdVT1qUl1X1dZ3tcVV tXH+dlcgKqakTYFouxtsMZxa6d5usPbIsfvEG/vnhoUaQoY3oug/uyy4mxvs6mbXxQ3WwfPy5jEP brCGUbrBrty8fzfYteV72m6wLg+ikIWOYJYthB9ZoemykAU8cpnLLcuJLMdEO5KCSSxc15VcCuEE hkv90F7Zjm/dYKfxVNajifFf8u7Xd2+avGDx6xLGySKb3ByHKY9yKdT9ZCy5gCfy7pTfdYF9DbuE KMJ9t1RTjw2e8tkNiWdXUEIQOYMqpDi+hu0mIyGfVb6yRH7FTQ6sZPLq4+f9hRHTFrd2XffX1fje 70pfZ3CLfURavhct3UQLtqCBgRYcaMGBFhxowR2rGGjBgRYcaMGBFtxDZQMteDi0oG03v/YrEexm 87l/N1ow/y2z0mn4U/jb9a60IDzWutLkzKLtMoxdaMGW16tZmLz549WFusMm+1J38Nc8Ny2k7ixv xEyvM3fnKeFwhlqBlI4MU5O6W99DulB3LvXb6MqBujOskrpz6/Lf9xzOvmEtnzaPF4RuQB1KpeEw XwgTXjXT5iyibmBw07BCJhxmiZD7rmBCSE7dMLBdX8JEhHxFkvVuODt2cVUZsPzx80zOwvRmjjTb 2/Pzjycm8mYzWcr2VbLgE4lB5GvEXyup9ztGp3OSwa0yRRlsqDiKLxZg12Jl2WI+R43AxVoPjkic EzBhBSpnR/JaKRRmJEqTKZE8HEP3r2u9VJVJSWIlon1TMYRKUbsa0iRGWfB5KiMeyhH5BJW/VXzi eTyVySLH7kFHvs6hSikmN2SW5CSQipKcz6G0KILtRaLi6cdxtjci0WtLLjDE0dcnYyAMl2X2RBh6 1O2TjXgz4ddqmg8kWQd+hZq02R104Ff2yq88HDniHBtGS7L7R6UgDhpOHwoCVrujFgJu3V/rYvmO 00zn3UHAPuwVmmL5HbJ8tSNg9fBoO8Zsm9vtCLj1IW5CwArO7YyAiyb7ImD/2LDOTVOJ5ZuFD0pH BMyOqXWugPMLxxwZumL5629qNwTMBrH88sbDQsAra/m0EbCktuP4zGaRFwWub3m2KUPhO7CGzAoi wMGBMEzTY9KgMhKhazuCu8wTjFuAlu/uircIGPeYk1xO5xN4s+7A4JUrr9Iky47P4lySswLIYmKs VYE3pdDWCnw/pvKq8GA52lB0Ke42STKJqm2AO7OQzwFiPj95DgA3xXK8SswVJSnhec7Dy2MEtmmC DylBNz3EvfHs/wCEEJ4CbErh+VHgF94jaDjBfwGuA5idANQtdeD+ODvbH3JtocXuD7iu+sZ0BK5r kvmHB1w/AVR8dYK//IppnV68b4OK7si4X+D6aSWTVGtv/NXePBng2h/sDcB1AK4DcB2A6wBcJ4BG PaNZiGwFuNrNIf8PCFzbj10fDrg2PnfNwJVpiI8XTeoBV5N2P7q9B+DaX3wcsP4QdVHeeFjA9ZtS ImcegFDHMqXBbUzZxn0ecm5RDmDWkZHFfeEbgG4j16YAZJnjB74AyOrRwGVmyBuAqwppgAZPvk4n ddzaFIHxx7tTIqpIihQtXlxr+GmukiDDJtwZtqqADPlmIqd4bnqtAGsqM5mrU9WiRoCxgDVTpVKO LcY5AswZQj4Sxan8X3VOWqJdhKDqoBaRLJFfx1AIGx6OUdWFA0SjwzFqj7gJxty9oNGeQRA+dZtl QDeDPuqb/RDbd4559xLR0Hud1qFh73UbkPZjuuB3rMI37IPI9mYAjNtbFi6Dec3ilo/teL+/nh00 E/FteqTjZ1bPI73tQ12nNayuOdV82ua7/nC0xiEIVbQ+dy3n8Roe6UWTerSGZT4iq7HybnViNUxj YDXKGw+K1VhdyyfOakRgnwCspp4Z+K4lWGTbhohk4PGQhQajLnWciAsP3kAmmCsti/sht2yHwnyE Kwt1xyF9kdYIjeJfsvx9ndxQnt8w7wL2rDgkMPXzifyKUxHP1Ak7chPJZKGYjK4UhypyVVAlE2Vo qaqyusLEnKfIf4g4VRj1hvzwfDR6/iPJ5AUSIxmZyzRK0qkUJMtTLAGFr4qnMSN8JkiwiABhQhPX APhkRpJZ4cKe5XKu+rCYKDcDGMjtAPN4KuujxIrG8cWYTOUUe4HOASFXvu8JtIbKGCPyPrk+Wp0M kkAHedkuV370VY9KzmYm5NdjfJigqzy8zOELcVm5HeBYl0M9InIST+MZzwsPeLmlv1l8MYPtI+Sz fHIDXRKLUGl2rI5gf47096rI8f34I7zdjXNxRux7ZoD25Y/QbGoP3Mzgj3BQLMngj3DwLMChAPcH 9Eew3I6h5ADcm1PoPBxwd+lh+CM0PyqNwN1sfos6AXe35cSuE3C3+6hA7h24rySQ7uZH3zbkAbg/ GnBfWcunDdzR1UB6QhiGJwLBTTsyBAv9wPVMy3Qd6kWuIRw7dIRv2b7JMRdREAbU9LkTecbKjC6B uwrSTk7ySbamC1m/tNk3AYFmBKAaAOhv8ubzXMDbCWAwy/gFWN5tmP0toEkY2OXt7UckWwCE5tlt VUcEkDqgymt+owB8FguJceZwDzyPuTzm4orPEIEi6L7gqZhginTo5PVYAphNAR2Plw2NoVgg5azE tBiZDnB4luQj8hJj0gvYflRXq6xCzFF48lLKOeDmmfLDWB8tukEAvIaJQRRdhsfDb6pYSShgyUt5 Q2AU8RWvQXlFLNxO5uRmcKBQFw4GPg8OFMtrjwGfe2nlDfB5gM8DfB7g8wCfl/DZNreej1fw2TEb 5/oB4bN5IPC5cXVb3PkbH69u8NnUhM+O8zjw2Tl2DW9le94On51j2/CcAT4XN26Az3aVQ8EZrYRC 7xcxb1y+p42YGbWZIUxmB660qGtYjmlRJ2LcMU0Ka2YHgrqmCGHwTARB5IchM02HUxbYEXdX/OqX iLlCySU8/upQ9gLTEUwWAk+MF4FIpqhUhoA1T+F/OREJAEQUHwOMmadxmJPreCJCAKzk7OX7dqT8 cral8hgPgUOZ5upgFiBpOMY71ttEYLpAsKoQcr0DpBRWm0ge1asakZ8BIsuvHAEztnLbag4GUtWx rMC8y87l0OKoLIXP2W1f5imATSjN15oq3AHCCY+n1VE09Iz8T72eFlT87s3rXz/fzQjUiotN1rLp bMXF/c+Pt58Qm48GZTtGpj8QlN2tN3uGspug6KYdcqO95ZkW6xg+6bm23WhjdLO3/J//mfx0+VGI aSPG22brbBnZZlun/HLvaOtUTfaydQDymGiyFAmj4Ivo9ghdRHfoc9iqHfrC9kemqZcxquh/f9Ed 27LbDLzB2jHM0tpx79/a+aZ0diiT1PR8ajkhdXzpe75BqccYdZlrmb6wuG85UWBL4YBBZFsh8z1h 2CZjEeVMsi3WDk/DcQyb4X/j+Qvypp58qUzJpAyWQk81WIBxoVRWi0KFQ5oyVG5raTV6aveBySKR ns8Wc5keA/STHCMRwR5BMF5UrdqaXCSwZY+nhWkSo1FzlVyiLC2sShSnWV64tPGiNPZmdtvDDKn4 mRQjcq7iG+FaFd3IiViefmTl6YcaZzHwtQxXpZm0QNFZ8tevH6tG9mi+PCitX+WpWjVd7s9bbu3C o6r3dCbYH8xGemC6v/ye4wtyDTu72kzh45xnJyittbSV8AN3Qo0TwznRyWJveX2yXTVbav2Idtvw 7Z7nDPAtt61GXq25SN9WPKM540m3E4DeY1uvovdYN1eh24u2uXhI7h0fsD2F1m15WnepbE9hYlCZ bzSnodnKcOObrEVwt20FNbxluZ6x7b4Sb3kG1cRb/xZm/ue/3/nyc7gLv12utia/bbteO0XeDfL1 VqtR8EUL8vVXqzGOTb+EfJSN/O4qq3cRn+XqAz6rf1gXzNjgHVbe+NiAb3X5njbgcxwXc4swUxgC wFxkU8+PuOkJ4QouIukC1jOl9KhwjNDzLcHBevI97ppW4MCfDg5h615agHNuSCDJHPZs+AmdsQpi OZ7B644BVaRM/IFocAIYBAVSa65krbDv9SKtyN/z0zN4LKya/1YckSnGVSEVvewOOohlsmDBU7Xj ZAX+y+aA4Na6EsAuLngaQ8kfMBgqxkd3VjptvSqcvd7KySRR0U5vlilM3nzNwYxQblpV0z8eVax3 Jv+zwILNcxRIaKxoZK1LIYwQCtVwJ0rvQMUJYFUAl6qX6D1WOIyJOEONWUBkOCEcA+yuk/TyGCOv JqWmrJLvmVUysstuidvZXc5qT0y6MzNeYUijM4YMg8kxvOOA99MN5Q6dT3+/jcF2Rtadi/fKp/fv zc5YsSdKaxZuqPPpttExbh/59GZLcF/2XSfjqnlkLcbVzr4DVZN6xhXr7np/17hybT063T02PcPu 73tvu7RNt/97t66ckV+63hv3aVxtXL2nbVwFlsF9z/NCQwrDdJk0HC58m4eWS6n0Qos6HvWYGUVc uC6n1HcDj9nUMJyA2kETm76iTQ+dVmfvU9g+KpF3/IancbDIJeFlvjY0cvD0vFKUhxJ3Kmr3Hygr uR7H4RgNHpnmVcN5oqyB9ZaTiLw9f3dadCznF2t9qJkrStQeJejLXG3LQnAlS5TbPccO5/NjMJTi q1oz13E+VkWu+GQhyRf46kWpzMZfno3IS5UQ7pcPr9/89PkXsPFyFaK+dOLnQkhxpOYB2yuHsEgn RQePyvFiBwN0eyjS04H1pNLeqZuwvmoFqqkuHSEaFqSoM0rUiL88g9b+8eUZCW5uu9fQnX/QfZL/ Ogr934brwmGK6nfqDezbD25qbdqgN5tajkm7Ju31mKMZaflGjP/6T/ov5o8vd3fTbB1Zo6mFdsOO plbVZF9TywKjSblpgqlljAyvp+uCQV+Y7gvbHrm6YY6Yj30HU8t325IED6YWK00ts4epVW3US0Pp GTzw2tbYhgV+2taYJ01uW4HreEHIPcM0JLRqgMFlGJ7juZ4IjCgCe8yj1DUc1w9Cj/JAGphE1xB0 m29DjIrHMz45yW7gaz2ZnCxm8dcX5FOS5KNX42kilBlRfvazm+kknl1mBBPMJoXFk8KdrUbYhxk5 RVL/qLKRcp5eSFW+1goYVKkEiy7ExLRoGvGqNbQ8JgWBFKpbl3GDhaMDck0XYDllR+S2w6VMEKoE 3W1S8VZFxUcEHTILd4baTRNkyGCAGHG5HCHafLIYB4lUN5BnK6aMZNA/JQoUX8zgQ1e4hL48//vs z3env77/7e/3H37+cHr64XcSTfhFZZ3Vxq68PNDZ4iqJxXLYecpReIlPRndWYyxDWIA4z6r+lgQb WnHojTFDXWuwvpRcE5FpiiTanYmPs9rkTm6w1Opwj0hWmMfxdA4VYxHUSsoLSzLk2GGYt5LXu623 aQkDmV/LcprVCFQ/lwu5RwOxTTvv4SSTcCf7plxDTpdCSm/rBFuVSmmLa8iK1bdxsg6iK/3N4dve nFa9edW1N97Iv1dzeM29xGBWH/+StfImfl31/FM2faFrlWx2A7Edy2n0OGgownyrXwpn23Fsr18c rO24Vu9WoEi/AF1VpOfwsUhPJyAsohlte99eLt2qgKfF1gvY7f/Abaii7wO4XkXvB3JDFX0fnc1V NHvoPFm/ozbvnr7BsGCne3uTpwaEbTVLqPcNbW7bbHepbG+hzZZl+HsTaLfgbWusbKuvlvosajlr tX4YaxSTa3kdo5E932Sap3l08WdwFXz869Pvn3fx1iofHl1vLd9j7YJg22mu1oe4keZCzmZnmqto si/NZR9T/5wWNJc1st3uNJcHewjSXPCX0pGxF55rhwgd322THh94rsPiub6pGB7my8h2Td/xAylM afkyDGjEIidwGIdPP540Rsz2GRM8skPLtH3hUEd6whMRDVYOeZc8VxVNk/O0LvL1PCO1K2TOw0t+ IZu0vmoBLsdhyiPkQcrirdQXVD36JLkonZJKjegyR7byx8KKpph2DJsspaQVIZWWGa5bG68re3My QRqGzBbTANrDeJ05TzOJ6mDqeE8lXytOUJGG+fIsmQjyy/vP1X1TPv/yjBTeVgcTrrO/uJzNlEx/ 0mURwGotHubk71QjIMfZEpCzC+tyT13Z2yFk1ZvTtt7Yj+DvtWm/brYQGy3ZFQvRopp5WjpYiN2s s/7O9GhpaFln/Z3p71pnXo/46fuwzvrnfoU5G9RiyhufgHX2TSWE9YUBRlcUCcYDaoW+Y3omM2wq ImmH3ODUFD5zeCQs6nrMdDmHW2zLDqXLQ0eaDdbZik9YZZ+t/PwqxVjLM5RaPSvMK7B3fvjj7OzH Qol1Ok/xKKlwWvqaF2lKAJNWeV2XOUDA9PnnGakqJjCBMJFb/PJflZVe80rGBdua3BS1oi0WYvdu a4WWZ+FYZirjyKbmjlR0tXJHT8gcysYBnh3eBhFUDlpgw1V+WEVseVXzNZ68oV/XiLwUIsauwltz c9tU5dKFx3ANgyYiFhsGpGzD2uTd6eyGPgZSGZ/z+SQuPONkJovUvFWi3MKlv0uPoIWgFk0BXcKz y+myg6X/3Z0uoe/dVe2NiGVb/pUDdzfb2Ua9f1P0Pp3Qepqigz+cvinaaHytmKLb/OaWpmhzgPQD m6K90/Ups0rLFO2fru+uKep3T9e3d0uU+abV3xJlreToYIkatDRF71npf8PyPW0705T/39619rht Y9H57F+hzn6YBMg4IkWJ0gD5sG26xSBtkyYtsgukWOhBTYS1LVeSk3GB/e/LS1Ky/JAsWZpXlxeT jEeiSUokpcPDw3ttzw1CQgPHtu0QmQyTIPY9HFs2jkJiExcHKPZQHMWOZwcR9i0vNDFG1IzZ7ibX YziTX2UXqLkFMvkFb9BPuX8BFP8KqrWDyS0ZrSE6geFvARgRAlBKxIRAX2A9wGkbVLTZCyFEWyns hfy0Mjkuh8qITyDKurnI1WFVNXWm0vRPjWspDINrqpCmUHWl5W6Jv3/47voa6lOwfMlRIfCmQFyK 7128ujCy1YLVaiKoxbZC5dZReQmZ8VVAxJg/xaHMPJknM38D9Yyk2AV7jwXcPZroeo/LX5AGYh2V VBihRmzV7fvN8oXOQHDv7VHL5LBIyKEuHqjw4VmYA3VGDjh4G57FQJ0RZDFQ4QNZDJQqQYsMvxcj NGq/rnQ4ixHuxUC1E2TRrL65N7WTDEowmkAJMmtUQ/TOzHPReAIlntlo0il4oI0W4aD16XhKZuPF q/BauvkJmTUrBPtmRjDxRmtNgu3mBa6Doi63JuoSb9hhoq62d2ydJ/G8ri4aXK/FWVcnnuSP8Lvr 7/+Zp4T+1JTREVEXdOuBoi5xZ4aKutqGVyNX4zULwTpwNaLIvlyNfWlS5SaCWFPs9d66SPCV5U4J HeomYv8Z0o2roY7mamTCB+dqdprvaXM1ge1YruP7mIaYmQQRhhizTR+HYejbgenZmEQxcUhAAieK YodEUUSZjSzbcS27KSrjghUvoYriv1WRzDZ8zYFT72GKnrN3WXq7hhW3rz64wfqcRLxHGn+sWLY2 ln7G71bBk70wlilwEolYLQvWSz/PgVoop0eCBMnSIyuCW0XK3ZGiWFFcIta9KlXWpmzB4XxJYLFP 8Cvv2dcMaKaYP9jF0tjU+AiLfMJdg9o+t5vmBSyI+cbrJBMz4Oq44ouEjis/cNVbNc79RVIkf6pt iqruDHYN8i+Co+iUf5ynX9heRqqY0h2GKC4CFxCrbDZ9B3/9At+YbpdXBcAowJPZJtmF2lNY7dFM +SCuydf2Sn+mWmcmC1XuMV7xwuf+rUgtEuevfn4u/XvweTBI62r+OpYsgxK36ieDV4oG9KubsCWs 26lJ5WR7p0XhInZbbCuKiKS+ZG6fzn306hbIMez4uPw0nU7VIcSfeGZ5WHbnV+tP51v9DXLb1OrT eS2Z3DPLDwiWspARRkR6uGZ+6/cq2o9X0/xY/WY8dR9pDx5z5PHyYz236ml+TPNjmh/T/JjmxzQ/ pvkxyY+JaXxXfowMDME6Cj9mPxJ+rHFENPNjtLGvduPH2hTmXfgx61R+jI4hZdrxo95NyqR91KuE D06P7TTf06bHPCeIvdghYWwSZHoOYhF/ugWmGQBjRrDnRMyz+BlCmBlGLMCO6dpWEDqBS+OwKQRr H8l8DjP8vNIxRSu2reKWp4R6qZQ0tTJf16A0KrJVLhxBVWLzVfE5zSrhkOQVSkWS/P5Gk7TFrCwM ngdvkYtivWQXdR+sQHPtHZWs2GJPk6QkRbd8EifYMCVDqiooRUZ1XbmSF/mLtRHx3mAAESgEVYrG OVD3YMbfgyOqjxBqoeU1vVK/GVp+VH3nr0av9HRrpOkVTa9oekXTK5pe0fSKplcUvYKQ21l+1OLg /P7oFeeR0CuN/a5FfnRylJqyyGH0CjmVXvEGBn2X46G//Mh2bS0/UgkP8ytY8Su2+E2n5mUW9nJh cALZcqAtnzbZQt3YNF3khKZLA4SigAbMiYMgIJFFQoR9ZuMosgNCHeRF2PYooa5rulFgkXjXh+vh gIC1TWP1w9cNAeoWtRB69eh6rQRLFXSw1NoINdDhjNT2+gA26vNHdLpYz5M/pTRGEhrgIEqFyDPS AKK3s6zkgPx6XdPYWGbskhfMsanxH7Y2EjGgYDN96S5qtWBVLUJZi89QixEZEcxft/fGiPR3IPUY t+E/XknJg4fdOxBbnTQ7Xu7KWew/OI9yFq5Jm30Jd4gjzWvdinXqUND2nI5Q0GtxqdkNCr7N5r98 /+9/Le0PJwcLPHI/m2CYwBQnwzBZZF8YRmHHPqJX/IdjT+J1dx7lXSJHADgCEWyGy8Ch/rQ/DjMt 7dpTJTyMw5DCYeTuoRf9K0GvMLCYa4UUYgA6QRzQmLjU540V2CGLXGZbMfV5RZiJPNM2CfZdRqkT ORC8mXhN0Au03uB4aZnx124dfcFvY+fsT0lebhOXEU+qwL9i7/5iuSpUSGBo8jYE9lE634QAKlVu +YuNSrvEQ1sVKP2HblaaZquIiSAtsmw/B6FuFYBG5KpkyfPk5nOhAgmCj9AyirFYqJNBjDcXV27S 5ydlzaQWGRTZolk4OOMXP0tvblh0N4reXfBTAay9E70B1vZS1HBPneVS1MPBsp+7Lf7cEyw7rTZ3 Dcuw10JK9YFljZxwAywjuB+S2yfrR8riUfD94JBkNIa4tTmOQl7RI4ZRp219qo6XqdkZL1uk2S1r J7xcvMmtbB5+G775ehp1Ku7qUOr06L3thtkbW7cRs5vNwWW6YXbaG7PzH/wrtoA6tdypiXs5fMW/ Iq90s4WHB/i2TvD36pjay5ZK+MABvndb72kjdteyCEXMNClyEMO2TQLLx9hyLAchH4e87TyMTBZZ hLh2FKIwDmnkhRGsu3he08bNNL8y/pHMGDCiYp+aUlmJvWe+CGII6Fn2JCPl0DeNVrN2D/pvF8Zv i+TWACEX8Kz5C+kKlQ93IWmrhfDOpSOsSG2OXBsrsbcTaiSc8L9OMgDH1d88oYx3KMA+R86bugsU n5X4hmcaw266MkiiAvHl9UiiFo5ABsJPbLpkC+UNtYygKL7E4X7p2GoroGLp7B/cd0k9WpZ+SSJJ 5s74EFEb+PyMY9OMd19DBBIQ8xG1exJqmK/zgs2FTA4CdMrNjuI2lQWATI+jJ/govrAf57J50vDj 2489uNyWJ9dgp/+ocUqxe6aaNOyeqEA1ahw7dxA78cfeLCke3XX+qLUZeXLQGL+7GTvU0BzxSNeF 8DaVzf3G726+spb43QN8YMgi+yKp7fjdPTznb4fv3l6DVp9+n5R//T757+RsTLvhT8JVAG14maUB ywr+j78PL2/SS7F4dSlubJ6FL08vA/ZpU9uG34jaZv13aWfIxhCqmZrEOjP5m9XCZ4Y93mU22yqH 6DbGWV6sAOc1pjt2/ola9/bnA42jkVPKgAZ2CGluf8faaX+MLHpmmGNf7CH7P29/CTCNTTd4+V50 g29FN3hZdoPJ5CY15Kxj3OePtoe17uN/7icL/rY9oYxj4x9hvDv+Lf5K0OP/HqxcloHWnfCJWJoV xnk8L84nsJ4jDj97rqAzPzx9BxPl2eLZuRTYfEwzcE4MM68f0m/On48NT7TdsXUf/zBBnrNpcdt7 DBwd/46zM/4t07L0+L8PE0P3w+s3V+INz9GAkSyS4jgeUImLJFqXn78I0mIy+Ztxk14ZixSi4/BD fIoLIrkirRJM9DNCmzZt2rRp06ZNmzZt2rRp06ZNmzZt2rRp06ZNmzZt2ka3/wH//+3/AAAFAA==