24252 2010-10-10 04:29:50 +0400 old version 2010-10-28 00:13:35 +0400 1 1 4 Development Sisyphus libatlas unstable all Linux CLOSED FIXED P3 normal --- 1 pip nobody led mike real.altlinux.org viy qa-sisyphus oldest_to_newest 113457 0 pip 2010-10-10 04:29:50 +0400 пора бы обновить. Скорость линейной алгебры в numpy зависит от этого пакета. В версиях > 3.9 поддерживаются фичи core-due и выше. Это есть в debian, ubuntu и gentoo 113509 1 real.altlinux.org 2010-10-11 08:31:16 +0400 Вот тут: http://sisyphus.ru/ru/srpm/Sisyphus/atlas/changelog сказано, что последнее изменение было сделано роботом. Попытка обновления с таким изменением вылетает с руганью на прикладываемый патч, который был создан роботом. Надеюсь, этот робот знает, что делать дальше... 113510 2 viy 2010-10-11 08:48:27 +0400 Евгений, вы совысем совсем не подумавши баг повесили (ночью спать надо) робот там не при чем, он пока патчи писать не умеет. Это старинный патч, его делал Алексей Турбин. то, что название atlas-3.7.11-alt6.qa1.patch, это потому что в git rules и в спеке так написано. а не прикладывается, потому что к старой версии написан... 113511 3 viy 2010-10-11 08:50:11 +0400 я тоже, кстати. Это же не мой баг, почему я его открываю... 113512 4 real.altlinux.org 2010-10-11 09:10:25 +0400 Извиняюсь, коли так. И я не весил баг, я на него отвечал. Предварительно попробовав обновить... Вообще, меня там не только патч, меня там и сам спек некоторым образом испугал. Так что взяться за обновление что-то нет желания (и так дел по горло). А пакет висит на @nobody :( А это очень плохо для такого важного пакета... 113754 5 real.altlinux.org 2010-10-15 08:07:49 +0400 Похоже, желающих помочь не найдётся. Попробую на выходных поковырять. 113755 6 pip 2010-10-15 08:37:28 +0400 > Похоже, желающих помочь не найдётся. > > Попробую на выходных поковырять. он собирается только вот в пакет запаковать хитрое дело. 113756 7 real.altlinux.org 2010-10-15 08:45:16 +0400 У меня он пока застопорился на этапе make (т.е. патч я уже поправил и добавил вызов configure). В общем, пока дело двигается, но со скрипом, так что возиться буду наскоками, ибо полностью сесть и начать его ковырять непрерывно нет времени. 114110 8 repository-robot 2010-10-21 18:07:29 +0400 atlas-3.9.25-alt1 -> sisyphus: * Thu Oct 21 2010 Eugeny A. Rostovtsev (REAL) <real at altlinux> 3.9.25-alt1 - Version 3.9.25 (ALT #24252) 114123 9 pip 2010-10-22 05:35:02 +0400 (In reply to comment #8) > atlas-3.9.25-alt1 -> sisyphus: > > * Thu Oct 21 2010 Eugeny A. Rostovtsev (REAL) <real at altlinux> 3.9.25-alt1 > - Version 3.9.25 (ALT #24252) видимо надо и numpy с ней пересобрать, а то object? -> Details about 'object'. ?object also works, ?? prints more. In [1]: from numpy.linalg import * In [2]: from numpy.random import * In [3]: m=randn(2000,2000) In [4]: %time U,s,V=svd(m) Ошибка сегментирования [vv@tao-vp ~]$ 114126 10 real.altlinux.org 2010-10-22 07:01:08 +0400 Не воспроизводится. Приаттачте какой-нибудь скрипт, попробую. 114128 11 pip 2010-10-22 08:07:44 +0400 (In reply to comment #10) > Не воспроизводится. Приаттачте какой-нибудь скрипт, попробую. На 64bit не воспроизводится? Возможно я не все библиотеки обновил, полный дист-апргрэйд у меня удаляет некоторые важные пакеты из x86_32. Может что то не дошло Я обновил libatlas+devel, liblapack+devel,liblaspack 114131 12 real.altlinux.org 2010-10-22 08:52:10 +0400 "На 64bit не воспроизводится?" Угу. "Возможно я не все библиотеки обновил, полный дист-апргрэйд у меня удаляет некоторые важные пакеты из x86_32" А это уже другой вопрос. Сейчас идёт процесс массовой пересборки гигантского количества пакетов, поэтому лучше переджать, пока всё устаканится. А точечные апгрейды, вероятно, ещё больше всё усложняют. 114167 13 pip 2010-10-23 09:43:59 +0400 segmentation fault, видимо из-за того что у меня еще интел фортран.Я отключил переменные ifort в bash_profile. Так что с этой стороны все должно быть чисто. Вы бы не могли показать результаты from numpy.random import * from numpy.linalg import * m=randn(1000,1000) %time U,s,V=svd(m) у меня выдал LinAlgError: SVD did not converge. Сейчас попробую пересобрать libatlas под себя > "На 64bit не воспроизводится?" > > Угу. > > "Возможно я не все библиотеки обновил, полный дист-апргрэйд у меня удаляет > некоторые важные пакеты из x86_32" > > А это уже другой вопрос. Сейчас идёт процесс массовой пересборки гигантского > количества пакетов, поэтому лучше переджать, пока всё устаканится. А точечные > апгрейды, вероятно, ещё больше всё усложняют. 114191 14 pip 2010-10-24 12:19:57 +0400 > "На 64bit не воспроизводится?" > > Угу. > Видимо системы разные:-) Сейчас поставил обратно старую версию и все работает. А новый пакет у меня даже не собрался. Ладно подождем пока кто-нибудь из юзеров нарвется 114208 15 real.altlinux.org 2010-10-25 08:38:46 +0400 In [1]: from numpy.random import * In [2]: from numpy.linalg import * In [3]: m=randn(1000,1000) In [4]: %time U,s,V=svd(m) CPU times: user 4.43 s, sys: 0.09 s, total: 4.52 s Wall time: 4.52 s "Видимо системы разные:-)" Я ежедневно обновляюсь до сизифа. 114216 16 pip 2010-10-25 10:40:09 +0400 (In reply to comment #15) > In [1]: from numpy.random import * > > In [2]: from numpy.linalg import * > > In [3]: m=randn(1000,1000) > > In [4]: %time U,s,V=svd(m) > CPU times: user 4.43 s, sys: 0.09 s, total: 4.52 s > Wall time: 4.52 s > > > "Видимо системы разные:-)" > > Я ежедневно обновляюсь до сизифа. Сегодня специально не пожалел времени и протестировал на виртуальной машине sisyphus на i586, и x86_64 на первой ваш libatlas работает и показывает прекрасный результат в виде 2.29 на этом тесте, на 64bit не работает, svd выдает диагностику что процесс не сходится, самое печальное для меня что eig не рабочий тоже. Ну да ладно хорошо, что хоть у кого то работает. Удачи! 114224 17 real.altlinux.org 2010-10-25 12:09:47 +0400 Может быть, я всё же позволю дать совет (Вам необязательно ему следовать, но обычно он помогает): сделайте dist-upgrade. Если часть пакетов при этом вынесет, их всегда можно установить позже (одним днём весь сизиф не пересобрать). В любом случае, точечные обновления, как не раз показывала практика, с лёгкостью ломают то, что до этого прекрасно работало. 114261 18 pip 2010-10-25 19:02:08 +0400 (In reply to comment #17) > Может быть, я всё же позволю дать совет (Вам необязательно ему следовать, но > обычно он помогает): сделайте dist-upgrade. Если часть пакетов при этом > вынесет, их всегда можно установить позже (одним днём весь сизиф не > пересобрать). > > В любом случае, точечные обновления, как не раз показывала практика, с > лёгкостью ломают то, что до этого прекрасно работало. Я так и сделал. У меня стоит интел фортран локально. Для того чтобы исключить проблему интерференции, я использовал виртуальную машину на которой был сделан dist-upgrade. Я пробовал i586 и 64бит. В первом случае работает а во втором нет. Процессор, интел core2 duo p8700. Кроме того, гарантированно работает старая версия пакета. Таким образом, врядли проблема состоит в моих локальных настройках. Предположения два: либо у вас локальные настройки, позволяющие держать две версии библиотек (это легко увидеть если посмотреть что использует нумпиевский lapack_lite.so), либо эта сборка не работает на моем процессоре, 114280 19 real.altlinux.org 2010-10-26 06:58:42 +0400 Я на всякий случай пересоберу NumPy, SciPy, да и matplotlib заодно. Всё равно все пакеты, использующие ATLAS и/или LAPACK, пересобирать придётся. 114281 20 real.altlinux.org 2010-10-26 07:14:31 +0400 Кстати, насчёт виртуальных машин, Вроде бы Вы ранее сообщали о нехватке памяти при работе с NumPy. Есть вероятность, что причины идентичны или каким-либо образом связаны. 114283 21 pip 2010-10-26 08:20:19 +0400 (In reply to comment #20) > Кстати, насчёт виртуальных машин, Вроде бы Вы ранее сообщали о нехватке памяти > при работе с NumPy. Есть вероятность, что причины идентичны или каким-либо > образом связаны. Было дело с виртуальной машиной на для i586. Тогда я памяти добавил и проблема ушла. Наверно я сумбурно объясняюсь, но чтобы небыло непоняток я подитожу мой опыт. 1)есть ноутбук - macbook pro5.5, cpu p8700 2.53Gz, 4G ram. На нем сизиф 64 бит, последний полный дистапгрэйд. Линейная алгебра не пашет, не сходятся ни eig ни svd. Я подозревал интерференцию с интел MKL. Однако, приэтом нет проблем со старой библиотекой (3.7) 2) Для исключения подозрений попробовал поставить сизиф с беты на виртуальной машине. Памяти ей дал > 2G. Имеем, на i586 все работает на 64 бит - нет. Для полной проверки нужен чистый пример без питона, просто фортран и слинковать с atlas 114284 22 real.altlinux.org 2010-10-26 08:35:02 +0400 "Для полной проверки нужен чистый пример без питона, просто фортран и слинковать с atlas" Жду :) 114320 23 4625 pip 2010-10-26 22:59:07 +0400 Created attachment 4625 лог ошибок автоматически создался при компиляции atlas 114322 24 pip 2010-10-26 23:04:42 +0400 Лог неудачной сборки приложен выше. Лог сборки rpm обрывается на execstack -fPIC -m64 -fPIC -DDCPLX -DFindingJITCPCE -I./ /home/vv/RPM/BUILD/ATLAS/BUILD/..//src/blas/gemm/ATL_gemm.c make[5]: Leaving directory `/home/vv/RPM/BUILD/ATLAS/BUILD/src/blas/gemm' gcc -DL2SIZE=4194304 -I/home/vv/RPM/BUILD/ATLAS/BUILD/include -I/home/vv/RPM/BUILD/ATLAS/BUILD/..//include -I/home/vv/RPM/BUILD/ATLAS/BUILD/..//include/contrib -DAdd_ -DF77_INTEGER=int -DStringSunStyle -DATL_OS_Linux -DATL_ARCH_Core2 -DATL_CPUMHZ=2520 -DATL_SSE2 -DATL_SSE1 -DATL_USE64BITS -DATL_GAS_x8664 -DWALL -DATL_DYLIBS -DATL_NCPU=2 -DATL_TRUST_OMP -fomit-frame-pointer -mfpmath=sse -msse2 -O2 -g -Wa,--noexecstack -fPIC -m64 -fPIC -o xzdfindCE zdfindCE.o \ /home/vv/RPM/BUILD/ATLAS/BUILD/src/blas/gemm/ATL_zdFindCE_mm.o /home/vv/RPM/BUILD/ATLAS/BUILD/lib/libatlas.a -lpthread -lm /home/vv/RPM/BUILD/ATLAS/BUILD/bin/ATLrun.sh /home/vv/RPM/BUILD/ATLAS/BUILD/tune/blas/gemm xzdfindCE -f res/atlas_zdNKB.h /home/vv/RPM/BUILD/ATLAS/BUILD/bin/ATLrun.sh: line 4: 2033 Segmentation fault $atldir/$* make[4]: *** [zdRunFindCE] Error 139 make[4]: Leaving directory `/home/vv/RPM/BUILD/ATLAS/BUILD/tune/blas/gemm' make[3]: *** [res/atlas_zdNKB.h] Error 2 make[3]: Leaving directory `/home/vv/RPM/BUILD/ATLAS/BUILD/tune/blas/gemm' make[2]: *** [/home/vv/RPM/BUILD/ATLAS/BUILD/tune/blas/gemm/res/atlas_zdNKB.h] Error 2 make[2]: Leaving directory `/home/vv/RPM/BUILD/ATLAS/BUILD/bin' ERROR 843 DURING CACHE EDGE DETECTION!!. make[2]: Entering directory `/home/vv/RPM/BUILD/ATLAS/BUILD/bin' cd /home/vv/RPM/BUILD/ATLAS/BUILD ; make error_report make[3]: Entering directory `/home/vv/RPM/BUILD/ATLAS/BUILD' make -f Make.top error_report make[4]: Entering directory `/home/vv/RPM/BUILD/ATLAS/BUILD' uname -a 2>&1 >> bin/INSTALL_LOG/ERROR.LOG gcc -v 2>&1 >> bin/INSTALL_LOG/ERROR.LOG Using built-in specs. Target: x86_64-alt-linux Configured with: ../configure --host=x86_64-alt-linux --build=x86_64-alt-linux --target=x86_64-alt-linux --prefix=/usr --exec-prefix=/usr --bindir=/usr/bin --sbindir=/usr/sbin --sysconfdir=/etc --datadir=/usr/share --includedir=/usr/include --libdir=/usr/lib64 --libexecdir=/usr/libexec --localstatedir=/var/lib --sharedstatedir=/usr/com --mandir=/usr/share/man --infodir=/usr/share/info --disable-dependency-tracking --without-included-gettext --program-suffix=-4.4 --with-slibdir=/lib64 --with-bugurl=http://bugzilla.altlinux.org --enable-bootstrap --enable-shared --enable-__cxa_atexit --enable-threads=posix --enable-checking=release --with-system-zlib --without-included-gettext --enable-multilib --enable-languages=c,c++,fortran,objc,obj-c++,java,ada --enable-java-awt=gtk --disable-plugin --with-native-libdir=/usr/lib64/gcj-4.4 --with-ecj-jar=/usr/share/java/ecj.jar --with-java-home=/usr/lib/jvm/java-1.5.0-gcj-4.4-1.5.0.0/jre --enable-libgcj-multifile --disable-libjava-multilib --enable-java-maintainer-mode --with-arch_32=i586 --with-tune_32=generic Thread model: posix gcc version 4.4.5 20101001 (ALT Linux 4.4.5-alt1) (GCC) gcc -V 2>&1 >> bin/INSTALL_LOG/ERROR.LOG x86_64-alt-linux-gcc: '-V' option must have argument make[4]: [error_report] Error 1 (ignored) gcc --version 2>&1 >> bin/INSTALL_LOG/ERROR.LOG tar cf error_HAMMER64SSE2.tar Make.inc bin/INSTALL_LOG/* gzip --best error_HAMMER64SSE2.tar mv error_HAMMER64SSE2.tar.gz error_HAMMER64SSE2.tgz make[4]: Leaving directory `/home/vv/RPM/BUILD/ATLAS/BUILD' make[3]: Leaving directory `/home/vv/RPM/BUILD/ATLAS/BUILD' make[2]: Leaving directory `/home/vv/RPM/BUILD/ATLAS/BUILD/bin' Error report error_<ARCH>.tgz has been created in your top-level ATLAS directory. Be sure to include this file in any help request. cat: ../../CONFIG/error.txt: No such file or directory cat: ../../CONFIG/error.txt: No such file or directory make[1]: *** [build] Error 255 make[1]: Leaving directory `/home/vv/RPM/BUILD/ATLAS/BUILD' make: *** [build] Error 2 make: Leaving directory `/home/vv/RPM/BUILD/ATLAS/BUILD' ошибка: Неверный код возврата из /home/vv/tmp/rpm-tmp.70936 (%build) Ошибки сборки пакетов: Неверный код возврата из /home/vv/tmp/rpm-tmp.70936 (%build) Зачем на 64 бит --with-arch_32=i586 --with-tune_32=generic?? Ошибка похоже исправлена (выделил ___) в версии 3.9.27: ATLAS 3.9.27 released 10/20/10, changes from 3.9.26: * Fixed several bugs to allow L2 BLAS to install using a low-res timer * Fixed bug in x8664 kernel description causing seg faults for _____CGER_____ * Fixed bug in r1hgen and r2hgen where first kernel's minM was ignored * Fixed bug in ATL_her/her2 where j-loop max was N rather than NN * Fixed bug so that h2gen.c generates ATL_GENGERK as a function that can handle all operations, not just least restricted kernel. * Fixed bug in ATL_syr & syr2 where nr computed incorrectly * Fixed cblas_[nrm2,asum,iamax,scal] so that they return with no operations if incX < 1 (this matches f77 behavior) * Fixed bug with extra spaces in configure's OSX libtool finding script ATLAS 3.9.26 released 10/18/10, changes from 3.9.25: * Much improved GEMV & GER performance for x86-64: + Addition of SSE/x86-64 GER/GEMV generators + Complete rewrite of GEMV tuning infrastructure + Change to GER kernel API to minimize parameter passing + Arch defaults for Core264SSE3 & AMD64K10h64SSE3 updated * ANSI C code generators for MVT, MVN, R1 and R2. + should improve non-x86/x86-64 performance * Started rewrite of all L2BLAS: + GEMVT, GEMVN, GER, SYR, SYR2, HER, HER2 built from optimized kernels - TRMV, TRSV, SYMV, HEMV just call reference implementation * Bug fixes: + Fixed kernel testers & timers to correctly handle alignments, particularly ALIGNX2A. * Basic support for POWER7 + VSX detection + VSX GEMM, GEMVN & GEMVT kernels provided by Mike Kistler of IBM + Arch defs * Fixed it so GEMM kernel files use $(GOODGCC) instead of flat gcc if 'gcc' is specified (so they inherit flags like -pg, -m64, etc). 114327 25 pip 2010-10-26 23:23:19 +0400 Пока не поздно... Может быть для 6.0 лучше остаться на последней стабильной версии 3.8.3. Нет гарантии, что выбранный релиз будет работать на новых процессорах 114339 26 real.altlinux.org 2010-10-27 07:12:44 +0400 Последние сообщения - это Вы с кем говорили? Не со мной - явно :) Я всё ещё жду ответа на #22. PS. Нынешняя версия ATLAS в сизифе вполне стабильна (проверено как на сборках клиентских пакетов, так и людьми, которые с ним работают), обновлять его раньше появления бранча 6 не планирую (тут наоборот, нужно всё оставить по возможности предельно стабильным...). 114340 27 pip 2010-10-27 07:38:22 +0400 У меня этот атлас не собирается, зачем тратить время на пример? Эта сборка не рабочая со всех сторон на моем ноутбуке. Может проблема в процессоре или в памяти. Мне сейчас некогда с этим разбираться. Поставил на hold старую версию. (In reply to comment #26) > Последние сообщения - это Вы с кем говорили? Не со мной - явно :) > > Я всё ещё жду ответа на #22. > > PS. Нынешняя версия ATLAS в сизифе вполне стабильна (проверено как на сборках > клиентских пакетов, так и людьми, которые с ним работают), назовите пожалуйста архитектуру на которой тестировалась. 3.9.25 не везде может работать. Смотрите чэнждлог от 3.9.27. > обновлять его раньше > появления бранча 6 не планирую (тут наоборот, нужно всё оставить по возможности > предельно стабильным...). (In reply to comment #26) > Последние сообщения - это Вы с кем говорили? Не со мной - явно :) > > Я всё ещё жду ответа на #22. > > PS. Нынешняя версия ATLAS в сизифе вполне стабильна (проверено как на сборках > клиентских пакетов, так и людьми, которые с ним работают), обновлять его раньше > появления бранча 6 не планирую (тут наоборот, нужно всё оставить по возможности > предельно стабильным...). 114342 28 real.altlinux.org 2010-10-27 08:15:35 +0400 "У меня этот атлас не собирается, зачем тратить время на пример?" Этот - который в сизифе? Если да, зачем Вы его собираете? "Эта сборка не рабочая со всех сторон на моем ноутбуке." Этот - который в сизифе? "назовите пожалуйста архитектуру на которой тестировалась. 3.9.25 не везде может работать" На i586 (точнее - Pentium4 c HT) и на x86_64. "Не везде" - это где? "Зачем на 64 бит --with-arch_32=i586 --with-tune_32=generic??" Это вообще откуда? В сизифной сборке нет ничего подобного. "Смотрите чэнждлог от 3.9.27." Предлагаете обновить? ;) Если нет, зачем эта версия вообще упоминалась? 114344 29 real.altlinux.org 2010-10-27 08:31:15 +0400 И всё же я жду результаты (см. комментарий #22) с atlas из сизифа. Всё остальное здесь написанное не имеет к теме никакого отношения. PS. Кстати, уже доступна версия 3.9.28. Можно будет на днях ещё раз обновить. 114347 30 pip 2010-10-27 08:56:06 +0400 (In reply to comment #29) > И всё же я жду результаты (см. комментарий #22) с atlas из сизифа. Всё > остальное здесь написанное не имеет к теме никакого отношения. Это не просто. Потому как нужно тестировать для матриц большого размера. Где-то больше 50x50. Пока ничего кроме диффузионного оператора с собственными занчениями в нулях функций Бесселя в голову не приходит. Для меньших матриц типа 2х2 и 3х3 вроде работает. Если у вас есть простой и доказательный пример, тогда я готов. > > PS. Кстати, уже доступна версия 3.9.28. Можно будет на днях ещё раз обновить. Мне честно говоря уже все равно, я считаю что в таких вещах лучше стабильность Зупустил пересборку через rpm --rebuild и вроде этих странных макросов --with-arch_32=i586 --with-tune_32=generic нет. Раньше собирал rpm -ba atlas.spec 114348 31 real.altlinux.org 2010-10-27 09:03:08 +0400 "Это не просто. Потому как нужно тестировать для матриц большого размера. Где-то больше 50x50. Пока ничего кроме диффузионного оператора с собственными занчениями в нулях функций Бесселя в голову не приходит." Так за чем же дело встало? "Если у вас есть простой и доказательный пример, тогда я готов." Я не занимаюсь высшей математикой, я программист. А трясти своих коллег-кандидатов - это как-то неконструктивно, у них своя работа, и пока им обновление atlas никак не помешало. Поэтому я и жду примеров от заинтересованных людей (в данном случае - от Вас), раз Вы говорите, что что-то не работает, Вам и демонстрировать это неработу ;) 114349 32 real.altlinux.org 2010-10-27 09:04:55 +0400 "Зупустил пересборку через rpm --rebuild и вроде этих странных макросов --with-arch_32=i586 --with-tune_32=generic нет. Раньше собирал rpm -ba atlas.spec" А в хэшере не пробовали? 114353 33 pip 2010-10-27 10:45:24 +0400 (In reply to comment #32) > "Зупустил пересборку через rpm --rebuild и вроде этих странных макросов > --with-arch_32=i586 --with-tune_32=generic нет. Раньше собирал rpm -ba > atlas.spec" > > А в хэшере не пробовали? Сейчас попробую, rpm --rebuild закончился неудачей после где-то часовой компиляции + lib=libatlas + shift + gcc -shared -Wl,--whole-archive libatlas.a -Wl,--no-whole-archive -o libatlas.so.3 -Wl,-soname=libatlas.so.3 -lm -Wl,-z,defs libatlas.a(ATL_thread_start.o): In function `ATL_thread_start': /home/vv/RPM/BUILD/ATLAS/BUILD/..//src/threads/ATL_thread_start.c:44: undefined reference to `pthread_attr_setaffinity_np' и т.д. до libatlas.a(ATL_thread_tree.o): In function `ATL_thread_tree': /home/vv/RPM/BUILD/ATLAS/BUILD/..//src/pthreads/misc/ATL_thread_tree.c:84: undefined reference to `pthread_create' collect2: ld returned 1 exit status .. Меня еще мучают смутные сомнения насчет библиотек из x86_32? Это может влиять как то? На работу с libatlas? 114354 34 real.altlinux.org 2010-10-27 10:53:40 +0400 "+ gcc -shared -Wl,--whole-archive libatlas.a -Wl,--no-whole-archive -o libatlas.so.3 -Wl,-soname=libatlas.so.3 -lm -Wl,-z,defs libatlas.a(ATL_thread_start.o): In function `ATL_thread_start': /home/vv/RPM/BUILD/ATLAS/BUILD/..//src/threads/ATL_thread_start.c:44: undefined reference to `pthread_attr_setaffinity_np'" Похоже на то, что в новых версиях появилась зависимость от libpthread. Достаточно в команду линковки добавить -lpthread перед -lm, видимо. Хотя это только предположения, у меня сейчас как раз идёт процесс сборки 3.9.28. "Меня еще мучают смутные сомнения насчет библиотек из x86_32? Это может влиять как то? На работу с libatlas?" Не знаю. Я про biarch ничего не знаю и не использую (нет необходимости). 114355 35 pip 2010-10-27 11:13:59 +0400 (In reply to comment #34) > "+ gcc -shared -Wl,--whole-archive libatlas.a -Wl,--no-whole-archive -o > libatlas.so.3 -Wl,-soname=libatlas.so.3 -lm -Wl,-z,defs > libatlas.a(ATL_thread_start.o): In function `ATL_thread_start': > /home/vv/RPM/BUILD/ATLAS/BUILD/..//src/threads/ATL_thread_start.c:44: undefined > reference to `pthread_attr_setaffinity_np'" > > Похоже на то, что в новых версиях появилась зависимость от libpthread. Я бьюсь над 3.9.25, src пакет взят из сизифа 114356 36 real.altlinux.org 2010-10-27 11:26:34 +0400 "Я бьюсь над 3.9.25, src пакет взят из сизифа" Зачем? 114357 37 real.altlinux.org 2010-10-27 11:29:20 +0400 Вам лучше взять репозиторий с git.alt:/people/real/packages/atlas.git и пробовать собирать оттуда. 114391 38 pip 2010-10-28 00:13:35 +0400 (In reply to comment #37) > Вам лучше взять репозиторий с git.alt:/people/real/packages/atlas.git и > пробовать собирать оттуда. Долгая история, я уже забыл свой username. :-) В хашере сборка прошла чисто и быстро. Я собирал 3.9.25. после установки тест с svd работает. Однако время исполнения теста больше чем в 3.7, а я ожидал прироста производительности где-то в 1.5-2 раза, кроме того результаты eig на моих кодах не являются правильными. Что ж, придется сделать тестовый пример. 4625 2010-10-26 22:59:07 +0400 2010-10-26 22:59:07 +0400 лог ошибок error_HAMMER64SSE2.tgz application/octet-stream 67606 pip H4sICJEhx0wCA2Vycm9yX0hBTU1FUjY0U1NFMi50YXIA7D1pc+LIkv2ZX1ERMxu2XxghCYmrgxeB QXazzRUIz3TPxgYjiwL0Wgejw0fH/vjNLB2Iy2DAPe4x9AxIpaysrMyszKysUrmtfaOcYesfXvHD w6cgSey3WFj85XlBliXxgyBIBQH+FWT+Ay/weTH/gfAffsAn8HzNJeTD/f3zcNue/6SfXwjJPvPJ wPN2pCJk7LikNmjV1DxX5kQ5s63uNoDFJogXuJQ8UDKhPvGnlOiO61LdJ96UmuaOuAgh6iel1SJV krsz7Jw33ZGKNUR1NIsSwx4ZuuYb9oTRNDM1H9hgEd8B+uyxMUGir4AnULJfS0ByrV//BBR/qrXb Sr8gqaoibiUbn/c0f+ohKfeaaziBR0YGcsxxDeptrQ/tXrUaUAN5NXUsmru/z/V77dzVbbPVyDFB h9eMq/36LqA5jssheLNTr4Xwv56HrVzkQIXMYESjx2r8OMS88Piq2VmuDMLEJ63m1fIT07jLMAq/ qsuP/MCmOe/JM+yxgyA37fYyiOfquTtT83ITalkhzG/PwdwzmL7wDIzLCBWuNoKY9J6aAoMSt0CJ DCq/BSqPUAN1sA7Kpx5qL1Oz2y/rILTg0TDxef2qtSozn7pjTade2FwdayDs9XbY62Ixhu4N1HUt z/ypS7WRhyDt9mCt+BaE0/7teSAmnb7wLFAknmdgUvLJbwNjrL9u1dZ1z9Rmmv4t59Ixdamt0/1M 0bPmSbNHxHTQQDk2ccbE011j5nvMVLuBbaPdoo9UD3ztzqRgIAKXmbIAnxyPHFSuQat/21kes2gb gA4O7HAIc/17rTlYgXrUfODo8EEz/LFhPsMofNAy7lwNTRxavjtK7gLD9DfXCNutqWAnWLuhCWF2 g7XKaZHurwXR7xIQ0OcNQONicQ7WG2zCNfP1NNhmfDN/AWOr1qvVP6+BCxUsBBoo6mANiA/hTdTL F4obwW9nM+qSOycANQMFMzV3AuaE6JoOvtAzvtNLcJDk7snf7m/WiKUlqs0/FKA42wgvq5JQlvK8 9BJSEbTuWBaOBI/6zFMHsyR4iLwKmYGrfAnW0Em1bhuKigQ2fz2PXNpFfKOu3OQgIvBddEcv53SD jg2bhgM31Q20oXXCLKsDomDjfA9OX4v1hnLNetKojUZD+AHMw2ZnoNwo/SrghxLVR9ugBrbqP5m7 GCuEqLn61PCB0xgHGSNq+8b4CYkfm9pkB6WIw585eYPWsKsOW4YdPEa3+HxYd1wqRgX13m370x9V URZ5EhVhzDS/FKLLW1UpSFfNgRrd39TU4WOpUJCOZYqZIW4Tw2PRIfaZuPSvAMKwEVonm9loUEZr BmYNxtDdf4BVublFxvrd/95YP6qIl2FVAiowhSHpTzWbLKIZTKkXovCIBsIIPMBx90Q004zxuB53 rH4T1m8QGAusoAtwqR/Pwc25+1sU2iZdCH2eYX/D7rBup57BmHcmNhgm1nVm95EZFcTELDipMHPB wNMcBoNLkgAGSsFkPBj+FHpD1KkTmCNiab4+JQ9TzQ+RRR9gsxtjOfPIFETpPNhzDB/JVPOI7RBt PEYBghllZC0ggaGNswoNgwRW3qzXlyll8kxRjHK1HTsLfLCp6aEhZogvI5sXOfjkU1/qHzJvpWsQ 0iS9cRf65i0gw55B/aYFDoj1KdUDmLAxrOCaLcM2LM3kwrpqe12n0n2CSH0CPzMQo+FhSAOUWYFJ oj6GaBpb0Yyc4G4rGvXzi6lBdq+l6POLKdqE6suq4O+isQy1/LmyE2pPwGNg0AcGxnU8LxvLbW4N 6nNZokmwHM83nxgWDyzTJYROfspiOGOf2jCHHLN41WdS6zg+RWsDnnRMnpyAeDPowvgJNRqHU4z/ kt09GKCUQK+HU3hQ+RQBiAyLjBFqKo5f4C53bIOBZqnX79arbOIUOz2IiFh8cUGYs2ZeHa8jt4iX sQ8Cj974vdZqRQ6j8RViqdh7dMDvVH89Zw1cRGWD/q06GHbbPTYHxWFbJRNdj25gYnATEhA1lB07 luFnxy6wPztz2IgkWWs8A12YVj0QWdaCb3BlXfh/QrK/a5fZrO2gqYdITv8GGHrNOkAVpPAyEw2s pFm4iZt9ndYa6dYar90aDtN53z6/dt/SrTVeu7UvaXX5sk5duhtafr6FaLaCuMEo+65mRyVJd7q7 EwnVOt1ub8CqQbxFfoHpo4uzzTFxZr5hGd/DsJTpXm3OMdvJ+i6lWdNxZtkIkoY6swtUnYni13OQ +QXe/xHdN6J77/Mqvxg0K2Ugo7UgjTSIvh3L9+1Ybrrdxk0szN1Z22okfLCoOcbQdJhIT/2qgu2B ZywjFvYdTAprr5OiKDYzF8wC1VvNzmelvwDNytZVYKmLOTw0Oi+bw8d6Ez5bQjaHvYhD+eZvDJ/m hgUhLBSw+36tg73Ce6pPnd0jRuY/NPAzk+/GDL4Dm/1iMDjTXAiGqAlu/hvdHSFOmmt9wmjxQ2pv /mj22D02wgpuO1hUjdpjE/d27bNCWOfxCgbpf4i4cz8Qrh8nhBj13pPnUws8YpTV2BlTmIXblJBw 6XieQUiDxXUiUS/kFsJE1kWUuRqGbvEixKGmQaNcLFNdM0rjwZV1tBlANK0PYxecRoVTY20+3dRM MqJjLTD9cEoQxjEYYrBckmHDuNt13WA3bhu63pjba89K340W7rxvC88W7h4XsICkGkt2utlRB4ld EAhPshqBCa1JhGPOWht0Ru0R00JMjzPuYiYN5w3Mrjs2cJg+grvBi1DE+Gye5j3iXNKDhqAJ6LEH QmdMS0pGUYmelOhRyfek5DuWvCxpdENt6ho6Dv0JBR2yKR3NZ8xoSsbxtGdXrCNnaPgVrJ+JU5mV zNte/8PsKyochLzDVvcmp/T73T4HVz9s/ZfnZbkQrf+KfEEUPvCCiEWn9d8f8AnzbL7mZO9nnOpr YArcEac0bonIFbi8nPX8URbsbFYz/SL5RYAQr0d+h3HShXm/yBNBrsjFisCT20Ed7uEiDGLITec2 x5BnwgJEkDWxIIsx0jmESxdE4iROZtUEnhfIea01ICFF7AnWES4ydWf25BqTqU/OoQ5r5BpCRqLC hPUBJ67XmKBmQeglado6lxlgGgb+GyOYF4F9BCtDw8mvE7hg99Dk6YAbrZru2CMDMUQ5JcxmeqTT hYHsgl32nz7CjNcn9B5myFivrYBnrHUGtatmqzn4SqDoujnoKKpKrrt9UiO9Wn/QrN+2IMLo3fZ7 XVXhMpm3P/5bQh2T+yoE4T9s/PN8Ptn/USxCuZDP88Jp/P+ID0w3Ppw+7/ezPP7V23a71v961Ahg y/4vAR4m/l8Q2fjnBek0/n/EJ/Ov437ePr5oBeNKual10tvZSDQMSPeaqEp90Ox2CJ+FfwQuBD4n FnIs9KgNiCBUxCLD8q93xz/2AeY1OwTYdaMQAXlUwe1XA6VNev3ulZKr3X4h9W6712wpMB0P+cVS BrhlBiaz8w0EcGnqgan5EFJqHilIn684lsgaXfduK5gCAAsFzyyAGeHSgee7gc62u7D1OZnoTzqu rBgzCtEl5TLJGlVtNtNwHYPYgXVHXVwpcenE8HxcKKyQvDgH7Ufl2RggvZ5VFcVSmSvk29etbk9l s9U3RJssQKAsR7Q1uh2FpOUSc7+8JDQxFNrga0/JNpSe0mkonQEZ3HaanZvNdQRWJwTq9ZV69Wx0 Rs7ni1oC9kO6SCMIR9oCDqFC2F5BxHLV6tY/k3Zt0L5tIWZlXjfs/aglhE8rBJSkKl8S+65a4i/J bFyFbyvAb5t9f2PfBiu34ftjhix/enPOVYiUL8pcOU/OhWKeK8jIVh1FAn2A/0YU00wgXd109G+4 7k4vIoSW1enAT4VYWlW4TNEllhhFEqNIQIKwZDxGKKArz+gSM5uJIlWSz8sSJ8jkvFhAooASnKJE K5iXBEJjDowQFiNdKZIGr0cS+GNOAEYVBU7Or5IkSpworCFpsMyl4vFIkoQyJ4tAUokTpDVckrni Oi4NBq9HEi+LnFAm54UyV1wjOFHgBDFFElsJSY1XNjQi9Yfxv3boiGB10HQqowklDWUQuamkUrTS n4BE67DFUkHKi9E+rTXNiluazVdwZafVysFsForq/a6qdjHLv1htoUK92/ua63Sz+JuqkRAdj28n sovLNOW30CRVcF8s7u9ImY2QBcuopPhpfj0quUJulPZvC4jyEX24qbQDGqNPHY8S1wl8XHTPV3Dt lz0bCo/CUNA4nTy4ho8L6HdPpM+Ruom7qn6faiZ9WlKcr4HtOqZZzYuX5Et0DYoW4H4DUhLShqgl ZJ7TOdDzIieWyHle4HhxRefkPFcqrIwCJHuw0iWBl5NODYbio1AYCvt0Kt2nwn6dEuUCDupzscQV hZVOSWWuJC92alng8haBF1Dg/Q3ydpl9MOxoXwYRIra434ba4+xpd56AQUEOoEWJ+cCHY5MoJrWA Gd6zfJDlEseDIxDEdfZEFGPDt5ENhbhzcmbx0TwwgEfLTl5ccfLegpMXF5w8Q7DCYnEnJw91w37j Hq473Fi6sCOGPEBAWK+pioq7iIeeZQnSo/BYkoaeRzn9knSuqiUw2VvFseD1y2JR4ng+DJhg4BRK OHJ29v2EnN9AW7oGQ2fi+ESSQTxiFBvuGiDIK34mDFwKzM8IW/xMSWbTlXNJ4Aqrw0OAQb+sGDsE CIeRJEOAzBeBm6XY7S6QJImcJO4SIEhH5BKEAKLEuJRf441DX71DgCAej6QCX+T4EuNSqbyGS0uC Wx3S4jxAKK0fds8HCKX9AgRxHiBsaHZbgFBaDBDEwwMEcR4gbKBpfYBQWocqCRDK61GtCxDKzwYI 5VSAkBcfpf18aTpAkGIfUpRf4kvlgoQx+7lQ4ErFFZ0riKiQLw0QTlnTf3D+19cmVOBMZ/Jj1n+F Il+UxZX1H5HnT/nfU/73787/Cqf875r8L+6N+R/xfytEwa1A6JPi946fyJ9b3g8Ge3OW0UfbXiNO 3kslH9lWHJL1UFbDaIfS2OGckIz8fmQk+M9CNNJ+aNIvGJ9lqGX4aQorpGmTcWCHKeE/65Y96lMP N6ydVTLb36NewJ5bQq5XZJhiPWguvshZYbsa8JUDcOV/+tbM1qwz3L8w0myYuDqBdwnzK98PX+kg f1rfcMvhLNX3FtXuD+l6uImxEspqBlz0h7gB+f/IeOLSGcneh0/+TXJAXQ43RrXvZg+f/tpQMdz+ tldt3IK/b8UDmsUt+PtWPKDZxr7NNg5qFjfl71vxkN7u22zjoGaT9xP3U+fo3cv9lDLZC7o35S+o yTavZ++pyxI9PzFctJt2J+CAvaSZ1Z4BeZg6mmU8AzDS/HV8TTzUy81rykHt4S3xTIwjucftrYNL j9tdfo//EN/KIgUm8aW3w0m2uaVq/NL3VkD0svsAx2967/469Q97s3nb+2vi8otr+75kRLL6DofB sOMdFrWC03dQqnR4EWsXkv3lQTNN37BoHADKx4jc9Bl5aTyGxCS06IQLVfWkra+urUd/E48sivJk c962zck6y4ZgWYCJXTh0VnPShDesCc5ytEGW3IzzEh+TW+tc9psa75jjWOfidlOtyKH6TzPKTY+Y wGAk9AN78DQ7onM9Ocaf1THulaMCrQQbvH/VxazZWQ8nsqCQ9BPVRvvmzULE5VIqY8YWyXxy9l+j M0IfZ6DZHsEBRc6gvbPwaAzNnQS4T4Pk2SEu4WPTwaOZbM+Y4N5PBD6IpvLbo0nI82+QKEl6JaJO Xv6nNE4YAMT6QRJFcTLbnfP8UMIXueh5a//e308fIyzdoYfLoVF2fFho8fOQzEZzOCUYm4E3Za8a QCx4ikD+oRFIkqvMLclcX9EFtu/p6gk47Z404l1pRFryh6Qf9VmwPD/8u5KPMSmn3OPPnntMJHk8 f/u2NDzB9OdCC2yLQjDDDai4gMYdsfuzF+d+EqqWuLAHqk2ZSSftkB6pe2dqJzf0XtxQJO+FoMTU rZMCvBcFQGHrPyIt7EWVj5oaPkUYb3o15KUB5A311SdPDaw9UrWpuku5WniSOj9oj2RtGvXC/kZj YjtMf13KDh+918yA7XY8G3u6Zo/PLsmI6qaG56azd+81H3Ui8CnDMQxsPEJ56LLtl0fs8YT6lmW7 dHJgZwXp9Xp7ym++7VVM0AcvsMhcHbbmMBfUKK7v7e0lMvFBExo7m7slRG+psvM5NI9Uo2M5TkuC p7An2+RyL/Y26Y01J6n/U2IIS/Monse7Rwgxr7rkT3HzASjLHt40hVIsSG8mdFggS36bZBXeJlnF U0T0TiOiWA1Iog9OhsvNiz2SFYnI82FEUjhaRPIK7nL93zdANxieGha7xddZH/eio8mithwS5SFf b408ajFzjX84FeQxcTUbSlzDf8Lhiu27lSjg5Cukg+dmVItl+PBF/hKfV3mOF0s8LyxD5QWAKoil BExAsNIymCACupIsiAkcXieHIyVwsgD4JF6K8YkcL/GAMtNgJw4AFRVC5oe8tW9btUYjOTkDO8Kq SAI7G2JsOjMYYLzEFflCvlzatX6+APTO6+NhblKpUOYPrp8MjUNz6UfbtnC86YNr7bA7YPGvh+bA 9Oe8kIkZ08b3SV8yyUrVPqDdHZN/SeU4/cdwYJbhcIOX4D5Nwv4h4fhc1SaWlUTkGFXjLpqIoyMy dh2LvEAEYZbAsmaa61FuWpEuV49T3Bcn/k1i5okOQ7u5+5W8tNMMYpGwCbXj3m7Kayo7v7y9EXWR f7W49mhCTwlIflsCSlF2eO75OexC+e2J6dj8XBD0IjtxFn7Tbn9mJ7GdHYpbLOZ34mb4Pu3R55gv 6vXROp0vF//OTj+jLCsjpx2Y4GhfkHVZj1cUxb89lbCJsvwbpWxRFjht6gRWn068g6Uh53eUBv75 o1MC5Z+UQEnUgMwVgqVQ5g+yM0yjADohnFQUjzOpiF8ha0VoS0dE+9IJWyoeICE15eNRw3D9eQBJ 63e/lfefaKcI3G+Cm+bX0v1cmIcTl6jb4ah2XbnMKOqg2a4N8DTawSeFdG7bV0ofT/qCAVu7aimk r9w01YHSD/98043SbrMsFU66q4SUKvGx81ZQJQI7BRR/voU/7Vq9Hp8XSmTMxlSFkiRwRWmOQygs 4MiHOMTncEiyWOBkYY4DDXoKR3hcOf5sxpHnRTwhOjPvccQKpcF6meo40riIQ64cnEZKCUsfoVaR j2R37VxKoyAXSCoR8pO8TLNuQ9bPRXZqzyKUunRsAtGnrYs/6xp+Q+0rtdbLtzDORb+wjdWDEerc 08FJId6bQsSCX6cOJ214n9qwogx44vRJGd6fMjC5LyvDd+o6J2V4f8rA5L6sDOGehJMuvDNdSG17 SVTB0zXzpAvvTxeY3BeVwXdPyvAelSGS+7IynJzEu9SFdU5Cd0xTm3mnqeW7nE1Esl+2DyeleL9G YoNSaI+zu1PK4d3pQyj2lZnFgzY7pabf49SCCX5RHby/8O9CeCd1eHdLV5Hgl61Dbfb/7L1rd6M6 ti5cX7d/hc7qcU7W6lMhiJvt9MgeIxdXVXY5To3E7k6ftfvNcoAk7vJtg52V8OtfCbANQvgKxEHT 3atisJCE5qOpqUdTmmeT+9YZAEI8/TATPaMh3k7HpwAJMZe3Z6LnQQLwICYekovb3dehAwd3ibi8 HUg+McOwQTmIOL+wJ8mVC3vSBCwIuHJB5c4BQwfAICQYOsUc9WjlcdTjNoePWVvXu/jDx6IniVhw ksh6J4lYhZ8kYmVykohKhKusPklEk9c7ScSQZVVffZKIomlYwxudJKJoMqnF/CQPpaZLuqbVjNra z2vx57GkVbFsVNd9XjXiz9clQ1UVTYWTRDhbUaydThKxtj5JxMrgJBEr85NE2I2FVrixcI+3Z1W1 mqTinbZnqXW9KmnyTtuzFKKGJE0rxfYsH1fI+nDbs6yPuT3L4m7PsmB71sefeVxkuj3Lgu1ZggKC uz3Lgu1ZIqMhAQbYniUmGJLbsyzYniUqGJLbsyzYniUoFhKe9xZszxIVC8kdORZszxITDLztWRZs zxIUC7xBAnbiiDyb4O3EsWB7lshKIgUUsD1LSDxwtmdZsD1L3KlFcnuWBduzRF264m3PsmB7lsD6 gbs9y4LtWQIvb3O3Z1mwPUtkPCQXt2F7lqjL25ztWRZszxJ1fsFuz7Jge5awKxeJ7VkWbM8SGAwF bc8y92V7lrl1vdd14S/A4dn8mH7aJtdP2wQ/7RLsED3/0bzLzE/bBD9tQQHB9dM2wU9bZDQkwAB+ 2mKCIemnbYKftqhgSPppm+CnLSgWEi54Jvhpi4qFpGuuCX7aYoKB56dtgp+2oFjgDRLgkivybILn kmuCn7bISiIFFOCnLSQeOH7aJvhpizu1SPppm+CnLerSFc9P2wQ/bYH1A9dP2wQ/bYGXt7l+2ib4 aYuMh+TiNvhpi7q8zfHTNsFPW9T5BeunbYKftrArFwk/bRP8tAUGQ0F+2t6++Gl7W9d7j/y0vY/p p+1x/bQ98NMuwVaRTP20PfDTFhQQXD9tD/y0RUZDAgzgpy0mGJJ+2h74aYsKhqSftgd+2oJiIeGC 54GftqhYSLrmeuCnLSYYeH7aHvhpC4oF3iABLrkizyZ4Lrke+GmLrCRSQAF+2kLigeOn7YGftrhT i6Sftgd+2qIuXfH8tD3w0xZYP3D9tD3w0xZ4eZvrp+2Bn7bIeEguboOftqjL2xw/bQ/8tEWdX7B+ 2h74aQu7cpHw0/bAT1tgMHS4R5eeAxo+LBrOR8N/32d9oq3PWNKcARiiAoN3uAi5B6gQHRUsKPwB BEAhMCg4p2BTH1sAhcig4JyGTd2oABQig4Jz4OmzDdv2BN05vBA9M3iM+69W76Vnge+UcJiIyD4J it7wxXZAUQgJilD2/Anp/Su+f8MAjJLYFf8dHpIw+2wEmJ+2M7RD3FB8hNhgrQ7YyyGmycExQX2X GxNca0CBrDwt85ncPwWoAFRWn6IZQAVwAjhZfrrm6MW+7971ZEDKBzdGSFHNH99O7y7lzAzYlAiT IWYAMSVBTDF4gflxSfCCi8ELjEglwUvO49HcWcg3Y+4fwJYptS1zeHHWaPu/zEzkbdGU4mL08AZA EgtIWQLIP8YJAAQAWm88S4bdjA5nAKFST7OKGcoARIKAKI9hDMAD4NllCMMAoFIzP4UMYQAiUUCU wxAG4AHw7DSEwUS+3GR0IUMYgEgUEOUwhAF4ADybD2HBIfIAnXJAZ4aQ7JZSOZ7IiaVUAE7ply/y X0YFEJUfRLktoQJ4ADw7LJ8CfEq+dJH30ikAqOwAymnZFIADwNl6yRTAU/LliryXSwFAZQdQTkul ABwAztbLpACeki9R5L1ECgAqO4ByWh4F4ABwtlgaBdiUY2G02GVR2Mle+kUJnPuyKIBIABDltSwK 4AHw7LAsCvAp+eIEznlZFABUegDlsywKwAHgbL0sCuAp+eIEznlZFABUegDlsywKwAHgbL0sCuAp +eIEznlZFABUegDlsywKwAHgbLEsCrApx7IoLnZZFLYZl35RIv9DdwFEAoAor2VRAA+AZ4dlUYBP yRcn8j5oFwBUegDlsywKwAHgbL0sCuAp+eJE3ofrAoBKD6B8lkUBOACcrZdFATwlX5zI+0BdAFDp AZTPsigAB4CzxbIowKYcy6L5HqLrBUFKzwEsHxYsgdFyeHHhg2YjcCyi1LYZWMyNXwCGqMDgWCUU FoAK0VHBgsIfQAAUAoMiGefcM0egKcQGBUUAAwpqfwIoRAYFZwbybDv2Y982J4CKD4uKLdGwED0z eIz7r1bvpWfZgAnRMBGRfRIUveGL7YCiEBIUoez5E9L7V3z/BjtBymJXZEV50vW3txAbrNXhmyIA GOFMDi4Jfjo+M+9bZ4AHUCAzWpxAYkIhweqN0/EpQAWgEpnTvhFIpEIFcAI4ieKEu/7q7wgDpHxw Y2TJdrDt1+w53PocM4CYkiCmGLzA/LgkeMHF4AVGpJLgJefxKBGIHWyZMtsyYRSLy1w85r1YMHYA khhAyhJAsYDsACAA0IYe9B4TlB0gVN5pVjFDGYBIEBDlMYwBeAA8uwxhGABUauankCEMQCQKiHIY wgA8AJ6dhjCYyJebjC5kCAMQiQKiHIYwAA+AZ/MhbB64HaBTAujMEJLrcRyJpVQATumXL/JfRgUQ lR9EuS2hAngAPDssnwJ8Sr50kffSKQCo7ADKadkUgAPA2XrJFMBT8uWKvJdLAUBlB1BOS6UAHADO 1sukAJ6SL1HkvUQKACo7gHJaHgXgAHC2WBoF2JRjYbTYZVHYyV76RQmc+7IogEgAEOW1LArgAfDs sCwK8Cn54gTOeVkUAFR6AOWzLArAAeBsvSwK4Cn54gTOeVkUAFR6AOWzLArAAeBsvSwK4Cn54gTO eVkUAFR6AOWzLArAAeBssSwKsCnHsigudlkUthmXflEi/0N3AUQCgCivZVEAD4Bnh2VRgE/JFyfy PmgXAFR6AOWzLArAAeBsvSwK4Cn54kTeh+sCgEoPoHyWRQE4AJytl0UBPCVfnMj7QF0AUOkBlM+y KAAHgLPFsijAphzLovkeoutCfOyPD5fbm8Zps5j42C7Exy4ZXnAxeIHhqCR4yTc+tguHmQowIuVw vI4Lh5kCcLaZMLlwDqUANk2OGgeAA8DZSuPASXBltopz1DgAHADOZhoHjhAsD2zyP0LQhaNzhJlL 4Zwm4QAcAM5Wk3AATsnnUjinSTgAB4Cz1SQcgFPyuRTOaRIOwAHgbDEJB9iUYxKOi5qEg+9EyedS ck6TcAAOAGerSTgAp+RzKTmnSTgAB4Cz1SQcgFPyuZSc0yQcgAPA2WISDrApxyQ83+1RFmyPKsFu uuK2R1mwPapkeMHF4AWGo5LgJd/tURZsjxJgRMrBddiC7VEAnG0mTBZsjxLApslR4wBwADhbaRzY 5VJmqzhHjQPAAeBspnFge1R5YJP/9igLtkcJM5fCOU3CATgAnK0m4QCcks+lcE6TcAAOAGerSTgA p+RzKZzTJByAA8DZYhIOsCnHJBwXNQkH34mSz6XknCbhABwAzlaTcABOyedSck6TcAAOAGerSTgA p+RzKTmnSTgAB4CzxSQcYFOOSXgO26O6DnJW5dLvPdD/upN+15W6PrQe+1P32eyaz7Y0Wtw4pzfO 3ogknfC2OZ5OeoNZoj+7/X7k8tV2Hvrd8KJvDsJvD9Ne3+oNH0fhteu+OfZj3zYnUiwMTTt+ubgy R+O3xZVnOyMputFUim0enF1NnPhVLKE56ve7Y9de/MzeCfT2Iuc/u+N5/dz/IdkN3cWvp+OzyX3r bPF+p+PT5I3I63Vfh84gkrs9iVTbnjRjV53wymLbzYq3mxVrNyvWblas3axou1mxdrNi7WZF281K tJuVaDcr2m5WvN2seLtZbLtZbLtZsXazmHazou1mxdrNirWbybabGW83M9ZuZqzdzFi7mdF2M2Pt ZsbazYy2m5loNzPRbma03cx4u5nxdjPZdjPZdjNj7WYy7WZG282MtZsZazePbTcv3m5erN28WLt5 sXbzou3mxdrNi7WbF203L9FuXqLdvGi7efF28+Lt5rHt5rHt5sXazWPazYu2mxdrNy+Ot6BdzmOi pcFrIzdi137qyDVtx+g1baDo9bPN4Nkc91+t3kvPsqN3esMX25mwhd6/4vs3vMgqilq/hcwIkJ7J jdPkjfiL3nfvenLiFnsDszfkRPv4Gd0/RHKb0cjM7RlJyEsdZsTNhptJWhaYlwXmZYHTsuC+DPdV 2Juhx2MvvZG4TcRtoJTm4TQOp2lSGobTLJxGSWkSToNwmiPRGGlNgXlNgXlNgflNgZNNgZNNgflN gZNNgZNNgflNgZNNgZNNgdmmwGlNwe043G6T0mk4XYbTYVK6C6ezcLpKSkfhdBNOJ0l0ETk+GJ1L bHz4yI3YNaN0PUbpeozS9Vil6yWUrpdQuh5X6XoxpeuxStdjla4XU7peUul6rNL1WKXrsUrXS1e6 Hl/penyl66UpXY+ndD2e0vXSlK7HU7oeT+l6aUrX4yldj6d0vaTS9dKUrsdTuh5P6Xp8peslla6X VLoeX+l6SaXrJZWux1e6XlLpekml67FK10tTuh5P6Xo8pevxla6XVLpeUul6fKXrJZWul1S6Hl/p ekml6yWVrscqXS9N6Xo8pevxlK7HV7peUul6SaXr8ZWul1S6XlLpenyl6yWVrpdUuh6rdF1W8bis 4nFZxeMmke8mke8mIe4mIe4msewmseyyWHaTiHOTiHOT0HKT0HKTGHKTGHJZDLlJSbtJSbtJkbpJ kbpJ2blJ2bms7CxWdhYrO4uVnZWUnZWUnZWUnZWUnZWUnZWUncXKzkrKzkrKzkrKzkrKzkrKzkrK zmJlZyVlZyVlZyVlZyVlZyVlZyVlF+V9K13nGJmO3Z30hk+bEX0V23webfjIZqkno6n5jMgN6cmx 5IpprSqNcpkP5Nmjvv1i9zH6Gxp0f9o0hwr98rv2r2PUoLQrfVmr5xDza+S8oT82y/YgRqI7o8kT 8OcfnD9fkyePgMCnyX3ZmywcBgAHYeEw4MAB1IPAeHjafhFtNrRE9UoUVEUNwFY4AvtjKCLWwzAY TfWMRlM/2yvyj/9mQd5GRnmvthkm06HNMxqs3pCgot+/n9epul2d2ALWqdRDbzirx6s7fXAnTlCF 2nZVINmFRgvooHx10HWKGlqlbkYzOa+ja4g4j4LURLnMcdG0uy9bwMIZkIogl6BcslyzsvqRo1lN D3vrVJaF/9HNdfuIzD9c26UF0lefFR53Xzh00cH/R0B7gA586K6jsQ/8p1DwFFr/sbAVrGJbwYq1 grUvrWAW2wpmrBXMfWkFr9hW8GKt4L1/K8BwsdfDBQEIMXaeg6nFNqCbZbAS4MufPkaXQ/Q4HZqT 3miIDr7aw8uheU6xfHC8Y85GvXqM/uw6QyLMYzQZjdCj/SfqOk/TgT2cuOhx5ND/iHkPYN1z2yYU 6gK1oZL92mgd/bUiHc0THI6RRVTe5QVCpA3v/H//iVC3P37uIvRgT8i/N9edduWEfBBi/41fzD5U h2Jfg8qRf5XZv6RsfP8qv8mkL8zv48i/TEr8hoOU6tKU2ixlBWPVOKYPpL5S8Mb0xS5bDZpy6Ysx rzd/R6Yyi9oEdY3WCSWaI5Yax1ulYo7n0tqQgz2SFpOnzS3U5NxpPjvcPDP+xJO813zuuWCKjSyZ YlBMH9avesG4BCPt/Cqk+fIoGtCSK1qSor1/HeO3Me7ezc4E5P0AAi+VwOW3scwV+PyHnXhbKQ1d 3CoUxuTSms2Y3ExH0WAI9Zef947AjdQqMwo3ZM0Cd/FiWILTux9n/2SZs3kF3p87K74tGP5sf9rC LL4tGBZtf9rCK74tGC5tH9oC5gF7TVD4ENmNT1tksR2jFnk+c04tmrdR14FVKwerNhdrFL8ss7ZI 9B7cml/6cnYNM2m7+OFuGccWT68q0QcqGNfl96DacJxq41RtNeMWbyshODf/lYF1A9aNy7rNNuPO Juixw0+AiPnoREzgrB6yIzhGxSR+AqGXTOgc/i3x0/YMXOzYCA7OUqpSGA8X1C9HJs4cu+beEXGL SmXNw9GcC6IYzn/cnrMs3Kz49yfhim4IhoLbm4Ywi24Ihn/bm4bwim4Ihnzbg4aA2cBe0xgUIbtR b/McTJD1nlNWM1FFpM4SVvMk78FX0cLXdAYLk86IJ9nQ39vHi63Uasop9rpCME70jYFwAsKJSzj5 xspiQupfAvNQEuaBSpPv6cX+AiIvl8g5XBP7y/ZUU6gzUiDGr0ZhPJNfuRxpJtfev/268zplTTLR 4zqLmUjfNhJbJMPC359hKrYVGH5pX1rBLLYVGHJpX1rBK7YVGGbp/VsBrP69JhsIQHbjlWYZAK20 92cnBJJaiJwllWYJ3oNTImVTjiVJKHHTcfyekulUZZ5QW5qQCoNSO3WlXig7xaGbtGiFVlNTi1YT gpcirwu0FNBSXFoqiBMwm7rSq3BEWi+3V9QlFRg89G3n8M/e5PnQHI8BFgVTEb7qo8wAy0QwPwD3 VCqBy2kCz4B5ioSrSaCLW4XCaCdasxxZp6EzUPaOdlpUKmveieZc0DS7dXOlsMzTrPj3p56KbgiG fNqbhjCLbgiGf9qbhvCKbgiGgtqDhgATf6+ZCYqQ3UioeQ7AQu07CzUTVUTqLA81T/IeRBQtfC0m Kky4moqiCbUxVmSGjkotnka7XFRBX55zmDjM2Vid2Fmkrq6RWpunrq9I/VqrztJiecUb+ml93qqm V9+bdKsnX2AZ8UZfjnkJEcg3+srAvgH7xmXffCNvMX33L4F/+2h0TKDVeAQc+wvI9oPKVk6VbQZk W6gFUrDEr0ZhfJtfuTzP9XKng/071mteqcxP9SI5F3Vg0W3nKnGmV1j8HhzpVXBDsAd67UtDmEU3 BHua1740hFd0Q7BHeb1/Q4BVv98neRGE7HiQ1ywHINz2/gCsUFQRqSeOv5oleZfTr0jh94/dB3ct 1i2aOka9pec9iOWtrch7EMtbX6MmcW7PWPEEmQkFaSs1vfbezJee+hqrjuFihCbGQVzkpYEBAwaM fw4Xtboi5/TQS/BTKsuBTFTdcYkx9hcQeblELqeKPIsTuAKdkQIxfjWKO36LVi5HvszawzCm1iin bZFWYXHyLpKRI619iRxZbCswRNm+tIJZbCswLNm+tIJXbCswFNk+YcEsHAxmFA3m3sDBLBwPZhQQ cAIb8GhLeDRr11ii1gg2yn4MxjSU1ELkLF9qvWMgTmvtQJzW2oE4ScrauBZJqy1Nq41rcuRMN7n2 3me6aYlqrWYwLdFCd1oQuhPIyxTy0oqF7rQgdGd5eCwrLXSnBaE7SyxwDm9pZRW601qE7rS4oTut 9wrdaeUbutP9szvev6Pc5pXK/Cw3knNRR1f94/RH4jS3sPg9OM6t4IZgD3Tbl4Ywi24I9ky3fWkI r+iGYI91e/+GAMt/v891IwjZ8WC3WQ5AWO39yW6hqCJST5ztNkvyLoe7kcKTpFVqUm3Oy9QNoyBu aekpa9Hqi3HQGnljIIuALOKftEaNj8hRTfQS2IOynMhFpJlyJBfzyw5ncgUA4pdX3PlbtBZ5kgVm t79/ZMG8UpmTBSTnomZE56fNBFkQFr8HZEHBDcGSBfvSEGbRDcGSBfvSEF7RDcGSBe/fEGD57TdZ QBCyI1kwywHIgr0nC0JRRaSeIAtmSd6FLCCFr3cUfJBwjbPgSUJtrNXCfXd1TXvvfXcqW6k1SIhF s4hBQZD3BQoCKAg+BUFNmsjUlF4CBVEWCoJIM+UYeOYXEHm5RC6nijwL1inQGSkQ41ejODKKVi5H Msocjd/2joxaVCprMormXNCM+/z6xz9ZMmpW/PuTUUU3BENG7U1DmEU3BENG7U1DeEU3BENG7UFD wBxgrwkKipDdyKh5DkBG7TsZNRNVROosGTVP8h5kFC18ze1WNOm9O7FW7raiCVVljI01d1zR9LVx JK2+ohKzcIaFOc+kbszSYzVaTXLFmlsImou+MdBcQHNxaS7fWFrMhf1LOHn7o5EbgQbkbcFifwE+ q1wi5/BZ7C/b81mhckiBGL8ahfFZfuXyPGz9dQ/5rEWlMj9s/bWwSfvpXZLPmhW/B4etF9wQ7GHr +9IQZtENwR62vi8N4RXdEOxh66/AZwHHsfSw9ddd+ax5DsBn7f1h66GoIlJPHLb++o58Fi18TT4r TLrO+UE0qaqsx2YFacex5Pry5LVxbRA92cdYUWljrL1iI/JAdfUDsQJqy9NrYy2aur7ksCSanB75 rgRHJWn6ezNydaZSaxz1/ioaJ0ffGDg54OT457y/xjg5/xI4uQ93uvdrGifH/gKcXLlEzjvQ/TUr Ti5UDikQ41ejuAPdX/Pl5HrdQfd170i5SK2yZuX8rAviHi5Pr07vWF5uXoH3J+aKbwuGmtuftjCL bwuGnduftvCKbwuGoNuHtgDTf69ZGx8iu1F0iyyAo9t3jm4uq6jgWZZukShnmo67Y9Ev/T4WiVBZ I7Wyzm7IeWp1nlpLJwyjNZFXxkQMUg9HNJwgExFRXpF+kX91nTcdK4sCams9oBrzB+orHqjNHdaw vCpvMuvzgzlqxntvKq0nar+U2tOSwhWC3fNfGug9oPe49F5gLi64geAaCL6PxvYEqo27izTxE1B8 JRO6nC70DEi+mYpIw1lKVQrj+YL65UH0/a5nOSD/mxgiFqn+oqp6RmPxPLM/ZvkfoJ6LpmM0oU5c E1vK9HV87THbQjwPpQGd/wMaA7e+PlmjRVnzdyF8M4oHFwAhMCBcHiKc0eQJACEmIHzZJ+AwADgI C4cBBw6gHgTGw9MuR9wsIjO5sYtwzIkqnCjaipqauFFrH7lP9jB7uz9g+eib7YsrgjtzRXByii4/ X7RwFnFJ3aI9i52145I6a8clJSm1RaBRrL77fnY1XqfVzrOOYFFGab8Dbh1Gw9hoGLFtFoycfwXM 64dlXhOi5brTsj+AwEsl8CTRzv6wtTEbags+urhVKMyKdXKNMkp3Wz+87Z3xGqlVXuarX8S7GbB+ 6ctNWMyk7eKHu2WGbDy9qkQfIPZsXX4PexbH7VlO1dbaE/Yg1qawoFeCaQumLde09dERGQWDa7B2 SmLt+OKcmSA4Zu8kfgKhl0zoHCM38dP2Zu5Mc6ThLKUqhRm7Qf1yNHfNsWvunbW7qFRexi4t4d1s XVr4useaBklnRqts6HsQkTdWfyHMT/rGYH2C9cm1Pik4IqOTfwlmSEnMECpNvhXC/rK9ERICiF9e YaaGX4scLQ3X3r814Xmd8rIzSAHvZmaQsteL5OenWyOQnz1RFWbXWlrC+SnoSv29N2Zp0QqtEelv 3mpCWDbkdcGwAcOGa9gQbETGO3oFO7E+mv3iqz7OPiz2BzBYSyVwOU3gGZirgVrgo4tbhcJMWFqz HC3YoTNQ9s6EXVQqLxuWlvBuRiwtfC0rNky42oylCbUxVuR1DmCgiQmg/2et8xciiTlnpaYkdhap q2ukXuvcBdfyW8MYr3Pmgp/1a606S4vlFc3hp/UN5Jpe3YNY3kmBLjHx6dsxbyGCmU9fGex8sPO5 dj4FR8RQ8C83svSXVlGtVYO6AVAKNQQDLccz/dlfYFb3QWUrp8o2AzM/1AopWOJXozBL369cnl6g 7nSwf06g80rl5gNKSng/F1BS+P1j7DCyJfZ+NHXM6E/PexDLW1uR9yCWt75GTeJGqLHiCaJFg7TE jK69txmtp77GKl9URmhieKOSlwZrGqxpvjMqAUfUWY1eArtaFq9Equ64RjX7C4i8XCKXU0WehRtq oDNSIMavRnE+qLRyOdra1h4eFmDlfViA9Y6HBVhrHxZgrX1YAElZG9fWC0JG0mrjmhxxa5Vr7322 gJao1mrb1xLteAELjhcAszfF7LVixwtYcLxAeSwgK+14AQuOFyixwDkWr5XV8QLRo7K4xwtY73W8 gJXv8QLun93x/rlBzyuVmx80KeH9HKFJ4UmDNzWpNrfp6obx3tutEnUSwimZvDEYmmBo8r2SCTii bo30EiyPsnivEmmmuK8yv+zgvxoAiF9ecb6qtBZ5Ghpmt79/hsa8UrkZGqSE9zM0SOHrbbkKEq6x 54ok1MZaLVwprmvaHjhcxiu1xpaqRbOIYb6Q9wXzBcwXvvlCwBEd1uglmC9lMV+INFO2WzG/gMjL JXI5VeRZWKyBzkiBGL8axRmytHJ5nlA0Gu/feZyLSuV2QhEp4f1OKCKFr3tCEUl6706slSvENKGq jLGx5ioxTV8bzxeJi+PiUteIE2+76pykaCuKcU4SeWOwfMHy5Z+TRMARPT+HXoIZVJZzkqhi5C4S s7+AyMslct7RWMwvOxyNFeiMFIjxq1HciVm0cvkeRb+XJ9HnfxD9u55Dv67lGyZdxzuSJvXPzlrD 7g3SjmPJ9eXJa+PaIOq3aKyotDHWXrEReaC6+oFYAbXl6bWxFk1dT0/tY5xuhQr8QDX9vW38erxO ax3GL95Z/GDkg5GfehQ/cxL/G+wv/4iHr6dY8+wvYM2XS+T84/azO20/etg+z5pPVKPIo/ZzteZ7 3UH3de/M+Uit8rLn/SKKMOi5HhR+6fexvfzKGqmVdbwz5qnVdc4Si9Zk9WFiQerhiG7IZ84UkFek X+RfXedNx8pah4pFHlCNdQ4W8x+ozQ/eDc8VW5J3OA2oacY7O7nQqkartGIWoCVlK8REwH9pmAnA TIA7E/DRETEjgmuYC3w0wzBQbVynlsRPMBsomdDldKFnMB+YqYg0nKVUpbApQVC/POYEv+tZDsj/ JoaIS6q/qKqe0Vg8z+yPWf4HqOei6RhNRsjqTmwp09fxtUcgd88ZTZ4C2EHn/4DGwMX5j+YdWqNF WfOX1s2X/dZ6JYROUWrCi/Y85D3Zw+z7YGBx0xfbFwbBmzEITt5HpTiLo1K8wh3hnOgZHxirxj7E 33MEO3iE4h7mmTC0xIaWiJ2wsE79K5iFfNhZSEK03DkI+8NOloKUUlhh5oOT64kTnh9CeO+shkit cvQjenh7P8vBLz5qO9TlfbAdmGqJYD34rwz2A9gPXPthFgJ+NrYE12BDlMSG8MXJtyISP21vR8xA lFJkYbZEUI8crQkaJXjvjIlFpXLbjUdKeD9TgpYesSRkQ98HSyJeKxEMCfrGYEeAHcG1I8Io7rPB xb8EK6IkVgSVJt+IYH/Z3oYIAcQvrzALwq9FjgaEa+/fEsa8TrmdSmW/4xJGJKJ9Xanvg+UQqZEI ZgN5XbAawGrgWg1BMO3ZYEKvwGYoic2QFondyyoSu7eIxO69V8x1L9+Y694+xVz33ifmevEWQzLq ubIi9SKluirf9wgfLi85zdIULxy4B+HAwSRJNUni4cC9zcOBg9v+HlgfqSGivcxCRHuRENHeu8WC 9nKOBe3tUyxo731iQb+D90RaMOglySPxnQuMqbzc20LA8MgehEcG6yLd3SIWHtmD8MilcrZIi5Xr ZRYr14vEyvXeLSiul3NQXM/aw50eRQbFLd7iWD8qLk0anpqwPwFsOfVawz4RLYStByFswTRJM01i IWw9CGFbIsMkLYStByFsyyxwjiGaVQhbbxHC1uOGsPXeK4Stl28IW2+fQth67xPC9h2ceKIxbPch MC1OVEoIVx6ITAv2Y6ovTywyrQeRaUvlzZMWmdbLLDKtF4lM671bZFov58i03j5FpvXeJzLtO9gP 64am9VMqY2WPQs4ma7WObSJW0FkPgs6CZZJumcSCznoQdLZUlkla0FkPgs6WW+RyqsizMEYXQWc9 ftBZ792Czno5B5319inorPc+QWffYZt7LF7qnnBcwgVx9SCIK1iS6bvcY0FcPQjiWqpd7mkRPb3M Inp6kYie3ruF7vRyDt3p7VPoTu99Qne+y4l7a0et3yB4pxmE2MRrhq0PUmvrBuT00zPhPpeE4/Rm wSnVfYmYafDfYY1zCMU7hhDsKrCrUk8hZA4hBLuqTGcQ8sNpehBOs9wi5586md2hk9EzJznhNL13 C6fp5RxO09urcJree4XTLN7I5kSxXLaazAm/qa6TXFknoqbJDXqpr8x/9gDHOk+vEDFv9yM6pW9x s/Vaw+IWMUalBzEqwepOt7qZGJVeLEYlGGEf3QhLD13pQejKsgtdThd6BrZ3NHSllxK60nu/0JVe fqErFxspzX3coWwWtkX5/8EWZdiinM8WZThlHixWxmK9OB8N/30f3d9oxnc1wmnzH9+E4Qg5dduy Cbxx+UXP38BsZraD2ZRSocavRpGbmM0PEnrdyzn0uldw6HUTQq9/aIvh9v1Cr5vFhl43Y6HXzRxD r5t7FHrdfI/Q6yaEXhcx9LoJoddhaGGHloidsDBMTQi9/qHnIQnRcqcfZlah181F6HXzvUKvm/mG Xjf3KvS6+V6h100IvS5q6HUTQq+D/ZBuPzCh100IvV4qGyI99LqZXeh1Mxp63Xy/0Otm3qHXzX0K vW6+T+h1E0Kvixl63YTQ62BHpNoR8dDrJoReL5MVkRp63cws9LoZCb1uvlvodTPn0OvmHoVeN98j 9LoJodcFC71uQuh1sBrSrIZY6HUTQq+XyGZIC71uZhV63VyEXjffK/S6mW/odXOfQq+b7xN63YTQ 67lZI0z9hTBHIO462COp9kg87rq5edz1pVVUa9WgbgCUQk2R1DjsZmZx2M1IHHbz3eKwmznHYTf3 KQ67+T5x2E2Iww5x2Nd3voA47GBtpPpexOKwmxCHvVSeF2lx2M3M4rCbkTjs5rvFYTdzjsNuWnu4 7aPIOOwmHHIAhxzkYJtAHHYwTdJMk1gcdhPisJfIMEmLw25CHPYyC5xjiGYVh91cxGE3uXHYzfeK w27mG4fd3Kc47Ob7xGE3IQ67kHHYTYjDDvZjumNPLA67CXHYS+XakxaH3cwsDrsZicNuvlscdjPn OOzmPsVhN98nDrsJcdghDnvGlgnEYQfLJNUyicVhNyEOe6ksk7Q47CbEYS+3yOVUkWdhjC7isJv8 OOzmu8VhN3OOw27uUxx2833isJsQh13IOOwmxGEHSzJ9y3ssDrsJcdhLteU9LQ67mVkcdjMSh918 tzjsZs5x2M19isNuvk8cdhPisOcQh92EOOwf+ExCsKvArko9kpA5kRDsqjIdSMiPw25CHPZyi5x/ BGV2J1BGD6DkxGE33y0Ou5lzHHZzr+Kwm+8Vh92EOOwQhx3isKcb3RCHHazudKubicNuQhz2Uhlh 6XHYTYjDXnahy+lCz8D2jsZhN1PisJvvF4fdzDsOu7lPcdjNd4nDfg5blGGLcj5blOHIebBYGYs1 HqfZjMdhNyEOeylMGI6QU7ctQxx2AUTP38BsZraD2ZRSocavRpGbmD9KHHYz5zjsZhFx2P28tGzy CkQYaxV165y709def53ZAU08sd0JLSKcGMytKHV7oYRZRgVuxV5NyyBvX3sHfc56rFYXtBTo3w9o sl3cNE6baI0WjWCA/nY/E72ZAMT81DwAhGiAiJyQGAHE3F8EACEaICK+QQtALByzARCCASLqhL8A xPw0M8CDYHiInFy3gMPTw+AF8CAkHnzRJwFh9//nERAhJiJ82fMg4QIiREWEywPEAKadogJiMOAC AqwIYQHBtSL+pw9WhKiIoLLnQcIBSAgLCYcLCQcAISggHC4cYOopLiD4U08XDEtREeFyDcsJWBHC ImLCtyJIiwIkhIVELwmJfncKjhGCQiKQfQISi3DWgAjBEBENXb4AxHgEpqWggPBFzwEEmJbiIoJr WjrgLCMoIByeswy5+QR4EBUPT1xAwDRDWEAM+IAAFSEuIjg6wgUPS1ER4XI9LBdRdwAQogEiEmEp AogxaAhRATHmaogx+EWIigeHCwegroXFA4e6XkQUBkCIBohI9OgIIN7AIV9UQLzxHPLJXTAphQUE z6R8A5NSVDw4XDiASSksHhQ+IH4CIoRFxE8uJAARwiKCA4gJrGSICogJdyVj8gAudMICgudCN4GV DGEBwV3JmIxBQwgLCK6GcICpFBUQDo+pJHdBQwgLCK6GcEFDiAoIl6shwIYQFxBcDQEbfsVFRHLD 75P9ZA8BD+LhIRA8C4YxaUvYtiUkHALRxwFhj91efwT6QUBAzETPagir9wgbfYVUEL7kWTgMnQEG OIgIB1/yPO0AkBBZQ3Bg0RvSuwAJ8SARSj4OhxcwIcREwwvHgpg4wD8ICYZA8CwYwHYQFQ08c/IB 4CCoKfnAgYP7AFakkGgIBM+AYQxgEBMMYw4Y3gAMYoLhLQmGCVgNohqRPKthMgY4CAqHMQ8O7sT9 n4nTBc5BRETMhR8HxfOMuQZQiAeKhfBjoCAosQAOHxYOt9vCwZc7QcLjyKGaIuo7+aczPzroes2K bFGBeUmPla6DnFVZ9HsP9D+i2rqTfteVuvPqdgfdV2lR+67rh/WYX76O3yKXfsB7NtI9E+icDXPN xDhmItwy8U3Z6JZsaMN4XDs2rBkT04qNaMSGs2FjmTCBLJgwBuwh9vETzJkDrJnji9nDa5mTS5lz K5lTC+Nn1jFHljEHVjHHFTGH1cSPKmFOqmDPKWA2qTNblJkNqsz2RGZzGrM1idmYwmxLYJzSWZfk mD8q44/IeKPFfZHirihJT4T4InRsDTK2BhVfg4hT0DEGMsZAxRiI+Aw0PgFJ2J8J22Mx7pCuf4xM x+5StbCxCthRcQR6rmKbz6PNi974gcloaj4ji9yUnhyrMuj+tH/X/nWMmnb3hb681XNsczJy3tAf K/KOaNIDPx90+G+kIDeRdYMOMTvkHeOe5toSTAXBTIW57GOGowuQEBcSbgomFuYYQEIwSMxEnwSE C3H2RAWEy4bZcxdTMgCEgICgok8AIpyUAyDEA4Qv+gQgwKoUFQ9ck/IJDpUTFQ9PiUPl3AgzC4gQ EBG+7HmQcAERoiLC5QECThETFhCDARcQYEUICwiuFREs0QIiREQElT0PEhDfXVxIOFxIQFQUUQHh cOEAU09xAcGfesLhk8IiwuUalhOwIoRFxIRvRcB5pCJDopeExMxpFyAhHiQC2ScgEfptAyLEQ4Qv +gQgQs99AIR4gPBFzwEEmJbiIoJrWjrgLCMoIByes0y4fwvwICQenriAgGmGsIAY8AEBKkJcRHB0 hAselqIiwuV6WIYbuQEQAgKCij4JCAjbKywgxlwNMQa/CFHx4HDhANS1sHjgUNfheS4ACAEBQUWf BMQbOOSLCog3nkN+eKYTAEJIQPBMyjcwKUXFg8OFA5iUwuJB4QPiJyBCWET85EICECEsIjiAmMBK hqiAmHBXMsLzXQEQIgKC50I3gZUMYQHBXckIz3gGQIgICK6GcICpFBUQDo+pDM95B0AICQiuhnBB Q4gKCJerIcCGEBcQXA0BG37FRURyw+8s4AvgQTA8BIJnwTAL+ANwEA4OgejjgFiEfAJACAaImehZ DRFG7gI8CKcgfMmzcIDozqLCIRnd2Y2F/QNICKkhOLCYx34ESAgGiVDycTi8gAkhJhpeOBbELPgr gEEwMASCZ8EAtoOoaOCZkw8AB0FNyQcOHGbhvwENokX1fUgakbPw7wAG0cAw5oDhDcAgJhjekmCY gNUgqhHJsxomY4CDoHAY8+BA/vc/E6cLnIOIiJgLPw6K5xlzDaAQDxQL4RNQPI4cio+ox9yfzvzA mOs1a7ONJTMr6bHSdZCzKot+74H+RwDdnfS7rtT1q2stgpT7ebLXNNL9oPsavXT9WA/zy1c/4v3s 0o+CzoY/Z6Jfs7GPmcC3TNhTJuglG/KQjXcXD3bGxrpiAh2xYW7YGCdsgAsmugFztj17snn8WGvm VGPmTFv2RFPmOEvmMEPmKLv4QWbMOVbMKUbMGTbMCSbx8yuY4wvYzevMzmVm3yqza5HZs8bsWGL2 qzC7FRhfdcZTmfVTjTkpMk5qjItS3EEl7p+QXJ6Or0zGFqZiCxNxYjrOS8ZoqRgtEZuWxqclcas0 YZQkBiR6g/xkSaMdNMNCkVVs83m0cSaVjR+YjKbmM3LJTenJsSqD7k/7d+1fx6hpd1+I0kNWz7HN ych5Q3+syDuiKg/8fNDhv5GCvETWDTqQ7JB3xCTwLDhs9KNbBBfnP5p3255HnDxt1FuMnoAI8RDh iz4JCBdiZYkKCJcNleUtLGgAhICAoKJPACKcQwEgxAOEL/oEIMgs2rx3pw8ACiFBMRc/DxhTAIbI wJjygfEEZ4eJCoqnxNlhXoRrBUQIiAhf9jxIuIAIURHh8gABh0UJC4jBgAsIsCKEBQTXiggWXQER IiKCyp4HCQjjLS4kHC4kHBMQISgiiOh5gIDJp7iI4E8+nSkgQlRETHmAgGMnhQWEy51rTMCwFBYR E75hCSeRigyJXhISz7DQJSoinrkLXc9AYgsLCC6J/QwktsCA4GoIiM0qKh4cLhwgNquweFD4gIBI nOIi4icXEoAIYRHBAwREXhQWEGOuUTkGo1JUPDhcOIBRKSweOEbl7AQJQIR4iAhkn4BEeIgIIEI8 RPiiTwAiPEYGACEeIHzRcwABS+DiIoK7BB6cJQWAEPB8mdGEC4cnwIOoeHhKAgIOoBIWEPzzp8Jj /wAQAgKCij4JiDdwkBEVEG88B5nZcZ6ACBERwV3vDE90BUQIiQgOICbgZisqICZcN9vwVGcAhIiA 4HGVE/CIEBYQXI+I8GR3AISIgOBqCAcmnqICwuFNPMPoDgAIIQHB1RAuaAhRAeFyNQTYEOICgqsh YAe4uIhI7gCfhXkCPAiGh0DwLBhmYb4ADsLBIRB9HBCLQG8ACMEAMRM9qyHCeH2AB+EUhC95Fg4Q 6FtUOCQDfXuxYJ8ACSE1BAcW84ivAAnBIBFKPg6HFzAhxETDC8eCmIV8BjAIBoZA8CwYwHYQFQ08 c/IB4CCoKfnAgcPzA1iRQqIhEDwDhjGAQUwwjDlgsAEMYoLBToLBhWFCTDC4nGHChWFCUDBwhgn3 DcAgJhjekmCYwORSVK6BN7mcjAEOgsJhzIODO3H/Z+J0gZoWERFz4bPTzHCBE0Ah4lRzJnwCiseR Q/ERdaz+05kfE3O9Zm22qMW8pMdK10HOqiz6vQf6HwF0d9LvulI3qK61OOdolml30H2NXrr+Yavz y9fxW+TSHMUurdHEvHenD/FbU+bWU7ADfhGKrO+HNI1cu7HLwUBKBFJnw2izMZSZALps+FQmdiYT OZGNm8cGTWMiZjHxkphoOfFYKUyoDDZQAnNKPnNGevyEbOaAZPZ4XOZsVOZkTPZcxPiheMyZaMyJ WMx5SMxpOOxZKMxBGMwxCMwmeGYLNLMBltn+yGx+Y7Y+MRtf2G0PMZ93xueZ8XiN+zvG3d2S3k5x R5eYn0NsnTu+zhlf5oqtcsRY7hjLGWO5YixHbJYbn+XEjdyEjZMY3+gN8pMljXZRNHO9WLHN59HG mVQ2fmAymprPyCM3pSfHqgy6P+3ftX8do6bdfSE6FFk9xzYnI+cN/bEi74jmPfDzQYf/RgoyE1k3 6Li0Q94RC8NcKGKwLz6mfXG7w/4cX/RmAhAunHsuKiBc9thzc2GMASAEBAQVfQIQoTkOgBAPEL7o E4CITMgAFOKBYi5+HjCmAAyRgTHlA+MJjicUFRRPieMJzQg9B4gQEBG+7HmQcAERoiLC5QECzqMT FhCDARcQYEUICwiuFRGs0wEiREQElT0PEhCSTVxIOFxI+Kv1gAgREUFEzwMETD7FRQR/8um77AAi hETElAcIONlWWEC43LkGxPoVFxETvmEJhx2LDIleEhLPsNAlKiKeuQtdz0BiCwsILon9DCS2wIDg aggH8CAoHhwuHBTAg6h4UPiAgGC/4iLiJxcSgAhhEcEDBAR3FRYQY65ROQajUlQ8OFw4gFEpLB44 RuVsPz8gQjxEBLJPQCI80gEQIR4ifNEnABEe6gGAEA8Qvug5gIAlcHERwV0CD072AUCIBwgqeR4c ngAPouLhKQmI8HAvAIR4gPBFnwTEn/NzDAEQggGCij4JiDdwkBEVEG88B5nZEY+ACBERwV3vDE/5 BEQIiQgOICbgZisqICZcN9vwpF8AhIiA4HGVE/CIEBYQXI+I8LRvAISIgOBqCAcmnqICwuFNPMMT /wEQQgKCqyFc0BCiAsLlagiwIcQFBFdDwA5wcRGR3AHuwpKn0GvgnDXPWTQowINgeAgEz4JhFg0M 4CAcHALRxwGxiAcHgBAMEDPRsxoiDOsHeBBOQfiSZ+EAccNFhUMybrgZiwkKkBBSQ3BgMQ8MC5AQ DBKh5ONweAETQkw0vHAsiFlkaACDYGAIBM+CAWwHUdHAMycfAA6CmpIPHDg8P4AVKSQaAsEzYBgD GMQEw5gDBhvAICYY7CQYXBgmxASDyxkmXBgmBAUDZ5hw3wAMYoLhLQmGCUwuReUaeJPLyRjgICgc xjw4uBP3fyZOF6hpERExFz47zQwXOAEUIk41Z8InoHgcORQfUV/7P535yUHXa9Zmi1rMS3qsdB3k rMqi33ug/xFAdyf9rit159XtDrqv0qL2Xdc/b3d++Tp+i1yao9ilNZqY9+70IX5rytx6Cg5BWESj 6/tRbSPXbuxyMIhfxh8OIqmzYbSZGMpsBF0mfCoTPJMNncjGzWOCpjEhs5iASfFwOUy0FDZWBhMo gTkmP35IOnNGNntCMnM8LnM4Kns0ZvxcROZYPOZQNOZILOZAJPY4HOYsFOYkDOYcBGYXPLMHmtkB y+x/ZHa/MXuf2J0v7LaHmM874/PMeLzG/R3j7m5Jb6e4o0vMzyG2zh1f54wvc8VWOWIsd4zljLFc MZYjNsuNz3LiRm7CxkmMb/QG+cmSRjsqmkAvVmzzebRxJpWNH5iMpuYzMslN6cmxKoPuT/t37V/H qGl3X4gORVbPsc3JyHlDf6zIO6J5D4J81GzyQZet2zYZCe+b11+Pmvi8S9r9tufZi0IadKTbrJSH 3vCgYo5XtfBkOrSP3DeXAHV05NhutPxYvaRdXtmvTOJV3atO8/TiAo0d+8R9l5ed1SDnF7UiL2q9 y4taubyoaW1SEfQ35LfMzXTY7A7M550aIprxQWCTg/39Ae3vNVoyBmabVOC+T+Ezm4qBzD/cnGuE XhdyRBGRktlSfzx5duyuRb4NKquVEL12pkPJfd5IFUXLXw8uuyjLuLLyM1K217qfPtKHiig66Phm 2b1JTQzberKl5wzKkMnH0DT/b9WI/6UfrCqfMNYMbOCqqqqfZKzpmvoJyUU0wJR0AgehTy8vy9Ot +v2Dfv7Sexxa9iPyUXx/fnr+rdG4+Nq4/1ZBCP2F/NIb2st/9M3RBsEKqtYMTVUqf7GHZFZS+QSf D9v/LTpXvxu92E4WCmBl/zfkRf/XdNL/dVnVoP8X0v/DXty+ub26b3ZarXtf7EjR5ArzWzv9t+aS 55qx56DHfYT+775T/1dkjOWg/8P4/779X1fS+3/it+aS55qx56DH7Xf/t5pqu9NqSOR7ZmUs7f+a UTWMiP2vGaT/qzD+F/Sp/DXbz/7nh4LPWePraSuY2ahSXVL0GfeKrr+g28Z5+/K6hZRDfKgj8gXL R4pxRIYnmTyCMD5WFT+XvwrXfuQzJ0i2JsjX46UfiCly1Ldf7L4646YpV56comRA3LMFhqsE1sQd PJqLValMcnYG6PBxTVqL87ZrNB9dwuK0ntXvPQTvom/3Lky+B5VfEanMT9sZ2v1IGei3oBQjk1KO gvxDiRz+Gyl+IYvlwmq2xUQciCz3jXpSNO+74Gj4cT2ILm4ap34lmj++neJ1nXiS0Ag9k31ExB3N Fji5A5yUAyd3OeKkA/oE9EkMJ500nIA+AX2yFCeDQbMJGPnwGGmO/rSd+50xQtDAQ0gHEPLhEdIZ j/NDyA3oENAhc4Tc8BACOgR0yBKEOMrPZgsgAkokgAiBw32LC5I2gARAsgBJmweSDmgSGGxWaZIO aBIAySpNcj+eTpr3DzJA5cND5azRPpUzQQrFxBKwADdfDrDgIsACBH05wHKXO1g6MAzBMMSApbME LDAMwTC0NlhgGIJhaBVYgKAF7m0BEy6rAvQsQCQKEQ6nAuQs8G6rtAhQswCRJVpk4vi+sQp5L/CE LZEn7OFFazTsDHuTi1736R79d+U/0OyzG4rmgFkGJJgClcZV9h2BBL755dJI74YiUEclUkfvgaIO WEdgHW0CpM4yIIE6AutodyCBdQTWUSYoAnUE1tEOKOoAdwTW0SZA6qSY2R3gjsA6ygpIYB2BdZQJ ikAdgXW0E4qAOwLraCMgdZYBCdQRWEe7AwmsI7COMkERqCOwjrZEUW/4YjsT8IMuj5NrFDXM4LYr iAKwpGEIzsMpKYYKAhC4UZfHR/a9lFAHlFBJMZQ7gOi5oS1QQaW1g2iYcHc8cu37rDVS8gTJEE2g jMRAU/5QaoNiKimU2oWrpTaoJRGwlDuQOmAtlXfCVri11AFrSRA05Q8lsJbKCqXCraUOWEtCYClv IN0AtwTc0lZouuGjCdQScEvZQAmsJeCWMsMSqCXgljIAEnBLwC1liSZQS8AtZQMlsJaAW8oMS6CW gFvaFUgQzwcCKUQxwom4MQMJ7DaCAAorQQKbiSBwwnKQQNweGG6iIOmkgASGGxhuVoIEhhsYbpaA xB18h20ZgmxPJWnsx8nus2SKmRQgAeFS+j2quaMI1qbE2KhaAJBAHZV+t2reKALHQmGso5ve03M2 SLpJQRLoIwHMo7xhBPaRKPZR/kgChSSAgZQtjCajqfmM1oPHUXdCcgvgdjd6sR3pOQFDMLBg58Z2 cOQpNTCyYOdGZlACX0TYuZEZlkAtwc6NDIAEsz/YuZElmkAtwc6NbKAE1hLs3MgMS6CWYOfG7kAC 1ybglrZDU5OPJlBLwC1lAyWwloBbygxLoJaAW8oASMAtAbeUJZpALQG3lA2UwFoCbikzLIFaAm5p cyB1HeSsyqrfe6D/+V5PUjcAnftGY0M2g1i08Tt38TudRJpOLA0NLhG76kSvbmK/3UR+c5SffgDK yGU7dtmJ/9qJ/3q/OAElcRPzbt4lbnZ4j3d4j3fij8crHq93vNqRWgfROJtKcxaTPHn3Lnm3w03b 4aTtcPPtcPPtcPPtcPLtcOvb4da3w61vh1vfDre+HW59O9z6dhL1jUQ1TcaoTAYdTIaQY2OCsVGd 2NA8bHgVNkoGG+eAPayePXCcPTeaPfmXPb6VPYKTPUmRPQuPPdCMPZQqcbZQ4hyZxJkhifMhEmcB JPZ9J/b4JvZqJrbdJbZPJXbCJDY0JFzTEx7GLB/PMqosLcZSG+wMlZ1jsIYiO9izvqastyDr8sW6 7bDeF+z6ObsIyixkVWzzebTZ0FHZLHXgk2uRO9KTY1UG3Z/279V/HaOm3X0hxgKyeo5tTkbOG/pj Rcb8gfDAz5IUoDr24yL7Bh3Ss8nftNC6T5Iq2I49NO0wD/Q3FNau9xDUrbZj3dgSDmJWJPn1yR4M wIj8oEbkJuYgi4SjGAJMHi6AQwNkEAzwsdEGbAiPjTYfG0BxATbaKXqjDXoDsMHRG5TpAWSIjAwf AVxc+FwhIENoZDSbKdiAJRDARoePjRvQG8Jj4yZFb9yA3gBscPWGo/wEaIgNDQoBPjKaQG0ANpqt NHQAuQHoaLZT0NEB3QHo6KTpjg7oDkBHh687wCAVHRp8exTMUUBGmjUKxihgI80WBVMUsJFmiYIh Ctjg2qHU5xuQITIyfARwcQGn0wA2QhSk4QPW2wAfFAUp+ACfUcCHj4I0fID+AHxQFPDxAee9AD4C FKThA/QH4KOTan/AsSyAjwAFafgA/QH46KTZHxD5CfARoiANH6A/AB+p/AcEaAJ8hChIwwfoD8BH Kv8BsZQAHyEK0vAB+gPwkcp/QMgjwEeIgjR8gP4AfKTwHy74BQmODXfAxwX4BQE23FS/IIhDBfiY oyAFH2CXAj7S/IIgYBTgY44CPj6AFwN8pPoFQWQnwMccBSn4APsD8JHmFwQhmAAfcxRw8QF+QYAP N9UvyAW/IMDHDAUp+AD7A/Bxk8p/gF8Q4MNN9QtywS8I8DFDQRo+QH8APlL5D/ALAny4qX5BLvgF AT5mKNg+rPQiCmUi+GAi4lwizFgithQTUCgRRyYRPCQRMSIRJoA9Gz55InjyFOjkyb/J016ZEz4T BzsmTvNLHOGWOLeLOayJc0YP51gWzkkcnMMXOPvtOVusObtqORspOXvnONulODtkOJsiOH7wHNdn jrcrx8GR8WnjuDJxvFc4DgucNWrOsiRnJYqz+MDhmzkUI4dV4hAJnLkjZ7rAsRA5RkFyHHif6MC1 3aIDJyPk5hRz+Hcj0yx/RaaFHPrL39AiajD6bVFWBjGN/fzjMYOpor05g0H/gw76hxcXN43T5kaD fxQM/sAfYiBmEjoQqU0YazABCIcXos2B+GyiI6LDQQREZRMYETc8HQGx2ERGRFJHQJwtkSHBCbDl QHQtwESbhwkIZCAyJjpcPQEBDITGBEdPgDkhMCS41gQYE0IjgqcjwJQQGBE8SwIMCaERweqIcMEc ICEkJGbS52EC1ITImODqCTAmRMZEh6snwJwQGhM8PXED9oTAmLjh2hM3YE8IjQmungB7QmRMcO2J G7AnhMZEUk+4wE8IjAmXy0+4wE8IjgmungB7QmRMdLh6AuwJoTHB0xPAT4iMiRuuPQH8hNiY4OoJ sCdExgTXngB+QmxMtHc4EmK+yTi+szS+qzC+oyy+m4jZScJsImD8xxnX4bjXaNxjMO4tFvcUYrxE GAcBZm2YWRZkVoSYxQCGB2YoQIb9YSb+zJyPMfcZS48Z5Bn9Hu/a77PbP5tN9OG+dtNCBMrjn8Fd 9LfFPvpsdtHPcvZLehoPBskiqhkVcUSzDw4aQIf/RgrnkIQMi/F1Paj1D7f5P3KMw9g/IGZ2HMB/ V1Dss8kAEMNGcD4AzTx+YNTYHPWVh/7P+y4Gq+CDnx1Bimr++HaKd0TJDBKpSLkDpJQDKXcZIwUA 8YFHHiLU/7r8ntvYE2YPSCkBUsbTCe31uUElzB+wUgKsON3hzxy1SpA9Y6k4oz/psNQGoxaM2jlQ QkykYwXMWjBrE1jZmikdLw56jc+xOZOpuP3FjLFxNcpTcBwgf8gjPzmc2a7nfTJZmhZyx87PHOk+ mn0BdF9QDJhHJTGlw8EnSxOJIiRqTdMiwKAug0H98/5n9hzxAi5h/oCVj48VKtTcgOJnDigpB0ru ne+5AoXmv70xHRslGS0YBXr8dT62DbwwJDOzgSNZGtlkGbN06S9q3CNg96Pu44fch8FDYJYu2BH3 vhp5SyxezwK/AB6ExAMVfgIQgAdx8cDCIYy2BHAQEA6+7Fk4uAAHYeHgDnaYgUSj1sXCzUVjxEWD uwVYK2wKwrG9d/bG9enyNR7pd8dUYiGn3u/aw5edZwBBnqHxnweYoP/n2v/nSrfn44H0igSpsG5P DqBwFMls6248q0xR3ZIUlpGT/Kw/+Nno2fRuv7u+WhN38GgqaOzYJxaavI1PAj3r9iz7pGk/Tu7R dNwfnXTGY9u5T0oRTZzu8KQ1apM/7njk2vekTt0ncmfYGfYmF+T7/aLaW6iDyXRoxyvuDEg3ntU8 9hrbv8V2r2Rk9UqhmRK8Bdgo5bdRWAgcBZI3edAkSPARfHgRwvfwIgZOehlBZoilh97TPVi8JUDT zoggl2eXX0llb694mW1hUQfBTueGdSR4mDvo9vuAO8BdgLvbK/I6qcjLEnqAsw+Hs9HcbJvbPesY Tu6b2xs+jo4mvYFNHllhy80/kfEwpqQ2m0asW9pmuR72x5Nnx+5a5Ntgl7lC0rJc8cBDb0ivnelQ cp/RhrnP5UfM8fV625H/xgE9cjd6sR3puVKp+N+QjyrUnaDW2UlN/oxaJ4omV3aZ8iRbI8s5T+lm PDDfAQ2+p/MdmO2A1bkWHmCuA6h7F9TBTAdQBjMdmOl8zJlOc/RnCVZ3Ut8CVndAi+//bCeEL6zu gOW5ASJgxgO4eyfcwZwHcAZzHpjzlHHOU7oZD8x3QIPv6XwHZjtgda6FB5jrAOreBXUw0wGUwUwH Zjr7MtNZb2de9BUTR97pGZ3P8SsilQm0WaQM9FuWp4CE2jL9YL1qtsWwu4abzRYE0vr4p4POZ8v/ EbEDWqNJ6sCfqUEZRVMH0CQEmvKHUhsUU0mh1C5cLbVBLYmApdyB1AFrqQRAmu8ceWdrqQPWkiBo yh9KYC2VFUqFW0sdsJaEwFLWQNr+CDJzNH5rRkJ+zO/cxe90Emk6sTSRqM1WLGqzFYvabMWiNlvx qM1WPGqzFY/abMWjNgeX9+PppHn/IPNuYt7Nu8TNDu/xDu/xTvzxeMXj9Y5XO1LrieO3r0KQGWnR 2N275N0ON22Hk7bDzbfDzbfDzbfDybfDrW+HW98Ot74dbn073Pp2uPXtcOvbSdS3N3yxnclCNvM7 HfZOJ5Gm04meuxcytNFr5vc283s7/nuHeb7DPN9hnu/En79hyr9hyr9hyr9hyr9hyr9hyr9hyr+J lc/0rNkNzN64i93osI902Ec6sUfcwfd4Mwc3OvEbHTZFh0lxw+Zxw+Zxw+Zxw+TBVoOtRVzaLiNt d8BWkq1jXNouI213wL4C+wY3TPk3TPnsC7Lvd8OUH0j7fQ60r2YS2T22gGP1Vcd+zGXlZoOFMVIF 27GHpp26RLZruCa2hINFtmTm80xznYzQAylyNLTR48hBf9DCD6RMQgmklJ69PLM5bnSepb+g6NBf /oZyW1D0849Ufx15ZPOaYcnkHYnBOAukkMSekVGoML8kGv0sx6Bni3htuQY9i4SFo3n+MSvmAPVc NB1T2VndiS1BMLr3CEZXtFwgQEpUqxt+w89L4Dd9aY6FXqm2/eJ219t5Hbv7u5aVH9B4e4cl5Nju UdhE6j7U57J126Zuls3rr7N6KVvV66E3PFg8vwXW/AxWw9ynoh67JmlHvxXOqbST+lzdrhLc3OPq /KUf0SdapsWEfKf54IvpyYZIGh+W7URrNSUHBUdHUfGbDCYgGJfQmIjH45rfhIhcYoMiFpRrcRdA ITQokpiA2FxCYyIenmt+EzbbCY0JN4GJV5vypQCKDwqKXTERip8Bhe04Y2cIoBAUFKH4F6AIaLpF rDGAhWgDSBwBGzlZ+ZiSugzTFec4mNltfFoTN2jjpgwziDHqiwUuqfYxMh27O6EU3kbV38Q3YPbI ZqlnvgEJznFzfnYtZlMpiNl8GIDKEJjZpOI3k2w3YEJotpuHCQcgITAkHB4iFICEyJBQkkw3WBNC M908a8IdAyZExsSYiwmwJkSGhMNDBFgTQkNC4XnYwMghtIfNC8+XAiAhsisF1+cKICG0y1XSkwLm oUJ7UvDmoZMHFzAhMiZcDiaAmxAaEzxuYjIGPSE0Jnh6wgE9Iba3Lg8ToCfE9tbd0dMqmKPEfS5i q+3xhdb4Elt8cSVGq8cZ1TiXFmNR4hPo+NQpbjTHzaX4QBlXkfHOsaH/V4FOXEohTlw4byeucBtx 93X88AbKSEBlFAOAyeLCtSeACnFRQcXPmi1ddwqbjMQ1W3zxJzDxOobRQ2BMUPGzmKAnVgImhMWE L34WE9YIzAlxIUGlzyJi6Axg8VVcSPjiZzHhgJYQGBIOR0uQe08ACZEh8cTDBMxDhcbEgIsJUBRi gyKpKVyz2wdQiOvgR8WfwMSf3TFgQlxMUPEzmOhZ3UH3FUAhKihC+W+3JM+so8bXz+IrJ3HOPM6W xniyOEUSmxzH50Vxi5ixheKDYFz9McDf39V2nONq+5aHdy/LcYdjtyt/zfaz//kFUeO+XLYub781 LpDfIqpUlxR9dqI5uv6Cbhvn7cvrFlIO8aGOyBcsHynGkSJjmTyCMD5WVZLNX4VrP/L5lOmHwPAo epS8dXV1fnr+rdG4+NqQyI0sypDJx9A0/2/ViP+VZaxho/oJY83ABq6qqvpJxqquaJ+Q/KmAz5SM MA5Cn15elqdb9fsH/YiqgdBZ4+tpax39o/D1j+LnIqQGKiBYxCJYBj1ibhYlgsbgCGwts2s+27b1 ZEvPaOzYJ9ZOwSPipc3irt1Mh196Q+u8sZNLHpv3BjHXoq++CJCh7xgJJ/KK3elrr0/DhVlm3+4O p2P/+9S1HZoom8A7m7y1X595EJwMIjz5GUbiHqVGwFnELds+PFJYGHnV742bVqM5e5HDR2RdkS+P vX6Wgatowx4FJa3xhlYmr5hS8NKYcDM4ZRQYLgJfTpguI7POEQaJWhqdK7NX2TVcUpCTT3AAl/Hh 4mubixDbBF6P/pgjjdahOOKDy1H4qLntIOOHUA3GvHv/fFZ/aEWTt/GJX78sB6BoiPhokcDHfewY 8VSWpBnOG6R9paONwsD7GKZVDSMogWYUHVUj9Brqw4he/O8KSn42Ml4SSmejwM/osD+ePDt21yLf BpXVEx567UyHkvu84bRn/vLEhOTOfSrtU4TaZ2EbXIV/W+Hf7+S/bn/83KUXD/aE/D2nzzbIs2GS 9uVVw3/0S/P6x23l5ASh4D/6Wfdv2tdKBbXn1cGGTP8osuz/NXRd9W9LwfXsb/AJv5KbCiZ/NSzX pGp92+wInujnUJHk2V2ZfN0yO6zUssxO0Y1ZHvUqJRU0FetSfdvsdKzMs/Of0lSjJulbth2WFW2e nV71s6spUtXYLjtF1mqJlzUkpbpddppcN+ZAkQOg1GWptuXL1nBdmWeH6352RBSKVqlcDnuTXreP zhsntEm+n31Gg8f+aHwya44tpVU1Fu3h44Dkp0nVbYVvaPIiu5qfXbUm1bZsXlSrGwnhY0nTK5Uz 253QxiD1j7WFX3dznKqvNot3GlV0mQSCjUy41EyyWh3BlbHQd2oVKavyso0dC4tYO5HIsIgFn/f7 8Nb/bq5vb6//3rgpZv2PjKeGzq7/GXoV1v9g/W8f1v9UWP/LfP1vGzPGyifyPVgvO3UBsF5KN/7f NG4ldzrIrowV47+sKxH/H6zR8Z/MsmH8L+JzeXEio5vrTvvkwHqirPsBOu20v50c/OO527ffjuxB b3JPprqofXpy0CZ/zk4OWgcBB3vVaZ5eXJxg9OOm8YX8aZ62T3TU+tI+/3aioMvw7xf/L0bfz65O 70hh5O9li/7tnGAjyKhF+eSrzomGrs5OajJq+f9+9/9tYo1cyujHl7Pz6+at/+30rPk9+BLeum1/ aV6T0uUgv+bF7PKUvFnjhnxpXpyekUqcNdqnLexfX95+p3e+d/wvpLibTovSoeRra/E1eNHFdfPi vH39g3y5q1XnP1Py9EST1KpeVzX7/8pq5SP3/9sGXU/Iyvhf1f8NRVFUje3/iiZD/wf7fx/sf8w1 fpQ62P+F+//dTIdXVxf2Yw4uf/4KMSwS575IfJ2yTrxiPXhzX5jBwLW7jvksmZXLIfJ978IGtdCj MxqgDSQQMOiDwbjruLb0fKx9Ti5Bb5vnxHYnk95g12xT3/5Y1Y4rG9fryR7OXhbR5psOzUlvNEQH X+1JEweryP2Je7Bb1lX5GP3ZdYYEu8eo9zQc+Z3YsSdTZ4heuv2pjUaP6ODRNbvDx4PPyLLNftch AvyzN3lG3QntFtOJ7edxPx1OXdu6d2x32p9kJ/OIfPS9kk+kYmkiuu159sHOueP6/kkp4+aMiTne mm1y8+vV1XfbGdr9g13zVqrqWo3pvrkTe7BZY2b91pm9tFqv5vbSMG6Xbtw2CQrHuw/ey/M91mRQ u0IZRIz4MVhFGXYmDIOuQIPu7oCJv3gTr99tVmSMNS233pP1axOtseF4serla7UP8/LEZL83u242 L65qxocB++zFL8+zefU6/kCvTl9czuS9NUP5MO9NOvrl0J00R09fyOibyetX62uah0820azv+vZP s9l5NjpO16P20WQ0Qo/2n6jrPE0HxH52/V2n5L9BN5eum+GLVNccqvYBwXSzTGt0Phq/tc4yeXtF xrWPNFC3poMb+ykbG0UxPtC730yH1C7N6s015aOoLTPT3q7U6vJHGq+upv1Ty8rm1esG3mO4k+op H0kT+Q5a2ZiOMtb3+M1VWcmPCYEdwftLGtNtvzEwoDg0RiC9vZfeXHBzmUlHi9uHY2RleqoSyfwp nnvIHN/Qbdq9IWriYO8zcnuejbouOkGG9v1Merd0N43TC3TZQq3O1VnjhjodfW1cXaGbxtfL23bj 5pYkxMZndHV6fn6CP6N+d3KiH1d+3FyfXba+oi/XN+gKnbYuUAt1WjfXzaZ/9/S8fX1ze0zrMHw4 QVXlMxpMTxBSP6Mh/at9Rj/pX8xkTLdwnqiqoUqGlvqwuuxhTTawVK8lHlbCh42lJeuKLtVx4mE9 fFhZ+nC9pktKPbVkfenD1aohGXriYW2dkjVcMyS9mlry8tY2FCzhemprr3hn3ZA0NbXkFaJSDLqL N+2d8bKHlXq1LqnJh411HlYVWZPk5DvX1nu4XpOURMlYXuthFVclQ0k8rKz3sFGTanLlttFsnLcb F76r8ufAa/nXEAS/VSpfLlsXtBd2fvwgHfrsukN65zXp4GfHldZZ7O75t+vbRgud3qJjVJN5eTSv /8HmEVZd3wyiMmm02vy9tdpmD1eJQsC09tH6RGuvK4s6nzVu26SmvlaiHuGN9vk3dNtot+mPRNPd nLa+NtDv5AVq8mflX7E3Gj/Sojeomqpokj5XN7q2RRZKVfbPWQizMLbJoq5JtUUtalvVAkvaXPUZ 8jZZ6Eo0i22aE9ewpM7bwtimOekAQA+GCLMImlPeJAu1XtclvMhim+aUq4qkLvr5Ns2JVZnR6lsI 1Yjqua3QqevRLLZBp4pxpO9Xt2lOTF6kOq9FTd6qp9Yk1aiE2uGE5uF/n6mIE0wUYJDot0r75p9U W3zvuL6JE6b/8YXmGdG6s2L8HSe+IqG7SeIDOxnkdD38SZPjbavJkqyEv9Xiv6l6cJiL/5sSx5Zc XfykxduJWj84/MmIF6YaEp79VGNgQkY0efVPOF4PVVUldZYjjpemVjUJz6vPvpoq4ersN42Rkhx5 jqkLGf7qs+dUhXm7Wmh80d+YN5dJo2iB5KM1pfL2q/lbfOwIRowfp21i/7Z8+RPQ1QIVEmAtQFqg InqP9sR8PvFtnvArnrVBPew6nDRKmKZGD3mJplG5+ajVlDSzfGhP11PSqGEaYlcHA80sjZYsqyZL mpKSJlJWLVYfnVNnRapq0TRGMo1B2sdvwwd6jgtdk6OkSm9gW3SGEmRDDHI1EN1l0E1JJVrhN1+A NMG8wzavTy8OT9uH7esf6Nf+qGsddieHD6PJZDRA87SkwNlPk9E4rItWpcr2tn15/p1m9I/L9rd5 bmfX7fb11ayM23+c/qDzJNr50a9jx3bt4YTk0rUsVuUsChx0TfMk7Abawvw16pXvjcYPmq2fXWim kOnb3Mj72mg1bk7pt+Cs0+MK7JXLZa9c1kf7aLAjBDxLOfzyNIstIel5Hqt18CgVyKM0KvptnYnB mzSTlgRPUnE8SaNgycyLNJqpsU8OpOlvu7PzaDRrLNc+ykvv6kgXzVypGfvjjZL+1q3poOPazjmZ p7nyzi+t7pMLzlJRz15691euafv5ympNz6tiMH0p2/TFpTNay36dhTIF4ZZLuOZo8EBku8vD2Q8c i6z1fRor095451FjkXG9tpfvm5/fPzhV7bdT1QwEaI6GUWXxNd4R/jgnN3tDm5rJ7q49AWvRWe/k 2aY1oCj7YzIYD7sDPziW1R0+2c5o6n6moSboO9FEfwx+0knhGEKwfAR0+eYFWhgavtPe4ofDMXJ3 CgEFFLsAZsxjv/u0tRkTPJyLGRNmre/ZsM594yzMmDBjLMt7+cJYxmDIiDvUUBSgORwCQyb4yhgy X0ekM7Sf7cHOHUHXdrdinAGNbUObYnI1/tJr2uuE/3p1pw/uxJmFOwyfRIe9LTTs+elt4/bIpT4b rkREG4+Gduiig/+PwP8AHfidYJ0oawf+Uyh4Cq3/2Mw28CVJ3iX2aiO0qGJl7TxRCqauUWAm0CKo V8IJRl9IqefPtvlzxoZA/pA/5A/5Q/4fOv9XZHKie0LukPvHyT1pJO4QyZjr4pgR/7Jf1uxj/2nf rVlSxb3SlmmTsMfh6NA1n21r2rcPe0N36PLuKXkPC4rQNfnISm/nVvRv2q9je+j2XuzD0XjSG/S8 Lp3T0ye6/d7T8LA/Go3dExqeWfR6w/C7lUQnjm370pgJyoaCP2DBYB5mZh6GZoy/TrywWyzXpEaN imIGlxwh6aIWDrvoZ5XR6LT2n0K1gEKF/CF/yB/yh/whf8gf8oclkF1z979DEWsUoSrAy7yrBCD3 jXL3U3z4ImDd7qNRKPu8breoYuWPw4E5np4Ebn72mLR9d2JbEupQB7/DAX1D8uPIQQeHA7o9k6Yc uhO7a0mVCEqO8TGyHWfkHKOHrhU6qv5ar8q/+ZEnwoyQ+2dvYj5nYVcUv3B1vclyCafSsKpYeE3w h1jfPJq6jq+LaDGwQAoLpGAL7r2ZT8dMMrpBG8GCLSxRA/pheT26vG5xl9et2fJ6ZI3ZSl9eN8s4 NzT3f3ndhOV1yB/yh/whf8gfrCnIHWxkWOlYas3u80qHCTuUYC0BSDcg4IGAB/oXCn53+rdIGtLk 0pDmjIaMcHFmOg3pldFw8/afhvSAhoT8IX/IH/KH/CF/yB/yh2UE2IIDu3xglw/kDrt8YO1rPyiU fV778mCXD+zyEXxlDvbWwNIe2EiwtwYWV6Fg2FsDuX/4RW2Pu6jtzRa1Iyu7HuP6BqGf9jf00yz4 NVqEwQ6iDM5/oAeOosMhqpFx0UaO7R5ZV1ffyatI7nSQffhBUqnB4Mke3rtkhrj5HHn+rETy6g8g pGHZQhoOBj8DYAZCrtDgZD3SNmGrWujRGQ3QBmLoTkje94PBuOuQDJ+Ptc+VhA7eNs+J7U7I6Lxr tsub4Fg11grFFq8c6SazN07EOmzi8y6ZCjX6kzVj1aZmXV0z2KFrdoePm0UVzEzwESHp+yekSO3S 5HRLrL+DnXPH9f0TVR5tGhN4vEnb5OZXMrjZztDuH+yat1JV12pR940GM8w81OdGb53ZS6v1am4v PfhJbAtEBrCvjdZ1p31xeQP25Z7bl7Eei+L917cz4wn8w+0NnSJkR4R/Ob1sNi7Q+fXV1WnrAh0j aqiSCkwn7oROT8hXZzSdnFhECQ4GocE4duwTC12daDXUov98p/8MHui/Q//fn/6/fasb/AmvzBMs y+g/0ZFlvxwNp/0+Uv7z/+AKHUpI7nT158+ui07QgV8HWiIp7jsaTMlsbkj/+Un/uTrzy6X/fkbf /QuLpCOpH+O1TBYvozAhmRq6J7/8kqzLQeXCfuySToTIdG5MdLGD/g/ykyNrhIajCXKn4/HImYTd 63/9r0qWU8nGXbtx0zptotsGhXQonWN2NnHbOT9v3N5+6TTRl8vW5S1JeH2DorORSqVy1rhtIwLv G38GilSl+pm2mkEsDUWXNUlX0NWX5vUPP+nXyxY6O70lkLw6bV+RfL+TejSaqE0yuT2mEgr0Xjib pcC3BgPtFb/WZdJfqpL5KyngNzTukhZ20UPX7ZmI4sytXFy3GmHe0UylWRWvvyxqeYsoCINWaMxa oUEqcHlxQvJHN0SdnRyk1uEAnXba304ObiR03qfa4x/P3b79doDapycHbfLn7OSgdRAsB5KXPL24 IJD4cdP4Qv40T9sEXK0v7fNv5Ooy/Psl/Pv97Or0Lvh72aJ/O6Qhg4xaHQrKzolGkUlu+m1MgUn+ bWKNXJIyvpydXzdv/W+nZ83vwZfw1m2biIGUHubXvJhdnpK3JWqQZHNxSnOhxEYL+9eXt9/P/GrM vtx0Wu3Lqwb52pp9xeGLLq6bF+ft6x8kzV2tOivOx8CJIhFIEGVp/19Z/UwvyHeVXiB0/qV5+vX2 5ODwFVHhDh5Ipzikg9ChOR4fhEurRH/8ONlgPTXoNFomnUaV6+qamjBhJ5lkdB+H83mT5lTNJifS JL6S/Iwo93h98l+X3z+jSfekhSYPJ63PVFkqRH0Ngz8/gz9UYcmffYUVXJrkL82KaEDtM1WBymdf Bxokg2m/a1knmCTrTk50dEI+/4kGj/3R+ERVdVWqVnVZlvenIhWrdUZ0e01Gvx5iSZZ/I9poRLTs +A0Fvyg19GvwQO23ys71loNqy7FaY7/WNOWKWmsS1nEWzbdRNWpMNTRDwjWj8GoobDV0qVZVV1aj na9QsCpLWMVQjXWr0c4DG9V4NchMQtJxvfBqMD1FU1TJyL0aqzoKxopUV4yVtWjnKxMydBO1USu8 GgxCZQ1LVbX41tDYasgSwQethioTiywjO4EYMpnkVKnctk9v2petr+i82ThtdX6ERm9lHSTLfqPV ZL/RanLQaLWw1WrcZltv4NYUw5B0o7Zmj6oFFcmjHnUivtq69civGlVNqtfrW1RDXo3kNeugyjWp qtRyrcOKPk0Kl2RqzRFzLpjXsaCtXLb+fv2dwvnr1VVwUzr/TOdYJwfWQfgQ/6cKmae06JPMJLhx O0/kT2Br8jFJ/GXkoMlzz0XOdEje/dlGD2S2SeaflHqZ2I6LHkfToYX+tB2bzGnPz+evFGby2Z+z fabTN/LyZD6HjUqbZvjUfbFRF41t53HkDLpD0ybGqaZWdamuauTVgxmTRBLbaGibtut2yaSF8rYu eqbPPtj2EJmOHThwostH9Daaoi6px3N3TMxdOneq0Bp3H0Ykud+2ru+eSdI5KGD8yDu9je0ZFUJd Pbv9/oEUsg1qJhMn00IbPYH+RjkGe9Cb3JOLcMHzBh0qlRCN5x2KroEPtcdH+m/v8eSQNPww+NMf USkQoJKh8jPFKsYzrf15gUVtjkUfxT6IUzD8GZnuKU1qumfBH1/O3f74uUu/PNi0sEijZbD6Os/r jwXd4rvtTsdoMkIWEbuUpZQ0ua5sN9iEkiLDjCZjY9c81HpV2z2Pur77uxjG7nlU67vnUd+9HpQQ 3T0Pddc8+r2hjYh617PKqJpVRvUMmmfnymyhKH29HartYLDOXAnNh4UFmZZBzh9Zw+sZanh9HQ2v g4YHDQ8a/uNr+MELdRC1yPdntN6i+JEUPhSsqlqt880eXj2muI4ZqWu4dvY3tPBltf4qjbLJZ9dc rCuiDX3XlVfTHD38m+Y3TxI0U1hhc5ty5m0dXi6K2y6z3iP6ne7o+yWorWS/2r+gf5GqklnZEP13 5T/I1Ngc0wIXCTYrSiKZkXzsPlMUKWZeSsR3KlrYNgXNfbB6oWUwGPR7D5kOihsiZGYCWfN6GNvV I5ZraPh0p6+9PskYWWbf7g6nY/87dcKmiYLiqpkUt95b+/WZvfH8hWvb18DP8GCRTWs0eaa5EOvj gbTpiGhOyhf8Qco6kBbJNjdFIoXFuix9kVi/trJ4LbYrHoBfUhF+SUGGPlV3Fbtqxa6+p7gvDR7H g+7k+cR3PBqQf5Vge81auzUOzUDzt84Gg/sHebbTA0QuisgxiFw0kd+ByAUS+fl5ZzBotUDm4sn8 /jtIXSSp+zLvYrDjhJT7Hchd1P4ORryY/R3kLmZ/hxmcmP0d5C7eLK4NMhdP5jBzF07Dt2HmLqrc YeYubH+HGZyY/R3kLmZ/hxmcmP0d5C7cLK4Na+4Cyhxm7qJp+DasuQsrd5i5C9vfYQYnZn8HuYvZ 32EGJ2Z/B7mLN4uDNXcBZQ4zd+E0PKy5Cyt3mLkL299hBidmfwe5i9nfYQYnZn8HuQsk9+n4Kn6E yXpPAzT2BRqL2MCyH+mUF84mKmr5XruXwYgXspNj6OQfvZPjTTo5WO4CdvI76OQfvZPfbdLJwVwX qpO3wFwXxVxv+Z0cg7kuaCcHc10Acz3SycFcF7CTg7kugLke6eRgrgvVyb+DuS6KuU5Fje/p/0De oshbAXkLJW8V5C2UvDWQt1Dy1kHeQsnbAHkLJe8qyFsoeddA3kLJuw7yFotvkUHgYgkcGDbBBA4U m2ACB45NMIEDySaYwIFlE0zgQLMJJnDg2QQTOBBtggkcmDbBPF2AaRNM4MC0CSZwYNoEEzgwbYIJ HJg2wQQOTJtgAgemTTCBA9MmmMCBaRNM4MC0CbbHCJg2wQQOTJtgAgemTTCBA9MmmMCBaRNM4MC0 CSZwYNoEEzgwbYIJHJg2wQQOTJtgp7sA0yaYwIFpE0zgwLQJJnBg2gQTODBtggkcmDbBBA5Mm2AC B6ZNMIED0yaYwIFpE+xcXWDaBBM4MG2CCRyYNsEEDkybYAIHpk0wgQPTJpjAgWkTTODAtAkmcGDa BBM4y7SBKMUJOwoBxAUIIB4JOwryFkTeCshbKHmrIG+h5K2BvIWStw7yFkreBshbKHlXQd5CybsG 8hZK3nWQt1h8iwwCF0vgwLAJJnCg2AQTOHBsggkcSDbBBA4sm2ACB5pNMIEDzyaYwIFoE0zgwLQJ 5ukCTJtgAgemTTCBA9MmmMCBaRNM4MC0CSZwYNoEEzgwbYIJHJg2wQQOTJtgAgemTbA9RsC0CSZw YNoEEzgwbYIJHJg2wQQOTJtgAgemTTCBA9MmmMCBaRNM4MC0CSZwYNoEO90FmDbBBA5Mm2ACB6ZN MIED0yaYwIFpE0zgwLQJJnBg2gQTODBtggkcmDbBBA5Mm2Dn6gLTJpjAgWkTTODAtAkmcGDaBBM4 MG2CCRyYNsEEDkybYAIHpk0wgQPTJpjAWaYNRClO2NE76OUfvZffbRJ2FOQtiLwVkLdQ8lZB3kLJ WwN5CyVvHeQtlLwNkLdQ8q6CvIWSdw3kLZS86yBvsfgWGQQulsCBYRNM4ECxCSZw4NgEEziQbIIJ HFg2wQQONJtgAgeeTTCBA9EmmMCBaRPM0wWYNsEEDkybYAIHpk0wgQPTJpjAgWkTTODAtAkmcGDa BBM4MG2CCRyYNsEEDkybYHuMgGkTTODAtAkmcGDaBBM4MG2CCRyYNsEEDkybYAIHpk0wgQPTJpjA gWkTTODAtAl2ugswbYIJHJg2wQQOTJtgAgemTTCBA9MmmMCBaRNM4MC0CSZwYNoEEzgwbYIJHJg2 wc7VBaZNMIED0yaYwIFpE0zgwLQJJnBg2gQTODBtggkcmDbBBA5Mm2ACB6ZNMIGzTBuIsvxhR5/G V37YURmELpLQWyB0UXs6RJUWsKeD0AXs6WDICdjTQehCCZ1O3MCME07kROJdBzlohST6vQf6X3fS 77pSN8hgZvqPopc4fnk3vzw/7wwGrRZ7ff99cce/7uJonuG9O849HC1skQ5z0t1x0rE1a7PXTM3a nJq1OTVrc2rW5tSszalZm1ezdou9jteszWmzNqfN2pw2a3ParM1psza3zdpt9pqpGafN2pw2a3Pa rM1pszanzdpMm03nNETshnyv3cuJu5H8Y8kwm+yOmyxyt8UW2vKTYabQFltoJBlmk91xk0XufmcL 9YMkBNwbc1fh3lW5dzXuXZ171+DerXLv1rh369y7sxib7G3+22H+62H++2H+C2L+G2L+K2L+O2L+ S2L+Wyr8t1RSZMh/S4X/lgr/LRX+Wyr8t1T4b6nw31Lhv6XKf0uV/5ZqClT5b6ny31Llv6XKf0uV /5Yq/y1V/ltq/LfU+G+p8d9SS+mR/LfU+G+p8d9S47+lxn9Ljf+WOv8tdf5b6vy31PlvqacoHv5b 6vy31PlvqfPfUue/ZUwZR5Qq5ipVzFWqmKtUMVepYq5SxVylirlKFXOVKuYrVcxXqpivVDFfqWK+ UsV8pYr5ShXzlSrmK1XMV6qYr1QxX6livlLFfKWK+UoV85Uq5itVzFeqmK9UMV+pYr5SxXylivlK FfOVKuYrVcxXqpivVDFfqWK+UsV8pYr5ShXzlSrmK1XMV6qYr1QxX6livlLFfKWK+UoV85Uq5itV zFeqmK9UMV+pYr5SxXylivlKFfOVKuYrVcxXqphVqndcpXrHVap3XKV6x1Wqd1ylesdVqndcpXrH Vap3XKV6x1eqd3ylesdXqnd8pXrHV6p3fKV6x1eqd3ylesdXqnd8pXrHV6p3fKV6x1eqd3ylesdX qnd8pXrHV6p3fKV6x1eqd3ylesdXqnd8pXrHV6p3fKV6x1eqd3ylesdXqnd8pXrHV6p3fKV6x1eq d3ylesdXqnd8pXrHV6p3fKV6x1eqd3ylesdXqnd8pXrHV6p3fKV6x1eqd3ylesdXqnd8pXrHV6p3 fKV6x1eqd3ylOr/9xBIdTywJ8cRyHE8s//DE0htPLPUw44Vjl+SqYpvPo81Iw8pmqSejqfmMLHJH enKsyqD70/699q9j1LS7L73hE7J6jm1ORs4b+mNFxq5jHj2QTI+e7MHg6HvjptVoHgQZVkmGrdHk mWY4GaEHG1mjoY0eRw76w5q6tkMfOZAWiXcsPSgWHf4bKcyr0cwblN3dKXefvA9ldXX2X4GsDoHQ z53Qz4Gmv7i4aZw20RotGgc4rVkoezMGiMuzFgBCUECEsmcBARpCXEBwNAT9FeDwweHgF/W1cXVF Wlc6Qv9dQbHPVmjxgRHHyld70joDsIioOwLRJ+FwDoAQFxDnCUjQn32HD0DER0YEO4AcXrRGbac7 dE/vF9/P7rcfWO7ueMBpA3BKDZx8YNMGfVM+2OSvbdqgbcoMm+xBQ10PL//rO6Dmg6Nmy/nwTPxJ UPzXJYBCYFBQ8XM1xb0JsBBaVxAAcLUFAENwfZEARrff67q2dQ9sfDnY+Obl6W3j4j5/Sv4UGDdg 3Fbh5pQ7nzkFzg04t62BA6wbsG7bAQc0DvBuG8EGGBZhV5l57AqwsALDIcnAEoh8vwQ4CKodiOhZ OIBuEBQMSc1w9uMHgEFMMFDRJ83ItjkGQIhqSFLhxyFhjvrKQ//nffcOUFEC0v3Ht9O77eARAoGB hzP6k95tAz5Ex8cMCWn6AwM+yoEPnI/+AHyIjY8U/TGeTqj6eIDh5ePDgwZL2XJ0CWCQgg1QHeXA Bs4BG0MARznA0coDHTKAoxTgkDPEhjvo9vvf78HLsATYuL0ilfl+n7eLoQ8ZQExpEJM7YB56TwCX Ugw+l1+LAMtw5PwEwJQEMK3rm3yGpG1DlszPxIsfiBY/DSt6FFLsqBvmoJP4IRfxkwviG9Lj+4yZ DaTM1sHEprHEZiHeJhHG/5/x6WY8dRn/y5hfXcyrKuZTE/GoiK2oM+upybU0zvJJkjDncKQJVixB hSTnv+yc533OqczspMjfjcxy0rfKaTId2tGswrftDUn37/eDnLVMcvazUrPLStkqq4feMPL8Fgdy +hmYY7RRlY8c2z2yrq5uGreSOx2gy9ZtmxpnzeuvR9LOr1P5a7af/c8vGGC+XLYub781LpDfIqpU lxR91rTo+gu6bZy3L69bSDnEhxiRL1g+UowjRcYyeQRhfKwqJJu/Ctd+5PNOXYCOrBmj/53e5MF2 J63WvVJ79f9fopdql/Cl2mWUVHuPJWVt9lLob8g/ODtueKjb1SDV8vhjlv0B6rloOqZHglvdiS3t k3HyCT65f0hDH0V7jHXVaZ5eXGRahkw+hqb5f6tG/K//UdRPGGsGJv+rVsl9rKiG8gnJRTTAlHQD B6FPLy/L0636/YN+cEWvKEqtLhmqqmhqRVWg24nd//9OpmafCu3/il6d9f+qagT9H+vQ/4v4qAgb iEz+MKphhFW9KhETKqQfX1r3+BXf465kol9uJHTep6TvP567ffvtlwqWdSQjhT5PH1V0Q8La/FFi kL3SKE3cR+v0IY0WS580VEMylEihqvKqpTwJHTb//t/utBoS+V5M/1dVQzdkpv+rclWG/l/ER1TG DJ01vp621uHLqny+TPVzAcasgDnxy2xOHHDH1EBBY8c+sbKbG7+EAaVgITb3hdjrlLXYVWuuI2ST p+6f7a4ljdZZUo0L+GjxtFnZ6eljdDlEj9OhOemNhujgqz1p4vOu+Wzf9jz74HjHzPXaMfqz6wyJ NI+R7yRAYe3Yk6kzRC/d/tRGo0d08Oia3eHjwWdk2Wa/69gW+rM3eUbdCQXKdGL7edxPh1O6ckr6 zbQ/CRAO8N5XeL/OYRBFOuilvRfc4MW1u475vJVamj+8jVZaPMwoJffF7LrbKaNInlhX11JG7ps7 sQebKaMMX3aQycsqivwRXpYMN+fkbd1dX1c16usNNE/2xC3wbdWanFe9VpudxDqd2ZpuH/d7Dwu3 j/xsXerC0p2+9vqzoufl6ttHSvUzjEVJfZj2+lZv+DgCh8MP6nC4phuhL/qjuMi39iN87E/dZ5Na t6HLmX/DN3fP3oiUnPC2OZ5OeoNZoj+7/X7k8tV2SF8PL/rmzIsvikh67bpvjv3YJyif3SDKYfRi t+OXiytzNH5bXHm2M1pcdS1rcUGM5f7sauLEr2IJzVG/3x279uJn9k73dfwQKdT9szue18/9nwk9 BnDx6+n4bHI/96Ak73c6Pk3eiLxe93XoDCK525NItclEI3bVmXkBsu1mxdvNirWbFWs3K9ZuVrTd rFi7WbF2s6LtZiXazUq0mxVtNyvebla83Sy23Sy23axYu1lMu1nRdrNi7WbF2s1k282Mt5sZazcz 1m5mrN3MaLuZsXYzY+1mRtvNTLSbmWg3M9puZrzdzHi7mWy7mWy7mbF2M5l2M6PtZsbazYy1m8e2 mxdvNy/Wbl6s3bxYu3nRdvNi7ebF2s2LtpuXaDcv0W5etN28eLt58Xbz2Hbz2HbzYu3mMe3mRdvN i7WbF8db0C7nMdGej4b/jtyIXfupI9e0HaPXtIGi1882g2dz3H+1ei89y47e6Q1fbGfCFnr/iu/f 8CKrKGr9FjIjQHomN06TN+Ivet+968mJW+wNzN6QE+3jZ3T/EMnN/yl524dTSuowI2423EzSssC8 LDAvC5yWBfdluK/C3qRSSX03/7W5TcRtoJTm4TQOp2lSGobTLJxGSWkSToNwmiPRGGlNgXlNgXlN gflNgZNNgZNNgflNgZNNgZNNgflNgZNNgZNNgdmmwGlNwe043G6T0mk4XYbTYVK6C6ezcLpKSkfh dBNOJ0l0ETk+GJ3HxoWI0vQYpesxStdjlK7HKF2PVbpeQul6CaXrcZWuF1O6Hqt0PVbpejGl6yWV rscqXY9Vuh6rdL10pevxla7HV7pemtL1eErX4yldL03pejyl6/GUrpemdD2e0vV4StdLKl0vTel6 PKXr8ZSux1e6XlLpekml6/GVrpdUul5S6Xp8peslla6XVLoeq3S9NKXr8ZSux1O6Hl/pekml6yWV rsdXul5S6XpJpevxla6XVLpeUul6rNL10pSux1O6Hk/penyl6yWVrpdUuh5f6XpJpeslla7HV7pe Uul6SaXrsUrXZRWPyyoel1U8bhL5bhL5bhLibhLibhLLbhLLLotlN4k4N4k4NwktNwktN4khN4kh l8WQm5S0m5S0mxSpmxSpm5Sdm5Sdy8rOYmVnsbKzWNlZSdlZSdlZSdlZSdlZSdlZSdlZrOyspOys pOyspOyspOyspOyspOwsVnZWUnZWUnZWUnZWUnZWUnZWUnZWRHZFbf8Md3/KaxLbPsfft1/sPk7Q 28b29HYk22CfASJtMVxsSc0s2yvyD92tOwnyrmWYt5/ZH/MC+Dsxarvtio29y+Hj4nUWUsj8lVqj yTPNi7zIA5HLaGijx5GD/iAlHuTwVsEr+f0lXxn5RRQoJb+8ssnJHLtmvmKiJRQoJVpc2YTk2jmr O1JAgSIipZVNQkNnoOQrIlpCgTKixZVuTHKng5yHJFJCkSMSPbGhZEKy8jbtrEJNO6t8ph1dust5 OCIlFDkekeJKJyQyD81ZSKSEIoVEiiud8T0a5zxHoiUUaXyT4ko4k81/IlvwPLZ0Qup1B93XfKXk F1GgmPzySiOnXQ+mi7fRv5HCOfcus+ap+ieyhPnzRZ7l67g5UqVu3lSpWyhV6paPKnXzp0rdgqlS t4xUqZs7VeoWS5W6JaRK3bypUrdQqtQtH1Xq5k6VusVSpW4JqVI3d6rULZYqdUtIlbp5U6VuoVSp Wz6q1M2dKnWLpUrdElKlbu5UqVssVeqWkCp1c6dK3WKpUreEVKmbO1XqFkuVuiWkSt38qVK3YKrU Bap0OVXq5kyVuoVRpV6OVKmXN1XqFUqVeuWjSr38qVKvYKrUKyNV6uVOlXrFUqVeCalSL2+q1CuU KvXKR5V6uVOlXrFUqVdCqtTLnSr1iqVKvRJSpV7eVKlXKFXqlY8q9XKnSr1iqVKvhFSplztV6hVL lXolpEq93KlSr1iq1CshVerlTpV6xVKlXgmpUi9/qtQrmCr1ykWVLswhM3fzzizWvjOBz07ls72c +WyvMD7bzJHPNvPms81C+WyzfHy2mT+fbRbMZ5tl5LPN3Plss1g+2ywhn23mzWebhfLZZvn4bDN3 Ptssls82S8hnm7nz2WaxfLZZQj7bzJvPNgvls83y8dlm7ny2WSyfbZaQzzZz57PNYvlss4R8tpk7 n20Wy2ebJeSzzdz5bLNYPtssIZ9t5s9nmwXz2WYZ+Wwzdz7bLJbPNoHPXspnmznz2WZRfLZ/hHAG ecWONl7E3tsu5zD03tbnI7v9lz7eOQDgsmM/jByP/cgeTSuO/ajmfuxH5q+0hv6o5nnsR14yWsL9 Vws49qMcclpw/3mJKZ37r+Z/7Ec5hDTn/vOSUSr3X8392I9ySGjB/eclonTuv5r/sR8lGZPm3H9u Q1Iq91/N/9iPcgjJytu0swo17azymXYL7j+34SiV+6/mf+xHSYQ05/5zE1Iq91/N/9iPkhjfc+4/ N+M7lfuv5n/sR2lmsvlPZAuex5ZOSBHuPy8pLeH+qwUc+/HB5ZQVd5p27EeWJKOf2YpjPzJ8nR3I 32Re2lZ5PfSG61DIJNmMNrb6eA5Nbbu2X7PQOcmdiOi3A18dsubzbNboDO9H0Vs5UfRWjhS9lTdF bxVK0Vvlo+it/Cl6q2CK3iojRW/lTtFbxVL0Vgkpeitvit4qlKK3ykfRW7lT9FaxFL1VQoreyp2i t4ql6K0SUvRW3hS9VShFb5WPordyp+itYil6q4QUvZU7RW8VS9FbJaTordwpeqtYit4qIUVv5U7R W8VS9FYJKXorf4reKpiit4Ci3ySIYfYUvQUU/RKK3gSKvjCK3gQvevCiBy968KIHL3rwogcvevCi By968KIHL3rwogcvevCiBy968KIHL3rwogcvevCiBy/6nCj6yGHjRo6HjefUL8xCqVKzfFSpmT9V ahZMlZplpErN3KlSs1iq1CwhVWrmTZWahVKlZvmoUjN3qtQslio1S0iVmrlTpWaxVKlZQqrUzJsq NQulSs3yUaVm7lSpWSxVapaQKjVzp0rNYqlSs4RUqZk7VWoWS5WaJaRKzdypUrNYqtQsIVVq5k+V mgVTpWa5qNLEYeM5mndmsfadCXz2moeNZ89nm+ByvsTl3AOX88Jczj04FQZOhYFTYeBUGDgVBk6F gVNh4FQYOBUGToWBU2HgVBg4FQZOhYFTYeBUGDgVBk6FgVNh4FQYOBUmL5dzL0eq1MubKvUKpUq9 8lGlXv5UqVcwVeqVkSr1cqdKvWKpUq+EVKmXN1XqFUqVeuWjSr3cqVKvWKrUKyFV6uVOlXrFUqVe CalSL2+q1CuUKvXKR5V6uVOlXrFUqVdCqtTLnSr1iqVKvRJSpV7uVKlXLFXqlZAq9XKnSr1iqVKv hFSplz9V6hVMlXpldDn3cnc594p1OffA5Xwpn+3lzGd74HK+xOXcnbiTAn3OJ7Y7oZnPSs/C7TzM M5djeWJ553skT7woPaN8Cgto+644snLCkZUjjizAES/qwrviyMwJR2aOODIBR7ytVO+KIy8nHHk5 4sgDHK3eHadlsztOW89u13bdHbe6OY5e3emDO3HQYW9Vq0jS0dFkOrQDa/XJHrwcnZ/eNm6PLLPr 2q5kuSY6JKBZXP13BS0+hy46+P+eTDLzOiD/osMndPiP7ufDw+HIfrVNd9I1f5I504/Lc3Q4MLTg 64H/FAqeQus/VtCLu7EXd8V5cS/24p44L27GXtzchxf3k140ldvL/9c40XBdU2Xy2+WKN+wNzf7U slcnpE2xTeIjczScOL0HUrdTy7onf75Uq/eXrXbja+PmpDeckDu3E6pHb6fD28lb36Yp283769v7 Zm84fQ0vT2/Ov92fjxxbCW+c/+hcfft/J4quyCi8dXvbUBZfcfi1c9swtLPL9m14/fX09v61ZtCW u/jHabMZ3r74Z/PybJamRXI/meXVvunctu+vr36gw2vS2qNBb3L46HQH9uF41KODwGoRjdDr4MW1 u475jGZfpFFFOlrcPhwja6fBJQ7UEBCAidwxwQXE4HE86E6eT1yX1H1A/iUPXivr9WV/9By8UENH GpHyLm4ap80tVFWYhVlxBpSFpNW26A+t+weZZBy9xsz1XfS6zaRvM+nbQfpXWmd3gtZD11F3Qip6 T55pSc+0I9ikFe+f7a7l9wR02Ec1GR220OErwgY6fEO0dZAc17ARXdtmU+5keLOdabWx6D/hvrm9 4eNoZi86tnvUxOdd89m+7Xn2TuZ6NP+ovc6UkaXZHi+yP6RD0waPHDFV4zeHvn3VIvIJQXFD8UNm f02M/GKQS8vtuugEGdr3M+nd0pljisltzQ3a1YKeiV/xPe5KZqS3kg6eTe7te2s6GLwp0dzbUrAI s7H2sK7+fn5+8gsZhH6h3780T7/envySvbL8haMQGnftsLxfMlUCsxH1cqPnchkhYFDPdVCnQzCL eBZns4H58OKs0T6VaR7NH99OcbxrAmZExQxeiRkMmAHMJEbO5Zi5S8FMhrZUfLxf0/ovbryH0R16 XYa9rr3t6N4GTS0wZrYb3QEzImNmu9G9ne3oDusEH5UTnhGsC26Yy4XmzfOukGm/90D/IxX1GV6p u+YDYep1FvDItTMdSu4z2kyFzhrw8BS1SNuier2KDocIV6uVSuWs8fWy1bpsfUVfG1d/Rz9uLq8u 25d/b6B247ZNbh/7XB+9QO3Tk9ZndHVCnv+MWifk+c+ob3VPsCxrn9GDPemeyJLsf9Bt+/Sm3bjY 6tkfpwSeGz2KdygWb1+sLNV2eNtarNhKhWn/4D6iPZDmeVvZxSmCN9fJasFtv5dZ2nkts4hBvbfv lVds3GOWHM+K2E9S760E9b53U3Gg3sF2gsk5YAYm54AZmJxnTr3v3VI7jO7Q62BhHTADC+uAGVhY B+odqHeg3jem3ts7Uu9tLqFcW4d5X/fRJAO+4km8faF460KXse7rPgqk+8fc2yCKv7uqvGpx0h38 3YF0B6sJpuWAGZiWA2ZgWg7+7jC6Q6+DJXXADCypA2ZgSR1IdyDdgXQHf3fwd9/lnJTgON3BCz3Z rIUGL85oOjnh0LLk9xOdvs3r9ERV0Nv0REN9PJj2T2oY9R773acTbKCnwQt9/hfSqekZMJb7RjOh z93L0QlOqGSum/jHTeNLo33+Dc1vKbNbv6DRmOalIKLJmgjXlV9Ii5M/pMFVBY7QguEjPEKLypBK dKdDtOaZ0FWHY4L34XA0QVTZr3PK6NGraY4e/i2Rmh7Qoz6ROzWf0WOPiGPkLIC5AO3vwQP/Qg3H ISkw+rX3NCTisX4LV7r8/hN0Qvn+r+zylzTKfTicX//e+tz+F1EB2a2J4XBNjCiS2UoXUR5LF8Vi SWFVLLtVseWLYtIo29PVxltUe1ZhUlNp60kmaPYPqNmj6m6m2RdoEMgfe9lS8Hq2FDDK0K+36dfr gAsWlQFYhQALVp4BWLkAC5anYXkauiYsTwNmYHkaMAPL07A8DSxUbHk6tsiwfHW6AGr+Y69Uo8Nz 5C8lhksBdNGa16SHA7rmRxeyKfnvLxHw1hVRZIEwzCWZ5pg8rRqSqlarso6uvjSvf9wuT1yVVGxg xVgrsSEpsqYbaiRxakpDmaXKeKfVkmXceTSJ6DIupqu4pGlXL+MGmfivUcO/5LCtS4TFLkwXsOZI Xr4DLJoUQp580MUuWKqCpSpYqoJdi9BlYdciYAZ2LQJmYNciLAvB6A69DpaFADOwLASYgWUhWBaC ZSFYFtp4WchPNgz/xlaGYksVKLpqE//lGGFVMyQd6zK5Sq7xJFMbmkTgJ+v6eqllSauSrBV2SSiZ VK9KSi3nNaE2uyYUCXMSWxMy6KJQYk1I5i4JtQtZEmp/1CWh9uolIf+kP2XtMwEVCMSTcSCe5XF4 YEEIzANYEAKCCCZusPwDmIHlH8AMLP/A8g+4e0A/hAUhwAwsCAFmYEEIFoSA8YEFoWIWhNrrLgjJ iLtOkVgPasfXgzSpasiGoaSs8MRSK1iTsFHTq+paqeW6pNVVrVbnrQfFk+qGhLWc1oOyWqOZ5/XH Yn3jAPVcNB2jCZl/dCe2lGXdl62jPIarI3TDjL93ph1CwF8wCdZObhfJtOA4ub1fRQmW73pD0jv7 fUR0QNqeLoKftKW9wcttyoGemaIh8x1AK7JyHZM5Hy7z7ULzJtyuKlkuVM0PAdyuKrfzqlwRkdOF yc0ykiq9R/Q7OnTRL4tDLn9B/0J/Q5Nne0hU8n8QOAclLBJsWAbJjORj95miSDHzUiIaP1rYNgXN 8nnshS0zX7OVntdd5pXYR+ky8efbf22Sw2RETwndwAgLCqPVzHxgUDPJavUBjwzY/THw6u83jVt0 2bptEzvuvnn99SgcPJStKkUMj4NK5a/ZfvY/vwDTXy5bl7ff6LnStEVUqS4p+qxp0fUXdNs4b19e t5ByiA+riHzB8pFiHCkylskjCONjVSXZ/FW49iOfT3v2oRZ0tFdYrfPWWcZl0CPKDU3z/1aN+N/g o37CWDOwgasqJveJ5avgT0guogGmxPhxEPr08rI83arfP+hHqVU+wUfcT6L//7Cdx9vpYCBNXicF 9X9Nl+f9X1cV0v/VqqFD/y/ic94fmT/vHTKZDvixq2ev8r+RSe8Gg71PEpC55HWnfdlqHP24uT5r Nq4qJ8EHoZOT5NfgQ2xgos8lw89GU6u6VEfo/HnkElv7J7HMBtwEX+2hTatjLdIomqT4aVSsyhJG wS+tNv1N1aVq8JuG65Ie/tZuUQNcV2mGCAUOsGrw24v/k1aXasFPPhcS/kRzRIpCr+lH12uS7P/k VITp/zf4u+ROBwWO/1pV1ef9X9MwHf8NRYH+X8TnL5W/BD286zjdN/TcddHjqN8f/UnnQwO7O6R/ H95Qb2jZr8eVv5BuQY8Zup5ODkePhyallT5TgumBagzb8hPgZAIllkBJJBiOggSkND+FSlL8cEYP fXuAxo7dH3UtohEmI5JTMrGWmhgzien/GqRE9NN2hnYf0RUH+3Xiv/Zg2p/0xn07/M0NXva035/d QH/2Js/I6Q5/3qD/JK3w3H2xKcE6cXrmpDcaup9Rd2iRhrTRaNh/8x+fuqQmf1J2wbSdSbc3pEVa PT+5n3JgTySE2s82IvPVyaxevaH/9ITcnlXRtZ2e7fo19atwIgfFmd0heiDJuv0+Kepx5JC7b+il 2+9R7m88ndB3RocZfUhWX+bwwOhr4wbPm8d/8fGkN+h5YU1GERmTn4heH1pulrW5vDjB/tB0chCQ Wc7P++7r+E0yD9Bpp/3t5OBGQud9unb0j+du3347QN9bp1eNILnlp7/HBwFfEzZrwJU2+hOXfL+9 bZB/SWM+DU9nX/45+3JHvgx6sz/09j87pD5+ZncdutGheXF61WmSH04vLs4DlxUZfSGpZDJfPv/n 5e11i+Z/V6uehN75fmc80SW9Jmv1uv1/ZeWz7m8KUWT/ggxW1XpVw/6FIimyrBg1cqH6F4ah01/U 95N4KOR4P6Q4JirAx0HG4ldZ8eNX7V7eSP5qqvyVmlFVNsWAnxcRseZjYGsIaFJNw2pN9wWtSQaW ZxDQJSJzYi2FEDB0TfcvKASqsoZr2r5CAH88CNTo+YCAgO0QMIqN8Q/TScQKoAkJIBybDJHBCOja fl7+kEkGSlLaIZUHmo7pWt8HGjZ2wQzeeeSoKkpdMyg0KtvZ/+1OqyGR7wXZ/1XZUDR2/i8D/1fM R9Q1DHTW+Hra2nEFQ9A1jPm62Rbr+v7CmWltsoTnoL8h31PBX8gL+QlqXZxYi3XFHV0MHAjQWZRj 3HWKb9wqH7gRcvBP1+465vNqvzPW58E5WjxsVi6HQdzMsH2J1eKMBpsvkjt43HWI9fJ8rH1OOo2t m2dqNY+N48rGdXqyh7NKIfqa06HPTaCDr/akiec2ysFuWVflY/Rn16Hk0DHyo4rSrufYk6kzpNzD lJiEj+jg0TW7w8eDz8iyzT6x7ayAQulOKJqnE9vP4346pCbfPend0/4kO9nQON/EKqXV1XMRT3W7 NozUK01Ct8SUPtg5d1zPTUg7tVv8rRuvz11ig/Ve7Dap9npvnZ53XdUiL03mGIPuBB38b+sA2a9j Mi64aPI2ttEB0TkHwYyk6zxNB6TdkOJza8HP1FHufnLwbnVRM6+LhvenXbavS/btUt+6Ltoe1UXP ui7Uy3lvGmaXyuTRMvV9apn6HrWMIu+NktmlMmr2lVGVPWqZ7SuTfctgtV6KyvC6Nuyz2uvp5Otc 9tGZJd0HsviB7gPJbB+Dszbz4b65veHjKEp8RKYGu+95D/M/WOT0B1MGf6fNDvsP50X2h9T7foNH jpiq8Ztjdz91Ip61N9icd13bJRXx0x1vs6CwF0vOtOIKW3Et2AtSRM21ec21zSu+8bpfLhXHm1dc Yytee4+K1zavuBGpuBVipXbvuva2dVd3rTsB/FqVr5w1vl620NnpLRkAvjduWo0majdu27fHNMfv gdcN03V/xb+hcdelHf2h6/ZMRLkVXvKww/yqrJk+xOmv6prpQ3j8qvHSX5D35L5XpXI7dV56vkKk 53O6oKZATZVfTYU9vYkPz5rX598bF6jdaV22vtLs8DEDe/Tr1YmiyvJn1Ar/9q2u/+039GQTe993 AY4ex6scswBcmYdW0yRci+ShHrNYWC8PLZKHdsyKZY08FEmb18PXG4lW+ozOv13fNlBSP/waNMVv RK+cn55/a4QPoC/XN6RdE+k/oyu/n7fIv5UNXDMCn/OQo2/5/5LXQGhsO2Yz9MtGg8f+aFyJu52n /Q090Wk70EfZvwjpwV9dU6WqtiwlDv/qVUVS6kvzrGpzB3K1ujSloYQpjVV56oGfvK7WJW15nnqY p4aluro8ZVhPFUv6itJrYT3pEdXL3yhsJV2X6trylIs8ldrylLP2xASCy1Ma89Lx8jyr8jzPmrI8 pTKvp7685WvGXJr68tJrs9LVVTKqGvPSl+MTVRftuaKVasq8leTl0qwtkLwiz3qYp6GsQkhtVk9F 0pe/UV1eIGRF6YtWwstLr0feaLnc62E9NUNSa3RguW0junPe1wRUIVHiZ/Tou5mH3leBapgPQko2 g1DNkEiv2nEQIl1H23EQMkge8o6DUB0TGMcHIWWDQchvitINQng2CJHWWQ7K2nwQIkBfqji0WZ66 LNWWqq3ZX9q4y4cWJZLSkJem1BfDAF76RmqY0pBlgtClbzQvXZVqS9WrMU9Zk2pLFYc+T6mveHd9 9ka16or2rEby1JcqIyOSp7H03av6ouWVmDKi7cdVRmGDBkmJpRf63oaOu1nOBvfG4TV4VSXHV91p SwDOcluIIdGN8/Wq/+LBST4ODs/yyZSyHh/TTUbD0QTRU7zW8fA7WhzScnCMWiPk0kNIfLeekbOo yaKWvwcP/As1HIekwOhX31/Ftn6rDH6S9PTVRtOJFcboCK+O/ro4N2b9I1qco+CQGgenHf1ioQ1O fJl7J85Pv3Ew+tnr9+mRSn+L5xzUfpuqsmfK+IsUz0+k6sFJVV/izpGH1rzFgiNkZi02z2+TSvjn 0MzbfKNTgpz5ITaL5zc4vSbnY4Gcgk4FcpKHAm15Lo+DpWe2E5CsnUG0k2S3HpLRsT3Opqf2OEcx OMeP7dnJe1gB72FRvYeVXbyHlY/hPayA93AmslHy8R5WwHt4K+9hJUfvYWWPvIeVPfIeVvbIe1jZ I+9hZY+8h5U98h5W9sl7WNkn72Fln7yHlX3yHlb2yXtY2SfvYWWfvIeVffIeVvbJe1gB72FhvYeV ufewssx7eF2f0i2XButSPbokt0UW/prlLIvynqhUq9dlLWDLhTk3SJaCMzbIhRy/iJ4aI8whOunN wbRN5BAdUc6KyXEFCZZweEs4u2ziiBZwUOmSxt0srg6KYnMR1cfvxWQQs83nUbaBepjUQSsurQRF yE/pybF4ic144p2iY8XbEhbWYGGtiIU1UIlpKlEDlZidStRAJYJKzFMlUnAoW813V++hVH4G4boK 35602+4kK6i68qrQrXHqtrVXErUnpn5YffItWn9lPiVQaP2V9PrjtBkBQmG6gzUjV6o0cuUBeez6 6sfJwZNpHgjcDgtmikZyPZjlFGsaNcemmbXIFm3D2UGZT+OcNi+/thoX/MbRoHFo46zE0Sb7WBfq c52drIlWX2c7a/IhdZuHctjYCmPHhxs7QEGCgsxKQS7f/rtQDitXR+pqNbJpdr73KiGC1RlpVUlR Nth9y5Hyr0Em3O1PynHKU/4mKIXugVI+xEbcKt1btNZG3Lqhr9pAGG7ErevKqm2B4UbcuqZIdX2d jbh1RV+5aVaZp1y1ZTesp1IjaFtnI25d0ySjts5G3LpaX7llN8yzqkr1tTbi1lVFqlbX2YhbV9SV m6XDemp1ScHrbMSta1iS1XU24voIWWsjLi29Vl9nIy7F0gq5hxtx67qxSkbhRty6unrLbtjyvgZY ZyNuvYolVV5nI27d0KSaus5G3Hp15Sbk+qzl5VXtGW7ErdO4b3idjbh+e2rRvW8kC+7WtzDP9fbh bjIa6HQrcBajQU2LrNuvsQ2WOxr4mZRzNJjtiK0bdam+zo5YMijoEsZr7IjFMl51MMLsb72GV+Sp zFNSWayzI7ZeUyWtvsaOWPJGxorxTYu8e726xo5YkrImybU1dsT6KVVljR2xJKW6Qs9V5cUbafoa O2KJjDRJxWvsiCV5arQ9I1rBSNsRO8t12ZbYTGdjVfrixU/IQiUoVfU6xjQ6mLLlFC1pXy/ZYZtp y+mqQszu92s6rapgQ/W9RoixgTUVU98QNbP5bjgy3ZDmvLk8bxOlT9ru8PrLYVSVL85p2HyEIVhX 4q5h2vY5aZJizHJqXafWOuRpyOzsQuK8IXdStu371avEpNczeL16VZY0zH+7SI1XvFtveEjGwfC1 wrT+8N06zoTbwEu1SkEz+LleUbBmaEmXKtpTlLqs+qHIYr/oUhWT/+3ah1YyJb4llSKWNLEp6WLT shQbR6UVLTdZk2uYLzdNMRRfbrqk6opWl8MLg8zgsJqp3FaSODwpKnMphp5qSo6eakrMLUPJwC1D ycstQ8ncLUPhHTagcBwAlIgDgBJzAFCiDgDKNg4Ayo4OAAo4AOzuAKBEHACUhAOAUrLDBhTeYQPz WGebV8oPdSZqfL0vl63L229Ed28fYU8XNb5ekfEfk/E/g26QZRnL43/KZLrOxv8ktogM8T+L+Pyl 8hffOkRdx+m++dvQHueRfgd2l25mQw9vxASz7NdjPyq9jI7RdSyyLzGxF4F9EcLJBEosgZJIMBwF CUhpfgqVpPjhjB769iAeR7qpJBNrqYkxk5j+r0FKDGkLRPey2a8T/7UH0/6kN+7bs8DGwcueEsNq FunYP9nAt+vRf5JWeO6++KfKTJyefxqB+xl1h1YQD3nYf/Mf94MZ/0mHdtN2Jt3ekBZp9fzkfsqB PQmiHyMyGE1m9eoN/adp0OxZFV1i2tquX9OQLQmKM7tD9ECSEQswDL/cHb7RAxp61K1uPJ3Qd84x ELSyLBB0RMbzqND5BnJew++D2Z6j5Llhbeu9OXVJrWo1JWCg6pJuqFVDntNRddk/BI7G+zZqZGJd m10YhuYnyy34t7JS5unh35Vcwr9vN0fngECDmfvOM/f8O+RCFlvvmINOGemUOJdOqWbVKVVgQbfy F8u5F84bHzrh+tbQrBOOYqYvPQhiYRzThKRPOjaxHAPDkIie5uVbksR+JKUdUnmg6ZgG5sq43ypZ 9VtFoDXRjzz/S8z/70YvtiM9Fzj/x7K6mP+rqkrn/1UN5v/FzP97j2Rm/4gu7q7/3ri5/1b5C7nq De3Fjfkd2r3V1hnCNXl+r9W6v2p9v79CWFENo24kfmlRz5Nk+hZSefe/I6zqydtfG0ipxpK3w2Jr tXr8gXZYKubdThbb5hfbnhWL1Zqize+3F29bM8joL7O/MG/bTnnb9uxteXfpyypGvYYXP8zeVtH8 DsP+QLLXdCNxO1ks/23bi7eVyeuqROL20Oo9Vj7Bp/SfhP5/sN0J6XZK7dX/fwH6X8ZVY6H/MblP epaigP4v4kOsKGJCIdQ8bSP046bxBRENj646CLXIf9/Jf1++tM+/IXQZ/GkFf0J7KjCFfP9PulqD ZISUms/I+neVwG8XhQ6iavCHGF6qSuP06KBh9rT/t9+9/6tg/5W5/2NVpuGLoP/tZf9vt6D/Q//f rP9XN+r/Gq5LugL9f0/7/x6M/wb0/xL3f5luyK1D/9+7/u821Xan1ZDI92L4X02VlaoS9n9Fxpj6 f6lEJ0D/L+Ijqn8qOmt8PW2t9k5VDnW+d2qg2oT0T537RG+x4cJ3it4kglvffrH7ajSIW+Cc7k4c dxAuVvnh3NyMwrkFBQbu6MiduINHM6MtJrOc19qZEXfFj77tGs033w4Qbz2333vI5tzh2av8Sjew BEvekTLQb0EpRialHAX5hxI5/DcxIGgh/kGYfjHVbIvxAy8EZ2+67ps5Gr8177v0NE6I75d/QIb1 VuKVwA1mVZSG4OvFLXVWo0U1f3w7xevsg+FD48iHRIAIyeTj5A5wUg6c3OWIkw7oE9AnMZx00nAC +gT0yVKcDAbNJmDkw2OkOfrTdu53xghBAw8hHUDIh0dIZzzODyE3oENAh8wRcsNDCOgQ0CFLEOIo P5stgAgokQAidBNIiwuSNoAEQLIASZsHkg5oEhhsVmmSDmgSAMkqTXI/nk6a9w8yQOXDQ+Ws0T6V M0EKxcQSsAA3Xw6w4CLAAgR9OcBylztYOjAMwTDEgKWzBCwwDMEwtDZYYBiCYWgVWICgBe5tARMu qwL0LEAkChEOpwLkLPBuq7QIULMAkSVaZOL4vrEKeS/whC2RJ+zhRWs07Ax7k4te9+k+DJkRfHZD 0Rwwy4AEU6DSuMq+I5DAN79cGundUATqqETq6D1Q1AHrCKyjTYDUWQYkUEdgHe0OJLCOwDrKBEWg jsA62gFFHeCOwDraBEidFDO7A9wRWEdZAQmsI7COMkERqCOwjnZCEXBHYB1tBKTOMiCBOgLraHcg gXUE1lEmKAJ1BNbRlijqDV9sZwJ+0OVxco2ihhncdgVRAJY0DMF5OCXFUEEAAjfq8vjIvpcS6oAS KimGcgcQPTe0BSqotHbQxOkO3fHIte+z1kjJEyRDNIEyEgNN+UOpDYqppFBqF66W2qCWRMBS7kDq gLVU3glb4dZSB6wlQdCUP5TAWiorlAq3ljpgLQmBpbyBdAPcEnBLW6Hpho8mUEvALWUDJbCWgFvK DEugloBbygBIwC0Bt5QlmkAtAbeUDZTAWgJuKTMsgVoCbmlXIEE8HwikEMUIJ+LGDCSw2wgCKKwE CWwmgsAJy0ECcXtguImCpJMCEhhuYLhZCRIYbmC4WQISd/AdtmUIsj2VpLEfJ7vPkilmUoAEhEvp 96jmjiJYmxJjo2oBQAJ1VPrdqnmjCBwLhbGObnpPz9kg6SYFSaCPBDCP8oYR2Eei2Ef5IwkUkgAG UrYwmoym5jNaDx5H3QnJLYDb3ejFdqTnBAzBwIKdG9vBkafUwMiCnRuZQQl8EWHnRmZYArUEOzcy ABLM/mDnRpZoArUEOzeygRJYS7BzIzMsgVqCnRu7Awlcm4Bb2g5NTT6aQC0Bt5QNlMBaAm4pMyyB WgJuKQMgAbcE3FKWaAK1BNxSNlACawm4pcywBGoJuKXNgdR1kLMqq37vgf7nez1J3QB07huNDdkM YtHG79zF73QSaTqxNDS4ROyqE726if12E/nNUX76ASgjl+3YZSf+ayf+6/3iBJTETcy7eZe42eE9 3uE93ok/Hq94vN7xakdqHUTjbCrNWUzy5N275N0ON22Hk7bDzbfDzbfDzbfDybfDrW+HW98Ot74d bn073Pp2uPXtcOvbSdQ3EtU0GaMyGXQwGUKOjQnGRnViQ/Ow4VXYKBlsnAP2sHr2wHH23Gj25F/2 +Fb2CE72JEX2LDz2QDP2UKrE2UKJc2QSZ4YkzodInAWQ2Ped2OOb2KuZ2HaX2D6V2AmT2NCQcE1P eBizfDzLqLK0GEttsDNUdo7BGorsYM/6mrLegqzLF+u2w3pfsOvn7CIos5BVsc3n0WZDR2Wz1IFP rkvuSE+OVRl0f9q/V/91jJp294UYC8jqObY5GTlv6I8VGfMHwgM/S1KA6tiPi+wbdEjPJn/TQus+ SapgO/bQtMM80N9QWLveQ1C32o51Y0s4iFmR5NcnezAAI/KDGpGbmIMsEo5iCDB5uAAODZBBMMDH RhuwITw22nxsAMUF2Gin6I026A3ABkdvUKYHkCEyMnwEcHHhc4WADKGR0WymYAOWQAAbHT42bkBv CI+NmxS9cQN6A7DB1RuO8hOgITY0KAT4yGgCtQHYaLbS0AHkBqCj2U5BRwd0B6Cjk6Y7OqA7AB0d vu4Ag1R0aPDtUTBHARlp1igYo4CNNFsUTFHARpolCoYoYINrh1Kfb0CGyMjwEcDFBZxOA9gIUZCG D1hvA3xQFKTgA3xGAR8+CtLwAfoD8EFRwMcHnPcC+AhQkIYP0B+Aj06q/QHHsgA+AhSk4QP0B+Cj k2Z/QOQnwEeIgjR8gP4AfKTyHxCgCfARoiANH6A/AB+p/AfEUgJ8hChIwwfoD8BHKv8BIY8AHyEK 0vAB+gPwkcJ/uOAXJDg23AEfF+AXBNhwU/2CIA4V4GOOghR8gF0K+EjzC4KAUYCPOQr4+ABeDPCR 6hcEkZ0AH3MUpOAD7A/AR5pfEIRgAnzMUcDFB/gFAT7cVL8gF/yCAB8zFKTgA+wPwMdNKv8BfkGA DzfVL8gFvyDAxwwFafgA/QH4SOU/wC8I8OGm+gW54BcE+JihYPuw0osolIngg4mIc4kwY4nYUkxA oUQcmUTwkETEiESYAPZs+OSJ4MlToJMn/yZPe2VO+Ewc7Jg4zS9xhFvi3C7msCbOGT2cY1k4J3Fw Dl/g7LfnbLHm7KrlbKTk7J3jbJfi7JDhbIrg+MFzXJ853q4cB0fGp43jysTxXuE4LHDWqDnLkpyV KM7iA4dv5lCMHFaJQyRw5o6c6QLHQuQYBclx4H2iA9d2iw6cjJCbU8zh341Ms/wVmRZy6C9/Q4uo wei3RVkZxDT284/HDKaK9uYMBv0POuiTt79pnDY3GvyjYPAH/hAD2w/5cxDRKyc2NjuxgdmJjcpO bEh24uOvEx98nfjI68SHXSc2yDqxEdaJDa9ObGx1wmEzfhn/tRP/Nf7sTfzZm/izN/Fnb6LPuvFy 3Xi5brxcN16uGy/XjZfrxst1g3LfR5tnoyRDvUU0JAHy+GdadPVdteQsZ7+kp/FgkCyimlERRzT7 MHj94b+RwhkEMyzGV/ig3D+cco8M0+PgtNZQ3f93BcU+m6j/GDYC/T9OHAQ6Nkd95aH/876LwTb4 4LYBKar549sp3hElM0ikIuUOkFIOpNxljBQAxAceeYhQ/+vye25jT5g9IKUESBlPJ7TX5waVMH/A Sgmw4nSHP3PUKkH2jKXijP6kw1IbjFowaudACTGRjhUwa8GsTWBla6Z0vFjIi8+xOZOpuP3FjLFx NcpTcBwgf8glHQ5ntut6DpOlaSHSUD9zpPto9gXQfUExYB6VxJQOB58sTSSKkKg1TYsAg7oMBvXP +5/Zc8QLuIT5A1Y+PlaoUHMDip85oKQcKLl3vucKFJr/9sZ0bJRktGAU6PHX+dg28MKQzMwGjmRp ZJNlzNKlv6hxj4DdXZniTkyhcyjM0gVzYfKF/5ZYvJ459gIehMQDFX4CEIAHcfHAwgFCH4sLB07Q YzjxVmQ4+GfdbjsDie5Kim0niu4Bim7eCbBW2BSEY3vv7I3r0+VrPNLvjqnEQk6937WHLzvPAII8 I5OV1mjyTLOZjNCDjazR0EaPIwf94Rd3IO362rHy9Gxaz2+OV3fiDh5NBY0d+8RFk7fxSYBjt2fZ J037cXKPpuP+6KQzHtvOfWLqjdDE6Q5PWqM2+eOOR659T+rUfSJ3hp1hb3JBvt8vqr1Fc0+mQzte cWdA+tys5rHX2P4ttnslI6tXCoeB4C1gDCj/GMBC4CiQvMmDJkGCj+DDixC+hxcxcNLLCDJDLD30 nu7BoigBmnZGBLk8u/xKKnt7xctsC4sl2Cw4N1yaC0PWHXT7fcAd4C7A3e0VeZ1U5GUJPcDZh8PZ aG62ze2edQwn983tDR9HR5PewCaPrLDl5p/IeBhTUpvNttYtbbNcD/vjybNjdy3ybbDLXCFpWa54 4KE3pNfOdCi5z2jD3OfyI+b4er3tyH/jYPp5N3qxHem5Uqn435CPKtSdoNbZSU37jFoU4ZVdpjzJ 1shyzlO6GQ/Md0CD7+l8B2Y7YHWuhQeY6wDq3gV1MNMBlMFMB2Y6H3Om0xz9WYLVndS3gNUd0OL7 P9sJ4QurO2B5boAImPEA7t4JdzDnAZzBnAfmPGWc85RuxgPzHdDgezrfgdkOWJ1r4QHmOoC6d0Ed zHQAZTDTgZnOvsx01tv5FH3FxJFiekbnH/gBYwJt9rfcAsZEA97wDi6rZlsMuyuzCfHWy3D64ny2 /B8RO6A1mqQO/JkalFE0QXRDMdCUP5QgkGpZodQuXC1B0FUhsJQ7kCA6fBmANN858s7WEsSSFwVN +UMJrKWyQqlwawlC1IuBpayBtP0RT+Zo/NaMhFSY37mL3+kk0nRiaaIR6+Ph6uOx6uOB6pmo9ExI eiYePROM3r+8H08nzfsHmXcT827eJW52eI93eI934o/HKx6vd7zakVpPHL99FYLMSIvG7t4l73a4 aTuctB1uvh1uvh1uvh1Ovh1ufTvc+na49e1w69vh1rfDrW+HW99Oor694YvtTBaymd/psHc6iTSd TvRcs2YkXH14zfzeZn5vx3/vMM93mOc7zPOd+PM3TPk3TPk3TPk3TPk3TPk3TPk3TPk3sfKZnjW7 gdkbd7EbHfaRDvtIJ/aIO/geb+bgRid+o8Om6DApbtg8btg8btg8bpg82GqwtYhL22Wk7Q7YSrJ1 jEvbZaTtDthXYN/ghin/himffUH2/W6Y8gNpv8+B4dVMImfHFnDcvurYj7ms3GywMEaqYDv20LRT l8h2DYfDlnCwyDbtvEBa+Oy4wB2Pak8pPXt5ZhRcfZalv6DoR1r/G8ptQTGM5L7y/MaoPCCG/D7E kJ/n+cesmAPUc9F0TGVndSe2BMG+3iPYV9FygQAUUa1u+A0/L4Hf9HDs7p4cu/u7lpUf0Hh7hyXk 2O5R2ETqPtTnsnXbpm6Wzeuvs3opW9XroTc8WDy/Bdb8DFbD3KeiHrsmaUe/Fc6ptJP6XN2uEtzc 4+r8pR/RJ1qmxYR8p/ngi+nJhkgFH5btRGs1JQcFR0dR8ZsMJiDYkdCYiMc7mt+EiEdigyIW9Ghx F0AhNCiSmIDYR0JjIh7+aH4TNtsJjYlNYyD5D0pdZpYSt08ZyyQ+JMWVURyGwdWrTQnb+aXtOGNn SC8DbuC+51MNmy3OhNWubJZ6tjiTmPRtPkFea2qpFDS1fBi8QKcXd2pJxW8m6QbAhNB0Aw8TDkBC YEg4PEQoAAmRIaEkqQawJoSmGnjWhDsGTIiMiTEXE2BNiAwJh4cIsCaEhoTCW+KEkUPoJc4X3mIW QELktSzuojdAQug17+RSFsxDhV7K4s1DJw8uYEJkTLgcTAA3ITQmeNzEZAx6QmhM8PSEA3pCbHcp HiZAT4jtLvWym7tUMEeJ+1zEVtvjC63xJbb44kqMVo8zqnEuLcaixCfQ8alT3GiOm0vxgTKuIuOd Y3+duJRCnLhw3k5c4T6u7uv44Q2UkYDKKAYAk8WFa08AFeKigoqfMVss1xoBJoQ1WwLxM5joud1B 9xVAISooQvmzE5yuO4X9IOJOcHzxJzDxOgY7U2BMUPGzmKCHCwImhMWEL34WE2BkigwJjo3pwsRD cExwUTF0BuC8Iy4ofPGzmHBATwgMCYejJci9J4CEyJB44mEC2AmhMTHgYgIUhdigSGoK1+z2ARTi OohT8Scw8Wd3DJgQFxNU/Ns57zAeF/GV9vgaK7O6Fl9WiRPqcSo1RqIx/El84hybMsWt5bidxIyQ cdUY7xT767SDc3Ta2fIQ5mU57nB8cuWv2X72P78g+teXy9bl7bfGBfJbRJXqkqLPTqZG11/QbeO8 fXndQsqhcqgj8gXLR4pxpMhYJo8gjI/VOsnmr8K1H/l8gk8GH9L9jqJHobtXV+c317e3139v3Ejk RhZlyORjaJr/t2rE/5KPohn6J4w1AxsK+aH6ScaqUdU/IbmIBpiSUddB6NPLy/J0q37/oB9RNS86 a3w9ba2jd1W+3q35uQipeXcPdrAycMMi2AM9ZvOIBo9w84ncAKbHTl0ATI/Sjf83jVuJTNqyK2PF +I81NTH+Y0WG8b+Iz+XFiYxurjvtkwP3iZ6SfYBOO+1vJwf/eO727bcje9Cb3A8GB6h9enLQJn/O Tg5aB+i/K6TvX3WapxcXJxj9uGl8Idk0T9snOmp9aZ9/O1HQpf9XQ1/8vxh9P7s6vSOpyN/LFv3b OVG0IKMWZVKuOiTx1dlJTUMt/9/v/r9NrJFLGf34cnZ+3bz1v52eNb8HX8Jbt+0vzWtSuhzk17yY XZ6SN2vc0LpdnJ6RSpw12qct7F9f3n6nd753/C+kPjedVvvyqkG+thZfgxddXDcvztvXP8iXu1p1 /jMp7ceJJulYUVRs/19ZrZBmVXA1bNjz09vGrR+W2R0MsPaKX2sapVAWjX0jofM+ZbqCVl/W2nK0 tXHY2nLY2nLY2jKvtWuL1sa0tfGHbm4yXlU1oj1ocyN0/qV5+vX25ODwFXVJyw4e+rZz+Gdv8nxo jscH/mMk0fXVj5MDnxldh3g7EGGKydH/tw3KpGY1+Vul/zUVG1WZ1f+KgUH/w/xvH+Z/mG/86jD/ yzHYXXz+N4tydzMdXl1d2I9o7Ngn7k7x7uIFHFSko9fBwLW7jvmMDsfI3WkXKyfzp3juwYCEbuyu hXpDMuqi8675bCO359lkBEMnyNC+n0nvlu6mcXpBOgSxFa7OGje0S3xtXF2hm8bXy1sywt+ShNj4 jK5Oz89P8GfU705O9OPKj5vrs8vWV/Tl+gZdodPWBWqhTuvmutn0756Ssfzm9pjWYfhwgqrKZzSY niCkfkZD+lf7jH7Sv5jJmCR7PFH1ek3StdSH1WUPa7pRk3DyYSV82FhacrWqSVo18bAePqwsLVmr VSVspJasLy25XtOlai3xsLZWyXq1KtX01JKXt3a1rksqTm3tFe+sV6U6Ti1ZXdFgulRPf2e8tNo1 XZGqcuJhY52HNSxXJTX5cG2thxVSspyoNpbXK9kwpJqaeFhZs2RZMozKbaNJRq3GhT+R+hzMqX4N QfBbpfLlsnVBe2Hnxw/Soc+uO6R3koGtdXZcaZ3F7p5/u75tkKHxFh2jmszLo3n9DzaPLCBq1DZ8 2JBUg9Y+Wp9o7avKos5njds2qamvlegMqkHmSWScb7fpj0TT3Zy2vjbQ7+QFavJnovejbzR+pEVv UDWDgEFfdHpjmywUPZIFqdMWWWhYUnAlfPMT+ib+99nrk+nYr2FVf6u0b/5JW+J7x/XVd5j+xxea ZwRRs2L8ub7fSGRmyfZBVcL18CfViFdJrUmGHP5GC4j+VtXDTkCpgdhPqk50/+wnjVEXsqTNfooX RnoGff/gp1q8LLkuqcrqnzBTRayGypz+Fi9Nr1clrTarvsxIwpDqszwVjcGKKinz35i64KpkzH5T FQb+miSrgXSjjRmRaQz7AeJ/nLbJ+N3yZUyAFfbXAE8Bmqp+eb1He2I+n/g6O/yKZ+9Sk6pqShpl gTy1Gk2jcvLRwvGck0ZZtJtSS0mjztLUw/FilkbjlEXaX09JMytLVsNRfpZG5+RDxuNYPgYnTVVS fPg/2O4EmV3XRo496Q1si1pYQRJVqtUD0V0GXZG8TCv8RgXoJ5h3yub16cXhafuwff0D/dofda3D 7uTwYTSZjAZonpYUOPtpMhqHdalpklqr3LYvz7/TjP5x2f42z+3sut2+vpqVcfuP0x/UzqMdHP1K DGzXHk5ILl3LSqiVeYGDrmmexKGuGFXaRN8bjR80Wz+7UM0S83M+SH1ttBo3p/TbdwLHRvO4kh0T iWNMpBpyY+oyJrKqcJhIctNXg5QaI/9+ICaSiKhW030mMpMI8ME0JuMp0TQ2JUKHQzLWokPbjxTv Xl19v71t0HWIoFg9w2nez6DYe98bj87GDB2NHtEGHouBF9xgMCE9nPZs6fmYTG8umwTO59dXV9TG OA6mrIPBdOJOfCUwGDij6eTEfbKHg0HA/wYTWXR1otVQi/7znf4zeKD/Dv1/f/r/9q1u8Ce8Mk+w LKP/REeW/XI0nPb7SPnP/4P92R3J3Z046E9/MncQBIen91z7O9X0mCp6TPU8phCn5dJ/P1OMkwuX pCOpH+O1TBYvozBhv/vknvzyS7IuBxUyUe9O+0QDjgbjXt920P9BfnJkjdBwNEHudDweORNE5Ey0 wf/6X5WFpHfHauOODnOnTRSwmaF0jllw3XbOzxu3t186zXBB1B8Yo+CsVAI93bklJiZl8pGCq58D zvwY1RU6CstBx/OTfiW25Nnp7eU5UX1topdC/YbaJJNg7vvddoZ2H6WvCvxKSvgNjSmR7aKHrtsz EQWaW7m4bjXCzKO5SrM6EvU9r+atb+kGzdCYNUPj9hiWJgpdmvhMjElVU+o4z3WK/dXxT4yOf0BE kIejoBs+TeeLzYUQTjVtI8LJkGVJrqY+vJzFUKpY0rTEw2sSTrV6OIOKPbwm4WTokoZTS15KOBHU GlI1+fCaE3IyM1CM1JKXt3ZNlaW6mtraq0g2RVLU1JKXi0pWuA22JuGkkt6elPO6hJMmaUri4fUI J5nMaXACnmsSTnK9JuFEyesSTkpN0upphJMPAh6fQEv58YVOFzbkE8iESp5NrLX4JFiry+GEzB9F mAkylmTM5ROUGunaOp9PIGgwNC6foBpaiO8EaaBjMilWVv/E8Am6rpOxIoVPUHVK63H5BJ3MnZU6 n0/QSZPgGp9P0Gvq4r1ZPkGvhtRpkrbRyRxbm7dXvC4anX9rcx5iXhuKBb8q6/AQIQKjPISiLech 9JohqfpyHkKvkXeqL+ch9NpMPuk8BM1H05fzEHpdlRR9OQ+hE9xVa8t5CL2uSVhdzkPo1dkAkc5D 6LoSojWVh9CxrzajPIQ25yH8zuwnWI+HmKXl8RA60c81I2MeIlJglIcwQt2qa5QXfE8eAjyiIh5R GdqoqlxX15ywM37Eg4E5HZrjwCSVTJpTNZucyEv7c3mCwdFofH3yX5ffP6NJ96SFJg8nrc90Tk91 8TD48zP4Q+fVFLXW7NI8CdhOVhX6CyP9aBdAJ+TznwTp/dGYDJHYkPSaTuYc+1ORits6Q3T9BP16 SAZi+TeiWUaIbvlDwS9KDf0aPFD/rbJzveWg2nJY68BWxmvWWqcOIHIWzSdQNdr5VkMnkyC5qhde DYWpRp2YaJryAVqjnQc2NAYbKm0NTdRqsD1F1aS6oqysRs4QNWQyQarWC69Gla0GXb+tFl4Nna2G IhmKSquhyoaclaFgaNnkRGzP9umN73Vw3myctjo/Qm62sg6U5c+z2dLP4A9ttODv7HrRbPHZ1PJm 06iCqVWra3YpWtTQr07m9aiT0Uf261EJCG+2mSqXrb9f+9OIr3Nn5fPP1OQmFvtB+BD/pwoxW1v0 SWZ5oHE7T/Q5WPM7rlTI1PScpg1s75QMv4wcNHnuuciZDkkjPduR2dbjaDq0/JWYIIupazvBLwqu Vv50epOJPUQPb4jh2D+jx17fHnYH9lKCXqq0acFP3RcbddHYdh5HzqA7NG1iW4XrEvQTGOM0sY2G tmm7bpfY3LQEFz3TZx9sUgvTsbsT25IQunxEb6Mp6jo2+XlMrDVKT1fom3UfRiS5LyeXvJxD0znI fXMn9oC8+9vYni049YbupNvvkyructgAa/dv4TYqHb2G8zUUMtA36FCphPA+71AED3y18/hI/+09 nhxiOpX2//RHFKQE+ZjyNgT8GM/GowhdEfCadNbtd0+/UwQaLKHPPiPTPaVJTfcs+OOTX93++LlL vzzYtLBKpp6t87z+WCxqHSACnOkYTUbIImKXspSSJteV7VRlKCmiJDUZG7vmodar2u551PXd38Uw ds+jWt89j/ru9aDLzrvnoe6aB1GWNlJrsp5VRtWsMqpn0Dw7V2Zb//pQbefgXh8bFrJeZPywGl7P UMPr62h4HTQ8aHjQ8B9fww9e0OEjIlMC6Rmtd34fPSrDfyg8wa11vtnDq8cUeuLcoq6hg9LfkDOg pfozmL9Ko2zy2TUX94poQzoFQq+mOXr4N81vniRoprDC5jblzNs6vFwUt11mvUf0Ozp00S9BbSX7 1f4F/YtUlczKhui/K/9B5tnmmBa4SLBZURLJjORj95miSDHzUtDiEy1sm4Jm+Tz2QstgMOj3HjId FDdEyMwEcuf1MLarRyzX0PDpTl97fZIxcs2+3R1Ox/53SgfQREFx1UyKW++t/frM3nj+wrXta+Bn eLDIpjWaPNNciPXxYCNrNLR9vuAPUtaBtEi2uSkSKSzWZf0wpNF+7WbxWmxXDAORrnVqJxy0mvNB q0GGPjl4Fbtqxa6+h1d0nftE9o/A5fleBvq+dTYY3D/Is/NzQdAfVdB4LUFjEPRHF/TdWoK+mwka RLgvIsz+nGwChEDk5+edwaDVApmLJ/P77yB1kaTuy7yLIzYbyF0cud+B3EXt7xjkLmR/B7mL2d9h Bidmfwe5izeLa4PMxZM5zNyF0/BtmLmLKneYuQvb32EGJ2Z/B7mL2d9hBidmfwe5CzeLa8Oau4Ay h5m7aBq+DWvuwsodZu7C9neYwYnZ30HuYvZ3mMGJ2d9B7uLN4mDNXUCZw8xdOA0Pa+7Cyh1m7sL2 d5jBidnfQe5i9neYwYnZ30HuAsl9Or6C40rEOJcmELV8r93LIG+B5I3jAgdRiqPV4Wyi0h9CFdPq IG9h5I3jAgdRiqPV76CXl/3EuZhWB3kLI28cFziIUgStDgfGCjMjbwW9HCbkgnZymKAJMEGLdHKY nwnYycFeF8Bej3RyMNeF6uTfwVwXxVynotbutXsM8hZG3jWQt1DyxgoIXCyBGyBwoQSuyCBwsQQO NptgAgejTSyBq2C0CSZwMNoEo13AaBNM4GC0CSZwMNrEErgORptgAgejTSyBG2C0CSZwMNoEEzgY bWIJvApGm2ACB6NNMIcX1mgDUYrjxggbEgTYkBBxYwR5CyLvGshbKHnP3BhB4KII3ACBCyXwmRsj CFwUgYPNJpjAwWgTS+AqGG2CCRyMNsFoFzDaBBM4GG2CCRyMNrEEroPRJpjAwWgTS+AGGG2CCRyM NsEEDkabWAKvgtEmmMDBaBPM4YU12kCU4rgxwpGrAhy5GnFjBHkLIu8ayFsoec/cGEHgogjcAIEL JfCZGyMIXBSBg80mmMDBaBNL4CoYbYIJHIw2wWgXMNoEEzgYbYIJHIw2sQSug9EmmMDBaBNL4AYY bYIJHIw2wQQORptYAq+C0SaYwMFoE8zhhTXaQJTld2N8Gl/Fg0qD0EUQeguELmpPBy91AXs6CF3A ng6GnIA9HYQulNDpxA3MOOFEroHIRRO5DiIXTeQGiFw0kVdB5KKJvAYiF03kdRC5aCLHMEsTT+ZA u4oncwVkLpzMVZC5cDIHCk48mQMHJ57MgYQTT+bAwoknc6DhxJM58HDCyVwBHk48mQMPJ57MgYcT T+bAw4knc+DhxJM58HDiyRx4OPFkDjyceDIHHk48mQMPJ5zMVeDhxJM58HDiyRx4OPFkDjyceDIH Hk48mQMPJ57MgYcTT+bAw4knc+DhxJM58HDinSQDPJx4MgceTjyZAw8nnsyBhxNP5sDDiSdz4OHE kznwcOLJHHg48WQOPJx4MgceTrzjfYGHE0/mwMOJJ3Pg4cSTOfBw4skceDjxZA48nHgyBx5OPJkD DyeezIGHE0/mwMOJF3MJeDjxZA48nHgyBx5OPJkDDyeezIGHE0/mwMOJJ3Pg4cSTOfBw4skceDjx ZA48nHiBsIGHE0/mwMOJJ3Pg4cSTOfBw4skceDjxZA48nHgyBx5OPJkDDyeezIGHE0/mwMMJJ/Ma 8HDiyRx4OPFkDjyceDIHHk48mQMPJ5zMicS7DnLQCkn0ew/0v+6k33WlbpBBizx+/yBLo+gljl/e zS/PzzuDQavFXt9/X9zxr7s4mmd4745zD0cLW6TDnHR3nHRszdrsNVOzNqdmbU7N2pyatTk1a3Nq 1ubVrN1ir+M1a3ParM1pszanzdqcNmtz2qzNbbN2m71masZpszanzdqcNmtz2qzNabM202bT8VUc lsEN+V67lzl3Me92pNjY05j/NGafvuM+fcd/OnK7xda8FaRjqthiqxhJhtlkd9xkkbvf2UL9U6ZI lXHibo17Fyv82wb3tiLzb/NLVPhFqvwiVX6RGr9IjV+kxi9S5xep84s0+EUa/CINfpFVfpFVfpE1 fpExSEREi7mixXzRYr5oMV+0mC9azBct5osW80WL+aLFfNFivmgxX7SYL1rMFy3mixbzRYv5osV8 0WK+aDEr2juuaO+4or3ji/aOL9o7vmjv+KK944v2ji/aO75o7/iiveOL9o4v2ju+aO/4or3ji/aO L9o7vmjv+KK944v2ji/a+e0ndvx6YoeFJ3aMemJHhCd2HHpiB4PA9pPjl1r8Uo9fGvHLavyyFr+s xy8xUxLGzLXCXKvMNVM1zNQNM5XDTO0wUz3M1E9h6qcw9VOY+ilM/RSmfgpTP4Wpn8LUT2HqpzD1 U5n6qUz9VKZ+KlM/lamfytRPZeqnMvVTmfqpTP00FklM/TSmfhpTP42pn8bUT2PqpzH105j6aUz9 dKZ+OlM/namfztRPZ7sGUz+dqZ/O1E9n6qcz9TOY+hlM/QymfgZTP4Opn8H2XaZ+BlM/g6mfwdSv ytSvytSvytSvytSvytSvytSvyioXpn5Vpn5Vpn41pn41pn41pn41pn41pn7kqmKbz6PNZsWVzVJP RlPzGbnkjvTkWJVB96f9e+1fx6hpd196wydk9RzbnIycN/THioxdxzx6IJkePdmDwdH3xk2r0TwI MqySDFujyTPNcDJCDzayRkMbPY4c9Ic7dW2HPnIgLRLvWHpQLDr8N1KYV6OZNyh9sVPuPjsVyurq 7L8CWR0CY5U7Y5UDD3Vxe9M4baI1WjQOcFqzUPZmDBCXZy0AhKCACGXPAgI0hLiA4GgI+ivA4YPD wS/qa+PqirSudIT+u4Jin63Q4gMjjpWv9qR1BmARUXcEok/C4RwAIS4gzhOQoD/7K5qAiI+MCHYA ObxojdpOd+ie3i++n91vP7Dc3fGA0wbglBo4+cCmDfqmfLDJX9u0QduUGTbZg4b61lz+13dAzQdH zZbz4Zn4k6D4r0sAhcCgoOLnaop7E2AhtK4gAOBqCwCG4PoiAYxuv9d1bese2PhysPHNy9PbxsV9 /pT8KTBuwLitws0pdz5zCpwbcG5bAwdYN2DdtgMOaBzg3TaCDTAswq4y89gVYGEFhkOSgSUQ+X4J cBBUOxDRs3AA3SAoGJKa4ezHDwCDmGCgok+akW1zDIAQ1ZCkwo9Dwhz1lYf+z/vuHaCiBKT7j2+n d9vBIwQCAw9n9Ce92wZ8iI6PGRLS9AcGfJQDHzgf/QH4EBsfKfpjPJ1Q9fEAw8vHh8dZo73t6BLA IAUboDrKgQ2cAzaGAI5ygKOVBzpkAEcpwCFniA130O33v9+Dl2EJsHF7RSrz/T5vF0MfMoCY0iAm d8A89J4ALqUYfC6/FgGW4cj5CYApCWBa1zf5DEnbnsk/PxMvfiBa/DSs6FFIsaNumINO4odcxE8u iG9Ij+8zZjaQMlsHE5vGEpuFeJtEGP9/xqeb8dRl/C9jfnUxr6qYT03EoyK2os6spybX0jjLJ0nC nMORJlixBBWSnP+yc573Oacys5Mifzcyy0nfKqfJdGhHswrftjck3b/fD3LWMsnZz0rNLitlq6we esPI81scyOlnYI7RRlU+cmz3yL26umncSu50gC5bt21qnDWvvx5JO79O5a/ZfvY/v2CA+XLZurz9 1rhAfouoUl1S9FnTousv6LZx3r68biHlUDnEiHzB8pFiHCkylskjCONjtUay+atw7Uc+79QF6Mia Mfrf6U0ebHfSat0rtVf//yV6qXYJX6pdRkm191hS1mYvhf6G/IOz44aHul0NUi2PP2bZH6Cei6Zj eiS41Z3Y0j4ZJ58y+pDMjqKocK86zdOLi09ZfmTyMTTN/1s14n/9j6J+wlgzMPlftUruY0WT8Sck fyrgMyWidhD69PKyPN2q3z/oB1f0iqJjTdK1mlGXK6pS+QQfcT7J/v93Mv34VGj/V/Rq2P8V2vVp /1cVGfp/EZ86wgZSFaShqo6wbhBFoKKQYntp3avKq3aPJRP9ciOh8z4lNv/x3O3bb79UsKwjGSn0 efKoosiGJNfnjxKj4xUbKY+q9CGFFkueVHUNS3UcKRS/4nvc5T4JHTb//t/utBoS+V5M/1dUo5ro /yqWFej/RXxEZYXQWePraWsdTqjK54Tqfi7AChUw73uZzfsCfpQaKGjs2CdudvO/l3UqReo+n4H2 cb/3sCC+82sJSuJ3p6+9/qzoebn69rGi/AwjKwJpwa9IWbOwV/rWSxBhYeu9qi+Qvv1i9/GirV/6 eOdXjmQcht9yn+zhYnkls2yvyD90DWuSTUSvaN5+Zn/MC+DzEzuuOsXe5fBx8ToL4GX+SmvAL8u3 Cl6p+zp+eMtXRn4RBUrJL69scjLHrpmvmGgJBUqJFlc2Ibl2zuqOFFCgiEhpZZPQ0Bko+YqIllCg jGhxpRuT3Okg5yGJlFDkiET9GEomJCtv084q1LSzymfauX92xzkPR6SEIscjUlzphGR2+zkLiZRQ pJBIcaUzvkfjnOdItIQijW9SXAlnsvlPZAuex5ZOSL3uoPuar5T8IgoUk19eaeS0q5N1vI3+jRTG GzxLktHP7I9Z/nyRZ/g6O5C/yby07Z2eNqLoLaDoC6PorZwoeitHit7Km6K3CqXorfJR9Fb+FL1V MEVvlZGit3Kn6K1iKXqrhBS9lTdFbxVK0Vvlo+it3Cl6q1iK3iohRW/lTtFbxVL0Vgkpeitvit4q lKK3ykfRW7lT9FaxFL1VQoreyp2it4ql6K0SUvRW7hS9VSxFb5WQordyp+itYil6q4QUvZU/RW8V TNFbQNEvp+itnCl6Cyj6JRS9CRR9YRS9CV704EUPXvTgRQ9e9OBFD1704EUPXvTgRQ9e9OBFD170 4EUPXvTgRQ9e9OBFD1704EUPXvQ5UfTIzJEqNfOmSs1CqVKzfFSpmT9VahZMlZplpErN3KlSs1iq 1CwhVWrmTZWahVKlZvmoUjN3qtQslio1S0iVmrlTpWaxVKlZQqrUzJsqNQulSs3yUaVm7lSpWSxV apaQKjVzp0rNYqlSs4RUqZk7VWoWS5WaJaRKzdypUrNYqtQsIVVq5k+VmgVTpWa5qNKFOWTmbt6Z xdp3JvDZqXy2mTOfbYLL+RKXcw9czgtzOffgVBg4FQZOhYFTYeBUGDgVBk6FgVNh4FQYOBUGToWB U2HgVBg4FQZOhYFTYeBUGDgVBk6FgVNh4FSYvFzOvRypUi9vqtQrlCr1ykeVevlTpV7BVKlXRqrU y50q9YqlSr0SUqVe3lSpVyhV6pWPKvVyp0q9YqlSr4RUqZc7VeoVS5V6JaRKvbypUq9QqtQrH1Xq 5U6VesVSpV4JqVIvd6rUK5Yq9UpIlXq5U6VesVSpV0Kq1MudKvWKpUq9ElKlXv5UqVcwVeqV0eXc y93l3CvW5dwDl/OlfLaXM5/tgcv5Epdzd+JOCvQ5n9juhGY+Kz0Lt/Mwz1yO5Ynlne+RPPGi9Izy KSyg7bviyMoJR1aOOLIAR7yoC++KIzMnHJk54sgEHPG2Ur0rjryccOTliCMPcLR6d5yWze44bT27 Xdt1d9yTaaLDi6Zye/n/GicarmuqrKHDyxUZ9IZmf2rZqxNK0tFWiY/M0XDi9B5I3U4t6578+VKt 3l+22o2vjZuT3nBC7txOaBvfToe3k7e+TVO2m/fXt/fN3nD6Gl6e3px/uz8fObYS3jj/0bn69v9O FF2RUXjr9rahLL7i8GvntmFoZ5ft2/D66+nt/WvNMEjrXPzjtNkMb1/8s3l5NkvTIrmfzPJq33Ru 2/fXVz/Q4TWZ6o0Gvcnho9Md2IfjUY8CBB0+ocN/dD8fHg5H9qttupOu+ZOk/HF5jg4HtKAReh28 uHbXMZ/R7Is0qkhHi9uHY+TuBLzJdGgHs4sne/ASAgIwkTsmuIAYPI4H3cnzieuSug/Iv+TBa2U1 UIKvI+QOXqgSlEa0KW4ap020RvPGEXAUZmFWnAFlKGi1XfpD6/5BJhlHrzFzfRe9bjPp20z6dpD+ ldbZnaD10HXUnZCK3pNnWtIz7Qg2acX7Z7tr+T0BHfZRTUaHLXT4irCBDt8QbR0ko/+uIN7nsM2m rIQJb2iexFpoYnTeNZ/JTLTn2ajrohNkaN/PpHdLZ45pPTeX6/npbeOWJroPpKUqr9o9lsyIBInQ s8m9fW9NB4M3JZp7WwpIu40R5V79/fz85BeimH6h3780T7/envySfQf6hQOSxl07LO+Xnay1NC17 udFzuWgNUPS5KnqqllnEszibKevDi7NG+1SmeTR/fDvF8a4JmBEVM3glZjBgBjCTGDmXY+YuBTM7 zMLZYS4+3q9pERY33sPoDr0uw17X3nZ0b4OmFhgz243ugBmRMbPd6N7OdnQH7vij8oQz0m3BF3L5 sby5vxUy7fce6H+koj7rJ3XXfCBMXVm91EOvnelQcp/RZip01oCHp6hF2hbV61V0OES4Wq1UKmeN r5et1mXrK/rauPo7+nFzeXXZvvx7A7Ubt21y+9jn+ugFap+etD6jqxPy/GfUOiHPf0Z9q3uCZVn7 jB7sSfdElmT/g27bpzftxsVWz/44JfDc6FG8Q7F4+2JlqbbD29ZixVYqTPsH9xHtgTTP28oui2i8 uU5WizD7Tb23gXrfjRxXXrERp97bmVHvLQ713kpQ73s3FQfqHWwnmJwDZmByDpiByXnm1PveLbXD 6A69DhbWATOwsA6YgYV1oN6BegfqfWPqvb0j9d7mEsq1dZj3dR9NMuArnsTbF4q3LnQZ677uo0C6 g7/7Pvu741d8j7vg7w6kO1hNMC0HzMC0HDAD03Lwd4fRHXodLKkDZmBJHTADS+pAugPpDqQ7+LuD v/sqbbgufxsc4zl4Mbuu3UKDF2c0nZxwaFny+4lO3+Z1eqIq6G1KtEMfD6b9k6qOeo/97tMJNtDT 4IU+/wvp1I7tHrnuG82EPncvRyc4oZK5buIfN40vjfb5NzS/pcxu/YJGY5qXgogmayJVr/1CWpz8 IQ2uKnCsEgwf4bFKVIZUojsdrDTPZHa0kg/goBfI939l15+kUe7j0fz699bn9r9IH8xsUYp02WBR ivTkwzek0DcjvXfpqlQsKSxLZbcstXxVShrtBRcEivEDKsaospopxhn4hHJnXraSup4pAoQs9Ott +vU64II1WQBWIcCChVsAVi7AgtVdWN2Frgmru4AZWN0FzMDqLqzuAgsVW92NcfTLF3cLINY/9kIv OjxH/kpcSOTTNV9ekx4O6JIZXQdWZwQ/b1kORdbXwlySaY6RqlVVqS4TXGF09aV5/eN2WWK9WpMU tVZTqmslViW9hnVVjyTmp9SwVJ+Xn/FGpSWroPNgDNFVUI0ugqrrrIIGmfivUdV/yWFXlAhLVRpd flLXWqqKJYWIIbBUBUYCLFWVYKkKiCPosrDpDzADm/4AM7DpD5aFYHSHXgfLQoAZWBYCzMCyEAge loVgWWidZSHsJxuGf2MrQ7GlChRdtYn/coywbtQlRVYNnbfGw0mNJaWqy9XaeqlVCcvVer3GLglx kmqSrua8JtRm14QiUUKia0KkJXk742TuklC7kCWh9kddEmqvXBIKDspT1j5ST4E4NhnHsVkexgYW hMA8gAUhIIhg4gbLP4AZWP4BzMDyDyz/gLsH9ENYEALMwIIQYAYWhGBBCBgfWBAqZkGove6CkIy4 6xSJ9aB2ZCFGUWQsYazWNS1lhYdJrUsq1mvc7UTJ1BhLNZU8VOWtBzEZG5Jcz2k9KKs1mnlefyzW Nw5Qz0XTMZqMkNWd2FKWdV+2jvIYro7QDTP+3pl2CAF/wSRYO7ldJFOCw+D2fhUlWL7rDUnv7PcR 0QFpe7oIftKW9gYvtynnYWaKhsx3AK3IynXMSC65bBeaN+F2VclyoWp+hN92VbmdV+WKiJwuTG6W kVTpPaLf0aGLfnk1zdHDvyUyDP6C/oX+hibP9pCo5P8gcA5KWCTYsAySGcnH7jNFkWLmpUQ0frSw bQqa5fPYC1tmvmYrPa+7zCuxj9Jl4s+3/9okh8loaj6jDYywoDBazcwHBjWTrEibbPTEkT8GXv39 pnGLLlu3bWLH3Tevvx6Fg4eyVaWI4XFQqfw128/+5xdg+stl6/L2Gz2WmbaIKtUlRZ81Lbr+gm4b 5+3L6xZSDpXDKiJfsHykGEfE+JHJIwjjY7VOsvmrcO1HPp/YDzVho7B0W+ets0/ZfugR24am+X+r Rvxv8FE/YawZ2FDID9VPMlawJn9C8qcCPlNifTgIfXp5WZ5u1e8f9KPUKp/gI+4n0f9/2M7j7XQw kCavk4L6v6bL8/6vGLT/a1QlQP8v4HPeH5k/7x0ymw0Iqqtnr/K/kUnvBqOtP0snk7nrTvuy1Tj6 cXN91mxcVU6CD0InJ8mvwYcYoapRk2Q/m7pS1ejX8+eRS4zdn8Q0GpAEuFqXsJ9A07EiKQh9tYc2 rY61SKMpkuanUfWan9z/pdWmv+n+Q/S3mqrTr/5v7Ra1gA0lzDvwQA1+e/F/qlXpNfn4ZESY5QvN EWmaVA2ewopBCyY/ORVh+v8N/i6500GB479WVauL/i9rdPw3VAX6fxGfv1T+EvTwruN039Bz10WP o35/9CedkAzs7pD+fXhDvaFlvx5X/kL6BT3n53o6ORw9HpqU1/lMGZ4HqjFsy0+AkwmUWAIlkWA4 ChKQ0vwUKknxwxk99O0BGjt2f9S1iEaYjEhOycRaamLMJKb/a5AS0U/bGdp9RCl/+3Xiv/Zg2p/0 xn07/M0NXva035/dQH/2Js/I6Q5/3qD/JK3w3H2xKcM5cXrmpDcaup9Rd2iRhrTRaNh/8x+fuqQm f9LpvWk7k25vSIu0en5yP+XAnkgItZ9tRCaMk1m9ekP/6Qm5Pauiazs92/Vr6lfhRA6KM7tD9ECS dft9UtTjyCF339BLt9+j5Nt4OqHvjA4z+pCsvszhgdHXxg2eN4//4uNJb9DzwpqMIjImPxG9PrTc LGtzeXGC/aHp5CBgk5yf993X8ZtkHqDTTvvbycGNhM77dPHmH8/dvv12gL63Tq8aQXLXT3+PDwLC JGzWgKxs9Ccu+X572yD/ksZ8Gp7Ovvxz9uWOfBn0Zn/o7X92SH38zO46dKdB8+L0qtMkP5xeXJwH PiMy+kJSyWTCev7Py9vrFs3/rlY9Cd3j/c54giWqAmua/X9l9TO5kLGhyeGFYii6pvoXuiRrimYo swtdI0/Ri/eTeCjkeD+kOCYqwMfBBxK/JteNjSHg5+XjACDAQgB/OAhgtWooAIFtITCKDfIP00nE DKAJCSIcm4yRwRDo2n5e/phJRkpS2iGVCJqO6WoboGZz1FS2sf/bnVZDIt8Lsv/Vqi7rzPxflYH/ K+Yj6iICOmt8PW3tuIQg6CLCfOFqi4V1f+XKtDZZQ3PQ35DvKuCvpIX8BLUuTtzFwt6Oa/yO74Dh 4J+u3XXI/JA6YKzv73DedcnUzAnSHW8zuuzFBIRWXGErrgUOC0XUXJvXXNu84ipbcUwqLhddcbx5 xTW24rX3qHhtw4oHMYvR2ent5Tn63rhpNcIQucc0w+8BjcGg/1f8Gxp3XdpXHrpuz0Q0lDQveYi5 X5U104ei/lVdM33Ywr9qvPQX5D2571Wp3E6dl56/VE9PHHChp0NPF6anN/HhWfP6/HvjApEJwmXr K80OHzOwR79enajEqKcRt4O/NOQ2/fYberInbrCookbPDFGOWQSuzKRekyUj6pWqHrNgWJ2HoUi6 EslDO2blskYehqTMHVl9xZFops/o/Nv1bQMlFcSvYVv8RjTL+en5t0b4BPpyfUNaNvHAZ3Tl9/QW +beywWQ3WMYLzd6W/y95D4TIJN1s4vD+4LE/GlfiK3lpf2eLe9QVGSX/IqTP1vwMqV5flhIHfzGW dalaXZpnVQtTapJSW5rSCBYE61pNqi3PUzfClFWpri5PGeaJDUnHy1MGS5V1tS5VV6SszUtXl5du hO1JIbv8jYwwz9rKNzLCehqyJBvLU4atZOiStvyNqnIo1JVvVFWClKou1ZeXXjOClPW6VFOWp5xh SZHUFfU0wpSypC8vvVoLU6qSvvyNaso8z6q8PKU2X11WlyO5HuaJZUnVludZm5deX156PZRRnbT8 8jzrs1aqSrUV9dTmLW/oy1PW5tJUdTq03LYR9QamjUL1kU1Gv9Gjv3AXsllhM82HISWjYUjGkqHt NgyRPHYfhyirK2Ftt4GIZEIAWo2PRMomI5HfHKUbibC+rk6ozUcibYWe0eZ5Yqm6tF+o8qyv1SR5 aUpFXuRpLO2Vyrz0qoSXaiQ1Uk+8dNTQ5vWsr6inEXmj+tKerssLPaMsbXldX9gAxtI3qkbqWV2q tw19Ic3lY1ZVj717RCPRlk5RSUHbB2mJwReuaYQLIlnOCvdnISF4VyXHd92v1dYqVqr+iweblBwc blPKbFeRc1AZ/CSP0rxH04kVnv8XXh39dbEnZf3tH85RsAHGwWnbSiy0wW6SOfE631njYPSz1+/T 7Vp/i+cc1H6bqrL7VXwe9vmJVD3YBfclzvseWvMWC7anzFpsnt8mlfD3uMzbfKMdSM58g8zi+Q12 xuS85cgpaMeRk9xwtOWeHwdLz2wnIFk7g2gnyWhPkJPZliBn0x1BzlEMzvEtQTstjCh7tzCynWku 16uSEWWItjLvVaLRZ3mU102MtJVRDQYqYZyhZClwHCAXcvwi6gkjjmdQenuwjbPwDCp548gqVnQw 397DfNvlQLRoAQeVroOszc7rQFFwSqOKbT6Psj3xg0kdO7M7LJVi4Kf05Fi8X3c6SCfePGAng51c hJ0MWi5Ny2mg5ZZqOQ20HGi5PLUcBYeSk8ue8jM4lqZwV54cHbEW78R3xdrN9emDtdhyh5bFy6wm G2qaKtXxZm4g0cb6NcxhxfJb5JFgAe7DeILgmqJLNWMdT5CaVl+5eq+FKRVJq6/hCYJrWJeq6hqe ILgmY0mR1/AEwdWaLqnyGp4guGqoklJdwxMEV+vqqnc3wlaSydReXcMTBFer9ZU+I2E9azXJqK7h CULquWr9beYJUjUUSVXW8QSp1lQJ19bxBKlpWKpq63iC1DRd0pR1PEFqSk1Sq+t4gtQ0Tarp63iC 1BRFquF1PEGC3r+OJ0hN1les0M48QWqqsgp1dXnd9qwbc3wuX3OeeYLQN1rhsVKf1bNKZRRZd6Wa gL/uGuqG9VxBNlHgdU2WiKbayHsirsCDHMqnwGcOFHWsSlptHQcK6vi23O1u5kBR16uSXl/HgYK2 Lq6v40ARyGEdB4q6UVvhyjdzoKgTNbLc+WvmQFHHValeXceBok46yHJHD32esi4pyjoOFHWsrRiS qvM88QpHrZkDRV01JE1dx4GibmhSVV/PgSJs+w0cKLa3LTHlyXejondgoonKUuXqhh4U279sBos0 O72sWpdroQsFkbEu16McvJIjB6/E2CklA3ZKyYudUjJnpxSeC4XCIU2UCGmixEgTJUqaKNuQJsqO pIkCpMnupIkSIU2UBGmilMyFQuG5UMCpqu91qqomw6mqJTj/TSn+/DddZc5/UAwNzn+D89/g/Lf9 P/9N2a/z39aYOzGOTErebm+Zzag0WZfx7EJXcC04/0mRjKqu15XwQpd1VTdyP/9J2eNT4AAE+w0C /BFB8N5E0scGwcc5Ca5suJkTkOvZ/3ejF9uRngu0/7GsYDb+Q1UH+78Y+7/3SCz7R3R7d/33xs39 t8pfyFVvaC9uzO9QuKutM6Toyvxeq3V/1fp+f4Vqmiyzd1sI1+rJ21ctVJV1I3H/O0mu67rB3v/a QLr2/7N3rU2N49r2882v0Jl76tAzRYzlZ0JVpiovmHQSQxFzx+fMmWJCYiDTeVUeMzS//kqynThm 00AIdNN7M9NJbMuSttbey9KyLRXcVKE+WKgfFwrtvV+mn5RpcKPoZPffK9MHDfUTQ+Xz5o51L3W2 VD+xFMpFlGkYTrHA1wdAQ33QUP8BQ/0HDPUzhub+NxQDkytaDQT5+h+XYkArAs8o3Kr/32L9D+46 Gf63dIsT/7/Fn+hWiD4FY62yz9jpWf2ICYZn7XPGPPGvKf4dHfnVXxhrRF9e9BV3MKJ7FurBAzv+ ZRTijeS3Uoyim8osSW4WbKkvE9l8o/Hvf/34dyn+v+P4lwv56C7F/7cZ/75H8U/x/7z4t553/ZeL dVkU/99o/H/9679tU/y/q/g3nhX/ju7KB9kp/r61+L9rt9UT7PXa8c4WAXlk/Q/HSa3/6xhcrv9h W7T+55v8vfFaDqNRes6i6KnFu77XrCTrve9gyqLRKHo0kd31z5bjo8G4X63v7JljmfnjJqee7lxb fDccXO5qeoyVjd3l7WAoMmZ3vWHYHS+n6vdyHs7U6qmqOGcnxT3NalWfxOKVwe72NVAZ7q2z8SaL G5nLYiIfe+lPxqG6f/iHKGtPWyfb7r3zuDBhavz+6/rp7Tv5cPTVYJjG0d1Bwx5EJT3BwrudmPhA wc6XCk7cKSndeXnpsfvm/xQdB2mYmkpgh966yuqPJPc9Npiz5VRaJ28i79qUl8/p8gJq6b8qtfRT 1NInatk5tfRX1NL/StTS/1rU0n9laum/ArVc93osP1G3w/tyjzZhebGn1jI6jf/US5Z8uU+3WL7x tBdZHk+YetHjWYkP5LOjMxG0+Vq5378QX0eue9Hw/Ppx/awkH2jJ1zoL2Rid5biz+CwcUKQUZp10 LlqD8fI23iyfVX+5qE5moRHvqJ6et3/5T7RgfLyr06kb6588/nneqTtWpeF34u3jcufituA4onVq v8oXC6LdtX+3GpUkjSdyLyV5+WfnHf/ipH0q4mQyGizyV7PuKMxPJwMJI8uPrqaj7uKmNJ+Luo/E pzjxRPy7Zvlfu/v5/HgS3oY90cXvfRI5nDaqIpUsPfpZq53Vy3Eljuvttmhd7WDznR7GnvIazqb3 Rw8SCcfo3fMVzyNveefeknWPfM2b+LPueF6+WP+uXGzvNkEAOY5PjvNdO87ruI1PfPP9uc3rs41P bPM9u83uncarjkaNj03ymnfuNVv2dhP47zvFxwY5BWKnkPCDTHHRI7dAzRXCAUC2IMdAzhf3HKM7 HHTnYf+CtLbvQ2trNcqdeu3i9QW3MilupLg95jdlcDxTJs2NNLetHYdUN1LdtnMcYhzS3Z7lNvNR dzhsXlDP+DvoGXfaojLNi9fuFiuXIY/5bjzm1R3mcnBN7vI9uEulcfwWzjKezD6Rw3wnDuOdnL3O Jak7Y7MtFj7rtysfK8q51Faj4m1spY/FgqH6fRwuvEp6o7reXMlEm0/bbD5EsXlvPHPTM3O7696N jnsCNyRsZjSrjA6RGV1mxgxqc6P4jaqJI83GeiO1v3J6mj7d702Tzd5kaFwOP110g2TPbPK33OOn dq0S8fuJVrumy4VMcxlkd9xLMb63R3+LBe6i6bbvPyv6DT0YriiWqPTdUalErVY9bQVxaSrE5MXy rn+l3kzSJk9h0c1XkA7iU3vbvi8ga3LXj16Niq7Ui8/TkqrnLl8jSPULNkujfsE77RfEniyxFM0g nLl6Wq1HfYOX9AWIJ7G71oTdJoyYpsZsh/Mpvc77XrZJPs/rc+aH08WNWic0PxzlHn8jVm7PlmNt fvPM92LX9uevgLdjn1PyIRvKKf6sQ7kwlmmyTng9CseLrpx2nF11l8NF1JD/FEWISDv450/r92N/ +ukn9lvqzdnfWX02m8wYN4u7Wjpk1TMyk/LumZuUauxqkZFVmUZS5rNOPnishi9YceRbe///rN7Z 7QoQj83/YRvu+v1/bsv3/7nh0Pv/b/HXqJX0eP7j/rW8FieTHkdzHR+EgukFbe4xv1za88VXpbTn xXMdR3OHlLiaN0TOUlz2S3Y0Q4i4aqgJQ0pWNH+IONqstMtyhmPx3fDk93nJiicr9uRVpi22Wbsi d3rqs6k+W9wSmzo7PapUT1od9atcaTWjH/Gujn/UOhGlx/m1aslmWVhWl/Mxt2rliqhEpe6XPa62 G52m3NM8Vz9EcWfnnt9oy3mUvfXPyND1dqtW9U9OobmWLc0sFI2C9fBcy+9k/o9OXfYydjX5x2Px b3OX6zwV/46Mf8Oh+T/f5A/rcleCCY7L3uOLXZl5/vBiV0iXu/pqc8aITmm7XQuv5JISpbudThij HdyORvOwO+vdyEUT73Y6YYzI/Hoz93hYcybHFYOxXP1ALRjA5oO7kHXnrMQcS/Qyv1q6s3q5JgJC XJnblfqZDAn1uOZZ/bjRERfUjkjInX3WLlerJb7Pht1FyT5cnSUv5fvRVf3o7ESeVm7ts/FlyTX2 2eiqZBmuWuk8d9TwanKd5PPTU1FK5eTcq8lo8yqHOa+ysVctvizitcMO5fLeH+Isflzn0Tr5NZuH sFQUykxR09GyxJio1Fh+i1p8kt88Y0FcOb2gOc7qZOOZJ3PNtmXt0/VJ19501nVWa8KKtGWRRnai 6nJytU7d9+VB0ZBnZe+4zn4TBhT0fRFxaYumV7LoZ1TNNItasbDKorBFFoZpaPYqC0vfphaFQrR4 fJSFsU0t3KLm8FUW1ja1cAqasTbE2aotTM1a12Kb5jRdPVULe6vm1F3NLK6yMLZCxNDcdRbbNKdR NDXHWGWxlXfahpwRNMliq+bkerTqvMrC2ao5HUsz9FUWW3ln0U15p7OVd+q2VrBXWWzlnYIe3XVb bBXsjqsVVlm42zSnYempGHG38k7L0fjKtdytvNMppvzC3co7hSHFtSFRc+rPq4WdppzCVt5pFTQz XklcjZH3oytIcukQw9kPMcH+mPPP/i2vIs3zOTs6OUvSnx7JPFMX6KQYNXhXFxgxMN8o1iwI49XC 8vKQoW9WybDWx6zNY5aIR3d13sYhnXOtwJPTNtuJW1oxOcvJNKGjGWZ8qLB5SJCYoT9+iBuZY6am r445GdMcaZpq4bRxqTbeuI5HV+/Tsi86SJ5qcwG0FTlJhG+ErqXKG1yFi95NSSI+jn/yhI+dmECA NEaSRs2fnkpj3s/HEl0R54E0ST6GofHCA2nMdb8husQkaSygznbctwDSpOpsbtTZBvMxNtrHuZ/G NWTnUaSRcxWzXncuZ5NcDEZhX3Zhk6JErClYGlFoSH+PfhkKQJlgFSStk3ItX/bz/skp+yAXq8t3 F/nLyWIxGbFVWlFgcmgxmcZ1KXLZV+z4jWpTZvRrw/9llVvlxPdP2kkZnV/Lp7IjLQOOfRAjmHk4 Xohcuv3+vQvyqsBRt9crbUaVLsx39FyzXj+V2ars4i6j6N936i0xdq3XRH/dq5+V5a9oMq7DHEl9 ryL17fqmyO7GnMuNMSfLj0WPneWjqVfv2u1mp1OXQv+L7viD4+hPUbEX6lagHO46NptcPX89+9Fo ISJcRrZ2c8iOyo2WcOfqSbstx0uHkSbQG42Wi/lCscBoNJssF6W763A8GsnStR7rzavC45RgwNol q8A8+dGUH6NL+TlWn5/U57Dfjb7irV6J6zr7mR30w78OxsvhkBk//4urUbQoY76Ysb/VoHkvmuBV 7puHTUn4XPI9l3TPlacXlKeLC1NTbdyJdCL11WZd7xevszjhsHs9L/3ww/267OVqYXRLrzcZTQfD cMb+xVRy1p+w8WTB5svpdDJbyCUIBSn84x+5l9xuz+JdD+TVrtxikWocg3SY9bHOebVa73SOzlvx QvPq+pj20Vwuouvzjhg1V8uduhh5u/uy1Rz9kBmWK664ejzruEx6LIbHlXKnURUM6At6SqZV9EUm nagfE62lLPPqHMRPuFm3/LaoX9wWXK33QRTwI5t2RQvP2WV3Pugx6W7zXO3Eq8d5pzPVkioKEl/V sqPG7lEr1JNWqHcU24r8Y759sA4PLjH5MAfrEQfrioN5zMF6zMF6zME6xMHOmoP5ioMdPWpj6ZjO bjlYv8fB+pc5mGc4mH+Bgw1NuISr6/Fap1bBdG21wVi8QuZe/pZJcEeXIijycvHufG863YsfKRA0 clraU4+LPOUJhb1d8rypF80nEmLmNvRo1FuOe9MoZLSezMndTU6iSRRJir7GZDI9KX1sNPfZolvy 2OKy5O1LspRd/3H09Sn6koSlvpLNXinqvWW6vIbqBQ/TXR1WEn8/ix7NcDItmbbUIBwBp/7tVCR3 51WYVLnYhzwXo4QfBRtNBMtOP7PoiCGORCe4P+ZeXO+42kmtbVVr/kCtjc1aO7ohhgPOLprvWdWw Mo1XdDWXF1+5GoVHIDRcMYaQCD5SC/91MTFNV/TTnTevRqY1RHNoNn/taliP1IJbciz9RUz8N3BQ h9ta4StUIxuuvKAVndeuBn/UNXQBimM/Wo1X9lBDjKELhQKOajwWKLptaLbJZS0EPPquegmiG7OT nHK5jl8+U7enqq162Ts/jbu8uad4ctRmVtRmVtxmVtxoVqbVNiWzLzebxUVn2nUKT4woK6rIa9Sj aGimVXxiPV6tGoZT0KwvX3seqMYXHTldB248VgfTEiM281Xr8EgwFQQaBjdUZy4a1WWdNtfw/u9E KXfHqwefqvtyhFXau9uLT4IP5cQoxZNnZobA9c4q0X4kbx2KxEeTGVvcDOZsthwL22/CSLycduVD 1ItwNmdXk+W4z/4OZ2FW+o+l+7Vovx8pajlfZnjd/StkXTYNZ1eT2ag77oWy0xppiJYwPRovaSJx yMZhL5zPu2LIIpcbmLMbee5lGI5ZbxZ2F2FfY6xxxT5Plqwr6nHTnYrOrhw55WSNu5cTkVy17VxN 3y/Szdj883wRjvblGyRhIoQMxmIUNRzuabldPr+7xWMj2sFtLGuyWAQ7Y3kjF3tj9Vzdw1GudqXu 5wyuSnkuFWf1NZxIFISjcuntwlc5T/oVKVXfWvmi8mLlxA/48D7rzcsyaW9eib4Uzt3h9KYrf1yG srDczpdCUnn9sRZb4LVQdoOSpReN7S42MVLiMmPp3HlpHmbRtV6eR9F+uS2O8/I83OLL8yi+vB5S Dn15HuZL81CvN5gF3d5VRu6uMiruoHleXJltn6+LafsVHq/buCzs+j7Hu2V4e4cMbz+F4W1ieGJ4 Yvj3z/Cjv9SahOL3o68ZJrdUtfik+C02ORvFc05+5pvmB6tVzmYjWap6DfMnbbKbfF6ay3oxx9te b3L55+aybH+lKtzbppxVW8eb6+K2y2xwxX5j+Tn7IaqtFt6GP7DfRVXFqGzM/pv7HzE0lrMKXLF1 gucVpYnMRD7hMFOUKGZVSuoN3HRh2xSU5HM1iHsGo9FLVzDc+SKtztsu0up+9ZUUC7tZSbHwtJUU C2+1SGthlysp0vQCbzC9QJShkuraG1vexlbzlSZsUcwvZ5WSsw/1CHJckHOCHBvkAUGOCPJq9VzN sEeYo8P8okmoY0JdYd7l1I9DiXtAuGONd+rE44x3wh1nvNMIDme8E+74RnE+YY4Pcxq5o2N4n0bu WHGnkTvaeKcRHM54J9xxxjuN4HDGO+GObhTn0z13hJjTyB0bw/t0zx0t7jRyRxvvNILDGe+EO854 pxEczngn3PGN4uieO0LMaeSOjuHpnjta3GnkjjbeaQSHM94Jd5zxTiM4nPFOuCPCfTltb05h8rSz yTW+FddQW3IBnZIuQQUXs0lDrV9YFzp14lEGOacgf+9Bzp8T5NRzRxjkAQX5ew/y4DlBTt11VEHu UXcdS3fdU0HOqbuONMipu46gu54KcuquIwxy6q4j6K6ngpy666iCvEnddSzddQk1v5D/Ed5Y8DYI b1R4m4Q3KrwtwhsV3jbhjQpvh/BGhbdLeKPCu0B4o8K7SHjj0lt0AhwX4KSwIQOcJDZkgJPGhgxw EtmQAU4qGzLASWZDBjjpbMgAJ6ENGeCktCF70oWUNmSAk9KGDHBS2pABTkobMsBJaUMGOCltyAAn pQ0Z4KS0IQOclDZkgJPShuwdI1LakAFOShsywElpQwY4KW3IACelDRngpLQhA5yUNmSAk9KGDHBS 2pABTkobstldSGlDBjgpbcgAJ6UNGeCktCEDnJQ2ZICT0oYMcFLakAFOShsywElpQwY4KW3I5tUl pQ0Z4KS0IQOclDZkgJPShgxwUtqQAU5KGzLASWlDBjgpbcgAJ6UNGeBZpY2gxLPsKC0gjmAB8dSy o4Q3ErwNwhsV3ibhjQpvi/BGhbdNeKPC2yG8UeHtEt6o8C4Q3qjwLhLeuPQWnQDHBTgpbMgAJ4kN GeCksSEDnEQ2ZICTyoYMcJLZkAFOOhsywEloQwY4KW3InnQhpQ0Z4KS0IQOclDZkgJPShgxwUtqQ AU5KGzLASWlDBjgpbcgAJ6UNGeCktCF7x4iUNmSAk9KGDHBS2pABTkobMsBJaUMGOCltyAAnpQ0Z 4KS0IQOclDZkgJPShmx2F1LakAFOShsywElpQwY4KW3IACelDRngpLQhA5yUNmSAk9KGDHBS2pAB Tkobsnl1SWlDBjgpbcgAJ6UNGeCktCEDnJQ2ZICT0oYMcFLakAFOShsywElpQwZ4VmkjKPEsOxpQ lL/3KA+es+wo4Y0Eb4PwRoW3SXijwtsivFHhbRPeqPB2CG9UeLuENyq8C4Q3KryLhDcuvUUnwHEB TgobMsBJYkMGOGlsyAAnkQ0Z4KSyIQOcZDZkgJPOhgxwEtqQAU5KG7InXUhpQwY4KW3IACelDRng pLQhA5yUNmSAk9KGDHBS2pABTkobMsBJaUMGOCltyN4xIqUNGeCktCEDnJQ2ZICT0oYMcFLakAFO ShsywElpQwY4KW3IACelDRngpLQhm92FlDZkgJPShgxwUtqQAU5KGzLASWlDBjgpbcgAJ6UNGeCk tCEDnJQ2ZICT0oZsXl1S2pABTkobMsBJaUMGOCltyAAnpQ0Z4KS0IQOclDZkgJPShgxwUtqQAZ5V 2gjK73/Z0etpWy07qhPomED3CHSskU6rSiOMdAIdYaRTRw5hpBPoqECXAzfqxqGDXCDenbEZewSJ 4eBS/usuht251o0ySLr+k/Qm39wMVpvV6vlo5HnZ7Yvmeo/a7vJ0nvG+ANjH04Wt03EgXQCky9bM z25nauYDNfOBmvlAzXygZj5QMx+qme9ltzdr5gNt5gNt5gNt5gNt5gNt5oNt5vvZ7UzNgDbzgTbz gTbzgTbzgTbzM222XMkQGzv0C+tCv7c3lf9GMp5NFoDJUnu9bKGeSsYzhXrZQlPJeDZZACZL7W1m C1WLJETaW2avAe41wb0WuNcG9zrgXhfcWwD3FsG9yRqb2d2wdRw2j8P2cdhADlvIYRM5bCOHjeSw lQZspfEAhrCVBmylAVtpwFYasJUGbKUBW2nAVpqwlSZspfmAq8JWmrCVJmylCVtpwlaasJUmbKUF W2nBVlqwldYDEQlbacFWWrCVFmylBVtpwVbasJU2bKUNW2nDVtoPEA9spQ1bacNW2rCVNmzlBhmn SJWDpMpBUuUgqXKQVDlIqhwkVQ6SKgdJlcOkymFS5TCpcphUOUyqHCZVDpMqh0mVw6TKYVLlMKly mFQ5TKocJlUOkyqHSZXDpMphUuUwqXKYVDlMqhwmVQ6TKodJlcOkymFS5TCpcphUOUyqHCZVDpMq h0mVw6TKYVLlMKlymFQ5TKocJlUOkyqHSZXDpMphUuUwqXKYVDlMqhwmVQ6TKodJlcOkymFS5VlS DUBSDUBSDUBSDUBSDUBSDUBSDUBSDUBSDUBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBS DWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBS DWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBSDWBS DWBSDWBSXe2+zgod11kR4jqrcVxn9YfrrLxxnZUeEl14Y1Ns5cLezeR5omHueakXk2Xvht2JPdr1 rJ8bdT+FvxV+P2StsPvXYHzN+oNZ2FtMZp/ZH49kPJ/1Di5FpgfX4Wh00KyfefXWXpShKzL0Josb meFiwi5D1p+MQ3Y1mbE/7pbzcCZP2dPWiV9YelQsy//JjIxpMvO6VHdflLsS72Os2pWPEVZ5EvRf XdB/BZm+VquetgL2hBbddHBZsxj73oZDNCoeOQRSh4ixzzoEMQRehwAYQh4ld3jn7qCKOq6326J1 tQP23xzb+NvKW5RjbPrKcbjwKuQsGLkjgv6+O/x/e2fbozYOxHFe76fwu0qVNsQBwm7vFQ3QTYHs qqQSUnVKQ8i1XHlSgL29fvpzSGmJ421LLiHg/EcLYXkwJP7NeDy2xwaAKC8QRgKJ8OXdhA8QcclE 8A3Iddta2oG7WLecH49fO+kbltFIBI4NcKQGJx9sbNgb+bDJ39rYsDYyY5M9NOHUQ/NtD9RcODUp +8Pevv6TVLw1QUWZqQjrX2grHA9clNtaMAKE9gJklN1iJMhwZ1N37U8cROTliMj3zdaw03byD8u3 EHWThBlxlC3ZzYmxlQqjlrCL00IYTkKSCuIIcTkJODoDe4RAnVQcnZ4iBGhKO0rtiYIzCOOWmYdk CJcx0jPBQ1ntA6t7fnQw8N2Z5q1mT6CijFQcApB0JWxvBSxK60yEtS8KmhhgQqqgibFc/F1Uhxcs ydThLZAkAyG4yyfpOz/FhuEMhOFkY6koktC+yUZSMW2c8SP7AXAqXXfM4PahiENBAUWJoaBiKEaA osRQCNecwReRe83ZoWeS7QIidK4lROe5znbGa89gdWRee5ajzUEQRnabkxM4sDiyg5OP1fGWM208 ++K4oyniLDKsSnm4a41MNeUI+DcaOEaC5T/hszYgASQMkj0OYksSmimAIhko4rbJycPKGIAH8KS2 PgweoCMJOnn5MACk3ID8lv8CSGSB5MS+C8ABOGn8FkxokAQcmpPfAkBKDsjv+C2ARBpITuu3AByA c4zVWW03odsyRqxOBnJed+z0AwIRCs/xATrkoCMHNtDmyMEGzYGNBeCQAw4rDzrgc8gBR5Yex3ru zmY9B8l1JWBjOGA/pufknVl3hwyIkYaY3IEZTz8BFykaH/NNLrC4AQmO20Od29k7vq1zfE/fww1d Yxt2cts1xrfqi++/Ft9WK75bErcNDrf/SWLji8R+B6I091wGcy4NNZdNmEsKG8vuGUvtGMvrJ0jq xiXz4vI4cal4uHwqXEoMLq8BvzCdX5PML0eNr0SMLy6LrxeKLwKJz+wXTdUWTs19biamcIZdcuKU YKqMeG6EYNQ7OZwpGMASj1gIYtGCIGMirpQIJiR7kHyv4cr3Pi+PU9Cr4969WW69z+Qre0b5FEyu 5u4X/0Pzz1ek77uPzPCSyTTwvc0y+Jd8/EXBMSPzIipJz6ykRqqSNtuFf1jUt7OdLpjFnc2ikuuZ lLwrqpZdUVqqosbTxcHnO2EblKIAb0WO+snVwF9Xvw4G7zpDZb2dE9Ma2qF7079/U1X+9+lcvcxW zr+8qF3vmpY5vOu0ye6K1JRbRWvsLy2575Jhx7DNe4to17VrStgDqlY1vaqpVGUfIZS+qtdZMS9L d/2YFKQCoUOTMf0FncnYX28sy9HUp92fRCdlS3hStow1ZZ9xTU2OOynyBwm/k3M8aul+wbOex8d9 8S/IdE22K7JZkom78ZVzck4qkEsWVofVQ2XcNXkZf4fKRK/Xd8emHj9GUqtQWteprqk6bVRUqtF6 o0LUU1yALdOwgJDK4+PP3/er1y9UNBUaDP0/0P8HP/hruJ3Plc3T5mT6r37Xf01vMv2vqyqF/hdS /1xH4RT1T5v6gf3Xd/VPm6j/U8jgfb/VbhPSb9mEPLzrdAnr8ZLBe0IsduuxW7drG3eEmNHBig7s Ld3+/UM0ckHDOy28U9mR3Ugjenb3eP+G6Nnov5quaop6g6bnTPXfLlz/Nfh/Mut/rdZUdAr9P0/9 ty3oP/T/OP2vH9f+04Zyg67nuep/8e1/rQb9vyj9bxyl/9otVW5uof8QCAQCgUAgEMhJ5T+1Iz3X AFAZAA==