From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mp1 ([2001:41d0:2:4a6f::]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)) by ms11 with LMTPS id qBw3GPVRE2CyNgAA0tVLHw (envelope-from ) for ; Fri, 29 Jan 2021 00:08:21 +0000 Received: from aspmx1.migadu.com ([2001:41d0:2:4a6f::]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)) by mp1 with LMTPS id QLzWE/VRE2CeaAAAbx9fmQ (envelope-from ) for ; Fri, 29 Jan 2021 00:08:21 +0000 Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by aspmx1.migadu.com (Postfix) with ESMTPS id A973F9402C2 for ; Fri, 29 Jan 2021 00:08:20 +0000 (UTC) Received: from localhost ([::1]:46402 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1l5HKl-0003rj-EO for larch@yhetil.org; Thu, 28 Jan 2021 19:08:19 -0500 Received: from eggs.gnu.org ([2001:470:142:3::10]:47400) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1l5HKZ-0003rB-Tn for guix-devel@gnu.org; Thu, 28 Jan 2021 19:08:08 -0500 Received: from mail-wm1-x32e.google.com ([2a00:1450:4864:20::32e]:38851) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1l5HKV-0002dv-V2 for guix-devel@gnu.org; Thu, 28 Jan 2021 19:08:06 -0500 Received: by mail-wm1-x32e.google.com with SMTP id y187so5888575wmd.3 for ; Thu, 28 Jan 2021 16:08:03 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:to:subject:in-reply-to:references:date:message-id:mime-version; bh=yLDovHZmwury9X8xzh2lYKCgBsHpy+lhwSA6/hXtYjo=; b=i16H78WjNmCDHb5f4bioz90JvWHYioJswrjeShf9V/w2770mp4+EaLyhw6omgAG659 twNggF9JNbMjWPt5bn2C2Tf3htLUNdOZijIeeLBQoF72fSIyjmrK4iQfE11BLvYLsyCb W9DrGiEw/6xIOmUQACmgAAXejvHhO6XHc1AH2IhrpgINycr/YrSj2aCKXA37cyJBhFXE nZY59IhfBwP5dLs12vl6EpllJP10XFVlUS6TaGQqDXw9BxWCNZRqV7zZJVh4HtZjq13J 6pyIq+aAGjD88Nys1T1cT3YGAGoN2CbOuyzcFyhGCX/zjco0x3cl5guleNhJbhHTas/u RGSg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:subject:in-reply-to:references:date :message-id:mime-version; bh=yLDovHZmwury9X8xzh2lYKCgBsHpy+lhwSA6/hXtYjo=; b=OXdt2t3DJf/dAQQRCo0R+gbSN+bdbuCNrBCVskohZ+aaAv3F8lPsGaEoYCjpOqcjDB ue/9xqeAvprDXHDPYwQarBp21sOcyGPH9YomXMcjMJ3Sday0GV1CbuqPXZBFrldzatnF DR9H7oe7CYQ1IvxcsI7SZHNLTXaoZKFhE18pbomM5t74MPrPRwtwBNOcTV1phqOThU7N z9vZsjayXvxkSotqMZRejGzF5y427G/nWR5gRmSmhLm7+DoVL05Z14bBwvZa6iEF5zwD vny8QbYEdFjp9lDyyPWWU78r8TM8QOZ1rs9Si4hSGz3my9/5HsPDNxz4Y3KRqcvNihKC pNsQ== X-Gm-Message-State: AOAM531aGlmQZVSMxyHiGuUf5YXf15lLMmp28bdGfLlOhV/VQ3unJBlo FLRKUbth8T+lhx4IMY535uFBi44I+GM= X-Google-Smtp-Source: ABdhPJxNN1kK0B7TXK9ZJ2BHNnCGvlxWXiY+0mHQXQnZ7aYRQWKFknDEgH9rj0TMINJG61yI7FT54g== X-Received: by 2002:a05:600c:22cf:: with SMTP id 15mr1345551wmg.19.1611878881967; Thu, 28 Jan 2021 16:08:01 -0800 (PST) Received: from lili ([2a01:e0a:59b:9120:65d2:2476:f637:db1e]) by smtp.gmail.com with ESMTPSA id f17sm9708385wrv.0.2021.01.28.16.07.59 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 28 Jan 2021 16:08:01 -0800 (PST) From: zimoun To: Magali , =?utf-8?Q?G=C3=A1bor?= Boskovits , Guix Devel Subject: Re: [Outreachy] 'guix git log' slowness? In-Reply-To: References: Date: Fri, 29 Jan 2021 01:04:12 +0100 Message-ID: <86eei4hamb.fsf@gmail.com> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="=-=-=" Received-SPF: pass client-ip=2a00:1450:4864:20::32e; envelope-from=zimon.toutoune@gmail.com; helo=mail-wm1-x32e.google.com X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: guix-devel@gnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+larch=yhetil.org@gnu.org Sender: "Guix-devel" X-Migadu-Flow: FLOW_IN X-Migadu-Spam-Score: -2.05 Authentication-Results: aspmx1.migadu.com; dkim=pass header.d=gmail.com header.s=20161025 header.b=i16H78Wj; dmarc=pass (policy=none) header.from=gmail.com; spf=pass (aspmx1.migadu.com: domain of guix-devel-bounces@gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=guix-devel-bounces@gnu.org X-Migadu-Queue-Id: A973F9402C2 X-Spam-Score: -2.05 X-Migadu-Scanner: scn0.migadu.com X-TUID: M8JTtlUNILcm --=-=-= Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Hi Magali, (It is a slightly edited version I sent you. Since the aim of Outreachy is also to interact with Community, let enjoy the French proverb: =C2=ABmore crazy people we are, more fun we have=C2=BB. :-)) On Thu, 28 Jan 2021 at 00:53, Magali wrote: > Another thing is that the command is a bit slower than 'git log' itself. > Thoughts on how that could be improved? The command is =E2=80=9Cslow=E2=80=9D. A first quick analysis about the me= aning of =E2=80=9Cslow=E2=80=9D. Basically, I have run twice: --8<---------------cut here---------------start------------->8--- for n in 60000 10000 5000 1000 500 100 50 10 5 1; do time ./pre-inst-env guix git log --oneline \ | head -n $n > /dev/null ; done --8<---------------cut here---------------end--------------->8--- to have kind of warm cache. And again for the equivalent Git command. Then, bit of Emacs edit processing to transform the output in the buffer of =E2=80=99M-x shell=E2=80=99 to something in the Python file: tguix =3D np.array([2.871, =E2=80=A6 tgit =3D np.array([0.013, =E2=80=A6 (Well, to be correct, it is not twice but a couple of times to have an average.) Let normalize by removing the additive constant and run a classic linear regression: t ~ B n ^ a =3D> log(t) ~ a log(n) + b where =E2=80=99a=E2=80=99 and =E2=80=99b=E2=80=99 have to be estimated. Well, I am surprised by =E2=80=99a ~ 0.5=E2=80=99 which means that =E2=80= =9Cguix git log=E2=80=9D runs with a sublinear time complexity. However, =E2=80=9Cgit log=E2=80=9D is li= near. Maybe I am doing something wrong. Initially I thought the tree was badly walked but a quick: --8<---------------cut here---------------start------------->8--- $ ./pre-inst-env guix git log --channel-cache-path guix /home/simon/.cache/guix/checkouts/pjmkglp4t7znuugeurpurzikxq3tnlaywmisyr2= 7shj7apsnalwq $ git -C /home/simon/.cache/guix/checkouts/pjmkglp4t7znuugeurpurzikxq3tnlay= wmisyr27shj7apsnalwq \ log --oneline | wc -l 72791 $ ./pre-inst-env guix git log --oneline | wc -l 72791 --8<---------------cut here---------------end--------------->8--- shows it is correct. Hum?! Well, I suspect noise on the data and the normalization is bad here. Running the experiment in a batch of 10 times then averaging them should give an analysis more meaningful. Hey, that=E2=80=99s a quick one. :-) The conclusion here, it scales well enough=E2=80=A6 for now. Therefore, the real the question is about the =C2=ABadditive constant=C2=BB= . On my machine, it is ~2.9s and this is where there is room of improvement, I guess. I exported =E2=80=99get-commits=E2=80=99 to have it in the REPL and I did: --8<---------------cut here---------------start------------->8--- $ ./pre-inst-env guix repl GNU Guile 3.0.5 Copyright (C) 1995-2021 Free Software Foundation, Inc. Guile comes with ABSOLUTELY NO WARRANTY; for details type `,show w'. This program is free software, and you are welcome to redistribute it under certain conditions; type `,show c' for details. Enter `,help' for help. scheme@(guix-user)> ,use(guix scripts git log) scheme@(guix-user)> (define (compute) (begin (get-commits) 'ok)) scheme@(guix-user)> ,time (compute) $1 =3D ok ;; 2.533936s real time, 3.099156s run time. 0.901027s spent in GC. scheme@(guix-user)> --8<---------------cut here---------------end--------------->8--- And I let you run =E2=80=9C,profile (compute)=E2=80=9D. Well, this functio= n should be optimized. IMHO. Initially, I thought about =E2=80=9Cstream=E2=80=9D but I do no think it is= the issue here. Well, I think that the =E2=80=99repo=E2=80=99 is open at each commit= when folding. Instead, it should be open before, keep alive and close at the end of the =E2=80=99fold=E2=80=99. Somehow. WDYT? I will give a closer look to =E2=80=99commit-closure=E2=80=99 because I am = not convinced it is useful here, neither. Bonus, attached the Python script and the plot. :-) Cheers, simon --=-=-= Content-Type: text/x-python Content-Disposition: inline; filename=eg.py Content-Description: quick-analysis.py # coding: utf-8 # import numpy as np import matplotlib.pyplot as plt n = np.array([1., 5., 10., 50., 100., 500., 1000., 5000., 10000., 60000.]) tguix = np.array([2.871, 3.006, 2.987, 2.892, 2.991, 3.038, 3.088, 3.315, 3.565, 5.634]) tgit = np.array([0.013, 0.006, 0.013, 0.016, 0.014, 0.029, 0.047, 0.12, 0.233, 1.283]) normalize = lambda x: x - x[0] logify = lambda x: np.log10(x[3:]) x = logify(n) y = logify(normalize(tguix)) z = logify(normalize(tgit)) X = np.concatenate((x.reshape((len(x), 1)), np.ones((len(x), 1))), axis=1) guix, res, rank, s = np.linalg.lstsq(X, y, rcond=None) git, res, rank, s = np.linalg.lstsq(X, z, rcond=None) predict = lambda sol: sol[0] * x + sol[1] plt.plot(x, y, 'bo', label='guix') plt.plot(x, z, 'ro', label='git') plt.plot(x, predict(guix), 'bx-') plt.plot(x, predict(git), 'rx-') plt.legend(loc='lower right') plt.xlabel("log(N) where N in `| head -n N`") plt.ylabel("log(real time)") plt.show() --=-=-= Content-Type: image/png Content-Disposition: attachment; filename=perf.png Content-Transfer-Encoding: base64 Content-Description: scale.png iVBORw0KGgoAAAANSUhEUgAAA74AAAQACAYAAADGGYWKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz AAAPYQAAD2EBqD+naQAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjIsIGh0 dHA6Ly9tYXRwbG90bGliLm9yZy8li6FKAAAgAElEQVR4nOzdeZzX4/7/8cc0bZYaS0qURMoSB1kK g8PRgkLKesKxnexJSqWkpKJDOLI7dodTWY42Wb85xRHZd8fSMpXCjKWmmj6/P16/JBWpmXnPvD+P ++02t7queU/zGk1un+dc1/W6cjKZTAZJkiRJklKqStIFSJIkSZJUlgy+kiRJkqRUM/hKkiRJklLN 4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRU M/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp 1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJ SjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJ klLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJ kqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJ kiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJ kiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hK kiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+ kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWD ryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN 4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRU M/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp 1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJ SjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJ klLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJ kqRUM/hKkiRJklLN4CtJkiRJSjWDryRJkiQp1Qy+kiRJkqRUM/hKkiRJklLN4CtJkiRJSjWDryRJ kiQp1aomXYDKzrJly5g9eza1atUiJycn6XIkSZKkrJTJZPjuu+/YaqutqFLFtcckGHxTbPbs2TRs 2DDpMiRJkiQBM2bMoEGDBkmXkZUMvilWq1YtIP6B1a5dO+FqJEmSpOxUVFREw4YNf3p9rvJn8E2x 5duba9eubfCVJEmSEubxw+S4wVySJEmSlGoGX0mSJElSqhl8JUmSJEmpZvCVJEmSJKWawVeSJEmS lGoGX0mSJElSqhl8JUmSJEmpZvCVJEmSJKWawVeSJEmSlGoGX0mSJElSqhl8JUmSJEmpZvCVJEmS JKWawVeSJEmSlGoGX0mSJElSqhl8JUmSJEmpZvCVJEmSJKWawbccjRw5ksaNG1OzZk1atGjB5MmT 1/jsCy+8QE5OzipvH3zwQTlWLEmSJEmVn8G3nDzyyCN069aNvn37Mn36dPLz82nXrh1ffvnlr37c hx9+SEFBwU9vO+ywQzlVLEmSJEnpYPAtJ9dddx1nnHEGZ555JjvttBMjRoygYcOG3HLLLb/6cXXr 1mXLLbf86S03N7ecKpYkSZKkdDD4loPFixfz2muv0bp165XmW7duzZQpU371Y/fYYw/q16/PoYce yvPPP1+WZUqSJElSKlVNuoBsMH/+fEpKSqhXr95K8/Xq1WPOnDmr/Zj69etz++2306JFC4qLi7n/ /vs59NBDeeGFFzjwwANX+zHFxcUUFxf/NC4qKiq9L0KSJEmSKimDbznKyclZaZzJZFaZW65Zs2Y0 a9bsp3GrVq2YMWMGw4cPX2PwHTJkCFdeeWXpFSxJkiRJKeBW53JQp04dcnNzV1ndnTdv3iqrwL+m ZcuWfPzxx2t8f+/evSksLPzpbcaMGetcsyRJkiSlhcG3HFSvXp0WLVowadKkleYnTZrEfvvtt9Z/ zvTp06lfv/4a31+jRg1q16690pskSZIkZTu3OpeT7t2706VLF/baay9atWrF7bffzpdffknXrl2B WK2dNWsW9913HwAjRoxg2223ZZdddmHx4sU88MADjB49mtGjRyf5ZUiSJElSpWPwLSfHH388CxYs YODAgRQUFNC8eXPGjRtHo0aNACgoKFjpTt/FixfTo0cPZs2axQYbbMAuu+zC2LFjOfzww5P6EiRJ kiSpUsrJZDKZpItQ2SgqKiIvL4/CwkK3PUuSJEkJ8XV58jzjK0mSJElKNbc6S5IkSUqdkhKYPBkK CqB+fcjPh9zcpKtSUgy+kiRJklJlzBi46CKYOXPFXIMGcMMN0LFjcnUpOW51liRJkpQaY8ZAp04r h16AWbNifsyYZOpSsgy+kiRJklKhpCRWelfXvnf5XLdu8Zyyi8FXkiRJUipMnrzqSu/PZTIwY0Y8 p+xi8JUkSZKUCgUFpfuc0sPgK0mSJCkV6tcv3eeUHgZfSZIkSamQnx/dm3NyVv/+nBxo2DCeU3Yx +EqSJElKhdzcuLIIVg2/y8cjRnifbzYy+EqSJElKjY4dYdQo2HrrlecbNIh57/HNTlWTLkCSJEmS SlPHjnDUUdG9uaAgzvTm57vSm80MvpIkSZJSJzcXDj446SpUUbjVWZIkSZKUagZfSZIkSVKqGXwl SZIkSalm8JUkSZIkpZrBV5IkSVJqDBgAgwat/n2DBsX7lX0MvpIkSZJSIzcX+vdfNfwOGhTzXmmU nbzOSJIkSVJq9OsXv/bvH7926ABPPAFXXAEDB654v7KLwVeSJElSqvTrB19/HeF3eQA29GY3tzpL kiRJSo0FC+Dii+Hmm1fM5eYaerOdwVeSJElSpbdoEVx7LWy/PYwYAUuWxHy1alBSsuaGV8oOBl9J kiRJldayZfDAA9CsGfTsCYWFULduvG/gQFi8OH5dXcMrZQ/P+EqSJEmqlJ59Fi69FKZPj/HWW8M+ +8Bjj618pveXDa/c9px9DL6SJEmSKpW3347V3QkTYlyrFvTuDd26wbBhsMceq4bb5eOSkvKtVRVD TiaTySRdhMpGUVEReXl5FBYWUrt27aTLkSRJktbLrFmxanvPPbHFuWpVOOecCLVbbJF0dWvm6/Lk ueIrSZIkqUIrKoqV3Ouvh4ULY65TJxgyBJo0SbY2VQ4GX0mSJEkV0pIlcNttcOWVMH9+zO2/Pwwf Di1bJlubKheDryRJkqQKJZOBMWPi3O7HH8dc06ax6nvUUZCTk2x9qnwMvpIkSZIqjClTolPzlCkx rlsXBgyAM8+MO3mldWHwlSRJkpS4jz6KFd4xY2K8wQZwySXRvblWrWRrU+Vn8JUkSZKUmHnz4s7d 226DpUuhShU4/fQ417vVVklXp7Qw+EqSJEkqdz/+GF2ahw2D776LucMPj3Hz5snWpvQx+EqSJEkq NyUlcO+9cffu7Nkxt+eecO21cMghydam9DL4SpIkSSpzmQxMmBBndt95J+YaNYKrr4YTTogtzlJZ MfhKkiRJKlOvvx6B99lnY7zJJnD55XDeeVCzZrK1KTsYfCVJkiSViS++iID7wAMxrl4dLrgA+vSB zTZLtjZlF4OvJEmSpFL1zTcwZAjceCMUF8fcSSfBVVdB48bJ1qbsZPCVJEmSVCqKi2HkyAi4X38d cwcfHI2r9tor0dKU5Qy+kiRJktbLsmXw6KOxhfmzz2Ju553hmmviiqKcnGTrkwy+kiRJktbZiy/C pZfCq6/GuH59GDgQTjsNqpo2VEH4rShJkiTpd3vvPbjsMvj3v2O88cbRubl7d9hoo2Rrk37J4CtJ kiRprRUUwBVXwF13xRbn3Fw4++yYq1cv6eqk1TP4SpIkSfpN338Pw4fH2w8/xNzRR8PQodCsWbK1 Sb/F4CtJkiRpjZYujdXdK66AuXNjrmXL6NR8wAHJ1iatLYOvJEmSpFVkMnF+t1cv+OCDmNt++1jh PfZYOzWrcjH4SpIkSVrJf/8bnZr/7/9ivPnm0L8/dO0K1asnW5u0Lgy+kiRJkgD49FPo2xceeSTG NWtCt27RvTkvL9napPVh8JUkSZKy3IIFMGgQjBwJS5bENuZTTom5hg2Trk5afwZfSZIkKUstXAg3 3ghDhkBhYcy1aQPDhsEf/pBsbVJpMvhKkiRJWWbZMnjgAbj8cpgxI+b+8Ifo1HzYYcnWJpUFg68k SZKURZ55JhpXvfFGjBs0gMGD4eSTITc32dqksmLwlSRJkrLAW29Bz54wcWKMa9eGPn3gwgthgw2S rU0qawZfSZIkKcVmzoR+/eDee+Nu3mrV4NxzY5tznTpJVyeVD4OvJEmSlEKFhdGk6vrrYdGimOvc ORpZbb99srVJ5c3gK0mSJKXI4sVw220wcCDMnx9zBxwAw4fDvvsmW5uUFIOvJEmSlAKZDIweDb17 wyefxFyzZrHq26FD3M0rZSuDryRJklTJ/ec/0al56tQY160LV14JZ54JVX3FLxl8JUmSpMrqww9j hfexx2K84YbQo0e81aqVbG1SRWLwlSRJkiqZefNiRfe226CkBKpUgTPOgAEDYKutkq5OqngMvpIk SVIl8cMP0aV52DD4/vuYO/JIGDoUdtkl2dqkiszgK0mSJFVwJSVwzz3Qvz/Mnh1zLVpEp+aDD06y MqlyMPhKkiRJFVQmA+PHQ8+e8O67MbfttnEX73HHxRZnSb/N4CtJkiRVQK+9FoH3uedivOmmcPnl cN55UKNGsrVJlY3BV5IkSapAPv8c+vaFhx6KcfXqcOGF0KdPhF9Jv5/BV5IkSaoAvvkGrr4abrwR Fi+OuZNPhquuiu3NktadwVeSJElKUHEx3HxzBNxvvom5Qw6Ba6+FPfdMtjYpLQy+kiRJUgKWLYNH HoktzJ9/HnPNm8M110DbtpCTk2h5UqoYfCVJkqRy9sILcOmlMG1ajLfaCgYNglNPhdzcREuTUsng K0mSJJWTd9+Fyy6Dp56K8cYbx7hbN9hoo2Rrk9LM4CtJkiSVsYIC6N8f7r47tjhXrQp//WvM1a2b dHUpVVICkyfHf/z69SE/3+X0LGbwlSRJksrId9/B8OHx9uOPMdexIwwZAk2bJltbqo0ZAxddBDNn rphr0ABuuCH+ApR1qiRdgCRJkpQ2S5bALbdAkyYwcGCE3lat4KWXYPRoQ2+ZOv54OPbYlUMvwKxZ MX/88cnUpUS54itJkiSVkkwGnnwSevWCDz+MuSZNYOjQWGi0U3MZKymBCRNW/75MJn6dMCGec9tz VnHFV5IkSSoFr7wCBx0ERx8dobdOHbjpJnjvvVhoNPSWg8mToajo158pKornlFVc8ZUkSZLWw6ef Qu/e8K9/xbhmTejeHXr2hLy8ZGvLOgUFpfucUsPgK0mSJK2D+fPj7t1bbokzvTk5cNppcaa3QYOk q8tS9euX7nNKDYOvJEmS9DssXBjNgYcMWbGrtm1bGDYMdtst2dqyXn4+bL11NLJanZyc+KlEfn75 1qXEecZXkiRJWgslJXDvvdGRuXfvCL277w6TJsH48YbeCqN58zW/L5OBESNsbJWFDL6SJEnSb5g0 CVq0iK3MM2dCw4Zw333w2mvwpz8lXZ0AWLQoriqaODHGNWuu/P7atePXd98t37pUIbjVWZIkSVqD N9+MJlVPPx3jvDzo0wcuvHDVXKUEfftttNN+8UWoUgU6dYKHHoruzQUFcaY3Px+uvjqW7pV1cjKZ 5RdaKW2KiorIy8ujsLCQ2st/wiVJkqTfNGMG9OsXq7qZDFSrBuedB5dfDptvnnR1Wsns2dCuHbz1 VqzqPv44/PGPSVe1El+XJ88VX0mSJOn/KyyEoUPjGOiiRTF3/PGxULjddsnWptX48ENo0wa++AK2 3DIOW+++e9JVqQIy+EqSJCnrLV4Mt94aVxEtWBBzBx4I114L++yTbG1ag1degSOOiL+wHXaIs72N GyddlSoog68kSZKyViYDo0ZFl+ZPP425HXeEa66BI4+M229UAY0fH+d4f/wR9t4bxo6FLbZIuipV YHZ1liRJUlZ66SVo1QqOOy5Cb716ser79tvQvr2ht8K67z7o0CFCb5s28Nxzhl79JoOvJEmSssqH H0YD4Pz82C270UYwYAB88gn89a9Q1T2RFVMmE3vPTz0Vli6FP/8ZnnwSNt446cpUCfjPWpIkSVlh 7ly48kq4/fa40aZKFTjzzAi99esnXZ1+1bJl0KMHXH99jHv0gGHD4i9RWgsGX0mSJKXaDz/AddfF ud3vv4+59u0jN+20U7K1aS0sXgynnQYPPxzj4cPhkksSLUmVj8FXkiRJqbR0KdxzD/TvDwUFMbf3 3rFb9qCDEi1Na+u77+DYY2HSpNiDfs89cPLJSVelSsjgK0mSpFTJZGDcOOjVC959N+YaN4YhQ6Bz Z3fHVhrz5sHhh8Nrr8VB7NGjo5mVtA4MvpIkSUqNadOgZ094/vkYb7YZ9OsH55wDNWokW5t+h08/ jZD76adQp078JGPvvZOuSpWYP+8qRyNHjqRx48bUrFmTFi1aMHny5F99/sUXX6RFixbUrFmT7bbb jltvvbWcKpUkSapcPv8cTjopstHzz0fI7dkzclO3bobeSmX6dNhvv/jLa9wYpkwx9Gq9GXzLySOP PEK3bt3o27cv06dPJz8/n3bt2vHll1+u9vnPPvuMww8/nPz8fKZPn06fPn248MILGT16dDlXLkmS VHF9/XU0+G3WbEXvoy5d4sqiYcNgk02SrU+/07PPxgHsefPgD3+I0LvDDklXpRTIyWQymaSLyAb7 7rsve+65J7fccstPczvttBNHH300Q4YMWeX5Xr168eSTT/L+++//NNe1a1fefPNNpk6dulafs6io iLy8PAoLC6ldu/b6fxGSJEkVxKJFcPPNMHgwfPNNzB16aDSu2mOPZGvTOnrkkfipxZIl8Mc/wmOP QV5e0lWVCl+XJ88V33KwePFiXnvtNVq3br3SfOvWrZkyZcpqP2bq1KmrPN+mTRumTZvGkiVLyqxW SZKkimzZMnjoIdhxx1jp/eYb2HVXGD8+Gv8aeiupG2+EE0+M0Nu5c/yFpiT0qmIw+JaD+fPnU1JS Qr169Vaar1evHnPmzFntx8yZM2e1zy9dupT58+ev9mOKi4spKipa6U2SJCktnn8e9tknbrP54gvY aiu4++44Etq2LeTkJF2hfrdMBnr3hosuit+ff37sWfdQtkqZwbcc5fzi/8aZTGaVud96fnXzyw0Z MoS8vLyf3ho2bLieFUuSJCXv3XfhiCPgkEPiZptatWKL88cfw1/+Arm5SVeodbJkCZx+OgwdGuPB g2Pl179QlQGDbzmoU6cOubm5q6zuzps3b5VV3eW23HLL1T5ftWpVNt9889V+TO/evSksLPzpbcaM GaXzBUiSJCVg9mw480zYbbe4zaZq1VgQ/PRT6NMHNtww6Qq1zn78EY45Bu65J4LuXXfFX6rL9ioj Bt9yUL16dVq0aMGkSZNWmp80aRL77bffaj+mVatWqzz/9NNPs9dee1GtWrXVfkyNGjWoXbv2Sm+S JEmVzXffxd27TZpEHlq2DI49Ft57D266CbbYIukKtV4WLIhOZGPHQs2a0cTq9NOTrkopZ/AtJ927 d+fOO+/k7rvv5v333+fiiy/myy+/pGvXrkCs1p5yyik/Pd+1a1e++OILunfvzvvvv8/dd9/NXXfd RY8ePZL6EiRJksrUkiUwciRsvz1cdRUsXBjXuU6ZAqNGeatNKnz5JRxwALz8Mmy6aVxf1L590lUp C1RNuoBscfzxx7NgwQIGDhxIQUEBzZs3Z9y4cTRq1AiAgoKCle70bdy4MePGjePiiy/m5ptvZqut tuLGG2/k2GOPTepLkCRJKhOZDDzxBPTqBR99FHM77BBHP485xt2vqfH229GFbPZsaNAAJk6EnXdO uiplCe/xTTHvC5MkSRXdyy/DpZfCSy/FeIst4Ior4OyzYQ2nu1QZTZ4cK7uFhbDLLjBhQoTfLOHr 8uS54itJkqRy98kncYvNqFEx3mAD6N4devYEc0HKPP44nHACFBfD/vvDv/8d25ylcuQZX0mSJJWb +fPhwgthp50i9FapAmecEVcTXXWVoTd1brstOpMVF0OHDjBpkqFXiTD4SpIkqcwtXAhDhkTjqptu gqVLoV07eOMNuPNO2HrrpCtUqcpk4MoroWvXaMt91lkwenQs7UsJcKuzJEmSykxJCdx/f1xPNHNm zO2xB1x7bdxooxQqKYHzzovVXoi//CuvtEuZEmXwlSRJUpmYODHO7L71Voy32QauvhpOPDG2OCuF Fi2Ck06Ku3lzcuDmm+Gcc5KuSjL4SpIkqXS98UYE3kmTYrzJJtC3L5x/PtSsmWxtKkPffhvneCdP hurV4aGH4nyvVAEYfCVJklQqvvwydrXef38c8axWLcJu376w+eZJV6cyNWtWHNp+++3oUPbkk3DQ QUlXJf3E4CtJkqS1MmAA5OZGuP25b7+NK1qnTo3jnRDbmQcPhsaNy71MlbcPPoA2beInH/Xrw/jx 8Ic/JF2VtBJPV0iSJGmt5OZC//4waFCMFy+GG26ArbaCl16K0HvQQfDf/8YuV0NvFnj5ZTjggAi9 TZvClCmGXlVIrvhKkiRprSxf6e3fH955B6ZNg//9L+bq1IF//AOOOMLmvVlj3Djo1CnuqtpnH3jq Kdhii6SrklbL4CtJkqS1dvDBcefuo4+umOvQIa5oreory+xx771wxhmxzN+2LYwaBRttlHRV0hq5 1VmSJEm/6YMP4Kij4MADo4/RctWrwxNPGHqzRiYDw4bBaadF6O3SJRpZGXpVwRl8JUmStEZz5sQ1 rM2bR77JzYW99473Va8e53yXn/lVyi1bBhdfDJddFuNLL4V77on23VIFZ/CVJEnSKr7/Hq68Epo0 gVtvjcW9o46Cc8+FV1+FgQOhuDh+/XnDK6VUcTGcfHJ0MwO47jq45hqoYpxQ5eCmFEmSJP1k6dJo UtW/f6z2QvQtuvZaePHFmB84cEWjq583vPr5WCny3XfQsSM880ys7t5zD5x0UtJVSb+LwVeSJElk MjB2LPTqBe+9F3PbbQdDhkDnztGp+bnnVg69yy0fL7/DVykydy4cfji8/nqc4x0zBlq3Troq6XfL yWQymaSLUNkoKioiLy+PwsJCateunXQ5kiSpgpo2DXr0iBVdgM02ixXcc86Jc7zKUp9+Cm3axK9b bBHXF+21V9JVVUq+Lk+eK76SJElZ6rPPoE8f+Oc/Y1yjBnTrFr2LNtkk2dqUsNdfh3btYN48aNwY Jk6EHXZIuippnRl8JUmSsszXX8NVV8HNN0dX5pycuJVm0CDYZpukq1PinnkGjjkmOpztvjuMHw9b bpl0VdJ6MfhKkiRliUWL4Kab4Oqr4dtvY+6ww6I57+67J1ubKoh//hNOOQWWLIFDDoHHHgO35ioF 7D8uSZKUcsuWwQMPQLNm0LNnhN5dd4UJE+Dppw29+v9uuAFOPDFC73HHxZleQ69SwuArSZKUYs8+ G/2IunSBL7+ErbeO64qmT4++RRKZTBzs7tYtxhdcAA8/HIe+pZRwq7MkSVIKvfNOrO6OHx/jWrWg d+/INhtssH5/dkkJTJ4MBQVQvz7k50Nu7vrXrAQsWQJnnQX33hvjq6+OEJyTk2xdUikz+EqSJKXI rFlxFdE998QW56pV41qifv3iRpr1NWYMXHQRzJy5Yq5Bg9gl27Hj+v/5Kkc//LBiS3NuLtxxB/zl L0lXJZUJg68kSVIKFBVFk6rrroOFC2OuUycYMgSaNCmdzzFmTPyZmczK87NmxfyoUYbfSmP+fDjy SHjlldgC8OijMZZSyjO+kiRJldiSJXEtUZMmMHhwhN7994epU+Ff/yq90FtSEiu9vwy9sGKuW7d4 ThXcF1/AAQdE6N1sszgIbuhVyhl8JUmSKqFMJlZgd9kFzj8fvvoKmjaNucmToWXL0v18kyevvL15 dfXMmBHPqQJ7+23Ybz/48ENo2BBeeglatUq6KqnMGXwlSZIqmSlTYsHu2GPh44+hbl0YOTIaWh1z TNn0JSooKN3nlID/+7/oRDZ7dvzEZMoU2GmnpKuSyoVnfCVJkiqJjz+OzsyjR8d4gw3gkkuie3Ot WmX7uevXL93nVM7GjIGTToLi4vipyZNPwqabJl2VVG5c8ZUkSargvvoqrlbdeecIvVWqwBlnwCef wKBBZR96IRYKGzRY82pyTk7snM3PL/ta9Dvdeit07hyh9+ij4emnDb3KOgZfSZKkCurHH+Na1e23 h7//HZYuhcMPhzffhDvvhK22Kr9acnPjyiJYNfwuH48Y4X2+FUomAwMGxH1Wy5bB2WdHx7P1vchZ qoQMvpIkSRVMSQncfTfssAP07QvffQd77hnNd8eOhebNk6mrY8e4smjrrVeeb9DAq4wqnJIS6NoV rrwyxldcESu/VT3pqOzkd74kSVIFkcnAxIlxZvftt2OuUaNY9T3hhNjinLSOHeGoo6J7c0FBnOnN z3elt0JZuDDO8z7+eCzHj+eFSxMAACAASURBVBwZIVjKYgZfSZKkCmD6dLj00ljVBdhkE7j8cjjv PKhZM9nafik3Fw4+OOkqtFrffLPiJxM1asBDD7kUL2HwlSRJStQXX0TAfeCBGFevHo2s+vSBzTZL tjZVMrNmQdu2ca9VXh488QQcdFDSVUkVgsFXkiQpAd9+G1uYb7wxmu0CnHgiDB4MjRsnW5sqofff hzZtYMaM2H8+cSLsumvSVUkVhsFXkiSpHBUXwy23xDVEX38dcwcfDNdeC3vtlWhpqqymToUjj4xv qGbNIvQ2apR0VVKFYvCVJEkqB5kMPPJIbGH+7LOY23lnuOaauKJoTffjSr9q7Ni4o3fhQth3X3jq KahTJ+mqpAqnAvQGlCRJSrcXX4xMcuKJEXrr14c77oj7eI84wtCrdfSPf0Qjq4ULoV276Ixm6JVW y+ArSZJURt57Dzp0iK3Mr74KG28MAwfCxx/DmWd6parWUSYDQ4bA6afHfb2nnhqNrDbaKOnKpArL /91KkiSVsjlz4Ior4M47YdmyuP7n7LNjrl69pKtTpbZsGVx8cXRFA+jVK0Kw2wakX2XwlSRJKiXf fw/Dh8fbDz/E3FFHwdChsOOOydamFCguhlNOgUcfjfH110O3bsnWJFUSBl9JkqT1tHQp3HVXrOjO nRtz++4bnZrz85OtTSlRVATHHAPPPQfVqsF998EJJyRdlVRpGHwlSZJ+w4ABsV25X7+V5zMZ+POf 4/aYBQtibvvtY+dpp07uPlUpmTMnWn9Pnx4HxR97DP70p6SrkioVg68kSdJvyM2F/v3j98vD73// C8cdB198EePNN49nunaF6tWTqVMp9Mkn0Lp1tAOvWxfGjYMWLZKuSqp0DL6SJEm/YXnY7d8fvv4a CgriTl6Izsw9ekSPoU02Sa5GpdC0abHS+9VXsN128PTTsaVA0u9m8JUkSVoL554L48fDiBEr5nbf HZ58Eho2TK4updSkSXGm94cfYM89Y6XXluDSOvMeX0mSpF+xcCFcc00stE2dumK+WrU4cmnoVal7 +GE44ogIvYceCi+8YOiV1pPBV5IkaTWWLYP774dmzWIbc2HhiuxRvTosWQKDBiVbo1Lo+uvhpJPi G+yEE2DsWKhVK+mqpErP4CtJkvQLzzwT/YNOOQVmzIAGDWLX6dy5MHBgXKc6cGCc+TX8qlRkMvET lu7dY3zRRfDgg1CjRrJ1SSnhGV9JkqT/7623oGfPuJ4IoHZt6N0bfvwxAu7AgSsaXf284dXPx9Lv tmQJnHlm3M0LMHRofCN6H5ZUagy+kiQp682cGQH2nnti4a1q1Whm1a8f1KkT9/j+PPQut3xcUlLe FSs1fvgBOneOzmm5uXDnnXDaaUlXJaVOTiaTySRdhMpGUVEReXl5FBYWUrt27aTLkSSpwiksjMZV 118fTawgMsjVV0OTJsnWpiwwf340sfrvf2GDDeBf/4qxUsfX5clzxVeSJGWdxYvh9tvhyisjewAc cABcey20bJlsbcoSn38ObdrARx/BZptFEyu/+aQyY/CVJElZI5OBMWPgssvgk09irlkzGDYMOnTw SKXKyVtvQdu2UFAA22wTh8p33DHpqqRUs6uzJEnKCv/5D+y/P3TqFKG3bl245RZ4+2046ihDr8rJ iy9Cfn6E3ubNYcoUQ69UDlzxlSRJqfbRR7HC+9hjMd5wQ+jRI968HlXlavRoOPnkuA8rPx+efBI2 2STpqqSs4IqvJElKpXnz4LzzYOedI/RWqRI3xnz8cZztNfSqXN1yS3ROKy6OS6EnTjT0SuXI4CtJ klLlxx/hqqtg++1h5Mi4aujII+NY5R13wFZbJV2hskomE3dlnXtu/P6vf43uzRtskHRlUlZxq7Mk SUqFkpK4h7d/f5g9O+ZatIhOzX/8Y6KlKVstXRqB9447YjxgQHyDeqBcKncGX0mSVKllMjB+PPTq Be+8E3Pbbht38R5/fGxxlsrdwoVw4onwxBPxTThyZKz2SkqEwVeSJFVar78Ol14Kzz0X4003hcsv j7O9NWokW5uy2DffxP1YL70U34gPPxzneiUlxuArSZIqnS++gL594cEHY1y9Olx4IfTpE+G3rJSU wOTJcRNN/frRmDc3t+w+nyqhmTPjjt5334W8vOjcfOCBSVclZT2DryRJqjS++Sa2MN94IyxeHHMn nxzNrLbdtmw/95gxcNFFkWuWa9AAbrgBOnYs28+tSuL996FNG5gxI7qoTZgAu+6adFWSsKuzJEmq BIqL4brrolPz8OERev/4R5g2DR54oHxCb6dOK4degFmzYn7MmLL9/KoEpkyB/feP0NusWYwNvVKF YfCVJEkV1rJlcTxyxx3hkktixXeXXWDsWHj22ejaXNZKSmKlN5NZ9X3L57p1i+eUpf79b/jTn+Ib tGVL+M9/oFGjpKuS9DMGX0mSVCG98ALsuy+cdBJ8/nmcqb3zTnjjDTj88PK7EWby5FVXen8uk4lF vsmTy6ceVTB33x2NqxYuhCOOgGeegc03T7oqSb9g8JUkSRXKe+9B+/YrtjJvvDEMGgQffwxnnAFV y7lDSUFB6T6nlMhk4sD5GWfEcv9pp8Fjj8FGGyVdmaTVsLmVJEmqEAoK4Ior4K67Yotzbm5ce9q/ P9Srl1xd9euX7nNKgZKS2N/+97/HuHdvGDy4/LYhSPrdDL6SJClR330XDauGD4cff4y5Y46BIUOi R1DS8vOje/OsWas/55uTE+/Pzy//2pSA4mLo0gX+9a/4yx8xIu7SklShudVZkiQlYulSuPVW2GEH GDgwQm/LlnFWdsyYihF6IVaeb7ghfv/LBb3l4xEjvM83KxQWQrt2EXqrVYvOa4ZeqVIw+EqSpHKV ycATT0Dz5nDOOTB3LjRpElliyhQ44ICkK1xVx44wahRsvfXK8w0axLz3+GaBOXPg4IPh+eehVi0Y Px6OPz7pqiStJbc6S5KkcvPKK3DppSs6INepE2d4//pXqF492dp+S8eOcNRRUXtBQZzpzc93pTcr fPwxtGkDn30GdetG6N1zz6SrkvQ7GHwlSVKZ+/RT6NMHHn00xjVrwsUXQ69ekJeXbG2/R25uLPop i0ybFvdnffUVbL89TJwYv0qqVAy+kiSpzMyfD1ddBSNHwpIlcSb21FPjeqIGDZKuTvoNTz8dS/0/ /BArvOPGJdtiXNI6M/hKkqRSt3Ah3HhjXHNaVBRzbdrANdfAbrslW5u0Vh58MO7mXboU/vSn6LhW q1bSVUlaRza3kiRJpWbZMrjvvujIfNllEXp33z0WziZMMPSqkrjuOvjznyP0nngijB1r6JUqOYOv JEkqFZMmxW7QU0+FGTOgYcMIwa+9BocdlnR10lpYtiy6r11ySYy7dYMHHqj4ndck/Sa3OkuSpPXy 5pvQs2es6kI0q+rTBy64ADbYINnapLW2ZAmcfnoEXYh9+T16rHp5s6RKyeArSZLWycyZcPnlsaqb yUC1anDeedC3b1xTJFUa338PnTpFx+bcXLj7bjjllKSrklSKDL6SJOl3KSyEYcPg+uth0aKYO/54 GDzYW15UCX31FRxxBLz6Kmy4IYwaBe3aJV2VpFJm8JUkSWtl8WK49VYYOBAWLIi5/Hy49lrYd99k a5PWyWefRbvxjz+GzTePJlZ+M0upZPCVJEm/KpOB0aOhd2/45JOY23HHWPVt394jkKqk3nwT2raF OXOgUaPY5tysWdJVSSojdnWWJElr9NJLsN9+0LlzhN569WLV9+23oUMHQ68qqRdegAMPjNC7664w ZYqhV0o5V3wlSdIqPvww7uF9/PEYb7jhiltevM5UldqoUXDyybF3/8AD4YknYJNNkq5KUhlzxVeS JP1k7lw491zYZZcIvVWqwNlnx2rvgAGGXlVyI0fCccdF6O3YMbY3G3qlrOCKryRJ4ocf4Lrr4urS 77+PufbtYehQ2HnnZGuT1lsmA/37w1VXxficc+Cmm+LqIklZweArSVIWKymBf/wjMkFBQczttVd0 aj744ERLk0rH0qURdO+8M8YDB8YF1B5Ql7KKwVeSpCyUycC4cdCrF7z7bsw1bgxXXx07Qat4GEpp 8OOPcOKJ8OST8U19661w1llJVyUpAQZfSZKyzGuvRaOq55+P8aabQr9+cba3Ro1ka5NKzddfx379 KVOgZk14+GE4+uikq5KUEIOvJElZ4vPPoW9feOihGNeoARdeGPfzbrppoqVJpWvGjLij9733onnV v/8NBxyQdFWSEmTwlSQp5b7+OrYw33RTNLMF+POfo89Po0bJ1iaVuvfegzZtYOZM2HprmDABmjdP uipJCTP4SpKUUsXF8Pe/w+DB8M03MXfIIdG4as89k61NKhNTpsCRR8Y3/I47xnVF22yTdFWSKgBb V5SDb775hi5dupCXl0deXh5dunTh22+//dWPOe2008jJyVnprWXLluVUsSSpMlu2LLYz77gj9OgR GaB582hm9cwzhl6l1JNPwqGHxjd8y5bw0kuGXkk/MfiWg5NOOok33niDCRMmMGHCBN544w26dOny mx/Xtm1bCgoKfnobN25cOVQrSarMnn8e9tkHTj45zvRutRXcdRe88Qa0a+cNLkqpu+6CY46BRYti xffZZ2HzzZOuSlIF4lbnMvb+++8zYcIEXn75Zfbdd18A7rjjDlq1asWHH35Is2bN1vixNWrUYMst tyyvUiVJldi778bVRGPHxrhWLbjsMujWDTbcMNnapDKTycQB9ssvj/Ff/gK33w5VfYkraWWu+Jax qVOnkpeX91PoBWjZsiV5eXlMmTLlVz/2hRdeoG7dujRt2pSzzjqLefPmlXW5kqRKZvbsuJZ0t90i 9FatCuedB598An36GHqVYiUlcMEFK0Jvnz6x8mvolbQa/p+hjM2ZM4e6deuuMl+3bl3mzJmzxo9r 164dnTt3plGjRnz22Wf069ePQw45hNdee40aa7hksbi4mOLi4p/GRUVF6/8FSJIqpO++iyZVf/sb /PhjzHXsCEOGQNOmydYmlblFi6BLFxg1Kvbv33BDhGBJWgNXfNfRgAEDVmk+9cu3adOmAZCzmgNV mUxmtfPLHX/88RxxxBE0b96c9u3bM378eD766CPGLt/DthpDhgz5qYFWXl4eDRs2XP8vVJJUoSxZ ArfcAk2awKBBEXpbtYL//AdGjzb0KgsUFsaB9VGjoHp1+Oc/Db2SfpMrvuvo/PPP54QTTvjVZ7bd dlveeust5s6du8r7vvrqK+rVq7fWn69+/fo0atSIjz/+eI3P9O7dm+7du/80LioqMvxKUkpkMvDE E3Fu98MPY26HHWDo0OjpY9MqZYWCggi9b74ZB9kffzzu6JKk32DwXUd16tShTp06v/lcq1atKCws 5L///S/77LMPAK+88gqFhYXst99+a/35FixYwIwZM6hfv/4an6lRo8Yat0FLkiqvl1+GSy+N21kA 6tSBAQPg7LOhWrVES5PKz0cfQZs20a68Xj0YPx722CPpqiRVEm51LmM77bQTbdu25ayzzuLll1/m 5Zdf5qyzzuLII49cqaPzjjvuyGOPPQbA999/T48ePZg6dSqff/45L7zwAu3bt6dOnTocc8wxSX0p kqRy9skn0LlzbGV+6SXYYAPo2xc+/TQaWBl6lTVefRX23z9Cb5MmMGWKoVfS72LwLQcPPvggu+66 K61bt6Z169bstttu3H///Ss98+GHH1JYWAhAbm4ub7/9NkcddRRNmzbl1FNPpWnTpkydOpVatWol 8SVIksrR/Plw0UWw884revecfnoseF11FdSunXSFUjmaOBH++Mf4h9GiRRxo3267pKuSVMnkZDKZ TNJFqGwUFRWRl5dHYWEhtX2VJEkV3sKFMGJEnNtd3pi/XTsYNgx23TXZ2qREPPBA3M27dCkcdlh0 cHMRQJWQr8uT54qvJEkJKymBe++Njsx9+kTo3WMPeOYZGDfO0KssNXx4XFm0dCmcdBI89ZShV9I6 M/hKkpSgp5+O3ZunnQYzZ8I228D998O0aXDooUlXJyVg2TK45JLo6AbQvXv8o6hePdm6JFVqdnWW JCkBb74Zr+snTYpxXl40rrrgAqhZM9napMQsXhwH2h98MMbXXgs9eiRbk6RUMPhKklSOZsyAyy+P BaxMJjozn39+hN7NN0+6OilB338Pxx4b2yCqVoW7746tzpJUCgy+kiSVg8JCGDIEbrgBFi2KuRNO gMGDbVArMW8eHHFE7PHfcMNoYtW2bdJVSUoRg68kSWVo8WK45RYYNAgWLIi5Aw+MHZz77JNsbVKF 8L//QZs2cXF1nTowdqz/OCSVOoOvJEllIJOJO3h794ZPP425nXaKq4mOPDLu5pWy3htvxMru3Lmw 7bZxZ2/TpklXJSmF7OosSVIpmzwZWrWC446L0LvllnDbbfDWW9C+vaFXAuC552L7w9y5sNtuMGWK oVdSmXHFV5KkUvLBB3DZZfDEEzHeaKPo3HzJJbDxxsnWJlUojz4ajasWL4aDD4bHH4/W5pJURlzx lSRpPc2dC+ecA82bR+jNzYW//jWOLF5xhaFXWsnf/x6d3RYvhk6dYPx4Q6+kMueKryRJ6+iHH+Bv f4tGVd9/H3MdOsDQoXGeV9LPZDJxl9fVV8f43HPhxhvjJ0WSVMYMvpIk/U5Ll8I//gH9+8OcOTG3 zz4RgA88MNnapApp6dLYBnH33TEeNCgur/bAu6RyYvCVJGktZTJx00qvXvDeezG33XZxP2/nzr6G l1brxx/h+OPhqaegSpXo9HbmmUlXJSnLGHwlSVoL06ZFo6oXXojxZpvFim/XrlCjRqKlSRXXggXR ynzqVKhZEx55JM4DSFI5M/hKkvQrPvssdmQ+/HCMa9SAiy6K+3k32STZ2qQKbcYMaNMG3n8fNt0U /v1v2H//pKuSlKUMvpIkrcbXX8PgwdGAdvHi2MbcpUscTdxmm6Srkyq4d9+N0DtrFjRoABMmwC67 JF2VpCxm8JUk6WcWLYqwO3gwfPttzP3pT3DNNbDHHsnWJlUKL70U25u//Tbam0+cCA0bJl2VpCxn 8JUkCVi2LLYz9+0LX3wRc7vuGp2aW7e2cZW0Vp58MhpZLVoE++0X25s32yzpqiSJKkkXIElS0p57 DvbeG/785wi9W28d1xVNnx67NQ290lq480445pgIve3bw6RJhl5JFYbBV5KUtd55Bw4/HA49FF5/ HWrVgquvho8+gtNOg9zcpCuUKpgBA+Kg+89lMjF31lmxdeKMM2DMGNhww0RKlKTVcauzJCnrzJoF V1wRq7rLlkHVqnDOOdCvH2yxRdLVSRVYbm7c4wXxD6akBC68EEaOjLmDDoI77nCbhKQKx+ArScoa RUVxZvdvf4OFC2OuU6dY5d1hh2RrkyqFfv3i1/79YenS6N48enTMHXEEPPVUcrVJ0q8w+EqSUm/J kliEGjAAvvoq5vbbD4YPh1atEi1Nqnz69YufHA0cuGKuc2d49NHkapKk32DwlSRVegMGxA7MPn1g 8mQoKID69eGAA+Lu3aefjnt5AZo2haFD4eij3Y0prZM774yfGv3c1Klxrrdjx2RqkqTfYPCVJFV6 y48dDh8e25l/Pl9SEr/fYosIyGedBdWqJVKmVPnddFOc6f2lmTPj3MCoUYZfSRWSXZ0lSZXeLrvE rz8PvbAi9HbqBJ98Aueea+iV1tnUqdCt25rfn8nE+5f/w5OkCsTgK0mq1EpK4KKLfv2ZV16BjTYq n3qkVBo/Hg45JNqg/5oZM+K8gSRVMAZfSVKlNmlS7LL8Nb4Wl9bDffdBhw6waNHaPV9QULb1SNI6 MPhKkiqlkhK45x44+eS1e97X4tLvlMnE/V+nnhpXFx122Np9XP36ZVuXJK0Dg68kqVLJZGDiRNhz T/jLX1Z0a/4tvhaXfodly+CSS6BnzxhfcgmMHQsNGqy5HXpODjRsCPn55VenJK0lg68kqdJ44w1o 3RratoW33oJNNoFhw6BWrV//uNq1fS0urbXFi+MesOuvj/Hw4fFWrRrccEPM/TL8Lh+PGBHt1CWp gjH4SpIqvC+/hFNOiVXeZ56B6tWhe3f49NNYkGrXLp5b02vxtm19LS6tle++gyOPhIcegqpV4f77 Y7V3uY4d48qirbde+eMaNPAqI0kVWk4mk8kkXYTKRlFREXl5eRQWFlK7du2ky5Gk3+3bb2HIkFhk Ki6OuRNPhMGDoXHjlZ8dMya6O/+80VXDhrEA5WtxaS3MmweHHw6vvRZt0EePhjZtVv9sSUl0jCso iHME+fn+dEn6Fb4uT57BN8X8Byapslq8GEaOhEGDVpzhPfjg6LOz115r/jhfi0vr6H//i5D7ySdQ pw6MGwd77510VVJq+Lo8eVWTLkCSpOUyGXj0UejTJ16HA+y8M1xzTSxEramnznK5uRGQJf0O06fH eYG5c2HbbaN7XNOmSVclSaXKM76SpArh//4P9t0XTjghQm/9+nDHHfDmm3DEEb8deiWtg+eeg4MO itD7hz/AlCmGXkmp5IqvJClR778Pl10GTz4Z4402ioZVl1wSv5dURh55JLo3L1kSWyUefxzy8pKu SpLKhCu+kqREzJkDXbvCrrtG6M3NhXPOiU7N/fsbeqUyddNN0SluyRLo3BkmTDD0Sko1V3wlSeXq ++/hb3+LRlU//BBzRx0FQ4fCjjsmW5uUepkM9O0b7dIBzjsv2qbbBU5Syhl8JUnlYulSuPtuuOKK WO2FONN77bXRfVlSGVu6FM4+G/7xjxgPHgy9e3uAXlJWMPhKkspUJgNPPQW9esV5XoDtt48Fp06d fM0tlYsff4TjjoOxY6FKFbj9djjjjKSrkqRyY/CVJJWZV1+FSy+FF1+M8eabx/ndrl2hevVka5Oy xoIFcOSR8PLLULNm3BnWvn3SVUlSuTL4SpJK3f/+F8cI//nPGNeoAd26RffmTTZJtjYpq3z5JbRp Ax98AJtuGtsv9tsv6aokqdwZfCVJpWbBArjqKrj55mgWm5MDp5wCAwfCNtskXZ2UZd55J0Lv7NnQ oAFMnAg775x0VZKUCIOvJGm9LVoUt6MMHgyFhTF32GFwzTWw++7J1iZlpcmToUMH+PbbCLsTJ0b4 laQsZfCVJK2zZcvgwQfh8stjRyXAbrtFp+bWrZOtTcpajz8OJ5wAxcWw//5xUfZmmyVdlSQlqkrS BUiSKqdnn4W99oqtzF9+GYtJ99wDr79u6JUSc/vtcOyxEXo7dIBJkwy9koQrvpKUCiUlsbOxoADq 1497cXNzy+Zzvf029OwJEybEuHbtuAr0ootggw3K5nNK+g2ZDAwaFBdlA5x5JtxyC1T1pZ4kgcFX kiq9MWMidM6cuWKuQQO44Qbo2LH0Ps+sWdCvX6zqZjLxevrcc2Ob8xZblN7nkfQ7lZTA+efDrbfG uF8/uPJKL8mWpJ8x+EpSJTZmDHTqFEH052bNivlRo9Y//BYVwbBhcP31sHBhzHXuDFdfDU2arN+f LWk9LVoEJ58c/zPIyYG//z1+IiVJWolnfCWpkiopiZXeX4ZeWDHXrVs8ty6WLInX0NtvHyF34cLo kzN1Kjz6qKFXSty338Z1RWPGQPXq8Q/T0CtJq2XwlaRKavLklbc3/1ImAzNmxHO/RyYDo0fDLrvA BRfA/PnQtCk89lj8WS1brl/dkkrB7Nlw4P9j776jsyzSPo5/Q6QqRaQLShFl7eiuDRFQkSooYoEt Yu+KDUUpAUREXAT1tfe1F1BEaSpNQMWCCCqigiBFlJJQE0ie949Z2iJISXI/5fs5h5PM5E5yec4m +/wyM9ecAhMmhIP2o0aFbR6SpD/kVmdJSlCLFuXvcwCTJ8Mtt4RVXYBKlSAjI/TJKVp0l0uUVBC+ +y6s9M6bB1WqhE5zRx0VdVWSFNcMvpKUoKpWzb/nvv8+dGYeMiSMS5WCm2+GW2+F0qV3v0ZJ+eyT T6BVK1i6NGzFGDUKataMuipJintudZakBNWwYejevL3GrWlpUKNGeG57liwJzWAPOyyE3iJFwuru 7NnQu7ehV4or770Hp54aQu9xx8FHHxl6JWknGXwlKUGlp4cri2Db8LtxPGjQH9/nu2YN9O0bGlT9 3//Bhg1hEemrr+CJJ6BatYKtXdIueu45aNMm/PA2bw4ffOA9YpK0Cwy+kpTA2rULVxbtv//W89Wr //FVRrm58PTTULduuH935Uo45hj48EMYPhwOP7zwape0E2KxcJ9Yp07hB/if/4Rhw2CffaKuTJIS imd8JSnBtWsHbduGjsuLFoUzvQ0bbr3SG4uF/jddusCMGWHuwAPDNUUXXBC2OEuKM3l5cNNNm7d2 3Hor3HOPP7CStBsMvpKUBNLToXHjP/7YF1+E18sffhjG5cqF1d5rroESJQqtREm7Ijs7rPK+8koY DxwIN94YaUmSlMgMvpKUwDIyQujt3n3bj918M4wZA19/HcbFioV7ee+4A8qXL9QyJe2KlSvDVo73 3w/3iD37LHTsGHVVkpTQDL6SlMDS06FHj/D+xvC7fDmceSZMmrT5uY4d4a67oFatwq9R0i749Vdo 2TJs1dh779Bu/Ywzoq5KkhKewVeSEtjGsNujR+jMXK5cWNFdty7MN24MAwbAX/8aWYmSdtaPP0Kz ZuFtxYrh+iJ/eCUpXxh8JSnB3Xln2M7cu/fmuYoV4ZlnwsLR9u75lRRHvvgCWrQIl2vXqgWjRoX2 65KkfGHwlaQENn483HILfPbZ5rm99oKFC8NbSQng/ffh7LNh1So4+mgYMQKqVIm6KklKKvbDl6QE 9M034Rxv48Yh9BYrFuaLFQtbnvv1i7Q8STvrlVfC1oxVq+DUU8Nfswy9kpTvDL6SlEAWLYLLL4cj joDhw0Nzq7/9DXJywlbn7OzwtkcP6NMn6mol7dDgwdChA6xfD+edF870likTdVWSlJTcCCdJCWDV qtCk6r77YM2aMHfWcy9UgQAAIABJREFUWXDAAfDAAyHsbmx0tWXDqy3HkuJELBa60N1zTxhfdx0M GgRFXI+QpIJi8JWkOLZhAzz5ZLiv99dfw9wJJ4QQfPLJYX7L0LvRxnFubmFWK+lPrV8Pl10Gzz0X xnffDbffbhc6SSpgabFYLBZ1ESoYWVlZlC1blszMTMq4dUpKKLEYDBsWXg9/912Yq1MnLBCdc46v kaWEtHr15i3N6enwxBNw0UVRVyWpEPi6PHqu+EpSnPnkE7j1Vpg4MYz32w969oQrrtjcxEpSgvn9 d2jdOvyAlywJr70WxpKkQmHwlaQ48eOP4djfa6+FcYkScOONcNttULZstLVJ2gM//wzNmsGsWVC+ fOhMd+KJUVclSSnF4CtJEfv9d7jrLnj44XD8Ly0N/vWv0JW5Ro2oq5O0R77+Gpo3D5dr16gBo0bB X/4SdVWSlHIMvpIUkbVrQ0fmfv0gMzPMNWsG/fvDUUdFW5ukfDBhArRpE37ADzsMRo6E6tWjrkqS UpLBV5IKWV4evPACdOsG8+eHuaOOCp2amzaNtjZJ+WTo0HBHb3Z2aME+bBjsu2/UVUlSyvLCOEkq RGPGwDHHwIUXhtBbvXq41eTzzw29UtJ49FFo3z6E3rZtYfRoQ68kRczgK0mFYPr0cMzvjDPgq6+g TJlwNdH334fzvOnpUVcoaY/FYuFy7auuCls7Lr8c3ngjdHGWJEXKrc5SAcvNDdfSLFoEVatCw4aG nFTyyy/QvXtY1Y3FoGhRuPrqsM25QoWoq5OUb3Jzww/344+HcY8eIQR76bYkxQWDr1SAhgyBG24I 4Wej6tVh8GBo1y66ulTwMjNDk6r774d168LceefB3XdDnTrR1iYpn61dCx07wltvhaD78MNw5ZVR VyVJ2oLBVyogQ4aEI16x2NbzCxaE+TfeMPwmo5wceOwx6N07XFMEoa/NfffB8cdHW5ukArBiRejc PHEiFC8OL73kL3dJikOe8ZUKQG5uWOn939ALm+c6dw7PKTnEYuGPGYcdBtdfH0LvIYeEBaAJEwy9 UsLLyAiXa29pwYJwfmXiRChWLNzRa+iVpLhk8JUKwMSJW29v/l+xWOjoO3Fi4dWkgjNpEpx0Epx7 LvzwA1SqBI88AjNmhIauHvGTkkB6eji3uzH8fvdd+MGfMSOML70UGjWKrj5J0g651VkqAIsW5e9z ik+zZkHXruG6ToBSpeCWW8K/0qWjrU1SPuvePbzt0SP85fLNN2HZsjB3440wcGB0tUmS/pTBVyoA Vavm73OKL7/+Cr16heatublQpAhccknYCVmtWtTVSSow3buHv3g98cTmudtvh379oqtJkrRT0mKx PzqFqGSQlZVF2bJlyczMpEyZMlGXk1Jyc6FmzXD8649+wtLSQnfnOXO82iiRrF4dujT37w+rVoW5 1q3DfbyHHRZtbZIKwTPPwGWXbW7QUKwYZGdHW5OkhODr8uh5xlcqAOnp4coi2PZ858bxoEGG3kSR mwtPPQUHHxwWfFatgmOPhbFj4Z13DL1S0ovFwl+4Lr5469Cbk7NtwytJUlwy+EoFpF270OV3//23 nq9e3auMEkUsBu+9B0cdFfrWLFwYVvJffhk+/RQaN466QkkFLi8vtOHv2nXzXK9eYaW3d++tG15J kuKWZ3ylAtSuXejqO3FiaGRVtWq4+cKV3vj3+edw661hVRdg332hWze45ppwVaekFJCdDRdeCK++ unmud+/Nja62bHi15ViSFHcMvlIBS093ZTCRzJ0Ld94JL70UxsWKhXt577gjhF9JKSIrK/z18oMP oGhROPNMOPro8Mtg3LjNf828447wvBezS1JcM/hKErB8OfTtCw8+GI7tAfz973DXXWF7s6QUsngx tGwJX34J++wDQ4ZA06bhbc2aW1/UXr16aOrg+RVJimue8ZWU0rKz4d//hjp1wtucHDj11LDV+YUX DL1SyvnhB2jQIITeSpXC6u7G0Nu+/dahF0L7/vbtw8clSXHL4CspJeXlhe3M9erBLbeEFd/DDw/N rN5/H445JuoKJRW6zz+Hk06Cn36C2rVh0qTQwj03F2644Y/vp9s417mz250lKY4ZfCWlnLFj4bjj wlbmuXOhWrVwXdG0adCixbZXUElKAWPGhIYMv/0G9evD5Mlw0EHhYxMnbrvSu6VYDObPD89JkuKS wVdSypg5E1q33ryVeZ99whne778P13PabVtKUS+/DK1ahUu6TzstbG+uXHnzxxct2rmvs7PPSZIK ncG3EPTt25eTTjqJUqVKUa5cuZ36nFgsRkZGBtWqVaNkyZI0btyYmTNnFnClUnJauBAuuwyOPBLe fRf22itcS/Tjj6GD8957R12hpMjcfz907Ajr18P554dfEmXKbP1M1ao797V29jlJUqEz+BaCnJwc zj33XK666qqd/px7772XgQMH8tBDDzF16lSqVKlC06ZNWblyZQFWKiWXlSvD9Zp168KTT4Zzve3a hZXfhx4KfWskpahYDG67DW66KYxvuCEc/P+ji7obNgzdm7d3DiItDWrUCM9JkuKSwbcQ9OrVixtv vJEjjjhip56PxWIMGjSIO++8k3bt2nH44Yfz3HPPsWbNGl7aeLmopO1avx4eeSQcz+vTB9asgRNP hI8+gjffhIMPjrpCSZFavx46dYJ77w3je+4JK79FtvOyKD09XFkE24bfjeNBgzwvIUlxzOAbh+bM mcPixYs544wzNs0VL16cRo0aMXny5Agrk+JbLAZvvQVHHAFXXw1LloTw+8YboTlrgwZRVygpcqtX Q9u28PzzIag+80xY+f2zrnbt2oVfJvvvv/V89eph3nt8JSmu7RV1AdrW4sWLAai8ZWON/45//vnn 7X5ednY22dnZm8ZZWVkFU6AUhz7+GG69NazqAlSoAD17whVXQNGi0dYmKU78/ntoYvXpp1CyJLz+ ehjvrHbtQmieODE0sqpaNWxvdqVXkuKeK767KSMjg7S0tB3+++yzz/boe6T9z1+fY7HYNnNb6tev H2XLlt30r0aNGnv0/aVE8MMPcO65m7cylygBd9wR5q+91tAr6b/mzg3bPj79FMqXhw8/3LXQu1F6 erj2qEOH8NbQK0kJwRXf3XTttddywQUX7PCZmjVr7tbXrlKlChBWfqtu0SFyyZIl26wCb6lr167c tLFJB2HF1/CrZPX77+H87iOPhON6aWnhyF7v3mHnoSRtMn06NG8eVmkPOABGjYJ69aKuSpJUiAy+ u6lChQpUqFChQL52rVq1qFKlCmPGjKF+/fpA6Aw9fvx4+vfvv93PK168OMX/qBullETWrg09Zvr1 g427+Zs3h/79w3VFkrSV8eOhTZvwC+Pww2HkyG3P6UqSkp5bnQvBvHnzmDZtGvPmzSM3N5dp06Yx bdo0Vq1atemZevXqMXToUCBsce7cuTN33303Q4cOZcaMGXTq1IlSpUrRsWPHqP4zpEjl5sJzz4WO zF27htewRx8NY8bAiBGGXkl/4M03oVmz8AujYcNwNtfQK0kpyRXfQtCjRw+ee+65TeONq7hjx46l cePGAMyaNYvMzMxNz3Tp0oW1a9dy9dVXs3z5co4//nhGjx5N6dKlC7V2KR6MHg1dusBXX4VxjRrQ ty/8/e/bv31EUop75BG45prQ7v3ss+HFF0NDK0lSSkqLxWKxqItQwcjKyqJs2bJkZmZSpkyZqMuR dtlXX4VOzWPGhHHZsqFx1fXXhyZWkrSNWCy0dO/TJ4yvuAL+7/9sQiUpUr4uj54rvpLizvz50K0b /Oc/4TVs0aJh4aZbN9hvv6irkxS3NmwIl3g/8UQYZ2RAjx5/fkevJCnpGXwlxY3MzNC0avBgWLcu zJ1/Ptx9N9SuHW1tkuLc2rXhiqG33w5nIB5+OKz2SpKEwVdSHMjJCcfx+vSBpUvD3CmnwIABcNxx 0dYmKQEsXx46N3/0ERQvDi+/HM71SpL0XwZfSZGJxeCNN0KX5h9/DHP16sG990Lr1u5OlLQTfvkl 3Gk2c2ZoBDBsWPjLmSRJWzD4SorExImhcdUnn4Rx5crQqxdccgns5W8mSTvj22/DdUXz50O1auGO 3iOOiLoqSVIc8uWlpEL13Xdw++3hGB7A3nuHAHzzzbDPPtHWJimBTJkStoYsWwaHHAKjRsGBB0Zd lSQpTqVU8M3OzubTTz9l7ty5rFmzhooVK1K/fn1q1aoVdWlS0lu8OKzoPvEE5OaG3jOXXhqarlat GnV1khLK8OFw3nmhodUJJ4SxLd8lSTuQEsF38uTJPPjgg7z11lvk5ORQrlw5SpYsybJly8jOzqZ2 7dpcfvnlXHnllZQuXTrqcqWksno1/Pvf4dzu6tVh7swzoX9/+Mtfoq1NUgJ65hm47LLwF7RWreDV V8PWEUmSdqBI1AUUtLZt29K+fXv2339/Ro0axcqVK1m6dCm//PILa9asYfbs2XTr1o0PPviAgw8+ mDFjxkRdspQUNmwIq7sHHQQ9e4bQ+7e/wbhxofeMoVfSLonFwt1mF18cQm+nTjB0qKFXkrRTkn7F 94wzzuD111+nWLFif/jx2rVrU7t2bS688EJmzpzJwoULC7lCKbnEYvDuu3DbbfDNN2GuVq1wP++5 54YtzpK0S/LyoHNnePDBMO7aFfr2tfW7JGmnpcVisVjURahgZGVlUbZsWTIzMylTpkzU5SgFfPZZ aFQ1blwYly8P3bvDVVeFqzUlaZdlZ8O//gWvvRaC7qBBcP31UVclSbvE1+XRS7m1lxUrVvDkk0/S tWtXli1bBsAXX3zBggULIq5MSlxz5kDHjpu3MhcvDl26hLt5O3c29EraTVlZ0LJlCL1Fi8LLLxt6 JUm7Jem3Om9p+vTpnH766ZQtW5a5c+dy2WWXUb58eYYOHcrPP//M888/H3WJUkJZtizsNnzoIcjJ CXP//Cf06eOtIpL20OLF0KIFTJsGpUuH87ynnRZ1VZKkBJVSK7433XQTnTp1Yvbs2ZQoUWLTfIsW LZgwYUKElUmJZd06uO8+qFMHBg4Mofe00+CLL+D55w29kvbQ7Nlw0kkh9FaqFLaSGHolSXsgpVZ8 p06dymOPPbbN/P7778/ixYsjqEhKLHl5YafhnXfCzz+HuSOOCFcVNWtmnxlJ+eCzz8L25t9+C39d GzUqvJUkaQ+k1IpviRIlyMrK2mZ+1qxZVKxYMYKKpMTx4YfhDO8//hFCb7Vq8PTT8OWX0Ly5oVdS Phg9Gho3DqH3mGNg0iRDryQpX6RU8G3bti29e/dm/fr1AKSlpTFv3jxuv/12zjnnnIirk+LTjBlh 8WXjVubSpcO53tmz4aKLID096golJYUXX4RWrcKl36efHrY3V64cdVWSpCSRUsH3vvvu47fffqNS pUqsXbuWRo0acdBBB1G6dGn69u0bdXlSXFmwAC69FI46CkaMgL32gmuvDZ2a77gDSpWKukJJSWPg wLCdZMMG6NAhXAZeunTUVUmSkkhKnfEtU6YMH330ER9++CFffPEFeXl5HHPMMZx++ulRlybFjays cGZ34EBYuzbMnXMO9OsHdetGW5ukJJOXB7fdFrrlQbj/7N//hiIp9Xd5SVIhSIvFYrGoi1DB8KJs 7Yr16+GJJyAjIxyvg9BU9b774MQTIy1NUjJavx4uvhheeCGM770XbrnFhgGSkpKvy6OXUiu+AJ9+ +injxo1jyZIl5OXlbfWxgQMHRlSVFJ1YLFyPefvt4dwuhJXde+6Bs8/2NaikArBqFbRvHzo2p6eH Tnn/+lfUVUmSklhKBd+7776bbt26ccghh1C5cmXStnhFn+are6WgKVPg1ltD41SAihWhZ0+4/HIo WjTa2iQlqd9+C02spk4NzQLeeANatIi6KklSkkup4Dt48GCefvppOnXqFHUpUqRmz4auXeHNN8O4 ZEm46Sbo0gXcfSOpwMyZEy79nj0b9tsvNLE6/vioq5IkpYCUCr5FihShQYMGUZchRea336B3b3j0 0dA8NS0tXEnUuzfsv3/U1UlKal99FS79XrwYDjwwbHM+5JCoq5IkpYiUapt444038n//939RlyEV ujVr4O67oU4deOihEHpbtAivQ596ytArqYCNGwennBJC7xFHwOTJhl5JUqFKqRXfW265hVatWlGn Th0OPfRQiv7PIcYhQ4ZEVJlUMHJz4fnnoXv3cC8vQP36MGAAnHZatLVJShFvvAF//zvk5ITw+/bb UK5c1FVJklJMSgXf6667jrFjx9KkSRP2228/G1opacViYRdhly7w9ddh7oADwqpvhw5ekSmpkDz8 MFx7bfil1K4dvPgilCgRdVWSpBSUUsH3+eef580336RVq1ZRlyIVmC+/DIH3/ffDuFw5uPPO8NrT 15uSCkUsBj16wF13hfFVV8GDD4ariyRJikBKBd/y5ctTp06dqMuQCsS8edCtG7zwQnjNWbRoCLt3 3hmap0pSodiwIQTdJ58M4969wy8nd1lJkiKUUhseMzIy6NmzJ2vWrIm6FCnfrFgBt90GBx8M//lP CL0dOsCsWTBwoKFXUiFaswbOOSeE3iJF4LHHQpMBQ68kKWIpteL7wAMP8OOPP1K5cmVq1qy5TXOr L774IqLKpF2XnQ2PPAJ9+sCyZWGuUaPQuOpvf4u2NkkpaNkyOPPM0LG5RAl4+WU466yoq5IkCUix 4HuW/wesJBCLwWuvQdeuMGdOmPvLX+Dee6FVKxdWJEVg/vxwR+8334TGAu+8AyefHHVVkiRtkhaL xWJRF6GCkZWVRdmyZcnMzKRMmTJRl6N8MGEC3HILTJ0axlWqhONzF10Ee6XUn7EkxY1vvoFmzeCX X8Kl4CNHwuGHR12VJMUVX5dHz5fKUgL49lu4/XYYNiyM9947dG6+6SbYZ59oa5OUwiZPhtatYfly qFcv3KN2wAFRVyVJ0jaSPviWL1+e77//ngoVKrDvvvvu8O7eZRsPSkpxYvFiyMgIfWJyc8NNIJdd Bj17htVeSYrMsGFw/vmwbh2ccAIMH243PUlS3Er64Hv//fdTunTpTe/vKPhK8WLVKvj3v0OjqtWr w1zbtnDPPWFRRZIi9dRTcPnlkJcXVnxffRVKlYq6KkmStsszvknMswSJZ8MGePrpsKK7eHGYO+64 EIBPOSXa2iSJWAzuvjvcywuhwcDjj9tkQJL+hK/Lo5dS9/imp6ezZMmSbeaXLl1Kenp6BBVJQSwW mqAeeSRccUUIvbVrh0WUjz829EqKA7m5cN11m0PvHXeElV9DryQpAaTU/1ttb3E7OzubYsWKFXI1 UjB1Ktx6K4wfH8bly0OPHnDVVeD/LCXFhexs+Mc/4I03wp1pgweHECxJUoJIieD7wAMPAJCWlsaT Tz7JPlu0wc3NzWXChAnU8+Ck8llGRmhG1b37th/r0wd+/x1+/TWs6gIULw6dO4fuzeXKFWqpkrR9 mZlw1lkwblz4a9x//gPnnRd1VZIk7ZKUCL73338/EFZ8H3300a22NRcrVoyaNWvy6KOPRlWeklR6 eli5ha3Db9euoUlVkSKhL0xaGvzznyEMewuIpLiyaBG0aAFffQWlS8Nbb8Gpp0ZdlSRJuywlgu+c OXMAaNKkCUOGDGHfffeNuCKlgo1hd2P4vfXW0Jl59OgwzsuDpk3h3nvh6KOjqVGStuv776FZM5g7 FypXhhEjoH79qKuSJGm32NU5idk9Lj707h26NG/pyCNDp+YzzoimJknaoalToWXLcCbjoINg1KjQ cU+StFt8XR69lFjxlaLy/vthZ+CWnn029IixkbikuDRqFJxzTrhE/Nhj4b33oFKlqKuSJGmPpNR1 RlJhmT49HItr2hS+/DLMbQy68+YZeiXFqRdegNatQ+ht2hTGjjX0SpKSgsFXykcLFsDFF4czuyNH hgZWALfdBhs2hG3PPXqERlaSFFf+/e/QaW/DBujYEYYPDw2tJElKAm51lvJBVhb07w/33w9r14a5 Qw+Fb74JYXdjo6v/bXj1R1cdSVKhysuDLl1C8AW46abQhKCIfxuXJCWPpA++06dP3+lnjzzyyAKs RMlo/Xp47DHo1Sv0gAFo0ADuuy+s+P7RPb4bx7m5hVurJG0jJydsU3nxxTAeMABuuSXamiRJKgBJ 39W5SJEipKWlsb3/zI0fS0tLIzfJkojd4wpOLAZDhoQ7eWfPDnMHHxxWfdu2DXfzSlJcW7UqNLEa PRr22guefjpsdZYk5Ttfl0cv6Vd8N97hK+WXyZPDgsiUKWFcqRJkZMCll0LRopGWJkk7Z8kSaNUK PvsMSpWCN9+E5s2jrkqSpAKT9MH3wAMPjLoEJYnvvw8rvEOGhHHJknDzzeFonP1fJCWMOXPCJeI/ /AAVKsC778Jxx0VdlSRJBSrpg+8f+eabb5g3bx45OTlbzbdp0yaiihTPliwJDaoeeyw0Oy1SJByJ 69ULqlWLujpJ2gXTpoW71hYvhpo1w529Bx8cdVWSJBW4lAq+P/30E2effTZff/31Vud+0/57IDPZ zvhqz6xZE7o09+8PK1eGuZYtw/jww6OtTZJ22dixoQnBypVw5JGhA1/VqlFXJUlSoUipuwpuuOEG atWqxa+//kqpUqWYOXMmEyZM4K9//Svjxo2LujzFidzc0OOlbl3o1i28RjzmGPjgg7Aj0NArKeG8 /no4w7tyJTRuDBMmGHolSSklpYLvlClT6N27NxUrVqRIkSIUKVKEk08+mX79+nH99ddHXZ4iFovB iBFw9NFwySWwcCEceGC45WPqVDj11KgrlKTd8NBDcP754eqi9u3DL7qyZaOuSpKkQpVSwTc3N5d9 9tkHgAoVKrBw4UIgNMCaNWtWlKUpYl98AU2bhq3MM2ZAuXLhLt7vvoOOHcO5XklKKLFY2LZy3XXh /auvhldegRIloq5MkqRCl1JnfA8//HCmT59O7dq1Of7447n33nspVqwYjz/+OLVr1466PEXg55/D 68IXXgjjYsXCa8Q77oDy5aOtTZJ224YNcMUV4dwGQJ8+cOedXjIuSUpZKRV8u3XrxurVqwG46667 aN26NQ0bNmS//fbj1Vdfjbg6Fably6FfP3jgAcjODnMdO8Jdd0GtWtHWJkl7ZM0auOACeOedsF3l scfCReOSJKWwtNjG1sYpatmyZey7776bOjsnk6ysLMqWLUtmZiZlypSJupy4kJ0NDz8cFj+WLw9z jRvDgAHw179GWpok7blly6B1a5gyJWxpfvVV8Ko+SYqcr8ujl1Irvhv98MMP/Pjjj5xyyimUL1+e FM/+KSEvD157LWxhnjMnzB16KNx7bzjXm4R/95CUaubPh2bN4NtvYd99w4pvgwZRVyVJUlxIqZY9 S5cu5bTTTuPggw+mZcuWLFq0CIBLL72Um2++OeLqVFDGj4cTToAOHULorVoVnngCvvoKWrUy9EpK AjNnwoknhtBbvTpMnGjolSRpCykVfG+88UaKFi3KvHnzKFWq1Kb5888/n5EjR0ZYmQrCN9/AmWeG rcxTp8I++0Dv3jB7djjutldK7neQlHQmTYKTT4YFC+Avf4HJk+Gww6KuSpKkuJJSL/1Hjx7NqFGj qF69+lbzdevW5eeff46oKuW3RYugZ0946qmwxTk9HS6/PMxVrhx1dZKUj4YNC3f0rlsHJ50Utjfb kl6SpG2kVPBdvXr1Viu9G/3+++8UL148goqUn1atCnfv3ncf/Ld5N2edBffcA4ccEm1tkpTvnnwy XFmUlxe2t7zyCvzB/8dJkqQU2+p8yimn8Pzzz28ap6WlkZeXx4ABA2jSpEmElWlPbNgAjz4KBx0E vXqF0HvCCeGI29Chhl5JSSYWC63pL7sshN5LLoEhQwy9kiTtQEqt+A4YMIDGjRvz2WefkZOTQ5cu XZg5cybLli1j0qRJUZenXRSLhV19t90G330X5urUCSu855xj0ypJSSg3F66/PtzLBtCtW2he4C88 SZJ2KKVWfA899FCmT5/OcccdR9OmTVm9ejXt2rXjyy+/pE6dOlGXp13w6aehaVXbtiH07rcfDB4c Glq1b+9rQElJaN26cJ734YfDL7mHHgorv/7CkyTpT6XMiu+GDRvo27cvF198Mb169Yq6HO2mH38M d/G+9loYlygBnTvD7bdD2bLR1iZJBSYzM/ylb/x4KFYMXngBzj036qokSUoYKbPiu9deezFgwABy c3OjLkW7YenSEHD/8pcQetPS4MIL4fvvoV8/Q6+kJLZwIZxySgi9ZcrAyJGGXkmSdlHKBF+A008/ nXHjxkVdhnbB2rXQv384uzt4MKxfD2ecAV9+Cc8+CzVqRF2hJBWgWbPCNUXTp0OVKiH82oxRkqRd ljJbnQFatGhB165dmTFjBsceeyx77733Vh9v06ZNRJXpf+XlhZ183brB/Plh7qij4N57Q/CVpKT3 6afQsmXY8lK3LowaBbVqRV2VJEkJKS0Wi8WiLqKwFCmy/QXutLS0pNsGnZWVRdmyZcnMzKRMmTJR l7PTxoyBLl1g2rQwrl4d+vaFv/8d0tOjrU2SCsWIEaFT35o18Le/wbvvQsWKUVclSdpNifq6PJmk 1IpvXl5e1CVoB6ZPD4F31KgwLlMGunaFG26AkiWjrU2SCs3zz4e7eTdsgGbN4I03YJ99oq5KkqSE llJnfBWffvkFLroIjj46hN6iRUPY/fHH0K3Z0CspJcRiMGBA6Ny3YQP84x8wbJihV5KkfJD0wfeV V17Z6Wfnz5/PpEmTCrAabSkzM1xNVLduaFQVi4VGpd9+C4MGQYUKUVcoSYUkLw9uvjlsewG45RZ4 7rlwdZEkSdpjSR98H3nkEerVq0f//v359ttvt/l4ZmYm7733Hh07duTYY49l2bJlEVSZWnJy4MEH 4aCDwlVE69bBySfDxx+Hq4rq1Im6QkkqRDk58M9/wv33h/F994WV3x30pZAkSbsm6c/4jh8/nuHD h/Pggw9yxx1gcLOcAAAgAElEQVR3sPfee1O5cmVKlCjB8uXLWbx4MRUrVuSiiy5ixowZVKpUKeqS E15GRmhC1b371vOxGHToAKNHw/LlYe6QQ8J1RW3ahLt5JSmlrFwZmliNHg177RW2v/z971FXJUlS 0kn64AvQunVrWrduzdKlS/noo4+YO3cua9eupUKFCtSvX5/69evvsOOzdk16OvToEd7fGH4nTYIL LgjneQEqVYJevUL/lqJFo6lTkiK1ZEm4rujzz2HvveHNN0MzK0mSlO9SIvhutN9++9G2bduoy0h6 G8Nujx7w++/hHt6hQ8Nc0aKhU/Mtt0Dp0tHVKEmR+umnEHJ/+CE0NHjvvXBtkSRJKhApFXxVeLYM vxsde2xoUFqtWjQ1SVJc+PJLaNECfv0VatUK7ezr1o26KkmSklpaLBaLRV1EYdl3331J+4ODpGlp aZQoUYKDDjqITp06cdFFF0VQXf6Lh4uyixWD9evDSm9OTiQlSFL8+PBDOOuscLb3qKNg5EioUiXq qiRJBSweXpenupQ62NqjRw+KFClCq1at6NWrFxkZGbRq1YoiRYpwzTXXcPDBB3PVVVfxxBNPRF1q UujTJ4TejeG3T5+oK5KkCL32GjRvHkJvkyYwfryhV5KkQpJSW50/+ugj7rrrLq688sqt5h977DFG jx7Nm2++yZFHHskDDzzAZZddFlGVyaFPn7DNuXfvsO154xi27fYsSUnvwQfhhhs2X1j+n/9A8eJR VyVJUspIqRXfUaNGcfrpp28zf9pppzFq1CgAWrZsyU8//VTYpSWV/w29EN727h3mXfmVlDJiMbjj Drj++vD+tdfCyy8beiVJKmQpFXzLly/PO++8s838O++8Q/ny5QFYvXo1pW03vEdyc7cOvRttDL+5 udHUJUmFasOGcGdbv35h3LcvPPBAuPNNkiQVqpTa6ty9e3euuuoqxo4dy3HHHUdaWhqffvop7733 Ho8++igAY8aMoVGjRhFXmtgyMrb/Mbc5S0oJa9bA+efD8OEh6D7+OFx8cdRVSZKUslKqqzPApEmT eOihh5g1axaxWIx69epx3XXXcdJJJ0VdWr6ze5wkRWDpUjjzTJgyBUqUCE2tzjwz6qokSRHydXn0 Ui74phJ/wCSlnNxcmDgRFi2CqlWhYcPC3Vo8bx40awbffQf77htWfJPwD6uSpF3j6/LopdRWZ4Dc 3Fzeeustvv32W9LS0jj00ENp06YN6Z65kqTENmRI6Jz8yy+b56pXh8GDoV27gv/+M2aE64oWLAjf d9QoOPTQgv++kiTpT6VU8P3hhx9o2bIlCxYs4JBDDiEWi/H9999To0YN3n33XerUqRN1iZKk3TFk CLRvHzonb2nBgjD/xhsFG34nToQ2bWDFCjjsMBg5MoRfSZIUF1Jqq3PLli2JxWK8+OKLm7o4L126 lH/84x8UKVKEd999N+IK85dbKiSlhNxcqFlz65XeLaWlhRA6Z07BbHt+6y3o0AHWrYMGDeCdd8I2 Z0mS/svX5dFLqRXf8ePH8/HHH28KvQD77bcf99xzDw0aNIiwMknSbps4cfuhF8Iq8Pz54bnGjfP3 ez/+OFx1FeTlhRXfV16BkiXz93tIkqQ9llL3+BYvXpyVK1duM79q1SqKFSsWQUWSpD22aFH+Prcz YrFwMfkVV4TQe9ll8Oabhl5JkuJUSgXf1q1bc/nll/PJJ58Qi8WIxWJ8/PHHXHnllbRp0ybq8iRJ u6Nq1fx97s/k5sLVV0PPnmHcvTs89hjslVKbqCRJSigpFXwfeOAB6tSpw4knnkiJEiUoUaIEDRo0 4KCDDmLw4MEF9n379u3LSSedRKlSpShXrtxOfU6nTp1IS0vb6t8JJ5xQYDVKUsJq2DCc4U1L++OP p6VBjRrhuT21bh2cdx48+mj4ug8/HFZ+t/e9JUlSXEipP0+XK1eOt99+m9mzZ/Pdd98Ri8U49NBD Oeiggwr0++bk5HDuuedy4okn8tRTT+305zVv3pxnnnlm09jt2JL0B9LTw5VF7duHALplz8aNgXTQ oD1vbLViBbRtCxMmQLFi8NJLcM45e/Y1JUlSoUip4LtR3bp1qVu3bqF9v169egHw7LPP7tLnFS9e nCpVqhRARZKUZNq1C1cW/dE9voMG7flVRgsXhjt6v/4aypSBYcOgUaM9+5qSJKnQJH3wvemmm3b6 2YEDBxZgJbtu3LhxVKpUiXLlytGoUSP69u1LpUqVtvt8dnY22dnZm8ZZWVmFUaYkxYd27cKK7MSJ oZFV1aphe/OervTOmgXNmsHPP4evOWIEHHVU/tQsSZIKRdIH3y+//HKnnkuLs/NZLVq04Nxzz+XA Aw9kzpw5dO/enVNPPZXPP/+c4sWL/+Hn9OvXb9PqsiSlpPT0/L2y6JNPoFUrWLoUDj4YRo0KdwZL kqSEkhaLbXkYSjsrIyPjT0Pm1KlT+etf/7pp/Oyzz9K5c2dWrFixy99v0aJFHHjggbzyyiu0286W vT9a8a1Ro4YXZUvS7hgxIpwbXrMGjjsOhg+HihWjrkqSlICysrIoW7asr8sjlPQrvgXl2muv5YIL LtjhMzXzcVWgatWqHHjggcyePXu7zxQvXny7q8GSpF3w3HNwySXh6qLmzcP54b33jroqSZK0mwy+ u6lChQpUqFCh0L7f0qVLmT9/PlXz6x5KSdK2YjEYMABuuy2M//UvePJJKFo02rokSdIeSal7fKMy b948pk2bxrx588jNzWXatGlMmzaNVatWbXqmXr16DB06FIBVq1Zxyy23MGXKFObOncu4ceM488wz qVChAmeffXZU/xmSlNzy8uCmmzaH3i5d4NlnDb2SJCUBV3wLQY8ePXjuuec2jevXrw/A2LFjafzf JiyzZs0iMzMTgPT0dL7++muef/55VqxYQdWqVWnSpAmvvvoqpUuXLvT6JSnp5eRAp07w8sthPHAg 3HhjpCVJkqT8Y3OrJOYheknaCStXhquQ3n8/rO4++yx07Bh1VZKkJOLr8ui54itJSl2//gotW8IX X8A++8CQIdC0adRVSZKkfGbwlSSlph9/hGbNwtuKFcP1RcceG3VVkiSpANjcSpKUer74Ak46KYTe 2rVh8mRDryRJSczgK0lKLe+/D40awZIlcPTRMGkSHHRQ1FVJkqQCZPCVJKWOV14JZ3pXrYJTT4Xx 46FKlairkiRJBczgK0lKDYMHQ4cOsH49nHcevPce2FlTkqSUYPCVJCW3WAy6doXOncP4+uvDfb3F i0dblyRJKjR2dZYkJa/16+Hyy8PdvAD9+sFtt0FaWqRlSZKkwmXwlSQlvowMSE+H7t03z61evXlL c1oaPP00dOoUVYWSJClCBl9JUuJLT4cePcL73bvD0qXQqhV88kmY69jR0CtJUgoz+EqSEt/Gld4e PWDFCnj3XZg1K8xddhk8/nh0tUmSpMgZfCVJyaF7d/j1Vxg4cPPcddfBAw9EV5MkSYoLdnWWJCWH CRPghRc2j4sVM/RKkiTA4CtJSgZDh8IZZ0BmZhgXKwY5OdCnT7R1SZKkuGDwlSQltsceg/btITs7 jLt3D+/37h3O/Bp+JUlKeZ7xlSQlplgshNuMjM1zGRnQs2d4f8uGV1uOJUlSyjH4SpIST24uXHNN WO0FaNQITj11c8jdaGPYzc0t3PokSVJcMfhKkhLLunXhXt6hQyEtDR5+GK68cvvPu9IrSVLKM/hK khLHihXQpg1MnAjFi8NLL0G7dlFXJUmS4pzBV5KUGBYsgObNYcYMKFsW3n47bHGWJEn6EwZfSVL8 ++47aNYM5s2DqlVh1Cg44oioq5IkSQnC64wkSfHt44+hQYMQeg85BKZMMfRKkqRdYvCVJMWvd98N 3ZqXLYPjj4ePPoIDD4y6KkmSlGAMvpKk+PTss9C2LaxdCy1awAcfQIUKUVclSZISkMFXkhRfYjG4 5x646KJw/+6FF4ZGVnvvHXVlkiQpQdncSgUmNzfcOLJoUehF07AhpKdHXZWkuJaXBzfeCA88EMa3 3Qb9+oX7eiVJknaTwVcFYsgQuOEG+OWXzXPVq8PgwV65KWk7srPD6u6rr4bx/fdD587R1iRJkpKC W52V74YMgfbttw69EK7gbN8+fFyStpKVBa1ahdBbtCi8/LKhV5Ik5RuDr/JVbm5Y6Y3Ftv3YxrnO ncNzkgTAr79C48ahedU++8B778EFF0RdlSRJSiIGX+WriRO3XendUiwG8+eH5ySJH36Ak06CL7+E SpVg3Dg4/fSoq5IkSUnGM77KV4sW5e9zkpLY559Dy5awZAnUrg2jR0OdOlFXJUmSkpArvspXVavm 73OSktSYMWF785IlUL8+TJ5s6JUkSQXG4Kt81bBh6N68vZtH0tKgRo3wnKQU9fLLoZHVqlVw2mlh e3PlylFXJUmSkpjBV/kqPT1cWQTbht+N40GDvM9XSlmDBkHHjrB+PZx/Prz7LpQpE3VVkiQpyRl8 le/atYM33oD99996vnr1MO89vlIKisXgttvgxhvD+IYb4KWXoHjxaOuSJEkpweZWKhDt2kHbtqF7 86JF4Uxvw4au9Eopaf16uPRSeP75ML7nHujSZftnIiRJkvKZwVcFJj099K6RlMJWr4Zzz4URI8Iv hSefhE6doq5KkiSlGIOvJKlg/P57aGL16adQsiS8/noYS5IkFTKDryQp/82dC82bw6xZUL58aGJ1 wglRVyVJklKUwVeSlL+mTw+hd9EiOOAAGDUK6tWLuipJkpTC7OosSco/48fDKaeE0Hv44TB5sqFX kiRFzuArScofQ4ZAs2aQmRnauE+cuO29ZpIkSREw+EqS9twjj0D79pCdDWefHbY3lysXdVWSJEmA wVeStCdiMejZE66+Orx/xRWhe3PJklFXJkmStInNrSRJu2fDBrjmGnj88TDOyIAePSAtLdKyJEmS /pfBV5K069auhQ4d4O23oUgRePjhsNorSZIUhwy+kqRds3w5tGkDH30ExYvDyy+Hc72SJElxyuAr Sdp5v/wS7uidORPKloVhw8L1RZIkSXHM4CtJ2jnffhuuK5o/H6pVg5Ej4Ygjoq5KkiTpT9nVWZL0 56ZMgZNPDqH3kENg8mRDryRJShiu+EpSMsjNhYkTYdEiqFoVGjaE9PT8+drDh8N554WGVscfH8YV KuTP15YkSSoErvhKUqIbMgRq1oQmTaBjx/C2Zs0wv6eeeQbOOiuE3pYt4YMPDL2SJCnhGHwlKZEN GQLt24emU1tasCDM7274jcXg7rvh4ovDavKFF8Jbb8Hee+95zZIkSYXM4CtJiSo3F264IYTU/7Vx rnPn8NyuyMsLX/fOO8P49tvDym/RontWryRJUkQMvpKUqCZO3Hald0uxWGhGNXHizn/N7Gzo0AEe fDCMBw2Cfv0gLW3PapUkSYqQza0kKVEtWpS/z2Vlwdlnw4cfhtXd55+HCy7Y/fokSZLihMFXkhJV 1ar599zixdCiBUybBvvsA0OHwumn71l9kiRJccKtzpKUqBo2hOrVt78NOS0NatQIz+3I7Nlw0kkh 9FaqBOPHG3olSVJSMfhKUqJKT4fBg8P7/xt+N44HDdrxfb6ffQYNGsCcOVCnDkyeDMccUzD1SpIk RcTgK0mJrF07eOMN2H//reerVw/z7dpt/3NHj4bGjeG330LYnTQphF9JkqQk4xlfSUp07dpB27ah e/OiReFMb8OGO17pfemlcDfvhg1hW/OQIVC6dOHVLEmSVIgMvpKUDNLTw+rtzhg4EG6+ObzfoQM8 +ywUK1ZQlUmSJEXOrc6SlCry8qBLl82ht3NneOEFQ68kSUp6rvhKUipYvx4uuQT+858w7t8fbr11 +x2hJUmSkojBV5KS3apVcO65MHJk2BL99NPwr39FXZUkSVKhMfhKUjL77Tdo1QqmToVSpeD116Fl y6irkiRJKlQGX0lKVnPnwhlnwOzZsN9+8O67cPzxUVclSZJU6Ay+kpSMvvoKmjeHxYvhgAPCnb2H HBJ1VZIkSZEw+EpSIsvICOd2u3ffPDduXLjXNysLKlWCKVOgWrWoKpQkSYqc1xlJUiJLT4cePaBP nzB+4w1o1iyEXoBLLzX0SpKklOeKryQlso0rvT16wKefhnO8sdjmuV69oqtNkiQpTrjiK0mJrls3 aNQIhg/fHHozMgy9kiRJ/2XwlaREtmEDXH45jB+/ea5YMejZM7qaJEmS4ozBV5IS1dq1cM458OST kJYW5ooVg5yczWd+JUmSZPCVpIS0bBk0bQrDhoUGV7EY9O4N2dnh7ZYNryRJklKcza0kKdHMnx/u 6P3mGyhRAtatC2F3Y6OrLRtebTmWJElKUQZfSUok33wTriv65RfYf39o0waqVt023G4c5+YWfo2S JElxxuArSYli8mRo3RqWL4d69WDUKDjggO0/70qvJEkS4BlfSUoM77wDp50WQu8JJ8BHH+049EqS JGkTg68kxbunnoKzzw5neVu1gg8+gP32i7oqSZKkhGHwlaR4FYtB375w6aXhrO5FF8HQoVCqVNSV SZIkJRSDryTFo9xcuO466NYtjLt2DSu/RYtGW5ckSVICsrmVJMWb7Gz4xz/gjTcgLQ0GDw4hWJIk SbvF4CtJ8SQzE846C8aNC6u7L7wA550XdVWSJEkJzeArSfFi0SJo0QK++gpKl4a33oJTT426KkmS pIRn8JWkePD999CsGcydC5Urw4gRUL9+1FVJkiQlBZtbSVLUpk6FBg1C6D3oIJg82dArSZKUjwy+ khSlUaOgSRP4/Xc49liYNAlq1466KkmSpKRi8JWkqLzwArRuDatXQ9OmMHYsVKoUdVWSJElJx+Ar SVH497/hn/+EDRugY0cYPjw0tJIkSVK+M/hKUmHKy4Nbbgn/AG68Ef7zHyhWLNq6JEmSkphdnSWp sOTkwMUXw4svhvGAAZsDsCRJkgqMwVeSCsOqVXDOOTB6NOy1Fzz9dNjqLEmSpALnVucCNnfuXC65 5BJq1apFyZIlqVOnDj179iQnJ2eHnxeLxcjIyKBatWqULFmSxo0bM3PmzEKqWlK+WrIkdG4ePRpK lYJ33jH0SpIkFSKDbwH77rvvyMvL47HHHmPmzJncf//9PProo9xxxx07/Lx7772XgQMH8tBDDzF1 6lSqVKlC06ZNWblyZSFVLilfzJkT7uj97DOoUCF0bm7ePOqqJEmSUkpaLBaLRV1EqhkwYACPPPII P/300x9+PBaLUa1aNTp37sxtt90GQHZ2NpUrV6Z///5cccUVO/V9srKyKFu2LJmZmZQpUybf6pe0 k6ZNgxYtYPFiOPDAsOJ78MFRVyVJkgqZr8uj54pvBDIzMylfvvx2Pz5nzhwWL17MGWecsWmuePHi NGrUiMmTJ2/387Kzs8nKytrqn6SIjB0Lp5wSQu+RR8LkyYZeSZKkiBh8C9mPP/7Igw8+yJVXXrnd ZxYvXgxA5cqVt5qvXLnypo/9kX79+lG2bNlN/2rUqJE/RUvaNa+/HrYzr1wJjRrBhAlQrVrUVUmS JKUsg+9uysjIIC0tbYf/Pvvss60+Z+HChTRv3pxzzz2XSy+99E+/R1pa2lbjWCy2zdyWunbtSmZm 5qZ/8+fP373/OEm776GH4Pzzw9VF55wDI0dC2bJRVyVJkpTSvM5oN1177bVccMEFO3ymZs2am95f uHAhTZo04cQTT+Txxx/f4edVqVIFCCu/VatW3TS/ZMmSbVaBt1S8eHGKFy++E9VLynexGHTvDn37 hvHVV8MDD0B6erR1SZIkyeC7uypUqECFChV26tkFCxbQpEkTjj32WJ555hmKFNnxQnutWrWoUqUK Y8aMoX79+gDk5OQwfvx4+vfvv8e1S8pnGzbAFVeEu3kB+vSBO++EHezQkCRJUuFxq3MBW7hwIY0b N6ZGjRrcd999/PbbbyxevHibs7r16tVj6NChQNji3LlzZ+6++26GDh3KjBkz6NSpE6VKlaJjx45R /GdI2p41a6BduxB6ixSBJ56Abt0MvZIkSXHEFd8CNnr0aH744Qd++OEHqlevvtXHtrxJatasWWRm Zm4ad+nShbVr13L11VezfPlyjj/+eEaPHk3p0qULrXZJf2LZMmjdGqZMgRIl4JVXoG3bqKuSJEnS //Ae3yTmfWFSAZo/H5o1g2+/hXLlYPhwaNAg6qokSVIc8nV59FzxlaRdNXNmCL0LFkD16qFz82GH RV2VJEmStsMzvpK0KyZNgpNPDqH3L3+ByZMNvZIkSXHO4CtJO2vYMDj9dFixAk48ET76CGrUiLoq SZIk/QmDryTtjCefhLPPhnXrQkOr99+H8uWjrkqSJEk7weArSTsSi4V7eS+7DPLy4OKLYehQKFUq 6sokSZK0kwy+krQ9ublw7bXQo0cY33lnWPndy76AkiRJicRXb5L0R9atg3/8A958E9LS4IEHQgiW JElSwjH4StL/ysyEtm1h/HgoVgxeeAHOPTfqqiRJkrSbDL6StKWFC6FFC5g+HUqXhrffhiZNoq5K kiRJe8DgK0kbzZoFzZrBzz9DlSowYgQcfXTUVUmSJGkP2dxKkgA+/RQaNAiht25dmDzZ0CtJkpQk DL6SNGJE2M68dCn89a8waRLUqhV1VZIkSconBl9Jqe3556FNG1izJmxzHjsWKlaMuipJkiTlI4Ov pNQUi8GAAXDhhbBhQ7i6aNgw2GefqCuTJElSPjP4Sko9eXlw883QpUsY33wzPPdcuLpIkiRJSceu zpJSS04OXHQRvPRSGN93Xwi+kiRJSloGX0mpY+VKOOccGDMG9toLnnkmbHGWJElSUjP4SkoNS5ZA y5bw+eew997w5puhmZUkSZKSnsFXUvL76acQcn/4ASpUgPfeg7/9LeqqJEmSVEgMvpKS25dfQosW 8OuvULMmjBoFBx8cdVWSJEkqRHZ1lpS8PvwQGjUKofeoo2DyZEOvJElSCjL4SkpOr74KzZuHhlaN G8P48VC1atRVSZIkKQIGX0nJ58EHoUMHWL8e2reHkSOhbNmoq5IkSVJEDL6SkkcsBnfcAddfH96/ 5hp45RUoXjzqyiRJkhQhm1tJSg4bNsDll4e7eQHuuiuE4LS0aOuSJElS5Ay+khLfmjVw/vkwfDgU KQKPPw6XXBJ1VZIkSYoTBl9JiW3pUjjzTJgyBUqUCE2t2rSJuipJkiTFEYOvpMQ1bx40awbffQf7 7gvvvAMNGkRdlSRJkuKMwVdSYpoxI1xXtGABVK8Oo0bBoYdGXZUkSZLikF2dJSWeiROhYcMQeg89 FCZPNvRKkiRpuwy+khLLW2/BGWfAihVhW/PEiVCjRtRVSZIkKY4ZfCUljscfh3POgXXrQgOrMWOg fPmoq5IkSVKcM/hKin+xGPTuDVdcAXl5cOml8OabULJk1JVJkiQpARh8JcW33Fy4+mro2TOMu3UL K7972ZtPkiRJO8dXjpLi17p18Pe/w5AhkJYGDz0UQrAkSZK0Cwy+kuLTihXQti1MmADFisGLL0L7 9lFXJUmSpARk8JUUfxYuDHf0fv01lCkDb78NjRtHXZUkSZISlMFXUnyZNQuaNYOff4YqVWDkSDjq qKirkiRJUgKzuZWk+PHJJ+Fu3p9/hrp1YfJkQ68kSZL2mMFXUnwYMQJOPRWWLoW//Q0mTYJataKu SpIkSUnA4Cspes89B2eeCWvWhG3OH34IFStGXZUkSdL/t3fn4VFUedvH76YTskAISlgCBAKIgDoo wiCRJWEzgAjC+CiyCCIoKgrygooLojiiPoAIKIgybLLoAEEW2ZwhIRAZwuaCiMiAsiQDKCZhC1nq /aOe9NCQBLJ0qlP5fq4rV1LVp6p+nUNp7q5Tp2ATBF8A1jEM6d13pUGDzOf1DhggrV4tVaxodWUA AACwEYIvAGtkZ0ujRkkvvGAujxkjzZsn+fpaWhYAAADsh1mdAZS8S5fMq7xLlpjLkyebIRgAAADw AIIvgJKVlib17i199ZXk42Ne5e3Xz+qqAAAAYGMEXwAl5z//kbp1k3bvlipUkFaskO65x+qqAAAA YHMEXwAl49Ahc8bmQ4fMGZu//FJq0cLqqgAAAFAGEHwBeN7u3VLXrtLJk+azeTdskBo2tLoqAAAA lBHM6gzAs776SoqMNEPvHXdICQmEXgAAAJQogi8Az1m61Lyn9+xZqUMHKS5OqlHD6qoAAABQxhB8 AXjG++9LDz8sZWRIDz5o3tNbqZLVVQEAAKAMIvgCKF6GIY0dK40caS4/84z5vF4/P2vrAgAAQJnF 5FYAik9GhjR0qDR/vrn81lvSiy9KDoe1dQEAAKBMI/gCKB7nzv13SLPTKX38sfToo1ZXBQAAABB8 ARSD06el7t2lf/1LCgiQPv/cXAYAAAC8AMEXQNH88osUHS0dOCDdeKO0Zo0UEWF1VQAAAIALwRdA 4X33ndSli3TihBQWJm3YIDVpYnVVAAAAgBtmdQZQOFu2SG3bmqH31lulhARCLwAAALwSwRdAwcXE SPfcI6WkSG3aSPHxUu3aVlcFAAAA5IrgC6BgZs2SHnhASk+XevaUNm6UbrjB6qoAAACAPBF8AVwf w5DGj5eefFLKzpYef1xatsycxRkAAADwYkxuBeDasrKkp56SZs82l8eNM0Oww2FpWQAAAMD1IPgC yN+FC1LfvtLKlWbQ/fBDadgwq6sCAAAArhvBF0De/vhD6tHDnLzKz09avFjq3dvqqgAAAIACIfgC yN3x4+Yzer//XqpUSVq1SoqMtLoqAAAAoMAIvgCu9uOPUnS09OuvUmiotH691LSp1VUBAAAAhcKs zgDcbd8utW5tht6bb5YSEgi9AAAAKNUIvgD+a+1aqUMH6fffpZYtpW3bpPBwq6sCAAAAioTgC8A0 d67Us6c5i3PXrtI//ymFhFhdFQAAAFBkBF+grDMM6e23pcGDzef1DhwoffGFVKGC1ZUBAAAAxYLg C5Rl2WR2K/cAACAASURBVNnSyJHS2LHm8gsvmFd+fX2trQsAAAAoRszqDJRV6enm1d3PPjOX33vP DMEAAACAzRB8gbIoNVXq3Vv6xz/Mq7vz50sPP2x1VQAAAIBHEHyBsiY5WerWTdqzR6pYUVqxQurc 2eqqAAAAAI8h+AJlyc8/S9HR0r//LVWrJn35pdS8udVVAQAAAB5F8AXKil27zMcUnTol1a8vbdgg 3XST1VUBAAAAHseszkBZsGmTFBVlht5mzaSEBEIvAAAAygyCL2B3S5ZI994rnT0rdewoxcZK1atb XRUAAABQYgi+gJ29957Ut6+UkSE99JC0dq1UqZLVVQEAAAAliuAL2JFhSC+8II0aZS6PGCEtXiz5 +VlbFwAAAGABJrcC7CYjQxoyRFqwwFx++23p+eclh8PaugAAAACLEHwBOzl3Tvqf/5HWrZOcTumT T6RBg6yuCgAAALAUwRewi9OnzUmsduyQAgKkv//dXAYAAADKOIIvYAdHjkjR0dJPP0k33mhOYtWq ldVVAQAAAF6B4AuUdt9+K3XpIiUlSXXqSBs2SI0bW10VAAAA4DWY1RkozeLipLZtzdB7221SQgKh FwAAALgCwRcorZYvN4c3p6aa4Tc+XqpVy+qqAAAAAK9D8AVKo5kzzdmb09OlXr3M4c2VK1tdFQAA AOCVCL5AaWIY0rhx0lNPmT8/8YQ5e3NAgNWVAQAAAF6Lya2A0iIz0wy8H39sLo8fb4Zgh8PSsgAA AABvR/AFSoMLF6SHH5a++EIqV0768EPzai8AAACAayL4At7uzBmpRw9p61bJz09assS8rxcAAADA dSH4At7s2DHzGb379knBwdKqVVK7dlZXBQAAAJQqTG7lYUeOHNFjjz2mevXqKSAgQA0aNNBrr72m S5cu5bvdoEGD5HA43L5atWpVQlXDK+zfL919txl6a9Y0H1dE6AUAAAAKjCu+Hvbjjz8qOztbH330 kW666SZ9//33Gjp0qM6dO6dJkyblu22XLl00d+5c13L58uU9XS68RUKC1L27Ocy5USPzcUV161pd FQAAAFAqEXw9rEuXLurSpYtruX79+jpw4IBmzpx5zeDr5+enGjVqeLpEeJvVq6WHHjIntGrVSlqz RqpSxeqqAAAAgFKLoc4WSElJ0Y033njNdrGxsapWrZpuvvlmDR06VCdPnsy3fXp6ulJTU92+UMr8 7W/mxFUXLkj33it99RWhFwAAACgigm8JO3TokKZPn65hw4bl265r165atGiR/vnPf2ry5MlKTExU hw4dlJ6enuc2EydOVHBwsOsrLCysuMsvmKwsKTbWnIU4NtZcRu4MQ3rrLemxx8zf06BBUkyMVKGC 1ZUBAAAApZ7DMAzD6iJKo/Hjx+v111/Pt01iYqJatGjhWj5x4oQiIyMVGRmpTz75pEDHS0pKUt26 dbV06VL17t071zbp6eluwTg1NVVhYWFKSUlRpUqVCnS8IluxQhoxwpyVOEft2tL770t51F9mZWVJ I0dKM2aYy2PHSn/9q+RwWFsXAAAAikVqaqqCg4Ot+bsckrjHt9CGDx+uPn365NsmPDzc9fOJEyfU vn17RUREaPbs2QU+XmhoqOrWrauDBw/m2cbPz09+fn4F3nexW7FCeuAB8yrm5Y4fN9cvW0b4zZGe Lg0YIP3972bQnTpVevZZq6sCAAAAbIXgW0ghISEKCQm5rrbHjx9X+/bt1bx5c82dO1flyhV8hPlv v/2mo0ePKjQ0tMDblqisLPNKb24DCQzDDHcjR0o9e0pOZ8nX501SUsz7eTdvlnx9pYULzUmtAAAA ABQr7vH1sBMnTigqKkphYWGaNGmSTp06peTkZCUnJ7u1a9y4sWJiYiRJZ8+e1ejRo/X111/ryJEj io2N1X333aeQkBD16tXLirdx/eLj3Yc3X8kwpKNHzXZlWXKyFBVlht6gIGndOkIvAAAA4CFc8fWw jRs36ueff9bPP/+s2rVru712+e3VBw4cUEpKiiTJ6XTqu+++04IFC/THH38oNDRU7du312effaag oKASrb/AkpKKt11pNn68eVX71Vfd1x88KLVsKf3xh1Stmhl677zTkhIBAACAsoDg62GDBg3SoEGD rtnu8hAcEBCgDRs2eLAqD7reodjePmS7ODid0rhx5s854XfnTikyUjp/XrrhBikhQWrQwLoaAQAA gDKA4Ivi1batOXvz8eO53+frcJivt21b8rWVtJywmxN+77pL6t5dysgwg/+ePVL16tbVBwAAAJQR BF8UL6fTfGTRAw+YIffy8JvzeJ6pU8vOxFZXhl9Jql9f2rvXvLcXAAAAgMcxuRWKX+/e5iOLatVy X1+7dtl8lFGFCv/9uVw5af9+Qi8AAABQggi+8IzevaUjR8xZixcvNr8fPly2Qm92tjRmjPT//p+5 7HSa6955x9q6AAAAgDKGoc7wHKfTfGRPWZSRIQ0eLH36qbncubO0YYP05ptXT3gFAAAAwKMIvkBx O3vWvMc5Z2bu3r2l5cvNn6+855fwCwAAAHgcwRcoTqdOSffeKyUmSj4+Up8+0sKF7m1ywm5WVsnX BwAAAJRBBF+guBw+LEVHSwcPSlWqSGvXmo8wyg1XegEAAIASQ/AFisM330hdukjJyVLduuYw50aN rK4KAAAAgJjVGSi62FipXTsz9P7pT1JCAqEXAAAA8CIEX6Aoli0zhzenpprhd8sWqWZNq6sCAAAA cBmCL1BYH34oPfigdOmSOXPzhg1S5cpWVwUAAADgCgRfoKAMw5yc6umnzZ+HDZM+/1zy97e6MgAA AAC5YHIroCAyM6Unn5Q++cRcfuMN6ZVXJIfD2roAAAAA5IngC1yv8+elhx+WVq2SypWTZs6UHn/c 6qoAAAAAXAPBF7gev/8u3XefOWOzv7+0ZIl0//1WVwUAAADgOhB8gWs5etR8Ru8PP5iTV61eLbVp Y3VVAAAAAK4TwRfIzw8/mI8rOnZMqlVLWr9euu02q6sCAAAAUADM6gzkJSHBvLJ77JjUuLG5TOgF AAAASh2CL5Cb1auljh2lM2ekVq2krVulOnWsrgoAAABAIRB8gSvNmSP16iVdvCh17y794x9SlSpW VwUAAACgkAi+QA7DkP76V2nIECkrS3r0USkmRgoMtLoyAAAAAEVA8AUkM+g+84z0yivm8ksvmVd+ fZj/DQAAACjt+KseSE+X+veXli2THA7p/ffNEAwAAADAFgi+KNtSUqT775diY6Xy5aWFC6UHH7S6 KgAAAADFiOCLsispSeraVfrmGykoSFq5UurQweqqAAAAABQzgi/Kpp9+kqKjpSNHpOrVpXXrpGbN rK4KAAAAgAcwuRXKnsREqXVrM/TedJOUkEDoBQAAAGyM4IuyZcMGqX176fRpqXlzads2qX59q6sC AAAA4EEEX5Qdn34qde8unTsnde4sbd4sVatmdVUAAAAAPIzgi7Jh8mRpwAApM1Pq21das8ac0AoA AACA7RF8YW/Z2dLo0eaXJI0aZT6yqHx5a+sCAAAAUGKY1Rn2demSNHiwtGiRufy///vfAAwAAACg zCD4wp7OnpX+8hdp40bJx0f629/Moc4AAAAAyhyCL+zn1Cnp3nvNxxYFBkrLl0tdulhdFQAAAACL EHxhL4cPS9HR0sGDUkiItHat1LKl1VUBAAAAsBDBF/axd6/UtauUnCyFh5vP7L35ZqurAgAAAGAx ZnWGPWzeLEVGmqG3aVMpIYHQCwAAAEASwRd28Pe/m/fwpqZKUVHSli1SaKjVVQEAAADwEgRflG4f fCA99JD56KIHHpDWrZOCg62uCgAAAIAXIfiidDIM6ZVXpOHDzZ+fekpaulTy97e6MgAAAABehsmt UPpkZkrDhklz5pjLEyZIL78sORzW1gUAAADAKxF8UbqcPy/16SOtXi2VKyd99JE0ZIjVVQEAAADw YgRflB6//y7dd585Y7O/vzm0uWdPq6sCAAAA4OUIvigdjh6VoqOl/fulypWlNWuk1q2trgoAAABA KUDwhffbt898XNGxY1Lt2tL69dKtt1pdFQAAAIBSglmd4d22bZPatDFDb5Mm5jBnQi8AAACAAiD4 wnutWiV16iT98YcUESFt3SqFhVldFQAAAIBShuAL7/TJJ1KvXtLFi1L37tJXX0k33mh1VQAAAABK IYIvvIthSG++KQ0dKmVnS4MHSzExUmCg1ZUBAAAAKKUIvvAeWVnSM89Ir75qLr/8snnl14c52AAA AAAUHokC3uHiRWnAAGnZMsnhkKZNk4YPt7oqAAAAADZA8IX1UlKk+++XYmOl8uWlTz+V/ud/rK4K AAAAgE0QfGGtpCSpa1fpm2+koCDpiy+k9u2trgoAAACAjRB8YZ2ffpKio6UjR6QaNaR166Q77rC6 KgAAAAA2w+RWsMaOHVLr1mbobdhQSkgg9AIAAADwCIIvSt769eZw5tOnpRYtpG3bpHr1rK4KAAAA gE0RfFGyFi6U7rtPOn/eHOa8ebNUtarVVQEAAACwMYIvit/48dKECVevnzRJeuQRKTNT6t9fWrVK qlixxMsDAAAAULYwuRWKn9MpjRtn/vzqq1J2tjRmjDRlirnu7rul+fOlcnzuAgAAAMDzCL4ofq++ an4fN07KypJ+/llatMhcFx1t3uMLAAAAACWE4AvPuDz85vjLX6Rly6ypBwAAAECZ5TAMw7C6CHhG amqqgoODlZKSokqVKllTRPnyUkaG5OsrXbpkTQ0AAACAhbzi7/Iyjpss4TkTJpihNyf85jbhFQAA AAB4GMEXnjFhgjnM+Y03pPR08/u4cYRfAAAAACWOe3xR/C4PvTn3+l55z2/OMgAAAAB4GMEXxS8r yz305shZzsoq+ZoAAAAAlFlMbmVj3EQPAAAAWI+/y63HPb4AAAAAAFsj+AIAAAAAbI3gCwAAAACw NYIvAAAAAMDWCL4AAAAAAFsj+AIAAAAAbI3gCwAAAACwNYIvAAAAAMDWCL4AAAAAAFsj+AIAAAAA bI3gCwAAAACwNR+rCwAAAADgXbKzs3Xp0iWryyg1fH195XQ6rS4D+SD4AgAAAHC5dOmSDh8+rOzs bKtLKVUqV66sGjVqyOFwWF0KckHwBQAAACBJMgxDSUlJcjqdCgsLU7ly3Bl5LYZh6Pz58zp58qQk KTQ01OKKkBuCLwAAAABJUmZmps6fP6+aNWsqMDDQ6nJKjYCAAEnSyZMnVa1aNYY9eyE+wgEAAAAg ScrKypIklS9f3uJKSp+cDwoyMjIsrgS5IfgCAAAAcMN9qgXH78y7EXwBAAAAALZG8AUAAABQ5h05 ckQOh0N79+61uhR4AJNbAQAAAChWWVlSfLyUlCSFhkpt20rePt9TWFiYkpKSFBISYnUp8ACCLwAA AIBis2KFNGKEdOzYf9fVri29/77Uu7d1dV2L0+lUjRo1rC4DHsJQZwAAAADFYsUK6YEH3EOvJB0/ bq5fscJzx05LS1O/fv1UoUIFhYaG6r333lNUVJRGjhwpyZx8auXKlW7bVK5cWfPmzZN09VDnN954 QzVr1tRvv/3mat+jRw+1a9dO2dnZnnsj8AiCLwAAAIAiy8oyr/QaxtWv5awbOdJs5wmjRo3Stm3b tGrVKm3atEnx8fHavXt3off38ssvKzw8XEOGDJEkzZo1S1u2bNHChQtVrhwxqrRhqDMAAACAIouP v/pK7+UMQzp61GwXFVW8x05LS9P8+fO1ePFidezYUZI0d+5c1axZs9D7dDqd+vTTT3XHHXfoxRdf 1PTp0zV79mzVrVu3uMpGCSL4AgAAACiypKTibVcQ//73v5WRkaGWLVu61gUHB6tRo0ZF2m/9+vU1 adIkPfHEE3rooYfUr1+/opYKi3CNvgT06NFDderUkb+/v0JDQzVgwACdOHEi320Mw9D48eNVs2ZN BQQEKCoqSvv27SuhilGssrKk2FhpyRLzu6fG9wAAAFgoNLR42xWE8X9jqR0OR67rc14zrhiHnZGR cc19b9myRU6nU0eOHFFmZmYxVAsrEHxLQPv27fX555/rwIEDWr58uQ4dOqQHHngg323effddTZky RTNmzFBiYqJq1Kihzp07Ky0trYSqRrFYsUIKD5fat5f69jW/h4d7dmYHAAAAC7Rta87efEX2dHE4 pLAws11xa9CggXx9fbVjxw7XutTUVB08eNC1XLVqVSVddrn54MGDOn/+fL77/eyzz7RixQrFxsbq 6NGjmjBhQvEXjxJB8C0Bzz33nFq1aqW6devq7rvv1osvvqjt27fn+QmTYRiaOnWqXn75ZfXu3Vu3 3Xab5s+fr/Pnz2vx4sUlXD0KzcppDQEAAEqY02k+ski6OvzmLE+d6pnn+QYFBWngwIEaM2aMNm/e rH379mnw4MEqV66c6ypwhw4dNGPGDO3evVs7d+7UsGHD5Ovrm+c+jx07pieffFLvvPOO2rRpo3nz 5mnixInavn178b8BeBzBt4T9/vvvWrRoke6+++48T7TDhw8rOTlZ99xzj2udn5+fIiMjlZCQUFKl oiisntYQAADAAr17S8uWSbVqua+vXdtc78nn+E6ZMkURERHq3r27OnXqpNatW6tJkyby9/eXJE2e PFlhYWFq166d+vbtq9GjRyswMDDXfRmGoUGDBqlly5YaPny4JKlz584aPny4+vfvr7Nnz3rujcAj HMaVA93hES+88IJmzJih8+fPq1WrVlqzZo2qVKmSa9uEhAS1bt1ax48fd5uJ7vHHH9cvv/yiDRs2 5Lpdenq60tPTXcupqakKCwtTSkqKKlWqVLxvCPmLjTWHNV/L5s3FP60hAABAIV28eFGHDx9WvXr1 XIGxMLKyzNmbk5LMe3rbtvXMld78nDt3TrVq1dLkyZP12GOPefx4+f3uUlNTFRwczN/lFuKKbyGN Hz9eDocj36+dO3e62o8ZM0Z79uzRxo0b5XQ69cgjj1x1c/2Vcrs5/8p1l5s4caKCg4NdX2FhYUV7 kyg8K6c1BAAAsJjTaX62//DD5veSCL179uzRkiVLdOjQIe3evds1A3PPnj09f3B4PR5nVEjDhw9X nz598m0THh7u+jkkJEQhISG6+eab1aRJE4WFhWn79u2KiIi4arsaNWpIkpKTkxV62bR3J0+eVPXq 1fM83tixYzVq1CjXcs4VX1jAymkNAQAAyqhJkybpwIEDKl++vJo3b674+HiFhIRYXRa8AMG3kHKC bGHkXOm9fFjy5erVq6caNWpo06ZNatasmSTp0qVLiouL0zvvvJPnfv38/OTn51eomlDMcqY1PH48 9/t8HQ7zdU9MawgAAFAGNWvWTLt27bK6DHgphjp72I4dOzRjxgzt3btXv/zyizZv3qy+ffuqQYMG bld7GzdurJiYGEnmEOeRI0fqrbfeUkxMjL7//nsNGjRIgYGB6tu3r1VvBQVh5bSGAAAAANxwxdfD AgICtGLFCr322ms6d+6cQkND1aVLFy1dutTt6uyBAweUkpLiWn7++ed14cIFPfXUUzpz5ozuuusu bdy4UUFBQVa8DRRGzrSGI0a4P9Kodm0z9HpyWkMAAAAALszqbGPMHuclvGFaQwAAgOtQXLM6l0XM 6uzduOILeFrOtIYAAAAALME9vgAAAAAAWyP4AgAAAABsjeALAAAAwPbCw8M1depUq8uARbjHFwAA AEDx8sLJPRMTE1WhQgXXssPhUExMjO6//34Lq0JJIfgCAAAAKD4rVuT+OMf337f0cY5Vq1a17Niw HkOdAQAAABSPFSukBx5wD72SdPy4uX7FCo8dOi0tTf369VOFChUUGhqq9957T1FRURo5cqQk96HO 4eHhkqRevXrJ4XC4lmFfBF8AAAAARZeVZV7pNYyrX8tZN3Kk2c4DRo0apW3btmnVqlXatGmT4uPj tXv37lzbJiYmSpLmzp2rpKQk1zLsi6HOAAAAAIouPv7qK72XMwzp6FGzXVRUsR46LS1N8+fP1+LF i9WxY0dJZqitWbNmru1zhj1XrlxZNWrUKNZa4J244gsAAACg6JKSirddAfz73/9WRkaGWrZs6VoX HBysRo0aFfuxUDoRfAEAAAAUXWho8bYrAOP/hlI7HI5c1wMEXwAAAABF17atOXvzFeHTxeGQwsLM dsWsQYMG8vX11Y4dO1zrUlNTdfDgwTy38fX1VZaH7jeG9yH4AgAAACg6p9N8ZJF0dfjNWZ461SPP 8w0KCtLAgQM1ZswYbd68Wfv27dPgwYNVrly5q64C5wgPD9c//vEPJScn68yZM8VeE7wLwRcAAABA 8ejdW1q2TKpVy3197drmeg8+x3fKlCmKiIhQ9+7d1alTJ7Vu3VpNmjSRv79/ru0nT56sTZs2KSws TM2aNfNYXfAODoOB77aVmpqq4OBgpaSkqFKlSlaXAwAAAC938eJFHT58WPXq1cszMF6XrCxz9uak JPOe3rZtPXKlNz/nzp1TrVq1NHnyZD322GMeP15+vzv+LrcejzMCAAAAULyczmJ/ZNG17NmzRz/+ +KNatmyplJQUvfHGG5Kknj17lmgd8E4EXwAAAAC2MGnSJB04cEDly5dX8+bNFR8fr5CQEKvLghcg +AIAAAAo9Zo1a6Zdu3ZZXQa8FJNbAQAAAABsjeALAAAAALA1gi8AAAAANzz4peCys7OtLgH54B5f AAAAAJIkX19fORwOnTp1SlWrVpXD4bC6JK9nGIYuXbqkU6dOqVy5cipfvrzVJSEXBF8AAAAAkiSn 06natWvr2LFjOnLkiNXllCqBgYGqU6eOypVjUK03IvgCAAAAcKlYsaIaNmyojIwMq0spNZxOp3x8 fLhC7sUIvgAAAADcOJ1OOZ1Oq8sAig3X4QEAAAAAtkbwBQAAAADYGsEXAAAAAGBr3ONrYznPX0tN TbW4EgAAAKDsyvl7nOcjW4fga2NpaWmSpLCwMIsrAQAAAJCWlqbg4GCryyiTHAYfO9hWdna2Tpw4 oaCgIKZWt1hqaqrCwsJ09OhRVapUyepycJ3ot9KJfit96LPSiX4rfegz6xiGobS0NNWsWZPn/FqE K742Vq5cOdWuXdvqMnCZSpUq8T+aUoh+K53ot9KHPiud6LfShz6zBld6rcXHDQAAAAAAWyP4AgAA AABszTl+/PjxVhcBlAVOp1NRUVHy8eEOg9KEfiud6LfShz4rnei30oc+Q1nF5FYAAAAAAFtjqDMA AAAAwNYIvgAAAAAAWyP4AgAAAABsjeALAAAAALA1gi9QRBMnTtSf//xnBQUFqVq1arr//vt14MCB a24XFxen5s2by9/fX/Xr19esWbNKoFrkKEy/xcbGyuFwXPX1448/llDVmDlzppo2bapKlSqpUqVK ioiI0Lp16/LdhnPNWgXtM84z7zRx4kQ5HA6NHDky33acb97jevqM8w1lCcEXKKK4uDg9/fTT2r59 uzZt2qTMzEzdc889OnfuXJ7bHD58WN26dVPbtm21Z88evfTSS3r22We1fPnyEqy8bCtMv+U4cOCA kpKSXF8NGzYsgYohSbVr19bbb7+tnTt3aufOnerQoYN69uypffv25dqec816Be2zHJxn3iMxMVGz Z89W06ZN823H+eY9rrfPcnC+oUwwABSrkydPGpKMuLi4PNs8//zzRuPGjd3WPfHEE0arVq08XR7y cD39tnnzZkOScebMmRKsDNdyww03GJ988kmur3Gueaf8+ozzzLukpaUZDRs2NDZt2mRERkYaI0aM yLMt55t3KEifcb6hLOGKL1DMUlJSJEk33nhjnm2+/vpr3XPPPW7roqOjtXPnTmVkZHi0PuTuevot R7NmzRQaGqqOHTtq8+bNni4NecjKytLSpUt17tw5RURE5NqGc827XE+f5eA88w5PP/207r33XnXq 1OmabTnfvENB+iwH5xvKAh+rCwDsxDAMjRo1Sm3atNFtt92WZ7vk5GRVr17dbV316tWVmZmp06dP KzQ01NOl4jLX22+hoaGaPXu2mjdvrvT0dC1cuFAdO3ZUbGys2rVrV4IVl23fffedIiIidPHiRVWs WFExMTG65ZZbcm3LueYdCtJnnGfeY+nSpdq9e7cSExOvqz3nm/UK2mecbyhLCL5AMRo+fLi+/fZb bd269ZptHQ6H27JhGLmuh+ddb781atRIjRo1ci1HRETo6NGjmjRpEn8glKBGjRpp7969+uOPP7R8 +XINHDhQcXFxeQYpzjXrFaTPOM+8w9GjRzVixAht3LhR/v7+170d55t1CtNnnG8oSxjqDBSTZ555 RqtWrdLmzZtVu3btfNvWqFFDycnJbutOnjwpHx8fValSxZNl4goF6bfctGrVSgcPHvRAZchL+fLl ddNNN6lFixaaOHGibr/9dr3//vu5tuVc8w4F6bPccJ6VvF27dunkyZNq3ry5fHx85OPjo7i4OE2b Nk0+Pj7Kysq6ahvON2sVps9yw/kGu+KKL1BEhmHomWeeUUxMjGJjY1WvXr1rbhMREaHVq1e7rdu4 caNatGghX19fT5WKyxSm33KzZ88ehu9ZzDAMpaen5/oa55p3yq/PcsN5VvI6duyo7777zm3do48+ qsaNG+uFF16Q0+m8ahvON2sVps9yw/kGuyL4AkX09NNPa/Hixfriiy8UFBTk+rQ7ODhYAQEBkqSx Y8fq+PHjWrBggSRp2LBhmjFjhkaNGqWhQ4fq66+/1pw5c7RkyRLL3kdZU5h+mzp1qsLDw3Xrrbfq 0qVL+vTTT7V8+XIe1VGCXnrpJXXt2lVhYWFKS0vT0qVLFRsbq/Xr10viXPNGBe0zzjPvEBQUdNWc BxUqVFCVKlVc6znfvEth+ozzDWUJwRcoopkzZ0qSoqKi3NbPnTtXgwYNkiQlJSXp119/db1Wr149 ffnll3ruuef0wQcfqGbNmpo2bZr+8pe/lFTZZV5h+u3SpUsaPXq0jh8/roCAAN16661au3atunXr W93AuwAADwNJREFUVlJll3n/+c9/NGDAACUlJSk4OFhNmzbV+vXr1blzZ0mca96ooH3GeVZ6cL6V PpxvKMscRs6sAwAAAAAA2BCTWwEAAAAAbI3gCwAAAACwNYIvAAAAAMDWCL4AAAAAAFsj+AIAAAAA bI3gCwAAAACwNYIvAAAAAMDWCL4AAJQh8+bN08aNG60uwxaOHDmiN998UxcuXLC6FADANRB8AaCM ioqK0siRIz2y73bt2mnx4sUF3m706NF69tlni7WW8ePH64477ijWfVolPDxcU6dOLdI+WrVqpSFD hmjfvn1XvTZv3jxFRUV5vIbCcjgcWrlypSXHzk14eLhOnDihJ5980upSAADXQPAFABSrNWvWKDk5 WX369HGtCw8Pl8Ph0Pbt293ajhw50i1oPf/885o7d64OHz5cUuWWuPHjx8vhcGjYsGFu6/fu3SuH w6EjR47kuW1iYqIef/zxIh2/cePGWrx4sfr166e0tLQi7cuuBg0aJIfDobfffttt/cqVK+VwONzW TZ8+XWfOnNHHH39ckiUCAAqI4AsAKFbTpk3To48+qnLl3P8X4+/vrxdeeCHfbatVq6Z77rlHs2bN 8mSJRWYYhjIzMwu9vb+/v+bMmaOffvqpQNtVrVpVgYGBhT5ujjZt2mjv3r0KCgoq8r7syt/fX++8 847OnDmTbzun06kvvvhCQ4cOLaHKAACFQfAFAEiSzpw5o0ceeUQ33HCDAgMD1bVrVx08eNCtzccf f6ywsDAFBgaqV69emjJliipXrux6/fTp0/rqq6/Uo0ePq/b/xBNPaPv27fryyy/zraNHjx5asmRJ nq9Pnz5df/rTn1zLOVfhPvjgA9e66OhojR071m27hQsXKjw8XMHBwerTp4/b1U7DMPTuu++qfv36 CggI0O23365ly5a5Xo+NjZXD4dCGDRvUokUL+fn5KT4+XpK0evVqNW/eXP7+/qpfv75ef/31a4bi Ro0aqX379nrllVfybXelK4cZOxwOffLJJ+rVq5cCAwPVsGFDrVq1qkD7LIzz589r8ODBCgoKUp06 dTR79my3148fP66HHnpIN9xwg6pUqaKePXu6XclOTExU586dFRISouDgYEVGRmr37t1u+zh48KDa tWsnf39/3XLLLdq0aVOhag0PD9dbb72Vb7256dSpk2rUqKGJEycW6rgAAO9C8AUASDKHd+7cuVOr Vq3S119/LcMw1K1bN2VkZEiStm3bpmHDhmnEiBHau3evOnfurL/+9a9u+9i6dasCAwPVpEmTq/Yf Hh6uYcOGaezYscrOzs6zjpYtW+ro0aP65Zdfcn09KipK+/bt0+nTpyVJcXFxCgkJUVxcnCQpMzNT CQkJioyMdG1z6NAhrVy5UmvWrNGaNWsUFxfnNoz1lVde0dy5czVz5kzt27dPzz33nPr37+/aZ47n n39eEydO1P79+9W0aVNt2LBB/fv317PPPqsffvhBH330kebNm3fV7yU3b7/9tpYvX67ExMRrts3P 66+/rgcffFDffvutunXrpn79+un3338v0j6vZfLkyWrRooX27Nmjp556Sk8++aR+/PFHSWYobt++ vSpWrKgtW7Zo69atqlixorp06aJLly5JktLS0jRw4EDFx8dr+/btatiwobp16+b6MCI7O1u9e/eW 0+nU9u3bNWvWrGuOFihsvXlxOp166623NH36dB07dqzQxwYAeAkDAFAmRUZGGiNGjDAMwzB++ukn Q5Kxbds21+unT582AgICjM8//9wwDMN46KGHjHvvvddtH/369TOCg4Ndy++9955Rv379q45Vt25d 47333jNOnjxpBAUFGQsWLDAMwzBGjBhhREZGurVNSUkxJBmxsbG51p2dnW2EhIQYy5YtMwzDMO64 4w5j4sSJRrVq1QzDMIyEhATDx8fHSEtLMwzDMF577TUjMDDQSE1Nde1jzJgxxl133WUYhmGcPXvW 8Pf3NxISEtyO89hjjxkPP/ywYRiGsXnzZkOSsXLlSrc2bdu2Nd566y23dQsXLjRCQ0NzrT2nnttv v90wDMPo06eP0aFDB8MwDGPPnj2GJOPw4cN5bpvze8whyXjllVdcy2fPnjUcDoexbt26PPeRn7lz 517VH7nV0L9/f9dydna2Ua1aNWPmzJmGYRjGnDlzjEaNGhnZ2dmuNunp6UZAQICxYcOGXPeZmZlp BAUFGatXrzYMwzA2bNhgOJ1O4+jRo64269atMyQZMTExBXpP16o3NwMHDjR69uxpGIZhtGrVyhg8 eLBhGIYRExNj8KcTAJROXPEFAGj//v3y8fHRXXfd5VpXpUoVNWrUSPv375ckHThwQC1btnTb7srl CxcuyN/fP8/jVK1aVaNHj9a4ceNcV/+uFBAQIMm8cpgbh8Ohdu3aKTY2Vn/88Yf27dunYcOGKSsr S/v371dsbKzuvPNOVaxY0bVNeHi42/2soaGhOnnypCTphx9+0MWLF9W5c2dVrFjR9bVgwQIdOnTI 7dgtWrRwW961a5feeOMNt+2GDh2qpKSkPOu/3Jtvvqn4+PgiPV6oadOmrp8rVKigoKAg13vzlMuP 6XA4VKNGDdcxd+3apZ9//llBQUGu38mNN96oixcvun6fJ0+e1LBhw3TzzTcrODhYwcHBOnv2rH79 9VdJ5r/HOnXqqHbt2q7jRERE5FvTokWL3PohZyj6teq9lnfeeUfz58/XDz/8cF3tAQDeycfqAgAA 1jMMI8/1ObPYXv5zXtuFhIRcczKgUaNG6cMPP9SHH36Y6+s5w3SrVq2a5z6ioqI0e/ZsxcfH6/bb b1flypXVrl07xcXFKTY29qpH8vj6+rotOxwO13DrnO9r165VrVq13Nr5+fm5LVeoUMFtOTs7W6+/ /rp69+59VY35fQCQo0GDBho6dKhefPFFzZkz55rtc5Pfe/OUa/0+mzdvrkWLFl21XU6fDho0SKdO ndLUqVNVt25d+fn5KSIiwvVhSG7/Hq/8t3elHj16uH1wc3lfFuV31K5dO0VHR+ull17SoEGDrmsb AID3IfgCAHTLLbcoMzNT//rXv3T33XdLkn777Tf99NNPrvt1GzdurB07drhtt3PnTrflZs2aKTk5 WWfOnNENN9yQ67EqVqyoV199VePHj9d999131evff/+9fH19deutt+ZZb1RUlEaMGKFly5a5Qm5k ZKS++uorJSQkaMSIEQV6735+fvr111/d7gu+HnfeeacOHDigm266qUDbXW7cuHFq0KCBli5dWuh9 eJM777xTn332mapVq6ZKlSrl2iY+Pl4ffvihunXrJkk6evSo655tyeyTX3/9VSdOnFDNmjUlSV9/ /XW+xw0KCvLYLNVvv/227rjjDt18880e2T8AwPMY6gwAUMOGDdWzZ08NHTpUW7du1TfffKP+/fur Vq1a6tmzpyTpmWee0ZdffqkpU6bo4MGD+uijj7Ru3Tq3K3HNmjVT1apVtW3btnyP9/jjjys4ODjX 2Zvj4+PVtm1b15Dn3Nx2222qUqWKFi1a5Aq+UVFRWrlypS5cuKA2bdpc93sPCgrS6NGj9dxzz2n+ /Pk6dOiQ9uzZow8++EDz58/Pd9tx48ZpwYIFGj9+vPbt26f9+/frs88+K9BszdWrV9eoUaM0bdq0 697Gm/Xr108hISHq2bOn4uPjdfjwYcXFxWnEiBGuSaJuuukmLVy4UPv379e//vUv9evXz62/O3Xq pEaNGumRRx7RN998o/j4eL388stWvSX96U9/Ur9+/TR9+nTLagAAFA3BFwAgSZo7d66aN2+u7t27 KyIiQoZh6Msvv3QNE23durVmzZqlKVOm6Pbbb9f69ev13HPPuQ3pdTqdGjx4cK7DXC/n6+urCRMm 6OLFi1e9tmTJkms+E9XhcLiuzrZt21aSeR9ncHCwmjVrlueVxrxMmDBB48aN08SJE9WkSRNFR0dr 9erVqlevXr7bRUdHa82aNdq0aZP+/Oc/q1WrVpoyZYrq1q1boOOPGTPG7Z7k0iwwMFBbtmxRnTp1 1Lt3bzVp0kSDBw/WhQsXXP3yt7/9TWfOnFGzZs00YMAAPfvss6pWrZprH+XKlVNMTIzS09PVsmVL DRky5LpmyvakCRMm5HlLAADA+zkM/isOACikoUOH6scff3SbSOg///mPbr31Vu3atavAAXDt2rUa M2aMvv32W/n4cDdOSZs3b57mzZun2NhYq0sBAKBYccUXAHDdJk2apG+++UY///yzpk+frvnz52vg wIFubapXr645c+a4ZugtiHPnzmnu3LmEXgAAUKz4ywIAcN127Nihd999V2lpaapfv76mTZumIUOG XNUu577ggnrwwQeLWiIAAMBVGOoMAAAkSXv37tXevXt5bA8AwHYIvgAAAAAAW+MeXwAAAACArRF8 AQAAAAC2RvAFAAAAANgawRcAAAAAYGsEXwAAAACArRF8AQAAAAC2RvAFAAAAANgawRcAAAAAYGsE XwAAAACArRF8AQAAAAC2RvAFAAAAANgawRcAAAAAYGsEXwAAAACArRF8AQAAAAC2RvAFAAAAANga wRcAAAAAYGsEXwAAAACArRF8AQAAAAC2RvAFAAAAANgawRcAAAAAYGsEXwAAAACArRF8AQAAAAC2 RvAFAAAAANgawRcAAAAAYGsEXwAAAACArRF8AQAAAAC2RvAFAAAAANgawRcAAAAAYGsEXwAAAACA rRF8AQAAAAC2RvAFAAAAANgawRcAAAAAYGsEXwAAAACArRF8AQAAAAC2RvAFAAAAANgawRcAAAAA YGsEXwAAAACArRF8AQAAAAC2RvAFAAAAANgawRcAAAAAYGsEXwAAAACArRF8AQAAAAC2RvAFAAAA ANgawRcAAAAAYGsEXwAAAACArRF8AQAAAAC2RvAFAAAAANgawRcAAAAAYGsEXwAAAACArRF8AQAA AAC2RvAFAAAAANgawRcAAAAAYGsEXwAAAACArRF8AQAAAAC2RvAFAAAAANgawRcAAAAAYGsEXwAA AACArRF8AQAAAAC2RvAFAAAAANgawRcAAAAAYGsEXwAAAACArRF8AQAAAAC2RvAFAAAAANgawRcA AAAAYGsEXwAAAACArRF8AQAAAAC2RvAFAAAAANja/wcbIq5L6PnqfAAAAABJRU5ErkJggg== --=-=-=--