Path: csiph.com!usenet.pasdenom.info!gegeweb.org!newsfeed.kamp.net!newsfeed.kamp.net!newsfeed.fsmpi.rwth-aachen.de!rt.uk.eu.org!newsfeed.xs4all.nl!newsfeed2.news.xs4all.nl!xs4all!post.news.xs4all.nl!not-for-mail Return-Path: X-Original-To: python-list@python.org Delivered-To: python-list@mail.python.org X-Spam-Status: OK 0.006 X-Spam-Evidence: '*H*': 0.99; '*S*': 0.00; 'subject:help': 0.08; 'received:80.91': 0.09; 'received:80.91.229': 0.09; 'received:gmane.org': 0.09; 'received:list': 0.09; 'python': 0.11; 'empty.': 0.16; 'iteration': 0.16; 'modules.': 0.16; 'objects.': 0.16; 'received:80.91.229.3': 0.16; 'received:plane.gmane.org': 0.16; 'code.': 0.18; 'bit': 0.19; 'trying': 0.19; 'result.': 0.19; 'things.': 0.19; 'starts': 0.20; 'memory': 0.22; 'header:User- Agent:1': 0.23; 'interpret': 0.24; 'fairly': 0.24; "i've": 0.25; 'code:': 0.26; 'header:X-Complaints-To:1': 0.27; 'tried': 0.27; "i'm": 0.30; 'purely': 0.31; 'lists': 0.32; 'subject:with': 0.35; 'but': 0.35; 'there': 0.35; 'c++': 0.36; 'skip:o 20': 0.38; 'to:addr:python-list': 0.38; 'track': 0.38; 'to:addr:python.org': 0.39; 'received:org': 0.40; 'how': 0.40; 'skip:o 30': 0.61; 'more': 0.64; 'here': 0.66; 'caused': 0.69; 'image-size:2**15': 0.72; 'leak': 0.84; 'received:139': 0.84 X-Injected-Via-Gmane: http://gmane.org/ To: python-list@python.org From: Neal Becker Subject: help with memory leak Date: Tue, 27 May 2014 15:56:43 -0400 Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="nextPart18709298.xrLdHrxNQQ" X-Gmane-NNTP-Posting-Host: exa2-in-fw-01-epn.hns.com User-Agent: KNode/4.12.5 X-BeenThere: python-list@python.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: General discussion list for the Python programming language List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Newsgroups: comp.lang.python Message-ID: Lines: 545 NNTP-Posting-Host: 2001:888:2000:d::a6 X-Trace: 1401220622 news.xs4all.nl 2923 [2001:888:2000:d::a6]:33823 X-Complaints-To: abuse@xs4all.nl Xref: csiph.com comp.lang.python:72125 --nextPart18709298.xrLdHrxNQQ Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7Bit I'm trying to track down a memory leak in a fairly large code. It uses a lot of numpy, and a bit of c++-wrapped code. I don't yet know if the leak is purely python or is caused by the c++ modules. At each iteration of the main loop, I call gc.collect() If I then look at gc.garbage, it is empty. I've tried using objgraph. I don't know how to interpret the result. I don't know if this is the main leakage, but I see that each iteration there are more 'Burst' objects. If I look at backrefs to them using this code: for frame in count(1): ## main loop starts here gc.collect() objs = objgraph.by_type('Burst') print(objs) if len (objs) != 0: print(objs[0], gc.is_tracked (objs[0])) objgraph.show_backrefs(objs[0], max_depth=10, refcounts=True) I will get a graph like that attached A couple of strange things. The refcounts (9) of the Burst object don't match the number of arrows into it. There are 2 lists with 0 refs. Why weren't they collected? --nextPart18709298.xrLdHrxNQQ Content-Type: image/png; name="objgraph-fCFPiC.png" Content-Disposition: attachment; filename="objgraph-fCFPiC.png" Content-Transfer-Encoding: base64 iVBORw0KGgoAAAANSUhEUgAAAccAAAGTCAYAAABd49lTAAAABmJLR0QA/wD/AP+gvaeTAAAgAElE QVR4nOzdeXgN1//A8ffNJglSgtROE/v6RdHqIvZaE2vQKLWrrYqopXatfYmlxF6lUkFKhaqlVPIV qRAiokGpJQiJJbKI3Pn9kV/uVyTIcm8m9+bzep48j5x75sxnZiSfnDMz52gURVEQQgghRKqbZmpH IIQQQuQ1khyFEEKIl0hyFEIIIV5i8XLBtWvXOHjwoBqxCCGyqHLlyjRr1kztMIQwOZqXH8jp1asX 27dvVyseIUQWWFhYkJSUpHYYQpiam+l6jsnJybRt25bly5erEZAQIpP8/PwYNWqU2mEIYZLknqMQ QgjxEkmOQgghxEskOQohhBAvkeQohBBCvESSoxBCCPESSY5CCCHES9K9yiH0T6vVsmLFCuLj4zly 5Agffvgh33zzjdphZUmlSpWwtbWlT58+2Nra0r17dxwcHFAUhe3bt/Pjjz9y8+ZN6taty8SJE6le vToAgYGB/PXXX1y4cIGDBw/SsmVLVq9erfLRCCHE60lyzAVeXl4cOnSIPXv20Lx5c7Zt26Z2SNnS qFEjxo8fn6Zs+fLlnDt3jjZt2nDmzBn+/PNPevXqxf79+ylVqhSNGzemcePGxMfHU7t2bZUiF0KI rJHkmAt8fHxwcnICoGHDhjRs2FDliLLHwiLtf5f79+9z8eJF1q1bpyubO3cu69atw9vbmy+//FJX bmNjk2txCiFETsk9x1xw584dNBqN2mHo3enTpxk3blyasr59+wJw/fp1NUISQgi9kJ6jAe3fv5+j R4+SkJDAhQsX8PDwwNHRkYYNG7J7926SkpJwc3Nj8uTJlCpVig0bNqDValmzZg0JCQnExsZy/vx5 vvvuO5ycnIiPj+e3337jwIED2NjYMHbsWGbMmMGZM2dwdnZm/vz5BAcH4+npyfnz5/noo49YsGAB BQoU0MV07949li9fzpMnT7h69Sq1atViypQp2NraZvn42rRpk67Mzs4OgDJlymT/xAkhhMokORpQ 27Ztadu2Lbt27aJmzZrMnz+fe/fu4evry/bt26lVqxb79u3jo48+Ijw8HIBt27axc+dODh06BECr Vq0YN24cu3fvRqvVYm1tzaFDh6hYsSL+/v7MmTOHffv2MWfOHBRFoV69eixZsgQ/Pz9mzJhBs2bN 6Ny5MwBxcXG4ubkxbdo0nJ2defLkCS1atODWrVts3rxZL8ccHh6ORqOhY8eOemlPCCHUIMOquczB wYH+/fsD8PDhQyZNmsSkSZP44YcfgJSJ32vUqKGr7+joSEREBAAFCxakVatWunI3NzccHBx0PbiH Dx/i7u5OsWLFaNGiBQDBwcG6tjw9PTEzM8PZ2RmAwoUL06xZM/z9/bl8+bJejs/Hx4ehQ4dStWpV vbQnhBBqkJ6jClIfbKlWrRrm5uZpPuvbty99+/YlMTGRPXv2cO3aNRISEnSfp9Y3M/vf3zWlSpUC wNLSMl1ZdHS0riwwMJCEhAQ8PDx0Zffu3aN69ercvn2bSpUq5ei4goKCiIiIYNasWTlqRwgh1CbJ MY9RFAUfHx+OHz/OiBEjeOedd7hy5cprt8noYZ/UsuTkZF3Z5cuXqVu3LvPnz9dv0EB8fDxLlixh 2bJl6Z5qFUIIYyO/xfKYVatW4e3tzaFDh7CystJr23Z2dly6dInnz5+nS2CPHz/WPUyTVYqisGDB AmbMmCEP4gghTILcczQwRVGyVH/Tpk2ULVtWlxhTV3nPajsZqVmzJjExMaxZsyZNeXBwMHv37s12 u99//z0dO3akcuXKurJnz57xxx9/ZLtNIYRQk/QcDSz1fuGL9w2fPn0KpCSQl5mbmxMSEsLRo0eJ jIzk1q1bAPzxxx9UqFABBwcHIO1waXx8PJDyQM7L+3j+/LmubPjw4fzxxx8sWbKEv//+m0aNGhEZ GUlERASenp7ZOr7t27dz8uRJFEUhICBAF1tAQABTp07NVptCCKE26TkaUFhYGPPmzQNSXpj/8ccf OXfunC5pBAUFsWLFCu7evavbZtKkSdjY2LB06VIcHR3x8PDA3t6en376iaSkJF17wcHB/PLLLzx8 +JDFixcDcP78eXbu3Mnjx4919c6ePcu+ffsAqFu3Llu2bKFevXocOXKEtWvXotFoWLJkSZp3ITMr ICCAadOmERAQwJIlS3Rfnp6exMXFpXnqVgghjIlGeWm8rkePHsTGxrJ8+XK1YhJ5UKVKlXI8abg+ 2hD/4+fnx6hRo/Qy5C6ESOOm9BxFpr34WkhWxcbG6jESIYQwLLnnKDKlSZMmAMyfPx87Ozu6dOmi u//5OkFBQZw+fZrbt2/TpEkTqlWrZuhQhRAixyQ5ikxJncEnq4x5FRIhRP4lw6pCCCHESyQ5CiGE EC+R5ChEBhITE/NUHSFE7jJYcoyJiaFq1apcuHDBULvIlgEDBjBo0CC1w1CFVqvF09OTefPm0aZN G5kgPAN3795l2rRp6RZxVquOEEIdBkuOhw4dolSpUroXwa9du6b3fWSnTUVR0Gq1eo9FXwxxnlJ5 eXlx6NAhJkyYwOzZs3P0aoYpOnLkCJ6enmzdulU3bZ+adYQQ6jFYcty/fz9t27ZFo9Fw+vRpFi1a pNf2s9vmhg0bWL9+vV5j0RdDnKcX+fj46JayatiwIUuWLDHYvoxR8+bNmTFjRp6pI4RQj0GS48OH D/H39+eTTz7hypUrjBgxQq9/HRuiTbXlxjHduXMnw+WtxP9kZrmt3KwjhFCHQX46Dx8+TIkSJahV qxbz5s0jKiqKCxcu4OHhQZMmTXB1dQVSFtpdvnw5T5484erVq9SqVYspU6Zga2sLwKVLl/jpp58o VaoUBw8e5MmTJxw4cABvb+9Xtvk6iqJw9uxZ9u7dy9SpU0lMTOT333/nwIEDWFlZMWrUKKZNm0ZI SAgff/wx8+fPp0CBAgQHB7N7926SkpJwc3Nj8uTJlCpVig0bNgApScfT05OiRYty+fJlChQowMSJ EylVqhSPHz9m1qxZ7N69Wzd12pUrV1izZg27du2iZcuWumWqsnue3mT//v0cPXqUhIQEXfuOjo40 bNjwlcel1WpZs2YNCQkJxMbGcv78eb777jucnJyIj4/nt99+48CBA9jY2DB27FhmzJjBmTNncHZ2 Zv78+QQHB+Pp6cn58+f56KOPWLBgQZr5WzN7TMnJyQQGBrJnzx6Sk5Pp3bs3CxcuJCwsjNq1azNz 5kwqVqzIvXv38PDw4MSJE9SrV48pU6ZQt25dnj59yrx58/D392f9+vVUrFgRgH79+tGyZUvc3d0z dQ6FEPmLQXqOBw4coG3btpibmzNp0iQgZbmk+fPn637hx8XF4ebmRosWLVi6dClbt27l0KFDDBs2 TNfO+PHjGTRoEEOGDGHDhg0ULFgQMzOzV7b5OklJSaxfv57PPvtM90K7VqvFwsKCAwcOEBISwu7d uxk7dizdunXDz8+PHTt2EBUVxenTp9m+fTuXLl1i3759fPTRR7rVLqKioujQoQOdO3dm/PjxfP/9 92g0Gjp27Mi9e/ews7Nj5syZaWJxcnJKM6T2umPKzHl6k7Zt2+oWOE5tv0uXLq89rm3btrFz507G jBnDN998Q0xMjO7BEa1Wi7W1NYcOHeL8+fP4+/szZ84chg8fjq+vL+PHjyc8PJwlS5bw1Vdf4efn h5+fny6erBzTkydPuHjxIj4+PgQFBXHo0CFGjBjBkCFDOHnyJL169SI+Ph4HBwcWLFiApaUlNjY2 1K1bF4CCBQtSpkwZRo8erUuMALdu3eLOnTuZPodCiPxF78nx8ePH/Pnnn7Rp0+a19Tw9PTEzM8PZ 2RmAwoUL06xZM/z9/bl8+TIAV69eJTg4GIC33nqL4cOHZzsuS0tLBg4cSIMGDXRlNjY2tGrVCoAK FSowZswY6tSpQ79+/QAICQnBwcGB/v37AynDxZMmTWLSpEm6BOvp6UmJEiV0s8CYmZkxduxYHj58 qLunZ2Njky6ejMoykpnzlB1vOq7k5OQ0q2o4OjoSEREBpCSc1PPm6OiIm5sbDg4Oumv+8OFD3N3d KVasGC1atADQXcesHlORIkV016NChQqMHz+e999/n6FDhzJo0CCioqLYtWsXACVKlMDV1ZWAgAD+ +ecfXRvHjx/XxZtqz549jB07NtvnTwhh2vQ+rHrkyBGKFi1K/fr1X1svMDCQhIQEPDw8dGX37t2j evXq3L59m0qVKuHs7MzYsWMJDAzkyy+/pGXLljmOz9raOs335ubmALrFhQHdavapk2Wn3huqVq2a rn6qkydPUq5cuTRl5cuXp2LFipw6dSrH8WbmPGXX646rb9++9O3bl8TERPbs2cO1a9fSrEmZWt/M 7H9/X6U+7GNpaZmu7MUnY7N6TKn7evkPik6dOrFmzRouXryoK+vfvz87duzghx9+YNq0aVy/fp3y 5cun2zazf5wIIfInvSfH/fv306ZNmzS/NDNy+fJl6tatqxvuy8jChQuZOHEi27dvZ+/evcyaNYtO nTrpO+R0svLQSmRkJHZ2dunKS5YsSUhISI5jycx5MgRFUfDx8eH48eOMGDGCd955hytXrrx2m4zO W2rZi4sz6+uYypcvD6Rd0Lly5co0bdqUnTt36oZ0MzPsLoQQL9LrsGpsbCx//vknn3zyyRvr2tnZ cenSpTS/2FI9fvwYSOmBLF68mO+++w6Ar776it9++02fIedYmTJluHHjRoZr6jk6Oua4/cycJ0NY tWoVy5cvZ9GiRVStWlWvbevrmOLi4gDS9TQHDBhAXFwcPj4+nDlzRiY+F0JkWY6T49OnT3Uv1R85 coRChQql+2Wk0Wh4+vRpmrKaNWsSExPDmjVr0pQHBwezd+9eALZs2YJGo6F79+66BXKPHj36yjbV 8NFHHxEdHc3Zs2d1ZYqicO3aNd39NkgZGnxxaPHhw4cAaSYkyO55yoysLoi7adMmypYtqxtuTn3F RB8L62bmmF78f/UqoaGhWFhYpLu//f7771OtWjVWrVpFtWrVMhzFiIuLy9OTQQgh1JXj5Ni5c2cW L14MpDyl2rp163T3r0qWLMnFixeJjIzkt99+Q1EUhg8fjpmZGUuWLGH06NFs3bqVhQsXsmbNGrp1 6wbAzp07dcNxqfcB//Of/7yyzcxITVAxMTHA/3oqL/ZiHjx4AKC7x5aasJ49e5auvREjRuDg4MDK lSt1v2x///13bGxs0kxTV716dYKDg9m8eTNHjhxh1apVQMov+NOnT+foPGVG6rG8eN/wdcdlbm5O SEgIR48eZdu2bdy6dQuAP/74g6tXr+rux744XBofHw/8L/G/uI8Xz29mjunF/1epLl26pGs7Pj6e FStW8MUXX6S756vRaOjbty8xMTGvHFJ1dXVN1z78rzf6usSZm3WEEOownz59+vQXC3bs2MGzZ89o 165dphr4/fffqVOnDpUrV2by5MmMGTNGdy8oVaFChTh+/DiBgYF07dqV4sWLU7JkSRo1asQ///xD UFAQFy9epGrVqkyZMkX3sMTUqVP573//y+3bt/H29sbFxYXPP/8cjUaTYZuvExsbi4+Pjy7hRkdH U7ZsWVavXk1oaChRUVHY29tTrlw55syZQ3h4OPfv30er1bJr1y4uXbrEvXv30Gq1VKhQgUKFCgEp D/h06tSJU6dOsXv3bi5evMi1a9dYsmSJrg6k9JZOnz7N4cOH0Wg0jBs3jn379tGuXTuqV6/O22+/ ne3z9CZhYWF4eXlx7tw5oqKidPdIly5d+srjcnBwICAggLCwMDp16kSjRo3w9/fn/v37NG7cmO+/ /57Q0FBiYmIoWbIkJUuWxNPTk7Nnz/LgwQNKlixJ2bJlWbBgAefPn+fhw4eUKVOGKlWqZOqYUv9f pb6S4enpiZOTE0FBQezbt4+DBw/Svn173RO3LytevDinT59m6NChGX7+cvuQkvhXr17NpUuXiImJ wcLCAjs7O+zt7VWp8yYRERHs37+fl36EhRA591ijvNTl6tGjB7GxsSxfvlytoIRIp1KlSrpJFDJj 69atmJub07NnTwNHph4/Pz9GjRqll6FuIUQaN01m/qoX38l7lYULF2a6R2ws8utxv87z58/59ddf dTMYCSFEVplMcgwLC1M7BFXkh+NOvb+Z0b3RF3Xt2pVHjx5RqFAhXFxc5F1GIUS2yWLHIk+7efMm U6ZMASAoKIiFCxemmf3mRbVq1SI6Opp69erx2Wef5WaYQggTI/cchTBScs9RCIO5KT1HIYQQ4iWS HIUQQoiXSHIUQgghXiLJUQghhHhJhq9yBAUF0bdv39yORQiduLg4bGxssrRCSn5z//59tUMQwmSl S47du3dXIw5hYMePH6d69eqUKFFC7VDeSKvV4ufnR+HChXnvvfcoUKCA2iHlSaVLl6Z9+/ZqhyGE SUr3KocwTRqNBm9vb3r06KF2KJly5coVXF1diY6OxsfHh/fff1/tkIQQ+Ye8yiHyJicnJ/7880/q 1q1LixYt8PX1VTskIUQ+IslR5FlFihRh7969DBgwgG7durFy5Uq1QxJC5BMmM7eqME3m5uYsX76c smXLMnLkSP7991/mzp0rD+oIIQxKkqMwChMmTKB06dIMGDCAqKgo1q5dm25RbSGE0BdJjsJo9OnT h3LlyuHi4sK9e/fYsWOHrLwhhDAIuecojIqzszN+fn74+/vj4uJCXFyc2iEJIUyQJEdhdD744AOO HTtGSEgILi4uxMfHqx2SEMLESHIURqlOnTr8+eefhIaG0rp1a92CyEIIoQ+SHIXRqlKlCnv37uX8 +fP07NmTpKQktUMSQpgISY7CqL377rv4+flx9OhRvvjiC7XDEUKYCEmOwug1adKEvXv3smXLFqZN m6Z2OEIIEyDJUZiE5s2bs3z5cmbNmsVPP/2kdjhCCCMn7zkKkzFo0CD+/fdfBgwYQNWqValfv77a IQkhjJT0HIVJmTlzJu3bt6dDhw7cvn1b7XCEEEZKkqMwKRqNBi8vL2xtbenTpw/Pnz9XOyQhhBGS 5ChMTtGiRdm5cyf//e9/mT17ttrhCCGMkCRHYZLq1q3LsmXLmDVrFn/88Yfa4QghjIwkR2GyBg0a RNeuXenTpw8xMTFqhyOEMCKSHIVJW758OYmJiUycOFHtUIQQRkSSozBpb7/9NqtXr8bLy4ujR4+q HY4QwkhIchQmr0uXLnTq1IkRI0bw7NkztcMRQhgBSY4iX1i6dCnXr19n+fLlaocihDACkhxFvlCx YkUmT57MrFmzuH//vtrhCCHyOEmOIt8YM2YMRYoU4ZtvvlE7FCFEHifJUWSZ2q9FZHf/1tbWzJw5 k7Vr1xIaGqrnqIQQpkSSYz6k1WqZMWMGHh4e1KhRg9GjR79xm3PnzjF8+HCcnJz4/PPPcyFKw+zf 3d2dOnXq8PXXX+sxOiGEqZHkmA/Nnz+fX375hfnz57N69WqioqLeuE2dOnUYMGAAV69ezYUIDbd/ MzMzZsyYwb59+wgICNBTdEIIUyPJMR/asGED5cqVA+Djjz9m27ZtmdquXr16hgwr1/bfsWNHPv74 YyZMmKCX9oQQpkeSYz508+ZNNBpNlrfLzjb6pM/9T58+nRMnTnDkyBG9tSmEMB2y2HE+cvLkSfz8 /IiPjyc4OJh+/fpRrVq1HN9/u3nzJjNmzKBYsWJcvHgRa2trFi5cqOudpnr48CFTp07l6dOnxMXF 8fjxYxYvXkzVqlWBlHuhc+fOJT4+nsePH/PXX3+xfv16qlWr9sp9nz9/njVr1lCuXDl2797Nw4cP CQ8Pf2PMzZo1o1mzZnzzzTc0b948R8cvhDBBisgXAMXb21v3bxcXl2y38+K2kZGRSrFixZRjx44p iqIoycnJipubm2Jvb6/cvn1bVy8hIUGpXbu2snjxYkVRFEWr1SrVq1dXqlatqquzYsUKpXLlyrrv q1Sporz77ruv3X+9evWUa9euKYqiKNHR0enqv87Ro0cVQDly5EimtxFC5As3ZFhV5Mj06dMpWbIk H3/8MZDywMucOXOIjo5O8z7h3LlziYqKYuTIkUDKEOmcOXNo3bq1rk5ycnKa+4rVqlXjwoULr91/ eHi47sGaokWLMmXKlEzH7uzsTNOmTZk7d26mtxFC5A+SHEWOHD16NN3wqZOTE5UrV+bYsWO6sv37 91OjRg0sLP43kt+5c2c8PT11348aNQpvb28SEhJYv349ERERxMfHv3b/7dq1w93dnSFDhnD37l1c XFyyFP/kyZM5ePAggYGBWdpOCGHaJDmKHLlx4wbR0dHpysuWLUtkZKTu+wsXLpCQkPDathRFYcOG DXz22Wc0atSIKlWqvHH/W7Zswc3NDS8vLypXrpzpJ29TtWrVivr167N48eIsbSeEMG2SHEWOVKxY kX/++QdFUdJ9lvqgDUCRIkW4cOECcXFxaeokJSXpkuacOXOYMWMGW7ZsoXbt2pnav6WlJVu3bmXd unUAfPrpp+zatStLxzB+/Hh27tzJ5cuXs7SdEMJ0SXLMZzJKYjnRunVroqKiOHnyZJp9RERE0KlT J11Zo0aNePToUZr7e4qiMG3aNJKSkgBYtmwZ77zzDgUKFADQLS/1uphXrlyJRqNhwIAB+Pr6AvDr r79m6Ri6du1K2bJlWbFiRZa2E0KYLkmO+UzqPbw33cvLyMOHDwF4+vSprmzq1KmULl2a2bNno9Vq AfD19cXW1pZx48bp6k2bNg1LS0tmzZqFi4sLM2fOpHnz5vznP/+hcOHCAJibmxMYGMi+fftYvXo1 169fB8DPz49Lly5luP+NGzfy/PlzACpUqADAe++9l6XjsrS0ZPTo0axfv55Hjx5laVshhIlS81lZ kXsAZd68ecrw4cMVQLG1tVVWrFih3LlzJ1Pbh4WFKb1791YApWDBgsqiRYuUuLg4RVFSXufo1auX 0qFDB8XDw0MZOXKkEhMTk66NgIAA5YMPPlAKFSqk/Oc//1F8fX3TfL5161bF3t5eqV+/vnLkyBFl 7969SokSJZQOHToo/v7+Ge4fUD744ANl6tSpyieffKJMnz5dSU5OzvL5efz4sWJnZ6d71UQIka/d 0CiKnsfZRJ6k0Wjw9vamR48eaoeSZ40YMYJ9+/Zx+fJlzM3N1Q5HCKGemzKsKrC2tn7j188//6x2 mAY3dOhQrl+/zoEDB9QORQihMpk+TrzxFYv8olatWjg7O+Pl5UX79u3VDkcIoSLpOQrxgoEDB+Ln 55fmHU0hRP4jyVGIF3Tp0oVChQqxdetWtUMRQqhIkqMQL7C2tsbNzY1NmzapHYoQQkWSHIV4Sb9+ /bhw4QKnT59WOxQhhEokOQrxkvfee4/q1auzefNmtUMRQqhEkqMQGfjss8/YunUriYmJaocihFCB JEchMvDZZ5/x6NEj9u3bp3YoQggVSHIUIgOlS5emdevWMrQqRD4lyVGIV3B3d+fAgQM8ePBA7VCE ELlMkqMQr+Dq6oqFhQV79uxROxQhRC6T5CjEK9ja2vLJJ5/g4+OjdihCiFwmyVGI13B1deXw4cM8 fvxY7VCEELlIkqMJ+uabb9BoNGm+ANzc3NKUVapUSeVI876OHTuiKAp+fn5qhyKEyEWSHE1Q0aJF dQnxVTQaDcWKFculiIxXkSJF+Pjjj9m9e7faoQghcpEkRxPUu3fvNyZHMzMz+vXrlzsBGTlXV1f8 /PxkaS8h8hFJjiaoZMmSNG3a9I2r2Xfv3j2XIjJurq6uPH36lEOHDqkdihAil0hyNFHu7u4oipLh ZxYWFrRs2ZLixYvnclTGqUyZMjRs2FCGVoXIRyQ5mqhu3bphYWGR4WdarRZ3d/dcjsi4ubq68ssv v/D8+XO1QxFC5AJJjibKzs6Otm3bYmlpme4zS0tLXFxcVIjKeHXq1IkHDx5w6tQptUMRQuQCSY4m 7NNPP03X07G0tKRTp04ULlxYpaiMU82aNalYsSK//fab2qEIIXKBJEcT1rFjR2xtbdOUJSUl0bt3 b5UiMm7NmjXj8OHDaochhMgFkhxNmLW1NV26dMHKykpXZmdnR7t27VSMyni1aNGCwMBAmS1HiHxA kqOJ6927N8+ePQPAysqK7t27p0mWIvNatmxJcnIyJ06cUDsUIYSBSXI0cS1btsTe3h6AZ8+e0atX L5UjMl5vv/021atXl6FVIfIBSY4mzsLCgh49egBQrFgxmjZtqnJExq1FixaSHIXIBzJ+EU5k6MiR I1y+fFntMLKsUKFCANSpU4cNGzaoHE3WWVtb07NnzzwxHNyiRQtWrFhBVFQUJUqUUDscIYSBaJRX TaMi0rG0tJSXwFWyc+dOunTponYYPHr0iGLFirFt2zZdj1wIYXJuSs8xC54/f86GDRtwdXVVO5R8 xd7ePs/8UfLWW2/RoEEDDh8+LMlRCBMm9xyFyKJmzZpx/PhxtcMQQhiQJEchsujDDz/k0qVLPHjw QO1QhBAGIslRiCxq3LgxiqIQFBSkdihCCAOR5ChEFpUoUYJ33nmHwMBAtUMRQhiIJEchsqFx48aS HIUwYZIc85jHjx+zZ88eNm/erHYoGUpMTFQ7hDwhNTnKm1BCmCZJjnlIZGQkEyZMoF+/fvz++++6 8h49etCzZ0+D7tve3p5y5coxY8YMFi5cyJ07d9LFNm7cOIYOHZqm/MSJEyxcuJDPPvsMe3v7fLOI cuPGjYmOjubKlStqhyKEMAB5zzEPKVWqFEuWLMHb2ztNuaIoaLXabLV55coVnJycMlW3SZMmTJs2 LV35gQMHOHDgAD/88EO6FT0+/PBDPvzwQ+Lj4ylTpky2YjRG9erVw8rKisDAQCpVqqR2OEIIPZOe Yx5jbW2drmzHjh38/PPPWW4rMDCQ2bNnZ7q+paVlhuWffPIJCxYseO22NjY2WYrN2FlbW1OnTh25 7yiEiZLkaKIiIiLo27cvSUlJemnvVYkzP5OHcoQwXTKsqrK4uDhmzpyJVrg3BykAACAASURBVKvF wsKCYsWKpflcURT++usvdu7cydy5c9N89ujRI7777juePn1KfHw8T548Yc6cOTg6OrJ582bu3btH SEgIw4cP5+OPP8bNzS03D83kNWzYkHXr1pGUlCR/PAhhYqTnqKJnz57RoUMHKlasyPz58/n222+p UKFCms9XrlyJq6srXl5eabZNTEykffv2VKhQgeXLl7N27VquX79O7969MTMz0w2n1q1bl5UrV0pi NIB69eqRmJhIeHi42qEIIfRMkqOKNm7cyI0bN+jfv7+urGPHjrp/W1lZMWLECBo3bpxu26VLl3L/ /n0GDRoEgEajYcqUKTRv3tzwgQsAqlWrhqWlJefPn1c7FCGEnsmwqop27txJlSpV0qxTWKBAgXT1 bG1t05UdOnSIatWqYWHxv0vYoUMHOnToYJhgRTpWVlZUrlyZCxcuqB2KEELPpOeoor///jvbC/iG h4eTkJCg54hEVtWuXVt6jkKYIEmOKipQoADXr1/P1rZvvfUW4eHhxMfHpylPSkqSWWxyUa1atQgN DVU7DCGEnklyVFHt2rW5du0ap06d0pUlJycDvPEVjAYNGvD48WOWLl2qK1MUhblz5+q21Wg0xMbG GiBykapWrVpcu3aNx48fqx2KEEKPJDmqyMPDA41Gw8CBA/nzzz+JiopixYoVAJw9exZfX18A7t+/ D5Bm/UAPDw8sLS1ZsGABn376KfPnz8fFxYXatWtTqFAhAEqXLk1oaCi3bt1i7969OZoH9OnTp8D/ krdIUbt2bRRFISwsTO1QhBB6JMlRRY0aNWLXrl04ODjQu3dvevXqxUcffYSTkxMDBw7EycmJjRs3 EhISAsC0adO4du0aADVr1uTXX3+lcePGHD9+nH379jF06FBcXV117Xt4ePD8+XP69+9PpUqV0Gg0 2Yrz999/Z+zYsQCcOnWKVatWERERkbODNxHvvPMOBQsWlKFVIUyMPK2qsqZNm9K0adM0ZS8uolu7 dm0+//zzDLdt2LAh+/fvf2Xbffr0oU+fPjmOsVWrVrRq1YrVq1fnuC1TY2ZmRo0aNSQ5CmFipOco dFKHb7PjyZMneozEuNSqVUte5xDCxEjPUQDoeq/Tp0+nSJEi9OrVi7fffvuN2wUEBHDq1Clu3LhB 06ZNqVmzpqFDzXOqVq3K4cOH1Q5DCKFHkhwFALt3787Wdk2aNKFJkyZ6jsa4ODk5cfPmTRITEzOc xEEIYXxkWFWIHHJyckKr1eoelhJCGD9JjkLkUOpi0leuXFE5EiGEvsiwai7KzLBbZuqcOXOGn3/+ meLFi/Pxxx/TsGFDfYYpssjOzo7ixYtz9epVtUMRQuiJJMdcEBkZyaJFi3jw4AEbN27Mdp0nT54w YcIEIiIiWLt2LRUrVnztfnv06IGZmRnbt2/P6SGIN3B0dJSeoxAmRIZVDezAgQPMmzePDRs2vHJK uMzUefToEV26dOHGjRvs27fvjYkRUqaT02q16crll7j+OTk5yXkVwoRIcjSwTz75hAULFuS4zsSJ E4mIiGD16tWZXsljx44d/Pzzz2nKAgMDdQshC/2R5CiEaZHkmAssLS1zVCc4OJjt27fTrVs3ypQp k+04IiIi6Nu37xsnNRdZ5+TkxNWrVzPsqQshjI8kRyPg7e0NpCTQ9u3bU7ZsWZo3b86hQ4deuY2i KAQFBfH1118DoNVq2bx5M/fu3SMkJIThw4fr2gW4e/cuX331FQMHDsTZ2ZnRo0frJhuPi4vD29sb d3d3Bg0axPXr1+nZsydOTk4MGzYMrVbLyZMn6dy5MxUrVqR///5pls0KCwvDw8ODZcuW0apVKxo1 amSI06QqR0dHEhISuHPnjtqhCCH0QJKjETh+/Djm5ua0atWKX3/9lb/++osCBQrg5ubGyZMn09V/ 9uwZK1euxNXVFS8vLyBlDtDU4dS6deuycuVK3NzcgJQVN9q2bUvbtm1Zt24de/bsYf/+/bi7uwMp idXW1hY/Pz/Onj3LsWPHWLZsGePGjcPb25thw4YRGhqKl5cXkydPxtfXN82kAl988QUjR45k9OjR 7Nixg8KFCxv6lOW68uXLA3Djxg2VIxFC6IMkRyMQGRlJgwYNaN68ORqNhpIlS7JixQoURWHJkiXp 6ltZWTFixAgaN26cqfbnzZuHmZkZrVq1AlJeTWjdujXHjh3j0qVLFCpUiLZt2wJQqVIlPvvsM95+ +206duwIQExMDAMHDqREiRK6ei+uUfn333/rvi9SpIhuhQ9TUrp0aTQaDbdv31Y7FCGEHsirHEYg OTkZOzu7NGVOTk5UqlTptatB2NraZqp9f39/4uPjGT58uK7szp071KpVi5s3b1K1alUsLFL+q5ib m+vqpN7/fPEBodSyF9eebNWqFUOGDOHEiRNMmjSJdu3aZSouY2JlZUWxYsUkOQphIiQ5GoGKFSsS ExOTrrxMmTJplrfKrkuXLtGgQQNWrlyZpe0yWh8ytezFRZHXrFnDyJEj2bx5Mzt37mTx4sV069Yt Z0HnQaVLlyYyMlLtMIQQeiDDqkagSpUq3Lx5E0VR0pRrtVocHR1z3P5bb71FWFhYhk+xPnr0KMft W1hY4OXlhaenJwCDBw9m7969OW43ryldurT0HIUwEZIcjUC/fv24e/cu586d05UpisK1a9do3rx5 ltrSaDTExsamKatTpw4PHjxg2bJlacpPnTqFj49P9gP/f+vWrUOj0eDu7s6PP/4IwMGDB3Pcbl4j yVEI0yHJMRekvhLx4lBjVup8+OGHODs7M3v2bN3nx48fJyEhgVGjRr2yzdTFi1+8/1e6dGlCQ0O5 desWe/fuRVEUxo8fj5mZGd9++y0DBgxg/fr1zJw5k2XLlumeWE1dzPj58+e6tuLi4gCIjo7WlaUm 3hd7odu2bdNtV65cOQDefffdV8ZtrCQ5CmE6JDka2O+//657OvPUqVOsWrWKiIiILNXRaDT89NNP VK1alX79+jF58mR8fHw4evQo9vb26fb55MkTNm7cSEhICADTpk3TLafk4eHB8+fP6d+/P5UqVUKj 0VC/fn18fX1p2LAhv/32G8uXL8fMzAwvLy8KFCjA7du3mTZtmi6+HTt2EB0dzZw5c4CUidC3bdvG o0ePdPVOnz6te50jNDSUjh07MnfuXDw8PJgwYQJ9+vTR1ynOM0qVKiX3HIUwERrl5RtZ4pU0Gg0b NmzA1dVV7VDyFXt7e7y9venRo4faobyWr68vXbp0IT4+XhY9FsK43ZSeoxB6Urp0aRRFkVlyhDAB khyF0BMHBwcA7t27p3IkQoickuQohJ6k3v998QElIYRxkuQohJ7Y2dlhZWWV5ulgIYRxkuQohB4V LVpUeo5CmABJjkLokb29vSRHIUyAJEch9EiSoxCmQZKjEHpUrFgxSY5CmABZlSOLFi5cyObNm9UO Q+RR9vb28iqHECZAkmMWfPXVV3l6pffTp09TuHBhqlSpkq3tL168iLm5eba3N5SePXtmeuFmtdnb 2xMeHq52GEKIHJLkmAWLFi1SO4TXKlmyJCNGjGDMmDHZ2n7q1Kl8//33/Pe//8XGxkbP0eUPcs9R CNMg9xxNxL///svdu3dp2LBhttsYMWIEsbGxbN26VY+R5S92dna6FUyEEMZLkqOJCAoKwsLCgvr1 62e7DQcHB3r06MHSpUvTLawsMsfW1jbdeplCCOMjydFEBAUFUaNGDWxtbXPUzrhx4wgLC+PQoUN6 iix/KViwoG6dSyGE8ZLkaCKCgoJyNKSaqnbt2jRt2pRly5bpIar8p2DBgiQnJ5OQkKB2KEKIHJDk aAK0Wi2nT5/WS3IE+PLLL/Hz8+PixYt6aS8/KViwIABPnz5VORIhRE5IcjQBf//9N48ePdJbcuzY sSNOTk6sWLFCL+3lJ5IchTANkhxNQFBQEDY2NtSuXVsv7ZmZmTFq1Cg2bdokK0xkkSRHIUyDJEcT EBQURN26dbG0tNRbm59//jlWVlasX79eb23mB5IchTANkhxNgL4exnlRoUKFGDhwIJ6eniQlJem1 bVMmyVEI0yDJ0cg9e/aMs2fP6j05AgwbNow7d+6wa9cuvbdtqlJnFpLXOYQwbpIcjVx4eDgJCQk5 evn/VRwdHfn000+ZMWMGWq1W7+2bIisrKyDljxYhhPGS5GjkQkJCsLa2pmrVqgZpf+LEiVy6dIk9 e/YYpH1Tk3rfV4aihTBukhyN3Pnz56levToWFoaZQ75atWp06tSJb7/91iDtmxozMzPMzMx4/vy5 2qEIIXJAkqORO3/+PHXq1DHoPiZNmkRQUJBMKZdJlpaW0nMUwshJcjRy586d09v7ja/SsGFDmjZt yoIFCwy6H1NhYWEhyVEIIyfJ0Yg9ePCA27dvGzw5AkyePJmDBw8SEBBg8H0ZO0tLSxlWFcLISXI0 YhcuXACgZs2aBt9Xq1atcHZ2ZsKECQbfl7GTYVUhjJ8kRyMWFhZGkSJFKFOmTK7sb9asWZw4cYLD hw/nyv6MlQyrCmH8JDkasfDwcKpVq5Zr+/vwww9p3rw5U6dOzbV9GiMZVhXC+ElyNGKXLl2iSpUq ubrPWbNmERAQwG+//Zar+zUmGo0GRVHUDkMIkQOSHI3YpUuXDPby/6s0adIEZ2dnZs6cmav7NSaK oqDRaNQOQwiRA5IcjVRCQgLXr1/P9eQIMHv2bAICAvjll19yfd/GQKvVYmYmP1pCGDP5CTZSly9f RqvVqpIcP/jgA7p3746Hh4c8eJIB6TkKYfwkORqpv//+GzMzM5ycnFTZ/9y5c7l27Rrr1q1TZf95 mfQchTB+8hNspC5fvkyZMmV0SyTlNkdHRwYPHsz06dN58uSJ3tt/+vQpu3fvZu3atXpv29Ck5yiE 8ZPkaKSuXr2Ko6OjqjFMnTqVhIQEFi9erNd279+/z6hRo+jSpQv79u17Y/22bdvSoUMHvcaQE9Jz FML4yU+wkfrnn3945513VI2hRIkSjBkzhsWLF3Pv3j29tVu8eHFWrVqV6fqKouSpVyek5yiE8ZPk aKTyQnIEGDduHIULF2bSpEl6bbdAgQKZrnvgwIFM9TBzi/QchTB+8hNshLRaLdevX88TybFQoUIs WrSIDRs2cOLECbXDyROSk5MNtr6mECJ3SHI0Qrdv3+bZs2d5IjkCuLm58fHHHzN69Gi0Wm2mtrl5 8yaDBg3i66+/xsXFBTc3N27cuJFhXR8fH2rUqMFbb72Fq6srkZGRus+SkpI4ePAgAwcOTLPN+fPn GTFiBPPmzeO9997L1Wn2EhISsLKyyrX9CSEMQBFG5/jx4wqg3Lx5U+1QdM6cOaOYm5sr69evf2Pd yMhIpVixYsqxY8cURVGU5ORkxc3NTbG3t1du376tqwcoFSpUUAYMGKDs2rVL+fLLLxVAqVy5spKY mKjExcUpc+fOVQoXLqy8/F+5Xr16yrVr1xRFUZTo6Gjl3Xff1ePRvp6ZmZny008/5dr+hBB6d0OS oxH68ccfFSsrKyU5OVntUNIYMmSI4uDgoMTExLyxXs2aNdOUXb58WQGUAQMG6MoA5aOPPkq3LaBs 3LhRV9amTZt0ydHGxkbZtm2b7ntfX9+sHk62JCUlKYCya9euXNmfEMIgbsiwqhG6desWJUuWzHMP fcyYMYOEhARmzZr12npHjx6lXLlyacqcnJyoXLkyx44dS1Nub2+f5vsBAwYAEBQUpCuztrZOt492 7drh7u7OkCFDuHv3Li4uLlk6luxKTEwEsvZAkRAi78lbv11FpkRGRlK6dGm1w0jn7bffZtGiRSxd uhR/f/9X1rtx4wbR0dHpysuWLZvmfmJGKleuDEBcXNxr623ZsgU3Nze8vLyoXLky27Zty8QR5Fxq cpR7jkIYN0mORigyMpJSpUqpHUaGBgwYQPPmzRk6dCjPnj3LsE7FihX5559/Mnw38U1zxaa+P1in Tp3X1rO0tGTr1q266e0+/fRTdu3alZlDyBHpOQphGiQ5GqHbt2/nyZ4jpCSvVatWceXKFebPn59h ndatWxMVFcXJkyd1ZYqiEBERQadOnV7bfmhoKLa2tnTt2vW19VauXIlGo2HAgAH4+voC8Ouvv2bx aLIu9Q8CSY5CGDdJjkYoL/ccIWXoc9q0acyePZuLFy+m+3zq1KmULl2a2bNn61798PX1xdbWlnHj xunqWVhYcP/+fV2dhIQEZs+ezeLFiylfvvxrY9i4cSPPnz8HoEKFCgC89957ejm+15FhVSFMg7yp bIRu376dp5MjwFdffcXWrVsZOXIkv//+e5rp1Ozt7Tl9+jRfffUVLi4u1KhRg/j4eAIDAylYsKCu nq+vL2vXrqVly5aUL18eKysrPDw8aNas2Rv3HxISgrOzMy1atODUqVNMnz493buQhvD06VMAbG1t Db4vIYThSHI0Mo8fPyYuLi7PJ0dLS0s2bNhAkyZNWLFiBSNHjkzzecmSJd/4kEz79u1p3779G/el KApFixZNV6aG1ORYqFAhVfYvhNAPGVY1MlFRUQA4ODioHMmbvfvuu8yaNYvx48dz7tw5g+3n2bNn qq1r+bLY2FhAkqMQxk6So5F58OABkP79v7xq/PjxNGnShN69e5OQkKC3drt3786mTZuAlHuwgwcP 1lvbOREbG4tGo0kzPCyEMD6SHI1MTEwMYDzJ0czMjB9++IHbt28zffp0vbVbvHhxPDw86NmzJz16 9NBNDqC22NhYbGxsMDc3VzsUIUQOyD1HIxMdHY2FhQV2dnZqh5JpZcuWZdGiRQwePBgXFxfef//9 HLf5/fff8/333+shOv16+vSpDKkKYQKk52hkoqOjKVq0qNEtpvv555/j4uJC7969dUPDpig2NlaS oxAmQJKjkUlNjsZo06ZN2NjY0L17d907iKZGeo5CmAZJjkYmJibGaO43vqxQoUL8/PPPBAYGMmXK FLXDMYjHjx9TuHBhtcMQQuSQJEcjExMTY7Q9R4BatWqxcOFCFixYkCvTueW2hw8fGvX1EUKkkORo ZGJjY42+ZzJs2DA+/fRTPv30U8LCwtQOR69iYmIoUqSI2mEIIXJIkqORiY+Px8bGRu0wcmz9+vU0 aNCAtm3bcvfuXbXD0RvpOQphGiQ5GhlTSY6Wlpbs2LEDKysrunTpopuw29hJz1EI0yDJ0cjExcWZ RHIEKFasGD4+Ppw7d46hQ4eqNh+qPklyFMI0SHI0MqbSc0xVt25dvL292bp1KxMnTlQ7nByTYVUh TIPMkGNkTC05ArRr146ff/6Zbt26UaRIEb7++mu1Q8qUbdu24e/vT4kSJShRogTFixfn6dOnxMbG cv/+fYoXL652iEKIbJLkaGTi4uJMcq1AV1dXli9fzvDhwylRokSGc6U+fPiQt956K8/MDnTw4EF+ +OEHrKys0Gq1JCUlATBq1ChGjRqFubk5RYoUoXXr1m9cnksIkbfIsKqRSUhIwNraWu0wDGLYsGFM mjSJIUOGsGvXrjSfHT58mJIlS+ap+VSdnZ3RaDQkJibqEuOLkpOTiYmJIT4+XoXohBA5IcnRyCQn J5v0ig+zZs3i888/x93dnRMnTgBw/Phx2rdvT2JiItOnT88zT7a2aNECrVb72jparZZJkyblUkRC CH2R5CjyFI1Gw+rVq2nbti0dOnRg/fr1tG7dWtczi46OZvXq1SpHmaJcuXKULVv2lZ9bWFjQpEkT GjZsmItRCSH0QZKjyHPMzc354YcfKF++PMOGDSMpKUnXQ0tOTmbWrFl5ZqiyTZs2WFpaZvjZ8+fP 8fDwyOWIhBD6IMlR5ElhYWFcvXoVrVabbujy4cOHrF+/XqXI0nJ2diY5OTlduZmZGZUqVaJTp04q RCWEyClJjiLPCQ4Opnnz5iQkJGSYeJKTk5kxY0ae6D02b978lfcdx48fn2eerBVCZI0kR5Gn/PPP P7Rp04b4+PgME2Oq6OhotmzZkouRZax06dJUqFAhXbm9vT19+/ZVISIhhD5IchR5SnBwMDExMW/s cSmKwowZMzJ8hSK3tW7dOs19RwsLC0aOHEmBAgVUjEoIkROSHEWe0rVrV27dusXkyZMpVKgQZmYZ /xdVFIU7d+6wYcOGXI4wvZfvO1pYWDBixAgVIxJC5JQkRyNToECBPPOen6G8/fbbTJ8+ndu3b7N4 8WIcHBwwMzNL15vUarXMmDGDZ8+eqRRpimbNmunuO1paWtK/f3/s7e1VjUkIkTOSHI2Mra0tcXFx aoeRKwoXLszo0aO5ceMGGzduxNHRESDNJAh3795l8+bNaoUIQKlSpXT3HZ8/f86oUaNUjUcIkXOS HI1MwYIF801yTGVlZcVnn33GxYsX2bJlC1WrVgVSemlarZbZs2fz/PlzVWP85JNPAHBxcdHFJ4Qw XjLxuJHJTz1HgGfPnrF9+3YSEhJ0ZaNGjeLChQvs37+fy5cv8++//zJ48GDee+891eJMTc6VKlXC y8tLtThSWVtb07NnT6ysrAzS/pEjR7h8+bJB2jZlrVu3pmLFigZp+9q1axw8eNAgbZuyypUr06xZ s/QfKMKoNGvWTBk2bJjaYeSanTt3KoB8ZeNr586dBrsuFhYWqh+fMX716tXLYNekZ8+eqh+fMX5Z WFhkdDpvSM/RyOS3nmNqjyw6OlrlSIyLvb29QYeanz9/zoYNG3B1dTXYPkzN559/btBrkpycjIuL Cxs3bjTYPkyNr68v/fv3z/AzuedoZPLjPUchhMhtkhyNjI2NjSRHIYQwMEmORsbe3p4HDx6oHYYQ Qpg0SY5GpkSJEkRFRakdhhBCmDRJjkamRIkS3Lt3T+0whBDCpElyNDIODg48efIkTyzXZEwePnyo dggiA3Jd8h65JinkVQ4j4+DgAEBUVBTly5dXOZq87cKFC2zatInDhw9To0YNfvzxRwB69OiBmZkZ 27dv1/s+7e3tKViwIAMHDqRgwYK4u7tTsmRJFEVh8+bNrFu3jn///ZcGDRowa9YsatWqBcCJEyc4 efIk586d49dff6Vdu3a6eE1NXroukPJaytq1awkKCsLBwYF///2XMWPG0LBhw3xzXeSapCfJ0ciU KFECgHv37klyfIOaNWvi7u7O+vXrqVGjhq5cUZRXLlD8KleuXMHJySlTdZs0acK0adPSlM2fP5/g 4GA6duxIUFAQR44coX379gQEBFCmTBk+/PBDPvzwQ+Lj4ylTpkyWYjM2eem6AEyePJnQ0FB++eUX LCws+Oeff2jfvj2bNm3KN9dFrkl6MqxqZF7sOYo3q1OnTrqyHTt28PPPP2e6jcDAQGbPnp3p+i+u 7Qgp1yo0NBRvb28mTJiAj48PI0aM4MmTJ/zwww9p6trY2GR6P8YsL1wXgIsXL7J27VomTZqEhUVK X+Gdd97B1dWViRMn6urlh+si1yQtSY5GpnDhwlhbW0tyzKQ3LZr8JhEREfTt2zdHiyqfPHmSqVOn pikbMmQIAFevXs1RfMYqL1wXgB9//BEbGxsaNmyYprxp06acOXOG06dP56h9YyLXJC1JjkaodOnS 3LhxQ+0wjJKiKAQFBfH111+nKQ8LC8PDw4Nly5bRqlUrGjVqhFarZfPmzdy7d4+QkBCGDx+Ot7d3 lvfZsWNHKleunKbsrbfeApCh8f+nxnWBlHu977zzTroJ2lOvV2BgYPYOyATk92siydEIVapUSVZE yIZnz56xcuVKXF1d062c8cUXXzBy5EhGjx7Njh07KFy4MGZmZrohorp167Jy5Urc3Nz0EsuFCxfQ aDR07dpVL+0ZMzWvy7///kuRIkXSlaeW5dc/QuWaSHI0Sk5OTvl2OC4nrKysGDFiBI0bN0732d9/ /82pU6eAlB/CsWPHGjSWrVu3MmbMmDQPP+RXal6X+Ph4ChYsmK68cOHCANy5c0ev+zMWck0kORol R0dHrly5onYYRsvW1jZdWatWrRgyZAhjxowhKiqKdu3aGWz///3vfwkPD2fChAkG24cxUuO6lCtX LsN3hp8+fQqAnZ2dXvdnbPLzNZHkaIScnJy4ffu2TASgR2vWrKFz585s3ryZBg0a4OPjY5D9xMXF 8e2337J+/foMn9QTaRn6ujg6Omb40ntqmaEWJjZm+eWaSHI0Qk5OTiiKwj///PPGuk+fPmX37t2s Xbs2FyIzXhYWFnh5eeHp6QnA4MGD2bt3r173oSgKM2fOZMGCBZQrV06vbZsqQ1+XGjVqcP369XTr LKb+bH3wwQd625epyC/XRJKjEXJ0dESj0bxxaPX+/fuMGjWKLl26sG/fvlyKzjitW7cOjUaDu7u7 braNgwcPAimPuMfGxuZ4H4sXL6Zr165Uq1ZNV5aYmMjvv/+e47ZNlaGvS/fu3Xny5AnBwcFpyo8d O0bVqlXTvU4g8s81keRohAoVKkSJEiXe+FBO8eLFWbVqVS5F9WoRERGq7fvRo0fA/+5XQMofDUCa pb+2bdum+0s1tVf37rvvAimvzoSGhnLr1i327t2LoihZjmPz5s2cOHGCY8eOsXDhQhYuXMi8efPo 3Lkzb7/9dvYOzojlletSo0YNBg8ezNy5c3Uzwdy8eZNt27Yxf/78HL/7Z0zkmqQl08cZqapVqxIW FvbGegUKFMiFaF7N39+fZcuWZWmWDX35+++/WbhwIQBBQUHMmzcPOzs7QkJCAJg2bRrjxo2jYsWK hIaG0rFjR5o2bcrp06eZMGECffr0AcDDw4NvvvmG/v37s3Tp0iz/cB47doxx48aRnJzMsWPH0nxW p06dDGcmMWV55bqk+u6771i9ejUjR46kSJEi3Lp1i82bN/P+++/r54CNgFyT9CQ5Gqn69etz4sQJ tcN4rfDwcLp165bh4+C5oUqVKnh5eaV7T2vYsGHp6kZHR7+ynT59+uh++LOjadOmMqPRC/LKdUml 0Wgy3Hd+ItckPRlWNVL16tUjNDQ0S1M1+fj4UKNGDd566y1cXV2JJsnmpwAAIABJREFUjIzk4cOH 9O3bF41Gg6urK5CS1Pr166crUxQFf39/hgwZwueff05AQAC1atWibdu2AJw/f54RI0Ywb9483nvv PapVq4ZWq2Xt2rXcuXOH4OBg+vXrx5YtWwxyLvKa1KGo7Hjy5IkeIxEvkuuS9+TlayI9RyNVr149 EhMTCQ8Pp3bt2m+sf/bsWYoXL86cOXM4fvw4S5cuJSwsjNDQUFatWpVmAuxq1aqxcuVKNm/eDEBk ZCT+/v54eXnRoEEDvL29adOmjW7IpW/fvuzevZsKFSowePBgWrdujZmZGYsWLWLx4sXUr1+fTZs2 GeQ85DVNmzYFYPr06RQpUoRevXpl6p5iQEAAp06d4saNGzRt2pSaNWsaOtR8Ra5L3pPXr4kkRyNV o0YNrK2tOXPmTKaSY/ny5Vm3bh0AnTt3Jj4+njVr1rBt2zb69euXrv6LM1SULl2aMWPGMGHCBKKj o1m8eDHm5ua6z8PDwwkICKBChQoULVqUKVOm5PwAjdTu3buztV2TJk1o0qSJnqMRqeS65D15/ZrI sKqRsrCwoGbNmpw5cyZT9e3t7dN8P2DAACDl5ntmpL6wXqdOnTSJEaBdu3a4u7szZMgQ7t69i4uL S6baFEKIvEqSoxGrX79+ppPjy1JnuI+Li8txHFu2bMHNzQ0vLy8qV67Mtm3bctymEEKoSZKjEatb ty7nzp3L1rtEqY9Y6+M1AktLS7Zu3aobtv3000/ZtWtXjtvVlzNnzjBx4kQWLVqU6Z6yMKzExES1 QxCvoSgK586dUzsMVUlyNGKNGjUiJiaG8PDwLG8bGhqKra2tbskkCwuLNK8bpD6unZyc/Ma2Vq5c iUajYcCAAfj6+gLw66+/AilJWK0n/Z48ecIXX3yBh4cHQ4YMYezYsa+dXaNHjx707NkzFyPMfyIj Ixk3bhxDhw7NVH25Jrnjyy+/xN7eXvdVrFgx3c9wRvLDdZEHcoxY/fr1KVq0KIcOHaJ69eqvrGdh YcH9+/fRarWYmZmRkJDA7NmzWbx4sW6x3bp16xIQEICnpyeOjo4cPXoUgODgYPz9/XU9zIz+4t+4 cSPDhw/HwsKCChUqAPDee+8BULZsWUJCQrhx4wZBQUF07tw5V2a4ePToEd26dcPa2pp9+/alWzg1 I4qi6GbkeNGVK1dwcnIyRJj5yoEDBzhw4AA//PBDpldykGtiePfv3+fcuXOMHj1aV2ZhYcHgwYNf uU1+uC6SHI2Yubk5TZs25ejRo4wcOfKV9Xx9fVm7di0tW7akfPnyWFlZ4eHhQbNmzXR1Vq9eTd++ fZk6dSpdunRhyZIl7Nq1i65du3L37l3dC7nHjx9n1qxZ9O/fnzJlygAQEhKCs7MzLVq04NSpU0yf Pp2BAwcCMHXqVMaNG6e7J5lbUz9NnDiRiIgI/P39M5UYAXbs2JGuLDAwkNWrV7Nx40Z9h5jvfPLJ J7Ro0SLNa0NvItfE8NavX8+kSZNo2bJlprfJD9dFkqORa9asGTNmzND1CjPSvn172rdv/9p23n33 XS5cuJCm7MVVP7p06aKbZPhlr7vnOXDgQF2izC3BwcFs3749TQLPjoiICPr27aubN1LkXE6X6ZJr ol+JiYls2LCBv/76i7CwMJo2bUrdunWz3I4pXhe552jkmjdvTnR0NGfPnlU7lDzD29sbSPlF3L59 e8qWLUvz5s05dOjQK7dRFIWgoCC+/vprALRaLZs3b+bevXuEhIQwfPhwXbsAd+/e5auvvmLgwIE4 OzszevRo3YTNcXFxeHt74+7uzqBBg7h+/To9e/bEycmJYcOGodVqOXnyJJ07d6ZixYr0798/zXB1 WFgYHh4eLFu2jFatWtGoUSNDnKY8T66J4QUFBZGUlMThw4eZPn06zZo1+7/27jssiuvrA/h3cQFB pAWjMYqIiLHXWBJbLIlRAWMEjAUVjBoLGBuWGBBLwIhKiEYsJAYblqjBX0yCRhPsLUUUe1BRUBFE iY1y3j/uuyxL3YVdBnbP53l4lr07e+fMzO6cnZk792LMmDG4f/9+se8xmO1CrErLzc2l2rVr05df fil1KDoRHR1NACgtLU3tvyZNmlC1atVo586d9PDhQ7p48SJ16tSJZDIZ/fTTT4WmT0lJoaCgIDIz Mys0LwDUv39/lbLbt2+Tg4MDRUdHU1paGiUmJpKdnR316NGD0tLS6NatW7Rx40YCQI0aNaKVK1dS QkICLV68mACQu7s7LV26lC5fvkwhISEEgFavXp1Xf6tWrejvv/+mtLQ0unHjBrVt21aj5VfEHR0d rbPtAoAiIyM1jqu4dWoI28TNzY3c3d11tk3c3d3Jzc1N47gePnxI//zzDwUEBJCdnR0BoHbt2tG9 e/f0frtERkZSMWnwNh85VnEymQw9e/bMa0DDRIvI9u3bo1evXpDJZKhTpw6+/vprEBFWrFhRaHoT ExNMnjxZ7Q7SQ0JCYGRkhL59+wIALC0t8e677+L333/H5cuXYWFhkdfvrJOTE7y8vFC7dm24uLgA ANLT0zF27FjUqlUrb7pTp07l1X/lypW859bW1pg+fXrZV0YVxduk4shkMtSrVw9+fn44dOgQHB0d ce7cOWzdurXQtIa0Xfiaox7o0aMH5syZg+zsbMjlvElzcnJgaWmpUtaoUSM4OTkhPj6+2PeZm5ur Vf/Ro0fx7NkzTJo0Ka8sJSUFLVq0QFJSEpo0aZK3HfL3JqS4/pm/gZCiLP94eX379sX48eNx5MgR zJ07V+2WnfqIt0nFev311xEYGAgvL68S73M0hO3Ce1I98P7772PSpEk4fPiwRi3O9JWDgwPS09ML lb/++uta6QTg8uXLaN++PVatWqXR+4pqqasoy38/aUREBKZMmYKNGzdi165dWL58OYYMGVK+oPUc bxPtUfRbqo2OGqryduHTqnrAwcEBHTt2VLkIbsicnZ2RlJRUqBVtbm4uHB0dy12/lZUVLl68WORw YYrR1MtDLpdj7dq1+OqrrwAA48aNQ0xMTLnr1We8TbRHcf9ikyZNyl1XVd4unBz1xIcffojdu3cj Oztb6lAkN3r0aNy7d0/ltBARITExEb169dKoLplMhszMTJWyVq1a4eHDhwgLC1MpP3XqFHbu3Fn2 wP/f+vXrIZPJMGLEiLzbZ3799ddy16sveJvo1tGjR1GzZk0MHjxYo/fp23bh5Kgn3N3dkZaWxg1z AHTt2hU9e/bEokWL8k7B/PHHH3j+/Dl8fX2LfZ9i4NX81zTq1q2L+Ph43LlzBzExMSAizJw5E0ZG RliyZAl8fHywYcMGBAUFISwsDCNGjACgHIg1/48VRSfv+UdSV+xM8v+y3rJlS9776tevDwB6c/+Y ogm/Ot0SArxNdK1Pnz4YOnQoLl++DAC4efMmli1bhvDwcNStW7fY9xnCduHkqCccHBzQvn17rfwa q+pkMhm2bt2KJk2aYPTo0Zg3bx527tyJQ4cOFRq6CxBfzm+//TZv8OaAgAAkJiYCAGbNmoXs7Gx4 e3vDyckJMpkM7dq1w549e/Dmm2/il19+QXh4OIyMjLB27VqYmpri7t27CAgIACB+Ie/YsQNpaWlY vHgxANER+pYtW5CRkZE33dmzZ/PGt4uPj4eLiwuCg4Mxa9Ys+Pv7Y+TIkbpebToXGxub15rw1KlT WL16Na5evVrktLxNKkbfvn2RmJiI3r17Y+LEiVi1ahW2bt0KV1fXIqc3pO0io4IXZliVFRwcjNDQ UCQnJ+tNq9Xt27fD09NT5RckK52trS2io6Ph4eGhk/plMhkiIyMxaNAgndSvj8aMGQMTExNs375d J/V7eHjg5cuXetN9W0XYs2cPvL29i+rlK4mPHPWIh4cHUlNT8fvvv0sdCmOMVWmcHPWIo6MjWrZs WWSnwIwxxtTHyVHP+Pj4YPPmzXj8+LHUoTDGWJXFyVHPjBw5EtnZ2UV2/cQY009///23SstRVn6c HPWMra0tPvzwQ6xbt07qUBhjOnb69Gm4ubmhbdu2uHbtmtTh6BVOjnrI29sbZ8+exZkzZ6QOhTGm AzExMejatSs6duyIp0+f4o8//lC7M3CmHk6OeqhXr15o06YNQkNDpQ6FMaYlRIQdO3agXbt2cHV1 hY2NDeLi4hAbG4uuXbtKHZ7e4eSop3x9fbFz507cvn1b6lAYY+WQk5OD77//Hm3atIGnpyecnJzw 559/5h09Mt3QjzvFWSFDhw7FrFmzEBERgUWLFkkdTrl98MEHUofACli2bBk2btwodRhVxsWLF9Gz Z0+1p8/OzsaWLVuwdOlSXLp0CcOGDUNUVBRatWpV7HuOHTvG3xUNPHjwoNjXODnqKTMzM0yaNAkr V67EtGnTiuw2rSro1KkThg4dqnZfnBXpypUrAMQoIJXN0KFDdXoNatq0aRqdlcjOzsaRI0fg4OAA BwcHncVVmb366qtq9ViUnZ2NyMhIhISE4NatW/joo48QHR2N5s2bl/g+d3d3bYVqMF599VX069ev yNe4+zg9lpmZiUaNGmH06NEICQmROhy9o9jR6ao7MH3x8OFDvPvuu0hOTsbBgwfRtGlTqUOqlJ4/ f46IiAisWLECd+/exZgxY+Dv76+VYdaYxrj7OH1mYWGB6dOnY9WqVbh3757U4TADlJqaij59+iAl JQW//fYbJ8YiPHv2DCEhIXBwcMDs2bPh5uaGq1evIiIighOjhDg56rlJkybBwsKCjxxZhUtOTkb3 7t2RmpqKQ4cO4Y033pA6pErl8ePHCAwMRIMGDRAUFITRo0cjMTERYWFhaNCggdThGTxOjnquRo0a mD17Nr755hskJSVJHQ4zEHfu3ME777yDFy9eIC4urlJel5XKo0ePEBgYiEaNGmHFihWYOHEiEhMT ERwcjNq1a0sdHvt/nBwNwLhx42BtbY1ly5ZJHQozAP/++y+6deuGnJwcHDp0yGAb4BSUlpaG2bNn w97eHmFhYZg0aRKuX7+OwMBA1KpVS+rwWAGcHA2Aubk55s+fj9WrV+PChQtSh8P02PXr1/HOO+/A zMwMcXFxsLe3lzokySUnJ8PPzw8NGjTAunXrMG/ePNy6dQuBgYGws7OTOjxWDG6taiByc3PRrVs3 ZGVl4cSJEzAy4t9F5cWtVVVdu3YNvXr1gpWVFQ4cOGDwpwjv3LmDpUuXYsOGDbC0tIS/vz98fHxg YWEhdWisdNxa1VAYGRlhzZo1+Ouvv7B+/Xqpw2F65sKFC+jWrRtsbGxw8OBBg06M169fh5eXFxwd HbFr1y4sX74c//77L/z8/DgxViGcHA1Iy5Yt4evri9mzZ+P+/ftSh8P0xPnz59GrVy/Y29vj8OHD ePXVV6UOSRJXrlyBl5cXmjZtiuPHj2PdunW4ceMGxo0bB1NTU6nDYxri5GhggoKCYGVlhZkzZ0od CtMDZ86cQY8ePeDo6IhffvkFNjY2UodU4RISEuDh4YGmTZvi5MmTWL9+PS5evAgvLy+YmJhIHR4r I06OBsbc3BwhISGIiopCXFyc1OGwKuzkyZPo27cvWrdujdjYWFhbW0sdUoU6e/YsXFxc0KJFC8TH x2Pbtm1ISEiAl5cXjI2NpQ6PlRMnRwPk4eEBFxcXeHl5ISMjQ+pwWBV0/PhxvPfee2jfvj327dtn UNfSTp06BRcXF7z55ptITk7Gnj17EB8fD3d3d27opkd4SxqojRs3Ijc3F97e3lKHwqqYQ4cOoW/f vujYsSN+/PFH1KhRQ+qQKsRvv/2Grl27olOnTrh//z727t2L06dPw8XFhZOiHuItaqCsra0RFRWF vXv34rvvvpM6HFZFHDhwAAMHDkTv3r0RExMDc3NzqUPSOcW4ib1794apqSni4uJw8uRJuLi4QCaT SR0e0xFOjgase/fumDZtGiZPnpw3/BJjxfn111/h6uqK9957Dzt27NDrFphEhJiYGLz99ttwdXWF jY0N4uLicPDgQR5g2EBwcjRwixcvRrNmzTB8+HBkZWVJHQ6rpPbu3QsXFxcMGDAA0dHRetsKk4iw Y8cOtG3bFm5ubrC1tcXRo0fzjh6Z4eDkaOCMjY2xYcMGxMfHY8GCBVKHwyqhH374AR4eHhg6dCi2 bdumly0xc3Jy8P3336NVq1bw9PSEs7Mz/vzzT8TExOCtt96SOjwmAU6ODC1btsT69euxZMkSbN26 VepwWCWyadMmeHh4YPjw4YiMjES1atWkDkmrsrOzsXbtWjRp0gTe3t5o27Yt/vnnH2zfvh2tW7eW OjwmIbnUAbDKYfjw4Th79iy8vb3RuHFjdOjQQeqQmMQ2btwIHx8fTJgwAeHh4XrV+OTFixdYs2YN wsLCcPv2bXh7e2PWrFlo1KiR1KGxSoKPHFmeL7/8Ej179sSHH36IBw8eSB0Ok1BkZCS8vb0xefJk vUqMz549Q1hYGJydneHv7w8XFxdcvXoVERERnBiZCk6OLE+1atWwefNmGBsbY/DgwXj58qXUITEJ hIeHY+zYsZg6dSpWrFihF4nx6dOnCAkJgYODA+bMmYNBgwbhypUrCAsL4/EmWZE4OTIVtra22LZt G86cOYO5c+dKHQ6rYCtXroSfnx/mz5+P0NDQKp8YMzIyEBgYiAYNGiAoKAhjxoxBYmIiwsLCeKxJ ViK+5sgK6dChA9avX4+RI0eifv368PPzkzokVgEWLVqE+fPnIzg4GP7+/lKHUy7p6ekICQnBN998 AyMjI/j5+WHy5Mk8uDBTGydHVqThw4fjyZMnmDhxIiwtLTFmzBipQ2I6tGDBAgQGBiIkJASzZs2S OpwyS0lJwRdffIFvv/0Wcrkcc+fOxSeffAJLS0upQ2NVDCdHVqwJEybg7t27GD9+POrUqYP3339f 6pCYDsyePRtLly5FaGgopk2bJnU4ZXL37l2EhIRgw4YNqFmzJhYuXAhvb2/UrFlT6tBYFcXJkZUo KCgIGRkZGDJkCGJjY/mGaD0za9YshIaGYvXq1ZgwYYLU4WgsKSkJCxcuxMaNG/HKK69g8eLFGDt2 rMF0hs50h5MjK1VoaChu3LiBwYMH48iRI3BycpI6JFZORIRJkyYhIiICa9aswccffyx1SBq5evUq Fi5ciOjoaNSuXRtfffUVRo0apdf9vbKKxa1VWankcjm2b98OZ2dn9OrVC9evX5c6JFYOubm5mDBh AtatW4eoqKgqlRgvXboELy8vNG/eHCdOnMC6detw7do1jBs3jhMj0ypOjkwtZmZm+Pnnn+Hk5IS3 3noLFy5ckDokVga5ubkYN24cIiMjsWnTJgwbNkzqkNRy7tw5uLi4oHnz5jh9+jQ2b96clyj1tRN0 Ji1Ojkxt5ubm2LdvH5o1a4ZevXpxgqxisrOzMWLECHz//ffYunUrPD09pQ6pVIrBhDt06IC7d+9i z549uHDhAtzd3XmAYaZT/OliGlEkyKZNm6JXr16Ij4+XOiSmhqysLAwfPhy7d+/G7t27MWTIEKlD KtGJEyfQt29fdOzYEffu3cPevXtx5swZuLi4cFJkFYI/ZUxjNWrUwO7du1GvXj3069cPly9fljok VoIXL15g8ODBiImJwe7duzFgwACpQyqWYjDhLl26IDMzE7GxsTh16hRcXFyqfG89rGrh5MjKxMbG BrGxsahduza6deuGs2fPSh0SK8Lz588xePBgHD58GPv370e/fv2kDqlIisGE+/TpAzMzM8TFxeH4 8ePo06eP1KExA8XJkZWZra0t4uLi0KFDB3Tv3h379++XOiSWz7NnzzBo0CDExcXhp59+Qo8ePaQO SQURYceOHWjXrh1cXV1hY2ODuLg4xMbGomvXrlKHxwwcJ0dWLubm5vjxxx/h6emJQYMGYdu2bVKH xABkZmaiX79+OH78OPbv349u3bpJHVKenJwcfP/992jTpg08PT3h5OSEP//8M+/okbHKgDsBYOUm l8uxYcMG2NraYtiwYUhJScHUqVOlDkurMjIyEBkZqTKM1/379wEAISEheWUmJibw9vaGlZVVhceo 8OTJE/Tv3x8JCQn47bff0L59e8liyS87OxtbtmzB0qVLcenSJQwbNgxRUVFo1aqV1KExVhgxpiW5 ubk0b948kslktGjRomKne/78OWVlZVVgZOV36NAhAkDGxsZkampa5J+xsTEBoMOHD+s8nk2bNtG8 efMoNzdXpfzhw4fUoUMHsrOzo3Pnzuk8DnVkZWVRREQEOTo6klwup5EjR1J8fLzUYTFWktucHJnW ff3111StWjXy8vKi58+fq7z25MkTcnZ2pg8++ECi6MomJyeHatWqRQBK/KtVqxbl5OToNJbHjx+T paUlAaApU6bklaemplK7du3otddeowsXLug0BnU8e/aMVq5cSQ0aNCBjY2MaN24cXb9+XeqwGFMH J0emGwcOHCBra2vq1KkTJScn55V7enpStWrVCABt27ZNwgg1N3XqVDIxMSk2MZqYmNCnn36q8ziW LFmStw5lMhnNmjWLHjx4QG3atKG6detSQkKCzmMoydOnTyk4OJhq165N1atXJ19fX0pMTJQ0JsY0 xMmR6c7Vq1fJ2dmZXn/9dTp37hwtX76cjIyM8nbqlpaWKomzsjt58mSpR46nTp3SaQzp6elUo0YN lXkaGRlRnTp1qE6dOjo7XTl9+nSytbWl9PT0YqfJyMiggIAAqlWrFpmbm5O/vz+lpKToJB7GdIyT I9OtO3fuUPv27almzZp5iVHxZ2xsTAMHDpQ6RI00aNCg2MTo4OCg8/kHBQWRXC4vNG+ZTEaTJ0/W yTxnzJhBMpmMjIyMKCgoqNDr6enpFBAQQHZ2dmRlZUUBAQF0//59ncTCWAXh5Mh0LzExkWrWrJl3 KrDgX3R0tNQhqu2zzz7La3hTMNHPnz9fp/NOT08nCwuLYpOzTCajJUuWaHWefn5+Kj9qatSokXf0 +PDhQ/L396eaNWuStbU1BQQEUGpqqlbnz5hEODky3crOzqbu3bsXmVAUpwStra3p3r17UoeqloSE hGKTk66v9eW/1lhSglyzZo1W5jd16lSSyWQq9cvlcvL39ydfX1+ysLAgW1tbCg4OpsePH2tlnoxV ErdlRESa3v7BmLoCAwOxaNEi5OTkFDuNsbExBg8eXGU6EGjRogUuXrwIxVdHJpOhWbNmOu2E/cmT J7C3t8ejR49KndbIyAhXrlxBo0aNyjy/adOmYeXKlShq92BmZgZTU1P4+fnBz88PNjY2ZZ4PY5VU EveQw3Tm5MmTWLhwYYmJERAjRkRHR1eZ7ue8vLwglyv7z5DL5Rg1apRO57lq1So8efKk2NcV8Tg6 OiIiIgKOjo5lnteMGTOKTYyA2F7Tp09HYGAgJ0amt/jIkenMhQsXMGbMGJw5cyZv552VlVXktEZG Rnj11Vdx+fJlWFpaVmSYGrt9+zYaNGigcuR48+ZN1K9fXyfze/ToEerXr4/MzMxCrxkbGyMrKwv9 +vXDvHnzyt392syZMxEaGlpsYlSwtbVFUlISzMzMyjU/xiopPnJkutO8eXOcOnUKDx48wPr169Gn Tx/I5XLIZDJUq1ZNZdrc3Fw8fPgQU6ZMkSha9dWvXx+dOnWCkZERjIyM0LlzZ50lRgAIDw/H8+fP VcqMjY1hZGQEd3d3nDt3Dvv37y93YgwICFArMQKiO73169eXa36MVWpSXe1khik9PZ22b99Ow4YN o+rVq5NMJivUyOSnn36SOsxShYeHk1wuJ7lcTuHh4TqbT8EWqsbGxmRiYkJTpkzRam8zISEhpd7D WfCvbt26lJ2drbUYGKtEuEEOk056ejpiYmKwa9cu/PLLL3j58iWICPXq1UNCQgIsLCykDrFYqamp qFOnDgAgJSUFdnZ2OplPcHAw5syZA5lMBlNTU4wfPx7Tpk2Dvb291uYRGhqKGTNmqJRVq1YNpqam yMnJydsugLi2+dprr8HBwQGtWrVCeHg4D0LM9FGSQSfHhg0bIjExUeowGGMGqmHDhrhx44bUYbDC DDs5ymQyfPrpp+jSpYvUobAqSNF6tGbNmjqbx4MHD2BlZQUTExOd1J+Tk4OLFy/C2toadnZ23MCm Ah0/fhwrVqxQ6xovq3BJBj+eY+fOneHu7i51GIwxA8NJsXLj1qqMMcZYAZwcGWOMsQI4OTLGGGMF cHJkjDHGCuDkyBhjjBVg8K1VWRW2aROwaxewZw9gZgZ06wY8fQpkZgINGwJ+fkCPHlJHWZhMBlhY AJMmiUdvb6BuXfHagwdAQABgZQXcuyfKAwMBuRw4fBg4cgQ4dw7YvRtwcxPLzhjTOk6OrOoaMQJw dRWJpHdvICZGlGdmAjNmAD17Aj/8AHzwQcXEc/Uq0LixetN27w4EB6uWPX4MdOwIrFkDvPceQAR4 eQFDhohk2LOn+Hv6FKhRQ9vRM8by4dOqrGpT3ICfvyNzCwsgPFwcdS1aVDFxHD0KzJun/vTGxoXL QkOBFy+Ad98Vz2UyYP58YO9eIDZWOZ25efliZYyVipMjq9qK69fT2Bho2hS4dEn3MVy6JI7uXr4s Xz0//gg4Oakuk7MzYG8PrF5dvroZYxrh5Mj0U04OcO0a0KiReP7oETBqlEg8gwaJskuXgNGjlWVE 4ghw/HhgzBjg2DGgRQvg/ffF9OfPA5MnAyEhQOfOwBtvALm5wLp1QEqKuBY4ejQQFVW2mK9eBapX L1z+2msVk+QZY3k4OTL98/gx4OsL3L0LLFkiyqytCx99vfEGsGqV8nlyskiOa9eKRBgdLa79KQZo HjUKmDkT8PcH9u8Xp3SNjMTpUABo1w747jtg5MiyxW1pCdy5U7jc1ha4ebNsdTLGyoQb5DD98Pvv QJs2IjH++69osLJmDTBggHKaohqx5C+rWxf49FOR/NLSgOXLVa9lXrokjiYbNABsbIDPPtPuMrRv D+zbB1y/rjziBYDsbJE4GWMVho8cmX7o0QP46y/gxg1xBPgPyQrqAAAQFUlEQVTFF+IU6IAByiM/ dSgayrRqpZoYAaB/f9FCdvx4cZuFm5v24gdE4xu5HPDxAVJTRdybNysTMmOswnByZPqnTh1gyhRx G8f+/eKaoDZERQGenuK0a+PGwJYt2qlXoWNH4NAhcc/mW28B7u7ifyMj1SNgxpjOcXJk+qtrV/F4 4oR26jM2Fkdy69eL58OHiwSsTV27ioR+5Yq4wT8zE3j2DBg2TLvzYYyViJMjq9pKGhPvwgXx6OCg LJPLRS80Cmlp4jEnp/R5rVolWrb6+Ch7ptm3TzzKZMD/D36sNenporecqVPV71yAMaYVnBxZ1aZI SI8fq5ZfuwaMHQvUqiW6aVNo3Vpcw/vqK5HYFi8W5efOiZaqivpevCg8r2+/FY1jAOU1wM6dxWO9 esDffwO3b4ujyfIOZPvypbidxMkJCAoqX12MMY1xcmRV144d4igOEP2OvvmmaJjTpo1oLNOtG/Dn n0Dt2sr3rFkDNGsGfP65SGKffy6OLD/6SDSy+eQTMd0ffwALF6reWvH336L7toAA0dgnMFAkYEDU k50trkk6OxffOYE6rl4Vy1GnjugYwMys7HUxxsqEb+VgVZe7u/jTRIcOytOtCv/+q/x/8GDRoXlR SjoaHDtWmSjL6sQJkQxfvgQiIkSLWcaYJDg5MiaF/Nc9FTp3Vp6mLUnBU8iMMa3j5MhYRevdWzz6 +4vOBEaNEl3EleaPP8T10ps3RR18ZMmYznByZKyiHThQtvd17y7+GGM6xw1yGGOMsQI4OTLGGGMF cHJkhuO//4Ddu7XXnZyhe/5c6gjKJiMD2LlTdAPIWDE4OepSbi6wYAEwa5a4t87PT+qI1LNpE/DB B+JePXNzMWxTt25A27biVofff5c6Qs2lpophrAYPBv73v9Knf/99YOBA3cdVUbKzgRUrxH2Yvr6A qytw/HjZ6rpzB5g4EfDyKvxabq5IOh4eYoST3r3F5/+//zSLh0h5O4ulJdCnj7jPtDTjxinH6ywu 9ilTxC1AP/2kWUzMsJABA0DR0dG6m8EXXxC1bSv+//13oo8+0t28tC0jgwggGjhQWfbkCdH48aL8 hx8qLpYrV7RTz/PnInY3t9Knfe89ov79tTPfomhrmdStZ/Jkom7diLKyxPNr14jq1iU6elSz+f34 I9HHHxe/HhctIqpVS3x+iIhSU4ksLYlGj9YsnsBAsf4DAsS2AEQ9t24VH9vhw+pt36dPi55OW+tI TdHR0WTgu+DK7LZBbxmdJMfsbOX/jRsTubpqt/6Kkptb9A7k5Uuxw2jXrmLiOHKEyN1de/Wpmxx1 SVvLpG4958+L5T58WLXcz4/ozTc1n+/Ll8Wvx7p1ifr2VS3r1YvIykr9eFJSiAYNUn1txgzxnvnz i47p2TOR3NTdvgWnU2cd5f9uawEnx0rtNp9W1ZZLl0RfmMHByrKkpPJ1Iyal4uI2NgaaNhXLq2uX LgFDhogeY/SFtpZJk3o2bBCnx7t0US3v0wc4fRo4eVKzeSvGvCzK06dAfLxqXImJquNRlhbP8uVi PM78fH3F47VrRc932TLlNGWhzjpq1gyIjNSvzyMrFifH8jpzRuykmjUT3ZL17i36/Bw9Wgw1dO6c +P+LL0TH1uPHiyR67BjQooW4tgWIazVLlogBb/38gLffViag//4TYwkOGiSGLvr3X3E97JVXxHWf 3FzgyBHxRba2FtdMCjaWSE4GJkwQfYi2awd8/LEYDklTOTliB6UYqf7RI3ETu0ymvNZz6ZJYZkUZ UcnLfv686Ks0JET0EPPGG2KZ1q0DUlKU6zAqquTYkpLEcs2eLfpW9fQUHYEXZedOsc2srESMycnK 17KygF9/Lbo7OHXW46NHYkft4yOmGzAAuHxZ82Uq7jOhaT2HD4vtZWKiWv7GG+Lxq6/EUFkymbiP MilJJChTU9G/bFGdsBdn2jSxjsaNE5+VgwdFvZGR6sdTu7byfwVra/GYf4QVhfPnRcIubuSS//4T 22PyZDHCiaKz+fxKi+noUfGd8/UV04WFFb6OyvSL1MeuUkJZT6vm5hIdOiROHwHitFFsrChXnYHy 1M2dO0QhIaKsfXsiX1+iadOIevcWr3/9tTgNq+DsTNShg/j/8WOinTvFexs3Jlq7lujuXaLly0XZ 8OHi/ffuEYWHi7KNG5V1PXlC5OhI9L//ieePHonrQn36lLaCVE89ZWQQTZxIZGxMFBOjLM/MLDxt /rLSlr1tW6LERPF/WppyuYuKoTjJyUSvvCKu7RIR5eQQeXoS2dqKdZW/vgYNiHx8xHXTqVOV6/XF C3E9KjiYqGZNUZ6fOuvx+XOili3FtiESn4mmTYmaNNF8mUr6TGhSj7U1UffuhctTU0Udvr5EDx4Q 1alD1LChOG06YwbRd98VX2dx887NJVqwQPn6hAnis6BpPAUdOUIkkxH9849qeXY20ahRYr0XFdeL F+Izt3Klsmzz5sLTqRtTairR558T2diIz1tQkPjMlgGfVq3U+JqjRskxJ4do716izp3FF8bVlejE iZJmoPoFVFyradiw8PWLsDAiDw/lc1dXIjMz5fOsrMINZG7eFGXvv1+4bPx4ZdmMGUROTqrzGz1a THfhQsnxW1sTtW4tYgaIatQg2rCh5B8CRZWVtOxmZkRbtiif79lTcr1FGT+eqHlz1bJr18T7fXxU 6+vWrfB7AaJvv1WWKRqB5KfOegwMFElG0aiDSCThKVM0X6bSPhPq1mNiUnTjIsU2UVy33LpVPJ8z R1zzK7iN8ytp3rm54jMpkxHZ2REdO1a2ePLz8SGaO7dw+VdfEf32W/FxhYWJGF68UJYV1TBL05ge Pyb68kuxrS0siGbOVP0RpgZOjpUaX3PUSKtW4laARo3EqZy9e4FOndR/v+JaTatWQLVqqq/5+gLR 0eJ06IYNYtiiZ8+Ur8v/v6e//O+rX1885j8VpCjL37H14cOirtGjlX9374qxDW/dKjnmHj2Av/4C btwQp8u++EKcnhowQJx+VFdJy96/PzBihDjteu+eOCWqqUOHlMuu0KiRONVW8NYTW1vV54phr06f VpZVr154Huqsx/37xelaeb6eGT/4QJy61FRpnwl1OTiIa4EFKU4HW1mJR09PoG9fsY0XLCjb9fKs LHE6+q23gK1bxWDS/fqJ65CaxqMQFycuWQQGqpbfuiVOV7/zTvHxbN0qrpHn/46YmhaeTtOYatYE ZswQlzi+/FKcpndwADZvLj4WVqVw36qaePBAfLFee63wDra8iMRguj//LK4xOTsDCQklv6eonZei LP/I9gkJQMeOwHfflS/GOnXEPWKNG4vrhevWifvdyisqSiSotWvFzmzNGnFtVRO3byuvS+VXrx5w 6lTJ71Vcqypq55ifOuvxwgXtdQhels9EUZycVMelVEhPF4+OjuJRJhPXsGNjxbYND9d8XhMmiB+N 9+8DRkbi0dcXmD4d+OUXzeIBxHW9+fOBbdsKNwSaO1ck8oLX13NylGUJCWKYstJoElN+JiZif2Bn JxJlRkbp82JVAh85aiIxEVi6VDS4adhQ7AiuX9dO3YsXi1/rUVFAy5baqVPB2loc6RZ1pPfokeb1 de0qHk+cKF9cCsbG4hf3+vXi+fDhYiBiTTg4iJ1TUWMuNmlS8nsVPyhKS2rqrEdra5EgCybarCzN e5TR1meiZUuxbrKzVcsVLT979BCPmZliO4wcCaxeDZw9q/m8oqPFejT6/13L5MlA3bqiwZim8RAB c+YAq1aptnZV2LwZsLcXg0Er/gBg3z7lc1NT1fE6i6NuTArZ2WL++c8m/fWXdn4sskqBk6MmzMyA SZPE6a1168TpHmdnsTM/f1512pIGxi1KWJhIuIpTPorm4prWU5S2bUUPMSEhquXHjgFbtoj/nzwR rSAVSpqvYrDg/C0H5XLVU7lpaeIx/xFscVatEgnKxwfYs0eU7dsnHmUyEVtp3n1XzD9/wiYS28rV teT3xseLZvwffljydOqsx44dxdFD/lt6iICAAGVSVXeZSvtMqFvP8OFiDMiCR9AHD4pTwIrbFz77 TBzhhYaKXmkmTlRv++X3yiuqrXdlMnFas3ZtzeNZskS09m3eXDnNixfKnm2OHy/8B4jenBTP27YV lwSOHVPWoVim/D9y1I3p+XNxZqNxY8DbW5w+vnxZnPFo3VqzdcUqN6mvekoJ5e0EICeHaPduoo4d lQ104uPFa//9J8refVc5/ePHoqxfv8J11a5NVL060b59RN98Q9SsmZh23z6iS5eUPdYMGKB8j6JF aNeuheeRv3HByZNERkai3NOTaNUqotmzRbzPnolpmjQRZQqK+b3zjmqcV68StWghWmmmpCjL27cX 04eFiZas06aJ53XripaGJS1769aqvZIARBER4nn9+qJV4K1bRLt2Fd9I5OFDMa/+/cV2IRINYZyd VVtLyuVEb7+tnObZMxHTmjWq9bm5FW6Qo856/Ptv0ZpX8XlYsICoZ0+i/J8zdZeptM+EuvUQiQZB ffsql/vmTbENFY1Zjh4lGjxYOf2qVWJe+Vt5Kjx5UrhxmMLChaqtSh89Ei2GFy/WLJ6ICNGaeeFC 5V9AgPisnztX/HIWbGhz9KiIx95e1J2SIlojA2L95t8upcW0dKlogFOjBtH06aIVdjlwg5xKjVur aqWHnNxcogMHxJc5IEB8eSdNEl9Ac3PRJP/UKXHLhaIsKIgoKUlZx+bNYifSrp34MsbEiC/mwIFi R6NoUWljQ7Rpk2hS/umnoqx6ddHSMj1dNJ0HxE5z2zZl/YcOEXXpIr7YDg6i5d+TJ8rXe/cWt4EQ EW3fTjRkiKhHJhO3D3TvLpJYs2ZEn3yiGjsR0enT4jUrK6IxY8RO0cFB7ER27Sp52QGRsD7/XCSq wEDlDmrdOlFnly6iF5OSJCeLLvoGDiSaNUvs7NLTVafZt0/sPN95R9wC8PHHqq0dFYpKjuqsRyLR OvPtt0UrxjZtVFvfarJMJX0m0tI0Wze5uUQrVohl/vRTsX3/+EO8tn+/+GHxySfKlsRRUWL55XKi ZcuU9fzvf0QjRojXbG2JQkOJEhKUr2dni7IuXcRn0c1NPC/YQrmkeA4cIKpWTcyj4J+iO8biFNWK NjZW/IC1sBCPJ0+KH01BQUTXr6sXE5FY3sBA8UNMCzg5Vmq3ZUTaOG9XNclkMkRHR8PDw0N7lWZm AhYW2quPScPNTZw2V5weZkzL3+3t27fD09MTBrwLrsyS+JqjtnFi1A8vXyp7AWIM4O+2geHkyJiC u7vyNg1FF2iMMYPE9zkypmBnJ8Ye/PlnMR6honMAxpjB4eTImMI334g/xpjB49OqjDHGWAGcHBlj jLECODkyxhhjBXByZIwxxgow+AY5Bw8exKOydL7NGGPlcLYsHbuzCmPQPeTY29vj9u3bUofBGDNQ 9vb2uHnzptRhsMKSDDo5MsYYY0Xg7uMYY4yxgjg5MsYYYwVwcmSMMcYKkANIkjoIxhhjrBI593+s +/lJ39xJQAAAAABJRU5ErkJggg== --nextPart18709298.xrLdHrxNQQ--