From owner-rtfm@auckland.ac.nz Thu May 4 18:27:48 2000 Received: from mailhost.auckland.ac.nz (mailhost.auckland.ac.nz [130.216.1.4]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA19795 for ; Thu, 4 May 2000 18:27:45 -0400 (EDT) Received: (from majordom@localhost) by mailhost.auckland.ac.nz (8.9.2/8.9.2/8.9.2-ua) id HAA18846; Fri, 5 May 2000 07:31:04 +1200 (NZST) Received: from caida.org (ipn.caida.org [192.172.226.30]) by mailhost.auckland.ac.nz (8.9.2/8.9.2/8.9.2-ua) with ESMTP id HAA18841 for ; Fri, 5 May 2000 07:31:01 +1200 (NZST) Received: from localhost (nevil@localhost) by caida.org (8.8.8/8.7.3) with ESMTP id MAA00594; Thu, 4 May 2000 12:30:41 -0700 (PDT) Date: Thu, 4 May 2000 12:30:41 -0700 (PDT) From: Nevil Brownlee To: Marcelo Pias cc: rtfm@auckland.ac.nz Subject: Re: Collaborative work (acct/meas for charging) In-Reply-To: <3.0.6.32.20000426191247.007a2510@pop.cs.ucl.ac.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-rtfm@auckland.ac.nz Precedence: bulk Hello Marcelo: [ ... following up on an earlier suggestion about defining an interface for changing a running ruleset on an RTFM meter ... ] > I think we could start defining two interfaces (Reader and Manager). See > my other email. I think this is a strong argument to carry this work in the > RTFM working group. What do you think? I've had a chance to think about this a bit more carefully. I agree it's a good work item with which to re-activate the RTFM WG. It would also fit in quite well with some work being done by Tanja Zseby at GMD Fokus (have a look at http://pam2000.cs.waikato.ac.nz/pdf_papers/P005.pdf). However, it isn't going to be all that easy to define these interfaces in a way which fits neatly into the SNMP Meter MIB (RFC 2720). The way the MIB is written now, to download a ruleset you have to - Locate an open slot in the RuleSetInfoTable. - Create a RuleSetInfoEntry by setting the status for this open slot to createAndWait(5). - Set flowRuleInfoSize and flowRuleInfoName as required. - Download the rules into the row's rule table. - Set flowRuleInfoStatus to active(1). Once the ruleset is safely installed on the meter you can start a task to use it (by creating a row in the Manager Info Table), at which point it's an 'running' ruleset. After that you can't make it inactive, or change it, until all the tasks using it terminate. Once the ruleset is inactive you can mark it inactive in the Rule Info Table, then change row within it. However, that's how it is right now. As I understand it, what you want to do is to introduce a new status for a ruleset, in which it can be modified even if it's running. That could be done by introducing a new object in the RuleInfoTable to indicate that the ruleset is 'modifiable' by the manager which created it. The next issue is, what kinds of modification could be allowed? Right now the MIB says -- Although an individual rule within a RuleSet could be modified, -- it is much safer to simply download a complete new RuleSet. The meter makes no attempt to verify that any ruleset is valid, in the sense that it actually does something useful - that's left to the RTFM manager, SRL compiler and/or person creating the ruleset. I can see two ways to handle this. To change rulest (x), A) Download a complete new ruleset (y) incorporating all required changes, then change the Manager Info Table (in one operation) so that all refernces to (x) refer to (y) instead. or B) Download information about the rows to be changed, then have a single operation which makes all the changes. (B) would require significant changes to the Meter. Also one would need to create rulesets in which individual rows could be changed without changing the logical execution paths through the ruleset. For example one could have a ruleset with large sections which test, say, DestPeerAddress, with the same mask for a set of specified addresses. One could make those sections bigger than needed (with carefully-chosen dummy entries), then overwrite them later. Great care would be needed to do it right. (B) is much safer and simpler, but it could be expensive downloading a whole ruleset if only one or two rows are to be changed. Also, the manager will need to inform the reader(s) collecting data from the changed ruleset, so that their data files have some marker within them to record when the changes happened ... So, how often do we want to be able to make such changes? Every hour, every 10 minutes, or what? And roughly how many rules within the set are likely to be altered on each change? As I said, this is going to take a lot of careful thinking to get it right. Still, it certainly would be useful, so let's get started. If we can discuss it for a while on the list, we (you and I) can write a draft explaining what's needed and how it could be done - that's what we need next. Cheers, Nevil ------------------------------------------------------------- Nevil Brownlee Visiting Researcher Phone: (858) 822 0893 CAIDA, San Diego From owner-rtfm@auckland.ac.nz Wed May 10 07:16:42 2000 Received: from mailhost.auckland.ac.nz (mailhost.auckland.ac.nz [130.216.1.4]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA22423 for ; Wed, 10 May 2000 07:16:40 -0400 (EDT) Received: (from majordom@localhost) by mailhost.auckland.ac.nz (8.9.2/8.9.2/8.9.2-ua) id VAA19963; Wed, 10 May 2000 21:16:28 +1200 (NZST) Received: from utrhcs.cs.utwente.nl (utrhcs.cs.utwente.nl [130.89.10.247]) by mailhost.auckland.ac.nz (8.9.2/8.9.2/8.9.2-ua) with ESMTP id VAA19948 for ; Wed, 10 May 2000 21:16:18 +1200 (NZST) Received: from zeus.cs.utwente.nl (zeus.cs.utwente.nl [130.89.10.12]) by utrhcs.cs.utwente.nl (8.9.3/8.9.3) with ESMTP id LAA18284 for ; Wed, 10 May 2000 11:16:10 +0200 (MET DST) Received: from cs.utwente.nl by zeus.cs.utwente.nl (8.8.8+Sun/csrelay-Sol1.4/RB) id LAA09623; Wed, 10 May 2000 11:16:10 +0200 (MET DST) Message-ID: <391928DA.8A3EF350@cs.utwente.nl> Date: Wed, 10 May 2000 11:16:10 +0200 From: =?iso-8859-1?Q?R=F3bert=20P=E1rhonyi?= X-Mailer: Mozilla 4.61 [en] (WinNT; I) X-Accept-Language: en MIME-Version: 1.0 To: rtfm@auckland.ac.nz Subject: fd_filter error Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-rtfm@auckland.ac.nz Precedence: bulk Content-Transfer-Encoding: 7bit Hi, I'm using the fd_filter program to filter a flow file generated by NeMaC. This file is quite big and I receive every time a "Segmentation fault". Can anybody explain this behaviour? regards, Robert Parhonyi PhD Student University of Twente, The Netherlands From owner-rtfm@auckland.ac.nz Thu May 11 14:59:42 2000 Received: from mailhost.auckland.ac.nz (mailhost.auckland.ac.nz [130.216.1.4]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA05958 for ; Thu, 11 May 2000 14:59:40 -0400 (EDT) Received: (from majordom@localhost) by mailhost.auckland.ac.nz (8.9.2/8.9.2/8.9.2-ua) id FAA01987; Fri, 12 May 2000 05:25:33 +1200 (NZST) Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31]) by mailhost.auckland.ac.nz (8.9.2/8.9.2/8.9.2-ua) with SMTP id FAA01978 for ; Fri, 12 May 2000 05:25:23 +1200 (NZST) Received: from riogrande.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP id ; Thu, 11 May 2000 18:24:58 +0100 Message-Id: <3.0.6.32.20000511182346.00795640@pop.cs.ucl.ac.uk> X-Sender: marcpias@pop.cs.ucl.ac.uk X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.6 (32) Date: Thu, 11 May 2000 18:23:46 -0300 To: Nevil Brownlee From: Marcelo Pias Subject: Re: Collaborative work (acct/meas for charging) Cc: rtfm@auckland.ac.nz Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-rtfm@auckland.ac.nz Precedence: bulk Hello All, I had some time to think carefully about this interface. First, I'd like to comment the email below. At 12:30 04/05/00 -0700, Nevil Brownlee wrote: > >Hello Marcelo: > >[ ... following up on an earlier suggestion about defining an interface > for changing a running ruleset on an RTFM meter ... ] > >> I think we could start defining two interfaces (Reader and Manager). See >> my other email. I think this is a strong argument to carry this work in the >> RTFM working group. What do you think? > >However, it isn't going to be all that easy to define these interfaces in >a way which fits neatly into the SNMP Meter MIB (RFC 2720). The way the >MIB is written now, to download a ruleset you have to > - Locate an open slot in the RuleSetInfoTable. > - Create a RuleSetInfoEntry by setting the status for this > open slot to createAndWait(5). > - Set flowRuleInfoSize and flowRuleInfoName as required. > - Download the rules into the row's rule table. > - Set flowRuleInfoStatus to active(1). > >Once the ruleset is safely installed on the meter you can start a task >to use it (by creating a row in the Manager Info Table), at which point >it's an 'running' ruleset. After that you can't make it inactive, or >change it, until all the tasks using it terminate. Once the ruleset is >inactive you can mark it inactive in the Rule Info Table, then change >row within it. The Meter finds that a ruleset is not being used anymore checking the fields CurrentRuleSet and StandbyRuleSet in the flowManagerInfoTable for the manager that owns the ruleset. If these fields contain 0, the ruleset is not in use. Is that correct? > >However, that's how it is right now. As I understand it, what you >want to do is to introduce a new status for a ruleset, in which it >can be modified even if it's running. Correct! Further, I want to define generic operations to make this changes on the fly. That could be done by introducing >a new object in the RuleInfoTable to indicate that the ruleset is >'modifiable' by the manager which created it. Right. I'd suggest two modifications in the Meter MIB. One is the addition, as you said, of a new attribute in the flowRuleInfoIndex: 1) flowRuleInfoChangeable - to indicate whether this ruleset can be changed by its owner or not. The second would be the creation of a new status for the flowRuleInfoStatus. This is to indicate that a ruleset is being modified by its owner (Manager): 2) flowRuleInfoStatus - activate,...., "CHANGING" Although we could assume any ruleset to be changeable and cut off the item 1 above. I'm not sure about this yet... > >The next issue is, what kinds of modification could be allowed? >Right now the MIB says > >-- Although an individual rule within a RuleSet could be modified, >-- it is much safer to simply download a complete new RuleSet. > >The meter makes no attempt to verify that any ruleset is valid, in the >sense that it actually does something useful - that's left to the RTFM >manager, SRL compiler and/or person creating the ruleset. That's very good. Actually, we could push as much as possible the intelligence to the Manager and Reader Interfaces implementation and leave the Meter with minor modification. > >I can see two ways to handle this. To change rulest (x), > >A) Download a complete new ruleset (y) incorporating all required changes, > then change the Manager Info Table (in one operation) so that all > refernces to (x) refer to (y) instead. > >or > >B) Download information about the rows to be changed, then have a single > operation which makes all the changes. > >(B) would require significant changes to the Meter. Also one would > need to create rulesets in which individual rows could be changed > without changing the logical execution paths through the ruleset. > For example one could have a ruleset with large sections which > test, say, DestPeerAddress, with the same mask for a set of > specified addresses. One could make those sections bigger than > needed (with carefully-chosen dummy entries), then overwrite them > later. Great care would be needed to do it right. > >(B) is much safer and simpler, but it could be expensive downloading > a whole ruleset if only one or two rows are to be changed. I'd say that we could leave this decision to the Manager Interface implementation. Maybe, the Meter could provide some meta data (attributes from RuleSetInfoEntry) about a specific RuleSet. Suppose we have defined an operation in the Manager Interface called addRule which is responsible to add a new rule within a ruleset. For instance, if the implementation of this operation decides it should upload a new ruleset as the number of changes to be made whithin the ruleset are close to the total number of rules, it could choose the option B. On the other hand, if the change is in a few rules of the ruleset, the addRule implementation can choose the mechanism A. What I'm proposing is a separation of the Manager interface from its implementation. > >So, how often do we want to be able to make such changes? Every hour, >every 10 minutes, or what? And roughly how many rules within the set are >likely to be altered on each change? We could offer changes within rulesets very often although a minimum time interval since last modification may be specified. And also, we could provide any number of changes within the ruleset through this interface. It'll be a task for the Interface implementation to use A or B mechanisms to accomplish this modification according to relation between number of changes and total of rules. > >As I said, this is going to take a lot of careful thinking to get it >right. Still, it certainly would be useful, so let's get started. >If we can discuss it for a while on the list, we (you and I) can >write a draft explaining what's needed and how it could be done - >that's what we need next. That's fine. After some discussion in the list, we could start writing the requirements. My other email explain some ideas. I would appreciate comments on this email. Cheers. Marcelo ------------------------------------------------------------------- Marcelo Pias, PhD Student funded by British Telecom Labs University College London (UCL), Department of Computer Science URL: http://www.cs.ucl.ac.uk/staff/M.Pias/ ------------------------------------------------------------------- From owner-rtfm@auckland.ac.nz Thu May 11 15:04:10 2000 Received: from mailhost.auckland.ac.nz (mailhost.auckland.ac.nz [130.216.1.4]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA06137 for ; Thu, 11 May 2000 15:04:07 -0400 (EDT) Received: (from majordom@localhost) by mailhost.auckland.ac.nz (8.9.2/8.9.2/8.9.2-ua) id FAA01992; Fri, 12 May 2000 05:25:37 +1200 (NZST) Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31]) by mailhost.auckland.ac.nz (8.9.2/8.9.2/8.9.2-ua) with SMTP id FAA01982 for ; Fri, 12 May 2000 05:25:32 +1200 (NZST) Received: from riogrande.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP id ; Thu, 11 May 2000 18:25:21 +0100 Message-Id: <3.0.6.32.20000511182409.007a8c90@pop.cs.ucl.ac.uk> X-Sender: marcpias@pop.cs.ucl.ac.uk X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.6 (32) Date: Thu, 11 May 2000 18:24:09 -0300 To: Nevil Brownlee From: Marcelo Pias Subject: Generic Access Interfaces for the RTFM Meter Cc: rtfm@auckland.ac.nz, m.pias@cs.ucl.ac.uk Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-rtfm@auckland.ac.nz Precedence: bulk Hello Nevil (and list)! I'm sending some ideas about two access interfaces for the RTFM Meter: 1. Generic Manager Interface (GMI) - high level interface for controlling the meter using the RTFM rulesets. 2. Generic Reader Interface (GRI) - high level interface for getting measurement data from the RTFM meter. My idea is to define the common operations that should be provided by each interface according to its functionality. This would mean using GMI for controlling the RTFM meter and the GRI for getting data from it. For each interface we do have requirements. However, I'd like to start this discussion with the GMI. Following some ideas I had about all this... I would appreciate comments on this. Cheers. Marcelo ----------------------------------------------------------------------------- 3. Metering Layer To introduce some motivation for this work we could begin with a specification of the RTFM Meter Layer. This layer would provide two service interfaces: one for control and another for data. Control Interface Data Interface (GMI) (GRI) -------- ------ ^ ^ | | --------------- --------------- | GMI Implem. | | GRI Implem. | --------------- --------------- | | V V +----------------------------+ | RTFM Metering Layer | +----------------------------+ -------------- | RTFM Meter | -------------- 4. Generic Manager Interface (GMI): Control Interface 4.1. Requirements In order to cope with a diversity of applications the Generic Manager Interface (GMI) should be able to: (a) Provide a high level set of operations to control the Meter; (b) Upload a new ruleset into the Meter; (c) Change specific rules within the ruleset quite often although a minimum time interval since the last modification may be specified. This change would be required when adding, deleting a rule or modifying a rule. The latter one may be implemented as a delete followed by a add operation. However, this will be a choice of implementation. (d) Minor modification in the RTFM Meter in order to support changing rules on the fly. This also means putting the core knowleadge into the GMI implementation. (e) Change the status of a specific ruleset; 4.2. Operations The GMI may have six recommended operations with all of them matching the requirements described in section 4.1. In this section we do not make any implementation assumption, only describe the interface. The operations are defined as: (a) RuleSet granularity: (1) addRuleSet(MeterID, RuleSet_Record): this operation uploads a RuleSet into the Meter. The initial status of the RuleSet will be 'activate but not running'. (2) deleteRuleSet(MeterID, RuleSetName): deactivates the RuleSet with the name "RuleSetName". (3) RuleSetMetaData getRuleSetMetaData(MeterID, RuleSetName): get meta data for this RuleSet containing useful information such as: number of rules, owner, status... (attributes of the flowRuleSetInfoEntry) (b) Rule granularity: (4) addRule(MeterID, RuleSetName, RuleNumber, RuleSet_Row): add a new rule (RuleSet_Row) within the ruleset with the name "RuleSetName" in the position "RuleNumber". Depending on the "RuleNumber" it might be added in the beginning, end or after a specific point. (5) deleteRule(MeterID, RuleSetName, RuleNumber): delete the rule "RuleNumber" of the ruleset "RuleSetName" in the meter "MeterID". (6) changeRule(MeterID, RuleSetName, RuleNumber, RuleSet_Row) change the rule "RuleNumber" of a ruleset with the "RuleSet_Row". Description of data structures used: MeterID - identification of the meter. It is likely to be the IP address or other. RuleSetName - object value as described in the Meter MIB. Main ruleset ID. RuleNumber - index of rule within the ruleset RuleSet_Row - Meter MIB object FlowRuleEntry. RuleSet_Record - a list of RuleSet_Row. 4.3. Implementation Considerations We could provide here implementation guidelines for each of the operation proposed in the section 4.2. ------------------------------------------------------------------- Marcelo Pias, PhD Student funded by British Telecom Labs University College London (UCL), Department of Computer Science URL: http://www.cs.ucl.ac.uk/staff/M.Pias/ ------------------------------------------------------------------- From owner-rtfm@auckland.ac.nz Fri May 12 08:36:30 2000 Received: from mailhost.auckland.ac.nz (mailhost.auckland.ac.nz [130.216.1.4]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01199 for ; Fri, 12 May 2000 08:36:25 -0400 (EDT) Received: (from majordom@localhost) by mailhost.auckland.ac.nz (8.9.2/8.9.2/8.9.2-ua) id WAA26791; Fri, 12 May 2000 22:43:19 +1200 (NZST) Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31]) by mailhost.auckland.ac.nz (8.9.2/8.9.2/8.9.2-ua) with SMTP id WAA26782 for ; Fri, 12 May 2000 22:43:15 +1200 (NZST) Received: from riogrande.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP id ; Fri, 12 May 2000 11:42:58 +0100 Message-Id: <3.0.6.32.20000512114145.007e3bc0@pop.cs.ucl.ac.uk> X-Sender: marcpias@pop.cs.ucl.ac.uk X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.6 (32) Date: Fri, 12 May 2000 11:41:45 -0300 To: rtfm@auckland.ac.nz From: Marcelo Pias Subject: ToC proposal for the Generic Access Interfaces for the RTFM Meter Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-rtfm@auckland.ac.nz Precedence: bulk Dear All, I want to propose a ToC for a possible draft regarding the definition of two generic access interfaces for the RTFM meter. In my opinion, we need to make discussions about this. So, I would appreciate comments, ideas!! Cheers. Marcelo ---------------- First ToC Proposal -------------------------- Generic Access Interfaces for the RTFM Meter Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . 3. Metering Layer . . . . . . . . . . . . . . . . . . . . . . . 4. Generic Manager Interface (GMI) - Control . . . . . . . . . 4.1. Requirements . . . . . . . . . . . . . . . . . . . . . . . 4.2. Operations . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Implementation considerations . . . . . . . . . . . . . . 5. Generic Reader Interface (GRI) - Data . . . . . . . . . . . . 5.1. Requirements . . . . . . . . . . . . . . . . . . . . . . . 5.2. Operations . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. Implementation considerations . . . . . . . . . . . . . . . 6. References . . . . . . . . . . . . . . . . . . . . . . . . . 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 8. Authors' Addresses . . . . . . . . . . . . . . . . . . . . . ------------------------------------------------------------------- Marcelo Pias, PhD Student funded by British Telecom Labs University College London (UCL), Department of Computer Science URL: http://www.cs.ucl.ac.uk/staff/M.Pias/ ------------------------------------------------------------------- From owner-rtfm@auckland.ac.nz Sat May 13 17:24:30 2000 Received: from mailhost.auckland.ac.nz (mailhost.auckland.ac.nz [130.216.1.4]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA11903 for ; Sat, 13 May 2000 17:24:23 -0400 (EDT) Received: (from majordom@localhost) by mailhost.auckland.ac.nz (8.9.2/8.9.2/8.9.2-ua) id HAA01107; Sun, 14 May 2000 07:00:10 +1200 (NZST) Received: from tokyo.ccrle.nec.de (tokyo.ccrle.nec.de [195.37.70.2]) by mailhost.auckland.ac.nz (8.9.2/8.9.2/8.9.2-ua) with ESMTP id HAA01096 for ; Sun, 14 May 2000 07:00:01 +1200 (NZST) Received: from ccrle.nec.de (root@tokyo1.berlin.ccrle.nec.de [192.168.100.30]) by tokyo.ccrle.nec.de (8.10.1/8.10.1) with ESMTP id e4DJ0Dd20795; Sat, 13 May 2000 21:00:13 +0200 (CEST) Message-ID: <391DA5F0.253FC000@ccrle.nec.de> Date: Sat, 13 May 2000 20:58:56 +0200 From: Juergen Quittek Organization: NEC Europe Ltd. X-Mailer: Mozilla 4.7 [en] (Win98; I) X-Accept-Language: en MIME-Version: 1.0 To: Marcelo Pias CC: rtfm@auckland.ac.nz Subject: Re: ToC proposal for the Generic Access Interfaces for the RTFM Meter References: <3.0.6.32.20000512114145.007e3bc0@pop.cs.ucl.ac.uk> Content-Type: multipart/mixed; boundary="------------F65A4A07804C1C69117016A9" Sender: owner-rtfm@auckland.ac.nz Precedence: bulk This is a multi-part message in MIME format. --------------F65A4A07804C1C69117016A9 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi Marcelo, Marcelo Pias wrote: > > Dear All, > > I want to propose a ToC for a possible draft regarding the definition of > two generic access interfaces for the RTFM meter. I very much like your initiative for defining an RTFM access interface for RTFM managers and readers. I see a big need for such an interface, particularly for management and accounting applications. > > In my opinion, we need to make discussions about this. So, I would > appreciate comments, ideas!! At NEC we built a Java class library modeling the RTFM architecture, something like a small version of what you are proposing. However, for the applications we have been dealing with (policy-based management, accounting for single value added services) also higher-level interface functions were required. The main point was, that the application programmers were used to deal with IP addresses, port numbers, TOS fields, etc., but not with RTFM rules and rule sets. Consequently we added another layer on top of our RTFM library which we called flow measurement interface. This interface allows an application to specify flows to be measured without dealing with RTFM rules or SRL. You'll find the javadoc interface description in the attached HTML pages. The core class is FlowSpecification which provides means for specifying flow measurements without writing rule files. Such a flow specification can be passed to a manager (class MeteringManager) which converts the specification into a rule set and installs it at one or more meters. Then it returns a reader interface which can be used by the application to access measured data corresponding to the specified flow. > > Cheers. Marcelo > > ---------------- First ToC Proposal -------------------------- > > Generic Access Interfaces for the RTFM Meter > > Table of Contents > > 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . > > 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . > > 3. Metering Layer . . . . . . . . . . . . . . . . . . . . . . . > > 4. Generic Manager Interface (GMI) - Control . . . . . . . . . > 4.1. Requirements . . . . . . . . . . . . . . . . . . . . . . . > 4.2. Operations . . . . . . . . . . . . . . . . . . . . . . . . > 4.3. Implementation considerations . . . . . . . . . . . . . . > > 5. Generic Reader Interface (GRI) - Data . . . . . . . . . . . . > 5.1. Requirements . . . . . . . . . . . . . . . . . . . . . . . > 5.2. Operations . . . . . . . . . . . . . . . . . . . . . . . . . > 5.3. Implementation considerations . . . . . . . . . . . . . . . I suggest to extend this document by a higher-level Flow Measurement Interface (FMI) which offers an abstraction of the RTFM architecture, hiding several details. Via this interface applications can specify flow measurements and can retrieve measurement data in a simple way, dealing only with flow specs and measured flow data. This can be based on the work at NEC mentioned above, or on some other suggestions. But anyway, I consider it very important to have such an interface available for applications requiring flow measurements. Of course, application may still use GMI/GRI directly: Application | | V | +-------+ | | FMI | | +-------+ | | | V V +-------------+ | GMI / GRI | +-------------+ Best wishes, Juergen -- Juergen Quittek quittek@ccrle.nec.de Tel: +49 30 254230-19 NEC Europe Ltd., C&C Research Laboratories Fax: +49 30 254230-99 Hardenbergplatz 2, 10623 Berlin, Germany http://www.ccrle.nec.de --------------F65A4A07804C1C69117016A9 Content-Type: application/x-zip-compressed; name="fmidoc.zip" Content-Disposition: inline; filename="fmidoc.zip" Content-Transfer-Encoding: base64 UEsDBAoAAAAAAIyTrSgAAAAAAAAAAAAAAAAHAAAAZm1pZG9jL1BLAwQUAAAACABYpa0oHp1u aHUDAADWDgAAGQAAAGZtaWRvYy9vdmVydmlldy10cmVlLmh0bWztV9+PozYQfq/U/2HOVftS gdlkpUopROJngkogIuxury8nhzgJVwIRWMntf9/BkGySyzWn21ZqpeUB7JnxN9/Ynhmhv3Mi O3k/dWGcTAKYPliBbwNRKH3q25Q6idMq7lUNvIpteM0FpW5IyFqI7YDS/X6v7vtqWa1oEtPY tZW12OT3Gl121upCLMjw++/0d4oS8v2UrbiiNPMGV35d0+n0MOIFr5jgC5g/w0e2Y4syhbKA GRMwYc9w14eeNrjvDe5/gdEk+VnrDTQNRfhqQRM/CVwcDMDOWV3DOEO8Kl0/o44elHrgh79B 7AZgkFo857xecy4INPtgEME/CZrWNYFx7HqnFqqUShSDzBqxDIweIrAi5z1YIzsKotgg+3Um pEEXm3F8YJaYcQKRB6H5aJnxqaoNw0TNBJ0UbDdn1QdRbslQgqBap+YQAzWtwAUrih0XfWkE nnwnGRvkTtN+JGC7QTA1HccPRyhq57Opacu5JlknsXw7gHRRFRq9F+o/uPh4Hq4LzNnMICHb WayyeZ7fkev0PiyzqhZVub/J84yadkGt38ADIDkwA38UGiTlheAVgcduLreitXG+kjDgo3tR mJxrvbIQqJX3RKeNfvhTMa+3v+JFca648ORzzUXMd+ilW/wFR9JGt4ZJxblOreHXeLwZlNld 0QXfVjxt8kbJs1qoTQqiN0nFdy4CRhLO0f6ESnNe/wydrFjwTwrL85tE/MbyX+Gw5vlWwepx k8IYDW8woG2uUHmX25HTJU93K6tstRbnlxTi6EnmVX+ouxO5qvu0i1vQ00S8qByfx9g7xDHz /8DsU3pNJrRkp7H72I4OktD9PTnGdOT7bT4uz7bdVMAqNnITg3T1yRp6MZaFWbubJp7HOaMj Rrnj1S7je0VgMnwJK4zgHO4imM8O5bzGGuCGztUS29XYpvNIBNsNE1eOxj18HzsGeGUFZp7D lKV/YtfCIiENdHqx4kqnaQ0fZH8L/K6xpFmVNi0jlfZNd1NzVqzUaP6Rp+LvrSux3KjLTaYe N7GRUJRQLy/3DhPscNPxGDrJcee+DXS25Wm2zLBOZGVxjn6meoWbCcfinhWrCStwh6sXJxeK V7h4qBEg5mxxCn8ifLle7QF0n3GXna/s3PNSiHLzX27eLcO3/v3Wv9/691v//v/17waj+fGS YN0/5V9QSwMEFAAAAAgAWKWtKE45f/a/CAAAADsAABUAAABmbWlkb2MvaW5kZXgtYWxsLmh0 bWztW21v4rgW/l5p/oM31V11dIeEaUerVS8ghQRodCGgkHZ2dnWFTDAlOyFBiVuGf3+PnQTy Wl67ZXYHqZQc28fneXx8fGIntZ/UvmJ+GbTQndnrosF9s6spSKhI0ucbRZJUUw0LPolV1Pbx nASESlJLF4QZpYtbSVoul+LyRvT8R8k0JKOlVGZ07nyqStOotjihE6Hx7qL2U6Wik+UAP5JK hV0zvfx/S1ajctQhLvExJRM0XqE/8TOeeBbyXDTEFPXwCn28QdfV20/Xt59+RZ2e+e/q9W21 CiL4CpWamtltwY9bpLkT8g0kUiyqdTX9v8hodVFdCOjKIcGMECoghr4uUPKNSlYQCOjOaLWT NUQu5VrqwpCJORwptrvZV7+gZkfpd/tGXVjObMorRIjq6w8amrJhon4b6fJDUzaSRaHxMpT0 oBMXP4+xP6LeQmhwJVBck+QGwJOb3RZq9g21BX1VBfRZU827uvCxWv2XgJRWtzuQVVXTOyAK r4cDWeHXVW61afBvFYG5UKTXrzemX7bg025Du648HNYFHT83sa8Qx/koFJs3mtp+QH1vudXO lGnVjGk3TD1CYBySu1pHrwsWcSnxBfQQXXMqwjrqjgYj+NTafd1Ml7Y9l0Kp4uAgqEmsvPGz Ow4W/wFHUQ/pQo4cxnsm/rNNlhXqEyKySQCk8P41NdN5rdkwoVJNajYiExhrpzFjQhY+sdgU qjh2QLcaoq7r72NOm3+KzDHIM1gUNS6hn9eBvvkUTXR7GgZmxFlUIG5shX4HFbeAlsL5InF/ Dn+p0QSKPNO3H2c07ajI6H/mc+umUWv1eKvoX9g4VJqcjJnokcd4HeMYar/DDKxcs9kQGjsw Wg/hr1iit34z15jW9h7WR4JXm41WSCqCSNZpmXUhilHNRtuA0DAM2ZRhPNIWpXVUsOOU6dH7 KK0qAyQ3IOkYW0ctXS0MsVGM5WEstOVy1B4JjTa3NyHsgLCTFWog1LLCHgh7WaEOQj0rNEBo ZIUmCM2s8B6E96HwzkgGXW5sMsjW7q756sMQMKrCS5UvqSoQFuv06XQuTee21Ha8pYopjucF tIwk62GrIIvFRcTaiNBG3Kplo0JuiDVVbSjY920SIDojKFgQy57aEF1sWMOxO0EQajx/Asv7 BNogb4owmhOI9CCZgiZxB8sv46ur9ykQV+8TMN5dKJ4LK9OTRT0fTeHvGGCMUUAWawwACiLz BV2hmuVNSKIuvxRfGoFhkpP0UKSK9h2TIr0FSqNRMmd2EOmNBgmGDCOX0KXnf+VjgaiHxgSG BwdPMD4i2gfTZU58xbI5EToQNfiSJxOfBMEHZLv0A3qhKPX1vpiqq3yzbYqOd5Wd+c47T8Jr MrUPcZ8yqh3sPopD6tvuY5LmvHgnilMtXmj+lxMbLzQ1KQx9majZKYuanR2j5n0ANywGwRPi h3Q/RrcoxpND2rZDrtZ4claGZOYa5OulWOsROvMmwOZ2vrLGwdqxkRRwtI9bPRKqkoAmZtJV jCdfcCiAAwa8QyhfXCZgg+2GS4s2QDg0JY5amFozCFqHYe7h4GsabCh5Y5RzMEK8QIdhGng+ 1Z/mY+KnkSXlb4xvAaYgl9uyxzBuEgNA1Gb3o7JF7WebrjZAM+JjcL6YIsSoqD0nLLvht8co 7niPEEPzEXgDJl/0yvEjhsXTglRSt/PAdHHRuKSlf9mwgMLto1I+nYa+VRgVs/I3mE6B9+Rb 5JTxEEClw+Fa8HbwDg6EYHtRHMyI3w7YCSKg6WM3mBIfUvbmipKEe+ZLXn3CRVBgytFN52jM ej8E0ABbXwkthrQueyNQi7D//X2SY2ADb64WJANtI34Dn6SxBYiCCSIqzbG1shxb22NnooAc bTCSVdUYyfoXzkriOkXHkEIjCz1j38Zjh7wWLwMYZjK1XRhsFoBuEXZXyUgbTtUAuZ5bmdgs rWErDNk35iZQjlRtaGq6YmoPrSwDybJzZuMETPCdypGcYiCSnQdyZHnzuRf3Jr+iU0SwSx2j oPy7oehk9DQLKGmeIw3NV/eU5hZPaZ67pxRSdDJ6lAJKlHOkQXl1T1G2eIpy7p5SSNEJ6Gn9 JitmihAuOQsKPNdZ8VyNfMMWLSBALE3aemVJW2/HpK3Hzm1s97GHXfy43j9gCtIF+xxkFOvM KVwfYhBUsxs+wU6F39pPgbr4wGJOXFqT7Abi+7PvEfYtdu5pUShDMYvj1QUy2gq6rv5yA97k Umy7QXwm9QF++Hwn5AM/uwJxZFbJUUiR9ZcZYZTg56TH7ZvvzFvoR7LjeOBb0ZHPMjqPyFXn UuSN/wTWQuxQegFtgD785NCQJvZ4DnNBptJBMy+gYumd+U4MbTsoSpxgFJOZbxfVPneST0Hc 9mOfYtIKDni+D76Kw5teFt70I8LbJdgV7o2Ghz69MEysOciF7pfO3MJh2GiMlBXoyLU8dDNg zxFRIPpRwg67sWV5Ty7l45442I9P7cuHwSgbBuPQ4zcWkaMgGv585b1vds0xW0++TxgHbJFh jy+Ur61mGWrzuA0R05D14aBvmCP2zOJ6XyQvfovkhG1YrXeLwu0A9qskW903G8tjzOWrL1f5 21OiKb1BEQ9M/lbggwR6ZsdxCE2lECCIzwEfRmAISyJdWJnsl4/GdsI66huje7UMclR6DsiJ DfHR5/BhfQCjTuvYG7DbpnxxzX8ARSVucib+AWaULpX3ZUvl/YEJAuchtZKvCTg+JeC3nQ82 hoRg/fxc/IDc+uG5J9aGp0oYBZBXOdGNaSggFHnTCy4JEnnUeLWp/fLjYfENmQV52ZhcsNwM bnTKH9DLZVAJQSKN/eN/xyWySa1rhXsns3vfZOyYxaWf3QwJTlUMqV3CNIVxYA/Os/M9fn8b XERHsSzdLJ2nO7J8Yo6/W4YjT+cEp/gtuJUI7Tnzp7XfHf22z9ij1Juno+F5vfATWvjjnZ8f 7/z8eOfnxzs/ze/pnZ+7iH2JvazJlUVvn/4fUEsDBBQAAAAIAFilrSh8410C8AIAAPcMAAAb AAAAZm1pZG9jL2RlcHJlY2F0ZWQtbGlzdC5odG1s7VdBb5swFL5P2n949bRdJmKa5DBlEAmw SdAIROC16y4VadwmEw0RWKH99zOGpEnaLl3VwybFB7D9nr/3PdvvIzFOSOiwizGFIRv5MP5u +54DSMP4vONgTBipDd2WDm6e3PKCC4xpgNBMiGUP47IsW2WnleU3mEU4oo42E7dpV8fXjXdr Kqao//6dcaJpAS/HyQ3XtGpc4ao3tUhjhwFf8DwRfAqTe/iVrJJpdgXZAuJEwCi5h9MOtPVe t93rfoHBiH3W2z1dl1PyUYMyj/lUdnpA+DLnVwrLnxdC2vDaaPhe8A0i6oOJCnGf8mLGuUBQ 7YOJBL8T+KooEAwj6m57tNSsQjFRXE2rxPA6AzskF2APnNAPIxOVs7lQDk1u5qZBzKyIQehC YJ3ZVrRtqtOwpGUkgyyS1STJL0W2RH0FIs0GtvoyUcv2KdhhRKiMpSM49wgbmuhU1z8icKjv jy1CvGAgp+pxPLYcNdYVaxapJwFJV5oCs/1A/QOVzXXlOt+KYxMFycpOcoen6Sl6mt7l9Twv RJ6VB3nuUNP3qHUqeABJDizfGwQmuuILwXMEZ81YbUXtQ15IGGQz3DBgu1Y3WwhpddKkKAxc 2fufFpNi+VVeFPKaEFZzYbIVz1dzXmoi57xVlYPcFBXfI3vBDbvPpJOB7X5Dodq1P9JwVXuK RsRXkkmz+Jl8lY+M+lAdW7HfJv35YsrvtCRND6buVZ5/k/uLOcx4utSkdhykMJSOBxjgulKw usl1jzSl09zJfH4zE7tXFKLwXFVVp2/QkVrVvOrFNeh2Ge7pxuMc2+s8Yu+nrD2tXdVBTXYc 0bO6t54J6A+2yWnD93Ux9s+23lSQGjagzESNOtl9N5KiENe7acnz2GW0wZhurp6WSmF+Di0I YRdwL51Hx7KrsSbQgDwpsY3GVl8eheDQgFHVG7aVim8VB1hjr45f2/CWc/Qm0j7JhMhu/2V1 rxkeBf4o8EeBPwr8/yfwFUb1y1yBNX86fgNQSwMEFAAAAAgAWKWtKANZf8zIAQAAdgMAABwA AABmbWlkb2MvYWxsY2xhc3Nlcy1mcmFtZS5odG1shVLJjptAEL1Hyj/UtJTTCAovh8gCJDYv Cgar6ZE1xw60bSIMI2iFeL4+TWNHnihyLkVR9frV9uynMA3Y6y6CNdvGsHvx400AxEDczwLE kIVjYm5asGz5WXRCIkYJIScp3xaIfd+b/cxs2iMyijQKjJM8V3MLD1e0WciCuJ8/2U+GkYh+ x4/CMIb/gVd/Iy+85mElatFyKQr4foEf/CcvmhyaGjIuYcsvMJnB1FrMp4v5V1ht2bM1XViW CikzkrINiyPleFUFQcW7TnQqirewHW+Sb0CjGBzSyUslupMQksCwAYdI8Uti3nUE1jRa3iNM HdUsDsmGsB4Jb737afgK/ipI45Q6pD+VcgQs04RBV74LhzxPCASxl2UO0YtcC16U9XHZ1FJD ffeuZxt918bhtU5RZdVsnh9H4Kc0jFQRi8B+E7K1QyaW9UVTMKptCEm6p97OHcvfF91IcR4r 2t51yFYezng4l7ismj7kkpvDAdWwHl1FzCH50JF+TNwbxEbvT1//JsreRF4eypzLsqn/w/gB +5h6K6Ro1da2vFY6ah8Q/4V8TPvSKQxVB3lIeYf6QDfeSYlMSwHHK6C+lvIGf5DHqJdR878B UEsDBBQAAAAIAFilrSg7WPnysQEAALYCAAARAAAAZm1pZG9jL2luZGV4Lmh0bWx1UlFv2jAQ fp+0/3CNVGnTmtgLPFQoIKUktEhAUepq2qNJLuDVsZFtiPj3dQKs26S9xHf33Xf+vnOSm+x5 yn6uc3hiywWsXx8W8ykEISE/BlNCMpadgWFEgRmurHBCKy4JyVdBsHNuPyKkbduoHUTabAkr SJFPw51r5JASqbXFqHLV5POn5CYMV9iu+RbDsMu7uf2Zp9kFh0dUaLjDCjYn+MWPvNIlaAUv 3MGSn+D7AGI6Gsaj4T08Ltk3Go8o9SVKzzPZnC1yH3zMyXR5aFA53umGL6/KCSex+uqbybU7 IVcNsyJd5i85g1JLOw5ient3T2+D3xBYU44DLmUpubVow9rwBqPObgDKh+Ngz8s373HWAf8S jasbUjeCzKRuM+74X8x+5gePXMV0yer5nPULi/2374NUonFeflfxyNofbCcsVBfX0MVoxVb5 VTgNG4SjwNYnByvUFtwOobdgoUbuDgYjmNdw0gewiB72fA9ab+iur3KDFyoHpdXZf1jyPd9I hBY3UErhL468mofCy1kI9dbdnKTwVOSz/+5gsrpOgyMa6x8rSkg6Scifzsnln3kHUEsDBBQA AAAIAFilrSgpQgHsbAEAAKYCAAAUAAAAZm1pZG9jL3BhY2thZ2VzLmh0bWytUsFuwjAMvU/a P3g57DK17oDD1BUkaANDKwWVTBPHQF3KVFrUROv4+zUFJg47TbvYL7bjFz/HuwvmvlgtOLyI WQiLt1E49YFZiO9dHzEQwSnRsx0YV3JPijQijxjLtD64iHVd23XXLqstihhj7luZ3uc9B9Nz tZ3ohA1ub7w7y4qoXsgtWZY5m76t58PgnIcJFVRJTQmsj/AhP2VSbqAsYCk1zOQRHrvQcdxe x+09wWQmHpyO6zhNqDGnpmIqQt6ABuIFe+E0eoWYh9BnSh9zUhmRZmDG7jNNXxo3SjF4ifn4 usJuo22XPluacDsHXh48mgcrGE38eTiP+6zOdrotMJn4F+/5PBLcIJERpFVZaDg0akAmFayJ CqgoLzdmenuRk1QEisi9XL8v1urw/DfrDc/D7YqEvmyzITZo1wmfVKldWXg4HPwrU6XTPab7 HY7zsg6klmfWqCys9JrZNtQe/ohjFDbKnqQ+/ZFvUEsDBAoAAAAAANCkrSgAAAAAAAAAAAAA AAAMAAAAZm1pZG9jL3J0Zm0vUEsDBAoAAAAAANCkrSgAAAAAAAAAAAAAAAAQAAAAZm1pZG9j L3J0Zm0vZm1pL1BLAwQUAAAACABYpa0oV1gbvR4HAACuJgAAHQAAAGZtaWRvYy9ydGZtL2Zt aS9GbG93RGF0YS5odG1s7Vpbb9pIFH6v1P9w6mhXXTVgEvKwYsGSwSZBCwYZt93uSzTAENw1 NmsPoaz643duNr5xSbbZXhQrMrbn9p1vzpzzjZ3mK2PYcT6MTLhxBn0YvW33ex1QKqr6vt5R VcMxRMFVtQbdEC1xhImqmpaiLAhZNVR1s9lUN/VqEN6pjq3aZqeyIEvvqqbOZe3qjMwU7eWL 5qtKxcKbEbrDlQq7Z/3yX1M3ZDlcYx+HiOAZTLbwEd2jWTCFwIcxIjBAW7iow2WtcXXZuPoV rgfOm9plo1ajj+hJdOr0nL5JLxrQ8VAUAXS9YGMggmiZGhc2+z3rd7DNPrSUiGw9HC0wJgow HloKwZ+IOo0iBW5ss9tSqlWV/u3qVXkZ76uljNljbp4a29EeGh+gfd0Z9od2S9ksXMIrSAtb yQFjR7cdGHbB0t+1dTtdJIzRacmADuKj+wkKb0mwUjTeCS1uqrpGzdXbfRPaQ9sw6Vg1Bd73 DOempVzUaj8p0DH7/ZFuGD3rmj4S9+OR3uH3NY7asfnZAAqXFlmtyx30M5Me3S5t19fH45Zi ofs2CjvY8y6Ucni3czeMSBhsjuLMQKvloNVZ9wAUHOj93rXVUqbYJzhU4J2851SIOkYKcJcf ZYBtfK9o8LM/iVa/NbtDy8nW6QY+EXWabY17TlNtU+yspiZbqY5RMuQhjoAelKW0GwX3OLx3 8aZCQoyrbKnQITmenpEBw5E4tFIKCOPyS4KZ4VWIp2y5VTw3IkfhGEn9JwTl+jP8qYI87yic Hqv5hEgW2FtVaAA6CuSGVjyCQxULTeULQVwZcuVJlw7duwXJejjYw/d8Uda1pjngreSPaCw6 Ta/iXNgpWnoZ2zHu/UmXbuWSLSMBdmSb70QDcR8/z7ESkvlSnS9dlUXW8QpP3blLfcIN/Jin tmaZf8gFJmjRd9QkZj8Oarm7iJGBxtNr02kpMlK2ta5NA1SCAWCPYXGO2NeNNYRsTzlr5OQm s5CZw4Jd9XK7xm8HA93+0JAge5Zl2uLys/jp9sy+kXmSGHA2DXwad9dTEoS30Xq5ROFW0TpD a+zYO28stFpisghmuwYD07kZGuXT9QibDNPRe/3YoFPhzzBBrnc6+rj+XvCFpZdNwy0wLaM0 C8s0zCSKnW+3S96cCzB0R98VSoFzyc5pWnjeZAuoSheQhMkUA+s+p1eonuDN6aqkP0wKVT3k 31WHk494SpjDfGanN5UKddC4y+qudZvbLFpL/AaXW4ajrdYTz53ClI/IlkmqFSunAgj7swjy ozZV3gVDxQCjMHRxBGSBIUrHAUD+DGiiCMIZFXIz2jEEc0BAJwyH9MmcDleVnexAURxjjEH3 oqABAomhPTj2dIaGqRVKmyp/zpzjHLhyKe11wAC6/t0A+VSlhrzPMx9vbIxmOHzNWeZ1ziHD eGas8xRtY8K6OwfXJ/z0S4wwP1LpIC/gIUcRSGH0FA1yKoVnxHORcXAegKR3y9iUc/ASTdsC vswLDfa3AbHO33acob13nERqlsa5Q1LzIic16wUVnJHMIoynJEOHHinJ4KCJh2/oTFFiO4EX hIoMj4l8zsTBNzwOMkm5ww1jgTsjGHJ5ZF9y5MPbwSY9tPQoOsihxZKkt7P47jV1xt3CZx7R 1l7HLsJJ52FJRN3HnROrIxoSAC9XZAvNaTDDqYH5bXVfrI4Tdn7rJIL9ITdLXCaf5L4Dbxlw yF/aUQ4Jzdiqn3J5XGz0uEuIqeCxbU1cr0qnDycBJacWHuqTd5h02c5RnxL33iVb5pv5Z0/k o9eY8AxG3CVmaYrvYGE36A9LeB8V+E4/+p/ophLkW2LbC/y7L8ixEyI/muOQqp72luBI8px/ /MRc++vlBIeMbbIbGCZi5B+X8RGa/oVJCeey4KuwvorHPpTvcrkrunX9BZWLBM9u52GwvOXC /Tavzp82teXenXB/GK8nJdlNSzJYBAlwYMDljqO4rWh/EQk0pb6Emb7Hf6+RF7Gruesjz/2H P6VOwPdY7HqBokWHKg927QfEnW93V7rnsRsSxPr55YsNckn+N3HaI8plpJXp5Fgoiz3yCTo5 K5RzzQ7q5HiD/C0Jn4vjMtngsPeLn4TplNlpbZs2t3lTpxVTW2t2K7fWcjOc2wazoCA3z3KH amgPkbPxRrkw7VK5FuZ9n279jmZPytbHTNyeXFmcw9LsWTKdLK2wOS2tXz69p6RL9koi/cbC xmQd+lEjfl9xNN2KxqohCMnuwcvJSKWxY3Qkie1kQpIWj6EkzmWPJyXJhifSUtgklDKS2zaU kJEV1ZKWXLPDjOzZLpxCxb6dxokc5HV7KQVZJX8qA9lWpxGQFfAPsT8n/cvMzyZA+Z625IVr 8sZV0vUfP7ROAkKC5eGY+3W/tQqEz59bnz+3Pn9uff7c+vy59flz6w/3uZX1wf6Vincm/1fs X1BLAwQUAAAACABYpa0oCDO1b+AOAABUhgAAJgAAAGZtaWRvYy9ydGZtL2ZtaS9GbG93U3Bl Y2lmaWNhdGlvbi5odG1s7R1pb9s49nuA/geOi120mNhOk2IxyNoGZMtJjI0P2GpnuouFoch0 rKkseSS6bhbz45eHLlKUTSlO2k5kFK5DkY/vJvn4SLV+0sc949OkD26M4S2YfOjeDnqgVm82 f73oNZu6obMH7xtn4Mo31zCAqNnsj2q1FUKby2Zzt9s1dhcNz79vGtPmtN+rr9DaeX/WXIa1 Gwu0qHVenbR+qtdHcDcx72G9Tv4mcOn/fU0Pn4Nr6ELfRHAB7h7A7+YXc+FZwHPBzERgaD6A dxfg/Ozy/fnl+1/A9dD4+ez88uwMF+EvBtQYGLd9/OMS9BwzCAC4crzdbAMte2lbJrI9F1dq RrVat4PRv8C0fwvatQA9ODBYQYhqgDCkXUPwK2paQVADN9P+VbvWaDTxv6Regz6jsNq1GSmm dDYjgrpj/RPoXvfGt+Npu7Zb2YhWCEltxx8wM7SpAcZXYKR97GrT9CNGlYafDHEnrvnlzvTn yNvUOhQIftxqah1Mt9a97YPueKr3cV9nNfDrQDdu2rV3Z2d/q4Fe//Z2oun6YHSNi9jfs4nW o3+fUayNKf3WAUYXPxq1zxPUX/fx5+oKt7vVZrN2bWR+6Zp+DzrOu5ocvfnS9gPke7uDeHKo nQmoXRDwAGDkgHY7uB61axZ0EfRr4GP4N2UFq6OnEL6iHxnCU/il1gF/d++CzT9bV+ORwde5 8lzE6rS6HapCrWYX405qdsJWTUOXdLmPRwB/MJfSauR9gf4XG+7qyIewQWwGd0nxGegcMhQT A1dKIUJ4eUxkFnDjQ4vYXd2xA3QQHT2u/4RI2e4Cfq2bjnMQnQGp+YSYrKCzqWNPdBCRG1zx AB5NZmhNagjslx5aXqjSvn2/QryGg+n4V2qUF51Wf0hbhf+xxgxo2ooFt5Ol9DyiYzb4Nzbd +jkxoxBZnnofLdfN5dpuEleqm8iM2NDtTKb9jww0ozoi9xCkIcRGbLv3Q9PFw4GfABz1fzN4 gDErYzaVI02uXqxngP3vdd9o10LP2u1cTbFDi3EAIIeszOCSB280BjxIgaxQK2LxccLPEHgh J3D2YTjUpp8uQ2wHo1F/yn7+KeD9emlDZzEPtuu16T/UOleD/q2eiC9T3fJc7M23FvL8pFFv PJoZ0z2t1hCtvFQvw75xM9blQi1BsN43tMFtRK1I2wIi03aKkRa1UaUsqp9LWMbY+YG/Dfoj XTruhwM/mR1NxXbJdIHyCeiaoSUPw7nVOflOs4yO1MQAG9gAQzTJHIWAz5sq4akMhYPNHP9H pmMNx3TvG+O736GFiMr9Sb5+rtexikewGxIwXcoFBiakSKdzP93obLZ3jm0Bi+JATE/WnFTE szHoLgIg4tFqUlgETwJy0zFWdhDCCxggGAATuBDtPP8zWOIOAPLAHQRraAZbHy4aQGPFdnAS t6Fz0KX9BYKNSeay2GVhMO4CYDBgbQafgwZAK/6pDy9PQGvrdE7IKOLYHVoj8La+BcFgAszF wodBIHu68XwE3O36Dvrc4wUMkO1STuRBSFfJA4N80w3oQ/SwgRjJJsVSc5w0AZbpEr7sbGdh mf4CU3jl+bQ97Zr0DINTRn3IqIcTgB2xtQJ3NgqAt6S1QywJQ9LAjBXEBaSQ8PCzTcTpLU+S KpdgYRNaLET4Tpjtem49VdZoNTedE4z+pjNYJu0ozERwZiC2OwUmpdT6DBFm3sJeLukIBGwX /L4NEME6gAJAy9u6ZC2C61BBYS4B06KljRPKmTS2cdtTqmy7qCpWKnAfr2uWuBU0rdUJ8Cxr 61McMOcwaghLwMWsRdaKFJIeI5CU6kjBY8vBxjKDEIsw8C4BsxK9U2TIfY2xnEJzAf031Hpp nVOQb8mnKeObIQLuFDMH0a+32Ef3xnq/I/Yk7YTopvoni0im91aTds6cMGPRzTTlFjgPSofG 0H2Go6bgQSXLtDagY0mmQapNvBISxtd9a6B3whroIrM849ZybJqQmsv28Cc1lzXMOwfeYH5j 9vQ8x/Nr4Qgbr+u4ofRnOpQSt0swBjOGMTeHFWYoefMv2vHU23Gd7pnSRkT9TRjb2ZKSCjNA WNgWG3yxjGMRC5OHUO8wEftmrtl52uvBZI65Pp1ro0+1TuoPpkSEBRQwlSwdKBkq5b4neJyB S9vFPoC4z0vs3R5SDp0ZPsw4rsYLEcBcH8yMwahnDD72OWGkH3xTwbw0oVAK5loijLDgmYSA R9/1Gk9o2EROq0yFE4LcXCQPvx9pvUxJdUXpdL+JRLqV/XBC2Gc/3W9uP1JpvUxJ9UTp9L6J RHqV/XBC2Gc/vW9uP1JpvTRJ9X/TekYiG/rn80jDc50HGr+AX00LSWTxl+a/MdVGs8l4aszJ NjJbXGbLnkQShm+6cYyPLWTIrxyH9cKkwLus/c8r6TyrdAa94SQjElL4ZHIIUoJgHb0YZhu9 LK9x2bOw2gS4JzxWuy602N7OS2L7fDydf9Cl3A8fPYsQoI3HZp9KwvMB7vUFe56E9XuHB3m1 SlrPLC2Z8TyX1dB+OE7H2/qswSthk4llEXzoGeNp7iZTvGEkzbD4AbaNegnex948Kq00meI3 dNfShagxwF8aW4uEe4d7HnFfb2uyzASmd1L4TCkC30qVzWX7nhh4XHeI10/SSvk9kM1/5S5I 5dw+0ohMsOIfBnW4Vpx2YGAjmr89qoXGukeyO1qWt4AyEdHyxznJY2tidv88f1u9iAamAISC RP6MU0AV/ctVPxl4ndc+JeVT0T0l1VPXvOMqHvsuNBpgl8+yxJTSB8TMuR9gIBhSlL+zBAIm CpnvPPpM5R4iwRbeYIPNlsYWe1yVvIZoT7ZYlO5GAw2LbzpjZOg+xVwx5DXxMCnWkz+fkeck ANog2XV/YQ4T1zuiGYYpPieFz8jtVLLjd6fiz+V1+AE+lAhf+MQSyaS3vhR3M2MTpoTnz+Bs Qmb/1f3MjE0DOTfDlT0Pm79nB/NEnDfS8+aQ81zZE3OeT1RviCouTLOFKXMwt90V9DH/F/Ol 763ndD94Lh4ZeNoZtXC6jCrAbHsnmVR34olzAGLEAUE83MjOnnXoHiWoYjmeC0/BqxP4x9Z0 AvJriQdVx/4fLcUip2dCyO+VGax6eA1NfrsespcPyS/NccgfyItWra9OdqaNxP9jNT2wYJp0 ZGnXUd41O++jkHYdHcv5nlZN7/ZlXesU4fw1U8ypFKnCLn+a1tbNBa4r7PvTsvAkT3jkhgVv ARU8WTeT0zdCK4JReGwnPHagd0pt9/MZ+SIRcapPPhlx8k9BQuJ2SqSUy/pRIa57mLhuSeK6 5YhTTAlUIU47TJxWkjitHHGK+cL7iaOpG/mE0SyOgkTRNmoEqR4PUDItfnfroJnxu1zlTI6H cTTzK2F6xYjvHoH47uOJV8gBVTLNYsRrRyBeezzxCgnkB01XlfBM6k9xky5DsMohEzmNsnyO LHmyDA9FymRNZUQdzu5QIoDu6h4kgO6/liOANlUhgFZUwpllLB1EmqUWlcOatVVBm9VUVZY4 C0VFZ+K8lNKqE0NQoUQh2UGJzJyRW5aGWY6w3HG8cJJfQbEd8mqKiSuPFedhj/eEoi3Ig7IO /hCY44n/lbDo5XJHhHXv3tSRH2jpm84cKbMAfqosj6w6SXbdJRrUyrm64nHJIeKu9r4ckfDz uFSR8JPZtFfFLjclRKy7N39ErJxJIxHtrlAuCLkmIWxJb1GYxBdgXFId1KMwmSCgMI4F6uQi i+y2R0MgpHXneNbnP7Yegh3h0YBck2EHydUb5JIIhra7dZwQ01PqP1g/IoDM1JHo7wJuIP7C 7iu8hyPcJuA+rWYKL45WKmeeyDD27SV3bbB7Pmz3UfQT0Fik95AhKufEf+oX56cX5/+N5CaC p6Vn4VNyl4tL/Gv6Hg67ARv8FpUAI+YhuQIkie6zSJbz0Ag7uTiPevF8KRb1pAa7OGS9cTAx Ql0RMyJdMVQGJl5g0/tEMJ+F9kwY7G6SbRBeJiIEBFKdiK3hvUlrRGCWwgUmngvZHS34iYON Bzt0AQQdakwXy39Ndi6wFlg+NANIeYx/Lcw727HRQyRVoXks40uRieGNOYBdWMN4KoQiGXcB eNPmBf/2NLdlHPtjbaOWv6i27Iot3/1DtakmNj1/f7gpnVRGdCaUxrr1tnGIVH6CwmNQV6d7 L5jiTEiDS+AU4wgHI82depY90Se840gok3o/cWjkvaA8q+aJ/H2qsyd3+vHoruz1eVbI3Wk1 ECyqgaAaCKqB4AccCKIljHSun58OAtRdnuCusAMn6xN2RWGOCQsQaKfkjjjiN1I4iYZuiqbO dSbrqMCwIeGSWmrei+IUv3Dl2SVkuhQZNLlxMhqzsNuOpgCx30489R5XyxXHpifbBaAPZAv2 3GY09Fu8GQtpF2+XDjgXb514GVEaip1K/GVRBCQgSriz6nbIQ7dDNnUWYZQHnY99cudpoorl DvyEHzGmeCikWPL4T/jJBBQLxBOLhBMLRxPfShqgle/tgiSS+sH97Ho798YLUP+rBTdMClwM Uh7IbyfnfjLpa3mnfn6gGH546KdM+F6auZ61Ekkuu8RKZCFvYjqS1ixvVQwfK6eyi6HjKURb 303ixmpAKu/8OO/M597naQ3LxpeoS7i/l66lpBZR0v0GD75C4II93BeyKBmlKBWYOBCLKBF+ KBxxKBdkKB5XSIcSquhBFT04wBHF6AGdYRdx9tQ1VI79OI6dP/mT597TZ4EOOfl0XSVXv+fg TwGtUAqCPC7KoRLhKBHdqFT5CKosORAuVebMEfFiU9xM8/0arnZEXEXLFSFVmnQUTdo73U2O uu91hUk1dR2pZr3VrLea9Vaz3jzPX019j+nlFea+4nUbBz2++uxX8XKNoipSzYNftF5nrhSQ qrVwycBerRbq7ldq8UaBAX0lG90t3SL6/rlkCFXf/lTf8SywyVl4X1N9K/NRu5fFNyzVxxJe PNXq4ZH2xx8kCV8wSqcU/JtC41eFhvbK71sVfif5nYeQt96/ayXsSj3za8kZhtWbyas3kz8W qerN5Ml1RdWbyRM2lSNNrl6s5+rN5NWbyb/rN5MTGN2x/okCuzGGZAryf1BLAwQUAAAACABY pa0olwVOQy4LAADvNwAAJAAAAGZtaWRvYy9ydGZtL2ZtaS9NZXRlcmluZ01hbmFnZXIuaHRt bO1b62/iSBL/jjT/Q61He0o0AecxWp1ygAQYEnQ8IuPM3NyXyDFN8I6xWbsJk9P+8VfV3X5h A0kuuZ0ZxRqB6Ud1VXXVr6sqPfVfjHHH+nLVhUtrOICr6/ag3wGtquufzzq6bliG7PhYO4Ze aC9YxLiud0eaNud8ea7r6/W6tj6rBeGdbpm62e1U53zhfTzWZ2p0bcqnWvNdpf5LtTpi6yv7 jlWr9Jvoiu9uy1D9cMF8FtqcTeH2AX637+1p4EDgw8TmMLQf4OQMTo/PP56ef/w7XAytD8en 58fH2IQfkqjVtwZdfDmHjmdHEcCQcRa6/t3Q9nHlEIfo8Zj6oD/6J5jdATS0iD94LJozxjUg dTQ0zr5x3YkiDS7Nbq+h1Wo6/kvH1USfoNXQJtQspNRjcdpj4wu0LzrjwdhsaOu5y8UAJWgj eWBitUwLxj0YtT61W2a2S8rUwp4hLuLb97d2eMODpdYURLC7rreaKHWrPehCe2waXVzrWIPP fcO6bGgnx8e/atDpDgZXLcPojy6wSf6eXLU64vex4NoyxacByC52jRqnKevvu/j0ejhv0JpM GtrIvm/bYYd53olWzt7NzA0jHgbrvXzmWDveYO2MyAMgc9Aa9C9GDc1hPu6mBp/Ub6EKOcbI MNwTTxnDJrvXmvA3/zZa/qPeG4+s/Jhe4HM5pt5uCgOq623knUY21SzdMkqW3KUjwAe1lDWj 4J6F9y5bV3nIWI08BpcU/PSNHDOCEwsHZRghXb4kM1O2DJlDXlf13IjvZcdIxr8iU64/Zd+q tuftZadPI1+RkznzllXEob2MXOLAPXzo0tF04QjyzVCep0w6dO/mPG/hYI4/C6c8a9a7QzFL fcnJkmjWizdgpyjpaSzHpP9vdN3qKbmRYjYvfchnC322cPWeF6wnS+a4Mxf33g38WB/t5pXZ /STXkOLHcu8jeR0hJJvMnrIwJTbq/svKE0v0mejqefKV25hcGRCEL7pWQ1Pw2m72TES1hAeA LSJtnC/bqI3GkCe4IZQyjGQHc/tfEO+sXLzJ9XDYMr+cK177o1HXlK9/yq9evzswci2JHO+d wEfMXjk8CG+i1WJhhw9aszMeTSwz3dHCrAXj82CaThh2rcuxUb5rz5DJ6Fqt/iAW6LHsTxm3 Xe/x3MfjtzJfcNv8Ed6A7sgoPcHVEU5Rjrk5Lz34hS7AaFmttFPFSKf0mVWLOHPJg2roQYpN ijaIfHnIgyGJoIJeil8UVNU827+rjW9/Zw4nu/mTPj5Uq2inMeVagUhbaEASUdIYIn4zrOZy deu5DjhifaRSMpmGYUTF/GkEmzzUdUGJeMQva86g7jZDZntV7i4YzBB5YMHsaBWyBUYAdd1t wgFxegh26BAEOBz7YMpmri9ixwqYvQ7Ghb+dARoG7q4fgY00kK0jfAkF5uCbP6Vm5brQonXF IFqiArbzx8oNWQRLFkouVoRYMLW5rUZLSoIjJ/A8ZCQS3TALgwVGrqwCQQiLALkThCMKbJc4 ErUDfM4WyMvdyrND7yFeP9aZIOnP3DsULaoIXkmWMPBIFoysIZgpSSLRKxeoVaC+ROb7PgSr ENzF0hNaE5hNU3BVIOXldHckmtXaSNVh7j2tulA7iU1/rFiE0kXiDHigNlsqBQVM+KGGCHgA t0pgNkWGejiE2c48ppJfzUEhOIsy+xIxqUkxiqgcRIcVorriruf+h8XKUMqRIom5oic7rQZW 2umiRCFDeclK7Ajfl94D0ZUEBG+x/jLTkp3NDEMCrj8LwoXUa7KLgJlKsr7atghBCbl0OU6h LbYdJ1j5PKI9rimrly4V/8jhhMBxBRIK4jdwoiStaIBAy8KE7XNAwuV1xxqbW9dJov3S42JX tH+yEe2fFRKRXNYiT8NM1NbBJxO1Wfatxy5xf9AQO4EXhJo6ZZIMJnecfBDHCUX1Kd8wkXzn YraN43hbqCGWN4N1dulmvTM2unTabwt4yoKF9xuNB4daET/pRGo3Dw7rulhC7ISAfHmiPe+z 5XmBozzPZ2uoO8GUFdcWrRAIpJY+i71o2AS39srj0tQpSye7J5IezAN0JPgetCnOGh/T9T5+ tKZTdMToCP0QMSg9hiacpuxQfBkVqUSSNCZbgY0Hl5GjlkHIi92bDMixTrBYrHyXP/zlu/0d 7WBGS8/cvYKeaedG9oL9/NuWxK2SZOGsABn47jorEtzfDPh/AMgfCpZfGu13JeyxVL9u5DSy YCasQ27F4zPjTIu0a+lbm0nWM30OLUvSFmG1DP6PIEkGCrl/0f2UV+IHOmJCLnHBlKwUXIbi Gdd6atmh0J6meTPVlSVf7rPB2s9zkXj+HHcVreVraSdlJcGKv6yjd0QUjPEiqPAQMLoUR6qM uF2MOEmyfd694anRjevPccMxYL2hnORGJGk3mynY6zryRsVNuNNkdVviy83EXzFejhmXyZTM Lou5Y/tFojbHw2TtCN5VMLy3vYjeMJm0Kdug9zvGRXZN73M7mncQb+ndDzgmQ+kbwjX94EHs Fe8qa9vlm9+J++7B6atmWWgfx/ayOvKI0D4f229M2xnax6WR7wnmT/ZH9oZgezvUJ5rOiF0S h2elrl+e4fjCwStaVW1FlUHKCyAHh4VAgs9DypaT4PLa/+ojJF1iYNL95rClQNrCLPUIRB0F k5UzF6slM5IijSrPGM1XD/StuRtBBTKGA+gkErwwwgI2mxFNG/Ph0rUPUh1kAuwaOt2AViGF HByiJk5+OzkCDQXh2mHML7e/uqIWIRmajIZXgDQRP5FRCuBoFqxdPicGVYym2CBKCR1ZsxAM TWuktrS+RX+DEXt1LktZRnPnnkEV3FmqJLmDpCnoI29SOHBsHyGD1gxuqUClFt26rQnRkAFq 2w8AySwDn9wkI7VcTBUh0j0SCpDq8FeLW9xG0oosoyidyJkFxRxRfUeJI/U6DRitzyFaLQXF pN4Bw367phQ2YQxaXhScQ6yz/1MWIPF873wpbjmRBJ4pepFWoBsSP+SvuFizDTuewO1LQsxT Ex317Ml31LMv7fmR4E1WBStieFyGlN8StYaxNccF0zQGS12tVpHA95KwR5jXfqDdOqBtI9AT m5K6aQx8RYi6sunKBa6QwpT0hdgAlFkjkhCLAhZ84hPRAuMkZy6aMwCKsoUr30eGU1rETZ5O FlRWkby6kYBFOjERID87heRIWJWIejECRoh0ZYk8JvTjwnBqPCkU59tLgFlS2FTP09H4+fj1 6KrZy2PY00tt6nldKNtqFbtX2magPzeUqQjuVdBM0c9DScarf1xE+7mBpyyXVcXHQjK7rfT4 A6WkQ1l5fE42+sIVuSJEZ2p0ZeD83MIkwXqG9GNKfyXPK1cDS55HFghLnp01w13j4zLiluHP PnPkI+aZK49FO86bxxcd6ZCJx2DMhAFTRYEaDrl3bTEp+du6PCNyppE7s9Tf2FvyAoEbZZZD rJ259wjCKf5DFKxCJwvwKLtqywDYEUwxbXV9+Rf5zOBcO02oJDPEdYHQ9iOJYA9LlvDmeerE jLLMKOERS72pY4fT6Gjj7GFiCCoHc1+ax70HGm8nM55w1InG/OkiAVYebq64aaFuxkzTabHJ 52dGWf+IgwE8q9dB+FVuckpB2H1+umxKT2Fl7OmgpCUZoww8Q0c1SLFNYS2pzKjYPXaj5E6u 9CBfGSegCEiZ6NOLEFuvNwreS/AlqRBQTfcJNxzFvJyEe2oN+UKuumlWcmUsuTNmlh20T75m fhtwHix2H7N/7U1zyeHbZfO3y+b/K1Nvl83fLpsrYm+Xzd8umz+J/e/2sjnRoP+EJoip/2z3 X1BLAwQUAAAACABYpa0oQZB72A4JAADHMwAAIAAAAGZtaWRvYy9ydGZtL2ZtaS9Vc2FnZVJl YWRlci5odG1s7Rtpb+JI9jvS/Ic3bs2qRx1skvSHVRaQAJsELUdk3N3bu1qhiimCZ4ydsYvQ mZn97/vqsPEFuYgmsxsrwlDHu+rddprfm5Oe8/XSggtnNITLT93hoAda3TC+nPYMw3RMOfFR b0A/IisaU2YY1ljTlozdnBnGZrPRN6d6GF0bjm3YVq++ZCv/Y8NYqNX6nM219ne15vf1+phu Lsk1rdf5bw5X3K2OqebhnAY0IozO4eoOfiK3ZB66EAYwJQxG5A6OT+Gkcfbx5OzjX+F85Hxo nJw1GjiEHxKoM3CGFn45g55P4hjgU4wIbUrmNMJpI5lvDgfjv4NtDaGlxezOp/GSUqYBF0VL Y/QbM9w41uDCtvotTdcN/Nuu08WcgNXSpnxYcGgkrHQn5lfonvcmw4nd0jZLj4kFislWesHU 6dgOTPow7nzuduzslOSngzMjRBKQ2ysSzVh4o7UFEJxuGp02ctzpDi3oTmzTQlwNDb4MTOei pR03Gj9o0LOGw8uOaQ7G5zgkf08vOz3xuyGodmzxaQKSi1Pj1smW9HcWXv0+7ht2ptOWNia3 XRL1qO8fa9XkzRZeFLMo3NxLZ460RoG0Uw4eAImDznBwPm5pLg0YjTT4rH4LUcg1Zobgvriq CLbprdaGvwRX8c3fmv3J2Mmv6YcBk2ua3bZQnqbRRdr5yrbaZThmBcp9MgK8UEpZNQpvaXTr 0U2dRZTq3FoQpaBnYOaIEZQ4uChDCJflIYmZ05uIutzi6r4Xs3vJMdP1L0iUF8zptzrx/XvJ GfCVL0jJkvo3dfRB9xJygQvvocOQhmYIQ5DfTGV5SqUj73rJ8hoO9uSLMMrTdtMaiV3qJjdL oFkrLridMqcnCR/TwT/RdOsn3IwUsXnuI7ZYGYuVZ4womp4XXI9IgN40SqTRbV/a1meJQTKf cJ0AHFv/cJJ5JZiU6acRWq0skiJAb3puOS1N+cluu2+je0ppA8hTl0LKBIldkMYTyAMrMKRO Nz2G3CGWWDutZm36aTTq2F/PFJ2D8diy5dff5a0/sIZmbiTl4Z0bBuh41y4Lo1m8Xq1IdKe1 e5Px1LG3B1PataJsGc63G0aWczExKxh8Gk+m5XQGw4Shh5I/p4x4/sOpT9bvJL5ke/k43AJr bFaGYRWHeZpiF/dto7eQBZgdp7OdVEnOCf/MikUETm5ZOlqWIpOnDBx8OWfBnEJAQEPDG8+I dJ8E1/rk6ifqMq4zv/OPD/U66mgCVc8B6ArOJQDFhSkSL9Np36yvfM8FV+BFCIWNfAmmQjSY x1DE3TQEFE4b3j57BNjSixWoFSXxOsI0LqBsE0Y/w5oDhkUYAYEYPYmPP/xwA3KAMggXNTES Q3xDXW/hyRwwXd10wzlt93HFVM5jEPLCoGmIcQgFTeCSAK5oDYjr0jimcx0GDIgfh4DqhToS xEilQh1n4eg1aN60a9AJFKacJHI4OJMRTZJUEtTUjoKTVLuOYLOkAQpETiKfgoItk5yWGgca 0V/WNEawupLp9pjwZKaUQgcZOQN5MGZ7l7MuyShx172JaVVJUIxzozkCkdLthGoSRkrA+OAD YVSFkXcB3UgxvxfqK9YcQarKJXqPMpo4ZRzcEXgBEx8/JoQVMVUiqcFjrjIhJewZMSjrkPaW nGfOeQjnrjyH8vsF51FRMLRAuNDSht17QPrQTz1nYu/Ek+bxlTFkXx5/XMjjT0slRq4ekSEy k4/18MrkYw658ukFnhQKthf6YaSp0JPWJrkY80HEGJ6vb+mGqaQ7l40VYvSu3EOgt8NNFrXS KESyS62L2cO7zEBG3/7172eodc4dcfXqtnOgZWAUXiae5TX7sX6iwkMk8XehpvIIioTLteEm KNoYsiLnlpjo4hn9vGOaeSsarlN7Ehoq4qOcftpnqiIx+uzdTn7jsSVGHV4McV8tJVoDFmJc QRdN5vrr0KYX06WMJr3p0bP1SGUvQpw5LYJdqanEUvL8IHPbfZ4/9eLFnP5P4MBHguRD++59 hXXC1Q+FskU2toTCyKO4Db15GtkLJdETzPlatTjttU/7nk/f77ZjtNXi6tRgD2yNPB89rJ3I zz/++A4iqIOePyuheC9OujyenvZhj+Yc663qMuh1H1em/thWHgc9HO6a+WHw+wsJn6MT0nfX UUQDJk9hzlnbHxIK7j2eecESixwsGGeLKFzNROU9K5bpL+v9C+1UoRXT9VVFAGinTj6GlHDg hKuWQbm/0D1I4u76YUB5VYrlNfFj/m3hBcT3fhWjaHai68K/L0m87GH05t+DkHmLu+23ju/z HyxMEqrvahviseI91cV7gvtlu6q6S8o72TV7QHWXL+8K2/ZWd0nL7DXlBsf3F3emIHt3fpBK OsP2wRPorMyaF6eILd+x4yOqY6cabOXW2kMybnW9eOKtrgfn3+q6Jw0vrkqy8dwitox4z08I YxxO1+5SiMT65tIbcRJVIMVqnhHF6bq0wal6Zmb7UHm6I7qaYZBoIDoJ2cEjKwp0seAtQRLX yhjeB3QDqjyH3yRo+M8RJGcw+1HRoItuXrbZd0n4w3NeeZ6l3T7hV8Sg8jFQF3RIuNn6NF2d YMpvyAX8pB8pgpAEIgDSLByhAnkgcihdoc42s0YNSH4cccopLztPGzd6iwxXLgnQ+3KikoZu Xh2I7Jp6vNPMe8tJl1VP2m+GKZ1Dvhm33zE8q09zYNewt7Gjrjf3ULhewD3sawc9znjjWd6Q 8pD/p+wY8y7+fGcVRjQhL5Ff9iob+SPtuiqPUt2SUiK1q1fyJ0qHVKvkKZnQrtqz7LWqq9Eq 93UI78OdYDVGWX4VjXdPAVu0RpuydRRsdXbXtlfxkG2k9Pz/6AHZ/gid1ONlBZUV+iMVckcD gaufhFetbvtK9vv0DYMLiSJyxz3g9hm2xJx9rBwnT6j5Q2Li+xnZC2zoG8N1gMuu01ckRfGM ZGWWPugpuf4kbX/uw9+nW8zDVOVxfdUqj1fstFYoF+8FS2dVXPxiLdmcMire8y0A9e5KxUso 6VsoSlbPfPv0KmQsXO0Pkn/sC6iSwrd3UN/eQX0uUW/voL69g/r2DurbO6iv7x1UDoP/g4kA pv6J5r9QSwMEFAAAAAgAWKWtKJIrMw3mAgAAHQ0AABsAAABmbWlkb2Mvc2VyaWFsaXplZC1m b3JtLmh0bWztV11vmzAUfZ+0/3DraXuZiMnHw5QBEh8mQSMQgdeue6lI4jaZSIjACu1+/Ywh WZJ2ylS10jTFD2D7Xp97ru17SLQLJ7Tp9ZjAkI58GH+1fM8GpGB81bUxdqhTG3otFdw8WbKC cYxJgNCc83Uf47IsW2W3leV3mEY4IrYy58u0p+Lbxrs14zNkvH2jXShKwMpxcscUpRpXuPJN TKexw4CtWJ5wNoPJA/xINsksm0K2gjjhMEoeoN2Fjtrvdfq9TzAY0Y9qp6+qYko8alDqUZ+I TszyRZIufgokN8uXwoK3Js33gi8QER90VPCHlBVzxjiCahd0xNk9x9OiQDCMiLvv0ZKzEkVH cTUt08Jb/lboXIM1sEM/jHRUzhdcOjSZ6bsGMTUjCqELgXlpmdG+qU7CFJaRCLJKNpMkv+HZ GhkSRJg1bBoiTdPyCVhh5BARS0Vw5Tl0qKO2qr5HYBPfH5uO4wUDMVWP47Fpy7EqWdNIPh0Q dIUp0Du/qb8jormuWOebcayjINlYSW6zNG2jp+nd3C7ygudZeZLnATX1iFq3ggcQ5MD0vUGg oylbcZYjuGzGcitqH+cvCYNomhsG9NDqZisurHaaFIWGK7vxYTUp1p/FRXGeE8JsLky2Yflm wUqF54y1qmIQmyLje85RcM0yqHDSsGU0FKpdexkaM7bO2bQqJSVdFPwkEWfn/yp0FqsZu1eS ND1JxKs8X4XDnKVrRQjKSQpD4XiCAa4LCMsLXvecpqKaq5ov7ub88OZCFF7JYusaGhnJVc2r XlyD7lfnkZw8zrGzzSP2vouSVDpVedRkxxG5rHvbmYB8o7ucdnyfF+P4bOtNBSFtA0J11IiW ZbiR0Iq43k1TnMchox1GsdNr5Vbo9Z/QghAOAY/SeXQsh9KrAwmcJ5W3kd7qcyQRbBJQInvD tvHocyJEvy3j7LlFL6L1k4zzbPkvy33N8Kz4Z8U/K/5Z8f8bxa8wql/wEqz5a/ILUEsDBAoA AAAAAFilrSgAAAAAAAAAAAAAAAATAAAAZm1pZG9jL3BhY2thZ2UtbGlzdFBLAwQUAAAACABY pa0obHheTeIHAACeGQAAFAAAAGZtaWRvYy9oZWxwLWRvYy5odG1s7Vlhj9rMEf5eqf9hXlet EvXA5C4fqisgATZ3KBxQcJKmX14t9h72G+O11gZCf31nZtfGcHe59m0rtVL4APbu7OzMszPP jE33J28+Cr4sfLgPHqaw+DicTkbgtFz3883Idb3AMxPv2x0Ya7GVhSxd1585TlyW+a3rHg6H 9uGmrfTGDZbu0h+14nKbvu+4j1a6HZWR0//tb7o/tVozeViIjWy16J708q8/8Ow83MlMalHK CNZH+EXsRaRCUBmsRAkP4gjvbuC6c/v++vb9n+DuIfhj5/q208Eh/DJKg0kw9fHiFgaLCdzL NMdBtxrtTiezD7D0p9BzivKYyiKWsnSAAOg5pfxWumFROHC/9MdNiTaPspaes6Jh9sitTB/O vS8wvBvNp/NlzznESckC1qle/YFVMFgGMB/DbPBpOFg2p4z9A5x5wE0ysV8L/XOpcqfPSnC6 6w766OFgOPVhOF96Pu7VceDzxAvue867Tuf3Doz86XQx8LzJ7A6HzP1qMRjxfYetDpb87QGa i1Oz3vXJ9N/5+BmPcd10sFr1nJnYD4UeyTR95zxv3s+PiS5KrQ6v2nlmWufCtBtSD4DGwWA6 uZv1nFBmpdQOfLL3DIWR8f5JgwE/3fF8FpzPjlVW4uwoFUXRdWm+/4dsXeR/xkDxfs0WAxsw ai/1PpGHVqmlbFMeICi8/8S72Lw77Aco1HWHfWsCofafMSOSuZYhZVErTYryVUO8Wv6/Yk6S RfJbS6Tpq4ZMSPJfsWHMn+dsWMo9mmEXvxACLIPbEks0dr3c0TUJ43JAmyvPZpANTZ1s4vI8 UmE5/8zJddPv+g+8yv6YxUZpMxsv6OOpT9cVdKvJ3zAFW9eUDsbYxdL/ZK6qkZn/16BGsbb3 1+1xeZbmHAGp7M4Peo4lqWF/vERuWBkkBxgD5xbVOmKEu4W8/pKa2RzONV348eQ8zjm2B/7M e5ZiLcdSyWENI38W+Hx1/w6/79UBgjgpuHJ4KtxtkYFgUsBcb0SW/J3SgwW7br2yln8zyPM0 wSRKsFottNpg9dsm2QYmxGKPIpRvIap0xqKAHOtgAaHSWha5yiKSLRWUsQQ8mm0BScY3yLTJ xqhFxr2CSBahTtZYIVHJo0pTdSja3fsbNGYhwq+oFa2ku+5wOh99+MvHeeAbpxf444swxq1Z js0QbAjuhAU2xKQQSUaDxBugHtEUtJF4Em0VWYRGWW+KKzgkZYyixW67FfqItmiQqL5tQGS1 ocgqrTi/00Ass1E6kcVt9yNX/+mkX0NUwJukFAjjWxoemX3p0v8WypwwMHdaK43MbRS4Z34a JHipW+t9FRF28erkHV1iI2LGreN0XwswdgSOOmRQyFxQ08I+t4E1InZ4eoW050ziZYyEj8Ih O0K4FIgynTsKC7OXe9rBHDR7fVWDXIp1StiTSZFEXFMMhK3crtG4xoImuowFWh5LjeBmpDkR FJ406SXI/CWsduvwhPYgTeFDRp7h+OnEz6cm2zyVFM3kQOOoRk/cQM2aA/j56dpoeyAmIght Y/nKuE7D40SmUXNghK6WeheWGHuN4QdZxqoWbOg1CjwG7nL9adQutwMmzPhQq1NAr/G7ThdK U+6C8HDRMYL4Uastj9eH1HCUU4UzjjKd8kWeqU4o2bQEkeaxWMsSaSXFdItRz4s6zQrLGUpH knaQRxB5LoWuJgrMQbQuVJGs0hTpB5sWabxI1YY2g41WuxwPtgBZUMgl2AVzS05CuWU3qdvP Jh/1NfDGHN59gu28DuPj22dzED0nq4lxvttEXSjj8sIEQ0hiZ1FRGmZGnu5IXVzJ1rxUydgE PQkYpnrKfU3eazBinQ/m4GopTbjbQkFQNTPOBNlO05XQnDLMnl06iT4957RTkW3a8/UvmI5d l4eN/tN+WD8gU2Wl2MTY9zWcWOBzLDMgWLnOoN65hZndv0I3kvArc1EGDp2ggzRR5Kk4msg4 x7MJefuJeqop6GS4w8SvxK4sl6omiV5s/fq+KkuPJnOqw/xeFTj1tVSiXwrA1/rmSy1V6JFI wUgYruc2gG2LxV7CWiIiJ41tGDTuWBbDno4ThxQmEzZWEfu4KxCTDT8Hp+httJPUFyRbTLs9 E66lf4ErEaeQxyqFu2LHqzbJXmZtOLe9gC0+Qq8lLtyiLipq8LjjsEwqNmeiLp5PbNuevwbj Zb9vlyFupyTLGuRWpxZhadOpUYvxOjzRNN5tmZ4tCo/E6C/Yu9By787wod7aHHBBTpPsK8aW +CrhqHZVz5WhGAUn4rVP1K54piWoIxnFtEwZVa74ZjfzdsSdKfue5JlNixhbTLI6TiJD5fxu xbwpaQMMOK02Ff3vkeWp3huuwF3pV+3KakFV2HifFXICtk5EPmOlt3Z3U7bsFOsiNvyGPmX1 iElM0ww2q5Rt/+rVpnQx3uyEPQdbSJIMg3drhPC2YkqsHwSxlq06xPgQKfJxPJJ7mapcagrd ipowVNvwuSp3pkBkijF8qSumk6RmcyNLc6IXFiEfb5RtsUV2rJ1CtE493omILsB0qm1xgspy oZyqj6uS32hpwFefTvNxyrxE4QdBBo0ehRBSdFTQ4wNVYcWpXD0qGOs39VuxE0hYSLJI6IhE U1m2oX7A5IclylDzaGOfM//NF1FrVZZq+7/8LspY+ON11I/XUT9eR/14HfX/9zqKm9i594WV 2f9G/gFQSwMEFAAAAAgAWKWtKJLXlGm4AQAA9QQAABUAAABmbWlkb2Mvc3R5bGVzaGVldC5j c3OtksFq40AMhu+BvINgbyHZJO2l1KdNNqEtpSxtYc9yLNtDx54wI7tNS959Z8Y2Hqe5BFYn o/H/Sb+k+QQesMZE7cDwQRKYnIhhMh+PxqP5BH5TKkqCnZJKmymkqmQDWCagOCfdapBZi7hi MmCTBKxA1aS1SOx3TpBQipW0wp77BzOCGHdvmVaVxfkC/jlWyQG+greZf7uFH1sfcGwRrxjL rjOv/Okzd4SJKLO1B7oIWZaytuEo4NyhfoMCq5oCwEsVDxkngI2NDvAospy/EZ7Ve1/+TAud EUf4mwumfi5bO2CoDCUgSpCU8ix30041FgRSGG6tbl3iVbAkr3Al3G5mRnzSLZRKFyijJpVi IeShS7rxNerWpNdfrr5nKtrSF6m9y80HFnu7PJWCsXlJegoGSzMzpEXqpc6/d90YtqLTuoOq y8WeT2rekayJxQ6n8EsLlIMSx37kT1iLDFmo0m5JBycenpb9aYV6TVIuTxfa3md7F9Hx3FkE 8meqz8oXNm5WXu7PciWrgdiZXgbT7my23gK3vc0IAvhiER2HtKaVy2nN/UZ+m4G1q4u7+z6F Hh2Ar/8reDz6B1BLAwQKAAAAAACMk60oAAAAAAAAAAAAAAAADgAAAGZtaWRvYy9pbWFnZXMv UEsDBBQAAAAIAKt2sSaWfRXkgwgAAMEIAAAeAAAAZm1pZG9jL2ltYWdlcy9PcGVuQm9va0lj b24uZ2lmnZP9V1IJGsevvAmCgZGGSoFmiuQYApk0bQEq0kSNmhr24lF8GW+jjSXuscbrQURw S+daY2bTTt6mdpZjNsttraxtSi++J4U2pdV1ji8Ts9sLtYWbs0249i/s88PznO85358+n/Ok bFElyPJEwErgLQAsLCx4vd7379+/e/dufn7e4/G43W6Xy4XjuNPpxDAMQRAIgkAQ1Gq1MplM KBTyeLygoCAmk0mj0chkMoFAAACARqWSicQPwccHTI5beOtGTxwik4jY+Wpwt0q7VQbqMp3/ aIZKttOoZG16MplEQEwpi21In4Nd1DP9qZ6nDthSCuq2adM344MNwshQHpeNnikKWsbALu4D d8e5Z29A+kzXWAM+ZETb9kIlse9eOzxP7c7+NvfsJdRqhC2FoE6FD1a6J2rQMxlBy/zmX95y 4e34GOLsb0ath9TKSOf17Hn3ZfeMFR9tcfYfwa5Ww+YMSK8CdQmQfrMwkimTsHlcWhCb7Hrc 4p5pRv9ahLRkw2YNpI/Hx4wu3OjsL8FHS5GTqag1AzYnIidTYMsGUCeEzWJQtwrSR2nTg8E8 LmpVICfXwRYhVL58YeGt1zv3/v2bd+9ee73P1epEBIFhuNrtXmR8CUVb3O5+BKmTydbieCcE FfJ4wR6PHcNOu903ULTB5erE8Xansw2GSyEoh8mko6gRxxGnsxnDLAjyucvVguMNi0LU6hgQ 3Ixhehw3wnDGoh+ZLAzHKzFsHwgm8HisRV0gGCcUsiEoXqtdTaORZDIODG/g8RggKISgWARZ B8NCEOTCcDiCRMBwsBUtsqJlrUieBc6YdLXfw5EurPrl/OAzj33S1XkPbx9wtllRYyuit8CF XZglH1RNum7ewzsHnNZWpEoHbuvCTldAOf3OS1a05SRSZ4Er0rXJXdhfKqDCNtS4B1RZ4Oo2 tEGj3WxF204isA7cXQ2XwkhdG9pSChXuATPnvHNW1GqBLXpI/3aR5tuFuTnvy5fzz555Zmfd Oh0I/L8T9hvJ58ON+bA+/ARA9QI+VK7axjd8vGIDHiew2c9YfIM/aebI4aP5yvOaYwqV+eTa HSEBwKvE4WMrUoab49Jdt5dHpzULhiUoQhfl7psi7ahjnFrR2v6KX25P8VtVcNjdleRaGiUo Un4bJxlwNG/8lRLLnc+lhCSUxPiHqhpaV+dh5sn0eA0nMCLaql56xToR/gm2/19P+qKXNcsu P+84IkUQQ9DBgt9Y5Rfq0ULHN3nsIvWpsHDHftSmv/Diu7vN5ujbnw9lRaWVax3tmX6KpJqn 6vuHnvTfEw00WRSh5zUPkX1mE5ob9jztcrI+8eqPmY/qretsur70EwhizNrff0OZP/QoXz35 lXeNPau0hB2U6YwN6Y3hrL7y5MChpwVFjaEbOjM+qvTP7WnZabUFmFc/rBzXFGKs/GzpJMO4 cldtlStnXXFjBR0q7p6hfLRXlJpKvGCLo4Re0V9lFAz4ixnxN6J3jNTW5toZq1+kzQ+KWn5R moODr/vQl1e9SlGH5pO0xICopCDHgYkk+wg/NR0jBGICNkPkfjH2WHRrb7EqJLPvUjLp9xU6 P2H2+TJ63AN1OSI+upHdKB5l5KvZK5uAJ+E9VS71Z4emu/KZHV0Bt0OXsiIASnZBzXBIi1zn EE/q2Q6pXpWvlk4xZhTs2ogZqvg68Petu74N1tGCVSZUddZblpt7J4tScLwuOrs+b4vkYSrb f52TZA/9sjaJTaUquoc0/JEIosZ0nH9VGJxbP00Rbqpidcv9Co7f5mRXzvSumuYoG8XOiDfb pHc7H/tKEpsAf+k9zdeDJ275j/0a3srxoYR02HrMZAIjY/vS7IbjyZzt8vT9UmeJxMF3bpQM SmsijM8lI8BsnKjxVT9DUpE2vTZEh8xQ/ChlFcMXs6fCh8Qiw+wF3lDqxrXS8WJuL39k+2yM OPW/WVGBBR6DuV6fur6J9I30cLJ/onRVo2MU2DEYf1fzhiEdjTE+5w8lPGOIb20MHhzHUo97 QkoQ7Eg1MQ8pC7r2k43qlY6Z2I29g53P5iTDEZiSPaEhNkjuBI2510PPzXrmg5tKi/9KAr87 aqeg40LEtGnHknHl29GB+EfyN8aJr6Rf/ifupuh1wY+fhvuboU3O0WQi4eIbxfC1w/wGKV5W 3rh+6Rfnfo8fNzkI7BHPMYZ4gPvmaF+N3J61esu5Gj9uRfG/DXJ6la4v8Nqu11TeVNLsNt6j hFWo5BFD6WANZu35m8QA2OfuNJUZGd4nxIRuBZ3FqiduCqN4/tRDwxDKuh98Gr/QJFbveUyZ 6A0iFXsUxXcSTvcJvo/eknrW94yoL/Rn2xEa9RefLaxarGPKJ/HPfO6Seops8BjjSfIPh7y6 rUnjx3s295faQx1HVRJzT+1K63Tgmioi5QrZJ6R7xG+7qJfqyoiTc4ksg7pfuqmVyowIZ+V0 5MwNUxo+izAGb0OUnJ+m/HYOzvgvF3DI8mPksmwFp0YQyvfroSqoM8z6eXucX4BJHKugGuUm +uGzpiShMaBBYIr9PrVOVNDjm3ajWx2D0HaMlFGPqikcetofEkPEvnXq/nCtkxomIAQ0J5nW ko7V5ou+C/i6jBS4j0/Z2eEMFMoBn1hDXUyJLq9Io2JRrpE04nDf1pIkllxuUomUAVdJBNbB JmJY7kFaYxmJVi8zZ9zo8RVyUggZNhq3c5qmPScwLsm1ETjzbMoqQw8t6ZUpql5O4goUVCWr lv2qnNxk6KVHbkhbU0KgJ9Qncn9GCH8UTBGkNgV9k8HIbPIQI7sBiiC1pqVoirB8wkCR2nxO 5UwRhX3yiMjudLp2ihjYbfh0Ta6RuYevJqoHgJbDAJGDAGS+Id0XuE+IOQEQqAbgNR84QOID lSzfDd3CJRsZcqDyn/tWPjhyv5RMAj7+H1BLAwQUAAAACACrdrEmrmXayrwAAAD/AAAAIQAA AGZtaWRvYy9pbWFnZXMvYmx1ZS1iYWxsLXNtYWxsLmdpZnP3dLOwTGRjYGP4ysCwf/9+Bob/ WordjdXH3J3OCQucy0k+x8JwjYPhPwfDe3aGY9XVx9gZzokI/Dc0+S/AARR8zsEwp2f5/6XH /ksI/Gdm2MvKMEVEoEVD8T8/xzUmhhYWhih2hntsDFvZGZL37z3GybCXjWEtA2MUA4OXgEAx C0MxG4OXt9ec1ORzDNQAij9ZGEG0DogA+YmBTcahoYDDgEWAK9HBaWUMg4yf+2bGCB3Vy54e t3KaGawBUEsDBBQAAAAIAKt2sSbbSuH/2QIAAJ0DAAAbAAAAZm1pZG9jL2ltYWdlcy9ibHVl LWJhbGwuZ2lm1ZPvS1NxFMYf07mrWX4zJV8kuV5E035wJSopKC9adg1Ss0nLSrs21EWmixDv C8sWUY6KEpO62FgMobSUsRbBZTLj9oPtWza7goEv+sGgYEIYEQbfrn9G59U5HHie53zg7D+w b2dpUxay8BtgjAGIRCKhUMjhcCwsLAyGfeFwOJFItLe3G3ufz1eQd0xRFJvNBoztyZabm5uP AV3nnPbD9R09b4ssfbeH4vy6QPsFLxgeXD0F1mPo1tR4qmrGH47PBCemQsFgx97t0/Hp49kr h+4rZRusJalZNy5fHDmxI6LJUSpQSt5QSdexyJCYx6QGTZNfUUOGDAwMsL/8IuO+JHBXAfvu YV/9vxlRVdzsRzQa7XWTRWZ5p3NuN+rtmJ3D0yBOSHipqSEVw6OwH4UowuUiHS5UiFAUkVLJ 6cQ1D6/JHio4KRGpJOsQGGoZxHkIGko1uZ+ilBE742WDEeNqExAUWJkRyqMz/zwjtSpK+2F1 E55ZJJ0T3LDakT8HIQheQr6mGhl5A/AoeDssIvJdhHctNUQRJSq5nbB4ePGSrO0X6GpCz0g0 DTpnWGHeDE2WtWQyufXQodbW1omJiXg8zmbway4w6y9kfoNxNatArKHBX1Tkrd38Ku/K+8Jb o2zr4OLgvVhJSaxkd6BjS1/9+r71aMbJIa8j9qhyKvriSHz2nJ7sTV4v/mwGbXg+mEsYv40R zjBNcFA8w8yvsXzCUqGa0J9L3BstLJvTl8GdBrsZc+kImiGFVS0DarpxWYodEAlxpcGVDrFK VE5LFP9PWf6kpQBmbFoaln4CXDeyubWVY42FZZfTMzLrblVbTJlZBUJ5S6OvLtM23vYtc+px +TKTtcVUPDAiBFo+5ux8XV1se3Y21n2leE3V8Sfn/bMVd+7GUnsdsZ+/+jaMNUa6CyrLf+Rt bpjsWrXcdsY73jTZafrU1uT9UB7p/JZjO2gtC5ijRSs47PoHUEsDBBQAAAAIAKt2sSbdWEdE vQQAANkFAAAdAAAAZm1pZG9jL2ltYWdlcy9jbGFzcy1pbmRleC5naWadU/tT1FUUv86kIbpM YgyIZPGQFmJIui0C+ULxsZC1MYkOKoqOV3koYSlqsxI+EAFfkeIdZXUXBF1kEY2WJE2Sio2r og2sIuRAgKRMuuJjlszTPV//g85vZ+45n/N5fL9ztXMiIpNvkonkGXlZlBBGCCdEEALYUUoo I5QTKggFfGSUMEYYJ0wQBjjLKeGMcE64IBxwVVAiGBGcCEEESCQClAAjwAkIAoDACC2x5LKc BryDxSjllApKAc/KjjHKOGWCMkAW8pEzyjnlgnJAUnJWMCo4FYIKkBzlKgVGgVMQFAApowZJ WrKUtAAV4CUJLbHkMqAg7GVxxgRjgPrkWdlxzrhgHFCuZCEfBWdCMAFSvSQlZxlwBoIBoBlo ltyWdkj9gN6gJAkmSUuWgFbhYYktoSUWoHP4LHtZgnNAI6U+eVZ2QnAB0lcpV7KQjxwEB0Cb MRWGs2g0oOvoHcNVtAMwBFTIEAlJA2aCPBgCIzRgRDjN8A4WyMSkkVKfPCs7AQDK1wBK3qAk CkpmoKQCiu+gOAuKd6C4A4p+UBSCogEUlqDwAOUSKFigTMPLnhCVSuXt7a1WqzUazaxZs3Q6 3dKlS9PS0jZt2pSbm1tUVGQ0Gi0WS319vc1ma2tr6+npcTgcL3f/d/k6XxlGiJ2EYIP/xKsu QKpH+sw7c7xhu8vY4CVNZ4yXd6mGka/m1Zh+LvCiqR1NYcbG3W94xo7Wni2zFflO3xaj1U1s Lh4996ejWof//JzQjHvi1LmKaMOkBZbHTnf1J8fCM7ynDun6W7bF8RuF93Qsxhw/Jqp2iKYk 7Fvkmzu8eUGZvdrValj7fXhvmzn1WGL6qNbK3w9nXS+/Um8dSDqX+e9ZVUhdUGfDW85UfWVx epdl3/0JgYUHHyXXrZ/4dsJoPnipJjo6++KUxP7o5iLnlnAfp3fvxUJNs0dH4ukUwVeNX3Ch XaOr9bqcmORjHcxor63+etqFjsfrDDV7/JPcxn+W2afNKE54PiUy6cHt5Ax4esQ6+arVVFVg HffE8nRwwG7QZ4dNXfHrn7bSa1VHtuq7eq8P78x02FZ7briT4zlh43vOu4uX182evPBza23W 8Rx/Z3GRvrU14PhGt3Wr//jCfYvXuIyhzNl3NpRMbXoz7d0o1hHra8pdluPi7umft9aRl7z7 4c41Kwobw9SLsz3ZiK5tHsMilu3Tmpv8trt22wO++fiF1+34HWGRPQXTDs1efijy/Lyy9c/1 h7t1HU/aOlfONK8+URw94tSqkqiElLLCZymBfnzz1XJ3zY74sQuNeYFmtXFmUHtFyf5bI9Ul 2Yagir1bmTnBdFL1W0oFCQk2mwz57WuMJ/yG2ks/nR/pe3R6/Heb+ysO7Wn/62SdaWiM5ZQh 9v7p8+ymOS4zMC+l3H4y9papti92kaVlTuVAdfmIOcFnGgL5gZ1N2saBhNy/U9PMLSaPKmur KS609CN9XOiupvzDIVU/5Fe+o9nT++Vr31584W3+cW83K79g7zM/aAhw/FNd3x+a2ll+wz3/ YWNM0IeTTK41ozIvvZ7VUfOLamV6S+2JLemPbAfcAs7OuGEaeNTsdnDUoHjfPyD1SsTdJTpj 1LOBwWsz7D468sF/UEsDBBQAAAAIAKt2sSbH1S5UoAUAAK8GAAAjAAAAZm1pZG9jL2ltYWdl cy9jb25zdHJ1Y3Rvci1pbmRleC5naWadU/tXzVkUP02DWPIYlsxVjFdPjzjq9sDMKlEkhpkp msLQKS4lUyPpsRi30kMz3VFOUW5SKm6ekUeMR3VzPGqpm8gSqWSFhFY32nPO138w+7e9zt6f /Xl8v4s8Fjo4rhtlYI560JfCCBGEKEIMIRAdxggThCnCDGEQjwQjQhChiDBEQMxSjChBlCLK EAWxyjBiBDGKGEMMOBICjIAgoAgYAhDAAppj8WU+DeKOKIIxxZhhDOIs7wjBhGLCMAHBgj9S ginFlGEKghSfZQQzihnDDDhHvoqBYKAYGAYQlIUGTpqz5LRAKBCXODTH4ssgBImeFyWEEQJC Hz/LO0oJZYSCkMtZ8EdGCWOEAVfPSfFZApQAIwDCDGEW3+Z2cP0gvBGSOBgnzVmCsEoc5tgc mmOBcE48854XoxSEkVwfP8s7xigD7iuXy1nwRwqMAgibRSpEzAqjQbguvCNiVdgBIgShkAgk QRpEJoIHEcACGkREYpqIO6KAJ8aN5Pr4Wd4xAJC+BpDyBilRkDIDKRWQfAfJWZC8A8kdkPSD pBAkDSCxBIkHSJdAwgJpGr70CBkbG8tkMktLSzs7O1dX1+XLl/v5+SkUioiIiLi4OJVKpVar NRrNpUuXtFptfX19S0tLV1fXl93/XRN7vzZASIemiUb8E4OMAJUMNnU/efjfP41GWf9adVJ9 I37EZH9D9EPuraSxeNPjqlPrXRLHu07KM1qRp1VZriza4376WXXKdL/yKJ9t+XeyrH2f3Tz2 u4VrmlWA3rRjpOWP6fNSZZ7WaYW1BXNSXPzaw4p+Uv1csMkvMtxGlzZLqWfVs/J1ZwaoMpWy 0umrsvxcy92WPDjRlOOmGHL21cdin6JYg1dbHCJaHh9U1LzrdD4ZqKtUjrPIrvJe53tjxvC+ FNmO7obbS52XaC762LZfWaHNzp/bO/n52RKzioDp+HjAI+c3q5MPnpuz+c6x9bjzilPE2qsX Bv/SatoQqnhxa03FkxNyx84X2bELvceZ9LU/v7/SFWK8Dpyvaa/IMUx1j8l7+r7ORJUdPWJe z2untqT+8f6WkZ2tO3Q54bq6etMIw9kma91KdtmbbTl9rs00xs7xKFV6j6wy3TZcP3CCj9Ks fnz86DE+zVs/dsqfhvYYp49V/rZX7r/bZKTF0/CX0efHxFlrLA5HDSxt+Da1qmzgs4jYsgPr kq72nvePtm2Nn+I+08thdWy+m4NaRdRB46Idr+vy/h66Wj4p3dzf0TzWruP8i0wHmUXbwRFp ZROzNpa2TvxnkZd9blRQZr7voeYBi5t2R19/GZirzPYISt/1LL8xZUKRvHD7/GGJFodSF7e0 7u1SDtkUmkks2jO10frv9sgqtAX51/bDRStFXKST7/7F+2QbEgPnF+bso9MfZsSd7khYVfjY rC/9ZC55X6h2eW2tPrK2PLEoozZfNvVAbIzj8+MDyk3MC0oXbpxWkFY6pO2U+9Qd489ci6s6 kpBRuiDlyoOWrJkePRV5b67299u8Lf+k97XZ2Rp88WhyodXORzF3csuOaaq/+m6rj+Ga5JAK g54lNdeidl0OrdA3Hu2qnFIw9G1Ve+SqlxeKdsszkhNjsvy1waFzZ2vNcpJPV45JSj519/ti z633Fg3aElbtbHhizn3ITHilyapTBG7Vtern/PXpm89na9z6Lm9rWB8b/a52ftZS20Zr+bzZ te+aPV/cfDMjaIPaVtUU/tC8O9mz9JrRIOXaBZtNhn3UBcdP2+4btfJzXZ3PGnlpg8twY4ew sICYmmb/24f+aEi3+HzhnM3rlO6HVgqN0+q4mcH2tV5txUSVgO/aVxe3aXa6KuHQ2VsRjf07 Ki6ZhHzyCFrWtL2jUv7m0/KrlYrIB/e8Qz4HGpYF6z883r5sdPeT+Jn9S/X7Qvp72wtKYj4Y 7ELO/wFQSwMEFAAAAAgAq3axJqM7gi0KBQAAHQYAAB4AAABmbWlkb2MvaW1hZ2VzL2NvbnN0 cnVjdG9ycy5naWadU/tT1FUUv07GSxhHnII2xfAFLqHSFZ9hbYiBloOvQcFFBb3A8lBRE0hR U0AT8kGwXBQSVJT3W7TNsUSR1atOGIss6xiIK22yuIKAWnO69+t/0PntzD3ncz6P7/fLgMVz 5218iqaiIfS2MEIEIYoQQwhEhzHCBGGKMEMYxCPBiBBEKCIMERCzFCNKEKWIMkRBrDKMGEGM IsYQA46EACMgCCgChgAEsIDmWHyZT4O4I4pgTDFmGIM4yztCMKGYMExAsOCPlGBKMWWYgiDF ZxnBjGLGMAPOka9iIBgoBoYBBGWhgZPmLDktEArEJQ7NsfgyCEGi50UJYYSA0MfP8o5SQhmh IORyFvyRUcIYYcDVc1J8lgAlwAiAMEOYxbe5HVw/CG+EJA7GSXOWIKwShzk2h+ZYIJwTz7zn xSgFYSTXx8/yjjHKgPvK5XIW/JECowDCZpEKEbPCaBCuC++IWBV2gAhBKCQCSZAGkYngQQSw gAYRkZgm4o4o4IlxI7k+fpZ3DACkrwGkvEFKFKTMQEoFJN9BchYk70ByByT9ICkESQNILEHi AdIlkLBAmoa3PUIODg4ymczd3d3b29vX1zcwMFCpVEZHRyckJKSmpmZmZhYUFFRUVGg0Gq1W q9Ppuru7LRbL293/Xa6vRo5AqA19LBrxT1jbAKq0Hedfdfq3AzZj5euaqzYqEBq35JR/deGN I85YZUgYE7boh488v7APqDmrzbT1SV6kjT99K3tm+K1Vw/HGgCyPWJMhwNHtXr7LXrvxei/3 5RkLj8qVqipj/dmvnOZnLMlyu1saeM5YpxpQtRaurw8br3n5l39txITsr48cf/agyqH62qzc FWZFw+Y+qzvyrXLlwd1vuhzs1ac7MmLeBe90p6qgtIzWojz9gFd4zTbb+zGJzl4h1w6MXXo3 pM0cqf5xhvXsvPb4J0eLFTVF017PCr6V5v0of1hnWfsw56Vuem9SXVxr4/F+q4U1c15k1a6e cTQjKT6m8o/iUVuSfS719bRMSQK1eZ3laVen65nRle0Rb25cL2i8Wu4zK6q3o210snL7npfl /b+7Ze/un5el3NdxbujR/sEnbqF7n6c0rU+87jk8Pgqo2ml9o5+7LM5FF+GUMGzZ1X7QsHNu aIoseU3X0skTIwtT7OSzO/eN/G7q4/TWD9Rh8QOvIkKPDV1xXpsR5pnjfGi6uUedvECf3bnd 19hQuEnRlBa+42beYtfgT5cFhKyw+Wba5ixP1aGJyr795w05Qdk9kw6g1EOd25nf4e7MGH2U LG/x/MTJJyydN7uDkxyLJx2zldu8l7bOavhJxlbz6j8T5XZRj/MPX7wU6eeanxh5XL483FRc FJpiKrnQdd1Usmb565WmUhet6acinxb/nlOfTZ92pvJX40DPhYIruaaypovfl1WX6O3G1G5S aMqqT9olGFMarW5PzDW8sEXlg3tOPqvTuniYLzWf+FZV17IwpKImPO/n0hPtusTzDWG51pWa svffxJX3WdmbLxsd7aNre154zNRYe8SUbHsQ9iovtmL2L1suW2pMHy5J/2RyVe2Evr+f33B+ J/38lbNF9XFN/zqWmRv9H67NCXas/6ev2TPNw7DrnatlXio8Y0q/n+uc9LZ7G2TqqmuKOXdW aWwm9W65vCwotv5z1/shXi1Bu3p3Kkrznz1v26QuP9Z8Ibd8UL9h5e3Bjq23Rw0ZdrTFNuhH dKIF/wFQSwMEFAAAAAgAq3axJkc2I53CAAAA/wAAACEAAABmbWlkb2MvaW1hZ2VzL2N5YW4t YmFsbC1zbWFsbC5naWZz93SzsExkY2Bj+MrAwODhxeDlzhAVyhAVxZCYzFBcwNBSxzBlIsPs OQxrVzJs3ciwdyvD3r0Mxw4xnDvJcO0sw72bDM8fMbx/xfD/C8P/n5zXLnH//yVYnC947phQ S5PQ/69C/38pdXQr/f9t/P+Ly7nT3nNmpJ4+l3ruZO2RY7XHDvf+f7Ds/+39QHD0GAM1gOJP FkYGBnUGHRAH5CcGNpkDkwsquHw1eB8u2RKZ4ifQccjAQpaPlcGjQjKnmcEaAFBLAwQUAAAA CACrdrEm3C5N95oCAACdAwAAGwAAAGZtaWRvYy9pbWFnZXMvY3lhbi1iYWxsLmdpZs2Ty09T QRjFf7eAXAShvRCtD+TKBh8NxUci6oIKPhAQq4kxQRZFW7SWUKREsFGpSkwTY4LgwkSiuJDo wlAVEzYoL0t8deGCYDQG3RTdXBeogCbjQOL/4FnMzJmcOd/5MjN79+0p3FaTRhq/gNIyyvZS dZiqKmqOE2ig7QI3rnOri4cPePqIgacMDBAdJvaS8bd8ek/8C8Y3xDRiNmX8XaqYswT8llhU a7ukiR+amMsNX80VvzeL6V2x1+VdNz2vY57Yy/Oj0fPRkWtiskd8eC7xIkpWLRORBNZiGIha 4hEcNgzwwUoS3CzABWqqnAXE6ZTsLHRROU0G6HJfZVSOCRRD64JqjKPS4+v8WSMTMcbUKVmC jzAIjw0cQXBCNWTBMyiUOgyB+Eq8gv4s6ibYAFZwwCeFtu+EFPhulrISoRRMFdPtQgmy3clM NSKrY+rZYL50mYIwCiUKO9MZ6dBl0K0wCSHSZRilQC7EEYQfMYTYpBhF7cKmxC/fH6zkXjDb rS0mpU8KbxOF/jsy8gyDp3jMv1bLFprvrmTcOl/qQC5nwAON0KzQqOK10nQRTx+6QstqvJ0E EmlagRdHPZwYQ6+mFkmpy0n37O7Q89W8Euwa9X3Sx2gyBe0/nc1/qpvzadbw9NICDSb8n2k0 K035qr0Et4a3l0ArPsFGgX2M/eCHgE4NFi/GcXx2lrvNPj++AFY7dSf4j7FmNlFebh62eTL/ J1CDHFBXlUYmwxnWgnPWgwWjl1My15aeNB/TU5Izt7REzHcVk2Yqqog4h0evJKemu83hi5eS MrLX2xwuF+WqRS90OjpDbzNzltpc7aGeEMucXfpphpSEJFW+12PdYVPSosSJVzXdh7Rli9L6 KV6380nem/VLVHb8BVBLAwQUAAAACACrdrEmarnDL7AMAACaDQAAGgAAAGZtaWRvYy9pbWFn ZXMvZG9jdW1lbnQuZ2lmnVP5P1T7/5+6FSr3cstauoVKZMlgJt0UwtjLbnDthzsje6gky8cI UbIewrUzZMtYw3Ul20GyZ2dQprFVyE3v7znuf/B9/XAej9fj9Xo9X8/n83XeWtqaRIL9+33B uC3cf4HH4SAcDsbhEBwOYBkej8NDODyMwyM4PMCKEB4HQTgIxkEIDgJYL4zHwRAOhnEwgoMB NorgcQiEQ2AcguAQgCLhAB4HIByAcQDBAYABY9AoFjqMdgNsDxYQHg/j8QgeD7C1aAZBeAjG QwgeAhgLtAhDeBjGwwgeBhgptBeB8AiMRxA8AlCO6CgeQHgA4wGCBwCjjGlASaMsUVoAU4Bt QqFRLHQYYIKwHA0YghAIApg+dC2awTAEIxAMMLkoC7SIwBCCQAhA1aOk0F4IwBBAIAAwMzCz 0GnUDlQ/wLzBJKFgKGmUJcCswhaj2Cg0igUw57AymqOBwDDAjET1oWvRDEFgBKC+onJRFmgR BggMAGYzdhUI68WMBpjrmHcQNorZAbAjYAohDAkjDbCbYDwgDBiDBtiJsG4I24MFQC+GGonq Q9eiGQIA2PsbwN69wd5Fwd7NwN5VwJ7vYM9ZsOcd2HMH7OkHewrBngawxxLs8QB7m8AeFtjr Bv/lOBw3N7ewsLCEhISioqK6uvrNmzdtbGyoVGpAQACNRouPj8/KyiotLW1oaOjs7BweHmYy mevr6//N/r/jIvbB3gIHJ8CVcZ0klf/VEsZ5TNK6ozyrNYJHTOc5qSK7LVoQT5noqMjRCsep RR0X3MjpjJcwplt3VuZ1J8sR9gsS453aUy/5LE90vizoy7yueup1UZzr2xxSvKRNl6+Lfrah V1KgjvEyUmiuqzTZxSgeKbPSUGvUTHQyq3QdeNZIeXlxtMZzviv9fr80uULh9pRIUTLdrD74 pyOnjg67WoU5j/HrWSeWz7c/FWs4ed/cg9ycJDSeiNS9XOo73eZgY/X1k0lx/nroHz31DNZI uWN3hn5DNXu81oc11fPvh+mpv8OOREQOra3Nd8QbK3SZ131eeitwwlyksf7OTDOvS19QY4N/ X5R6NFD60fjq29+zQlEqfU27Ez1c7Q2ZhplSgeyNkdjpvuZm1OhDEjPBB/hdZ0I4zxXOhB5V +jATxqPSEWdy5Fay+QNBSIL3Lt8tIm+EME0im6zBe8AhUOyNy2wEyyvZgXo+yHWOJivKzFKn 3RUUi1F8v8TrzMMunI3hXng0+uTqziNm+C9c8Jm4nvUha1s9imKic24cjfxM/7IQ08HW21L4 6WyLlogjWTLJAnaQX3BO0aOPnEs0pTujj9dSu4Di3njZRTdoLsAp7uF7jazYsF7SnLPvHyRm 5vWDkhIp3r1357Niz5GFcp5nj53PJWnrFOYrkXeSEwvyKQ7RhYtwvmVpNThMyZXQN58tqVU5 v1yYa5TKKmtldxWXIfuobp7/mCmbJHmdYn263SkpNZnQr0J9YTc0pSzN4PVjsWtmaFIrtcwk 6kodq+i1ddBItZRDrFS91GogczZ8pXF78cRMys4ma7V5dz99LvszRdezxcZZ6nYR52e6Z+vF q25yrTxHwp3b7PWVvAyNT1lT6qXqtgdyNr2WnMxERiS97GVDais1GoKE143EpcbOZOC7N+Wj fh+ot06/fN7U5w1h/uSG25WtCSlD1j8UvrfJGowReoE+WwHKUkor9yOXHmYM2gBI1oq+DPsm v77Ajo8z8FdezKAo3ctLD8ojVA4rSSkTUnfuAmFCgc3AZNJkcbp+RZn/sSu1GQKsZKH55IzU kQd68/oKeVR5g/KEvM7JyxkJYQLGjU1noxQH73fpw0iacG9Ou+svXTX9RT/JVC9FtEzBrG4Z 9uDjRxlNhQUxkGQAy5KQQVh/vN/y308Ml2811FeyNg/iSrK/S06y8rvP09u4V2o7lLnZ+Vsl /3TtVhtVXyNPzz25oeIyIlYa6LZoJVP1Y3T94bUfC59F3Da2mpkfSwkHfj/9NahAwrWnz1Sp eNrOiPbbBIknIKs1JV2hlrHhFRvWaHya83JS/HdrzxtcJ+0XdLavBtethU/NCRx1IbUK83dc OG6S9XNcfKvIDo+lxMssHgCGCPt2nxIJ4rNcxerErKPrkTdvJINP7depirwauq8tDhdIlFNt 1h219Dr5j4zK8wfdOvcxLDp1TrEuy6k/2vrxfP+7/CZ7y5IQoYT+fpmDb3hovJCpQ87KKeIZ 5a5bJDeqXeTmI+UZEffbPsZjowE/nX2VXnjv1xLzMYevVVtCqwH3P5Nn7lmQVGKNzObE7EUE 8qu5o+REPazFIyl+nTIUxDemhXz+hP+52phK2YFjYlsmfEpvX8REvQhkLibdSNbZ7yNrmPZ8 Vl5I7YbqQKrbQD8Hvv33O3VGd2aoyKXbg/StxGOhmY8hiTYhbSM/zySXd48vNVpqas3pBfQP vIfqYdBHfjDg55S2dNjrW7fBQbzh6KvU8f1bXaM9Rh4ePhOSu3+ShjsHEv0p82owR2N8+YX0 fW2FklzLhI71jYzxA/nandsGSxW9ftwHmR9eJ9EH5XSDXOKybrRaKn40jXkqZOl45Bqh+vkh nuy+Bgutl5F1cy3H5XpfT37ZEBz0S18nVfgKviW+74vwvpNTFa/LpEne2p41SfH88s5UjCGS sTj4ApHM3gykGNfvHKd7P75AfiE4wPhR4frAmXqcg1r6N8Rn7zjIsk37ktfKf7EiNj05hKkz 2kXmV5U/8NaB8G+x+K+kC1Wc0jZqukd6yUtOFx88573A+m14Y0k4V+QpdSt+7Ac47WtbMTni NvZ1bDLbsDq/kcvRtk6DTjExyOU7Ljrvap0mXmPUrS89UxVOnn7jw2jcLXEJiS6Z1UyufTd5 3FtDt3BVU4DRl7eS+3f31HKqv9GSaan/SJoN23vEJeeeR5BQw8SXKZ3mt84lD+8esvu6ItD0 RWMtZIE5dUA2pYXVJjuU2tWk+WtKc6iwHL/IyZmj8SPO76TdHxm+b9hvdbFpIb3sN3DX4ei4 1/OAG3Ri3lLhpwdPbe2WZe8WkYhP/zROMDCd0nt2pXlRjctp4PYl+Xg95T82Kxs3HogvRCnO 0YJMYDHG86zUTUdfykq+Y7/Xn/edCQ6a7tlbFXxx9jWG9/tjQYp209r/nGuYWnBdV/cCT6L+ xaXj7YlU1ZARv3InKapSsaF+92RA++9L8z/q0k0cl6tiqxbnkNJ8D7OJxYznpMOsLz3WfhsZ Oa/mLlB3qotb+9hrXMzXt3XfcPwrkThgqb9GCYB7r8iYGCUtOucJuHWof9Z1mYT8nqX0Xh+2 LvXhkHVSt+jaEvOr5LbR6kyXifl1OCErSYM9W5RUUz1VIcrHJvM3FI4k8PvfyyfPEbVPZLyL 9Il3ibli0r06Wu6X/V5PZF794Ub/P9GLmfaP66XbG8ZzzuXY9ntTxaQbZl7xV5YNsYfKh8fH 26QUmj6nr0XIlwyAYULV1U33vB/jZnOgJ8ApXc81j45nihDgnPB3niXV5H77JAX5CL11Oavv Ref6iDWEn30MV98vdvtuPtwSYRs7BtZX618eND/FUvVQ1tGI+MqGYrNTl4hTVWaZQ7d+zt9X tuNl92T5anuP2zR74IPdqmTXZ8T32ZXdyV8y61p0C+ycG9PnpcjEPqGezVPXLlNeaPqPtYlQ 1Tw+ZgY0f1MeWxlMKa4fj+Eoy7h6+xoj2vcN7L3dw+Vu17y1Oxb9R8XgfdpcemvS+m/mim91 bdl5b4ImZE3lF2IQqHd+ajLVoOq6tBB9p5S/K7cpcYT95w7HMTCs0+zI8jrJVOGwea29tUvT mmKGcoANZNoCzAbfb+1b9LNtZKzEGE/PLtZkbq657H680/zlW/f0N1EX56i/VlXHMsC1Tf47 yxwq9QybdpfAgxJWhwqtOpZcdgii380DD/GGtCgGJ+VbfhQQpauGhCidPWkREv2VTPmf02mL 0GStO3Nngs9ZiImP6hxulQnVsIqjhUptC0fmWRDVpOOzQvn2myhui98ND7tqYaaWEirHaZGw FSaYrW+lFvxkxEBUzHO2TUWaqKWzLUPYJHB91ePg8/0+QA7xC5fIJVrV3HT3cb+0P3hVnHZ9 1EOB/2qgMW3fh7Cf22i6JJoRw6+FRun8wq/gGvGk/9LKT6rNbdorgxEX5zWDOywfcp73fsh4 lDb0aOpsmKpY5FMJoaQPcpQPt1QDIqfiIo8ck009a9zeenN6PrI+9BTfAd00ksEzMSKZGF5o ceIviaDKR1dDlgJacowYxOhcAYGIs/hR37AmUjSd09GL5O5l9jjB11K0Q8E5zLjoTDhiEZPo GjNqcWusNiZYLWJiO2bGlWefKu7K/wFQSwMEFAAAAAgAq3axJkMd9aDEEQAAoBIAABgAAABm bWlkb2MvaW1hZ2VzL2VkaXRvci5naWadU/k/VP/3n+odg1RosRSFLCPrtQ29VWSvKPvawkzD oCJC2YYZY9+Xa9+NsW9jjXdJthuRPaRFtjYpa9zvXJ//4Hse95fzOOc8z/P5PK+rq6+jqnLH cD9+3zrqfwGgUHgUCkShIBQKRjIAQAF4FACiAAgFwEgRD6DweBQeROEhFB5GekEABeJRIIgC IRQII6MQgILwKAhEQRAKgplIKBhAwXgUDKJgCAXDCDACzcRiDjO7YWQPEngAAAEAAgAYWcvM 8HgADwJ4CMDDCAtmEcQDIAiAEADCCClmL4QHIBCAIACCmRyZowCMB2AQgCEAhhHKiAYmaSZL Ji0YUYBsYkIzsZjDMCIIyZkB4vEQHg8j+phrmRkI4kEID8KIXCYLZhEC8RCEh2CmeiYpZi8e BvEwhIdhxAzELOY00w6mfhjxBpHEBGOSZrKEEauQxUxsJjQTC0acQ8rMnBkQCMKIkUx9zLXM DIJACGb6ypTLZMEsgjAEwjBiM3IVPNKLGA0jriPe4ZFRxA4YOQKiEI8gIaRh5CYIDzwCjEDD yImQbjyyBwmYeTGmkUx9zLXMDIJheO81wHv3hvcuCu/dDN67CrznO7znLLznHbznDrynH95T CO9pgPdYwns84L1N8B4WvNcN/y9HoTg5Ofn5+SUkJJSUlLS0tIyNje3s7IhEore3N4VCSUhI yM3NraioaGlp6enpGR0d/fz588rKCnPWwsji//2dR94/8i+womFUJdspvaqc58FoHoxtd1Vu R+hREYMMveq8zghewHmquzq/K+aMVvgp/ZqCngQJE7ptT01hX4qcQ2+Gfm3R63RVj6Wpntri gexLweynDepog/l6CRi73rqSt8XG+QaZBvX0kVLLWufp3vrSscrbHeGnDRllE7WEt3S7Pkb5 u4YHn3ozDRsqplu8V5em+xoq37cHHuAQvNpY9aGDyiNlDzVWf+qKFTHMutpUM9eXChBnoKba +YEcrQjBa82Kd1FFJqUV4b71y2OGNh1Z19RpXycbPd63lrc0/EhuDeAQ0lBQXfnUDfAb9bdu dsz15xvKov2aPedHKokXs/xaHr9/jmuAv3DmPF0nz7eXwgL+s95L/G8rsysbnv23/s9cPlNV 1Q5xo4n1vd93MDQtaD1p7D3pK5+oYNAmyHo2kEdM4gz5EL5xLGCfuV6ON4sBOtefP5l6IlSQ sT4byBtP+BjhkEv9QN58TbtNkmpGf4q25xP/GKk6cyIrBPuLIGqjJlhSEGCojLWKv67TlJto KsOCS7Ls9cQl2zLqcSm3s3i/pD4m8wqDhGbclxgnBjiadqtZUtyU2O9UbIWfKSm2vPd9cd4C vSu5YO5y2JlmpiJIl7j5QHmJkB2h03wzU+gGZrHw1MHe+KJUry3bXPmHS4s0rq1EiZKsLGfn DJoXG8a4kgH7FtN1BzGl6QyxpOXy29/5BUpb8WgXow4Ww1vVL4WJX71605rPJ73ePfatFrom JV1vaL18nqFjL+msNX+aEM+YWvoiTT1K5nB8ONeGLqw8VGJYEfptbb5c8yKLgSj1SHTxj6rV 0e3JZ6zn6LfjpPOkZHyWMeI/STYyXXIvTt9sfJDVzfK3yolrGy3UKWk99bP7pFPrQ6fTJ9E5 L05+vPfzhVRpxkrPXz9J5zZpCyrQftzBUeF1wU5rwptLoTqPruue3LEaUpIvf3Tv+sCxP2/T r+3Ej5gSKX8KrLRa/zTeelyuZP4zfMduIpLiyrhHODKyNrn6LPP81ESvq0nZo1XCY5rzPPzv g/fD+76ruLx/Yboxayt4eG2GqkVc1HmgLLOYz4bFbNawqvj6tOQotPkU//Td3cz1VpfYWFgT k1Rbcq9wO7sclywTt1xlof9krv6HZIt78bxbRU/Ddv/s97MHZf/6vkqZ2l7J4cq++QuykFL9 Pcih67fSZwGnrl67V+l3d+wivOM5tXpsd8PlSXbF1kKKT9vw1wahse2WTtl26zkVbXiH/K0d A+f1CBG/sv5XtjEg8c/5M2aHU5UPaoQFtbvnOJzZOMZucNT4RKQBS6dYgTeWcPE8/fZ+z5rL dM9YR06ek5q7Hcf2OTuFpKqdbdNQ7Ti6xE5hPP+Qd32hMrgcus22+qEd9vXO59Z5Sf4zmJv7 y1E+OYoWptA30WyqqnhulD3wQkZelEaBdvidstsx7CLcByz4RNwrdIRLb8TRujvFoyHdMmr+ qXm9LrWhG5HekMjcxl0dF9wKzuYY35lnvNpL4yux1KufT/U56OpgfsVFDefP3XXsxOK4dJ1c 4qJCJDrPDHKgYg/hhDlFdVepvxJppo7nfBpemE32uDUcwGFM+ITNVd2TG18WSn3SytLg57i5 cGOKy8PHAOBaTa37nCCa0gPh1VYfZmiJiY6JvnamK+hQu4uUG9iKXX1X02cPFat26fe7x3uk QDViwgKF/V4y8UZfFBOw8/oDvpM8GQvD0eenm2A9r18mmZsX7mJbfAZI/L9zJg7RPNc2B+R2 36axe6JRH/WNQ6x/532IpD3i3XoT7WuaP8tPyJczGIoX3l8wvkIzuMhuFBc9XCAnLWHgsQpZ V/wu6nyJv/q41+Tp4dA7J55PXHF51ZfGP1KEyXSa8+01LcLU6wrlo1/gLRQ4BZPyPZ86P0/r HUl1rpeSoqM53pqNCI3WkfKHnSzGfAcr3TsjrqQZy3I8OoP5dSgBK5x36NMjs3z8UtodUMKG 9YlxQ5vSPQ+z3BhbsbExJ4XY4+lLnAO+2XWMxKrU8uI7R97JRj53IbsluojbvXpZxcKoST4Y cRvomwwyJ0ZEvHZ0EJKyVoGHTtRJiCY+uVk6ObrllZE+VxbIsih7Qv5EzRmOch6P0qQVFKOh PapCJZjDXDGtPoErjV7EMBw7vBhhb6z97WwJx5nFNJXcVoVy5051+/qs9YbP2nzepX0zOvZr z9qiKp++3oLGh1VKv3fKEv+op939k1RzeVr6/hPR2xsshpX9IUKk6q3+bWXe2n8On4ki+c6i tb43BTY/IAIu5/QPd4frmj14In3+LiPeUl50SU4x7OPUweD6TtBvpb6XjqnT/uqmwCKm99Lh k8Cn1Cq1GXOSgznmiAzoIqbxMeAH76yIh1LPy1+gAy5CvVpYv9dEBpdU/nV45OgN3OUhotgE r4UQizzsYefPY7bvn49HE3zKLfsVXN6pTomv9VW7mZ/nexhxUe5tU3eIJoE60zL55IvVQMB3 N9v7T6Y0ALDd27AmgtraL/+2djCKcv+OHU+OYJt8XYOZg0BX8x35OCXo7uEyba3TS1wCvwdj GY/sOHosh35taf0V/h2WWnJ3u2Y8q260yKex6WsIZrugz41Rw5n4Bm9oU3sg/VtJ2/b4JXuL N1d1YrtGf1WoAZaDb8lrdSPfviZoPKn/6a+sMj393XqAZ4w0qFh9x9XQiKg83cN43Ppp2o0g mza75vWtrPvfe+wWlqO1LI0v27EPfN5uf1ytaLh60p6gpX4h+vM11RrAPsszxfZTPyy20WlU 4Raibm2U5tbYFHWB6AwIDI8VMXqdXF19ImwzNg4fSnawdUEJbfercza/4Tv8JSZ46r2eSsmQ wMvf88emF95E+YJYmXsxImqD1u+bz8UVCF1/vLVwR3xxeIfb4+qPlTdSFO/JhMOr9AG3L/cz 1Y3GZk76ya4uHRlonim0X094QzrBiNscS1f3rnSfFg3VbY0v7/8RqyJY9PSHuoGwrM+/7ZOL Ozrbcyq7vYlX/LuudamP68YMQkY9P4uzW69snvZtqvUfjZHV/4o7IjNps7vm+npHTzVma0B2 JvC/I/6LZkObjknT68ZG/hfIt05khVX/GkzZxXLARX7Bu+uT4jvwuHrWVu07jb/jRRpWypx+ g/lCC0sxg8bV2TvrJhd2n95s3Tx4H16UbV+eawjwGg1YpwQUz3nu93b7ZyHgk65/4rrfn4ZA cVzgMjnwYVcAd+qNYzTrmWLfyJSgd+sBFWJmXaFPz6BtBcZFTluSLFY8xBd8/9x5fKSbJO1N kpAg3Xvu/+Bx0AFC8HposLdCgDrBZX4+uDeZ9Bv3VLGbX2fB5JKlnyY66DsrSS71pOu6s5Ye ziQ1yIx60oIWwqbncc07xKjR3yOZHBQadG2cjNN0k+Kl+KhQpkJCQs+ZG3aHuOtROFVdAkIp TrRAn26K1woZQIc+TA3lsaT8Ohl4AU11hyWonpeptfOh1eYhRL3QZlzQFUKYDjrM42zYM90w owYy90JYtkRwrkT4yhzJlEbdLA6IkqDEUMNee1KuN4RTqfYVvJb3eSNCeal1hEuXPUM5EiPC GkOx4xHp48G1jRHHu6mWXWEahHBrUvh1MXJBY+Aw+mSOd5QggURtjJoiRJ0pCF++F/mRJvYZ rTlOo2TyRjo6RLxUPV7WTf5JlbixEeUYHL1OjUnjjfquGrXTeOn1sP+EXuDiwrlltMyqxPE/ luEH9GPVhqL3OQWzbeI+n8QdW6Q6eMfxsgU61sed8gnr5g2paowU2YzuR0tsEqilrJHn2Mji bLGSkrFSfPZvYBqZdSJAsCl2AE1p24h4Pa6mEWbT4B3dNJ5wmCp+UDK8XjzqhE+CvqopZ8nl vxJJ9ztVhSUjToclgQteCXVJstjEK02JCd6JXLQoe6sIoCTBeJNElIx0tQq7PxEu70RZFU92 mEh63ETFboYq8SU9nCDdnEtsTE0M1k/1wIarOwUYsMVftYr13Qz6kJq6q0eS0QflJlJaaZGK fKkPekDWjZTAniRRq+RYMJwwkfpjPRVrmXjQIdbTKSZvMyGBEO0UFsu1GCsNgrONYOP+9Fdm SY/ck+rDEi6Fxec4pdCwpMTQ9KaehGTJOD9VsF0/2XUoirswZpua2sVmfs0pcxq2At9jk8gL 6Z/YMiaaohl3vXqd0jATGZVWWPJNhwWr4BnJbLEJg6mDmbVscW/1Y1Y0JT+GZX8uyVYIk42r TSsL869oyl7lu8rbkclCz3LKy5wolH1plbHJlzEtDt4JSWRdyi3Rz54tEnuln8fFLhP35Zxk cxbeSsuGIirrmysfTpaZzFX2za9dJxwpSZ6zCiX32HLy+x9Jy5HypVzpjWSI2emGp5zbylg1 K7i5nn/tT4E/WEAZdRZOu2WO8cLQC1V7C02tb95iL5JbyjBw1oJ9cF/IhZUbuYJiLpl8jl6+ xS+GpS7wCyXuK04Oyn3Kz+e/VfyEHRv7wZCU57A6ZKgAnyjEH6BpJxXf7bxyfkI1HMO9E8Yd j9GMP1BSyyOlW1vSXINNSZNcclfap0RzMKEtixTKBNEEc+ml9KKb9SrmgM6lNWwWPcdIjKdq CZdvQq83EF9bLQ2vKS3Siq0SLjXmwYU3Fe131oDCBbydy3h6klXvlHWZlNss0QcMeCZ9L2vN lc3wx8xiKtitSihA+aljxekvKr6pFfGAF0e56Qv06OlenZLx8qQXSutpFYIqBW/4SwcOVBI9 yns7XaeT7e6vaR0SABQepZ87K5+nJpM1XApbR+/rw2V74lNY7j1dUygUCxHdd5rriZJJbyWf YfFZrC3HkutZERECo2pukqsN3m9KSahxaK45T/SU5tdeUpJXVNS+2FKrvOwokCyrOVLlMiSu VaAtonCl6FWtgoFZ5ohdoDLh6Ls84XeOB/oMt9PkLIk2IkSbswL12o5Ozu8AU/U6A18LDnXR IDHhSDqAflfJbVj7DWstvapNO2HmxmrlWlq9JVJVhKmUVHCbNrchpWsqlroGGzYa5CZV04Pd DzZEqNvcJZ7tKXQWIlqoCTQN427TV6UNWlwfGYpoxjezPalh72PE9RWN9xpUE8ndJRfq+3Q1 w1uaSn0wNzSf4qq+rQLHibRTYq3rWpcTw0vmh1qKueTiAMcf3BX9pWXnSrGhHGUj73gb1XIA 5ZLX5JZnJH7p9+oVrVJtR0wlNZXLTlHaUm+0ctMrl2vVppfbKPvab18pEnMGuont2zatqum5 G+8Mdg1Vx/uy55aBHZv/ivifAc61HNB/P6ybuJ+eL9N8dvwq7Vv68458tVMuwu7Bz4UgvhqR FyKtisUCz8UztDCtL2T+vpA71aFwoUPRtkP5Ave+XJT6/wFQSwMEFAAAAAgAq3axJhDs1fiB BAAAngUAAB0AAABmbWlkb2MvaW1hZ2VzL2Vycm9yLWluZGV4LmdpZp1T+1PUVRS/oqEgmPZQ W1SiJJYFzeHGCMtABiQwhCIjKAo4qHMBeahMwEbijJrLOJaKvLzIEAvotjStsBmj8VBMibUL Fq4QUEwqMAkyDKkxOA6ne77+B53fztxzPufz+H7DIjb6+SebyXtkmrwsSggjhBMiCAHsKCWU EcoJFYQCPjJKGCOMEyYIA5zllHBGOCdcEA64KigRjAhOhCACJBIBSoAR4AQEAUBghJZYcllO A97BYpRySgWlgGdlxxhlnDJBGSAL+cgZ5ZxyQTkgKTkrGBWcCkEFSI5ylQKjwCkICoCUUYMk LVlKWoAK8JKEllhyGVAQ9rI4Y4IxQH3yrOw4Z1wwDihXspCPgjMhmACpXpKSsww4A8EA0Aw0 S25LO6R+QG9QkgSTpCVLQKvwsMSW0BIL0Dl8lr0swTmgkVKfPCs7IbgA6auUK1nIRw6CA6DN mArDWTQa0HX0juEq2gEYAipkiISkATNBHgyBERowIpxmeAcLZGLSSKlPnpWdAADlawAlb1AS BSUzUFIBxXdQnAXFO1DcAUU/KApB0QAKS1B4gHIJFCxQpuFlT4izs7NKpVKr1b6+viEhIdHR 0YmJienp6TqdTq/XFxcXGwwGs9nc3NxstVp7e3uHh4enpqZe7v7vcpuZN4eQPuKNDf4TZMEs ueSwIryhuv3Ygtc1CZ0NhhCyeI5fpXrdO7dOLqf7/uhs3B16YuWrh50iLHXW4nlhNS3/Lhm+ Xb5uz9KMmbMXxXm/qIFQk4+xvWyD2/On/d8bbh0PKMx3qtia1lUTflqToS3xjL4Q8/GPQls6 3lOydb59jmNv/bYvg9vLjmvKTFuavE8n3dPG7e1uSE3siNppG9t8bUtrReBC3Wh8VeHcyMcL dQ33bzjcX9Q72Duxw5TmnH7zW535YWuZ152JF9NPRn6rjg0+/suZnJQ7X2X6bIx63hg/GJzy U3dLyT+D1sQ1yzbVBxxMG6oJme7+8+7Upu6fPd7dnLSCHhz/tTZsW1sA7RrNTpy9WxpUMPPp QF+ji+uki9+zEdvXhe+7VrbsfzLoaZxeW1Uwayt3Txk6TGL2uhZcbs+rzLRLCo3N23nEuCr/ TKDNNTjZ/cquL8YzC13CXR6U79Dn+OsMhWrzm9VH7S2vJB/L8SzffXLl8/W1qV5Lwx9+Vh/R 2Ze6x+3cSr3H4c6hgqDs8F0FE+6e1QcsDmV/7Z/+oUOVFKhNqyl5MeP34NCH2f6rs94q6iz7 oMdJzypcv3s0ev6A3TMWmd5zYrR8wymbLTMgzrq6aEl7vDE2a8Dzb0OGvdZYc8JnQG1IrdCn FLWW9rNKveNtW81it7G02nO5qkcXjlXEqutq2dW63DcC81MvFlTcyB8zXm5zvG46FTTgaTIx TYrJt2p5/TeNr42bGvPO3ov7qLVb622xqFT9R66O1pssvw/s87JcHxpjh2xrNeqm4dyba8wd Rz9/u3KkabzYnF4fWdX8VBMQ0zJ31fbJtu0zEV7XyPxDXUa7ZQmXjPYen0xecVz/OMuyKCwh awMJ+A9QSwMEFAAAAAgAq3axJvXdFqeaBQAAqwYAACEAAABmbWlkb2MvaW1hZ2VzL2V4Y2Vw dGlvbi1pbmRleC5naWalU/k31lkYv01Jy+jIaNPbdiqUaLmjo5rmFEPNNFRTQ4tlInfiJYPw qqgZwmvJ1qSbZUjJdiwteBk1CaGr1ZIOyjJ1KlO9U0rEM/d++xPm+e0593k+z2f5fjd8a2W2 ao9slD56jz4VRoggRBFiCIHoMEaYIEwRZgiDeCQYEYIIRYQhAmKWYkQJohRRhiiIVYYRI4hR xBhiwJEQYAQEAUXAEIAAFtAciy/zaRB3RBGMKcYMYxBneUcIJhQThgkIFvyREkwppgxTEKT4 LCOYUcwYZsA58lUMBAPFwDCAoCw0cNKcJacFQoG4xKE5Fl8GIUj0vCghjBAQ+vhZ3lFKKCMU hFzOgj8yShgjDLh6TorPEqAEGAEQZgiz+Da3g+sH4Y2QxME4ac4ShFXiMMfm0BwLhHPimfe8 GKUgjOT6+FneMUYZcF+5XM6CP1JgFEDYLFIhYlYYDcJ14R0Rq8IOECEIhUQgCdIgMhE8iAAW 0CAiEtNE3BEFPDFuJNfHz/KOAYD0NYCUN0iJgpQZSKmA5DtIzoLkHUjugKQfJIUgaQCJJUg8 QLoEEhZI0/CpR0hLS0tPT8/Q0NDU1NTCwmLLli0ODg5yuVyhUISFhSUmJmZkZBQUFFRUVNTX 17e0tPT29qrVar6r+T9q3ocxoxBqRUvEfyD+Cc1xgArHyzYWpV8LGffFYvu6oozr4WNGbQzd WHymJmo6dm+vK87cEKNjsPLzD8Vn18Xrb8u1r7/gWh2HrKoOKL31G5PNfJ8fVvpkNKZOOlKw YPyl7K2x5lOqogO3u907sTn5u9TTcdksbXvgClfVl3mthQZKkw53baOmKIfdVqmrS4x3ZHv2 JHS4Tel+GOt2raFyqGRui2r0+/L+iWVLHsn9Js+xKhtIb8tVVk4IZiVFtjXxGkPRMsUylxMu xp3WKQH4aX5M866Wcj1f+6pz1rs7hwK7X0Rm1tT6y8oLX2RV7CxMY8k+6rZLURU2Wsk+drdv hIRrBO8q6Qq/c2xNq0eZ3/uqwEwbeOd4yKy+r/HQ+auOJQvePHntqnlkRPF62xN5oo++yeBb RV3pudb7wbVdtqmHk1b5ORKsTDoZolFrsGcHGf453fujZWl7gPny6fMseozfJUZoJWQlh9t2 nJ+jGKE7ukI0F0fIAmT7yxwdjIINncLQioWzjqWrItJirgYlPQ5NrRyYFr38swcuyv423ZlB mxL6HQOnnh57NtDMdtVehbY+7Yn2zTF0Tfx7rdvMX9dPULketmsoi7f3eXnqEfUYNHOKOhqz 6F7sLmfV79TzYE6a/2pHS8ct4Xrju+O8dBY9Tp1+cOf8uH0/amYl2A4ucjjqeyZnX2iM6fPs nJtqS7fz5pP0DLL0IuHZvtyIqRHuWcrj7h2ZipOqn/aXVh1MORXkT+UBm3XWvMhId1YmZWeW 2LpHGo5Nzg34q7whp+TGbFVavs5Kd2P5bf88cs5zktGzy5rHLRsrXtcOmXz/b0OFXWXTnYmF V5rsH8xXFU1eMzdxr3mFPDvOOavPVN0p99o98LXsh2qd0evUf2q33vS6MtMpX109YNK8LLkv LM+tyn9Gike8hrfMrm6pXbt3Nb6/JP6GSWt0cc3Qb9GYfTV61oFbFq9SZGWHRsoNCzfdrPO8 axXlcbFphcXay83TXIaLW3Y++nixUXfI423zjNh2j7NLeh6+Xf+N1z+X27aHD/vX6FJ5U6vC ybqkBt+MWHipZmJe5Z3bzteNA6qnFQQleFl5O5i1uF7X6q/1C89+c8XnamRqQ3TAfu1bvqOt V3bf3emoqLQxSrvQkVZqs/WBwx+dXs+yNDvmPI2+8HLw5WzVrME+WUvB6j32XbeG1L9Yhw6u C7wadfjh7Q2b7eSBm5bm9z/Y7Of9rs/71XDH6+ClIwO9SV4jH3Q1zqI1/wFQSwMEFAAAAAgA q3axJm8lT39UAAAAZgAAACIAAABmbWlkb2MvaW1hZ2VzL2dyZWVuLWJhbGwtc21hbGwuZ2lm c/d0s7BMZGNgY/jMwHDo0KH9+/cz/GeY8XuG0Vkjve5wpW4lhhSGmB0zGLABxZ8sjAwMjAw6 IA7IDAYWYYMTkhMEXLoafXccK2RxqxAJ6nBjsAYAUEsDBBQAAAAIAKt2sSbkJzhuUAIAAHYD AAAcAAAAZm1pZG9jL2ltYWdlcy9ncmVlbi1iYWxsLmdpZu2T30tTcRjGn+najqH5hRRGFGzL iy0iOsMLR1B2pNSTV0paxxHIWSNbMONARKeLsiMUjn7IoGGHUGIIgSTBWAiHxaqvBds3YXEC L7zpx6gLdyNYrfg2L/oveq6elwee98MLb2//ya7wWDOasQVwzgEUCoVcLheLxTY3N/P5fKVS SSQS9XBf+6hpmsPDw8DSsVY9Go2OAlcn4srQyAHfdOLKHDhy2ezjR+bdWzcKVC8yiTHyjqm2 jRpHpYpXFJTqK6y+iPDfYo0Lnyp4aIJ/T/LPmS1OLAv3Upg0SI373tuCYWBEwdo6nmVxTsVr auUsLCxCOQtZhqaRyxpOyDBNmTE1HsedpEj1JJPijMhM1W1IHIMcchUSRZjqKYYwJwoXdS4M ViCZCPI6S9LmmSongxbCKQQNInKfaguSgaACzzqkLEQVHmrV6cT6hRYhKvDJ8GhE1LYNMWWV qUYcvqQo39Rpr8R2E3ZJZU7YAriAqhtU1+n4+Hi5XOYfsZbxliKR537/SvvUqvc+fs2ma+nZ UihUCh2aHvFP+xHF8uny2oS9MbnhBisWi7HS08iLdBvhYicnQr2zIsBMLvAM5R7CG2HtQKqN GAEfbxXsBhhOKG6su5B1Q81btAmWqw7uUACZEM0JzQX5lGyeVxn+6598P50OwIWD28P2T0AY Qouwt29pzt8tOIKhM/NjPVONrUevBy7MO7pb/vyYD/R56c4mR9OezrdPeh7s7782EAh88R5v aA5c/DYTnbvdoe1aDh7+OjAz2fCmo+vDy9W0w4EjfwFQSwMEFAAAAAgAq3axJvs8q4VfBQAA cAYAACEAAABmbWlkb2MvaW1hZ2VzL2ludGVyZmFjZS1pbmRleC5naWalU/tXz2ccfxqmbJ0x m0slYvJd0myPRsiOOn1V1CL7JvluFM9R33LrusSQWl+5rVrtSRfpwlyqL5V0c5AuzlOnTBfd pH3TEBV1VJb3nufjT9j7t/d53u/X+3X5fNY4yJdZbdPRMUVv0PvCCBGEKEIMIRAdxggThCnC DGEQjwQjQhChiDBEQMxSjChBlCLKEAWxyjBiBDGKGEMMOBICjIAgoAgYAhDAAppj8WU+DeKO KIIxxZhhDOIs7wjBhGLCMAHBgj9SginFlGEKghSfZQQzihnDDDhHvoqBYKAYGAYQlIUGTpqz 5LRAKBCXODTH4ssgBImeFyWEEQJCHz/LO0oJZYSCkMtZ8EdGCWOEAVfPSfFZApQAIwDCDGEW 3+Z2cP0gvBGSOBgnzVmCsEoc5tgcmmOBcE48854XoxSEkVwfP8s7xigD7iuXy1nwRwqMAgib RSpEzAqjQbguvCNiVdgBIgShkAgkQRpEJoIHEcACGkREYpqIO6KAJ8aN5Pr4Wd4xAJC+BpDy BilRkDIDKRWQfAfJWZC8A8kdkPSDpBAkDSCxBIkHSJdAwgJpGt73COnr6xsYGMhkMktLS1tb WxcXF6VSqVKpQkJCIiMjY2Nj09LSsrOzi4uLq6urGxsbtVrtwMAA3236H2UyMl4HoSZkLv4D 8U9M1AWUo2dkn3v2VrjuVLMtVbmPbdBknWVJ9ppzd6NnYJ+2Ko3X3ag51sdKZuzNqI6Vue4b P3ND5toji5Q2YQ7+XSxRz72i3Dtmfk3MyvDb4802brVPtbN7udFhibdThoOHReCxKWZ/ZW3c 4al/LECliHZdHzrWgs2dr213OFfz87XHLXkL125+Ias/W3t9XOsqI4/TOR43dQeeNKxV5N4p 9v/ygxPqZRplyfLP5dn3pl/6u6zp49QJJdcHtJWB4+YZ91oN3qo/13e1tI3219WluqoOLjcM 8M5JsI49WZr4RvVI/kP4FXO3wt787HzThLaCgOe5F/zAumglHtI2Oc2WexgNv2xz1aggoNfg 2bo7J/zedvse+HpXRXO7W1/7irGnvZVDu66UegT19/fmB3ccvKxXVbfdK9Cu85DOu6zk8Iqh +Q37NYuGO4MGAid2hj0aO99x2GiSrmHE6JT49og5iuDWiEmkqnP34ov5bepuvaCuX1mNW1qE zGZT596hhh3Twiw78qb+ZjJ6/seAscwq7S/WMY2JRy2ce5Jjre5VeqrtHOOajzp0rOkK9FxV oI1cb0MzvX/6cDjDs/zwjfkJLfWyJ6fcO9Szf/c1HslKTL7ytDNZUbR0S1KN+4z0U3ry4fo/ 5EEjs5KjPB0eZRzd7NOZbjZXnR4V9i4kMytt0p+y2NsOBv9kzjUzyww41ALlStednvcuxHo7 39BeyrRZc0mdZHI+M0etUC+4HFqmMI1Tys12yk98GmGqSrFwnGzZ7Gm4vqBtX/Fc1O7z3L3w wRK5RcGQpsj7+MnRh+npV4uae5ImpDQvsu8JLfYr6+6Iz7mh/YDtLhxO8dhT0JrykGhk76yS YzQ2UafLotqLv7o53YvFFE4e259QafiJ756KacHHF1cbtU6ct6dor2NfSajjF/tumiSYv7o1 Lt33mwoc3epf23nSaLDadqr54rv9Ry4GlAxWbHpVu7TpX1xkfDuw/oF1+rq86s2zS4dKtilq eiw/+0i5pPHhhmkFDSvTZuU+sE918q9TdKgHq0buxy9sDTN0Kr9W4VLZ7uczYYrf66oDC7IL 5SHhYd+mfbcgrKUrzs3ifuvM6JHXW+ODX4587zWztq8nI8G5dfWZM9mjz1Z314w+z22KC+rN a9r19kXhfuPTaMV/UEsDBBQAAAAIAKt2sSYkUzuxZgAAAGgAAAAkAAAAZm1pZG9jL2ltYWdl cy9tYWdlbnRhLWJhbGwtc21hbGwuZ2lmc/d0s7BMZGNgY/jMwHDo0KH9+/cfq57yn+H//6Xv oxgUiwUM53gl9xh6HRdtvaAwYVfo5VsF1941vWMAA8WfLIwMDIwMOiAOyAwGFlGDE5I3JDz7 2gIDvJokHCdJHFXszWJhsAYAUEsDBBQAAAAIAKt2sSa7jWn2VgIAAIADAAAeAAAAZm1pZG9j L2ltYWdlcy9tYWdlbnRhLWJhbGwuZ2lm7ZPRS1NhGMaf5dpOYfhVCisK3K5cRHXEwBERHCj1 5EUsUjpaYGcuxoJZByI4XpSdLspRXixadigtlhRYEoxFcBgt+aTavozBCbzwpmAQ4W5GWgu+ jhf9Fz1Xz8sDz/vjhber51hnaKgRjVgFOOcACoVCLpeLRqO1Wi2fz1cqlUQi4YS7WwZM0+zr 6wPmjjTpkUhkALgyEldO9u/xjycuT4Ejl80+fGDevn61QPUikxgj75lq26hzVKp4R0GpvsCc RYT/Eetc+FrBPRP8e5J/y6xyYlm4k8KYQerc/8kWDAP9CpaW8TKLsyrmqZWzMDML5TRkGZpG Lmo4KsM0ZcbUeBw3kyLVk0yKMyIzVbchcYQ55CokihDVUwwhThQu6lwIVyCZCHKHJWnzTJWT sIVQCkGDiNyv2oJkIKjAtwwpC1GFj1oOnehcaBaiAr8Mn0ZEbd0QU1aZasThT4ryNZ12SWw7 YRdU5oYtgAuoekF1ncZisXK5zL9gKdNaGhx8FQgstNxYbJ3A78l0PT1Zam8vte8b7w+MBxDB m1PlpRF7ZWzFC1YsFqOl54Ov082Eix2cCE5nRYCZnOEZyn2EN8DaiFQzMdr8vEmwN8BwQ/Fi 2YOsF2reoptgeRxwlwLIhGhuaB7Ix2VzWGX4r3/y/3K7AA/2rg/rPwHhDLYIu7rnprZJo2sk HJ4jj9rIkx/BjsWh+bbRn73N3cLj6aDv4IutawfO0cPxDxO96RORHXc9Qsy8f6m40+XZ/Hk0 FG2cvLX/Y60WIuenU8P21NueiWeup50NG3DoL1BLAwQUAAAACACrdrEmBxqE+h8FAAA0BgAA HgAAAGZtaWRvYy9pbWFnZXMvbWV0aG9kLWluZGV4LmdpZqVT+VNURxCeWEbxwCsJIgK1FqJc HuUIimK8CtYDI6BUUEBClIGSAEIUcSUIqFxqKmEFdkDCoYhsllVBWNklGg0iq6MYKUFDOFRA Aoq1pQS1TDozzz8h/VvXdH/9He/J13stdQ9tRnPQCPpQGCGCEEWIIQSiwxhhgjBFmCEM4pFg RAgiFBGGCIhZihEliFJEGaIgVhlGjCBGEWOIAUdCgBEQBBQBQwACWEBzLL7Mp0HcEUUwphgz jEGc5R0hmFBMGCYgWPBHSjClmDJMQZDis4xgRjFjmAHnyFcxEAwUA8MAgrLQwElzlpwWCAXi EofmWHwZhCDR86KEMEJA6ONneUcpoYxQEHI5C/7IKGGMMODqOSk+S4ASYARAmCHM4tvcDq4f hDdCEgfjpDlLEFaJwxybQ3MsEM6JZ97zYpSCMJLr42d5xxhlwH3lcjkL/kiBUQBhs0iFiFlh NAjXhXdErAo7QIQgFBKBJEiDyETwIAJYQIOISEwTcUcU8MS4kVwfP8s7BgDS1wBS3iAlClJm IKUCku8gOQuSdyC5A5J+kBSCpAEkliDxAOkSSFggTcOHHiFzc3MrKysHBwdXV9e1a9f6+PgE BwdHRkYmJCSkpaUplcqSkhKtVmswGIxGY2tra09Pj8lkgv9Xs96O/gihNuQi/gPxT4w1A3R+ nPW6C8XXDpt94hTUtEeWnG5u5zRadrH0xjHLBWtYk293Y9rklfGnLP1ON/3osGWaB/0h7Ga2 LFTtYZldeiMfT9CzR3Hlq4vmpGg/zao6uynT68Cy2erX+d7ZcvmT48bXvS0ZWwp2RL371vmO eoJyEbvl7+JX5f573IO/cuYFnHO7MylsRk1E4OXIl9p96radIWqvB6X+M08u/FMfSq06nUcG Q46k2E1OWnbAdP/Xo87br9TrwrsaA+w3DikumSKaVTGKsokKxy4dDgxJ8n63ePfVXZ52jTbb hnerKiKvDk3c5hZTEnd4btrDg/Gv6mqb9Q0zOtxD/7DMMQS/T4yP6Lj7SwysHDqoMA61Ou9s 09go2gcKWuZe3VCkSHxc4/ns0InLum5dv/GZKn7EFF6guHf0ZmeMKso+V/F34dbu2OX+4bLo caMsbFLrHS93R99Z/TA0dWrW4tBkm391vtHDw23F3k5Z8s50Wf6DJ3IXrZn18YeeAUFBS3b0 l8RVPLLsyViT9Obp93NIjp2n10/7S0+6DgRMT5gR/UYWOX/7tvxYXe9Ib45H7jj71C1+mV17 hirOhRQsqnLsVF6osQjbuvmWPO/oxlHrifLQDrn94aVxmc+2ZtQusShJdxvoV2VZmeUqp8U9 6ovOvpJXftzXL6C8ONltfHm5aoNT4bnYW46zC/d3uNtm/2aCuqCcZH1YsCJp+G3h6Qr1bgft WF+7bzZdnzTT72LjhjQXV/2KS8+rbyhUszSZ+vGDlclhxu5TmjXHNL5dtgZN1dOBvhd1LCtQ W/141JQoTYevPvxMy1kz9f6Umtvz9Q21E6J2QWqB1tA/pv78FbPBn8vyvk7qK9YstC+Iuf7x l4lavUU4W3Ct2dZjz80+p+fk/Gy5TttkPf2Yqa6h7MzFa2OqDVURg58v+6rItsg7tm7eyKmW 206b66sbUkpzq+++TOt9dX/eQUPcPce5iZWrfFb+M6TwDq/ct/qLkPqydbKd75+27spbkd0S 6HeitproXsQ3E/+J7e17275rWfVZe+VIZ+SBqbdHupJfndzbfWR69JvHGTrrKWj5f1BLAwQU AAAACACrdrEmffllo1wEAAB7BQAAGQAAAGZtaWRvYy9pbWFnZXMvbWV0aG9kcy5naWadU/tT 1FUUvzrGq6DSSIaIwgoQ0SFvlIk8EgUBLYYZhiAWB6G74EIoxEMciolEXcFoCJaLwQDq4kJs mILR0jRDQYJ3Kh5ujAMMJSzIK3Ewg0lP93z5Dzq/nbnnfM7n8f2GhoXseDNRR9zJA7JalBBG CCdEEALYUUooI5QTKggFfGSUMEYYJ0wQBjjLKeGMcE64IBxwVVAiGBGcCEEESCQClAAjwAkI AoDACC2x5LKcBryDxSjllApKAc/KjjHKOGWCMkAW8pEzyjnlgnJAUnJWMCo4FYIKkBzlKgVG gVMQFAApowZJWrKUtAAV4CUJLbHkMqAg7GVxxgRjgPrkWdlxzrhgHFCuZCEfBWdCMAFSvSQl ZxlwBoIBoBloltyWdkj9gN6gJAkmSUuWgFbhYYktoSUWoHP4LHtZgnNAI6U+eVZ2QnAB0lcp V7KQjxwEB0CbMRWGs2g0oOvoHcNVtAMwBFTIEAlJA2aCPBgCIzRgRDjN8A4WyMSkkVKfPCs7 AQDK1wBK3qAkCkpmoKQCiu+gOAuKd6C4A4p+UBSCogEUlqDwAOUSKFigTMNqT4i9vb2zs7On p6evr29wcHBkZKRKpdJoNHl5ecXFxeXl5fX19Uaj0WQy9fb2ms3miYmJxcXF1d3/XW7L69YQ 8jvxxgb/CWsbIF/buuxrHQ8iVhu84q8fcSOf2j8eULLvckP3GQePQyPDd+t7tK5umS5hAw2h ZZsiwp9IvXeh78wLB9n+Xh+90Lilmj88Rjf9WrsuduXq8nZ1REPQhkf7tVmG3Y1eJxr29C29 OFQbpZ7x3vzalFm/x9DXmT+YEvqN0/dND6sd06JbD95o902pnLhlCv/YfECbMzdSUVBk6fZb uj1Wd9Kvq2DLb7O//FD4484Au6PbVTfUT1udrXZ6Nbm5LDDQ35DjY+lvXNZbbc1vSa3KfeXK nKjOSb6sSzWF+Pcb0gdN/s+5R1Tlpmp/zjhZO388a36i6/yU+dtdry/0NG+EebvsD0YGxnTF kNsx+l3KXPXnmTHv5O868GfI6NRHpQnqu3NfjY6FPNl5Zehmea/HaOHk4jMuGf/aekRlWVus Y3KzS99QJbeteRAf76Bt1KW1d/89fMQlasf4oYf3c2qKVm7q1hcVOKpdtD4LjonZ9ocnY0/4 mHVJpYNbbf447vzJqcpsbtuRVBaQ6VRVEuhlM5y1957TS8e8tVXDCUPJhslKt6OxddqgKOs6 zT9t3fEVdpar71ceNt7RV6q67vSfG/pJP3nikWNo0mePrW21qDJ6whrPZeUvP1VTuD78QiaN un87MdN1pTHm1LM7py9Wve11qeG088x0RczzGy9drE4In6l5OXFGrU+vjZsybJt5N2WkqSI8 fCq2JP2L1BbLZoNjS6fFyzv0fPU1dnbtXjvvpusOs2lxNa6a6Le+ZMtzbcPuftO7K9js/LVb cRHz7eOVmoWgidpmY8eAcXabaahzy1+dS+b3ooOI339QSwMEFAAAAAgAq3axJvgcAZYzBQAA RwYAAB8AAABmbWlkb2MvaW1hZ2VzL3BhY2thZ2UtaW5kZXguZ2lmnVP7V8xpGH93WbS0a3Xc mtK2Ugkd9jWbXM6qWTbTWazVhZoU9o1SId1copOdSC6rVl6JUrKZjMYypkbOkaTRG7lUuuBU U3al2px1K/bZ9/36D/b57Tnv83yez+X7/X7RwlnuIT1oEnqNPhRGiCBEEWIIgegwRpggTBFm CIN4JBgRgghFhCECYpZiRAmiFFGGKIhVhhEjiFHEGGLAkRBgBAQBRcAQgAAW0ByLL/NpEHdE EYwpxgxjEGd5RwgmFBOGCQgW/JESTCmmDFMQpPgsI5hRzBhmwDnyVQwEA8XAMICgLDRw0pwl pwVCgbjEoTkWXwYhSPS8KCGMEBD6+FneUUooIxSEXM6CPzJKGCMMuHpOis8SoAQYARBmCLP4 NreD6wfhjZDEwThpzhKEVeIwx+bQHAuEc+KZ97wYpSCM5Pr4Wd4xRhlwX7lczoI/UmAUQNgs UiFiVhgNwnXhHRGrwg4QIQiFRCAJ0iAyETyIABbQICIS00TcEQU8MW4k18fP8o4BgPQ1gJQ3 SImClBlIqYDkO0jOguQdSO6ApB8khSBpAIklSDxAugQSFkjT8KFHyNLS0tra2tnZWS6XKxSK pUuXqlSq8PDw+Ph4tVqdnp6em5ur1WqNRqPJZKqvrzebzX19fR92/3fZvx38EUINaKpoxD8x dBig8xY2XsU515KHWbkE7hmJkHrkROXxqlr7G/vG4bCWKt2p6ym2ilSHN5vyPQ85/nTW/OZu 263MUQHlNs7RBTVZDt5tLdSqgx2bm1ww2OWB493MaTvG7C9c/ufd014q6zqLmJB7+d4nfeWF dWvuq1WaINspMaH+B12qw+YO943w1C8w3bJUZjwLMPqWtZUZLod6FDmdj4wY7h/cWrjsE6ue gdcblxiSHvXLnxt0bdeG3CTEpu+FB1377R1Vi9s/K00zvY+NOUdf+p1Ycv1UQaD7y3UNJ6Y9 vaoefaGr6kpAu7ZsoGtD2sr0ONW5ba/XtNYky/xqVvSqSn7XRb2yGZi1oCMstxliXctexT9r 9Oj1WTXvRXjT5bw5B3Y+cKvoaaWZOw0t7xIOVdRdnLD1cUlcdmzziuV01/LRnU/iBjXVBSTO TvILTjj4fsiq6M9XTpIFjznhZBc0vzH0UcSXSv/VW0NV+pzUH27cHBuaWnZ0XOK7Or/cvT32 BrvtzjVuj5JTWo/ap/TOkGclf9HvkJ8oe/Fw4pbJ5UfN0TMP73nyq39QwqojI76ZvDozzaPk x3WLv27MiXpumdCe7mFdYr85+Or4vN+Upip7PzvtutOHo5+pO2LXfNrfnvZz/6KQYzF73WxO 3vzL0PlLUqRF/hFXd7fCzN22VR3xmZ0mc1JG4vj1eTsLOtfn6LudrOTF5ZBqi3T1yv2B+s6z mqL3EVuddrVNVrqcyfZJCTs7blSFRnFJn9FVxJpmd2nXdodF/GG3omtDUbNSk5MXoixwyo6J dW/0uUO6Ii62nNqmKX3o633O2BDU3F16UT886oqhUik7lL+I5iRpS/N7/Ut7ZFGlfz82Fl8f 0TBlxuWXFwa0N959xaaXqR23FehqO7N05fduV29YGHkp9F6FY1aRq2msU8fmSrte+YVK2b7j msKKz5gLGa/bPr1S8TRwyu5lJc+nVuhjHYq3nPx4XnrtlbDqmfIJ5Wnnb/f2y+rWbzriIl9c 01S9pT7Qe+6r7yK03XpPxZkdtaXGqyqcEPt0v8/9QRYO2XFeB/6dNitqo2aewct2gtYwv961 5027Z4LlW3OGX+TbjoNDM9Cc/wBQSwMEFAAAAAgAq3axJtxeMgp4EQAAVxIAABsAAABmbWlk b2MvaW1hZ2VzL3BhcmFncmFwaC5naWadU/c/lf/7vxFJkRGZ5RhlJRyOQ+VtRCGyOWTntlL2 ShkHOWacnOM2s45D9soqIlmHSpmpZNOQVEbl/p7b5z/4Xr9dj9d1Pa/nuO9L+npYNWdthvvA FvC/QgMACAAQANAAAEY6NBpAgwAaAtA0AA0jjyAaAEEAhACQBoAwMguhAQgEIAiAaAAEI6s0 NEADARoE0GgADaYjATAagEEAhgCYBsAwAoxA07Hoy/RpGLmDFIhGQ2g0DY2GkbP0DgTRIIQG aWgQRljQHyEQDUFoiIaGYIQUfZYGomkQmkZD02A6R/oqGgbRMISGaWgYRigjGuik6SzptGBE AXKJDk3Hoi/DiCCkpxcEgjQQhBF99LP0DoJAiAZCMCKXzoL+SINAGg2kwXT1dFL0WRCGQJgG wjBiBmIWfZtuB10/jHiDSKKD0UnTWcKIVchhOjYdmo4FI84hz/SeXjQIghEj6froZ+kdjQbR YLqvdLl0FvRHCKZBMIzYjKQCIrOI0TDiOuIdiKwidsBICIhCEEFCSMNIJggPEAFGoGEkImQa RO4gBdMToxtJ10c/S+9oMAzvfw3wft7wfqLwfmbwfirwvu/wvrPwvnfwvjvwvn54XyG8rwHe Zwnv84D3L8H7WPD+NPy/HgDY2dkFBQWlpKRUVFR0dHRMTEzs7e29vb1DQ0Pj4+OJRGJhYWF1 dXV7e/vAwMD4+PjCwsLGxsb/dv/fhdo77uvs7+zu7+zrIerrf8vXzT/Q0y1AlJ0NQO0cOMQK TAByyBzyuwCse0DNIeHLtQ+f4Vl5ZDooHJrNPpziBrmX68R0ffjQXjP9dcV9aaI6iR3YH8UY YIGv4tpAfakhmf11eY5+A2U4B+tDESc3lL0s0MSziRg0Ul/nLIyK8GZ/s2shmxQb5MmYiDQH WhyvMISavK0FlBjPXIxNyLQmux0SMxvY/KIVNJvflmdoXWpZGfTIf3hoQlS21upY8GpMgq8j PuF6pYiU66e25ohPgxeu7Hx5/9RO1ryZ0UqxuztZ5fCXH2Qn17i0MedU2bamt0PpW23BOS2b ns+TBBrWGEk/u94/MXRkku953DWdb+SIvluluD4HMnNvDUdsbYy04gzyh+3/jD39PLgXFOIt dOP3xwHm7M6bv9fnpnuI6xjv7VBy9PWaReuC8363BcBMHtaoP6uPre7svp8YDZ5X2R4PSScs i1oykULstLp2SA4xQ25km3vMmaWzmi8YqCdcp3Du19TWbELsXMe5nufaVxp5iOtwLpWmxqrq qdjcwp85thB3mntqJj0BtwL5ioqVz/gsVgaUOI+j3N+FsSlOuuJrF3k/kMzruPkNUwRwFC33 ufhXAXnSHhSi6YnyIhyzXqlk1kwQbiI1VE+Son2cBM3dQd9NmPW7wrEMBp7GLKMKk3QFxKPK JAWLg1V+q+XkB7pRyg0jTx2TSPi86l50e+KZ1Mkgtj3d5QS4r8qyInKjSjf10cQiRqb6MLhb 7pE7srhaIxsn41UOaFjn1Xcb9Xn4vBCp8AgY2XshF56czfbeMITbEFd7k/1zlfaLjB1Pr0dG mG/Os7Yv5ttYmMVqeH8kNmWk29VnyXd8XLKZUF/6LlSTIefcWqpmprf0qCtv991q+xFV6/wn XDzHyrrXrG5APd3atNrMbb3Ksy8EVZPJ5BPoCLcBvoZ2hSGLps+8BXLiBzc1xT9v/xgO9heu viFIZkO3SWzmZtLsfUqzX6vKpkDmV/qSjdSWsgJGxqyt7VXGcbeGf6nZRvoYYx1cq5rGrhWc CZ4CG9iDJz173je/8+pLaX7v/eZMyMRe+McLIR/u4o2CZ4PgvMeTi6GbRm/NcUEc6islPGHq GTondhZIOvJqi5DjyA42+5ZD2EqW3cjOSklxqtrag8c3W7HpDWJ/vjRO3PzztWW5+s+3jq1U jhVi+8227z0B+WGfMwrm2bHJnUubOMc5aPSIgdWtf78m3Wv+/eYb8R3uVXevFMAtnFQlBU4N cvCSZis6NbSU8dInfqHf3oL/bS/XwHt/t9Zh+B/T0BY3n8Y3KUqEsXa/55rF6QDeA9fYA50s zTCFOwNAFD/tIYtqgpYAjTHi9rSAmzYL1eZOryincpaB/LELm35lhxs8BP5hL/cITitEnZRw Yu21hlR54xUi96Q5/Dk9tA2i40zHtDO43HyBjiN1mjrJmfWNmcJXmdNjI8rZe+VCe+TD6+JP xF5hcJ/X4YH6FQ7/vF9payk9T7tEuPF80Gm6J6dXlFW50Ws07i0X6xzrKxUr7sPHOy/3a8+Z JJ9Tfuhj3HhSrNT0TH3Mgv9oSMGbGT0R2RXxSs3A8te1Rcak2IKM8EsGZjFXWN60P0RZDxis DTOm53F/co5o4d+6uJy+TbguLwmiJDkOEKM9nO4ah3V9if1BNhBMViIP0ED1TWik0UTGTt9I O97P0zWQ/cyYix7Ptx/aO4emFAUy9J2rJWNYzIwZxzaKdRnrzydkgSrLnjpHYxRI54VhVPbo 6lPf8IaEjPMP4gV1RzS4CQbHKzIUDw1ybmz7YfrrJkRX9U3ZMafcCRLzGps84sp76LzZa4dr 9FSMneT7VLvqxIVWG18F/aBeqzqUdMQS9zp772X+JI2KUd09cb3UIGpvprQo1mr0yt6DhFeJ HFIC090svb4UTIoMx7q0HA5z1vBljgCziLJpefjY3d9vURo1bLwlIuwknrk3Rzp3LWs7lMpc JWWv1Gcr+96fr17s93ZYN+B4spZZYzH0SG2QYfIZC19tuIicy7/wCW+TtNv3Uyqvn/YWWi9U uWaVKcu2Gt7o1xVk6MBa5XHJcHo0+3cj1ajKy7rdVovzTE5202H3PVfMf83Tv5v3/nxwV8x5 OXnz93VYpk9OynBmTnCrpX/YgWNr0W6KkSz/SUDeyM/Q/l1bZkWHT3WAovjMqtVWu0OvdEBT DmeDmWTbB+aacFmvq+RfpA5s+M2++PYPu+MWHbr4dfzCH/Tu9FbnWMdS6K8/XVOYVOE95toY dunZv9z8d1nusse8aZ9lw29UMd31tRS7UfbNdrtH9fnZGyrCjt9O8T+XGnGyQVc6b2JGeqRT FQHdSnNuMlb7NKkudfOxy0HiZK+Z1tm4ELn5z4mZEZ5+ztpHTlHKqJyob0yWzd8PXD9nhG21 GaknPb6BOpITHyFhz/qU5ecxuYpVyZUvpplDe9kZspw7w1vt3d4C6sWxyTgWN70hIrljudE/ a3jCQ1f6oNGKttDuqwRMoOxSbtdD9aEMa+UhMSEdkvTcwXN/CEXRcl+NC+UPWKa8DQhTI4E3 SzOGqMZNFY/dcS3xFsb/dlz9SgIElU5Kp/2C5k2OPhA6n03V55Jr8nzQxo++tzLwjPt3TNu5 61j8ZD9zK3t1LrHqqEr4mWbKT+1zU6ibRLB89TzmZnhV1lWOUxkN4kMPhg72MtbNAQVXfhz9 08H9U11o5AKHEectqEvvifq7T+BCWH+ZhJiY9KS0DIXAbU4K+7g7E/b3sHMo3iQmR5JTVvRI lfBdZf6Pt/uU9bhjRkvgkcj5nfHDKqfVn6X4BsW/0C9kSQnkizthH88yH0pIr8e5rKfhI1rl I02J1T2v/gIR5YS1YO31Hn85LdN5p+YwmrnGdwPxVoWErp2hkKcZIouTy1Ijqlpu1RVb3ykr kxqtzWPlnx/kPxiXs3Qf+hrwq3EGXBs0VXs70MdWpZr4ORhSmAyamSXf5tZ/s7Eon/bBTEZM ofxkb6nEV56iuuPnZo/+10zbCGDvPJn7wf7431nXa6GZFyGdxWE5N80B6u7Mvx+ze0FVVF2x GzWz1zPSoJVLJ4PPEvXL6yOT1o7ig1JP+Foo3LtjKOnAjxpfu8e4EfHt6N+gD2dbflI6rUjx H8mhe4/bpztN13y7CgzTNWQHejzFlvvWM/Pt4B6XtEcptyrF0zo7uzdeGgS3jc39FzC0t1Xm tGOFMlcUteIU+bvl5P07cpQ3auKxi5VZlE8326HLTuH1+mfrUF5NXsKT0SIrF+RGDc1MUT7d d+qYhSSkeJv8osILT8Z8j1Hgxyth8co2eEwoHpuFV2/Bn5/EYzxk1x56naLiQhiUZDyEStIF NFpir0zGGm/HmmDxE/cFxfh5JSaFdKSsQ5xRv0bjHLbjfrmg7FCsJ2ziQaqbQ5Q8FzbGR+kw L9XRDqvohL3nIBW/qBnPEXrH4aEdk1bUDQ/hW6yOIX66kY0al7FxIZM34DAbe6Xae3aSvFPc CQvMQhge3vsTJqEtCfBYQsSK67+D147x8MphOT+lE7JtHHMT7EnpcdebouRNExNDE8cXE1ws LddVCHsHeQhSSdSVhPamxMdU++oW0QWnxI5t+5R+j2ZspNALR5Jpcl1osllmMpk5mXY58jZj yiirQ65NCron0Volebou6V1/QmO/C1tW0hmGpEWPVKbYVD5q6rxH8td+q/LtpI/8ycOSqfaX Un9hk/KwLgX99tdD41nC7hMlWddXCKsryWgid8MrNm619GO4dL6wdH4oXbA1XXgq/cROuigh /epiSlBp2tPetOhlGzjBzvxyxiWm1GF8hi1DCsyKI2LCLniuZCgyEZPFibiiFFnPtK1CYkoP 0XSeeOhZqtaU1VZh4k5oymUC5zntNMMwRyNc0lRvsmkYpwEu01TF2xiXYUL/uZfSHDwzD0vf Z9e//zqUdGT8wbYUyVafFH0p86aAk44A+axnxunLZLQnOWTKoLcwGacWaUfQ1hTI+qmcdXFH UG9A+d6OJlie9fYAKbyVGKEfe5cgSNSHKg+QbDxTAlbv67+B8ndIDDjX/5qhxcwE1Z3s2rC0 76z3Kgmcjv7ZPyU4nVvJrvrZ7kHQBKtciudpEzR5UUE1tkE1rkfVUgdDO4R9RVANJube0cl7 K5DXT8h7qZY3M5UH96thPBpyb/Vg5soxS7PsUHncpC/H83m1CnH1PPz5PKZzL3jUvxLOV77J f0Ys2JIu6DUv+CNQ8E/tIZBY0FdcsOepXkpUr9Ap2Gl9+Ej8IXMiR69nVtMOnyZPLodSrtuv LD0e6PtmwvzcPb2GnKypnFlU0YcfwuYvlE7ZPlidSjzuxWnUq2CtfOr8ZlGqQPEnbHEGTik3 WkSSLfdcYlGEafFffwNJweyratlZ6YXfiyQKbZSEB8+cGESL7hqasJ3VkmGyiiqW08kq8ix8 Wn5bwdzd2bZUwa/UaL74e0Pp3xiKXDHlSXex4iqlhFokO10kP1jir0+ZCaJE2CrxS0ZiSKUw jwTVtLg4uJuigKcmUKlXeqgiXmUXCGWEQSV/A+rdwRxjQaGMRFTUBO9lCvcdvbIsl/JLhPIj DGXfYyjfnH3ELTzBeF2X6ZIotors6aJOTIW5n9Aotw81usJ+mSrHVL5RrOhhIHHgtHIq2VM7 3FNj0Ug/MES7TWg3Q+hZSFHBmiRDwyOWQu+JaLPi0eCgQDODUtEQC33fMqnXG3pldbI63Kyf nXz0+6pabSMLB43yUY/WUOdSLfguKp3LuSh4D6t/4JPYGzHxOxPMv7L5nlbwvrBV6q8oe15f vsF1K9s6SNbat8tW/0BD8P2fx7/81G9UpdwbrLg8qMDvLSrTV8eCN4ajR28VOhtP+EbFdvvk Z9fcdqnrNHP/ulRDfO6ziKkFcmq/CAoySnpumHK+5boiwhge5FWiPlfx2cy9xUyQoFW9ZlCd j6pWeiz7XllwS9Hspnbzay6JtNc1L8DSCjaoUkael+nRs0/1r8MtXq1VkluU0JcsRrOF0xTF +FsapO0oB5trClBNGO1K5R/NJwUfa7w73aBwuESxlY2rjt2MOjRZ9cxSMXZSwCkRhZ8+/WCt 2eJdw6AXqlhWobSdkppUev+R4ujYKUP1nBhqfXrSaeuWtsp35du2LVcHlfIUSwMMSlJ224yY AKW6Y5rpHPm9Ha+Eniiboo5JPRmXfTLy6NjEbR5Ro+iOtDVOyU9c/FxPF+yeLt1+umLHNZ/z lMeP28ruOAMDcO7/AFBLAwQUAAAACACrdrEm5E/hVb0AAAD/AAAAIAAAAGZtaWRvYy9pbWFn ZXMvcmVkLWJhbGwtc21hbGwuZ2lmpY+hC8JAHEY/QcYMysUFg5dkYjnThkUuKF4cBlmTixoE 44rg4pKsaRE1Go1jbfGCYdFqNBlMP9nf4Bfee/WbTMeev7Bg4QPkeV5EiZFLw5TRUQlJCAjq DVnAL6LUwCcWkojIDl6QR7iUlHR9Ewsy+CncmAniurRlDDeE84S8Q2g4RZZlEDeIEFzB2TCx qYIdlTY6XoInQuHv8W+9VrlfofoEqz3aru2O02qc+OA0x6HZPddmvct+6j1WOwx/UEsDBBQA AAAIAKt2sSbin1V0pwEAAA8CAAAaAAAAZm1pZG9jL2ltYWdlcy9yZWQtYmFsbC5naWa1kb9r E2Ecxh9BzgyFvouQxeHGgMs79a5JTXpNUvvG2jTV1Fx/KReypBDkxls0uQ6FA5dAl1uLi0WH 1kxHIMqJQV63GxS6KD3okiUoEuFr8kf4TM+zfHh4ntW1oqY/m8McfgFEBGAwGPR6vUajMR6P +/1+HMetVuvO7R3f96vVKvAuO+/U6/VB6HyRhpTss7SiCBNCPMKHEGHofJJTFKO/fEKJHzFO fNC1Rz9PfxMLArzqou2yCalfo4TrYtvEt0u8vcC+hY9h0Avw+gxmDULAttlzGwUB3xdSWs0m jj0eOp40mpIJaTkRDEKFIEYwQuih05XQiZnEHUpUYhg+UuRFdDoiVgmgd5FyGSfVihKGi5SJ 5CWMC3ALyTCYVuPTAc7ATagCSZtxe2aYLyxpuU2oHhf4n1L/3LwBKLg7C7NPcOvwhdLuuEcL mq5rC0eL6Uxm6V52sZNbzhgr+UK2mGtrmfzq/bW0KD1o60v5lfWHorhRfrNZ2FyunG+9VzQh Hm+fP6mZj5RcqVja2Nnare119g8OyrXvTzvDnFkum3tD90q5Gr5E+h9QSwMEFAAAAAgAq3ax Jlo6OfYTBQAAKAYAACAAAABmbWlkb2MvaW1hZ2VzL3ZhcmlhYmxlLWluZGV4LmdpZp1T+VfU VRR/dtBwIY4ZqXNSQ2IZEEpfmmBHTUIW9YB6FBGxjHjgjGPihhDiMXUCQWUTehrLGAKecVAQ BmXAMjQH56FC7JtonMElUrQRcbu99/U/6P52z7v3cz/L9+vrv3iu54YbyAkNoTeFESIIUYQY QiA6jBEmCFOEGcIgHglGhCBCEWGIgJilGFGCKEWUIQpilWHECGIUMYYYcCQEGAFBQBEwBCCA BTTH4st8GsQdUQRjijHDGMRZ3hGCCcWEYQKCBX+kBFOKKcMUBCk+ywhmFDOGGXCOfBUDwUAx MAwgKAsNnDRnyWmBUCAucWiOxZdBCBI9L0oIIwSEPn6Wd5QSyggFIZez4I+MEsYIA66ek+Kz BCgBRgCEGcIsvs3t4PpBeCMkcTBOmrMEYZU4zLE5NMcC4Zx45j0vRikII7k+fpZ3jFEG3Fcu l7PgjxQYBRA2i1SImBVGg3BdeEfEqrADRAhCIRFIgjSITAQPIoAFNIiIxDQRd0QBT4wbyfXx s7xjACB9DSDlDVKiIGUGUiog+Q6SsyB5B5I7IOkHSSFIGkBiCRIPkC6BhAXSNLzpEbKxsZHJ ZC4uLrNnz/b29g4KCgoLC1Mqlbt27VKr1RkZGRqNpqSkxGAw1NXVtbS09PX1DQ4Ovtn932U/ bDUCoVY0QzTin3jbGtCZ0R/4nc2/tM96gus6Y5A9pzZyfopf6YkryZOwossY3VubNHbBor8T HznUZbisXG9V1/CRd5rr6N9T/M8VBmTLIr0uT17ZZ8ydoRpTUTwhvGG/70onn7ptxX9mf3Fo sfuFzIIbx1blKSxtmVGt+0LmqVWeFvONxIgrhT4vmnvXlClPzYk3lLo3nw/t7jh6YUi2vmD3 2Eabcdl3mqvjL91ZGrdm5oaaiOQ5o15WPuq6unUk5MwYXxCYG7Igr+blDnN/zvbbh3euu/DE 749on7yjuzssipIKm2y1/LlH5JGAwddZRq/KW/rySOevUn6OUXZV5tgu6AnF5tTS7KnLNtec j26/WesNCbvDXjy+26s7ntsRc2a48N5w2a9Vr+aZhvpMD83ze/TPLdoDERMjy9tbuqKa/bNC w59N9j0W01yXPe2HxmHnqRF3967O31u53Nd+2/L2iU0J48vsQlSW7iKNP3J1XhcsL+zPj4dX Bb2HXK9PzNs85YT1tESPRZ+GJNsGRP210WtZvyZt6Vbr7sRPsozh6oUBUSf3vFP2rO2IZ9Xl /O8XjpKbM+7HttJg59Wxt9MDX+sdFMFxbY704hV5f1qjz6nwVLs0Y9He7T2evSkxuiSHXKXj qbuaFdfa8w44rUhyOaHS0zDFJn3byUiNbq4887sSeW76YbvIoqJvdrZ9W7DJS1ac/tlWmFQY 2NSk0Oqmdx90U1U9XastTq26f+z0mcIk+Q7jW/cLyk7fi5uKKg+atedsl5kelLes7nTTGRRL 6tMdPWRf/1KcpJTnXDTLBmIfJOx3r7hJTAOGzrjOzdUV15fUxwa/e95+o2KMl/1FS0enzvBi erVbzbiiJVtqm8jYh7WjTXOis/7doM/PajRrPfSjPo53i05cZVVaa79Ne/bqlGnV8t+Or3eb VevYmtJQ1fl+9SyTfMLOn0z/JJzW1n/p4DTz2ntPXrqpbfe0qBqCfHWptR/2r/WoWRGps9T7 3WoZKA/90V2vsLL7/HFbVFFNY+DjBO3Tzi2hmaqu7a31qbqkdt1QT/zQyIGhhSOa0bz/AFBL AwQUAAAACACrdrEm7neww0IEAABkBQAAGwAAAGZtaWRvYy9pbWFnZXMvdmFyaWFibGVzLmdp ZqVT+1PUVRS/Nlb4w85UpuKmvRwei4XDXJZgIksYaMOpoWgo4jU7xeWxmEGxMtILBEwQiJDH XWDZXUEiaEEmBoLBCVNx5cKgPHsMmAPIartJMwhGM6d7vv4Jnd/O3HM+5/P4fl99LTI4RM+J N1kj94sSwgjhhAhCADtKCWWEckIFoYCPjBLGCOOECcIAZzklnBHOCReEA64KSgQjghMhiACJ RIASYAQ4AUEAEBihJZZcltOAd7AYpZxSQSngWdkxRhmnTFAGyEI+ckY5p1xQDkhKzgpGBadC UAGSo1ylwChwCoICIGXUIElLlpIWoAK8JKElllwGFIS9LM6YYAxQnzwrO84ZF4wDypUs5KPg TAgmQKqXpOQsA85AMAA0A82S29IOqR/QG5QkwSRpyRLQKjwssSW0xAJ0Dp9lL0twDmik1CfP yk4ILkD6KuVKFvKRg+AAaDOmwnAWjQZ0Hb1juIp2AIaAChkiIWnATJAHQ2CEBowIpxnewQKZ mDRS6pNnZScAQPkaQMkblERByQyUVEDxHRRnQfEOFHdA0Q+KQlA0gMISFB6gXAIFC5RpuN8T olKp1Gq1r6+vVqsNDw+Pjo5OTEw0GAy5ubnFxcVVVVVWq9Vutw8MDDgcjunp6YWFhZWVFfh/ 9cy9zZsImSHP4X+A/8TDHkA6t+zSdVmGjnls9Uu43GU9QMiuqAbdWdvF0kf2+RV6XtVHlj0d XvL75azmV8p3P1ET6Oj2ijilCbedCp5oGa33MJQnOCt9xpqC9KZJx9e+46eff7kksH/c51p+ VN7+HEegfur0G/Ws0USXfv4+duiBbXl3Mt4tiHfYtCMxaXE9waP2oo1t7O1vM1xzL8W9dXhm MKTqfM6GtmO+oHD+SXN7bVvCuQqfK6+HrnUmXkzaue/Sb7NZcyMRmo/HQk2BaRfqD8aPLd/r ujpcphq2H1R3Oyfbdna6Z03Lhu6wCaN7UL2aOdV9uNe1e/WFO9Vxx0vN7n/XeybqAoY7THV/ f7C8v+J96HX1bCzesq7ZWgcb+xZvT8W8N1PWEPrXctPUBvt8ftB4a+REkPqzuyqdJfOSdjg5 /wev4cbkYw/OJuWuTxkth0wp3o1FmWW11i9V533+KDwSsTT36bO2tOtfbK9Z0hd42nXmLGe6 h/n4tcrNsyc17ru2/Mf9+szlQR/qRit1xtg9BQG/zuz4qmHVs6V0MaJooXrSUpjyzdEUx6M1 ZYc0Z2qKHcENJ0tO6CzJ72RoWH1y+ZattSQz5GZ9dlhha+2Nxl+aK/pzHjpjzsv+MbXho3Nx qVbjYyWpdZ+41/dUP2VMv2mOCfjOaTkSecXZ7N+95mWLz06JsTZH9utLV4ec3k2TdQf82sNK 7EdbYttuhNy271hJ19T0+ft5d14w/8PKftre/uebOu8FZ+VI5Mje1rNRBpfdntTu6hh397l6 rxfv9Q/blEJe/A9QSwMEFAAAAAgAq3axJgdNZKbDAAAA/wAAACMAAABmbWlkb2MvaW1hZ2Vz L3llbGxvdy1iYWxsLXNtYWxsLmdpZnP3dLOwTGRjYGP4ysCwf//+Q8eqz5xzOndO4OS55GvX GH7/Z3j+nuHwMYZjx6qPn2P4/1/g/x/D3/85Hj9nmD6H4f+rnv9Pln7/L7B3L0P/FIamFoHf /xXPX+NoaWEIj2K4dY9h/VaGuGSGI8f2bt/LsHwtQ1Qkg5cXQ3GxQGExg6sXw5w5XufOJWdn M3T2GP7/z0A5UPzJwgiidUAEyE8MbDIODQUcCkt4OBK5lGaGMEidZZvMGKChHOSpcSqnmcEa AFBLAwQUAAAACACrdrEmJ5gh3MwCAACdAwAAHQAAAGZtaWRvYy9pbWFnZXMveWVsbG93LWJh bGwuZ2lm1VPfS1NxHD2muRus9g0KVqxypuGkiCtRSVG50uz6YM60Wr+907IVsxaR7UGpVUo3 S1iWdrGhSC9Jmo1FMZQVN9Tt9kvWD5goCqsg7cEoM/g0/4zO0zkcOBwOnJ278jZmH9NCi18A EQEIBoN+v7+iomJqaupub1tvb28sFnM4HHG/ra1t2eIDsiyXlpYCXVt1LpvNdgC4UGW3Fu/N NF7nV/Q4zntBuH+tDFSTnibEQ03pjqIiqaCou737Q3h02tf31u/zDfQPnNm2vjIzZej90Gmb eEi3oPWenLPKlJWobbhcG41GHx7eEFRcIdWsqqxfFSMRzBBik3ihQFFcr9R4MGtqaqK//Axx YzHckUHfJBrv+EUsEMBND0Kh0CU3myHj6wjndmOvFZ+H8ciHwyJeKgF/AA86Yd0PQYDTyc44 kStAlgVVFe121Eu84pJUs11lgiq6IjATLARhEmYF2YrLoyKbmJV4F3GWGMwyTCRFqGOSmCWA bA9MbsaTUYxwZjdMVuiHYfaBF6FXAvF2fHzpTvBWGAXonYx3zhImC6Iquu0wSrxwSlQ5TE5M TKwtLKQP+Dnc8+6J9LkjhWp2/6k9S7kIHzz4ODXVa1lDfYMt4azU66n2hpUl9rWw4Uird9vI mxoaAsKuExO229/zLj1ZsqSqOSv0rOR91BvVQC21nx35TRd/Dow4j1PweWVlJb+OGEccYhxk 6QF1KKRnlIjAXHgWMXeGkXRcJAnWZPg0EJPj9ROsgMCYMwnOZAgFglwuqvifYZxOSgA0WD0r Zj8B7iJ0nCG/K0XSZ9le1hZVJd/YXH3y41NeTGkwVJ/78WV0TNEZ5ojjXxPS6tjS7XX6hRnt ebrlO29ppxt3zBtM39I0ftTcUx9O5K9qq+dcqVhkKpYNfTnjmsa0PS17GsvKRxav3ufJ/9RS 7rUYij0ZzWW5ppweTShzPodN/wBQSwECFAAKAAAAAACMk60oAAAAAAAAAAAAAAAABwAAAAAA AAAAABAA/0EAAAAAZm1pZG9jL1BLAQIUABQAAAAIAFilrSgenW5odQMAANYOAAAZAAAAAAAA AAEAIAC2gSUAAABmbWlkb2Mvb3ZlcnZpZXctdHJlZS5odG1sUEsBAhQAFAAAAAgAWKWtKE45 f/a/CAAAADsAABUAAAAAAAAAAQAgALaB0QMAAGZtaWRvYy9pbmRleC1hbGwuaHRtbFBLAQIU ABQAAAAIAFilrSh8410C8AIAAPcMAAAbAAAAAAAAAAEAIAC2gcMMAABmbWlkb2MvZGVwcmVj YXRlZC1saXN0Lmh0bWxQSwECFAAUAAAACABYpa0oA1l/zMgBAAB2AwAAHAAAAAAAAAABACAA toHsDwAAZm1pZG9jL2FsbGNsYXNzZXMtZnJhbWUuaHRtbFBLAQIUABQAAAAIAFilrSg7WPny sQEAALYCAAARAAAAAAAAAAEAIAC2ge4RAABmbWlkb2MvaW5kZXguaHRtbFBLAQIUABQAAAAI AFilrSgpQgHsbAEAAKYCAAAUAAAAAAAAAAEAIAC2gc4TAABmbWlkb2MvcGFja2FnZXMuaHRt bFBLAQIUAAoAAAAAANCkrSgAAAAAAAAAAAAAAAAMAAAAAAAAAAAAEAD/QWwVAABmbWlkb2Mv cnRmbS9QSwECFAAKAAAAAADQpK0oAAAAAAAAAAAAAAAAEAAAAAAAAAAAABAA/0GWFQAAZm1p ZG9jL3J0Zm0vZm1pL1BLAQIUABQAAAAIAFilrShXWBu9HgcAAK4mAAAdAAAAAAAAAAEAIAC2 gcQVAABmbWlkb2MvcnRmbS9mbWkvRmxvd0RhdGEuaHRtbFBLAQIUABQAAAAIAFilrSgIM7Vv 4A4AAFSGAAAmAAAAAAAAAAEAIAC2gR0dAABmbWlkb2MvcnRmbS9mbWkvRmxvd1NwZWNpZmlj YXRpb24uaHRtbFBLAQIUABQAAAAIAFilrSiXBU5DLgsAAO83AAAkAAAAAAAAAAEAIAC2gUEs AABmbWlkb2MvcnRmbS9mbWkvTWV0ZXJpbmdNYW5hZ2VyLmh0bWxQSwECFAAUAAAACABYpa0o QZB72A4JAADHMwAAIAAAAAAAAAABACAAtoGxNwAAZm1pZG9jL3J0Zm0vZm1pL1VzYWdlUmVh ZGVyLmh0bWxQSwECFAAUAAAACABYpa0okiszDeYCAAAdDQAAGwAAAAAAAAABACAAtoH9QAAA Zm1pZG9jL3NlcmlhbGl6ZWQtZm9ybS5odG1sUEsBAhQACgAAAAAAWKWtKAAAAAAAAAAAAAAA ABMAAAAAAAAAAAAgALaBHEQAAGZtaWRvYy9wYWNrYWdlLWxpc3RQSwECFAAUAAAACABYpa0o bHheTeIHAACeGQAAFAAAAAAAAAABACAAtoFNRAAAZm1pZG9jL2hlbHAtZG9jLmh0bWxQSwEC FAAUAAAACABYpa0okteUabgBAAD1BAAAFQAAAAAAAAABACAAtoFhTAAAZm1pZG9jL3N0eWxl c2hlZXQuY3NzUEsBAhQACgAAAAAAjJOtKAAAAAAAAAAAAAAAAA4AAAAAAAAAAAAQAP9BTE4A AGZtaWRvYy9pbWFnZXMvUEsBAhQAFAAAAAgAq3axJpZ9FeSDCAAAwQgAAB4AAAAAAAAAAAAg ALaBeE4AAGZtaWRvYy9pbWFnZXMvT3BlbkJvb2tJY29uLmdpZlBLAQIUABQAAAAIAKt2sSau ZdrKvAAAAP8AAAAhAAAAAAAAAAAAIAC2gTdXAABmbWlkb2MvaW1hZ2VzL2JsdWUtYmFsbC1z bWFsbC5naWZQSwECFAAUAAAACACrdrEm20rh/9kCAACdAwAAGwAAAAAAAAAAACAAtoEyWAAA Zm1pZG9jL2ltYWdlcy9ibHVlLWJhbGwuZ2lmUEsBAhQAFAAAAAgAq3axJt1YR0S9BAAA2QUA AB0AAAAAAAAAAAAgALaBRFsAAGZtaWRvYy9pbWFnZXMvY2xhc3MtaW5kZXguZ2lmUEsBAhQA FAAAAAgAq3axJsfVLlSgBQAArwYAACMAAAAAAAAAAAAgALaBPGAAAGZtaWRvYy9pbWFnZXMv Y29uc3RydWN0b3ItaW5kZXguZ2lmUEsBAhQAFAAAAAgAq3axJqM7gi0KBQAAHQYAAB4AAAAA AAAAAAAgALaBHWYAAGZtaWRvYy9pbWFnZXMvY29uc3RydWN0b3JzLmdpZlBLAQIUABQAAAAI AKt2sSZHNiOdwgAAAP8AAAAhAAAAAAAAAAAAIAC2gWNrAABmbWlkb2MvaW1hZ2VzL2N5YW4t YmFsbC1zbWFsbC5naWZQSwECFAAUAAAACACrdrEm3C5N95oCAACdAwAAGwAAAAAAAAAAACAA toFkbAAAZm1pZG9jL2ltYWdlcy9jeWFuLWJhbGwuZ2lmUEsBAhQAFAAAAAgAq3axJmq5wy+w DAAAmg0AABoAAAAAAAAAAAAgALaBN28AAGZtaWRvYy9pbWFnZXMvZG9jdW1lbnQuZ2lmUEsB AhQAFAAAAAgAq3axJkMd9aDEEQAAoBIAABgAAAAAAAAAAAAgALaBH3wAAGZtaWRvYy9pbWFn ZXMvZWRpdG9yLmdpZlBLAQIUABQAAAAIAKt2sSYQ7NX4gQQAAJ4FAAAdAAAAAAAAAAAAIAC2 gRmOAABmbWlkb2MvaW1hZ2VzL2Vycm9yLWluZGV4LmdpZlBLAQIUABQAAAAIAKt2sSb13Ran mgUAAKsGAAAhAAAAAAAAAAAAIAC2gdWSAABmbWlkb2MvaW1hZ2VzL2V4Y2VwdGlvbi1pbmRl eC5naWZQSwECFAAUAAAACACrdrEmbyVPf1QAAABmAAAAIgAAAAAAAAAAACAAtoGumAAAZm1p ZG9jL2ltYWdlcy9ncmVlbi1iYWxsLXNtYWxsLmdpZlBLAQIUABQAAAAIAKt2sSbkJzhuUAIA AHYDAAAcAAAAAAAAAAAAIAC2gUKZAABmbWlkb2MvaW1hZ2VzL2dyZWVuLWJhbGwuZ2lmUEsB AhQAFAAAAAgAq3axJvs8q4VfBQAAcAYAACEAAAAAAAAAAAAgALaBzJsAAGZtaWRvYy9pbWFn ZXMvaW50ZXJmYWNlLWluZGV4LmdpZlBLAQIUABQAAAAIAKt2sSYkUzuxZgAAAGgAAAAkAAAA AAAAAAAAIAC2gWqhAABmbWlkb2MvaW1hZ2VzL21hZ2VudGEtYmFsbC1zbWFsbC5naWZQSwEC FAAUAAAACACrdrEmu41p9lYCAACAAwAAHgAAAAAAAAAAACAAtoESogAAZm1pZG9jL2ltYWdl cy9tYWdlbnRhLWJhbGwuZ2lmUEsBAhQAFAAAAAgAq3axJgcahPofBQAANAYAAB4AAAAAAAAA AAAgALaBpKQAAGZtaWRvYy9pbWFnZXMvbWV0aG9kLWluZGV4LmdpZlBLAQIUABQAAAAIAKt2 sSZ9+WWjXAQAAHsFAAAZAAAAAAAAAAAAIAC2gf+pAABmbWlkb2MvaW1hZ2VzL21ldGhvZHMu Z2lmUEsBAhQAFAAAAAgAq3axJvgcAZYzBQAARwYAAB8AAAAAAAAAAAAgALaBkq4AAGZtaWRv Yy9pbWFnZXMvcGFja2FnZS1pbmRleC5naWZQSwECFAAUAAAACACrdrEm3F4yCngRAABXEgAA GwAAAAAAAAAAACAAtoECtAAAZm1pZG9jL2ltYWdlcy9wYXJhZ3JhcGguZ2lmUEsBAhQAFAAA AAgAq3axJuRP4VW9AAAA/wAAACAAAAAAAAAAAAAgALaBs8UAAGZtaWRvYy9pbWFnZXMvcmVk LWJhbGwtc21hbGwuZ2lmUEsBAhQAFAAAAAgAq3axJuKfVXSnAQAADwIAABoAAAAAAAAAAAAg ALaBrsYAAGZtaWRvYy9pbWFnZXMvcmVkLWJhbGwuZ2lmUEsBAhQAFAAAAAgAq3axJlo6OfYT BQAAKAYAACAAAAAAAAAAAAAgALaBjcgAAGZtaWRvYy9pbWFnZXMvdmFyaWFibGUtaW5kZXgu Z2lmUEsBAhQAFAAAAAgAq3axJu53sMNCBAAAZAUAABsAAAAAAAAAAAAgALaB3s0AAGZtaWRv Yy9pbWFnZXMvdmFyaWFibGVzLmdpZlBLAQIUABQAAAAIAKt2sSYHTWSmwwAAAP8AAAAjAAAA AAAAAAAAIAC2gVnSAABmbWlkb2MvaW1hZ2VzL3llbGxvdy1iYWxsLXNtYWxsLmdpZlBLAQIU ABQAAAAIAKt2sSYnmCHczAIAAJ0DAAAdAAAAAAAAAAAAIAC2gV3TAABmbWlkb2MvaW1hZ2Vz L3llbGxvdy1iYWxsLmdpZlBLBQYAAAAALQAtANoMAABk1gAAAAA= --------------F65A4A07804C1C69117016A9--