diff --git a/.github/workflows/pull_request.yaml b/.github/workflows/pull_request.yaml index 7815ade..3824cb9 100644 --- a/.github/workflows/pull_request.yaml +++ b/.github/workflows/pull_request.yaml @@ -75,7 +75,7 @@ jobs: - name: Report test coverage if: ${{ matrix.coverage-report == true }} - uses: rhs/pr-comment@1.0.4 + uses: enlyze/pr-comment@v1.0.5 env: GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} with: diff --git a/.gitignore b/.gitignore index e72a38f..ddb46ad 100644 --- a/.gitignore +++ b/.gitignore @@ -129,3 +129,6 @@ dmypy.json .pyre/ .vscode/ + +# Mac OS-specific storage files +.DS_Store diff --git a/README.rst b/README.rst index afcc523..2020fe8 100644 --- a/README.rst +++ b/README.rst @@ -64,3 +64,9 @@ so:* .. code-block:: console $ export PYENCHANT_LIBRARY_PATH=/opt/homebrew/lib/libenchant-2.dylib + + +Examples +-------------------------------- +You can find examples of how to use the Python SDK here: `Notebooks `_ + diff --git a/docs/api_clients/production_runs/models.rst b/docs/api_clients/production_runs/models.rst index 73188d9..03581d4 100644 --- a/docs/api_clients/production_runs/models.rst +++ b/docs/api_clients/production_runs/models.rst @@ -11,7 +11,7 @@ Models :exclude-members: model_config, model_fields :show-inheritance: -.. autoclass:: Appliance() +.. autoclass:: Machine() :members: :undoc-members: :exclude-members: model_config, model_fields diff --git a/docs/api_clients/timeseries/models.rst b/docs/api_clients/timeseries/models.rst index 19d57ee..dd03852 100644 --- a/docs/api_clients/timeseries/models.rst +++ b/docs/api_clients/timeseries/models.rst @@ -11,7 +11,7 @@ Models :exclude-members: model_config, model_fields :show-inheritance: -.. autoclass:: Appliance() +.. autoclass:: Machine() :members: :undoc-members: :exclude-members: model_config, model_fields diff --git a/docs/concepts.rst b/docs/concepts.rst index 9f9b54f..5e2c6c4 100644 --- a/docs/concepts.rst +++ b/docs/concepts.rst @@ -13,15 +13,15 @@ The concept of a *site* refers to a physical production site. Depending on its s your organization might have one or many of them. In the ENLYZE platform, each site has a name and an address, which makes it easy to identify for humans. -.. _appliance: +.. _machine: -Appliance +Machine --------- -An *appliance* refers to a machine that your organization uses to produce goods. For +A *machine* refers to a machine that your organization uses to produce goods. For example, a CNC-milling center, a blown film extrusion line or an injection molding -machine all represent an appliance in the ENLYZE platform. Just as a physical machine is -located at one production site, an appliance in the ENLYZE platform is associated with +machine all represent a machine in the ENLYZE platform. Just as a physical machine is +located at one production site, a machine in the ENLYZE platform is associated with exactly one :ref:`site `. .. _variable: @@ -29,9 +29,9 @@ exactly one :ref:`site `. Variable -------- -A *variable* represents a process measure of one :ref:`appliance ` of which -timeseries data is captured and stored in the ENLYZE platform. One appliance may have -many variables, whereas one variable is only associated with one appliance. +A *variable* represents a process measure of one :ref:`machine ` of which +timeseries data is captured and stored in the ENLYZE platform. One machine may have +many variables, whereas one variable is only associated with one machine. .. _production_order: @@ -45,7 +45,7 @@ MES and then synchronized into the ENLYZE platform. They are referenced by an id which oftentimes is a short combination of numbers and/or characters, like FA23000123. In the ENLYZE platform, a production order always encompasses the production of one -single :ref:`product ` on one single :ref:`appliance ` within one +single :ref:`product ` on one single :ref:`machine ` within one or more :ref:`production runs `. .. _production_run: @@ -54,15 +54,15 @@ Production Run -------------- A *production run* is a time frame within which a machine was producing a :ref:`product -` on an :ref:`appliance ` in order to complete a :ref:`production +` on a :ref:`machine ` in order to complete a :ref:`production order `. A production run always has a beginning and, if it's not still running, it also has an end. -Usually, the operator of the appliance uses an interface to log the time when a certain -production order has been worked on. For instance, this could be the appliance's HMI or +Usually, the operator of the machine uses an interface to log the time when a certain +production order has been worked on. For instance, this could be the machine's HMI or a tablet computer next to it. In German, this is often referred to as *Betriebsdatenerfassung* (BDE). It is common, that a production order is not completed in one go, but is interrupted -several times for very different reasons, like a breakdown of the appliance or a +several times for very different reasons, like a breakdown of the machine or a public holiday. These interruptions lead to the creation of multiple production runs for a single production order. @@ -71,9 +71,9 @@ for a single production order. Product ------- -A *product* is the output of the production process which is executed by an -:ref:`appliance `, driven by a :ref:`production order `. In -the real world, an appliance might have some additional outputs, but only the main +A *product* is the output of the production process which is executed by a +:ref:`machine `, driven by a :ref:`production order `. In +the real world, a machine might have some additional outputs, but only the main output (the product) is modeled in the ENLYZE platform. Similarly to the production order, a product is referenced by an identifier originating from a customer's system, that gets synchronized into the ENLYZE platform. diff --git a/docs/conf.py b/docs/conf.py index caac768..175b12f 100644 --- a/docs/conf.py +++ b/docs/conf.py @@ -33,7 +33,7 @@ ] templates_path = ["_templates"] -exclude_patterns = ["_build", "Thumbs.db", ".DS_Store"] +exclude_patterns = ["_build", "Thumbs.db", ".DS_Store", ".ipynb_checkpoints"] # -- Options for HTML output ------------------------------------------------- @@ -49,6 +49,9 @@ ("py:class", r".*\.T"), ("py:.*", r"httpx\..*"), ] +nitpick_ignore = [ + ("py:class", "ComputedFieldInfo"), +] autodoc_default_options = {"exclude-members": "__weakref__, __init__, __new__"} autodoc_member_order = "bysource" diff --git a/docs/examples.rst b/docs/examples.rst new file mode 100644 index 0000000..036b345 --- /dev/null +++ b/docs/examples.rst @@ -0,0 +1,7 @@ +Examples +======== +In our examples section, you can find Jupyter Notebooks and other material to help you get started: + +`Notebooks `_: + +* `Introduction to the ENLYZE Python SDK `_ diff --git a/docs/index.rst b/docs/index.rst index 08dccf7..5d9247a 100644 --- a/docs/index.rst +++ b/docs/index.rst @@ -14,6 +14,7 @@ User's Guide installation quickstart + examples concepts client models diff --git a/docs/models.rst b/docs/models.rst index 566c64c..3a2a9cb 100644 --- a/docs/models.rst +++ b/docs/models.rst @@ -8,7 +8,7 @@ Data models .. autoclass:: Site() :members: -.. autoclass:: Appliance() +.. autoclass:: Machine() :members: .. autoclass:: Variable() diff --git a/docs/spelling_wordlist.txt b/docs/spelling_wordlist.txt index 4612c8c..d97518f 100644 --- a/docs/spelling_wordlist.txt +++ b/docs/spelling_wordlist.txt @@ -21,3 +21,4 @@ url uuid virtualenv whitespace +Jupyter \ No newline at end of file diff --git a/examples/notebooks/introduction_enlyze_python_sdk.ipynb b/examples/notebooks/introduction_enlyze_python_sdk.ipynb new file mode 100644 index 0000000..51869e6 --- /dev/null +++ b/examples/notebooks/introduction_enlyze_python_sdk.ipynb @@ -0,0 +1,2947 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "075c41c6-ea51-4542-a1c2-7a403a389b27", + "metadata": {}, + "source": [ + "# Introduction to the ENLYZE Python SDK\n", + "---\n", + "This guide introduces you to the concepts of the ENLYZE platform and shows you how they are implemented in the Python SDK.\n", + "We will use the SDK to query information about different locations and machines, and ultimately query time series data of one machine to further analyze and process it. \n", + "\n", + "In the second half, we will query production runs and OEE KPIs and prepare the data to be exported to Excel or used by Power BI.\n", + "\n", + "All is based on our demo organization, which represents an imaginary company with several locations, exemplary machines and dummy time series data." + ] + }, + { + "cell_type": "markdown", + "id": "a09447e1-85e5-4942-8855-3281ba81cee9", + "metadata": {}, + "source": [ + "## Installation\n", + "\n", + "The [enlyze-python package](https://github.com/enlyze/enlyze-python) is an open-source package that can be directly installed via pip. The full documentation of the SDK can be found [here](https://enlyze-python.readthedocs.io/en/latest/) and also provides a detailed installation guide for macOS/Linux and Windows.\n", + "\n", + "In short, all you need to do is run the following command in your shell. You can install it globally or in a virtualenv.\n", + "\n", + "```bash\n", + "$ pip install enlyze\n", + "```\n", + "\n", + "Installing the enlyze package will install all the required dependencies and would be enough to interact with the platform and process the data. But we will explore some more options within this tutorial, so if you want to follow along, you need to install the following additional packages:\n", + "\n", + " # loading secrets from .env files\n", + " pip install python-dotenv\n", + "\n", + " # visualization\n", + " pip install matplotlib\n", + " pip install seaborn\n", + "\n", + " # saving data to excel\n", + " pip install openpyxl" + ] + }, + { + "cell_type": "markdown", + "id": "0479401f-b1cb-4ca4-82c5-f465147ac2d1", + "metadata": {}, + "source": [ + "## General imports" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "319ab2b0-4a7c-4d57-8d64-9de3d46f4bcb", + "metadata": {}, + "outputs": [], + "source": [ + "%load_ext dotenv\n", + "%dotenv ../.env" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "38f10cf7-c76c-4c05-818c-964e96a4ed08", + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "\n", + "from datetime import datetime\n", + "\n", + "# To plot pretty figures\n", + "%matplotlib inline\n", + "import matplotlib as mpl\n", + "import matplotlib.pyplot as plt\n", + "mpl.rc(\"axes\", labelsize=14)\n", + "mpl.rc(\"xtick\", labelsize=12)\n", + "mpl.rc(\"ytick\", labelsize=12)\n", + "mpl.rc(\"legend\",fontsize=12)\n", + "mpl.rc(\"axes\", titlesize=14)\n", + "\n", + "import pandas as pd\n", + "# To display large DataFrames\n", + "pd.options.display.max_rows = 100\n", + "\n", + "# Set the color schema\n", + "import seaborn as sns\n", + "sns.set_style(\"whitegrid\")" + ] + }, + { + "cell_type": "markdown", + "id": "ed277287-997d-4fb3-844c-26169850c140", + "metadata": {}, + "source": [ + "## Client setup" + ] + }, + { + "cell_type": "markdown", + "id": "283dd0fc-e802-4b4d-a621-c154f860302a", + "metadata": {}, + "source": [ + "The only thing you need to interact with the ENLYZE platform is an API token. Reach out to us if you haven't received one already. This token will give you access to all the data of your organization, so please keep it safe! In case you have lost your token or you think it might have been compromised please reach out to us as well.\n", + "\n", + "**Store your secrets and config variables in a special file** \n", + "You don't want to leak your token on Github or any other version control system. One way to do this is to create a `.env` file in your project root folder and store your token and other secrets in it. Make sure to add it to your `.gitignore` file, so it is kept out of version control. Here is an example:\n", + "\n", + "```\n", + "# example .env file\n", + "ENLYZE_API_TOKEN=\"XXX\"\n", + "```\n", + "\n", + "You can then use a package to load these variabels automatically. Here we are using a package called [python-dotenv](https://github.com/theskumar/python-dotenv). \n", + "It loads up all the entries in this file as environment variables so they are accessible with `os.environ.get`/`os.getenv`.\n", + "\n", + "This happend a few cells above with the commands:\n", + "\n", + "```python\n", + "%load_ext dotenv\n", + "%dotenv ../.env-demo\n", + "```\n", + "\n", + "Now, since the token is stored in an environment variable, we can use it to set up our client:" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "e7569b68-8c4b-4ec8-90ba-1854be649ab1", + "metadata": {}, + "outputs": [], + "source": [ + "from enlyze import EnlyzeClient" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "41606c72-52fd-4156-b45b-bdcd2835a67a", + "metadata": {}, + "outputs": [], + "source": [ + "enlyze = EnlyzeClient(os.getenv(\"ENLYZE_API_TOKEN\"))" + ] + }, + { + "cell_type": "markdown", + "id": "1be58d63-b9e6-4037-abd5-870e2fdac2be", + "metadata": {}, + "source": [ + "## Sites, Machines, and Variables\n", + "\n", + "These three are the core models to interact with the platform. An organization consists of one or multiple sites, which again hold one or multiple machines. Each machine has then a set of variables. A variable represents a process measure of one machine of which timeseries data is captured and stored in the ENLYZE platform. One machine can have many variables, whereas one variable is only associated with one machine.\n", + "\n", + "To get variables, we need to traverse down this tree until we get and select the variables we want, and then use these variables to query their data.\n", + "\n", + "There are different ways how this can be done, but for now, we will simply go from top to bottom." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "a2f84359-c596-4d6c-b774-cf3f974dc72b", + "metadata": {}, + "outputs": [], + "source": [ + "from enlyze.models import Site, Machine, Variable" + ] + }, + { + "cell_type": "markdown", + "id": "ca474120-c510-4f54-9345-f11808395324", + "metadata": {}, + "source": [ + "### Sites\n", + "\n", + "A list of all sites can be fetched by using the `get_sites` method:" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "e4b1d442-2521-49a8-8a67-468e8e96ea1e", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "[Site(_id=49, display_name='Platform', address='Maybachstrasse 156'),\n", + " Site(_id=32, display_name='Köln', address='Heliosstrasse 6a'),\n", + " Site(_id=35, display_name='Stuttgart', address='Mahlestrasse 67')]" + ] + }, + "execution_count": 6, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "sites = enlyze.get_sites()\n", + "sites" + ] + }, + { + "cell_type": "markdown", + "id": "bb02f9a1-3c75-46e0-b925-a55cc6827aeb", + "metadata": {}, + "source": [ + "
\n", + "Side note: data analysis in Python 🫶 pandas\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "0e44713b-9a76-45c6-abbb-813a6c781f93", + "metadata": {}, + "source": [ + "If you are new to processing data in Python, you will inevitably come across pandas and pandas DataFrames. [Pandas](https://pandas.pydata.org/docs/index.html) is a great, open-source library providing high-performance, easy-to-use data structures and data analysis tools for the [Python](https://www.python.org/) programming language.\n", + "\n", + "Due to its many advantages for data processing and widespread use in the Python ecosystem, it is already part of the enlyze package and is installed alongside as a requirement.\n", + "\n", + "Especially in [Jupyter Notebooks](https://jupyterlab.readthedocs.io/en/stable/getting_started/overview.html) it makes working with and visualizing data straightforward. You will notice that models (Site, Machines, Variables etc.) have different attributes and are often returned in a list. For a few models and properties, printing the list in its raw form is still sufficient. But we soon reach the limits, and it becomes confusing and unmanageable. For this, we have this small utility function `models_to_dataframe` that takes a list of models and returns them as a DataFrame. This makes it a lot easier to explore data." + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "80593ce2-f90c-407f-92cc-12d04d8f75f3", + "metadata": {}, + "outputs": [], + "source": [ + "def models_to_dataframe(models: list):\n", + " \"\"\"Convert a list of enlyze models to a DataFrame\"\"\"\n", + " return pd.DataFrame([m.__dict__ for m in models])\n", + " " + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "001cd893-d5a1-4c1a-a24a-b8c3468e03f8", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
_iddisplay_nameaddress
049PlatformMaybachstrasse 156
132KölnHeliosstrasse 6a
235StuttgartMahlestrasse 67
\n", + "
" + ], + "text/plain": [ + " _id display_name address\n", + "0 49 Platform Maybachstrasse 156\n", + "1 32 Köln Heliosstrasse 6a\n", + "2 35 Stuttgart Mahlestrasse 67" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "models_to_dataframe(sites)" + ] + }, + { + "cell_type": "markdown", + "id": "39500780-f10d-40b8-b1b8-4fa931c42171", + "metadata": {}, + "source": [ + "We will pick a site by selecting an element from the list." + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "95f3ba77-25d1-4dff-add5-d15f65d84416", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Site(_id=32, display_name='Köln', address='Heliosstrasse 6a')" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "site = sites[1]\n", + "site" + ] + }, + { + "cell_type": "markdown", + "id": "734921aa-f1c8-4254-9569-74f7e90e7db4", + "metadata": {}, + "source": [ + "### Machines" + ] + }, + { + "cell_type": "markdown", + "id": "1aae646d-d47b-434e-a90e-daec386a6802", + "metadata": {}, + "source": [ + "We can get a list of all machines by using the `get_machines` method. Optionally, we can pass a site to the function to filter by it." + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "4f26f753-8a64-4950-9dab-ca7f6409e81d", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
uuiddisplay_namegenesis_datesite
0fafdee59-933b-40bc-b3ce-c59c7363e358PLC Rack2023-03-21Site(_id=49, display_name='Platform', address=...
122a71d1e-8262-4c27-b5fb-66598027936aW&H Varex2022-03-18Site(_id=32, display_name='Köln', address='Hel...
2cc0d2dcb-564b-48cd-a342-71765a536058Macchi2021-01-15Site(_id=32, display_name='Köln', address='Hel...
3141e0927-62b3-4e76-8398-ad82d20f397fKiefel2020-08-26Site(_id=32, display_name='Köln', address='Hel...
4f5c0b9d4-e89e-43ad-b81c-16bd5cbe3646Reifenhäuser2021-01-21Site(_id=35, display_name='Stuttgart', address...
50eba5355-1dc2-49d5-b539-2ca191dadb76Alpine2021-01-15Site(_id=35, display_name='Stuttgart', address...
\n", + "
" + ], + "text/plain": [ + " uuid display_name genesis_date \\\n", + "0 fafdee59-933b-40bc-b3ce-c59c7363e358 PLC Rack 2023-03-21 \n", + "1 22a71d1e-8262-4c27-b5fb-66598027936a W&H Varex 2022-03-18 \n", + "2 cc0d2dcb-564b-48cd-a342-71765a536058 Macchi 2021-01-15 \n", + "3 141e0927-62b3-4e76-8398-ad82d20f397f Kiefel 2020-08-26 \n", + "4 f5c0b9d4-e89e-43ad-b81c-16bd5cbe3646 Reifenhäuser 2021-01-21 \n", + "5 0eba5355-1dc2-49d5-b539-2ca191dadb76 Alpine 2021-01-15 \n", + "\n", + " site \n", + "0 Site(_id=49, display_name='Platform', address=... \n", + "1 Site(_id=32, display_name='Köln', address='Hel... \n", + "2 Site(_id=32, display_name='Köln', address='Hel... \n", + "3 Site(_id=32, display_name='Köln', address='Hel... \n", + "4 Site(_id=35, display_name='Stuttgart', address... \n", + "5 Site(_id=35, display_name='Stuttgart', address... " + ] + }, + "execution_count": 10, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "models_to_dataframe(enlyze.get_machines())" + ] + }, + { + "cell_type": "markdown", + "id": "4c378054-420f-405f-a033-1270e948d538", + "metadata": {}, + "source": [ + "Filtering by site will only return the respective machines:" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "0957caa7-f777-4ebe-878d-dc92d799ccc0", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
uuiddisplay_namegenesis_datesite
022a71d1e-8262-4c27-b5fb-66598027936aW&H Varex2022-03-18Site(_id=32, display_name='Köln', address='Hel...
1cc0d2dcb-564b-48cd-a342-71765a536058Macchi2021-01-15Site(_id=32, display_name='Köln', address='Hel...
2141e0927-62b3-4e76-8398-ad82d20f397fKiefel2020-08-26Site(_id=32, display_name='Köln', address='Hel...
\n", + "
" + ], + "text/plain": [ + " uuid display_name genesis_date \\\n", + "0 22a71d1e-8262-4c27-b5fb-66598027936a W&H Varex 2022-03-18 \n", + "1 cc0d2dcb-564b-48cd-a342-71765a536058 Macchi 2021-01-15 \n", + "2 141e0927-62b3-4e76-8398-ad82d20f397f Kiefel 2020-08-26 \n", + "\n", + " site \n", + "0 Site(_id=32, display_name='Köln', address='Hel... \n", + "1 Site(_id=32, display_name='Köln', address='Hel... \n", + "2 Site(_id=32, display_name='Köln', address='Hel... " + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "cologne_machines = enlyze.get_machines(site=site)\n", + "models_to_dataframe(cologne_machines)" + ] + }, + { + "cell_type": "markdown", + "id": "c6a57053-edac-4fcc-b3df-795ff9441d42", + "metadata": {}, + "source": [ + "**Selecting a machine by UUID**\n", + "\n", + "A more reliable way to select a machine from the list is filtering by `UUID`. Attributes like the name may change over time, but the assigned `UUID` will stay unchanged." + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "ec91bd4f-0819-46a1-8738-c09839315bd0", + "metadata": {}, + "outputs": [], + "source": [ + "from uuid import UUID" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "d95c523e-78f9-4c68-ae0f-e1fc77cca9bb", + "metadata": {}, + "outputs": [], + "source": [ + "MACHINE_UUID = UUID(\"22a71d1e-8262-4c27-b5fb-66598027936a\")" + ] + }, + { + "cell_type": "markdown", + "id": "2daff658-0ed4-4564-8a9f-b182cbbc436a", + "metadata": {}, + "source": [ + "We can create a function, that returns the machine, if it matches the UUID:" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "c920fd8d-bc74-4c0b-ab78-d6f5651dcba1", + "metadata": {}, + "outputs": [], + "source": [ + "def get_machine_by_uuid(uuid: UUID, machines: list[Machine]):\n", + " for machine in machines:\n", + " if machine.uuid == uuid:\n", + " return machine\n", + " " + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "e0507860-468a-4923-a28b-861361355b5e", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-66598027936a'), display_name='W&H Varex', genesis_date=datetime.date(2022, 3, 18), site=Site(_id=32, display_name='Köln', address='Heliosstrasse 6a'))" + ] + }, + "execution_count": 15, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "get_machine_by_uuid(MACHINE_UUID, cologne_machines)" + ] + }, + { + "cell_type": "markdown", + "id": "4cdacd41-eb81-4cf8-b8b8-383e421cf4d5", + "metadata": {}, + "source": [ + "Or we can use Python's built-in `filter()` method, more compact but less readable:" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "1ec26d71-2a44-427b-a4fb-ed05e92a539c", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-66598027936a'), display_name='W&H Varex', genesis_date=datetime.date(2022, 3, 18), site=Site(_id=32, display_name='Köln', address='Heliosstrasse 6a'))" + ] + }, + "execution_count": 16, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "results = filter(lambda a: a.uuid == MACHINE_UUID, cologne_machines)\n", + "machine = next(results)\n", + "\n", + "machine" + ] + }, + { + "cell_type": "markdown", + "id": "09c87deb-7054-441d-893d-8b501abbb23f", + "metadata": {}, + "source": [ + "### Variables\n", + "\n", + "The last step we need to make is getting the variables for a machine. We use the `get_variables`method, which always requires the specification of a machine." + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "b45c1a3a-65e1-4ba0-8a5e-11d569ae4e9e", + "metadata": {}, + "outputs": [], + "source": [ + "variables = enlyze.get_variables(machine)" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "81293554-d4a6-4d41-a9fe-5a858dfc8f30", + "metadata": {}, + "outputs": [], + "source": [ + "variables_df = models_to_dataframe(variables)" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "8624057f-f070-46a0-b1a6-33ae30cf3745", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
uuiddisplay_nameunitdata_typemachine
003e13e36-3a5f-4d9b-babf-58da68cde23fExtruder C Schneckenumdrehung1/minVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
1063e9d4a-1687-4d7e-95e3-fdd56be1c68cExtruder A SchmelzedruckbarVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
207420ec1-76da-4288-b9ee-b14891005144Extruder C SchmelzedruckbarVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
3094853e6-702d-43ab-a612-15ea0dfc90c3Extruder I Schneckenumdrehung1/minVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
40b8af028-7bea-4b71-8719-48bb8c57505dAnzahl MessungennVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
\n", + "
" + ], + "text/plain": [ + " uuid display_name unit \\\n", + "0 03e13e36-3a5f-4d9b-babf-58da68cde23f Extruder C Schneckenumdrehung 1/min \n", + "1 063e9d4a-1687-4d7e-95e3-fdd56be1c68c Extruder A Schmelzedruck bar \n", + "2 07420ec1-76da-4288-b9ee-b14891005144 Extruder C Schmelzedruck bar \n", + "3 094853e6-702d-43ab-a612-15ea0dfc90c3 Extruder I Schneckenumdrehung 1/min \n", + "4 0b8af028-7bea-4b71-8719-48bb8c57505d Anzahl Messungen n \n", + "\n", + " data_type machine \n", + "0 VariableDataType.INTEGER Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "1 VariableDataType.INTEGER Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "2 VariableDataType.INTEGER Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "3 VariableDataType.INTEGER Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "4 VariableDataType.INTEGER Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... " + ] + }, + "execution_count": 19, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "variables_df.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "c3f9cd8a-5880-40d0-9fdf-d38fdf79065e", + "metadata": {}, + "outputs": [], + "source": [ + "variables_df = (\n", + " variables_df\n", + " .sort_values(by=\"display_name\")\n", + " .reset_index(drop=True)\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "id": "0e5a8384-0812-4a30-979d-68f013658e0c", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
uuiddisplay_nameunitdata_typemachine
084802f7e-96a5-4e81-a61d-2c9c133bc58c2-Sigma%VariableDataType.FLOATMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
1163fb0cc-0e24-4092-8873-775ea9daccadAbzugsgeschwindigkeitm/minVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
20b8af028-7bea-4b71-8719-48bb8c57505dAnzahl MessungennVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
3f0e3d6cc-d302-4b0f-a660-8217f1b5d1eaDurchsatzkg/hVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
4378f5c4f-958e-4d72-98dc-88b6ac700b4bExtruder A Durchsatzkg/hVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
\n", + "
" + ], + "text/plain": [ + " uuid display_name unit \\\n", + "0 84802f7e-96a5-4e81-a61d-2c9c133bc58c 2-Sigma % \n", + "1 163fb0cc-0e24-4092-8873-775ea9daccad Abzugsgeschwindigkeit m/min \n", + "2 0b8af028-7bea-4b71-8719-48bb8c57505d Anzahl Messungen n \n", + "3 f0e3d6cc-d302-4b0f-a660-8217f1b5d1ea Durchsatz kg/h \n", + "4 378f5c4f-958e-4d72-98dc-88b6ac700b4b Extruder A Durchsatz kg/h \n", + "\n", + " data_type machine \n", + "0 VariableDataType.FLOAT Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "1 VariableDataType.INTEGER Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "2 VariableDataType.INTEGER Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "3 VariableDataType.INTEGER Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "4 VariableDataType.INTEGER Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... " + ] + }, + "execution_count": 21, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "variables_df.head()" + ] + }, + { + "cell_type": "markdown", + "id": "8a99baeb-a034-4763-b980-9e1970989c0b", + "metadata": {}, + "source": [ + "We will select variables of the Extruder A and all melt pressures for the next steps:" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "e9c5ef33-4a87-4af4-8312-094b9f558b2f", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
uuiddisplay_nameunitdata_typemachine
4378f5c4f-958e-4d72-98dc-88b6ac700b4bExtruder A Durchsatzkg/hVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
53bbcd24b-684c-49fc-b3d4-8e77f9f13a07Extruder A FolienstärkeµmVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
6dab58100-7bf2-4a37-ba9c-7130897051a8Extruder A Förderratekg/U/minVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
7063e9d4a-1687-4d7e-95e3-fdd56be1c68cExtruder A SchmelzedruckbarVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
85c5eb4a7-a6ed-47ff-a971-722e6334801bExtruder A Schmelzetemperatur°CVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
9c1a6fde4-d5e8-4b0f-a064-ae961d2bf962Extruder A Schneckenumdrehung1/minVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
\n", + "
" + ], + "text/plain": [ + " uuid display_name \\\n", + "4 378f5c4f-958e-4d72-98dc-88b6ac700b4b Extruder A Durchsatz \n", + "5 3bbcd24b-684c-49fc-b3d4-8e77f9f13a07 Extruder A Folienstärke \n", + "6 dab58100-7bf2-4a37-ba9c-7130897051a8 Extruder A Förderrate \n", + "7 063e9d4a-1687-4d7e-95e3-fdd56be1c68c Extruder A Schmelzedruck \n", + "8 5c5eb4a7-a6ed-47ff-a971-722e6334801b Extruder A Schmelzetemperatur \n", + "9 c1a6fde4-d5e8-4b0f-a064-ae961d2bf962 Extruder A Schneckenumdrehung \n", + "\n", + " unit data_type \\\n", + "4 kg/h VariableDataType.INTEGER \n", + "5 µm VariableDataType.INTEGER \n", + "6 kg/U/min VariableDataType.INTEGER \n", + "7 bar VariableDataType.INTEGER \n", + "8 °C VariableDataType.INTEGER \n", + "9 1/min VariableDataType.INTEGER \n", + "\n", + " machine \n", + "4 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "5 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "6 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "7 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "8 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "9 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... " + ] + }, + "execution_count": 22, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "extruder_a_variables_df = (\n", + " variables_df.loc[\n", + " variables_df[\"display_name\"].str.contains(\"Extruder A\")\n", + " ]\n", + " .copy()\n", + ")\n", + "\n", + "# a filtered copy of the DataFrame\n", + "extruder_a_variables_df" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "id": "28651ac9-b781-46d3-a3c2-3beb841d3ddf", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
uuiddisplay_nameunitdata_typemachine
4378f5c4f-958e-4d72-98dc-88b6ac700b4bExtruder A Durchsatzkg/hVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
53bbcd24b-684c-49fc-b3d4-8e77f9f13a07Extruder A FolienstärkeµmVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
6dab58100-7bf2-4a37-ba9c-7130897051a8Extruder A Förderratekg/U/minVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
7063e9d4a-1687-4d7e-95e3-fdd56be1c68cExtruder A SchmelzedruckbarVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
85c5eb4a7-a6ed-47ff-a971-722e6334801bExtruder A Schmelzetemperatur°CVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
9c1a6fde4-d5e8-4b0f-a064-ae961d2bf962Extruder A Schneckenumdrehung1/minVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
10c9709564-395c-403e-be34-96bb2c41551cExtruder B Durchsatzkg/hVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
118057cb67-e711-4b43-93cc-0342ec7dae44Extruder B FolienstärkeµmVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
121a07c75f-6c00-4d21-8512-8578008928faExtruder B Förderratekg/U/minVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
13aa62c591-bd89-4423-8ea4-d973290f0dadExtruder B SchmelzedruckbarVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
14f84a0232-bd3b-4978-8d21-a10922846926Extruder B Schmelzetemperatur°CVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
158755d5bb-e3ea-4ab0-994b-2b7cca767b76Extruder B Schneckenumdrehung1/minVariableDataType.INTEGERMachine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6...
\n", + "
" + ], + "text/plain": [ + " uuid display_name \\\n", + "4 378f5c4f-958e-4d72-98dc-88b6ac700b4b Extruder A Durchsatz \n", + "5 3bbcd24b-684c-49fc-b3d4-8e77f9f13a07 Extruder A Folienstärke \n", + "6 dab58100-7bf2-4a37-ba9c-7130897051a8 Extruder A Förderrate \n", + "7 063e9d4a-1687-4d7e-95e3-fdd56be1c68c Extruder A Schmelzedruck \n", + "8 5c5eb4a7-a6ed-47ff-a971-722e6334801b Extruder A Schmelzetemperatur \n", + "9 c1a6fde4-d5e8-4b0f-a064-ae961d2bf962 Extruder A Schneckenumdrehung \n", + "10 c9709564-395c-403e-be34-96bb2c41551c Extruder B Durchsatz \n", + "11 8057cb67-e711-4b43-93cc-0342ec7dae44 Extruder B Folienstärke \n", + "12 1a07c75f-6c00-4d21-8512-8578008928fa Extruder B Förderrate \n", + "13 aa62c591-bd89-4423-8ea4-d973290f0dad Extruder B Schmelzedruck \n", + "14 f84a0232-bd3b-4978-8d21-a10922846926 Extruder B Schmelzetemperatur \n", + "15 8755d5bb-e3ea-4ab0-994b-2b7cca767b76 Extruder B Schneckenumdrehung \n", + "\n", + " unit data_type \\\n", + "4 kg/h VariableDataType.INTEGER \n", + "5 µm VariableDataType.INTEGER \n", + "6 kg/U/min VariableDataType.INTEGER \n", + "7 bar VariableDataType.INTEGER \n", + "8 °C VariableDataType.INTEGER \n", + "9 1/min VariableDataType.INTEGER \n", + "10 kg/h VariableDataType.INTEGER \n", + "11 µm VariableDataType.INTEGER \n", + "12 kg/U/min VariableDataType.INTEGER \n", + "13 bar VariableDataType.INTEGER \n", + "14 °C VariableDataType.INTEGER \n", + "15 1/min VariableDataType.INTEGER \n", + "\n", + " machine \n", + "4 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "5 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "6 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "7 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "8 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "9 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "10 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "11 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "12 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "13 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "14 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... \n", + "15 Machine(uuid=UUID('22a71d1e-8262-4c27-b5fb-6... " + ] + }, + "execution_count": 23, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "melt_pressures_df = (\n", + " variables_df.loc[\n", + " variables_df[\"display_name\"].str.contains(r\"Extruder [A-B]\")\n", + " ].copy()\n", + ")\n", + "\n", + "melt_pressures_df" + ] + }, + { + "cell_type": "markdown", + "id": "8bf00c18-7670-4ba5-9c0c-ca88273708dd", + "metadata": {}, + "source": [ + "The resulting DataFrames `extruder_a_variables_df` and `melt_pressures_df` are a copy of the filtered `variables_df` DataFrame. In this case, creating a copy is not necessary because we are not further changing the data, but it's good practice when you work with DataFrames.\n", + "\n", + "If you don't use `.copy()`, the assigned variable will only hold a view of the original DataFrame and changing it would also change the data in the original DataFrame. This can cause some unexpected behaviour and difficult to trace errors, so usually pandas will throw a `SettingWithCopyWarning` error if you do so.\n", + "\n", + "
\n", + "SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame.\n", + "
\n", + "\n", + "Check out the [pandas documentation](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.copy.html) for more details." + ] + }, + { + "cell_type": "markdown", + "id": "a3033a2d-b022-4559-8f98-cae16bf42043", + "metadata": {}, + "source": [ + "## Time series data" + ] + }, + { + "cell_type": "markdown", + "id": "03ebad8e-8e54-448f-9d6f-053487ca7d76", + "metadata": {}, + "source": [ + "Now that we have selected the variables we are interested in, we can use them to get time series data.\n", + "\n", + "Pandas built-in filtering allows to quickly select the variables we want, but `get_timeseries` expects a list of `enlyze.models.Variable`'s." + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "id": "c64809fc-d6dc-4309-9f7a-bd256dc82f67", + "metadata": {}, + "outputs": [], + "source": [ + "extruder_a_variables = []\n", + "\n", + "# iterate over all variables\n", + "for variable in variables:\n", + " if variable.uuid in extruder_a_variables_df[\"uuid\"].tolist():\n", + " # if a uuid is in the uuid-column of the DataFrame\n", + " # add the variable to the list\n", + " extruder_a_variables.append(variable)\n", + " " + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "id": "593d289f-6958-4927-85f1-adecd309fb54", + "metadata": {}, + "outputs": [], + "source": [ + "melt_pressures = []\n", + "\n", + "for variable in variables:\n", + " if variable.uuid in melt_pressures_df[\"uuid\"].tolist():\n", + " melt_pressures.append(variable)\n", + " " + ] + }, + { + "cell_type": "markdown", + "id": "8ec97ec7-8e4b-4553-ac20-1e668108eabd", + "metadata": {}, + "source": [ + "### For small periods and quick explorations - get_timeseries()" + ] + }, + { + "cell_type": "markdown", + "id": "66e8c2ce-ceff-43d0-9d98-ceb4b23f2f9e", + "metadata": {}, + "source": [ + "The `get_timeseries` expects a start and end time, ideally with a timezone specified, and a list of variables.\n", + "\n", + "It then returns a `TimeseriesData` object, from which the data can be retrieved as a pandas DataFrame or dictionary (records). By default, the column names will be set to the `variable_uuids` but if `use_display_names` is set to True, it will return the DataFrame with human-readable names.\n", + "\n", + "
\n", + "Note: Display names are not guaranteed to be unique for a machine.\n", + "
\n", + "\n", + "If two or more variables with the same display name exist, there will be multiple columns with the same name. This can lead to unexpected behavior in later steps. However, display names can be adjusted in the app at any time, thus solving this problem.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "id": "bd3d8705-511b-490d-bfcd-7af70d7bb174", + "metadata": {}, + "outputs": [], + "source": [ + "start = datetime.fromisoformat(\"2022-05-17 05:00:00+01:00\")\n", + "\n", + "end = datetime.fromisoformat(\"2022-05-28 18:00:00+01:00\")\n", + "\n", + "ts_data = enlyze.get_timeseries(start=start, end=end, variables=melt_pressures)" + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "id": "971125fe-5763-4525-bc92-7905abf7206a", + "metadata": {}, + "outputs": [], + "source": [ + "data = ts_data.to_dataframe(use_display_names=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "id": "cbee3902-5048-4bf6-9691-59e57a80910e", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Extruder A SchmelzedruckExtruder B FörderrateExtruder A DurchsatzExtruder A FolienstärkeExtruder A SchmelzetemperaturExtruder B FolienstärkeExtruder B SchneckenumdrehungExtruder B SchmelzedruckExtruder A SchneckenumdrehungExtruder B DurchsatzExtruder A FörderrateExtruder B Schmelzetemperatur
time
2022-05-17 04:00:00+00:00387.0127.050.02.0211.03.071.0502.067.090.074.0210.0
2022-05-17 04:00:10+00:00388.0127.050.02.0211.03.071.0502.067.090.074.0210.0
2022-05-17 04:00:20+00:00388.0127.050.02.0211.03.071.0502.067.090.074.0210.0
2022-05-17 04:00:30+00:00388.0127.050.02.0211.03.071.0502.067.090.074.0210.0
2022-05-17 04:00:40+00:00388.0127.050.02.0211.03.071.0500.067.090.074.0210.0
\n", + "
" + ], + "text/plain": [ + " Extruder A Schmelzedruck Extruder B Förderrate \\\n", + "time \n", + "2022-05-17 04:00:00+00:00 387.0 127.0 \n", + "2022-05-17 04:00:10+00:00 388.0 127.0 \n", + "2022-05-17 04:00:20+00:00 388.0 127.0 \n", + "2022-05-17 04:00:30+00:00 388.0 127.0 \n", + "2022-05-17 04:00:40+00:00 388.0 127.0 \n", + "\n", + " Extruder A Durchsatz Extruder A Folienstärke \\\n", + "time \n", + "2022-05-17 04:00:00+00:00 50.0 2.0 \n", + "2022-05-17 04:00:10+00:00 50.0 2.0 \n", + "2022-05-17 04:00:20+00:00 50.0 2.0 \n", + "2022-05-17 04:00:30+00:00 50.0 2.0 \n", + "2022-05-17 04:00:40+00:00 50.0 2.0 \n", + "\n", + " Extruder A Schmelzetemperatur \\\n", + "time \n", + "2022-05-17 04:00:00+00:00 211.0 \n", + "2022-05-17 04:00:10+00:00 211.0 \n", + "2022-05-17 04:00:20+00:00 211.0 \n", + "2022-05-17 04:00:30+00:00 211.0 \n", + "2022-05-17 04:00:40+00:00 211.0 \n", + "\n", + " Extruder B Folienstärke \\\n", + "time \n", + "2022-05-17 04:00:00+00:00 3.0 \n", + "2022-05-17 04:00:10+00:00 3.0 \n", + "2022-05-17 04:00:20+00:00 3.0 \n", + "2022-05-17 04:00:30+00:00 3.0 \n", + "2022-05-17 04:00:40+00:00 3.0 \n", + "\n", + " Extruder B Schneckenumdrehung \\\n", + "time \n", + "2022-05-17 04:00:00+00:00 71.0 \n", + "2022-05-17 04:00:10+00:00 71.0 \n", + "2022-05-17 04:00:20+00:00 71.0 \n", + "2022-05-17 04:00:30+00:00 71.0 \n", + "2022-05-17 04:00:40+00:00 71.0 \n", + "\n", + " Extruder B Schmelzedruck \\\n", + "time \n", + "2022-05-17 04:00:00+00:00 502.0 \n", + "2022-05-17 04:00:10+00:00 502.0 \n", + "2022-05-17 04:00:20+00:00 502.0 \n", + "2022-05-17 04:00:30+00:00 502.0 \n", + "2022-05-17 04:00:40+00:00 500.0 \n", + "\n", + " Extruder A Schneckenumdrehung \\\n", + "time \n", + "2022-05-17 04:00:00+00:00 67.0 \n", + "2022-05-17 04:00:10+00:00 67.0 \n", + "2022-05-17 04:00:20+00:00 67.0 \n", + "2022-05-17 04:00:30+00:00 67.0 \n", + "2022-05-17 04:00:40+00:00 67.0 \n", + "\n", + " Extruder B Durchsatz Extruder A Förderrate \\\n", + "time \n", + "2022-05-17 04:00:00+00:00 90.0 74.0 \n", + "2022-05-17 04:00:10+00:00 90.0 74.0 \n", + "2022-05-17 04:00:20+00:00 90.0 74.0 \n", + "2022-05-17 04:00:30+00:00 90.0 74.0 \n", + "2022-05-17 04:00:40+00:00 90.0 74.0 \n", + "\n", + " Extruder B Schmelzetemperatur \n", + "time \n", + "2022-05-17 04:00:00+00:00 210.0 \n", + "2022-05-17 04:00:10+00:00 210.0 \n", + "2022-05-17 04:00:20+00:00 210.0 \n", + "2022-05-17 04:00:30+00:00 210.0 \n", + "2022-05-17 04:00:40+00:00 210.0 " + ] + }, + "execution_count": 28, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.head()" + ] + }, + { + "cell_type": "markdown", + "id": "dfd17fd7-6f21-47dd-9753-db8702b99674", + "metadata": {}, + "source": [ + "Note that the index of the DataFrame is the time-column and the timezone is always UTC, regardless of the timezone specified in the request.\n", + "\n", + "Using the `reset_index()` method will use the index as a column and set a default index instead:" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "id": "f17ebc8c-edd0-4479-a037-5e48c655d0fd", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
timeExtruder A SchmelzedruckExtruder B FörderrateExtruder A DurchsatzExtruder A FolienstärkeExtruder A SchmelzetemperaturExtruder B FolienstärkeExtruder B SchneckenumdrehungExtruder B SchmelzedruckExtruder A SchneckenumdrehungExtruder B DurchsatzExtruder A FörderrateExtruder B Schmelzetemperatur
02022-05-17 04:00:00+00:00387.0127.050.02.0211.03.071.0502.067.090.074.0210.0
12022-05-17 04:00:10+00:00388.0127.050.02.0211.03.071.0502.067.090.074.0210.0
22022-05-17 04:00:20+00:00388.0127.050.02.0211.03.071.0502.067.090.074.0210.0
32022-05-17 04:00:30+00:00388.0127.050.02.0211.03.071.0502.067.090.074.0210.0
42022-05-17 04:00:40+00:00388.0127.050.02.0211.03.071.0500.067.090.074.0210.0
\n", + "
" + ], + "text/plain": [ + " time Extruder A Schmelzedruck Extruder B Förderrate \\\n", + "0 2022-05-17 04:00:00+00:00 387.0 127.0 \n", + "1 2022-05-17 04:00:10+00:00 388.0 127.0 \n", + "2 2022-05-17 04:00:20+00:00 388.0 127.0 \n", + "3 2022-05-17 04:00:30+00:00 388.0 127.0 \n", + "4 2022-05-17 04:00:40+00:00 388.0 127.0 \n", + "\n", + " Extruder A Durchsatz Extruder A Folienstärke \\\n", + "0 50.0 2.0 \n", + "1 50.0 2.0 \n", + "2 50.0 2.0 \n", + "3 50.0 2.0 \n", + "4 50.0 2.0 \n", + "\n", + " Extruder A Schmelzetemperatur Extruder B Folienstärke \\\n", + "0 211.0 3.0 \n", + "1 211.0 3.0 \n", + "2 211.0 3.0 \n", + "3 211.0 3.0 \n", + "4 211.0 3.0 \n", + "\n", + " Extruder B Schneckenumdrehung Extruder B Schmelzedruck \\\n", + "0 71.0 502.0 \n", + "1 71.0 502.0 \n", + "2 71.0 502.0 \n", + "3 71.0 502.0 \n", + "4 71.0 500.0 \n", + "\n", + " Extruder A Schneckenumdrehung Extruder B Durchsatz Extruder A Förderrate \\\n", + "0 67.0 90.0 74.0 \n", + "1 67.0 90.0 74.0 \n", + "2 67.0 90.0 74.0 \n", + "3 67.0 90.0 74.0 \n", + "4 67.0 90.0 74.0 \n", + "\n", + " Extruder B Schmelzetemperatur \n", + "0 210.0 \n", + "1 210.0 \n", + "2 210.0 \n", + "3 210.0 \n", + "4 210.0 " + ] + }, + "execution_count": 29, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data = data.reset_index()\n", + "data.head()" + ] + }, + { + "cell_type": "markdown", + "id": "e2ac6c18-da31-4ebc-869c-b4c2b0911d3c", + "metadata": {}, + "source": [ + "To convert the timezone back to the local timezone you can use pandas Datetime functionality:" + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "id": "879a3abc-8732-4395-8957-52d5e5c2b154", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
timetime_localExtruder A Schmelzedruck
02022-05-17 04:00:00+00:002022-05-17 06:00:00+02:00387.0
12022-05-17 04:00:10+00:002022-05-17 06:00:10+02:00388.0
22022-05-17 04:00:20+00:002022-05-17 06:00:20+02:00388.0
32022-05-17 04:00:30+00:002022-05-17 06:00:30+02:00388.0
42022-05-17 04:00:40+00:002022-05-17 06:00:40+02:00388.0
............
996992022-05-28 16:59:10+00:002022-05-28 18:59:10+02:00338.0
997002022-05-28 16:59:20+00:002022-05-28 18:59:20+02:00338.0
997012022-05-28 16:59:30+00:002022-05-28 18:59:30+02:00338.0
997022022-05-28 16:59:40+00:002022-05-28 18:59:40+02:00338.0
997032022-05-28 16:59:50+00:002022-05-28 18:59:50+02:00338.0
\n", + "

99704 rows × 3 columns

\n", + "
" + ], + "text/plain": [ + " time time_local \\\n", + "0 2022-05-17 04:00:00+00:00 2022-05-17 06:00:00+02:00 \n", + "1 2022-05-17 04:00:10+00:00 2022-05-17 06:00:10+02:00 \n", + "2 2022-05-17 04:00:20+00:00 2022-05-17 06:00:20+02:00 \n", + "3 2022-05-17 04:00:30+00:00 2022-05-17 06:00:30+02:00 \n", + "4 2022-05-17 04:00:40+00:00 2022-05-17 06:00:40+02:00 \n", + "... ... ... \n", + "99699 2022-05-28 16:59:10+00:00 2022-05-28 18:59:10+02:00 \n", + "99700 2022-05-28 16:59:20+00:00 2022-05-28 18:59:20+02:00 \n", + "99701 2022-05-28 16:59:30+00:00 2022-05-28 18:59:30+02:00 \n", + "99702 2022-05-28 16:59:40+00:00 2022-05-28 18:59:40+02:00 \n", + "99703 2022-05-28 16:59:50+00:00 2022-05-28 18:59:50+02:00 \n", + "\n", + " Extruder A Schmelzedruck \n", + "0 387.0 \n", + "1 388.0 \n", + "2 388.0 \n", + "3 388.0 \n", + "4 388.0 \n", + "... ... \n", + "99699 338.0 \n", + "99700 338.0 \n", + "99701 338.0 \n", + "99702 338.0 \n", + "99703 338.0 \n", + "\n", + "[99704 rows x 3 columns]" + ] + }, + "execution_count": 30, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data[\"time_local\"] = data[\"time\"].dt.tz_convert(\"Europe/Berlin\")\n", + "data[[\"time\",\"time_local\", \"Extruder A Schmelzedruck\"]]" + ] + }, + { + "cell_type": "markdown", + "id": "0239a449-efc0-4cae-a233-5987c6a45968", + "metadata": {}, + "source": [ + "### For large periods and more fine-grained control - get_timeseries_with_resampling()" + ] + }, + { + "cell_type": "markdown", + "id": "88bd9db6-6561-480f-bc96-ccddbe68cbcd", + "metadata": {}, + "source": [ + "The `get_timeseries` method returns the time series data in the currently highest resolution (10 seconds). For a long period and many variables, this quickly results in a large amount of data. Due to the size of the data, there might be a noticeable delay in receiving the results, and in some cases, it may even cause a request timeout.\n", + "\n", + "The `get_timeseries_with_resampling` method allows for a higher control over the way in which the data is retrieved. By specifying a resampling method and a resampling interval, the amount of data can be adjusted for the use case. Resampling happens on the server-side and thus reduces the payload.\n", + "\n", + "\n", + "A separate resampling method must be specified for each variable. This is necessary because not every resampling method is available for every data type. For example, categorical data types such as strings can only be aggregated with `first`, `last` or `count`. You can find all available resampling methods [in our documentation](https://enlyze-python.readthedocs.io/en/latest/models.html#enlyze.models.ResamplingMethod)." + ] + }, + { + "cell_type": "code", + "execution_count": 31, + "id": "9c004353-b0e8-41e2-9fe5-a30ee22a3010", + "metadata": {}, + "outputs": [], + "source": [ + "from enlyze.models import ResamplingMethod" + ] + }, + { + "cell_type": "code", + "execution_count": 32, + "id": "df79349c-f999-4754-8174-3433f3599c9e", + "metadata": {}, + "outputs": [], + "source": [ + "# pandas has some very handy Timestamp and Timedelta functionalities that are worth checking out\n", + "end = pd.Timestamp.utcnow()\n", + "start = end - pd.Timedelta(weeks=2)\n" + ] + }, + { + "cell_type": "markdown", + "id": "81aa8b85-a83c-4518-b94e-a83022211436", + "metadata": {}, + "source": [ + "
\n", + "Tip: You can access the Docstring of a function in Jupyter Notebooks, if you call it with a question mark right after the name, for example: `enlyze.get_timeseries_with_resampling?`\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "a92d1975-8cc2-4ade-8ab7-330c36598fc5", + "metadata": {}, + "source": [ + "For our example, we will use the same resampling method for all variables and a `resampling_interval` of 10 minutes" + ] + }, + { + "cell_type": "code", + "execution_count": 33, + "id": "783d0c11-b5e2-4e7b-8cbb-fcb9f777012f", + "metadata": {}, + "outputs": [], + "source": [ + "ts_data_resampled = enlyze.get_timeseries_with_resampling(\n", + " start=start,\n", + " end=end,\n", + " variables={v: ResamplingMethod(\"avg\") for v in melt_pressures},\n", + " resampling_interval=600\n", + ")\n", + "\n", + "data_resampled = ts_data_resampled.to_dataframe(use_display_names=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 34, + "id": "ca1b68ac-d0c0-45b7-aa4f-918c401ab171", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Extruder A SchmelzedruckExtruder B FörderrateExtruder A DurchsatzExtruder A FolienstärkeExtruder A SchmelzetemperaturExtruder B FolienstärkeExtruder B SchneckenumdrehungExtruder B SchmelzedruckExtruder A SchneckenumdrehungExtruder B DurchsatzExtruder A FörderrateExtruder B Schmelzetemperatur
time
2023-11-22 15:40:00+00:00-10.8333330.00.00.018.00.00.0-6.9444440.00.00.018.0
2023-11-22 15:50:00+00:00-10.8833330.00.00.018.00.00.0-7.0166670.00.00.018.0
2023-11-22 16:00:00+00:00-10.9333330.00.00.018.00.00.0-7.0000000.00.00.018.0
2023-11-22 16:10:00+00:00-10.9500000.00.00.018.00.00.0-7.0333330.00.00.018.0
2023-11-22 16:20:00+00:00-10.9333330.00.00.018.00.00.0-7.0000000.00.00.018.0
\n", + "
" + ], + "text/plain": [ + " Extruder A Schmelzedruck Extruder B Förderrate \\\n", + "time \n", + "2023-11-22 15:40:00+00:00 -10.833333 0.0 \n", + "2023-11-22 15:50:00+00:00 -10.883333 0.0 \n", + "2023-11-22 16:00:00+00:00 -10.933333 0.0 \n", + "2023-11-22 16:10:00+00:00 -10.950000 0.0 \n", + "2023-11-22 16:20:00+00:00 -10.933333 0.0 \n", + "\n", + " Extruder A Durchsatz Extruder A Folienstärke \\\n", + "time \n", + "2023-11-22 15:40:00+00:00 0.0 0.0 \n", + "2023-11-22 15:50:00+00:00 0.0 0.0 \n", + "2023-11-22 16:00:00+00:00 0.0 0.0 \n", + "2023-11-22 16:10:00+00:00 0.0 0.0 \n", + "2023-11-22 16:20:00+00:00 0.0 0.0 \n", + "\n", + " Extruder A Schmelzetemperatur \\\n", + "time \n", + "2023-11-22 15:40:00+00:00 18.0 \n", + "2023-11-22 15:50:00+00:00 18.0 \n", + "2023-11-22 16:00:00+00:00 18.0 \n", + "2023-11-22 16:10:00+00:00 18.0 \n", + "2023-11-22 16:20:00+00:00 18.0 \n", + "\n", + " Extruder B Folienstärke \\\n", + "time \n", + "2023-11-22 15:40:00+00:00 0.0 \n", + "2023-11-22 15:50:00+00:00 0.0 \n", + "2023-11-22 16:00:00+00:00 0.0 \n", + "2023-11-22 16:10:00+00:00 0.0 \n", + "2023-11-22 16:20:00+00:00 0.0 \n", + "\n", + " Extruder B Schneckenumdrehung \\\n", + "time \n", + "2023-11-22 15:40:00+00:00 0.0 \n", + "2023-11-22 15:50:00+00:00 0.0 \n", + "2023-11-22 16:00:00+00:00 0.0 \n", + "2023-11-22 16:10:00+00:00 0.0 \n", + "2023-11-22 16:20:00+00:00 0.0 \n", + "\n", + " Extruder B Schmelzedruck \\\n", + "time \n", + "2023-11-22 15:40:00+00:00 -6.944444 \n", + "2023-11-22 15:50:00+00:00 -7.016667 \n", + "2023-11-22 16:00:00+00:00 -7.000000 \n", + "2023-11-22 16:10:00+00:00 -7.033333 \n", + "2023-11-22 16:20:00+00:00 -7.000000 \n", + "\n", + " Extruder A Schneckenumdrehung \\\n", + "time \n", + "2023-11-22 15:40:00+00:00 0.0 \n", + "2023-11-22 15:50:00+00:00 0.0 \n", + "2023-11-22 16:00:00+00:00 0.0 \n", + "2023-11-22 16:10:00+00:00 0.0 \n", + "2023-11-22 16:20:00+00:00 0.0 \n", + "\n", + " Extruder B Durchsatz Extruder A Förderrate \\\n", + "time \n", + "2023-11-22 15:40:00+00:00 0.0 0.0 \n", + "2023-11-22 15:50:00+00:00 0.0 0.0 \n", + "2023-11-22 16:00:00+00:00 0.0 0.0 \n", + "2023-11-22 16:10:00+00:00 0.0 0.0 \n", + "2023-11-22 16:20:00+00:00 0.0 0.0 \n", + "\n", + " Extruder B Schmelzetemperatur \n", + "time \n", + "2023-11-22 15:40:00+00:00 18.0 \n", + "2023-11-22 15:50:00+00:00 18.0 \n", + "2023-11-22 16:00:00+00:00 18.0 \n", + "2023-11-22 16:10:00+00:00 18.0 \n", + "2023-11-22 16:20:00+00:00 18.0 " + ] + }, + "execution_count": 34, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data_resampled.head()" + ] + }, + { + "cell_type": "markdown", + "id": "609024e5-d407-49f8-9f28-0aa81463ca7f", + "metadata": {}, + "source": [ + "### Data preparation and exploration\n", + "\n", + "Data preparation and exploration is a bigger topic that will be covered in later tutorials, but here are a few examples of how to get started. " + ] + }, + { + "cell_type": "code", + "execution_count": 35, + "id": "d05a9b60-e954-416e-9258-d96084281d6f", + "metadata": {}, + "outputs": [], + "source": [ + "# quick step to sort the columns alphabetically\n", + "data_resampled = data_resampled[sorted(data_resampled.columns)]" + ] + }, + { + "cell_type": "code", + "execution_count": 36, + "id": "8f7fc1a4-b41e-48b9-b402-18d429f2c550", + "metadata": {}, + "outputs": [], + "source": [ + "# reset index and convert time to local timezone\n", + "data_resampled = data_resampled.reset_index()\n", + "data_resampled[\"time\"] = data_resampled[\"time\"].dt.tz_convert(\"Europe/Berlin\")" + ] + }, + { + "cell_type": "markdown", + "id": "f9223dfd-bbf6-4057-a4c8-791adac84e6a", + "metadata": {}, + "source": [ + "#### Simple plot" + ] + }, + { + "cell_type": "code", + "execution_count": 37, + "id": "b678e183-e184-426f-bfd0-447fab8cc841", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAABkcAAAHACAYAAAAYx7+IAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd3hUZfbA8e+dXlIhCb2X0LsUQVFkFXtf2wJioeuqKODqoriLChZEAlhZy7pIXbHvT0UQFbGCIKD0TkivU++9vz+mkDIJCYQwgfN5Hh/JzJ0779wyc+973vccRdd1HSGEEEIIIYQQQgghhBBCiLOE4XQ3QAghhBBCCCGEEEIIIYQQojZJcEQIIYQQQgghhBBCCCGEEGcVCY4IIYQQQgghhBBCCCGEEOKsIsERIYQQQgghhBBCCCGEEEKcVSQ4IoQQQgghhBBCCCGEEEKIs4oER4QQQgghhBBCCCGEEEIIcVaR4IgQQgghhBBCCCGEEEIIIc4qptPdgBOlaRp+vx+DwYCiKKe7OUIIIYQQQgghhBBCCCGEOI10XUfTNEwmEwZD5XND6mxwxO/3s2nTptPdDCGEEEIIIYQQQgghhBBCRJGuXbtisVgqXabOBkdCUZ+uXbtiNBpPc2tEiKqqbNq0SfaLEBWQc0SIysk5IsTxyXkiROXkHBGicnKOCFE5OUeEOL5oPk9CbTverBGow8GRUCoto9EYdTtAyH4R4njkHBGicnKOCHF8cp4IUTk5R4SonJwjQlROzhEhji+az5OqlOKQguxCCCGEEEIIIYQQQgghhDirSHBECCGEEEIIIYQQQgghhBBnFQmOCCGEEEIIIYQQQgghhBDirCLBESGEEEIIIYQQQgghhBBCnFUkOCKEEEIIIYQQQgghhBBCiLOK6XQ3oLaoqorP5zvdzTjjqaoKgNvtxmg0nubWCBF9TuYcMZvNcl4JIYQQQgghhBBCCFEDzvjgiK7rHDlyhNzc3NPdlLOCruuYTCb27t2LoiinuzlCRJ2TPUcSEhJo2LChnF9CCCGEEEIIIYQQQpyEMz44EgqMpKSk4HA4pEPxFNN1HZfLhd1ul20tRAQneo7ouk5xcTFHjx4FoFGjRqeqiUIIIYQQQgghhBBCnPHO6OCIqqrhwEj9+vVPd3POCrquo2kaNptNgiNCRHAy54jdbgfg6NGjpKSkSIotIYQQQgghhBBCCCFO0BldkD1UY8ThcJzmlgghRM0IfZ9JDSUhhBBCCCGEEEIIIU7cGR0cCZEZDEKIM4V8nwkhhBBCCCGEEEIIcfLOiuCIEEIIIYQQQggRiU/T2ZBfjKrrp7spQgghhBCiFklwRITpcjNwVpH9LYQQQgghBMzafZhhP/1B2t6jp7spQgghhBCiFklwpI6aOnUqqampFf43cODAKq8rPz+fyZMn8+OPP56y9q5YsYLU1FQOHDhwyt4DYNKkSaSmprJw4cIqLT9kyJBS261jx4706dOHW265hffee++UtjVSW6ZOnVor7zV//nxef/31WnkvIYQQQgghotncfYGgyFO7D5/mlgghhBBCiNpkOt0NECcuOTmZtLS0iM+ZzeYqr2fr1q2sXLmS66+/vqaadloUFBTw+eef0759exYvXsyoUaOqVJ9h8ODBjB8/HgC/309OTg6ffPIJU6ZMYevWrTz88MOnuum1bs6cOUycOPF0N0MIIYQQQoio4lI1QvOr97o8LEvPocCvsrGgGE2HS5Pj2ZBfzNqcQi6qH8urnVtKTTghhBBCiDpKgiN1mMVioUePHqe7GVHjww8/BOCRRx5h5MiRfPfddwwYMOC4r6tXr1657finP/2J5ORk3njjDS6++GJ69+59KposhBBCCCGEqAW/F7lZfDibQYkxFKkaGwqK2ZBfTIGqllqu1Ve/VrqeTYWu8L8/zMhjj8tLK4f1lLRZCFFzXKqGAtiMkkBFCHFqfJ1TwBsHM5nRrikNrFUftC5Or7MyOKLrOi6fevwFa4HdbDylI402b97MTTfdxJVXXsnTTz8NQFZWFldccQUdOnRgzJgxjBw5EoARI0bQt29f3n77bYYPH06DBg3wer189dVX9OzZk7FjxzJixAjeeust+vXrF36P4cOHA/D2228DoGkaCxYsYMmSJeTk5DBw4EDOOeeccm37448/eO655/jhhx8AGDBgAFOnTqVZs2YArF+/nhEjRjB9+nRefvll8vLymDt3boUpw5YvX86AAQPo378/LVq04N13361ScKQiEydOZNGiRbz77rv07t2bAwcOcNFFF/HUU09x3XXXhZebOnUq33//PatWrQpvj7Lb7l//+heFhYXMnj2b//u//6OgoIA2bdpwzz33cMEFF4TX5fP5mDVrFitXrqSoqIhevXrx2GOP0aJFCwCys7OZMWMG3333Hfn5+bRu3ZpRo0ZxzTXXhNfxww8/8NJLL/Hrr7/icrlISUnh2muvZcKECRgMBlJTUwFIS0sjLS2N33//nSFDhnDw4MGI2+GLL76gadOmJ7wdhRBCCCGEqEmarmMI3kMV+lWKVI1fC4rJ96vsKPbwa0ExvxcaaLNpN1uL3MSbjOx3e3FrOvP3V72uiNWgMCA+hk4xdmJMBuqZTXyakUei2ch/j+YCsMvlkeBIHbG5oJjdLi/9E5wkW6TTqi474vFx2OOjuc2CS9PQdJ2tRW78uo7VYGC/24uq62R5/SRbTHyTW8jHGXm0sltZ268DRpntJYQ4BW7YsBMAk6LwUueWp7cxosrOuuCIruvc8NI6ftqbc7qbAkCfFoksHTvghAMkfr8/4uNGYyDo0qVLF+6++24WLFjA1VdfzYABA5g2bRqapvH000/jdDqZNm0aTzzxBNOmTSsV9Pjkk0+46qqrWLBgAZqmVblNc+bMYdGiRYwbN47u3bvzySef8Nxzz5VaZvfu3dx88820bt2amTNn4vf7WbBgAbfccgsrV66kfv364WXT0tJ49NFHcbvd9OzZM+J7bt++nU2bNjFnzhwArrnmGubPn09mZiZJSUlVbntJsbGxdOvWjZ9++qnary277VRV5Y477mDPnj3ce++9tG7dmv/+979MmDCBN998kz59+gDw8ccfM2jQIJ5++mkyMzN56qmnuP/++1mxYgUADz30EFlZWUyfPp2YmBhWrlzJlClTaNiwIf3792fbtm3cfvvtDBs2jNmzZ6PrOh988AFpaWm0bt2ayy+/nMWLF3PTTTdxww03cOONNwKBbez1esPtz8zMZNKkSfTp04dGjRqd0ParSLbPT7rHR0u7FbuM2hFCCFEH6LqODuHOWCHOFEWqyqPbD3JRvTiuSEk43c2JaK/Lw+rsAgYmxvBrgYt/7DzEEY+Pdg4bXl1jj8tbwSsVDuQUAnDUe+yeqYnVTD2ziZ5xDva7vXyZXRB+blqbxtzcqB7ZPj9GFFraLeXu025vEri38Oq7+Sgjj+/zivgkI4++CU7+3LBezX54USO2FroYvmkXB9y+8GOjmyYzvW3jOp8STdcDSeAURaHAr6LpOvHmY908mq6jBJ8v8qvYjAYUwKtpePTAd8BBl5d8v0ojq5lMn58Yo5E2DisH3F4yvX6a2Cxowd/BdK8PTQen0YDNYCBfVXn/aC6bCopJsZixGw20dVgxKgr5fhWrwUCS2YRL01B1nUK/RpbPT7rXR1ObBQUoDM7oyPL52VrowqAotLBZyPb5OeTx0T3WwWGPjz+K3cHPDK5q9E+UtMvl4bAn8N5CCHGq7Cr2nO4miGo464IjAHX78ueYgwcP0rlz54jPTZ48mTvvvBOACRMmsGrVKqZPn87o0aP5/PPPmTNnDg0aNACgbdu24f+H/g2BuiXTp0/HYglcOKxfv/64bcrPz2fRokWMGjUqXNPivPPO4+jRo6xduza8XFpaGna7nTfeeIOYmBggMHNk6NChvPbaa0yZMiW87K233sqwYcMqfd/ly5eTkJDAkCFDALj22muZO3cuy5YtY+zYscdtd0WSkpL49dfKp9ZHUnbbffnll2zcuJF58+YxdOhQAPr378/+/fv57rvvwsGRBg0aMH/+/HDNmL1797JgwQIKCwuJiYnh+++/Z8KECeF19O3bl4SEhPD7bNu2jXPPPZdnnnkGgyEQeBg4cCCrVq1i/fr1XH755eEUYg0bNgz/u1OnTuG2e71e/vKXv5CcnMzzzz+P0Wis9uevzP7gDex+t5f2TluNrlsIIYSoaS5V46IffifZYuK/PdtKgEScUV7en8Giw9ksOpzNweTup3w0dahz1qgo+DWdXS4PxapGvl/lu7xC9rq8JJiMeHWdVVn5uDSN7Apm/Ic6SkMaW82YFIXGVjM9Y+0YM4+yNzYRh9HIuQkxxJuM/CkprtRn/PehrHBwxKIojGuWjKIo1DMf/za5tT0wW2TO3nQAVufkS3CkFrjVQCe7X9exGw14NZ1sn58dxR72uDyYFAWfrpNoNlGsarg1jUe3l58h/8qBDP57NIeBCTF0i3WQ5fNjBOJMRmxGA0c8PpxGAwYUDAo0t1vYXOAi3mQkz6/i0XTiTUasBoXDHh+NbRYyvT7cms5+txdj8DCLNxlJNJtQgINuH4WqShuHlX0uL3tcXswGsCgGOsXYyPGpxJgMOI1Gthe5Oejx0tZho8CvohOYyRRnMnLI42NroZsEs5F8v4oCNLFZ2FXswafr2AwKfl3HHyyeY1SgntlEhrfswEojfLMl4nZWIFx753T4vejY+X3Ik1fu+dBZrAM2g4JX02lsM9PYGgiqNLdZ0AFzMEiTYjXzfnC21x6XR4IjQohTSm4X6pazLjiiKApLxw44I9JqJScns2DBgojPlRztbzabmTlzJjfeeCOPPPII11577XGDDQCtW7cOd7pX1caNG/H7/Vx44YWlHr/00ktLBUe+++47+vbti81mC89+iYmJoU+fPnz77belXtuxY8dK39Pn8/H+++8zdOhQ3G43brcbp9NJ7969WbJkCaNHjw4HCqpL1/UT2j9lt91PP/2E2WwOB28ADAYD7777bqnXdevWLRwYAcLprPLz84mJiaFfv37MnTuXLVu2cN555zF48OBSgaRrrrmGa665Bo/Hw+7du9m7dy9bt25FVVV8Ph9V8cgjj7B9+3beffdd4uPjq/3Zq8p9gqN9hBBCiNq0tdDFLpeHXS4P63ILGZgYe7qbJMRxPbP7MG8czOKR1o24tXH9Cpf7Iis//O/1uUWcmxhTpfUX+lU2FbroEmMn1lR6II2q66zLLeTHvCJiTEbSPT52FHvI8flZn1eESVFo67Cy3+2lUK3e9aBFUbi/ZQOua5DIH0VurAYDnWPsmBSINRnDwUtVVdmQn06Pjs0rHejTvEQHaZLFVK3r/tZlUmmVnJVwJqvu/ZFf0ylUVfL8Kkc8PhLNJtblFpLt83PE48Ov67R22Dji8ZLu9XPU4yPebMSIQobXTz2LEbeqh1Mn/VJQjHoSvfZxJgND6sXx3tFcMrx+3juay3vBTvPaUnK2UsgvBcURl91d4cwocHmOnT8lgwlurfQGUnUiBEaOMSsKKRYTR71+TMF969K0wHllNJKvBgIwfh3qmY3YDQYKg0Eqo6LQP8HJkHpxpHt9qDr8WlBMhtdPjNFAE5uFTK8fZzAoGkpPl2AyssflQSEQ8PHrOs1sFupbTNgNCpk+Pw6DgYZWM7/kF1PfYuKceCfmYPvqmU1YDQpuTSfOZCyVaq8iBf6dfJldwB6Xl0GJlS4qhDiDaLqOR9NPeeYSVT/23Ws4Y4blnx3OuuAIBAIkDkvd/+gWi4WuXbtWadmOHTuSmprK5s2bywUuKuJ0OqvdptzcXAASEhJKPZ6cnFxuuY8//piPP/643Drq1Ss94srhcFT6nqtXryYrK4tly5axbNmycs+vXbuWwYMHV6H15aWnp9OwYcNqv67stsvNzSUhIeG4QZqynzW0fCit2ezZs3nppZf45JNP+N///ofBYODcc8/liSeeoEmTJrjdbv7xj3+wcuVK/H4/TZs2pWfPnphMpvCU68q88sorvP/++8yZMydcm+RUqUJzhBBCiNNup+vYtPglR3IkOCLqhGVHcsjy+Xng9/0MrR9HSoSioB9n5PJT/rEO2f8ezWFAgrNcx7em66h6IA3PPreXYlXj7t/2cMjjo1+8k8H1Yll6JBsFhcMeLzrlO2dL8gVrAwA4jAZ0HcwG6BcfQ1ObBYuiYFQUesc7MCsKvxe5uTIlgQ35xXSLdYSDEi3sJ1/no1mJ4Eh1R5KfnxhLY6uZQ55AUCTlDLi/LMutaux0eWhgMbOtyMUXWfm8eSiLbjF2HEYDBX4Nl6axz+2ho9POsKR4NODNg5noBNLpFlUzAHYiLIpCK4eVFjZLuOZEjs9PjMmISYE/igIBboAUi5lLkuLDAZHusXYaWMw0C9aucKkaHk0nxWqmWFXJ8PrJ8PrZ6/LQxmGjmc1CnMmASVHY5/YSZzLS2GrmoMdHrNFAvMlEM7sFPRg4OOzx4dUCM13qm00YFcj2qTSxmWloNZPp9WM3Gthd7MFhNJDl82NWFNo4rCRZzGwtdFHfbCLOZKRAVcn0+rEZDHSMseE0GokJvsalajS3W4gzGXFrOiYlkPPegEKhqnLY46O904amg0EBNI0tmzbRvVs3LCYjVoMBLXiDpgPpwSCVs0Rw0a/pGBVqPRVZZb+7luDtdVVmdba0W4EC9rg8rM0u4PdiN8OS4mUWiRBnuOt+2cEul4ePe7c/ped7ZokgtHR31S1n3hWciGjx4sVs3ryZDh06MGPGDAYMGEBcXFy11hG6CCpbf6SoqCgcDEhMDAzByMrKok2bNuFlQkGTkNjYWM4991xGjRpV7n1MpuodlsuXL6dZs2bMmDGj1OO6rjNx4kTefffdEwqO5OXl8dtvv3H11VcDxz6/qpaedVRcHHmUT0mxsbHk5uaWG2m1ZcsWdF2vMD1apPU89NBDPPTQQ+zatYsvvviC+fPnM336dF555RVmzJjB//73P1544QXOPffccLClKoXpV61axezZsxkzZkyVZhadiLIBmtBoIyGEECJa7SiRM3h5ejZ/aVyfzsGOwUiyvH42FRbzc34xmg5+XaeNw4pf10n3+FifVwQERqk/m9oM6wnObhWnj67rFKoaMUZDII+/qpLtU1F1nf0uLx5dJ8PrI9ZoZEBCDPVrudNc13XSvcdmMQzftItbG9VnWFI8DYJBEl3XeWb3EQBSnTZ+L3Lz9qEsPsrI5cJ6cTSzWUixmNhS6ObTzDzygql7vGWu5dbnFYWP6ZISTUYG14tF1cFmVOgaY0cHzkuMZY/Lg6oHggnnxDvDYysr6nC9OCkwk7llDQRDympiOxY0GlHJDJvIr7Xw44BOrM8r4tpfduCPkpE/uq6zvdhDm2Ddh9BjkbZvhtfH70XucHHrdK8vnIqp0K/yVU5BqXotId9F2OcVHQslxRgN+HSdrjF2mtutNLKaUYBtRW7aOqw0sQZmDux1ebAYDDSxmsn3q9iDsw7y/CrnJ8bQ0GoOd/o7DAasBgMmQ8X3FD5Np9majUBglsTF9ePoHecg1Wnj+Q7NK23z6Ta0/vHv2dsd5/n6mMoFE1VVxa4EApTG4O9QyQBD4wgdiJVt47qgpT3wmdL2HSVt31EAnt19hNV9O9AwQgBZCFH35fj84d+se7buZV7HFqRYzKfk+6zktVe2r+LZeiL6SHDkLHDw4EFmzpzJDTfcwMSJE7nsssuYMWMGM2fOBKhyTYlQbZAjR46EH8vLy2Pnzp1069YNgJ49e2Kz2fj000/p27dveLkvv/yy1Lr69u3Ljh076NixYzgYous6Dz74IC1atDhuKq2QjIwM1q5dy1133VWqmHzIsGHDWLFiBenp6eEaK1X10ksv4fP5uOmmm4Bjnz89PT28jM/n49dffz3ujJA+ffqwcOFCvvrqq3CgRtd1Hn74YVq0aMGLL7543PYcPHiQW2+9lYcffphhw4bRunVrWrduzYYNG9i7dy8QSN/Vr1+/cE0SgM2bN5OdnV0qqFW2vX/88QcPPvgggwYN4r777jtuW05U2ZtGt6rhNNVsTRMhhBCiJu0oUdfAr8OVP28n3mTkvhYN6BJjBwJ5hTO9fpal5/B1TkGlo+ZLurh+PFdGaRHsuk7XdVyazvZiN7uKPTS3WYgxGcnz+SlQNXa7PMSbjDiMBg57fBSrGrHB5w8G/y71nxYYeQ3gCqb4Cc0UiNR5XFLXGDsxJgOFfo0rkhP4a8vqXZNWV6bPX+oY3FjgYmPBAab8cYA4k4EesQ5a2a1sLXJjNSgs6d6GQeu3UqAG6nssT8+pcN1mRSHeZAwHCD/JzMNuUHigZUO6xNhpaDVTrGp0i3VgrqDjoVPwvIkGFoOB5zs0I8Pj5/oG1c+zY1AUkoL1Sap42p9Smq4zces+VqTn0NRmpleck6MeHz/lF5NsMQUCEDYLWwpduFSdXS53OBhyPAoQazJwe+MkUqxmnEYDccHr+ESTiW1FLt4/motH07kkKY6+8TE0sAZmPJgUBavBUGFQ+URVNUVKyWNRA5wmIx/1bl+jbRHRr1ts+YwUuX6VjzJyubNpcoRXCCE0XSfbp5JoNnLE48On6zgMBr7KKaC+2YRKIADxU34xvxYU0zvOwRXJCZgUBVXXqW8xhWe0NbWaa3XmmU/TeWHPsf67dblF9Fq3hRijgd5xTlKsJhpazOT5VXYWe0h12ogzGcn1qzS0BOo0xZqMNLNZ+KPITZGq0cZh5bDHx2+FLto6rHg0nbU5BeT5VZJKDIbJrCA4UqxqeDWNhCrUNhO1R/ZGHeb1etmwYUOFz6empmKz2XjkkUew2+1MnjyZ+Ph47rvvPp588kkuueQShgwZQmxsYJrq6tWriY+Pp0OHDhWur1GjRsybN4+YmBgUReHll1/Gbj92g+N0OrnrrruYP38+DoeD/v37s2bNmnLBkfHjx3PzzTczZswYbrnlFqxWK4sXL+bzzz+vUqAg5L333sPv93P55ZdHfP6aa65h6dKlLFmyhHvuuSfiMtnZ2eHtqKoqWVlZ/O9//+PDDz9k7Nix4dRl8fHx9OzZk7fffpsWLVoQHx/PW2+9hdvtPm7qrwsuuICePXsydepU7rvvPpo1a8bKlSvZuXMn//jHP6r0WZs0aULDhg355z//SWFhIc2bN2fz5s2sWbOGMWPGAIGaJZ988gmLFi2iTZs2bNu2jQULFqAoCi6XK7yuuLg4fv75Z3744Qfatm3L2LFjcTgcjBkzhs2bN5cKpDRv3rxcqrMT5SsbHNF0qp+8TQghap9X03hhbzqt7Vaub5BY6yklxOmh6TobgmmH7mmewvtHc9nr9pLnV5m+81CFr2tgMdEjzkEDixmPprOr2EOMKZB9+PzEWB4PvnZ9XiHdYu2YFSXiKN1okO7xscvloaXdQkNL7d7Uhux1eXBrOu2D6ZTCs3mDOaR/zi9iR7GHn/OLKFI1Nhe42O8OpHc6lf3VJYMiFkVBJZAzP8ZoJNli4rDHx7YiN5sKj12DbSp0cWH92IiddDVlVnBGiNNo4MrkBNK9PrJ9fjYWuMj3a3yVU8hXOYUAXJuSSAOrmXmdWvB5Vj4NLGaKVI2VR3NoabeS6rRxSVI8zW0WfLpOW4c1PLrcr+lsLXLR1GYhsQ7f5N/aqHozRsoylTgeT4d1uYUsOpxFS7uVQr/GimBw64DbxwF3bni5Qx5fOAVYSS1sFprZLDS0mmlgNaPrxz5LjzgHveMCga4GFjO+YMqqSM5NjOGOOtDBHCUTfMRp0Df+2J1nisXEmGYp/GPnIT7PypfgiBAlbCoo5vUDmaR7fWzILybHr2JUwIhSbgZpWT/nF/PqgcyIzzWwmFjaoy3tnbZT0exS3KrGjRt28kN++RmNharGmpzytZ++yS2s1nuUXb6wRI2oYlWjSFVLpSbUdZ1LfvydDK+fT/u0PyUzYsWJqbtXsYKMjIzwrIZI3nvvPX7++WfWrVvHCy+8EC6uPXz4cD744AOmTZtGr169aNeuHVdccQXvvPMOa9eu5cMPP4y4PqPRyIsvvsiTTz7JAw88QFJSEiNHjmTXrl3s3r07vNwdd9wRDhy8+eab9OzZkylTpvD444+Hl+nQoQPvvPMOs2fPZvLkyei6Tvv27Zk3bx4XXXRRlbfBihUraNeuHe3bRx7507t3b5o2bcrSpUsZP358xFkya9asYc2aNUDgZjsuLo5OnTrx4osvcskll5Ra9umnn+Yf//gHjz76KDExMdxwww307t2bpUuXVtpOo9HIq6++yrPPPsucOXNwuVykpqaycOHC8KybqkhLS+P5559nzpw55OTk0KhRIyZOnMjo0aMBmDp1Kj6fjxdeeAGv10vTpk0ZN24cO3bsYNWqVaiqitFoZOzYscyfP5+7776bxx9/nIMHDwJw2223lXvPp556iuuuu67KbayMTysbHJGi7EKIumFZeg7PB0ceLT2Sw9/bNKKZzYJb00m2mEqlovBpOr8Vuvij2M2vBcUYUGhqM9PeaSPVaTttHcyi+r7NLeSgx0ecycCklg15pE1jVF3njYOZvH80lzx/INWmDtgMCoMSY7kmJYEuMfZK93Fjm4XRv+3htQOZvHYgE7tB4bkOzbk2JaFGjg2fpmNQwKgo6LpOsabhMBg44vVhUhTqmU24NY18v4pH0znq8ZHl82MxGIg1Gvghv5hPMnL5rdCFq8Rvt9MYmHUwKDGGVnYrveIcNA/e2OnBuhRGJTASN93ro77ZRLKl8lQlfk0Pd94XqhqFfhWNwAyFbJ+ftTmFLD6SDYA92Ck7tH4ce10efi0RdKhIPbORJlYLe1wezIZA4V+HwUBysKPXo2k0spqJMQby+RtQSHXacBoDo9xD/3cYDSQEZzvHmQJ5/jcWFJNsMdPSbqlwBOBRj49vcwtRdZ2X9mewqdDFvw9lMSv11ARH/gimx4JAIeUXOh5LGfRVdgEbC4rxBr+jOsXYmdA8BQikrgqlrwKY1rbxcd/LZFDoegqDPHVFaFJCVWdgVMfsPUd442AmF9SLY1ZqUxQCxbXNioJRCbznhC17ywU9prZqSKcYO5sKXDSwmukcY+eo18cv+cXk+1X6JTipZzKRZDHRsRozeaxnwG9XKLWSOPsYFYUHWzbk+T1HeLFjcxpZLfxj5yG+zS0s14kpap8nWPMn3mTkuT3p/JxfxPjmKQySWm+1Itfn58vsAj7JzOP9YE2mklQd1BJDTrrF2Dnq9ePWArMpusU66Bpj5/2juex0ecjw+srNpE73+vkiKz9icETXdb7ILmBFeg5TWzUMX1+eqFm7j5QKjIxumsztTZJoajOzrcjN5gIXmT4/Rzw+nMZAHal8v0qhqlHfbCLT5yPeZGS3y8sBt5d4kxG7wYBf10kwG+ke6wjX1Oob7wzPytzt8oQHqWR6/Tjtx75XClWN7cF0vZO27Wd5z7Yn9RlFzVH0qlRpjkKqqrJhwwZ69OhRYVoot9vN7t27adWqFTbbqY9MisAXWnFxMQ6HQzp+RDnZXj/73cei6U6jgba1MGogmpzsOSLfa+JMV5Xf99q2Ij2H8Vv2Vvh8jNFAotmEpus0sVnYWewhq5I8s/XMRjo57TS1WWhlt5LnVznq9dHAaqahxcxOlwcjgQ73Ix4f+X6VeLORLK+fJIsJj6Zz0O3FF5zmbjEoeDSNLjF2nmrflN+L3PSJcxJnMqKiY1UM7HV7qG82YVYUijUNBYVYkwFv8KZF03VMioLJoGBRFPkND3p0+wFeO5DJXxrV59kOzWpsvVleP92+3Yxa5iq8td3KxOYpeHWd3S4PLWwWki2BnPupThsJZiMb84vxqhrb9u3DnNwAH4H6Dm4tkOrp14JAvRO7wUCMyUC6x49P14kzGcj3B27inEYDPk0/7ui/kCZWM0e8vnLtDa0LAsdrsaphVCi1XJLZRKLZyCGPD7shEGyIMRnwaDq5PpVsn5+THSrR0GKmud1C5xg7bR1WWtittHNYcRgNxw3O1KaVR3MY89te+sY7eb/X8aoEVM9HGbnk+1XiTEbu3LwHgIvqxfFO99Y1+j51RW3+lhxwe+mzbgtWg8Lewd0rXG6/28uELXuZ2DylVCAqkhyfn6E//M7BMkEPk3IsCGM1BAKdoXRvzWwW6ptNXJ4cz8TmKfI9Xsaa7AJe2Z/BzNSmUoCb6Lzeqg26rlOkasSYjOi6Tt/vtrLf7eWtrq2Oe16KU+uqn7ezpdDFU+2bcs/WfQB0j7Xzvz6pp6U9deUc+TqngLl7jzIrtWm52kKR/JRXxOdZ+UxonkKMycjWQhf/OZzFfw5nhzv7IZAS9PYmSXRw2jjk8XHv1n10j7XzQsfm1DMHUmVVVM8KoCgYaLAZFFyazmsHMkjbd5Q7myQxo31TPsrIZfaedM5NiOGn/CIOuL2kB2fkPtyq0UmlIPVoGt2++Y08v8pdTZPY7/byZLumNKml7/4+637jgNvHR73a0bvEjLXdxR4GrN8a/vvK5ASuSIlnSL04YutouvloPk+q0zaZOSKEqDWhvhKLQcGrBUaySlF2IUQ0y/D6SgVGOsfY8Gg6h4I1CQwERgEVqoHAb6gjK8FkJNFspG98DEkWE3tcHv4ocrPL5SHbp/J1NadtV8VXOYUMXL8NCKT40dArHMmsQDCoUn6B+mYTF9WP5eL68XSNtWMzGMJFnCPxahqHPT5yfCqarmM2KHRw2iusNxDtvsou4J87D3F1g0R+CBZwHJQYU6PvUd9i4s2urZm+4yC3N0ki2+fnpf0Z7HJ5eOD3/VVciwH2Z1T4rEdTyQ3ObAHI92soBH6LS9782g0KKRYz9S0mfJpOnl+lodXM9Q0S6RXnINFsoqnNgkvV2Ov2sC63iB/yitjj8vBrQXGpdcGxwEi8yUi+XyXT5w/nXS5SNTLLZ/XBpEA9s4lYoxGnyYBZUUj3+kgwmWjjsGI3GHg3OHukc4yNAQkxtLZbGZYUj91oIMFkrBMdwTHBGzO3evIzZzcXFHP3b3v4e5vGFPg1/rot0JFUsnjz39s2Oun3EcdX1bRas/cc4fu8IkZs2s3hC7pXesyuzi4oFRixBr+vS36nezQ9HBiZ1b4pI5okncSnOPMNrhfL4HoyAv1spygKMcFOSEVRGFo/jn8dzGR1doEER06jfL/K9+Gi2fvCj28scPF7kZvUs2xAZVVouo5BUbhhw04gMBNh2XFmIhSpKnds3h2ewVHfYuLL7GPppdo6rHSPdXBLo3qcmxATnhnfGyLWyKvsd8xpMobry8ZDOCi95Eg2n2bmhX/jNkeYBXwyxcx1XefV/Rnk+VUaWExMb9uk1vubks1mDrh95eqOlB0890FGLh9k5AKB2dGNrWZaO6y0dlipbzZhNxhItphwGg1Ygv82KQoJJiMpVjNbC11keP3scnlINBu5OqX6tdNEgARHhBC1KHBHZzMY0NHwaTpFfpW4Opynui77z+Es4k1GLk9OON1NOav9Nz2Hp3cdZmZqUy6oF3f8F4haU6xq3Lf1WEd1R6eNT3unYjYoeLVAkWan0ciO4kCBPremkeH1k2IxcU68E0uEvOweTePXAhd7XR5+KwxM565nNlHPZGKv20O2z09bhy38/i3sgZHAR4OzRo54fBgVhU5OGxZD4AZf1QMdZKHR4kC5GQElRxtD4Ns4UmAEAhfuS47ksOTIsaLMTW1mEkwmvJpOE1tgJoNZUVAU+CW/uNy0+UuS4niza90bsf7CniM8HZwKXzJlU8lRXzVlaP24Uh3Z45ul8PKBDD48mksjq4VWDgu7ij3k+VVcqsYBj5d8v0a3WDvxRiNaYQHtU5IoCBYMb2Q1YzEotLJb6Z8QQ47PT7Gq0dIemEGR7fPT3GbBajCw0+VB1XVSnTYMVH5zG2I3GujgtNPBaWdUsBO2WNU47PFiQMGtafiDqbU6OG3YjAaKVY3fi9zk+1UaWwNprAqDIwmtBoUEs4lkcyC1j6GSNqh6oNhlts/PG11b06yOjvq2BQOGrhpIKzpuy152u7zcsXkPjUsELz/Pygfg3uYpdHBGT+HzM5nxBNJqbS1yV1qYvuRM6xsbJjKicRKfZeZxXcNEmloD9V92FXvY5fLQO85Ja4fkLRfiRHQJnof7SpxzovZtrSRN5lU/b+eVzi05PzGGbUVudhZ72FLkoqPTToLJiF/X6RxjJ6WSgTxuVSPL56eR1Yxf19GhwvpJdcFX2QWM3LQ7nBoT4JeC4nLL7Sx2s+xIDpcnx7Mqu4DV2QXh2Rmh61wFuCI5gesbJHJxUlyl12MnIxQcKVA1CsoMErk2JYHhjZNYnZ3Pi/uOUqiqkVZRjq7rjNi0G5eqcX3DRBYeyCTD6+eINxB4+Uvj+qdlIG6oMHuGt3QwJDP4d3uHjXHNk/kmp5B1wRS+Lk1jp8vDTpcHso7/HmVnatsNClcl10x63rOR9EgKIWpN6LtbIZCvO8vrJ8PnJ7aOjPg8k+wsdvPAtkCn7zvdWnNR/ap3ynuDKXkMSuBCc6fLg8Wg4FZ1cv1+4kxGWtmtHPb4cGsahX4Nk6IQZwpcgB7x+ok1GnAF88rajQacxkAHrz/4n0/TcWs6e1wenMGURbEmI0e9PhQCedRjTAZijMZAehhdJ8ZoJNZkwG4wkO71cdjjo1OMHY+moeuQr6oYUUixmIg/zQE5Tdd572gujaxmxgVnJYzatIe/tkjBp+v0jY/BZlBIsgTy9Xs1nSyfnyK/SlObBafREB6JI06NIr/K9Rt2sqGgGLOi8EGvdnSPPVZHwmIwhIMf1crXbjBwTryTc+Kd3FDDbb6pYT3eP5rLO91aYzMqOIwGmlktFGtaILCha2h64Lu4wK+S71epZzZhNQSW9evg0zQ2Fbr4LDOfL7Lz2ePy4NeDhX0J3Gj8Uewu994mBeJNgZFNqq7TP75mZ1qcSrqu87/MfHL8flZGyLHc2GqmaSU33DXFaTLyQMuGPNCyYYXLhL4zw9PE2zau8hT2krN/amoEpsNooI2j4nU5jAZ6xp18TQpj8Bx0aVqdDYzAsZopZYOJJ6JkR17EIttS5LPWGDh2DRsayRtJyf3+XnpOpcGRHcHvWbvBwPS2TahnDgTdS+odbzolgVshziYNg7+NRyJ8j4ra81uwo75PnAO3prOtyMW/u7Vm5q4j/FJQzM0bd9LGYWVHsF5DJI2tZhLNRlrbbRT4VQpUlQNuLwpKYKCJFqhn4tU0fLrOlckJzEptVufSGWm6zuM7DuLSNJ7dcyT8eJGqsdflKfX7P237Ib7Izmf23vTwYxZF4ZE2jcj0+jEpCjc2rFcrAfZI12+jmybz54aJdAnWL9tSFDgOQmlgjyfD6+ez4KCQkrPy7QaFv7ZowL0tTjw118lIDgZHMr2lv1dCM0da2C3c0qg+tzSqj6oH0iWreiBN506X59jgKE3jaDBw4tF0Mrx+VD0wyztUf6y5zYLFoHB/y4bSp3YSJDgihKh9SuAHI9vnp9Cv4dI0HFGWn/BMtz73WHGy237dRbLFhAK4NQ27wYCCQqeYQIdXoB6Cn3iTEZ+us7vYgy844qausgWL8vq0QDCmYbAY7x1Nk7ixYb1T/v4rj+aWq2Hh0rTwiHVIL/+iMmKMBhpZzeH9YjcYSDAbsRoMGBWFQYkxNA9ehCZZTOwuDqRzamqz0Ch48xCoSVD+3MvwBmoDZPr87HV52VnsDnesFqsaMcZAvYADbi+5fpW2DiudYuy0dwRGi58MXdfxncTBpep6xJHwXk1D1QOj3yO9Z6bPT33zsdHrs/YcYUNBMYkmI/M6taBHDXTwnmovdGjGzPZNy+0DJ4F9bObYvo4zGWlS5vUWAKOBQYmxDEqMZXpwiXSPj/1uL/n+QBnGwx4fjuB7uFWNPvFO2jmsdfaC/KX9GUzfeajUY6v7prL0SA4H3V7uaJIUNZ8t0vF7tmhch4MiIaFz032SM0d8WmAQQVlf9e3A+d8HUus1roWAnggwlfh6UPVjBdrLyijRSfLivqPk+FWGJcWXGqCi6zqrsgv4JT8wAnhOMLe7EOLUaBT8rjwswZHT6veiQEB4QEIMk1s1IsfnJ8VqZkBCDBf98Ds7ij3sKPZgURQ6xdgxKoGOcbvRgKbr7Cj2cMjj45DHx2+F5QfxhOSVSDf636O5xJqMzEqtuZpyteHjjDy2FEX+jM/uOcLcji2AwLXCF9n54ee6xNi5oF4s16QkhIMRtamprfx1yfn1Yku1JS54X1rVmSP7y8z4uiYlgZsa1qNXnOO0DoZMCr53utfPc7uPkOq0cUVKQnjmSGhmCQQGAIWKz7dyWDmP46d/zPP5KVI14s1GnNKPViPkSksIUWtKzhyxBouyFvo1XKqOQ77Ta1Uop2tIySmf+cHSuEeyS98kVJQFP9FkRFECF2C24KjxbJ+KSYEUi5m44HTn/GBqmBSLGQ0dh9GAzWDApWoUa4HZJQYUzIZA/m6zotDEZsGvB4r2FvhVEsxGjIpCkaqFU7MUqoGRE6GOWwADgYuOo14/BghOnVbwBVO+uDUdd4nPXBAchZS27+gpDY5syC8m0Wzkk8y8Uo/f0zyFXL9KkRqoCfBDXhE6cNDtDRcqtpVoPwTqXGyvZPTUivScCp8rKTCCSqel3YLdaGB7kbvcVOeqMiqBgtIdY+w4DAYsBoUNBcU0s1loarPg1XRyfX5y/SpWg4KCQoLZiEsNBC5CxaQzvQba/PgHiWYTuT4VHT08qmu/20uS2USOT6W+xUgjqwVV19H0QH7a3wpdtLBbcBqNZHh9xJmMHPb4KFI1dKBBMH2P3WAg1+/HaTSSFUw/lGgy0s5p45DHywF34Pif26kFQ6oxs+p0UhQFm7HmO/EbWM2V1hypC4pUlY8z8hhaP47EEjdLRarK3H2lg5GJJiOpDht/b9O4tpspznDWYK/5ydYc2VbkiljEvr3Txj/bNeGnvCIGJUpthdpSMmWHX9cxU/57+KvsAtbmlK419fahLN4+lMW9zVOY0roRBmDRkezwzF4I5H8XQpw6oZkjWT4/Hk2r06mW6rJQbYbGNgtmgxJOkWU1GJjfqQX3bN1HoV/lne6tI6aMzPer/F7kJsfnZ1exh4Rgp3ETmxmzomBUFNo4rGwpdBNjNLC92M0dm/fwzuEsJjZPCXdO1yZd19nj8hJrMpbqLA/Rgvc3JoNCptfPjF2H2O/yRqxbeHlyPB9l5LH0SA43NqjH+fViw3XzAF7p3JIrkuNPWcqsqnAajRggfP2SaDJyUZk6UHHBjv58f/WDI1ckxzO3Y4uoqHmYbAkcv/86mBl+bEe9rmT6AveX9U8ycBNvNhFft2/Noo4ER4QQp43dYKAQrUZyb4vqCV0s/btba7rH2jni8WFQFAwEUnXYDAZ2FLuxGw3EGo0kW0wUqxpmRaF+cJZJ/WA6nrgIadFyfH4sBqVWRzJ4tED7PJqOUQlMM832qSSajRgUBT0426VQ1cj1+cMXXTajgdXZBTy6/SB7XZ5KU2KcjL0uD1f8/AeWYEAoxG4wMLpZcvgiqqQCv4pBCaTsMCsKOoGZEaquc8TrY3uRh2y/nyRzYP9keP0c8gQuEjcUFJPlVSlSVTK8fto5rTgMRnL9fg64feHzLjSCamuEEUhGBRpazLR32kj3+LAaDMSaDBSpGhaDQgOLmQSziT+K3GwtdJHjV9le7CkXtPm1oOI8wpEpwXVEDv6EgnlHvJFHh+12HbtQTi+T67Xs39m+YxffOSWKQQL0j3eWu2gXddNTuw7z2oFMzk+MYUmPY8UqZ+9JL3UMAHQtkT5NiJpkq6G0WhuCecUbWEzc3Kg+P+UVMTJYB+aupsnc1TT55BoqqqXkNYMWoSh7kary5407w3+/3LkF/8vMJ8vrZ01OAS/uO8rS9BwsisLeYEdPtxg7gxJj6ShFiIU4pRJNRiyKglfXeWDbfmZ3aBaxZpw4tXKD12KJEWa0d4t1sKZvB3Rdr/D6LM5kLJd6MJJQqs92ThvnJcawNqeQl/dn8FjbxsFaekrgPsvjI9GokKUFBqsl2azhAV7e4G94cXCAnwFwGg3EmYzEmIwUB9NbtXXYKA7eh+X51XCq2o0FxdgMBva4PGwqdGFWFLrE2ClSNVo7LPSIdfBrgYvv8grJ86vEGo3klgkW1DMb6R3nDKeUuqlhPVIsZv51MJP7t+1jdd8OPL37cPi5qyIUUz8dzAYlXPNwSutG5fZnbDANd1WDIweCv5lXJifwapeWNdfQkxQp2NV73ZZwjbIkmREadWSPCCFqTeh+MfQTaCuRlkXUnkyvP1Doi0Be1wSzqVTHfKh+wuCT6BROPA0/+KGRXvYSI+frl7gwURQlXO8mrsyFd0ublcd3HMSt6Rzy+MIF42rSmuwC/Dr4g8e71aCw+pwOGBQiBkaAiDlwQyNS25iMleb6L6nszUQowOLV9PCIm53FHvy6ThuHlfpmEz5dJ9FsKretKnuPdK+fLYUutha58Woabk0nPniTkOXzY1YUbMEizOkeH83tForUwKwhqyEwqqu93ULWzu3Et25LnqpTrGoUqIEbA6MSKOaX7vXhNBrRgoXIQ7PREs1GmtosHPH40Aik78vx+XEYAjVaGlrM7HV7UPVAMC3BZCTbp9LQaqapzcKWQhcH3F4aWM04jAY6OaWT/EzxzqFsAL4qMXJb03XeOhQY0fVGl1YYFfgwI4+RTeqfljaKM18oOOILfgefaJHQDcGUSzc1rMfDrRvVWPvEiSmVVivC8++l55b6e2BCLFenJAKw+HA2j24/UCqlT0u7hf/2aiupMoSoBUowMAKwPD2HFIuJx9qWTTwqTrWc4MyRBHPF33s1fU0+onESa3MKef1gJq8HR/hbgu/hDQe6jfD97zX6vmX5dD1cTP2PYjefZuaXej4UGOnotNE91sH/ZeXxZLum1DebwsGRVKeNgQkxfJGVzz63l77rtpDjV7EZFKa0qriWXW0LXPcEtq0pwv4M3XcWVLHmyIHgb2cre3SlXm0RoT2hAYkGoHsdSNd8tpHgiBCi1pQdS+cIdhIUaRoFfvW0FUML1DjQA8EbhfBoEKMSCOi4NR2rITBqXwEKVBWTouAwGIIFxANpFEIzOA0o+PVjrzEGU0T5dT1cT0FRNUyGY7McArUvAjfYiqLg03SU4OwHhWPTT13B9D9/FLnwe1RQArlyXarOXpeHJjYL2T4/e1wefs4vJtZkxGpQ+L3ITUGwmPd/Dgc6CVOdNhJk1AIQmK7c0h4o8rer2FNhcCRUt6JY1dhS6EJFp63DRj2zCa+mYVAUcnx+Cvwq9S0mGlmPrefTMqm0prdtQqtaSpdR9mbCGJxebjEcC4ZVp6h4Re/R0GqmodV8UmmoVFVlgwF6JMRUudB0WZUVua0fYSRPSJ94J32kuO0Z57UDGRFnKO4s9pDv17AbFIbWj8NkUPhTUvxpaKE4W5RMe+dWNZwneN0Tmo0nN9fRoWxarbJ2uUrPgqxXovPvpkb1uDIlgZ/zi1D1wLVZktmEKQrSgghxtrg0KT6c8nZ9mdTDonaEAgAJptq7N70iOZ4xzZJ5/UAG/uBXtzfCd3goFZRJgXiTCaMSqC9lMQSyLpgNCrpOcFaJBigoSuB3vr7FRLLZhNVgoIXdgk/X6eC0kWg24dF0rkyOp1DV2FnsQdV1tha52evy0CnGTgenjU4x9nA/SWOrudxMxWtTElCCRbkVRWF2h2bcsGEnOcHtOSwpPqpqtplLtD9SJuBQcCS/ijVH9gZ/X0/FwMaT0aNMTZdXO7ck2+enmc1CqtNGkyhrrziB4IjH46FXr174/aXTUjgcDn755RcANm3axKxZs9i8eTNOp5PrrruOiRMnYrEcOwAyMzN56qmn+Prrr/H7/QwePJipU6eSkpJykh9JCBHtSs4ciTcbyfOp7Cr2EGsykGwxlwqSqLpOltePV9fDr/ME61PoEAxCGDAQGH1QpGrhG1OrQcGgKJgU0PTARU0oEOHWdExK4ALIq+lEuA46hRSopFZEZXSvjyyfn79t3sMB7eRunAckxJzU6880rYPBkT9v3Ekbu5XGNjOZXj8H3F7iTEZcmlYu/U5V1lkQvLgLpYJa1qMNTW0WWp6G3LZCnI0e3X6w1N9FqorTaAynJuoa65COSFEr7CVStbg1nRMNxWaHcrNb5eY6GpRMwKNGuKD0lgjO/q11o3KpOx1Gg9SIEeI0eq5DMwamx/Do9oNSmP00yQml1apk5khNUxSF6W2b8LfWjXCrGl5dx6Pp+HWdRlYzX2Tk4d+7iyE9ukGwXmp1Zq9UJ1VzaGDXNdVov0FRWNC5ZanHBibGsrRHG/6+/SBbi9yMaJxUjTWeeiUDIpFmjsQEB8YVqxp+Ta/0+tyn6fwYDGZ2Ow0F5itjUBTua9GAF/amM6NdE66MkrRmomLVDo788ccf+P1+nnnmGZo3bx5+3BC82N+/fz+jRo2iR48evPDCC+zcuZPZs2eTm5vLE088AYDf7+fuu++msLCQxx9/HL/fz3PPPcedd97JihUrMJulsszpUFkOR1H3ROP+DN8ulmhXM5sFI16yfSoFfo0CvwebMXDx4dd1ivxaxFF4JRVGLEsKxdXrxw43zUAgoKIS2I5mg4JPCwR1dHTsRkN4RgkEfuStBgVVDwZfdB0FBZem4TAECpT79cCPuz8YiDEEZ5KEXm8xGDAFX+vXA8W3NT20LoKzUhQwGrAaFKyKgSbWwFf4YY8PgwKxRiNuTaeh1UQDi5lkixlrcHZKqtNGQ6uZg24vjaxmClWNm05h4fG6aEj9OP4vODV5p8sTTj0GRCxQ3tRmxqIYyo0ItQXrsGR4/aWeMysK97dsIB0gQpxme1xeOsfY+TUYHCk7ukuIU8WgKOHc9u6TqLfmLTEIRJx+iqKERxJHKicTmpE8qWUD7m3RoJZbJ4Q4nnpmE1cmJ/Do9oMc9fqO2ykrapZbPVaDNOE0ZJKwGgzh9MwlXZIUx4YDYDcaTmg2++kqfj4oMZb/65NKuvfUpIo+GSVnjpgjbJ+S6ZwLVJVEQ8Vd1r/kF1GgatQzG+kae3IZEE6FB1s2ZFhSPN2jsG2ivGoHR7Zt24bJZGLYsGGlZoKEvPrqqzidTubPn4/FYmHw4MHYbDb+8Y9/MHbsWBo3bsynn37Kli1b+Oijj2jbNlAUs2PHjlxxxRV88sknXHXVVSf/yc5wU6dO5b///W+FzyclJfHNN99UaV35+fn885//5MYbb+Scc86pqSaWsmLFCh5++GG++OILmjZtekreA2DSpEl8+OGHTJkyhTvuuOO4yw8fPpzvv/++wueXLVtG165dq/Tew4cPB+Dtt98GIDU1lYkTJ3LPPfdU6fU15ciRI0ybNo1p06aV2tZTp07l+++/Z9WqVaX+fTw1+zkCN4clfwaNikIzu5Vki0amz0+2z49b1UrVIbEYFBLMRvTg7A+rQcGrBdJYGVFQ0VGDqa3sBgNGJfAe3uDID5XAj68hmDIrNA3WZlCwGAxYFAVL8AJYoebymUYKUOm6TnFxMQ6HPRxMqc77uY2gWcx80bcVNlug3kS+X8UAxJymtGRnihGN6/N7sLD47U2S8AVnJenB49ZmMKDpOgMSYnAYDeFijXtdHuwGA06TAQMK9mAtncMeL1sL3diCga/OMfYTTqEihKg5WwpddI6xc9AdzFNcS+nthIBAai2v/+SCI75gZ3ukjgVxeoSuRyMN6DkWzJIiz0JEqyTLsXRJGT5fqdS44tQqWYvhdKXZPtOYDUrUBUag9GyRSHXXzAYFu0HBpenk+9VK65iuyw3MGhmYEHvCNdxOJZNBoYekP60zqh0c2bp1K61bt44YGAH4+uuvGTx4cKnnhw0bxvTp0/n666/585//zNdff02rVq3CgRGAtm3b0qZNG9asWSPBkSpKTk4mLS0t4nPVmX2zdetWVq5cyfXXX19TTTstCgoK+Pzzz2nfvj2LFy9m1KhRVep07tSpE4899ljE59q0aXPC7Vm8eDENG9Z+8atvv/2WNWvWlHt8/PjxjBgxoty/a1Nl8z9sRgNNjRYaWszk+VU8moZOIBBSz2w6bSMvTsbxjr+aGpFU1YLZonIGReGp9tUP3raoID1WI6tFbqyEiEI/5hVxY8N6HPEGgiMNKqlDI0RNsxkM5KOFZ5+eCE/wtRYZ2Rw1jMEas5HSakkwS4joZ1QUGlrMHPT4OOyW4EhtyvEfK8ZeF+/5RdWVDI6YKtjVyRYz+9xethd7KrzPBvgpPxAc6RMvAQhx8k4oOGI0Grnjjjv4+eefsVgsDBs2jMmTJ2MymTh48CCtWrUq9Zp69eoRExPD7t27Adi5cyctW7Yst+7mzZuHl6kqtZJCPaqqogcLIOu1W1DglNN1HbPZTPfu3StdpqrrCv3/ZLdTyXWdqveoyIcffgjA3/72N26//XbWrVvHgAEDKn2Nrus4nc4a345AeJ21fexVtK2bNWsWfrzkv6u6zpr4HKFVKJW8t1EJFaos3eF/ppzDFZ0j1Xm9ruuoqlrp958QdVXouJbjW5wqP+QVoaoq6cG84skmY5073uQ8qbtswYBGkc9/wvsvVMPCFLweEOXV9jkSGLWq44twfRaaDW1G9peIHvI7Ul5DayA4csDloXuM7XQ3p87TdZ0l6bm8cSiLAlXFGqwP6tJ0HEaF3rEOPCXuieOj7HpMzpGaVzI4olRwDTM4MYa3D2fzaUYu7W1mtha5aWG30M5x7JzUdZ1f8gPpcbs7bbKPTqNoPk+q06ZqBUd0Xef3339H13VuvPFGxo0bx6ZNm0hLS2PHjh3Mnj0bgJiY8kV+nU4nhYWFQGCEf4sWLSIuU1RUVJ0msWnTpkqfN5lMuFwutJOYuh6NQoGf4uLiSpfbsmULt99+O5deeinTp08HIDs7mxtvvJH27dtzxx13MGbMGABGjhxJ7969efXVV7n77rtJSUnB6/XyzTff0L17d+68805Gjx7NK6+8Qp8+fcLvcffddwOBlGoAmqYxd+5cVqxYQW5uLv3796dXr14AuN3ucJt37NjBiy++yC+//AJA3759uf/++8OpoH788UdGjx7N3/72NxYuXEhBQQHPPPMM/fv3j/hZly5dyjnnnEO3bt1o1qwZ//nPfyoNeoTaChx3O6qqyvLly1m2bBn79+8nMTGRYcOGMWbMGKxWa8R19erVi9GjRzN27FgA8vLymDt3LqtXr6awsJD27dszYcIE+vXrF36fXr16MXXqVLZt28aqVavw+/0MHDiQyZMnU79+fSBQ1+e5555j48aNeDwe2rVrx913382gQYN4//33efzxxwEYOnQoV155JdOnT8ftdvPqq6/y+eefc+TIESwWC126dOG+++4jNTUVgMcee4wjR47QokULPvnkE1JSUliyZAkAPp8v/LkWLFjA66+/zmOPPcaVV14JwM8//8z8+fPZsmULFouF888/n/vvv5/ExMRS29GnAyj4/T6K1bO72J3L5Tqh13k8Hnw+H9u2bavhFgkRXY73+y5EVcVgoBCFMTaNV9wKW4rczP5hI+keBVDI2vEHG+poths5T+ogrwFQ2PzHHxhPYNKSqoMaHEDyx2+/caSOHru1pbbOEV0N7tctW8ktM6E3szjw3JGDB9mQcaBW2iNEVcnvyDHW4Ln6y+49NDl4ZgzMO53ediu8663gR8oH+9x5pR6yeT1s2LDh1DesmuQcqTl+T+AcA9i3excb9pdfpp0fwMi/D2fz78PZ4cefcqh0C143HdUgw2fEiI62azsbZMLRaVfXz5NqB0cWLFhAvXr1aNeuHQDnnHMOSUlJPPTQQ6xfv77S14dSzFQ2Yrq6uf67du1aYXEkt9vN3r17sdvt4dz8wQaAr/LO8FpjdpQqTl1VRqMRRVEqTG8Wer5Pnz7cfffdLFiwgOuvv57+/fszefJkNE1j1qxZOJ1Opk2bxhNPPMG0adPo27cvDocDg8HAZ599xpVXXsmCBQvQNA2TKXC4WK1WHI5jU9cMwfy5DocDXdd58sknWbRoEePGjaNbt258+umnzJ07FwCbzYbD4WD37t2MGjWK1q1b8/TTT6OqKgsWLODOO+/kvffeo379+uGgw6uvvsojjzyCx+Ohf//+pd47ZPv27fz222+88MILOBwOrrvuOubPn09xcTFJSUkVbkeDwYCu6xG3Y2gbAjz66KOsXLmSu+++m969e7N161bmzZvH9u3bee2111AUpdR2CDGbzTgcDjweD+PGjSMzM5P777+flJQUli9fzj333MNrr71WKuAzb948hg4dyvPPP8+BAwd46qmnMJvNPP/882iaFn79rFmzMJlMvPXWW9x///18/PHHXHzxxaSnp7NgwQLmzp1LamoqDoeDhx9+mK+//pqHHnqI1q1bs3//fubMmcOjjz7Khx9+GCgkaTTyyy+/4HA4SEtLw+VyERsbW+pzvP7667z22mv885//DKdh++GHHxg3bhz9+/dn9uzZ5OXl8eKLLzJ27FiWLl1a6twzeXzg9WM2mXFYq5767Uyi6zoulwu73X5CtU0MBgNms5m2bduW/l4T4gyhqiqbNm2q9PddiOowfPMbqBp/6doBy5Ec5u7P4FnXsZv1wT26hWsI1RVyntRdCT/vYF+hi6atW9OjXly1X+9SNfjmNwB6d++KU/Z/RLV9jli+3UKRX6V9hw60d5a+PrNv2g05hbRp3pweDRMrWIMQtUt+R8pr9vsBSM8hvmEjejRPOS1tWJtTyHd5RUxslhyuY1gXHfb4WP7974DO/c1TGJjgxKcHUin5dTjo9vJroQuLQWFdbhFmg8LUlg3pkVh+oPXpIudIzYv5aTsUuQFo36ZtxP3dVddZ+ssONhW6Sz1e1KAJPZolA/B5Vj78tpd2Dhv9erY/9Q0XFYrm8yTUtqqoVnDEYDCUGuUecsEFFwBw4EBgJEyk2R+FhYXhjtaYmJjjLlNVRqOxwh0Q6twO/QcEAiP/Ggb7Kw/k1Jpm/eGOT6sdIFEUhUOHDtGlS5eIz0+ePJk777wTgAkTJrBq1SqmT5/O6NGj+fzzz5kzZ064Hkao9kvbtm3DQS9FUTCbzTzxxBPhwEEo+FVqe3IsoKUoCvn5+SxatIhRo0YxceJEAM4//3yOHj3K2rVrw6+dN28edrudN954IzzTaMCAAQwdOpTXX3+dKVOmhNd76623cumll1a6PVasWEFCQgIXXXQRiqJw7bXXMnfuXJYvXx6euVHRdvzhhx8ibsfnn3+eyy+/nB07drBs2TImTZrE6NGjARg0aBApKSlMnjyZtWvXMnjw4FLboeT6FUXh/fffZ9u2bSxZsiQ8m2Xw4MEMHz6cZ599luXLl4df0759e55++unw37/++iuffvopiqKQnZ3Nrl27GD9+fPi86969O2lpafh8PurXr0/z5s2BQC2Vpk2b4vV6cblcPPjgg9xyyy0A9OvXj8LCQp566imysrJITk5GURT8fj9PPPFEuVopiqLw7rvv8uyzz/LEE09www03lNpOrVq14uWXXw6fiz169ODyyy9nxYoV3HbbbRVu+7NZ2fOouq+r7LtPiDOBHOOipoTmDpuNRqa2acymQjercwrCz9urUact2sh5UveEOru8unJC+85folaJ3WTGKHVHKlVb50i4jpzBUO79fMFdZjPJ+Sqij/yOHBNrDmyHYk2v8jbJ8vqZt+8o1zRIoFvsydc+mPTHAQ56fGwqdPNO99Ynvb7T5b8Zmfh0nX7xTqa0aXy6m3NS5BypOSVrrloq2K5G4OXOrZizN50h9WPZUuhmzt50Dnj94eX/cHkB6Bhjl30TJer6eVKt4Eh6ejpr1qxh0KBBNG587AvO7Q5E9JKTk2nQoAF79+4t9bqsrCyKiorCxa1btWrF1q1by61/3759dOvWrdofovrOjJuI5ORkFixYEPG5Ro0ahf9tNpuZOXMmN954I4888gjXXnstw4YNO+76W7duXeHMlIps3LgRv9/PhRdeWOrxSy+9lLVr14b//u677+jbty82mw1/sABXTEwMffr04dtvvy312o4dO1b6nj6fj/fff5+hQ4fidrtxu904nU569+7NkiVLGD16dHhWRySdO3cOpxwrKRRk+P777wG4/PLLSz1/+eWX8/DDD7N+/XoGDx5caRvXrVtHcnIynTt3Dn9egAsvvJBZs2aRl5dHfHw8EAgslNSwYcNwCqakpCTatm3L3//+d77++msGDRrE+eefz8MPP1zhe1sslnDKs/T0dHbv3s2ePXtYtWoVAF6vN7xsQkJCxCLyX375JVu2bKFPnz78+c9/Dj/ucrnYuHEjd955J7quhz9bs2bNaNOmDd98803E4MhZHhcRQghRi0LBEYVAXYC5nZrTNTjyvrPkFBe1zBa8JnWfYMpfbzA4olBxMVNR+4zB+0t/hAwJ3uBjFglkCRHVYoIde4Vq5d/PRz0+ClSV1nYrt2zcya+FLtbk5PN5n9STHgB4MFgP7YvsfK75eTvn14vlsuR40j1+VF0nz69SqKqYFAUDCi5No8CvUqxqOI0GUp028vwqWT4/BX6N+hYTRX4Vg6LQ0m4hzx96LeT4VXJ9Kv3infh1nXxVpU+ckzy/ys5iDw0sJvJVDQVoZDXT0Wmr9PMVqxoGwGY0sLPYA8AF9ao3+Fmc2cwljh9zJb+JrR1W5nQM9McV+LMAOOA+1m+1NTj7pIPTfiqaKc5C1QqOqKrK3//+d8aOHcv9998ffvzjjz/GaDTSp08fBg4cyOrVq3n44YfDHev/+9//MBqN4dRBgwYN4sMPP2THjh3hWQs7duxg586djBs3rqY+W2SKEpipUcfTakGg07tr165VWrZjx46kpqayefPmcoGLijidzmq3KTc3Fwh0speUnJxcbrmPP/6Yjz/+uNw66tWrV+rvSGm0Slq9ejVZWVksW7aMZcuWlXs+NLOjIk6ns9LtmJeXF/EzmEwmEhMTKSgoiPSyUnJzc8nIyKBz584Rn8/IyAgHR+z20l/wodRfEJg1sHDhQhYsWMBnn33Ge++9h9lsZujQoUyfPj28jrLWrl3Lk08+ya5du3A6nXTo0CG8XUumuaton//2229ccMEFrF69mlWrVjFkyBAA8vPz0TSNV199NRyAKSmUGi1EMrcKIYSobaHfOUPweivZYub7/h2ZtfsIlydH/t0U4lSxBjsD3NqJXRWFitdaDSc2A1WcGsbgrlAj7NZQQMss+0uIqOYMzuwrrKCIr6brzNp9hLR96ag63NuiAb8WBgYx/lbo5peCYnrFVb8PJcRbJmj+XV4R3+UVMWv3kRNeZ026oUEiaZ3K1w4G8Ggaw378gwJVZV2/juwPdmQ3s1VvsK04s5UsyG6s4k9i6BjaXyI4si143nWUQU6ihlQrONK4cWOuu+46Xn/9daxWKz179uSnn37ipZde4rbbbqNVq1bcddddfPTRR9x1112MGjWKPXv28Pzzz/PnP/85PNvksssu46WXXuLuu+9m0qRJADz33HO0b9/+uOmTaoSigOXEf7TqosWLF7N582Y6dOjAjBkzGDBgAHFx1ctzHLoBK1vcvqioKNypHirAnZWVFZ4pBMeCJiGxsbGce+65jBo1qtz7hGqbVNXy5ctp1qwZM2bMKPW4rutMnDiRd99997gzOyoTCjhkZGTQpEmT8OM+n4+cnJxyRccjiY2NpWXLljz77LMRnw8Voa+KBg0a8Pjjj/PYY4+xbds2Pv30U1599VUSExN57LHHyi2/b98+JkyYwNChQ3n55Zdp1qwZiqLwzjvvlJrNU5mbbrqJ6dOnc8sttzB9+nT69u1LTEwMTqcTRVG4/fbby82sgfKBnhDlDJm9JYQQIvqFrlpKziFtbrdWeIMvxKlkM57czBGfJrMQopExdJ8UaeZIcF9b61htIyHONjGmwMyRogpmjrx6IIMX9qaH/55T4t8Av+SfXHAk0xvIwmBSYGXPdqw8msvKozmke/20tFuINRqJMxmxGw0UqxqFfpVki5lkiwm70cC2Ihc5vkBgJ9FsJMFkQkcn1mTEq+nsdXmJMRpI9/owKwoJZhNWg8LvRW7sBgP5qkqm14/NEJiBcsTjw2QItMut6SxLz2Fyq4Y0t1vLtX15eg5/FAdG8+91e8Md2c0lOCJKKBkcMVVxwEBTWyD97a5iTyBbiQ7bgzOTOjglOCJqRvV6oYHp06fTrFkzVq5cyYIFC2jYsCH33nsvd911FwBt2rRh4cKFzJo1i3vvvZfExERuv/127r333vA6LBYL//rXv5gxYwZ///vfMZvNDBw4kIcffrjaHePi+A4ePMjMmTO54YYbmDhxIpdddhkzZsxg5syZAFXOCxeqDXLkyLGRC3l5eezcuTOcDq1nz57YbDY+/fRT+vbtG17uyy+/LLWuvn37smPHDjp27Bje57qu8+CDD9KiRYvjptIKycjIYO3atdx1110R6+EMGzaMFStWkJ6eToMGDaq0zrJCn+Ojjz4K1xwJ/a2qKr17967SOlavXk39+vVLpaR76aWX2Lp1a4VBk7J++eUXJkyYwEsvvUS3bt3o2LEjHTt2ZM2aNRw6dAigXAqxzZs34/F4GD16dDhVGBAOjOgRbuLKCtUkefzxx7n++ut57rnneOyxx4iJiaFTp07s2rWr1Owbt9vNvffey+DBg8Ozw0q+l9zOCyGEqC2hAfrSlyyigSXYGeA70ZkjwY52syId7dEkNAI2Ulotny4zR4SoC2JCM0f8pYMjfk3niZ2HeOVABgDDG9fn7UNZ5V4fCkxURtf1Cmf9ZfgCwZEks5ne8U56xzuZ3rYxLk3HcZqLs1/503Z+yC/i65xCbo0QHPngaG743wfcXg56gjNH7BIcEcecSHCkidWCzaDg1nRGbd7Nvc0b4NN1YowGmZkkaky1IxEWi4Xx48czfvz4Cpfp06cPS5YsqXQ9jRo1Ii0trbpvL0rwer1s2LChwudTU1Ox2Ww88sgj2O12Jk+eTHx8PPfddx9PPvkkl1xyCUOGDCE2NpAHcvXq1cTHx9OhQ4cK19eoUSPmzZtHTEwMiqLw8ssvl5od4HQ6ueuuu5g/fz4Oh4P+/fuzZs2acsGR8ePHc/PNNzNmzBhuueUWrFYrixcv5vPPP+fFF1+s8jZ477338Pv9EWctAFxzzTUsXbqUJUuWcM8991R5vSW1bduWa6+9lhdffBGXy8U555zD1q1bSUtLo1+/fpx33nnHXcd1113Hv//9b0aNGsXYsWNp1KgR3377La+++ip/+ctfMFexGGynTp2w2WxMnjyZe+65h6SkJL799lu2bt3KiBEjAMIzgj777DPOP/98OnfujMlk4plnnuGOO+7A6/WyYsUKVq9eDUBxcdVTzHXo0IGRI0eycOFCrrzySnr16sUDDzzA6NGjmTRpEldddRWqqrJw4UI2btxY7ntC0moJIYSobVrw18cgoXkRRU70mshbIq2WiB6hmSORukY9muwzIeqCmArSar15KDMcGBnZuD7jm6dEDI5k+/zlHivpqMfHpT/9wbmJMcztWH72akZw5kiy5Vg3naIoOKqaf+gUGpQYww/5RXyZXcCtjeuXez5UYwTg5/wiVD0wGKCBpWr9HOLsULIge1WDIzajgSfaNmHyHwf4NDOf38P1RiqvgSNEdcg0jTosIyODm266qcLn33vvPX7++WfWrVvHCy+8EE4PNXz4cD744AOmTZtGr169aNeuHVdccUU4zdKHH34YcX1Go5EXX3yRJ598kgceeICkpCRGjhzJrl272L17d3i5O+64g/j4eN566y3efPNNevbsyZQpU3j88cfDy3To0IF33nmH2bNnM3nyZHRdp3379sybN4+LLrqoyttgxYoVtGvXjvbt20d8vnfv3jRt2pSlS5cyfvz4Ks+SKWvGjBm0aNGC5cuX8+qrr5KSksKIESMYP358pcXeQxwOB++88w7PPfcczzzzDAUFBTRp0oRJkyZxxx13VLkdVquVhQsX8txzzzFjxgzy8/Np2bIlTzzxBNdddx0A/fr149xzz+W5555j3bp1vPLKKzz33HOkpaUxbtw44uPj6dGjB2+//TbDhw/nxx9/JDU1tcptuOeee/j000959NFHee+99xg0aBCvv/46aWlp3HvvvZjNZjp37sy//vWvcsXlQ+QnTAghRG2RmSMimpzsfbxX0mpFpXBwpLKZI7LPhIhqkQqy67rOGwczAXi0dSMmtmjA0WDR9LJyjhMcmbM3nYMeH0uP5PBih+blOnYzvIH1Jlmir5vu4qR4Zu9N57OsPAr8KrGmY/0qblUrVQ9idXagJmsLuyVc700IKD2DsqrBEYARTZLI86vM2HWY3a7AsdYxRoqxi5qj6FXJqROFVFVlw4YN9OjRo8IOb7fbze7du2nVqhU2m+Siqw26rlNcXIzD4ZAorihnn8tDjk+lkdVMivXsHEVysueIfK+JM11Vft+FqCpd12m0eiMAmwd2icoOhxMh50nddc/WvSw9ksPf2zRmQvOUar/+65wCbtiwk/YOG1/1izzbW9T+OTLk+21sKXKzuHsbBteLLfVcp683ke1TWd03lQ5O6cwR0UF+R8r7Ob+Iy37aThOrmZ/O7QzAzmI3A9dvw2pQ2DywC7EmI7k+Px2+3hx+ndWg4NF0BifGsrhHm4pWz59++J1NwULSG8/tTIMy98Nz96YzY9dh/twwkRcjzCw5nXRd5/zvt7G92ENLu4UJzVP4S6P6KIrCtiIXF3z/e7nXVFbAvS6Qc6Tm3bV5Nx9m5AGwvn9HWkRI0VaRDK+Prt/8Fv772dRm/CXCLCZRu6L5PKlO2yRZrRCi1oQisRI3E0IIURtKjgCSQdsiGoQOwxMdn+aVFE1RqdKZI6HZPlInRoioFmnmyM/5gRTU3WIc4dkS1jKZI1oHO3grmzmi6zo7ggXLAfa4POWWOZZWK/oGESqKwt9aN8KkwB6Xl4d+P8A7h7MB2FbojviaHnGO2myiqANOdOYIBM6L3iWOqd5yfIkaJFdoQohaUyenqQkhhKizSta8loteEQ2UYHjkhGuOSFqtqBQKjkQqyB6qEyP7TIjoFms6VnMkFMD+KRgc6RV/rCO2bHC6jSMQHMmqJDhSoGq4SlyU7I4QHDkaTKuVbI7OWa6XJifw5TkduCwpkK59yZFsfJrO83vSAejoLJ3VoEesdF6L0owlAiLmExgxe0nw2ANo75QsGqLmyH2iEKLWya2hEEKI2qCV6IKWvNfiTODRAyOapaM9uoTqJWtlYiO6rh8LaMl3kBBRLTRzRNXBHTxvN4SCIyVGqRsUpdT53MYR6KTN9h0r5P5HkZtndh/mf5mBFELpZeqU7HV5KStSQfZo085p45/tmgDwQ14RL+8/yh/FbuqZjSzuXjqlWE8Z2S/KKBkcMZ7Ab+KdTZO4NiWBp9o3PaHXC1GR6P3WFUKccepmhSMhhBB1lcwcEdHmZG/lw2m1JEVTVAmlB1HLzAny68dmCUlAS4jo5jQe+14tVFWsBoXfiwIpozqVqRdkMSh41cDZHUqr5dI0bt24EwWFr3ML8Gg6RgV+6N+JdG/p4Mi2ovKpqKI5rVZJjW0WesU5+Dm/mH/uOgzAuGYppFjN3NEkiYUHM5nSqqF0XotySh4RphM4PJxGIws6t6yp5ggRJsERIUStU2TuiBBCiFqglfi33KOLaBA6Dk90vEiofoVZOtqjSmh3lE2r5dWPfQvJPhMiuhkUBYfRQLGqUaRqFKteXJqGRVFoVaZwtNVgCNcmaWw108BiIt3rZ1V2QanlVB3ePpRFu2AKIIXA9/+mwuJy75/pC6bViuKZIyGXJyeE67E0tJgZ1SQJgL+3acyF9WIZUj/udDZP1AHVrTkixKkU/d+6QogzhkwcEUIIUZtKFr02SGBeRIGTPQo9Ur8iKhmDe7ZsWi1viQdkto8Q0S8mGBwp9KscCqbCauOwYirznWsr8bfTZOCj3u35Kb8IV7C2SAenjcMeH+O37GXuvnSuSUkE4MJ6sazKLuCA20eOz09isL6IT9PDabmifeYIwNUpCbyw5wiKArM7NCMmWKzebjTwpxJ1IYSoSNlzSojTSYIjQohaFLhBlJ9BIYQQtaHkzBG5BxPR5ERTjXq1wFFtNUhHezQJp9Uqs2NDM30UjtUlEUJErxijkaP4KVS1cEqt1AiFn0sGqJ1GI01tFpraLKWWyQkWaPfrsCw9B4B2Dhs7iz3sdXt58Pf9HPX4yfb5aWQNBESMCtQzG0/JZ6tJTW0W1vXvhN2g4DRFf3tFdCg5WURmjohoIsERIUStCd8uyu+gEEKIWqDJzBERZUJHoX6C82mluHd0qiitVmimj9WgoMg+EyLqxQTrjhSqGpsLXQB0jrGXW65kgLpkrZKSEs0mesQ62FBwLIVW11g7CWYjM3cf4aOMvPDjO10eAOqbTRjqyHdFUh1I/yWilwzxENFEvs2EELUmdLtYNy73hBBC1HUyc0REm5Otu+bRJK1WNAoVHi6bVitcI6aOdHYKcbZzmoLBEb/KpoJAcKRbrKPcciVnglUUHAGYmdqU94/mckVyAkYFusTY8Wo6mwpcfJ1bwNUpiVxYL5Zndh/h9yI3f25Yr2Y/kBBRSgYMiGgiwREhRO0J3jDKz6AQQojaULKjUkaoiWhyonXYjqXVkqupaBJOq0XkguwWSYMmRJ0QYwykiDri8bErOJujS4SZI/4Sp3plwZHusQ66lwmu2IwKC7u2KvXYJUnx+DQdWyXrEqKukysXEa3km1eE6Sea/FjUabW53+UIE0IIUZu0Er88MkJNRIPQYXii10ShNE1mKe4dVY6l1Sr9uFdm+ghRp4TSan2cGUh51dxmoX6E9FH+EqMvaiL4aVQUCYwIIcRpIt++ddTUqVNJTU2t8L+BAwdWeV35+flMnjyZH3/88ZS1d8WKFaSmpnLgwIFT9h4AkyZNIjU1lYULF1Zp+eHDh1e6HTdt2lRq2eHDh5f794mYO3cuqampJ/z6mvLFF18wZcqUWn9fuT0UQghRG0Lxf7ngFdHiZK+BQnV0THIxFVWMxynILmm1hKgbYoLFxb/PKwLgypSEiMuVrS8khBCi7pK0WnVYcnIyaWlpEZ8zm81VXs/WrVtZuXIl119/fU017bQoKCjg888/p3379ixevJhRo0ZVaZRop06deOyxxyI+16ZNm/C/Sy5T0fJ1zRtvvFGr7yeXkEIIIWpTqOaIDNoW0eZE+9XC9duksz2qmMI1R0rvWDUczJL9JURdEFNm9saVyQkRl5PgiBBCnDkkOFKHWSwWevTocbqbETU+/PBDAB555BFGjhzJd999x4ABA477upiYmCptx7Zt20b8t6g+uaEXQghRG0IdlQaZsyiizIl2q+lSvy0qhbpTy6bVUkPPyw4Tok4I1RwBsCgKnSPUGwEJjghxIuSnUESrszLLgK7rFPuKo+K/U13vYfPmzXTu3JmpU6eGH8vKymLAgAGMGjWK7777jhEjRgAwYsSIUmmjHnzwQe6991569OjBqFGjWL9+Pampqaxfv77Ue5RNMaVpGgsWLOCCCy6ge/fujB8/nry8vHJt++OPPxgzZgy9evWiV69eTJgwgf3794efD73fu+++y4UXXkivXr345ptvKvysy5cvZ8CAAfTv358WLVrw7rvvnthGi0BVVd555x2uvPJKunXrxgUXXMCzzz6Lx+MJLzN16lRGjhzJY489Rq9evbjssstQVRWPx8NTTz3FwIED6dmzJw8//HCp14X8+OOP/OUvf6F79+707duXKVOmkJ2dHX5+xYoVdOrUiaVLlzJw4ED69u3Ljh07Iu4rgAMHDjB58mQGDRpE586dGTBgAJMnTyYnJwcI7Lfvv/+e77//vtR+zc3NZdq0aZx77rl07dqVP//5z6xbt65GtqMuc0eEEELUIpk5IqKNHIpnporSaqkSoBWiTokxHesi6+C0Ya7gAqJsIFQIIUTdddbNHNF1nRGfjGBDxobT3RQAeqb05M1hb57wSHq/3x/xcaPRiKIodOnShbvvvpsFCxZw9dVXM2DAAKZNm4amaTz99NM4nU6mTZvGE088wbRp0+jXr194HZ988glXXXUVCxYsQNO0iO8TyZw5c1i0aBHjxo2je/fufPLJJzz33HOlltm9ezc333wzrVu3ZubMmfj9fhYsWMAtt9zCypUrqV+/fnjZtLQ0Hn30UdxuNz179oz4ntu3b2fTpk3MmTMHgGuuuYb58+eTmZlJUlJSpe3VdT3idgxtQ4Bp06axcuVK7r77bvr06cOWLVuYN28eW7du5bXXXgsv9+OPP2K1Wpk3bx7FxcUYjUbuv/9+1q5dy/3330+LFi1YvHgxH3zwQan3+uGHHxg1ahT9+/fnhRdeIC8vjzlz5jBixAiWLVuGzWYDAkGahQsXMmPGDHJycsJpv8ruK5fLxYgRI0hMTOSxxx4jNjaWX375hbS0NGw2G0888QSPPfYYDz30EBBIE9a2bVs8Hg8jR44kMzOT+++/n5SUFJYvX85dd93Fa6+9VqWZOFUht4dCCCFqQ2jmiCK/PCLKnOiAkXBarZpriqgBoRowWpndGoqVGGWHCVEnlJw50jk28qwRKB8IFUIIUXeddcEROHNS+hw8eJDOnTtHfG7y5MnceeedAEyYMIFVq1Yxffp0Ro8ezeeff86cOXNo0KABcCxFVNu2bUulizKbzUyfPh2LxQJQbsZIJPn5+SxatIhRo0YxceJEAM477zyOHj3K2rVrw8ulpaVht9t54403iImJAWDAgAEMHTqU1157rVSR8FtvvZVhw4ZV+r7Lly8nISGBIUOGAHDttdcyd+5cli1bxtixYyt97Q8//BBxOz7//PNcfvnl7Nixg2XLljFp0iRGjx4NwMCBA0lJSWHy5Ml89dVXDB48GAgEq5544gkaNmwIBII2//vf/3j88ce55ZZbwtvjyiuvZMeOHeH3eu6552jVqhUvv/wyxuAFWffu3bn88stZvnw5t912W3jZsWPHcsEFF5Rqa9l9tXXrVho2bMjMmTNp1qwZAP3792fjxo18//33QGB/h7Z9KK3YkiVL2LZtG0uWLKF79+4AnH/++QwfPpxnn32W5cuXV7otj0euIYUQQtSm0M+OzBwR0SJ0H3LCabVqrimiBoVmjpRNtaMG95h8BwlRNzhK1BwZkBBT4XI+ubEVQogzxlkXHFEUhTeHvYnL7zrdTQHAbrKfcLAmOTmZBQsWRHyuUaNG4X+bzWZmzpzJjTfeyCOPPMK111573GADQOvWrcOd7VW1ceNG/H4/F154YanHL7300lLBke+++46+fftis9nCszZiYmLo06cP3377banXduzYsdL39Pl8vP/++wwdOhS3243b7cbpdNK7d2+WLFnC6NGjMRgqziDXuXNnpk+fXu7x5s2bA4SDCZdffnmp5y+//HIefvhh1q9fHw6OJCQkhAMjEJhJAoSDNgAGg4FLLrkkHBxxuVxs3LiRO++8s9QslmbNmtGmTRu++eabUsGRSNuj7L7q2LEj//nPf9A0jT179rB371527NjBrl27KpxtBLBu3TqSk5Pp3LlzqeUuvPBCZs2aRV5eHvHx8RW+vqrk/lAIIURtCI3iPivzyIqodLLXQMcKsp9sS0RNMoTSalE2rVbg/0a5+hWiTrCViGReUUExdpCaI0KcCJnJLaLVWRccgUCAxGF2nO5mnDSLxULXrl2rtGzHjh1JTU1l8+bN5QIXFXE6ndVuU25uLhAIEpSUnJxcbrmPP/6Yjz/+uNw66tWrV+pvh6PyfbV69WqysrJYtmwZy5YtK/f82rVrw8GLSJxOZ6XbMVQvpexnMJlMJCYmUlBQUGpdkV6bmJhY6vGS68rPz0fTNF599VVeffXVcu9vtVpL/R1pe0TaV//617946aWXyM3NJSkpiS5dumC320u1t6zc3FwyMjIqnJGUkZFxUsERSQUhhBCiNmnhUdvyyyOiy4n2q+nhVHEimlSUViuU2k++g4SoG/5UP567myZxXmJsqVkkZV1YL5ZPM/NpZa/eYFIhhBDR56wMjpyNFi9ezObNm+nQoQMzZsxgwIABxMXFVWsdoRkuZeuPFBUVhTvnQ0GArKyscD0MOBY0CYmNjeXcc88NFw8vyWSq3mG5fPlymjVrxowZM0o9rus6EydO5N133600OHI8oWBARkYGTZo0CT/u8/nIyckpF/goKfRcZmYmjRs3Dj9ecns4nU4UReH2228vNzsFwG6vONdpRT744AOefvppHnroIa677rpwwOmvf/0rmzZtqvB1sbGxtGzZkmeffTbi802bNq12W4QQQojTRWaOiGhTYzNHJDwSVUIzQ8qOJg/dNUlaLSHqBpNB4R/tjn/P+3yH5vQ8mMUNDSvuCxBCCFE3yL3iWeDgwYPMnDmTG264gZdeeomCgoJSgQRjiaJjlQnVpzhy5Ej4sby8PHbu3Bn+u2fPnthsNj799NNSr/3yyy9L/d23b1927NhBx44d6dq1K127dqVLly688cYbfPbZZ1X+bBkZGaxdu5bLL7+cfv36lfqvf//+DBs2jDVr1pCenl7ldZbVt29fAD766KNSj3/00Ueoqkrv3r0rfG3//v0BKt0eMTExdOrUiV27doW3RdeuXWnXrh1z586tUq2Xsn766Sfi4uK46667woGRoqIifvrpp1LBrbLpxvr27cvhw4epX79+qbZ88803vPbaa1U+VioiqSCEEELUptAvnvzuiGhRU4eiHNLRJRT8KFukOfS3pNUS4sxSz2ziry0b0MQmM0eEqKpbGwf6ps6Jq36WGiFOJZk5Uod5vV42bNhQ4fOpqanYbDYeeeQR7HY7kydPJj4+nvvuu48nn3ySSy65hCFDhhAbGwsE0lPFx8fToUOHCtfXqFEj5s2bR0xMDIqi8PLLL5ea2eB0OrnrrruYP38+DoeD/v37s2bNmnLBkfHjx3PzzTczZswYbrnlFqxWK4sXL+bzzz/nxRdfrPI2eO+99/D7/RFnXABcc801LF26lCVLlnDPPfdUeb0ltW3blmuvvZYXX3wRl8vFOeecw9atW0lLS6Nfv36cd955Fb62RYsW3HTTTcyePRu/30/Hjh1ZuXIlv//+e6nlHnjgAUaPHs2kSZO46qqrUFWVhQsXsnHjRsaPH1/tNnfr1o1Fixbx9NNPc+GFF3L06FFef/11MjMzS6XFiouL45dffmHdunV06tSJ6667jn//+9+MGjWKsWPH0qhRI7799lteffVV/vKXv2A2m6vdFiGEEOJ0Cae0kY5JEWVOtiC7BPyiiyk0w75cWq3A/2XmiBBCiLNdrzgnG87tTH2zdEWL6CJHZB2WkZHBTTfdVOHz7733Hj///DPr1q3jhRdeCHeKDx8+nA8++IBp06bRq1cv2rVrxxVXXME777zD2rVr+fDDDyOuz2g08uKLL/Lkk0/ywAMPkJSUxMiRI9m1axe7d+8OL3fHHXcQHx/PW2+9xZtvvknPnj2ZMmUKjz/+eHiZDh068M477zB79mwmT56Mruu0b9+eefPmcdFFF1V5G6xYsYJ27drRvn37iM/37t2bpk2bsnTpUsaPH3/CMx9mzJhBixYtWL58Oa+++iopKSmMGDGC8ePHV1rsHeCxxx4jKSmJf//73+Tl5XHeeecxduxYXnjhhfAygwYN4vXXXyctLY17770Xs9lM586d+de//kWPHj2q3d5rr72WAwcOsHz5cv7zn//QoEEDBg8ezK233srf//53du7cSZs2bbjtttvYvHkzd999N0899RRXXnkl77zzDs899xzPPPMMBQUFNGnShEmTJnHHHXdUux1lSc0RIYQQtSn0uyMdkyJahIIaJxwcCb5QDunoYlQip9VSw3WPar1JQgghRNRpaJUBtyL6KLp+ouUATy9VVdmwYQM9evSosMPb7Xaze/duWrVqhc1mq+UWnp10Xae4uBiHwxGuUSJEyOYCF6qu095pw15Jgbsz2cmeI/K9Js50Vfl9F6KqNhcUM/THP2hoMbNhYOfT3ZwaI+dJ3fX37Qd49UAmf23RgIdbN6r26ydt28c7h7OZ2qoh97VseApaeGao7XNk5q7DzN6bzh1Nkniy/bF6Be+l5zB2y14GJcSwrGfbU94OIapKfkeEqJycI0IcXzSfJ9Vp29nZOymEOE0CsVgJmwkhhKgNUgxZRKsTHZ8mBdmjW9m9Gqo5It9BQgghhBDRSYIjQohaUyenqQkhhKizNElBJKJMKKghNUfOLBWlS1OD/zfKDhNCCCGEiEoSHBFC1Dq5PRRCCFEbtHC+f/nlEdHhZI9EqTkSnUL7o+yMIC00c0T2mBBCCCFEVJLgiBCi1oRvF+X+UAghRC0I9VPKBa+INic7c0REl4rSnIVmr0laLSGEEEKI6CT3ikKIWif3h0IIIWqD1BwRUaeC9EtVpUv9tqhWPq1W4BGj7DAhhBBCiKgkwREhRK2R0Y5CCCFqk6S0EdHmWPqlE3v9sZojckxHk4r2RmjmiNQcEUIIIYSIThIcEULUHsmTLYQQohbJzBFxppGaI9GpwoLsusz0EUIIIYSIZhIcEULUitIFKuUWUQghxKmnhTsm5XdHRIfwzJGTnE8rR3R0qXDmSPD/MnNECCGEECI6SXBECFErJDQihBCitoV+e2TmiIgWoUDdyRZkl0M6OpVNlxYK0EpwRAghhBAiOklwRITpJ5r8WIhqkvtDIYQQtSGU718ueEW0ONlroGM1R066KaIGVTQjSJXvICGEEEKIqCbXaXXU1KlTSU1NrfC/gQMHVnld+fn5TJ48mR9//PGUtXfFihWkpqZy4MCBU/YeAJMmTSI1NZWFCxdW+TU//fQTY8eOpV+/fnTp0oULLriAv/3tb+zfv79a7z116lSGDBlS3SbXmPXr15Oamsr69etP+XudyP480dBbbR07Qgghzjxa8NfHID3JIsqc8MwRqWERlSqaERSqOSKz14QQQgghopPpdDdAnLjk5GTS0tIiPmc2m6u8nq1bt7Jy5Uquv/76mmraaVFQUMDnn39O+/btWbx4MaNGjUI5TmfIunXruOuuu/jTn/7EjBkziI2NZd++fSxcuJAbbriBpUuX0rx581r6BGcPuT8UQghRG2TmiIg24WsgmbB9RqnoliO0myWtlhBCCCFEdJLgSB1msVjo0aPH6W5G1Pjwww8BeOSRRxg5ciTfffcdAwYMqPQ1L730Et26deOFF14IP9avXz8GDx7Mn/70J/71r3/x2GOPncpmnzUka5sQQojaFiqGLP2SIlqc7KF4rOaIHNTRqMKZI7XfFCGEEEIIUQVn5XWarutoxcVR8d+prvOxefNmOnfuzNSpU8OPZWVlMWDAAEaNGsV3333HiBEjABgxYgTDhw8HYPjw4Tz44IPce++99OjRg1GjRlWYtmn48OHh1wFomsaCBQu44IIL6N69O+PHjycvL69c2/744w/GjBlDr1696NWrFxMmTCiVyir0fu+++y4XXnghvXr14ptvvqnwsy5fvpwBAwbQv39/WrRowbvvvnvc7ZOZmRlxH6SkpPDoo4+WSk+m6zpvvPEGl156Kd26deNPf/oTr7/+ernXr1ixgksuuYSuXbty1VVXsWbNmlLPde3alR9//JHrr7+erl27cskll7Bq1Sp27drFyJEj6d69O3/605/46KOPSq330KFDPPDAA/Tt25fu3bszcuRItmzZUuFnGzJkSIVp10IpqjweD7NmzWLw4MF06dKFK6+8ko8//rjUejRNY/78+ZXuz7lz5/KnP/2JtLQ0+vbty6BBg8jLyyM1NZW5c+eWWnbJKy/RsUOHUo+tWbOGm2++mR49ejBo0CCmTZtGfn5+xM+Vn5/P1VdfzZAhQzh06FCFn18IIYTQwh2T0pEsostJF2SXQzqqhGuOlNmxoZojMnNECCGEECI6nXUzR3RdZ++tt+H65ZfT3RQA7L160eKdfx83/VNF/H5/xMeNRiOKotClSxfuvvtuFixYwNVXX82AAQOYNm0amqbx9NNP43Q6mTZtGk888QTTpk2jX79+4XV88sknXHXVVSxYsABN0yK+TyRz5sxh0aJFjBs3ju7du/PJJ5/w3HPPlVpm9+7d3HzzzbRu3ZqZM2fi9/tZsGABt9xyCytXrqR+/frhZdPS0nj00Udxu9307Nkz4ntu376dTZs2MWfOHACuueYa5s+fT2ZmJklJSRW29YILLuC1115j+PDhXH311fTr149mzZoBcOONN5ZadtasWbz55puMGjWKgQMHsmnTJp599ln8fj9jxowB4PDhw7zyyiv89a9/xeFwMHv2bO69915WrVoV/kx+v59JkyYxceJEGjVqxLPPPsuDDz5IUlISt9xyC2PHjiUtLY0pU6bQu3dvGjZsSHZ2NjfffDN2u52///3v2O123nzzTW677TaWLVtGmzZtyn22tLQ0vF5v+O/MzEwmTZpEnz59aNSoEbquM2HCBH7++Wfuvfde2rRpw2effcb999+P1+vlmmuuAeCZZ57hrbfeqnR/QiB4s2bNGmbPnk1ubi7x8fEVbveSvvzyS8aNG8dFF13ECy+8QG5uLrNmzeLgwYO8/vrrpZYtKiri7rvvJj8/n7fffpvGjRtX6T2EEEKcnUL9lJLvX0SLigp3V5VMxI1uZfdPqO6RfAUJIYQQQkSnsy44ApwxQ60OHjxI586dIz43efJk7rzzTgAmTJjAqlWrmD59OqNHj+bzzz9nzpw5NGjQAIC2bduG/x/6NwTqlkyfPh2LxQJQpULf+fn5LFq0iFGjRjFx4kQAzjvvPI4ePcratWvDy6WlpWG323njjTeIiYkBYMCAAQwdOpTXXnuNKVOmhJe99dZbGTZsWKXvu3z5chISEsIF0a+99lrmzp3LsmXLGDt2bIWv++tf/0pBQQHLli3j+++/B6Bhw4YMHjyY22+/ndatW4c/11tvvcVf/vIXHnroIQDOPfdcMjIy+OGHH8LBEU3TmDdvXjhYYbVauf3229mwYQMXXXRReJmxY8eGgy/5+fncf//9jBw5klGjRgEQGxvL9ddfz+bNm2nYsCFvvvkmubm5LFq0iCZNmgBw/vnnc9lllzFnzhxefPHFcp+tU6dO4X97vV7+8pe/kJyczPPPP4/RaOSbb75h7dq1zJ49m8suuyy8r1wuF88++yxXXHEFxcXFvP3228fdnxAI+kyZMoU+ffpUuq/Kmjt3Lh07diQtLS0cJLRYLMyZM4fMzMzwch6Ph3HjxpGens7bb79N06ZNq/U+Qgghzj4yc0REmxMdEBUSmpkgR3R0qWh/aDJzRAghhBAiqp1UcGTixIls2bKFVatWhR/bu3cvTz31FD/++CNGo5Fhw4bx0EMPhTvAITD6+9lnn+X//u//KC4upk+fPjz88MPhjuhTSVEUWrzzb3SX65S/V1UodvsJ3yQlJyezYMGCiM81atQo/G+z2czMmTO58cYbeeSRR7j22muPG2wAaN26dTgwUlUbN27E7/dz4YUXlnr80ksvLdWZ/t1339G3b19sNlt49ktMTAx9+vTh22+/LfXajh07VvqePp+P999/n6FDh+J2u3G73TidTnr37s2SJUsYPXo0BkPkDHIWi4UnnniCe+65hzVr1vDdd9+xfv16Fi9ezIoVK3j++ee5+OKL2bBhA36/n4svvrjU6x999NFSfycmJpaaxRHqwC8oKCi1XMkZMKEZJd27dw8/lpCQABBOLbVu3To6duxIgwYNwtvLYDBw/vnn8/7771e6fSBQh2X79u28++674Rkd69atQ1EUBg8eXGoG0pAhQ3j//ffZvn07GRkZ+Hy+4+7PkMr2VaSRjm63my1btnDPPfeUOg8uu+yycMAmZPLkyWzevJknn3wyPLtHCCGEqExo3qvMHBHR5sTTaslMhGhWdkZQqOaIUXaYEEIIIURUOuHgyMqVK/nss8/Co9gh0JE7cuRIkpKSePrpp8nOzuaZZ57hwIEDpdLjTJo0iY0bN4aDJmlpaYwYMYKPPvqoyql4ToaiKCgOxyl/n1PNYrHQtWvXKi3bsWNHUlNT2bx5c7mO7oo4nc5qtyk3Nxc41rkfkpycXG65jz/+uFx9C4B69eqV+ttxnH21evVqsrKyWLZsGcuWLSv3/Nq1axk8eHCl60hOTuaGG27ghhtuAALBm4ceeojHH3+coUOHhj9X2baVVbatoQ7/smnJSgYLQ+x2e4Xrzc3NZe/evRXOFHJVEux75ZVXeP/995kzZw6pqaml1qnrOr169Yr4uqNHj4aDM4mJiaWeK7s/Qyo/Zsp3A+Tl5aHreqk0ahVJT0+nc+fOzJs3j2HDhp3Q8SmEEOLsoskoexFlKqpNUVXHao7IUR1NQvujfFqtAJm9JoQQQggRnU4oOJKens6MGTNo2LBhqccXLVpEbm4uK1asCHciN2jQgNGjR/PTTz/Ru3dvfvnlF7788kteeeWVcId1nz59uOiii/jPf/7DuHHjTvIjiUgWL17M5s2b6dChAzNmzGDAgAHExcVVax0VdfQXFRWFO6pDnehZWVmlZlCEggshsbGxnHvuueE0UiWZTNU7LJcvX06zZs2YMWNGqcd1XWfixIm8++67EYMjGzduZNy4cTzzzDOlCq8D9O/fnzvvvJOnnnqKnJyc8LbKzs4uNcPp0KFD7Nu3j969e1erzdUVGxtL3759mTx5csTnK5rhs2rVKmbPns2YMWPKzRaKjY3F4XDw1ltvRXxtixYt+PXXX4HA/iz5ucvuz8qoqlrqb3dxcfjfMTExKIpCdnZ2qWU8Hg/fffddqdk0oVRs1113HbNnzy43a0cIIYQoK5Tv3yAdySJK1NSRKEd0dKm4IHvoO6h22yOEEEIIIaomcq6h43j00UcZOHAgAwYMKPX4119/Te/evUuNrh80aBBOp5OvvvoqvIzD4WDQoEHhZerVq8c555zDmjVrTqQ54jgOHjzIzJkzueGGG3jppZcoKCgoFUgwGo1VWk9otsORI0fCj+Xl5bFz587w3z179sRms/Hpp5+Weu2XX35Z6u++ffuyY8cOOnbsSNeuXenatStdunThjTfe4LPPPqvyZ8vIyGDt2rVcfvnl9OvXr9R//fv3Z9iwYaxZs4b09PRyr23ZsiUul4u33norYsH53bt3k5ycTL169ejWrRtms7nc51i4cCEPPPBAlbfhierbty+7d++mVatW4e3VtWtXVq5cybJlyyK+/x9//MGDDz7IoEGDuO+++yKus7i4GF3XS63zjz/+YN68efj9/irvz4rExMSU2/a//7oh/G+n00nHjh3Lre+rr75i9OjRHD16NPxYUlISqamp3H777bzzzjts3LixSm0QQghx9gp1VJ7QBa8Qp9AJp9WS2VBRqcKaI8H/S80RIYQQQojoVO2ZI0uXLuW3337jww8/ZNasWaWe27lzZ7k6AUajkaZNm7J79+7wMk2bNi3Xmdu8eXM++OCD6jan3Kj0ss/puh7+70yi6zper5dffvmlwmVSU1Ox2Ww88sgj2O12HnroIeLj47nvvvt48sknufjiixkyZEg46LF69Wri4uLo0KFDeHuV3G7t27enUaNGzJs3D6fTicFg4OWXX8Zut4e3scPh4K677mL+/Pk4HA769evHV199Fe78Di03fvx4br75ZsaMGcMtt9yCxWJhyZIl4WLxJfdZZfvvvffew+/3c9lll0Vc5uqrr2bp0qUsXryYe+65p9RzcXFxTJ48mccff5xbb72VG2+8kWbNmlFQUMDnn3/Of//7X5555hkgMCNmxIgRvPHGG1gsFs455xw2btzIokWLmDx5MoqiRDzWyn6GSJ+pKo+NHDmSlStXcvvtt3PHHXeQkJDAxx9/zNKlS3n44YfLvTYnJ4exY8ficDgYM2YMmzZtKhUAat68Oeeffz59+vRh/PjxjB8/ntatW7Np0yZefPFFzjvvvPAsoHHjxjFnzhzsdnuF+zPS8QJwwQUX8NFHH9GtWzeaNGvOW8uWc2T//lLL3nPPPYwfP57777+fa6+9loyMDGbPns3QoUNp164dmzZtKvVeEyZM4OOPP+aRRx5hxYoVmM3miMdGtKpoW1Xn9bquo6pqpd9/QtRVoeNajm9RE/zB3z6FM+uYkvOk7gr9/mu6dkL7TwtdR2gn9vqzRW2fI5quBf+vl3rP0HcQZR4X4nST3xEhKifniBDHF83nSXXaVK3gyMGDB3nqqad46qmnItZeKCgoiFgHwOl0UlhYGF4mUr0Fp9NJUVFRdZoDEO44rYjJZMLlckWcGVCXqapKRkYGN998c4XLLFq0iI0bN7Ju3TpmzpyJ2WymuLiY6667jpUrVzJt2jQ6duxIkyZNGDZsGO+88w5r1qxh6dKl4e1VXCIFEsCsWbN49tlnmTRpEvXq1eO2225jz5497NmzJ7zsHXfcgcPh4D//+Q9vvvkm3bp14/777+fJJ5/E7XZTXFxMs2bNeO2115g3bx6TJ09G13XatGnD888/z6BBgyguLsbj8QCBFEtl2xGybNky2rRpQ9OmTSMuE/p8S5cu5fbbby8XlLvqqqto0KAB7777Ls8//zy5ubk4nU66dOnCyy+/TJ8+fcLrHT9+PLGxsSxbtozXXnuNxo0bM3nyZG644QaKi4vDwbiS7XC73QB4vV6Ki4vxer3hx0PLRfqcZV8XGxvLwoULSUtL47HHHsPr9dK8eXOmTZvGNddcU257/frrrxw8eBCA2267rdx2efzxx7nqqquYM2cO8+fP56WXXiI7O5uUlBRuu+02Ro8eHW7L8OHDMZvNle5Pn88X8Xj561//isfjYdasWRiNRvpffAm3TbyXl/75RHjZfv368cILL/DKK68wYcIEEhMTGTZsGGPHjq1wm02ZMoW//vWvzJ8/n7vvvjvisRHtKqsTUxmPx4PP52Pbtm013CIhosvxft+FqIrdXgUwUFRQwIYNG053c2qcnCd1T7o7cExmZmSyoTCj2q/PLTYACgcOHGDD0f013r4zTW2dI4c9gf2anZPNhg1Z4cfTXcH9nZ7OhrwjFb5eiNNFfkeEqJycI0IcX10/TxS9isOXdV3n9ttvp169esyePRuAqVOn8v3337Nq1SoAunTpwp133sn9999f6rW33HILDoeD119/nTvuuAOXy8WiRYtKLTN79mwWLlxY5Q2qqiobNmyga9euFaY0crvd7N27l1atWmGz2aq0XnFydF3H5XJht9ulUKQoxafpbC0KBH26xVZcfP5Md7LniNvtZvfu3bRo0UK+18QZSVVVNm3aVOnvuxBVtTQ9h/t+P8CFiTH8u2ur092cGiPnSd317J50Zu87yshG9XiyXZNqv37Ub3v4v6wCZrVrwm2Nyg9WEwG1fY68ciCT6bsOc21yPGkdm4cf/9v2g7x5OJsHmqcwqWWDU94OIapKfkeEqJycI0IcXzSfJ6G29ejR47htq/LMkXfeeYfff/+dDz74AL/fDxybFu73+zEYDMTExESc/VFYWEiDBoGLwZiYGDIzM8stU1RURGxsbFWbE2Y0Giv8kEajEUVRwv+J2iPbXJSjHEuSLcfGiZ8joddV9t0nxJlAjnFRI4LfswbFcEYeT3Ke1D2GUGXu4G959QVebzScmcd0Tautc8QY3K9Kmf2iK7K/RHST3xEhKifniBDHV9fPkyoHR/73v/+Rk5NTqpB6SOfOnZk4cSKtWrVi3759pZ5TVZUDBw5w8cUXA9CqVSu+/vprNE3DYDhWHnPv3r20adPmRD+HEKKOkLCIEEKI2hJKqmqQHx8RJZSTvBIKTfmXQzo6lU3KoAb/NsoOE0IIIYSISlUOjkyfPr3crJB58+axefNmFixYQEpKCoqi8Prrr5OdnR2uSfL1119TXFzMwIEDARg0aBAvvfQSa9euZfDgwQBkZ2fz448/MmbMmJr6XEIIIYQQ4iwX6qeU4IiINlXKa1zZ6+SYjiqhoFfZ/RoK0Bpl1rQQQgghRFSqcnCkdevW5R5LSEjAYrHQtWtXAG699Vb+/e9/M2rUKCZOnEhubi7PPPMM559/Pr169QLgnHPOoW/fvjz00EM89NBDJCQkMHfuXGJjY7nllltq6GMJIYQQQoiznRbsqjRIT7KIEid7JOrHspSKKBKKfZQNjoRmjsj+EkIIIYSITlUOjlRFvXr1eOutt3jyySd58MEHcTqdDBs2jMmTJ5daLi0tjaeffppZs2ahaRq9evXihRdeID4+viabI4SIIic6QlIIIYQ4UZrMHBFR5mQnEOhIZ3tdEvoOkpkjQgghhBDR6aSCI08//XS5x9q3b88bb7xR6evi4+N56qmneOqpp07m7YUQQgghhKhQKKWNdEuKaHOyabVOtnaJODUqTqtV2y0RQgghhBBVYTj+IkIIIYQQQtQ9WjCljUFGbYsoEToS9ZOcUiuHdHSpaL+q8h0khBBCCBHVJDgihKhVcmsohBCitoT6KeWCV0QLqTlyZgoHR8rMHQkHR2q5PUIIIYQQomrkOk0IUTuk6IgQQohaVlMzR7z79uFLT6+JJgkBlO9Ery4JjkQXpYLvmNBelpojQgghhNDcbtTCwtPdDFGGBEdEmH6y8/vFGetsPDbOxs8shBBnmnBB9pNYh5qXx+5rrmXnJcPQvN4aaZc4e4VqhZxszRERncrun2NptWq/LUIIIUQ00IqK0Fyu092M007XdXZffwM7hlxE7vIVuH//nayF/5IBWFFAgiN11NSpU0lNTa3wv4EDB1Z5Xfn5+UyePJkff/zxlLV3xYoVpKamcuDAgVP2HgCTJk0iNTWVhQsXVvk1P/30E2PHjqVfv3506dKFCy64gL/97W/s37+/Wu89depUhgwZUt0m15j169eTmprK+vXra3S9X3zxBVOmTKnRdUa77du3c8stt5zuZgghhDhJ4YLsJ9Ex6d72O1pxMbrbTdHatTXSLnH2OtkJBKEZJ9LXHl0q2h/q/7N33+FRlF0Dh3+zLZtNbyShhZ6EXgMRkSKiKIoiKKJ0pItSBPxABF8RkSbSFWnKSxXBAiqIIEp7RWnSi7QA6T3b5/tjs5MsSSAJJQGe+7qULTOzz+zsZnefM+ec7GiJWhwxQRAE4QFnvniRpLVruTLqbU4/1pILPXqSvHEjZ558inMdniX+889J3rhRWf5hy6Cwp6djPnsWe2oqV8eN43zH54n9+GPHvzNmkPjlV1waMJDY6dNJ/mYjxmPHkGUZa1wcttRUrElJmM6dQ7ZayTxwgNSffibzwIGS3q0HgqakByAUX1BQEHPnzs33Pq1WW+jtHD9+nE2bNvHiiy/eqaGViLS0NLZt20aNGjVYs2YNvXv3LjDF3WnPnj3069ePJ554gsmTJ+Pl5cXFixdZsmQJnTt3Zt26dVSsWPEe7UHptGzZsjuynfvpTMcff/yRv//+u6SHIQiCINwmpazWbUxMms+fVy4nf/MNnm3a3PL7hSDcyu1mjojXYOlSUEN2Z/aaOFyCIAjC/cYaF0fKd9/jVqMG+pqR2BITSdm0icwDf2G9dg17VhaGJk1wq1GDtK1bMZ086bp+bCyZ+/cDYAPiZswEwKtVK9L/+IOYkaMImfgefl273utdKxHWuDjlstrPD1tSEgC25GQSPl+s3Je+c2eRtqsNq4h7rVqUnTFDfD8sJhEcuY/pdDrq169f0sMoNb7//nsAxo0bR8+ePdm7dy/R0dE3XWfhwoXUrVuXTz75RLmtadOmtGzZkieeeIKlS5fy3nvv3c1hC4IgCIJwlygN2W/jd4L5/Dnlcvq2XzgRWRO3yEiC3hyGe506aAICHKUYbTaw27GbzWT89huJy1dgPHkSrzZt0JYNxXT6DCoPD2ypqZhOniT43fF4t2t3ezso3HcKmkQvLNGQvXQq6HjYs/8KiZ4jgiAIQmlkvniRtG2/YL1+DUnvjv9rr6Ly9OTKW8MLNUmf9vPPpP38s+OKRgNWq3Kff58+JH/9NfaUFJd1rElJxIwcBcC1iZMeouBIPAC6ypWp/PV6bGnpqD09SN2yhZRN32K5ehV7WhqyzYa9CFk1lgsXwWoTgZHbIIIjD7ijR4/y8ssv8+yzz/LRRx8BkJCQQIcOHYiIiGDAgAH07NkTgB49ehAVFcWXX35J9+7dCQ4Oxmw289tvv9GgQQMGDhxIjx49WLFiBU2bNlUeo3v37gB8+eWXANjtdhYsWMDatWtJSkqiefPmNGnSJM/YTp06xYwZM/jf//4HQHR0NGPHjqVChQqAo0xUjx49mDRpEosWLSIlJYU5c+YUWDLs66+/Jjo6mmbNmhEWFsbq1atvGRyJj4/Hy8srz+1lypRh/Pjx+Pn5KbfJsszy5ctZs2YNV65cITg4mK5du9KnTx+XP0IbNmxg0aJFxMTEULlyZUaOHEnLli2V+9577z2WLl3KlClTOHXqFGXLlmXMmDFUqlSJSZMmcfDgQcqUKcNbb73FM888o2w3JiaG6dOn8/vvv2Mymahfvz5jxoyhZs2a+e5bmzZtuHLlSr73/fLLL5QvXx6TycTs2bP54YcfSEhIoHLlygwaNIinn34acBzb/dmR/vDwcOXYJycnM3PmTLZt20ZaWhqRkZEMHz7c5fkODw9n4sSJHDp0iK1bt6JSqXmkfXt6vPEmU+fP4ZtvvkGWZdq2bcuECRNwc3NT1nv33Xc5cuQIP//8M+7u7jzzzDOMGjVKWQZg27ZtzJ8/n9OnT+Pt7U379u0ZMWIEBoMBgDlz5vDtt9/SsWNHVqxYgU6n44cffsDNzY158+bx008/ERMTg06no169eowePZrIyEjmzJmjZGSFh4czdOhQXnjhBR5//HGmTJlCp06dlDGMHTuW/fv3s337duX5uvF9s3Tp0nyPgSAIgnD33YmeI6Zz5/Pedvw4lwcOAsAtIgJLTAz21NR810/dvDnf26+OG49HVBRqX9/bGJ1wv7rtzJFbLGe+fIXU779DHxmJZ/b30DvNfOkS2uBgJJ0OW1oa6ny+U5c29owMVB4e2DMysCUnoylbFkmSMJ48RcLnnxM4aCC6KlWwJScTv2AB+vAIfDq9UOgJhxuPq10pqyUIgiAId4fp9GmuT/0YQ1QUgf1fv+Xyst1OwueLiV+0CDkz0+W+hEWL8q4gSY6zMzQaPB99FLfwcLTlyuJWtSqpP/6EPS0NXaUw/Lp2ReXhQdKq1bg3bIh77VoEvTEUW1oamqAgTjWLxp6Sgj09407t+n3FGu/IHNEEBqIyGFBlz135du6Mb+fOynKyxYLdZCL91x1kHvgT/x49sCUloTIYkG120rZuBQk0/v5oK1ZENprwiG5WIvv0oHgogyOyLGM122+94D2g0aluK7pnzRWVzU2tViNJErVr1+b1119nwYIFdOzYkejoaCZMmIDdbuejjz7Cw8ODCRMm8P777zNhwgSXoMeWLVt47rnnWLBgAXZ74Z+v2bNns2rVKgYNGkS9evXYsmULM2bMcFnm/PnzdO3alSpVqjB16lSsVisLFizglVdeYdOmTQQEBCjLzp07l/Hjx2M0GmnQoEG+j3n69GmOHDnC7NmzAXj++eeZP38+8fHxBAYGFjjWVq1asXjxYrp3707Hjh1p2rSpEpzp0qWLy7Iff/wxy5cvp3fv3jRv3pwjR44wffp0rFYrAwYMAODq1at89tlnvPnmmxgMBmbNmsWwYcPYvn27sk9Wq5WRI0cydOhQQkNDmT59OqNGjSIwMJBXXnmFgQMHMnfuXMaMGUOjRo0ICQkhMTGRrl274u7uzrvvvou7uzvLly/n1VdfZf369VStWjXPvs2dOxdzrsax8fHxjBw5ksaNGxMaGoosywwZMoS//vqLYcOGUbVqVbZu3crw4cMxm808//zzvPfee7z99tsAvPfee1SrVg2TyUTPnj2Jj49n+PDhlClThq+//pp+/fqxePFilwDJtGnT6NChA3PnzmXb9u18tWIFh/fupV7NSKZPn87BgweZM2cOlStXpl+/fi6voXr16vHJJ59w9uxZPvnkE+Li4pQMn++++45Ro0bx7LPP8tZbb3HlyhVmzZrFmTNnWLp0qfKeiomJYefOncyaNYvk5GR8fHwYNmwYf/75JyNGjKBixYpcuHCB2bNnM3LkSH744Qe6dOnCtWvXWL9+PWvWrCEkJKTA91l+ivu+EQRBEO4851nbqmJ+17KbzWQdPAhA2MqvsCYkkPLtt6Rv+0VZxnTiRL7rejRvTsYffwDg/cwzuDdsgGw0ARA7bRr2tDSMJ07i0axpvusLQn4KG1S5On48mXv3AlB9129ogoLy357dTtysT0hatQp7VhZ+r7yCb+cXcataFSm7TK/jR7qZrL8OoK9Th/RffiFp3TqMhw7jXq8ePi924tqE9/Dp+ByhH3yA8cRJ1L4+aENDke12LFeuYLl8GbfwcGSjEVQq7OnpuIWHI0kS9qwsJI0G2Wolfv4CdJUr4/PC82T8/gfG48cJ6NsH6/XryHYZe2YGWX/9hVu1amgrVMCWmIjp1ClQa5A0alCpyPj9Dyzx8egPHya+0wtk/rEb35dfwnLpMgmffw6AytNTOSvTq107zJcuYTp+nNTvv0dfsyaSuztZ2XW8r02ciPczz6CrVAn/nj1Qubsrz58lJoakNWtJzLTBI48rPWGcchqyS9jNZpLXrcO7XbsCj4cgCIIgFIUtPZ0Lvftgi48n4/ff0YfXUE6KsFy9SuoPP6AtXx7j8ROYz59H0qgdZbEK0QTcr0d3/F99FV1YGLbUVCSdDpVe77KMoVGjPOv59+iuXFa5uyufm9oyQZhSUrCl5mSSqIMKnq+7H1mTkjAeOYLHo48iqVxPz7LFOzJHNLfYZ0mrRa3V4vNsB3ye7ZDnfvfate7cgAXgIQyOyLLMhml/ce1cyq0XvgdCq/rwwqiGxQqQXLlyhVq18n9TjB49mr59+wIwZMgQtm/fzqRJk+jfvz/btm1j9uzZBAcHA1CtWjXlX+dlcPQtmTRpEjqdDqBQjb5TU1NZtWoVvXv3ZujQoQC0aNGC2NhYduVqYjp37lzc3d1ZtmwZnp6egCNzpG3btixevNilAXi3bt146qmnbvq4X3/9Nb6+vkpD9BdeeIE5c+awfv16Bg4cWOB6b775Jmlpaaxfv17JkAgJCaFly5b06tWLKlWqKPu1YsUKXnvtNSVY8MgjjxAXF8f//vc/JThit9uZN2+eEqxwc3OjV69eHDx4kMcff1xZZuDAgUrwJTU1leHDh9OzZ0969+4NgJeXFy+++CJHjx4lJCSE5cuXk5yczKpVqyhXrhwAjz32GE8//TSzZ8/m008/zbNvuTNKzGYzr732GkFBQcycORO1Ws0ff/zBrl27mDVrlpIp0qJFC7Kyspg+fTodOnSgWrVqyvFxlnBbu3YtJ06cYO3atdSrV08ZS/fu3Zk+fTpff/218rjVqlXj/fffB6Buo8asW7cOq9XC9OnT0Wg0PProo/z000/89ddfLmP39/dn4cKFaDQaWrZsiUqlYsqUKbzxxhtUqVKF6dOn06JFC6ZPn66sU6lSJXr16sXOnTtp1aoV4AhEjRkzhsaNGyvPQ0ZGBuPHj1f2OSoqivT0dD766CPi4+MJCQkhJCTEZZ8vX75cwCsorxvfN4IgPLxkm42sv//m2uQPMTRoQOCQwWhyBf+Fu89ejBJE6bt2Yc/KwrtdOzL++AN79tlu7g0aIKlUSiks8+UrmC/8S+a+/WhDQ4hfsBBrbCx+3boROHgQmsBArImJjjPDbvghmfLD95iOHUe2mPMbgvAAU8pqFTN3RGnIfpMXtSzLGI8cUa6fbvEYXk+0peyMGah0Oqzx8aj9/ZFUKlK3bFECBgBJX31F0ldf4RYeTqU1q0n88kviFyzMc2apU9ahQ2QdOgRAyqZvSdn0baH3RVu2rCODIyXv77Jr772HbLEAEDdzZqG3mZsKSPzMsW/XJriWyc1drkIpB5LNeOyYy3XZYiElu4Fs3CefoAkKQluhAmpfX7L+/htbUhKW5m2ygyOubEqAFlK/+47r//mAtJ9+JmzF8mLtkyAIgvBwk61WbKmpaPz9AUhcvlyZdAeIGT+eSitXYjx+gqvvvltgZrOTJiiIoJEjsCUk4tftFTL27SNzz168nmyHoWFDZTm1t/dtj13l6cgwzX1ikdrH57a3W5rEzphByvqv8X35ZcoMf4sL3Xtgz8xEW7485nOOUr3qm5zALZSMhy44Ag9OQ7ygoCAWLFiQ732hoaHKZa1Wy9SpU+nSpQvjxo3jhRdeuGWwAaBKlSpFnuA9dOgQVquV1q1bu9zevn17l+DI3r17iYqKQq/XK2fle3p60rhxY3bv3u2ybmRk5E0f02Kx8O2339K2bVuMRiNGoxEPDw8aNWrE2rVr6d+/PypV/gU1dDod77//Pm+88QY7d+5k79697Nu3jzVr1rBhwwZmzpxJu3btOHjwIFarlXY31AYfP368y3U/Pz+XLI7y5csDjmbxueXOgHFmlDgDDQC+2SU2UrM/yPbs2UNkZCTBwcHK86VSqXjsscf49ttb/wgdN24cp0+fZvXq1fhkf/js2bMHSZJo2bKlS2ZEmzZt+Pbbbzl9+nS+z/2ePXsICgqiVq1aLuu1bt2ajz/+mJSUFOUxcu+nSq3G29eXqhGRaDQ5f3p8fX3zPD/PPvusyzJPPvkkU6ZMUUqwXbt2jQEDBrg8fpMmTfD09OSPP/5QgiPg+vrR6XR88cUXAFy/fp3z58/z77//8uuvvwK4ZNoUV3HeN4IgPHhkWebK8BHKpJvp+HGS1q0joFcvgkYMFzVh75GiZo5YYmK49Hp/ADQrvyIj+7uL1xNt85z9pStfDl35cnhml/v0eeEFkGWXs8qdP1xvpNI5ykTKJlMR9kZ4EDjf+8Uuq6UE/Fxf0/GffY71+nWC3xmL5dp17BmuJSvStm7jQrdXMR49CoAmOBhNSDDGQ4eVZXRhYUju7phOnMB08iQn6+efsa0JCcGzxaNk7NmLpQgnkNzIEhNT4H3OwEhhuNWo4cgeKQRNaCge0dEYGjfm6rhxN23+4t+zJxl79uTZtjUuzqWpKoBUwBGVlbJaEuaLlwDI3L8f47Fj6AsojSsIgiA8mGRZxnL5Mtry5XO+D+SqNiFbrcgmE5bLl1F5eGC5dg3jsWNYr8eS+eefaAICHNdjY/Hp2BGP5o+QsnETAMHvjifpv6swnz3L2XZPujyuOjAQW2IimoAAvJ9+Gm3FCrhVrowhKgrsdqRc8xderVrhlWs+5U5SZZ94azqX08/Pnlb43hr3g5T1jpOFk9esQRMYiOn0aQAsuUreawJF9mhp89AFRyRJ4oVRDR+Islo6nY46deoUatnIyEjCw8M5evRonsBFQTw8PIo8puTkZCBnct8p6IbU8eTkZDZv3szmfOpw+98wkeDsIVGQHTt2kJCQwPr161m/fn2e+3ft2qX0/ChIUFAQnTt3pnN2nb+9e/fy9ttvM3HiRNq2bavs141ju9GNY3Ue2xvLKzmzMXJzzzWZcqPk5GQuXLhQYKZQVlZWget+9tlnfPvtt8yePZvw8HCXbcqyTMNcZwPkFhsbm29wJDk5mbi4uALHEhcXpwRH8ttPt5vsp5Mzq8nJGUBKSUlRjsWkSZOYNGlSvuPO7cbX8a5du/jwww85d+4cHh4eREREKMdNLm531Js8niAID6fk1atdzkbWlA3FGnOVhM8/R1s2FL9XXinB0d098Ys+w3zpIiHvvosqV5+oklLUniNJq9col+MXLlJKDhia3rqO743ZITfj/BFqF8GRh86dCovm3o7x5CkluyJp5Ur8+/QBQFe1Kh7R0SR99ZVjuezACID1+nWXkhqVN21En/098dr775P031XKfQH9swOGZcqgCwvD0DQKlU6H3WjkTJvHsSUmEjLxPeyZWcR+/DE+zz+PrmoV4mbkZHxoK1ZEGxKC+fIl5Cwj9vR0ZIsFryfa4hYRgT48nGvv/we1nx+STofl0iXKvP02SWvXYDx0GLeakY4sLA8P3GvXwXj0KFlHjhD21Ve4VamMLMtcmzgJa1wcvp07Yzp/nss2K0EXLuL/Uhcy9u3HrUZ1lwkft6pVSN64keRVq5Xbqv/xO+Zz58jYv5+Afv0oo3kbbDZkWUbSaLAmJGCNi8N8/l+s16+j8vBAXzMSZi8E8sZabM6/QVJOrXFwnOlbdurUQh1rQRCE+4Esy3nmtvK7rSDmf//FlpHhKG1Yik8iunGf0nftwnTyJN7PPkfWXwewpaXhVrkyCUuX4Va1iqO/hJc3aj9f4mbOckyWSxJu4eHYUlMcTbotFlSensh2e4GZmjdK2bSJlE2blOsezZrhXr8+F7q9imwyIel0yLKM/6uvEjx2TJGOxd2i9nLMD1mu5JwckV/26P1MV6WKkiESP29evsu4Val8L4ckFMJDFxwBx4S11u3haou3Zs0ajh49SkREBJMnTyY6OhrvIqbFFTTRn5GRoUwIOxuYJyQkuGRQOCe0nby8vHjkkUeUMlK55c4YKIyvv/6aChUqMHnyZJfbZVlm6NChrF69Ot/gyKFDhxg0aBDTpk3L0+S9WbNm9O3blylTppCUlKQ8V4mJiUqpLXD0tLh48SKN8qmzeCd5eXkRFRXF6NGj872/oEyF7du3M2vWLAYMGJAnW8jLywuDwcCKFSvyXTcsLKzAsVSqVMmlpFVuzmyZ25GUlORyPT47TdTf3185FqNHjyYqKirPuj43Scu8ePEiQ4YMoW3btixatIgKFSogSRIrV650yWy6kfO1b7PZXG7PLOQXF0EQHi7JX3/NtckfAhAwaCABvXqh9vEh7tM5xM+fT8qmbx+44IgtNZVrEyeSunkLABo/P8qMHFnCo8o5O191i9+CtvQMrNeukv7rduW2jFyfC4ZG+Z9IUFxSduBINhf+7HjhwVLc8zFyN2S3xMSgCQkh6b//dVkmcckSwFGTOmT8OILH/R9pP/2ELTUV05kzJK34MmdhlYrQDz5QAiPgCIaYL1/GnpmJ52MtCejXN0/mFDgCgmErlmO5fl3JoPJq9wTakBAkjYaAvn0dJbpqhCu9dexGI8gy1uvXsSUn455dwhTA8/HHkSTJkTWiUiGp1fi+2Cn/50F2lheTlH9DJ01U7jfYHuPCwYOE9O2LWq3GPVeGtpN7vXrYjSYlOKLy8kLt748hIABDkyY5C6rVSjBKW6YM2jJlcM91kpDp/HnlgBZUVkstSVhzlT1J2byFoBEj0QaXyXf/BEEQSiu72Yyk0SCpVJjOnSPr8GGssXEkfPEFQYMHoa9VC9Rq4ucvIGPXLtSBgah9ffB64gn0NWsiG41Y4+JReXogWyxoy5bFGhfHtYmTIHuuKWjkCDxbtsQSE4NK737b/dlku53EpUvRliuPIaoJSBKa7LmrPMuazZgvXkQVGor266+JWboMXYUK2NPTMF+4SGZ2OXZ1QAAqTw8sFy4CEDt9Rp5tpW/fnuc2x4PIeXrW5S73mB/fLp1J//0P1H6+mM+dR9JoXNbRVa6MpFJR5YcfMJ08gaFpU0fD71yfkyVN5eEMjuRkUchGI3azGdUDUn3DdsNclsrDg6rbtqJydydp9Wp0YWF43qXMHKH4HsrgyMPmypUrTJ06lc6dOzN06FCefvppJk+ezNTss5XU6sIFipxZANeuXVNuS0lJ4ezZs9StWxdwlFHS6/X8+OOPLhPXzrJFTlFRUZw5c4bIyJwSS7IsM2rUKMLCwm5ZSsspLi6OXbt20a9fP5dm8k5PPfUUGzZs4Pr163myESpVqkRWVhYrVqwgOjo6T+mt8+fPExQUhL+/P3Xr1kWr1fLrr78q/SsAlixZwubNm/n9998LNd7iioqK4rvvvqNy5cou2RgffPABFosl3wyKU6dOMWrUKB599FHeeuutfLe5ZMkSZFlWjh84gk1bt27lww8dE3sqlcolIBYVFcWOHTsICAigbNmyyu0LFy7k+PHjBQZNnArzkbx9+3b6Z5+lCPDTTz8hSRLNmjWjbNmyBAQEcPnyZaWvDjgyRkaPHk3Xrl2pWLFivts9evQoJpOJ/v37uyzjDIw4f2jf+FpwPufXc51habFYOHz4cIEl2wThYWBNSkLt45PvpNnDyhITw7VJ74PVitdTTxH0xhvK8+P78kvEz59P1qFDWGJi0Ob6G3o/s2dmcmnAQLL+/lu5zXjyZAmOKIfd+Xe9gE+fzD//JGbcOOWHbX70tWujucO1gSU3xw9AUVbr4XO7UxPOyfesw4c4M3Qg3s88Q+oPPwDgXr8+WQcPAo4a3gGvv+54TEnCO9dJMv49emA8+g+eLR5FlU/GqzYkhIqffVao8bhVq4Zbrp6FulwnyUgqFf49ergs78yw0lWqlGdbSqAjuxH8zdypSR5dpZyTgXQVKxZru5JWi5R9YG4MjuTOXrPFJziW1+uRjUauvvMOmpAQvB5vg1d2b8Ib2dLTscbGoQuriFTAbzbZZkM2mRyTYIDleiwZf/yBV7snUOeTxe1kOn2azAN/4dulM5JajeXqVdTe3sprIuvwYSQ3N9wqVybxyy8xNGqEOjCItJ9/Rl+zJtpyZZF0biR/vR6PZtEYGuYtw2Y3mZCNRtQ+Psg2G6aTJ5EtFozHj+PzwgvIRiOSmxtJK/9Lxr69lP3oIzT+/sg2Gxm79yBptag8PXGrXg3j0aPoa9fGePgwWYcPo6taFcuVK1hiYrBeu07G3r3oa9cioHcfzBcuoK9ZE314DeyZmZj//dclECcIpVXc3Hkkr1+PZ8uWeLZqiedjj4FKRcqGDegqVkRXqRKaoMKV5LHGx2NLSsJ09iyebdrkmXw2njqFNjQUSaMhfedODE2bovHzI/Ovv7BnZODZogWy2QxaLfaUFJJWr0FXuTIx77yDV5s2eD3ZjitvDHPZ5vUpH+UZhy0+Hlt8PAlnzhb+eZgxMyf7UJLwbt8e1Go8WzyK6fRpfJ5/nrRtv2A6ewbrteu4N2qIf48eqD08XEpEAZjOnCFj3z5ip2XPU2g0qH19CejVE+OJk1jj4tCWLYuk02H85x+XLEstUFDIwpaQgC0hodD7BODeuBHe7dohWyxIOjdUHh7oKlZAFxaGNTERe2oqmQcOoK9ZE49mzbAmJWG5fBnZYsGjWTMl+8OenRmStPK/XP/gAzyaN1d+bzjLvpZGzrJaN5bVtKekoCrk67o0k222PJkwQW++qQTiAnr1KoFRCYUhgiP3MbPZzMHsH0D5CQ8PR6/XM27cONzd3Rk9ejQ+Pj689dZbfPjhhzz55JO0adMGLy9HU6QdO3bg4+NDREREgdsLDQ1l3rx5eHp6IkkSixYtcikH5eHhQb9+/Zg/fz4Gg4FmzZqxc+fOPMGRwYMH07VrVwYMGMArr7yCm5sba9asYdu2bfk2Fy/Ixo0bsVqtPPPMM/ne//zzz7Nu3TrWrl3LG2+84XKfj48PY8aM4b333qNbt2689NJLVKhQgbS0NLZu3co333zD9OnTkSQJf39/evTowbJly9DpdERFRXHo0CFWrVrF6NGj7/oEea9evdi0aRO9evWiT58++Pn5sXnzZtauXcs777yTZ/nk5GQGDhyIwWBgwIABHD161CXAUbFiRVq2bEmTJk0YPHgwgwcPpmrVqhw+fJhPP/2UFi1aKCXEvL29+fvvv9mzZw81a9akU6dOfPXVV/Tu3ZuBAwcSGhrK7t27+fzzz3nttdfQFuIH7a0cPHiQUaNG0bFjR06cOMGcOXOU4wMwfPhwJkyYgFqtpnXr1qSmpjJ//nyuX79eYLkvgFq1aqHRaJg2bRp9+vTBbDazYcMGduzYAeRkgjizU77//nvq1atHhQoVaNCgAV9++SVhYWH4+PiwYsUKjEbjLcu+CcKtZB08SPLXXxPw+uvoCgjslTTZbid9505sCQnYzWbsaemk7/qNrD8PoK1QAUPjxo66t7d5VldpZktP59qE9/Bq94TLJOON4ubOQzabcW/ciHKzZrpMtGmDgzE0bkzmn39yacAAys+fjy7779qdJNvt9zRgde0/H5D199+oPD0JfmcshiZNUN+iDOW94vzkk+x2Mvbuxb1ePaUniLMvzI29A9yqV8N0+oxyvdwns+74uJwTFPId6HUl3F9uOziSPdme+vUGx7/ZgRGAiku+IP3338k6eAj/nj3Q3nBikJOufHmXIMbDTFOmDNry5bHGxlKmgAztW5E0mgJ7jjgDtJm7dmH85x8AAvr2JX7ePDKy+yymbNiAe8OGlJ87B9OJE1wd/y7u9euhq1SJhKXLkLOy0JQNxa1qNQyNG+Netw5qPz/02b/Zrk2cRPL69QT07YO2XDlip8/AnpHB1f/7P/S1aqEtG4rKwxO/7q9hPPoPqd9/j91kwnjY0W9GNptQeXtz9f/GOWrPa7V4tGhB+s6dSG5u+Dz3LMm5Sg7mJ/7TOVT/43c02aVwTWfOkLJxI4lfrUQ2GvNd59rESY4zmnOlUZ1+pDlu1athvnCxWH8fM3b+RsbO35TrksHgeHy7nZBJk/B7+aUib1MQ7qWk1auxxceTvGYNyWvWoAsLw3L1qsv7wbNNGwyNG5Oxdw/68HAC+vfHGhdP5p//Q+VuIGn1aswXLyBnZin9pwxNmzqyFU6edJ281WjQhoZiuXQJSa/H0KgRGX/8AYB7w4aYzpxBzsrK0wcq9fvvSf3++zzjVwcGKs3Bnd+nVB4eBA4eTOaBA1gT4kF29JWSNBq0IcFk/vU3toQEdNWqYmjSxKXUIQCyTGp2OfbU774DIOHzxS6LZO7fT8KChUhaLWGrVqGPjCBhyRKX8o4KqxVbfHy+mR4FUXl44F6vHmp/f3QVK6Ly9kJfsyb2tDTcGzYk68ABMnbvQV+rFta4WNJ3/kaZkSPQVa2KLbuKijU2DkNUkwKD8M6gV+7MRW1wsMtnuXNdZ+lav1e7oS1bFn2t+6OHlSq7rBY3VOTI2LMHn+eeK4ER3Vm21FQl+8mn43O4N2qEb5cuJTwqoTBEcOQ+FhcXx8svv1zg/Rs3buSvv/5iz549fPLJJ0q5oe7du/Pdd98xYcIEGjZsSPXq1enQoYNSWuj7fD7kwJFh8umnn/Lhhx8yYsQIAgMD6dmzJ+fOneP8+fPKcn369FEmj5cvX06DBg0YM2YMEydOVJaJiIhg5cqVzJo1i9GjRyPLMjVq1GDevHk8XsCZU/nZsGED1atXp0aNGvne36hRI8qXL8+6desYPHhwniyZrl27EhYWxooVK5g5cybJycl4eHhQt25dli9f7pKN8vbbbxMQEMDq1atZvHgx5cuX591336Vr166FHm9xBQcHs3r1ambMmMHEiRMxmUxUqlSJyZMnK31Scjt58iRXslMVX3311Tz3T5kyhU6dOvHZZ58xe/ZsFi1aREJCAsHBwfTu3ZshQ4Yoy7766qscPXqU119/nSlTpvDss8+ycuVKZsyYwbRp00hLS6NcuXKMHDmSPtk1rvNX+PoRPXv25Pr16wwdOhQ/Pz8GDhzIgAEDlPu7dOmCh4cHixcvZs2aNRgMBho2bMj06dOVAEp+wsLCmDFjBnPnzmXQoEH4+PhQv359vvzyS7p3786ff/5JeHg47dq1Y9OmTYwdO5bOnTszceJEPvroI/7zn/8wfvx4PD096dy5M40aNWLdunWF3i9BuJElJoZ/uzrKKxlPnSJo2DDHl121Guv1WFRensgWC+YzZ/Dv1QvUaoz/HHNMvteri6RWY8/KImP3bjyaN8eeloY6IOCOTow7yiVNUn6U5NmHS5dIuXSJlG++IeT9Sfi99GBOPKRs+IbUzZtJ3byZuEqz8Wz5GNqKFfFq1QpN2bLYMzKwXrtGysaNAASPGpXvj5/QD/7D+c5dMJ0+w9kn2hE4eBDG4ycImfAu2tDQYo9PNptJ//137OnpXJ/8Id4dnyP4nXewp6dj/OcYppMn0NesidrPj/jPPkO2WCgzciS2xETc69bFePw4qVt+xOfZDrhVr+44E9lmu2WKu+X6dVKyf6iWnz8Pj3zKHZYk51nbpn+OcvH/RuDz/POU/WgKAOYzZ5TAiMcj0WTs3gMaDb4vd+X6Bx8o27gbk8iSsyG7WWSOPKyK3ZA9e02Vm+t70xDdDJXBgHe7dni3a3ebo3t4SJJEpXVrkc2WYpe4knKVBL6xf52zrFZKrjK2vl06o69di4TFX5B14AAAWX/9xelHcsr83nhWrTXmKtaYqy7l/twiItDXrqU0f01Y/EWesRn/+UcJyjg/n250/cMpLtdli0UpBSNnZt4yMOJ06fX+hK38irRfthMzalSh1smvvlzu4HS+NBr0NWtiPHwYlZcXhkaNUBkMGKKbkf7rDpcyNrlr91//4ANHlozIdhVKKVtKihJYcDJfuJBnufTt25XXecZvu0hYugys1ptuO3PfvvzvsFqxXLoEOMobOQMj4Pi7VFi6alUJ6NMX304vYDx+HPPly3i1bYvl8mUknQ5tcDABffOfK7CbTGTu3Yu+Vi00gYGYz5wl888/qbh0CZKbGxd79MwTnAHQ162LNjSUrCOHscZcdeyDxcK/nTs7AqOFKIGtr10b7/ZPkfbrr6h9ffGIjkZXMQxdhfIY//2X03Y7NStXRhcQgPomZem92rbFq21b5XrgwIHKZWfWgFvlO99nQpIkvNoUrqdwaVBQNmPM6DGo/fzxbPHoPR7RneUsqaXy9hZ9xe4zknwnOhCXAJvNxsGDB6lfv36BZaGMRiPnz5+ncuXK6IvQJFMoPlmWyczMxGAwlIqahkLpkWmzcTrDhEYlUcuz4Kbs4eHhDB06NE+mz4Pidt8j4u/ag8GWnkHcrFmYL1wg4zbL8klarcsPBp9OnQj9z/sFlt8oCtlm40KPnsrkjXujRqj9fJHUGrQhIbhFRqDSu5O2dSupP/yA2s+Par9sU0p7FEdhPt/vNfO//3L2qfaFXt7ribaUnzOnwPvTfv2Vy4MG57ndLTISz+aPYDx+Asv1a7hVqYrp7FlkqwV7egbudeogW61oy5bFdPIkklaLNqwi2GXSt2/P29BQq3X8WC7iVz1Ds2aYTp0ClQp9ZCTG48epuOQLJI2W1B9+wL1+PZJW/hddWBgZe/diOnkS90aNqLTyqyI9zr0w8cwVFl6K45XtW+i/zjE5GXniOACJ//0v19//D4boZoQtXerSwyD+88+JmzGT0I+m4Pv883d8XFcnvEfy2rUEvjGUoFwnJBRWaXyfCIWz6FIs752JoVOwH/Nr5t/f7WaeOXCKA6mZfLBkDs3/58g88GjRgpAJ796VTLT71b18j9iSk/n0rTFM7TmINn6e/Ld+Tpmx1vtPcDzDyPRPPqDRSUeQIuLwIaXsiy01lbhPZpO0Zo3LmbRuNSORNFq827fH7+WXuD71Y1K3bEHt6YktORl7PpN+KoMBe2YmgYMH4/XUkySvW0/m/v2Ycpc5zJWpoQkNRTYalckcnxdewL1eXeLmznOZoNWUKYP3009juX4Na1wc5aZO5eq7E5Dc3LDExDg+j9zdkbOy8GjRwiWAEzh4MEn//S+25GT0NWtSdvo0rLFxWK5exXrtKp5tHsd84V9kiwVbYhL2rCzHWezXrqLx8+PqexPRBJchsH9/bGlp2NPS8e7wDPoaNcg6cgRtaKhL2UPZbidm9Jh8z2gHCD90UDnj+mEnPkdKn8y//uZCt25ogoMp+/HH2DMyiF+0EOOhw3mWVQcEoI+MxHT6NNbs0s9u1athiY3DXkCDa5XBQNlpH5O2fTsp2dmH3h064F6/Pu716nJp4CBkiwW/rl3J+OMP1IEBuNetiz4iAm2FCqgMBuLnzMVuMjmy2GrXInnjRvy7d8ctV7/Z22VLTsYaF4db9eqO6+npIMuk//orV9+dQPD//R8e0c3QZvcPBcdJZvFz5pC2dVuB2w0aMcKRZVGzJpJGjS0pCX3dugX+JhfvkTsvecM3XP2//8v3PkOTJoR9mbcfri05GdO587g3qF/sOcaUTZvI/OtvNIGB+L3aDY2/P+bLl4HCnQRlS0/HnprqUhI5vwb3mX/+yYXXuqMLC6PqTz8Wa6z3m9L8PinK2ETmiCAIgiDcY8lr15K0cqVyXVuxItarV5HtdmVyRB0UiC0p+ZZngt14JlXKhg2kb9/uCFLkU0u+sIwnT3F1wrsYDx1GZTBQfsECPJrmnxXg1fZxsg4exHLlCum7fsf7yQfnrGVbegYXcmXFBfTri/HkKZfJn9wkd3eC8unzlJtX69ZU3vA15zu96HK76fhxTMePK9fNN9RmTs8uAejizz/zH4ebm9LPQu3r65g0y8hwpHvfIliSuXevctm5n+ef61jg8urAQELem3DTbZYU566qdDklH51lx4xHHROVhoaNANceBgG9e+Pdvv1dKz0kGrI/vO5UzxGy399hK7/C0KjRbW5VuC2aXH9fbswccf4Nyr7dEBXlUg9f7e1NyIR3KTP6bYzHjmG5fBnv9u3z9F0JnTTRpeH89Y+mkrhsGZqgIDxbtcK/dy+0oaHYEhPRlnPUmg8ZlzMBZUvPQGVwR1KpSN/1O6ZTp/Dv1RNUKqwxMZjOncfjkWgktRq/rl0dvTyux+JWozoqd/c8J1xUXOLIUpHtdpAk0n76mStvvZXz2ahSUWnNatzr1MG7/VNY4+PxiI4GwK1KFZdt6cPzrwAAjpMyNP7+qLMrILjcV6dOntsklYpy06dRdtrHWGNi0JYrhy0lhVNNm2U/EbY86wjCvZT688/oKlTALTwcOSsLSafDlpqK2scH8znH9z63qlWV79xebVpjS093lLYym1H7+yO5uSkZUPbMTFK+/Q5tuXJ4PNoce0YG5nPnSN+1C8+WrVC56VAHBpJ18KAjMBAejtfjjxMyYQJYrS6/Far+9COSRuPoDTVieL7jLzvVtafI3ejlo/b1Re3rm3M9O9vA57nn8H722XwnyPU1alB+zhzl5BMA/759KJOdxWaNiUFTtqzruqW0nPGDTOXp+ts0YNBAVHp34mbNIvN//+N0q9a4161L8Lj/Q+XmxvUpH5G6eTOyxUKFRQvxbNmySI8n2+0k/XeVS0Z4/Lx5qP39sSUmojIYqLzxG2wpKcR9OgffLp1J3bIFn44dsVy5gscjj6Dx9+dC796Yjh1HX7cuwaPfJnnDN6Tv3ElAv35k7t+P8ehRPJo3x/zvvwCos7OFhPuHCI4IgnBPiXwi4WFmPHYMu9FI2jbXs5pCJ76HW0QEKr0ey9WrSGo1ukqVMF++7Eibl2UkNzfUvr6OSQ9JxcU+fXCrUoXAQQMxnjhJ5p9/YktOxnjkCLbkZP59rTtlRgzHs0ULANJ//4PU77+nzOi3Uen1N83ukK1WLg8ahCUmBslgIPTDDwsMjICjpIjXE0+QuGwZMWPHYmjYoNDNIksz2Wzm2sSJSqq+tlw5goYNA7WarEOHcKtaldQtWxzlqvz9yfr7b9zr1StU7xh9zZqUGTsG69WrJC53nCXl0bw5stWKW9Uq2I0mjEcOY7kei8+zz2Jo0gTL1atgt2GNjUNbrhxqPz+Mx4+TdegQhoYNlLIK5efPx6P5I1jj4pBUKjShocqPQdlqxZaWRvrOnWT8sRtDo0bIViuy2YyhSWPStv2CtmxZ0nfsIP2GfmEFqbTyK3RhRT8D/l6wZ08lq3OVUbFcuYKuQgUs2eUnczdkdpI0mrvak0ElGrI/9IqbvO9cTZJl3CIjRWCkFJC0GiTngbnhuDr/Bqnsdvx796bM6Lfz3YZKr8fQsCE0bFioxywzZjS+L3VBV7myy2Sfqlz+TXjVuSakPFs86lK6RFuunBJQudlt+XFO0Ho92Y6y06djPncO2WbD89HmSvDCrXp15QzwoipuGRpJkpTxq3L1x5RFcEQoBtlmw5qQgCYwUHnN5842taWmovL0RFKpyDpyFMvVGFR6PabTZ5A0auxmM6lbtmA6eSonQKfVQq4TnLyfeQZtWUd5Vd0NAUS1pycUUI5IZTDg1/Vll2Xd69bFvW5dl+W8WruWXlK5ucENWVQFlTwqTW6VORA0/C0kNzd8O3d2CbwW5u+ZcPfdmOHqXrs2nm3akLh8ObbERKzXrpF27RppP/+cZ13jiZNFDo6kbd3mEhhxsiUmAo7g4tl2Tyq3OwP8aVvyz/owHj7Mhde6K9djc5XOyl26UhcmAm/3GxEcEQShVDmZO/VfEB4g5gsXOP/Syy6ZILqwMCQ3N9wbNnScpQUuaem68uXx79Ej3+2F78+pHezZsiUM6I9stXJt0iSS163HdPw4l17vT7mZM1D7+3OpXz8g54tb2akf4f3MMy610sFR7/hC9x5YYmJQ+/tTeeM3aMvcug6715PtSFzmaBx7aehQwpYtc5mQuN9YExK42O91JZOjwuef4xHdTHm+DNkTWH65+k4VdTI9oFcvwFEKzZaSUqx+HT7PdlAu+/fpg/nffzE0aFDgeCSNBo2fH77PP59vuSjnZJb30+25NGAg+ogIDFFNkE0mEpYuw3TiBG41IzGfPYdsMuH15JOlNjACOT1H5OyGpAAZv/+O7pVXlOBISfxglkRD9oeWlH2aSPF7jmRvR5bRF9BzT7i3cn+O2rMbsSrXnZkjdrsjA+MOlR2WJClPBkZJkiQJnw7PlPQw8pcr60W+RTau8HCwJiYim0y37PUm22yk/vADsTNnYb12DUmnQ+XtDRLY4uJReXtjaNiQ9N9+Q+3riz4ykox9+26Z9Q24BEYAUn/4Ac9WrQDQhoYUd9ceeho/P5esOaF0cate3aUfjNrbG0mS8GzTWumf5UKS0Nepg/Hw4Ty9uAojY/du5XLQ8OFk/fUX6Tt3Fnv8yrD0emSjMd/71AEBBL/zzm0/hnBvieCIIAiCINwDqVt+dPmxZGjWzNFo8A72Z5I0GoKGD8d86bJSGunKiJH5LhszZiyx02fg2/VlPKKi0FYMQ+Pny7UPJjv6TQABffsWKjACYGjQgLIzpnN17DsYDx3mTOs2lBk7Bl1YGLpKlbCnpKDy8SFx+XK0ISH4vvQSkkrlKBVgMiHbbFguX0Fdtiyqg4eQ69RRJjSy/vkHS0wMstFExt49BA0ejCUmBm1YGNa4OCwXLiDp9Wj8/TGePo3aywvjP8dwb9QQld4d49Ej6GvVImnNWsznz2ONjcWrbVs0IcFogoKQjSayDh1C5emJvlZNVG5uxIwZCzga6gW98cZdbRCoDw+/I9vR+PkpTR9vl9rLi0r/Xelym/dzzymvV3tWFqmbt+D91JP5rV5qOKcppVxBiLhPZuMWEYHl2jWgpIIjjrM17aIh+0Pndv/ky7nCKm4iOFIqSGq1khkt228sq5WdOSLLLuW0hHsodwN2kTnyULNnZHB96sckr12LysODKls2F/g9N/Ovv4mbOZPMXOVLZbPZpR+PPTVVKXlqS0x0aWYOYGjcGLWfL7LNjiYoiNQtW7CnpuLfswcqb2/8unYl88ABrgx7E8gpn6opE3wH91oQSg9Jo0FXKQzTMcfJZ0oZyHfeQRsSispdj3+fPiSt/C/x8+ZRZvRosNu4eviwclJTUWQdPAiAxyOP4N+7F1LfPsgmE6azZ0nfsZOAvn0w/fsvWQf+QuXpSfK6dbjVqIFb1apYrl1D0mnR16iBZDBgOnkK49EjgESZMWNQe3uR8MUSrNeuofL0xHz5Ehm791Bh4UKXsnDC/UEERwRBuCeKe4akIJRWluvXyTp8GK82bTCdOsX1jz9GX6MGHo88gt1kQtJoyNi7l6y/D+JWo7pLg0BJr6fMqFF3NDDipPH3J2zZUhK/WumSRmyIikLl6Yn12jWMx44BYI2LI37OXOLz2U7ZqR/h/dxzRXpsn2eeQePvT8y4cVhjrnJ1bMFnzSQsXYra1xfTyVN5zrzRA+e+cNQzt6Wm5jkDz9lE8pY+L/iu5HXrCrWJCgsXYmjYoHCP94BzKd3i7o7vi51KcDSFY3eWvciemFT7+WG9fp0Lr3RzLKDVlkgJOJE5ItyJzBG3GsUrVSTceTmldlwzR2y5ymqJ4EjJkCQJNBqwWpFt9luvIDyw0nbsUPpR2DMySNm0iYB+/RyBjd27Sfv5ZzL/PohsNmNPTQVAMhgI7N8fv64vY758Beu1q1hirpLyw/cYDx3GvVEj/F97lYz9+7HGxuH1RFuMR47iFl4Dv5decnn8wMGDsSUnu5R78m7XjqxevUhctky5TRNcuBOTBOF+5Ne1K9c/mEzQiOFKg3OVhwdBQ4coy/i/9ip+r3ZDkiQysk/4Kyg4Yjp7Fuz2POUbbcnJygl/Zad+hCr7M1jSaFxKz7nXqoV7rVoA+HZ6ocBxe2VnduVWZvhbymXZYsGekSECI/cpERwRBOHeENERoZRL/+03YmfOouyHk9HXrHnTZU3nznGhR09s8fF4tX+K9F+2I5vNZO7Zq/SPyM145Ajg6DNRccVyVG5ueZqt3mleT7Ql7pNP0FWtQoWFC10yClI2bSLr6D/oKlcic88el8CNysODkAnv4tOx4AbcN+MRHU21n38mdto0Eld8WWDzb8uFi1guXCxwO85asLei9vdHV6kStqQkLNeuIWdlud4fFIg9NQ3ZZHJMgqtUWK9fd1lGUzYUlc4NbcUKWK9dV75I+3TsKAIj97ncE8na8uUJ++pLro5/19HLB9AEBCiTmveSpPQcEcGRh1WxvxY5e44go/YVDT9LC0md/Xfkhs8851WVbL/rn/tCwSSVyvHWsYmyWg8zZ8DDKW7GTBIWfQYqVZ77UKnwfbETAf37K30S3H19obZjEtW/R3fsmZlKDz/v9u1z1s2nbCmANrgM2nwCH/rIiBuWE5kjwoPL76WX8O3UKU9p5xs5T8pyZpeYz58n5bvv8GzdWumPYzeZOPeMo8Rwjf/tR+3lpayfsWcPyDJu1avdkxOhJK1WBEbuYyI4IgiCIAhAzNh3sCUmcr7Ti4RMfA/fl192OVPe+QPIlp7Oxd59lLT6/Bq2SXq948uZVqM08y7z9ij8e/a85RfBO0UbHEy1Hb8iabWOpou5+HTsqAQ//Lt1Q7bZuD51KnKWkeD/e+e2e4VIGg3B77yDf9++2NPTMZ09i6FhQ7IOHsTQtBnYrKR89z3Ybehr1QJJhenMaTyaNiXt99+54OFBZUnCrXx5tGXLIun1mI4fR+XlhSYggKwjRzFENUHt45NnYlu2WBwltypUwHTmjKOHi0qFLTkZtY+Po6aNzaYcB1tamqOJprNhuSxjS05G4+dX7IbJQumROzii8vRE4+9P+TmfEr9gIYlLluD9ZMmUBXO+J0VD9ofPncoXlGSQdGKyvbSQVI4ykDd+btiUniOirFaJ0mjAbBYN2R9ystVx/NX+/rjVqEHm3r3Y09OV+31eeAHvp9uT9fdBDFFN8GjW7KbbcwZGbpfbDeVVNYUsaSsI96ui/B7WBgcj6XTIZjMxb4/GrUYNKq1fh0qnw/zvBWU54z/H8GjWFADz5SukbNwEgEeLx+7s4IUHkgiOCIJwT935IkKCcGdIuQII1yZOImP3HvxeexW1lxdJq1YrafhO2goV8O/Rg6SVK7FcuUL5eXNR+/mjD6/hMgFiPHECW1ISHtHR92xfnJxn1dyKpFYT8n93vnmhtkwZKFNGaRrr9fjjyn3+r73qsqwzQ8O3fHn+PXgQ7/r1Uedqoqp55JGc7WanYOdH0mqVBuG5mxW79OLI9YU89xlG4DhLybns3Sh7JtxbsnKWPag9PRyX1WqChg4hcPCgEskagZyyWuk7dvDvy10pO2MGuvL3vveJUHKKG3vNvZrIRCg9lLJaN/QcsTvLaonMkRIlqdXZmSMiOPIwc5ay9GzxKGWnTiX+s8+JmzkTgAqfLcLzscey729xT8ely/6eDIBKddsnKQnCg0TS6fDu0IGUDY6yyqZTpzAeOYKhUSPM588ry13s1QuPRx7B9+WXufLmm8rtzve1INyMCI4IgiAIAqAtVxbr1avK9bSffybt558LXD6gT2/8XnkF/+6vIdvtBU6y6iMi8r1dEIS7z9m82pE5ckMgrIQCI5DTkB0g69AhLg8dSpWN35TYeIR7R2ncXczCWspasiwm20sRldJzpIDMEVn0HClJUvbJFiJz5OEmWxzBEed70a/bK6Rt3YquciU87nFAJDeVToffa6+RtHIlXk+2K7FxCEJpFfTWm1ivXSNj927AUeLa0KgR5n/PuyyXsXu3soyToUnjezZO4f4lgiOCIAiCANjTMwAov2A+moAA4ufNx3j8ONbYWFCr8W7fHve6dZFtNmwJ8fi8+KKybklOsgqCUDA51yVVITOp7oUbJ0lNJ05gS0vLk8kk3L9kmw3ZbFbOAJbtduzp6bedkaYE/BDBkdKkwJ4jzuNll0UZtJLkDI5YRXDkYWbPLmXpPEFB7elJ5XVrb7bKPRMyfhxl3h4lgqiCkA9tmTJUXPIF1z78kKQVX2I+ew4AU/a/ANry5bFcvuyyXuCQIUpwXBBuRgRHBIUsy6KEiCAIDy1nzWGNnx/udetSYdFCAEznzoNsd/SuEAThvpLTcwRU2WW1SgNnQ/bcsg4evOelPIrCfPmyYzLebgeVCrWvb55+RneL3WzGdOo0+lo1b/ld1XTmDKbTp9HXqUPqli1k7t2HJjgY9/r1kLOy0FasSMKiz7DGxeFWrRrqoEDcKlcm5Ycf0AQGovbyxnzpEmo/X2yJSbjXqYO+Vk3c69ZFcnfHcvkKGXv3gCTh360bWf/8Q+LiL9DXqYPa2wvjseNkHTyINS4OlY8Pvi++iMrTA/O586Ru3kzGxA8hKKzYDdmVUnFy3iCbUILU+fccyWnILoJZJUmZHBMN2R9qzrJapfVv5736TBWE+5VbFcfv8cRly0j75Rcsly4BUH7hArxatSJj717iZn9KmZEjcKteHZU46UgoJBEcuU+NHTuWb74puPxCYGAgf/zxR6G2lZqaygcffECXLl1o0qTJnRqiiw0bNvDOO+/wyy+/UL58+bvyGAAjR47k+++/Z8yYMfTp06dQ6xw4cIDPP/+cv//+m4yMDAIDA3nkkUcYNGgQFSpUKPRjjx07lv3797N9+/biDv+OKm3judNtjcPDwxk6dChvvPHGbW1n37599OjRgxUrVtC0adM7NDrhfuQMjtx4drlblcolMRxBEO6AnIlkGbVn6fmBlN8ESOaBA3ctOGK5epWkVavRViiPW5UqJH71FZaYGGzJydjT0tHXqoU+vAa+L72EtmxZsg4fQdLpULnpyDp0iPQ//iDt56156vVrK1TA+6kn0UdG4hYZiT09g6zDh8AugyzjEd0Mt+rVXdZRJo8tFuyZmaBSYTp1CrfwcGSrFXt6OpYrMchmE7aUVLIOHyb122+xpaTg+fjjaPz90NesidrXF2tsLNqKFUnZ9C3GY8ewp6djS0zM9zlw1qp2eV6uXHG5bspnPeORIwU+rwkLFiqXM//8M8/99pQUEpcscbktfds2eKVvgdu8FefzJzJHShdn5ohdvrHniJNoyF6iNM6yWvZbLCg8yGSzBSi9wRFBEG7OrXo15bIzMKKvWRPPli0B8GjWDI9mzUpkbML9TQRH7mNBQUHMnTs33/u0RfixdPz4cTZt2sSLuUrE3I/S0tLYtm0bNWrUYM2aNfTu3fuWZxfu2bOHfv368cQTTzB58mS8vLy4ePEiS5YsoXPnzqxbt46KFSveoz0QBKGkyLKMrYDgiCAI9y8lc6SUl9UCSN/5G0Fvvql8d7FnZKDy8EC22bBcuYK2QgUsFy+SdfgI5tjraK7HkhoTg6FWbXQVKyBpNI5ay3v2YjpzBkmjxhoXj8rbi8z//YmclVXgeDJ27SJj1y4SFn+BpNMpZ9feiuXSJRI+X1zwAlotPk8/jXujhphOnSZl0ybsRiNqgwHUamwpKaBSgcVSqMdL/+WXWy8kSah9fLAlJwOg9vdHExiI6dw53GvXxnjyJHJWFu6NGuH1+OOYzp0l/dcd2NPS0IaGoi1XFu+nn8ZuMmFLTCLzzz+xJSZiOn3asXmdzhGgyDVmlacnXk88gWyxoI8IR+3nhy0piayj/5Dx++9K8B0cGR+3QwkuiUyEUkWlKiBzJFffI3G8So6kzp72EJkjDzXZWVYrn+xNQRBKP/eGDQl6602MJ09iuXwFfWQkAf1fFxVwhNsmgiP3MZ1OR/369Ut6GKXG999/D8C4cePo2bMne/fuJTo6+qbrLFy4kLp16/LJJ58otzVt2pSWLVvyxBNPsHTpUt577727OeyHjvjYEkoj2WQCq+MH841NmwVBuH8pTa9LXVmtnMyRMmPHEDdzFqbjx0n84gtsaekkrVyJPT0dj8daYE9JJevQIcekvMWipMPogGvZ21D7+KApE4Tp9JkCH1Pl4QGyjGyzoatSBbdq1XCrUR1tcDC21DRSt2wh68ABR58MLy9kiwXZZMK9QQPc69fH0LAB+shIJJ2OtF9+IWnVarDbcAuPwHzxIqYTJ5DNZvS1HcEa878XMB47RsqmTaRs2uQyFltKSq4reXsAqH19Qa1G7eWFW40aSG5upH73HQCaoCBlfdlsRjIY8H7qKceyERF4NI1CExqK+dw5x/MSGOhY1mpF0mgc+2WxoDIYinTMjCdOYEtMxBAdnRPAMhpBpUJ1i7OQLdevYz7/ryNT5cdfHeMpZpAkJ3MEMdleiii9xwooqyXJ3PJ1Itw9zuMjGrI/3JyBf/FeFIT7kyRJBA4cWNLDEB5AD2VwRJZlrKb8EufvPY2b212Nch49epSXX36ZZ599lo8++giAhIQEOnToQEREBAMGDKBnz54A9OjRg6ioKL788ku6d+9OcHAwZrOZ3377jQYNGjBw4MB8yw91794dgC+//BIAu93OggULWLt2LUlJSTRv3jzfcl2nTp1ixowZ/O9//wMgOjqasWPHKqWsnOWOJk2axKJFi0hJSWHOnDk0b9483339+uuviY6OplmzZoSFhbF69epbBkfi4+PxyqcOYZkyZRg/fjx+fn7KbbIss3z5ctasWcOVK1cIDg6ma9eu9OnTx+UYbtiwgUWLFhETE0PlypUZOXIkLbPT/DZs2MD48eNZtWoVH374IceOHSMwMJDXXnuNvn1zSiyYTCZmz57NDz/8QEJCApUrV2bQoEE8/fTTRR6P07Fjx+jRowcNGjRg3rx56HQ6kpOTmTlzJtu2bSMtLY3IyEiGDx/u8ryFh4czYcIEjh07xtatW7FYLDz22GO8++67BGZPOLRp04aoqCjlNebc19yl1BbNm8v3P2ymx9BhjPpsARcuXKBKlSpMnDgRSZKYPHkyJ0+epGLFiowbN85lDPv372fGjBmcOHGCkJCQPAGry5cv8/jjjzN27FjWrFnD1atXmTBhAi+++OItX2dO586d47PPPuPPP//E09OTF154gbfeeguNRqNsf8qUKXTq1ElZ58bSZd27d6dixYqEhYXx3//+l4SEBGrVqsX//d//UbduXWW9Xbt28dlnn3HmzBlCQ0N54403mD17Ns8999xtlwkTikc5q1eSUBncS3YwgiDcMbkzR9SlKXNEmzMx49OxI+azZ0let57Y6TNclsv4bZdy2Tmpo69TB3WZIFKvx6JPT8caG4stJUUJGGgrVMC3SxfsmRkgOwIn+po1MUQ1gezvB/l9T/B/7VWs8fFY4+LQVamCpNUiW635TiL5de2KX9euLrfZUlIwX7qs9AWRZZnMffvJ2L2b9N9+w3TihGPdHt0x1K8Pag2aMkHYkpMxNGwIkoTp5EmsiUl4tXvCZYyyzebo45GQQKXVq5AMBlR6vWPmWZKUpue53dgnStJosp97bbGCCvqIiDy3qfT6Qq2rDQ5GGxxM2q+/KrPlxe854gyOSKLJaCkiFdRzxHm/KKtVsrLf/6Ih+8Mtp+eI6O0hCIIg5HjogiOyLLN6wmhiTh0v6aEAUDa8Jl0nTS12gMRqzT81WK1WI0kStWvX5vXXX2fBggV07NiR6OhoJkyYgN1u56OPPsLDw4MJEybw/vvvM2HCBJegx5YtW3juuedYsGABdnvh67POnj2bVatWMWjQIOrVq8eWLVuYMcP1x/758+fp2rUrVapUYerUqVitVhYsWMArr7zCpk2bCAgIUJadO3cu48ePx2g00qBBg3wf8/Tp0xw5coTZs2cD8PzzzzN//nzi4+OVCfz8tGrVisWLF9O9e3c6duxI06ZNlUnzLl26uCz78ccfs3z5cnr37k3z5s05cuQI06dPx2q1MmDAAACuXr3KZ599xptvvonBYGDWrFkMGzaM7du3K/tkt9t566236NWrF2+99Rbr16/n448/pkaNGrRo0QJZlhkyZAh//fUXw4YNo2rVqmzdupXhw4djNpt5/vnnCz0ep7Nnz9K3b1/q1aunBEZMJhM9e/YkPj6e4cOHU6ZMGb7++mv69evH4sWLXYITs2bN4oknnmDmzJlcunSJKVOmoFarmTlzZoHPbX4Srl9j6SczGD1iBAaDgf/85z8MGzYMrVbLwIEDCQ0N5aOPPmL48OHs2LEDvV7PP//8Q58+fWjWrBmffvoply9fZsSIEfluf86cOYwbNw5PT0/q1atXpNfZlClTGDhwIP369eOXX37h888/JyQkhNdee61I+/jTTz9RtWpVxo8fjyzLTJ06lTfeeIPt27ejVqvZu3cvI0aMoHXr1rz55ptcuHCB9957D1MpCdg+rGxpaYDjzGrl7E9BEO57uXuOlKamjLrKlfDr3h1dxYpo/PwIfucdUKvJ3LcfbYXy2DMyHWWeypdHE+CPe/36SBoNhqZN0YaEYLPZiDt4kMj69VHJMhm7d5O65UdQSYSMH59vsKAwNIGBSqYFFK0uu9rHB3cfn5x1JQmPZk3xaNaUMiOGF2obhgJ630lqNZVW/Re70YS2XLlCj6m0kXQ6pNvswqYERzQiMFKaKJkJBQRHRBm0kqUEEu0iOPIws5uzy2qJQKUgCIKQy0MXHAGUs+bud1euXKFWrVr53jd69GglE2HIkCFs376dSZMm0b9/f7Zt28bs2bMJDg4GoFq1asq/zsvg6FsyadIkdNlfHvbt23fLMaWmprJq1Sp69+7N0KFDAWjRogWxsbHs2pVz9uPcuXNxd3dn2bJleGafyRkdHU3btm1ZvHgxY8aMUZbt1q0bTz311E0f9+uvv8bX15c2bdoA8MILLzBnzhzWr1/PwJuk3b355pukpaWxfv169u/fD0BISAgtW7akV69eVKlSRdmvFStW8Nprr/H2228D8MgjjxAXF8f//vc/JRhht9uZN28eVbPPVnRzc6NXr14cPHiQxx9/HHD8aBo8eLASfGnUqBFbt25lx44dtGjRgt27d7Nr1y5mzZqlZIq0aNGCrKwspk+fTocOHcjMzCzUeAAuXbpEr169iIiIYP78+crx3LRpEydOnGDt2rXUq1cPgMcee4zu3bszffp0vv76a2UbNWrUYMqUKcr1w4cP8+OPP970mNxIBkxGIwPH/h8d2rUF4MyZM8yYMYPJkyfTuXNnADIzMxk2bBjnz58nMjKSRYsWERAQwIIFC5ReOn5+fgwfnneipX379i69c0aOHFno11mPHj0YPHgwAM2aNWPbtm3s3bu3yMERq9XKF198oTxeRkYGY8aM4fjx49SuXZu5c+dStWpV5syZgyr7h3RAQECBAZ+7RZZlEhYtQu3rm+fs3weR3Wx2NG2TZUeZlaAgzJevkPH7LrTlyiu1h0vT5KkgCHeOJIPKoxRljkgSIeP+T7muMhgInTixeNvSaPB87DE8H3vsDo2udNLc5GSX+0XuLBy5mEES5+S7SvNw/owrrW6dOSImZEuUsyG7yBx5qOVkjoj3oiAIgpDjoftWLUkSXSdNfSDKagUFBbFgwYJ87wsNDVUua7Vapk6dSpcuXRg3bhwvvPDCLYMNAFWqVFEm0gvr0KFDWK1WWrdu7XJ7+/btXYIje/fuJSoqCr1er2S/eHp60rhxY3bv3u2ybmRk5E0f02Kx8O2339K2bVuMRiNGoxEPDw8aNWrE2rVr6d+/vzIJfSOdTsf777/PG2+8wc6dO9m7dy/79u1jzZo1bNiwgZkzZ9KuXTsOHjyI1WqlXbt2LuuPHz/e5bqfn58SGAEoX7484GgWn1vuDBidToe/vz+ZmZmAo0m8JEm0bNnSJTOoTZs2fPvtt5w+fZq4uLhCjScjI4NevXoRFxfHypUrcctV43zPnj0EBQVRq1Ytl8dp3bo1H3/8MSkpKfhknwF6Y2+bkJAQsm7S2PVmIurWUy47s3qcwRkAX19fwBGQAjhw4ACtW7dWAiMA7dq1Q51PKYkbXytFeZ01btxYuSxJEuXKlVPGUBTVqlVTAiOAEoTMysrCbDbz999/079/f5f3/VNPPcXo0aOL/Fi3w3z2LHGfODKt3KpXR1+rFpKbm6PWvFv+qeay1Yr5/Hl01aohSRJ25w8MtRrZZMKekYH50iVkoxH3Ro2QtFrsmZmYTp9GV6kSyDIJiz7D58VO6GvUcGzTbMaWkYElJgZ9RETO5ILNRsbuPbhVr4Y2JKTI+5e+axeZ+/eDSk3m//5H1l9/5dypUiHp9cjZ77nc7MU45oIglF5yrkuiAatQ0iSdTmnIfttltURJrVLF2ZC94J4jInOkJEkq5/dL0ZD9YSabsn+7iO8DgiAIQi4PXXAEHBOf2kLWCC7NdDodderUKdSykZGRhIeHc/To0TyBi4J4eBS9cWlycjKQM8HtFBQUlGe5zZs3s3nz5jzb8Pf3d7luuEXDzB07dpCQkMD69etZv359nvt37dql9PwoSFBQEJ07d1ayF/bu3cvbb7/NxIkTadu2rbJfN47tRjeOVWnYeUNZMv0Nrz+VSqX82E1OTkaWZRo2bJjvY8TGxpKSXVf8VuNJTk6mSpUqpKamMm3aNObMmeNyX1xcXIHZR3FxcUpwxP2G8hy5x1tUBq+8Z+7euP3cUlJSXHq/AGg0mjy3Qd7nvyivszu1j/ltBxyvgeTkZGw2W56xq9XqPO+Zu834zz/K5QuvumbHuFWvju/LL6OvWZO0rVtJ/+03ZLMZe1qaozZ8s2a4ValM6uYt2DMyHA2C86Hy8sKeHRhU+/lhS0oCIHH5cjTBwai9vVyaB6sDAvBo1gy1jw+pP/2ELSEBtY8Pfj17kPHHbiS1Gk1wMKaTJ7FnZuLTsSMZ+/biXrsOhqgmpP30M9aEBCwxMZjPnSt45+32fAMjAPoC3g+CINyflLO2ZVlMJgslLndZrdttyK4Sr+fSRV1QWa3sYJYseo6UJOXvv2jI/lATmSOCIAhCfh7K4MjDaM2aNRw9epSIiAgmT55MdHQ03t7eRdpGQRP9GRkZSiDFOembkJDgkkHhDC44eXl58cgjj9C7d+88j6MpYpmAr7/+mgoVKjB58mSX22VZZujQoaxevTrf4MihQ4cYNGgQ06ZNy9PkvVmzZvTt25cpU6aQlJSkPFeJiYlKqS2AmJgYLl68SKNGjYo05pvx8vLCYDCwYsWKfO8PCwvjr+yz4G81Hl9fXxYvXsy3337LxIkT2bZtG23btlUep1KlSkyfPj3fx3FmvRSW7YYfG5kFTD4XNU/K19eX+Ph4l9tkWVYCRDdzp15nztd+YfexIAEBAWi1WhITE11udwZO7iXjsWMF3mc6fZrrH3xQ4P2Ze/eSuXfvLR/DnitjyhkYcbJev471+nWX22wJCaT+8IPrbSkpxH86h/zEz5sHQNafB0hctizfZXRVq+L3ajewyyR//TVqb298nnuOxOXLcW/YAN8XO3P5jTfw6dgRSavF87EWt9wvQRDuHznNq3lgyqoK9687MSFnd2YiiOBIqaIqoKxWbmJCtgQ5G7KL4MhDzRkcUYn3oiAIgpCLCI48BK5cucLUqVPp3LkzQ4cO5emnn2by5MlMnToVIN/yRPlxlgq6du2acltKSgpnz56lbt26gKNclF6v58cffyQqKkpZ7tdff3XZVlRUFGfOnCEyMlKZpJZlmVGjRhEWFnbLUlpOcXFx7Nq1i379+rk0k3d66qmn2LBhA9evX1fKGzlVqlSJrKwsVqxYQXR0dJ7SW+fPnycoKAh/f3/q1q2LVqvl119/dSm/tGTJEjZv3szvv/9eqPEWRlRUFEuWLEGWZeV5BUcQaOvWrXz44YeFHo+HhwceHh68/PLLbNy4kUmTJtG0aVO8vLyIiopix44dBAQEULZsWWUbCxcu5Pjx4wUGTfLj6enp8roARzmsOyE6OprffvuNrKwsJStj165dWArIVsjtTr3OnK/967km8y0WC4cPHy6wZFt+1Go1DRs2ZMeOHbz11lvK7du3b3cpbXYvZB11ZI4E9OuLbJexXr+OR/PmeDzanLQtW0jb/iumkydxq14dv+6voQkKAquVzAMHSPlmI7pKlfB4rAW6sEpk/XUA4z/HcG9QH+8OHUjfuZP0HTvx7dIZz+bNsaVnkLxmDek7dqCtUAGfF55H4+uL6ew5LDExWOPj8e38IsZ/jpH263asV6/h8/zz+Dz3LJf6D8B48iTezzyNR/QjWOPiMJ04QermzbjVqIHa35/MAwfAYsGzTRu05cuh9vHBdOIkgUOH4lajuhLc8n/tVWX/fV/spFyuvsP175MgCA+O3M2QRXBEKGmSTqekjNxeWS0JlWjIXqpIBWWOiLJapYLk/L4ugiMPNdnZkL2AEsKCIAjCw0kER+5jZrOZgwcPFnh/eHg4er2ecePG4e7uzujRo/Hx8eGtt97iww8/5Mknn6RNmzZ4ZTcg3rFjBz4+PkRERBS4vdDQUObNm4enpyeSJLFo0SKXMkIeHh7069eP+fPnYzAYaNasGTt37swTHBk8eDBdu3ZlwIABvPLKK7i5ubFmzRq2bdvGp59+WujnYOPGjVitVp555pl873/++edZt24da9eu5Y033nC5z8fHhzFjxvDee+/RrVs3XnrpJSpUqEBaWhpbt27lm2++Yfr06UiShL+/Pz169GDZsmXodDqioqI4dOgQq1atYvTo0UWaIL+Vli1b0qRJEwYPHszgwYOpWrUqhw8f5tNPP6VFixZKOaiijEelUjFp0iRefPFFpk2bxvvvv0+nTp346quv6N27NwMHDiQ0NJTdu3fz+eef89prr7n0+LiV1q1bs2jRIhYtWkS9evXYvn07ewuRWVAYQ4YMYdu2bfTt25e+PXqQmJTE7Hnz0Gq1yFYrst1e4Fl6t3qdybKMnB2UsF6/juXaNSStFtlqc/TTkGVM58+jt9upX7s2X65YQfmAAHx8fflq9WqM2QEb8+XLgKPptyRJWBMTQaVC5eambN/pjTfeoEePHrz55pt07tyZmJgYZs929P6QcPyoLm4fopuRrVZsaWlo/PywpaWRdegQAL6dOzv6geTi37Mn/j175rsdQ5MmBA4c6HKb56OumVf+r76K/6s5gQiVhwdBw94gaJjre9D9hl427nXq4Nf1ZZfbKq1ehd1kQn1Do/SyM6bnZPSkpGBNTMStcuV8xywIwsPLmesqcXf+tgpCUdy5niMSKo2YaC9NlMn3gspqIcpqlSjRkF2AnH6J4r0oCIIg5CKCI/exuLg4Xn755QLv37hxI3/99Rd79uzhk08+UfpHdO/ene+++44JEybQsGFDqlevTocOHVi5ciW7du3i+++/z3d7arWaTz/9lA8//JARI0YQGBhIz549OXfuHOfPn1eW69OnDz4+PqxYsYLly5fToEEDxowZw8SJE5VlIiIiWLlyJbNmzWL06NHIskyNGjWYN28ejz/+eKGfgw0bNlC9enVqZDd3vlGjRo0oX74869atY/DgwXmyZLp27UpYWBgrVqxg5syZJCcn4+HhQd26dVm+fLlLNsrbb79NQEAAq1evZvHixZQvX553332Xrl27Fnq8haFSqfjss8+YPXs2ixYtIiEhgeDgYHr37s2QIUOKPZ6IiAh69OjB0qVLefbZZ2nSpAkrV65kxowZTJs2jbS0NMqVK8fIkSPp06dPkcY8YMAAEhMT+eKLL7BYLLRq1YrJkyczaNCg23ouwJHh89VXXzHlP/9hxNtv4+/ry4juPZi+5AusiYkYjx3DnF2mSjabscTGYktJAbtMFXd3ln/6KZ8uXszoUaOQgWoVK/LJO//HYxUqONa9dAkAW2Ym1lzlu5x9NOwZGQBMGjCAqV98wXsffoiHwcDzbdpQr3JlNmzbhi27JJZssSADlpgYZTvO7ZsvXMDo7UMdTy+mjx7NglWrGDxoEGWDgxn7+uuMnjoVXVoaphMnHKn/jrkPJJUaSaNG0mqR3N2xZGVhz8jAePIksloDVgu29Ax0YRUxnz+PymDAfPGS47mIiSHzrwNIWi2mU6exJSSgr13bceai1YqucuU8gZHSRtLpUOfzAyb3JKfaxwd19t83QRCE3JzzlCq7DHfwRAZBKI7cPUeK23REmWwXZbVKFUntzFB2vV0pg4YkjlkJch4f7CI48jCTzY7fd5JWBEcEQRCEHJJc3K7KJcxms3Hw4EHq169fYFkoo9HI+fPnqVy5cp4G2MLdIcsymZmZGAwGcYam4CLVauN8pgm9WkW4R9Hej7aMDMy5AnB3mqTVIqnVSHo9stXquC5JyLKMyt0d2WjEbjQiaTTIdjtYbY4z0Ox2ZKsVlcGApFY7Aiqy7AiS2O3K/bnt+N//CA4IIDJXr5gzFy/SefhwPhk7llZNmtx0rCa7nYtxcag/mIzq6tXb2u+gt97MkwUiCCWtMJ/vglBY/Y6e5/u4FIatXsLIMcNxq/JgZJiJ98n9yZaezuw3RvFxj4E87uvBygbVi7yNxtsPcFlSs3jjl3SYPeMujPLBcK/fI6tmL2B43WjqZqXx89M5/cvK/XoQG7D+veE8+pso41lSLg0YSPrOnYROnuxSWvVh9jB+jpx6pDm2xEQqf7sJfQEnVwqC08P4HhGEoirN75OijE1kjgiCcE8VNWQmyzLW2FgA1L6+qAwGZLMZSa9HkiRs6emOYERmFrLdBpKEpkwZR1DDbEY2mpBtNlQeBlR6PUgSss2OpFGDSoWk0eSUQrgLZKvVUQIju8bxvrVr2Tx/PqPeeouwsmWJTUhg4ZIlVKlUidZduqBTqXLKhMmA3YZstWJPT3cEYiQJKTnZsY92O7bUVMc+ZWaiLV8ea0ICurAwVG5uqAMC8IiOxnL5MrrKlfB87DHSftmOJSYGtxo18On43F3bb0EQhNLAeQaQlJ2NJwglyVFWK7vniL2Y56fJsiOzVC1+xpUmBTVkd15TacXxKlHO42O7tz3+hNJFNGQXBEEQ8lPkb2l2u52lS5eyZs0arl27RqVKlejXrx/PPZczyXbkyBE+/vhjjh49ioeHB506dWLo0KHocn0IxcfHM2XKFH7//XesVistW7Zk7NixlClT5s7smSAIDwTLlStKaSuNvz8qg8Hlfmc5Jdlux5aUhMrDwxEEKSUkjcYxH5fdw2XMO++g0elY+MUXxMbG4uvrS4sWLRg5ciT67Mbv+cruNSMbjWiMRsKWL3PJiJPN5kLVz/Xv/trt7I4gCMJ9SUK+q4FwQSiM3A25Zdl+kyULpky2i4bspUqBDdmd92tEcKQkKSXNREP2h5ozOCIasguCIAi5Fflb2uzZs/niiy8YNmwYderUYefOnbz99tuoVCo6dOjApUuX6N27N/Xr1+eTTz7h7NmzzJo1i+TkZN5//30ArFYrr7/+Ounp6UycOBGr1cqMGTPo27cvGzZsKFIjaEEQ7hPFqOAn2+2O/iGASq9HcncvcFlJpUITEFDs4d0rer2e4cOHM27cuDtaek40FhQEQcjL+dHjyBwRqSNCyZKknL4Tsr2YwZHsF7UkGrKXLkrmiOvNOZkj4niVKGdDdlvx3nfC/U+W5ZzgiPjdJAiCIORSpOBIVlYWK1asoHv37vTv3x+A6Oho/vnnH7788ks6dOjA559/joeHB/Pnz0en09GyZUv0ej3/+c9/GDhwIGXLluXHH3/k2LFj/PDDD1SrVg2AyMhIOnTowJYtW1yyUARBeDAUp3iEPSMDZBlJq0VXtaroYyMIgiAUifLZI4uG7ELpcNvBEed2RCZCqaKU1SrgG6+qlNXhftgoZehEWa2HlmyxKJdFcEQQBEHIrUi/EnU6HatWraJPnz4ut2u1WkwmEwC///47LVu2dCmh9dRTT2G32/n999+VZSpXrqwERgCqVatG1apV2blzZ7F3RhCEB4stNRUAlZeXCIwIgiAIReacqJSQxeeIUCook+TF7DnizBwRZbVKF0mV/fcl12F1KbEl/v6UKKXsmVWU1XpYyZmZymVRVksQBEHIrUinHKnVaiIiIgDHl72EhAQ2bNjA7t27ef/99zEajVy5coXKlSu7rOfv74+npyfnz58H4OzZs1SqVCnP9itWrKgsIwjCw8VuNGK+eAlJo3Y0GzcYsGcHR9Te3iU8OkEQBOF+5NqQXUxOCqVAdsaHvZg9R5xE5kjpIsl5/77kDn+JvLUSlp05IoueIw8t87//Ajh+Z4rMEUEQBCGXYn+r/uGHHxg5ciQArVq14rnnniMtLQ0Az3yaCnt4eJCeng5AWloaYWFh+S6Tkd14ubBsN/mCY7PZHLUls/8T7j7n8yyeb+Fm8nt9WGNjkc0mZLPjy6tKr0e22ZA0GlQGwwPzmrrd94jz75nNZrvp3z9BuF85X9fi9S3cCXZ7TuaITZZRPSCvK/E+uX85M0fsNnuxjp/d2XNErRbH/ybu9XvEmaVmz/3YN2SOiONVgrLLKtqtFnEcsj1snyOZJ04CoKtW7aHZZ+H2PGzvEUEojtL8PinKmIodHKlbty5fffUVJ0+eZPbs2fTr148ZM2bcdB1nOYObTQoWteTBkSNHbnq/RqMhKysLezHr+grFk5WVVdJDEEoZkwwgYbfbyczMREpPR0pMRA4KQtZoUGVniaDVgsWC3WgEwO7vT+YD+Hoq7nvEZDJhsVg4ceLEHR6RIJQut/p8F4TCSMlUARLIMseOHUO+dq2kh3RHiffJ/cea/ZskMyOdgwcPFnl9m80GWkhJSyvW+g+be/UeSUpKgspgtVqV42KTARzBMLPJJI5XCdImJ6EFrl6+wqXCHAeLBdWFC9grVwa1GqzZvUpuJ2NLlpEuXEAuUwbc3R23ZWU5Lt+iL5aUlITs5+e4Yrc71lOpHL+bsretOnMW2dNDGatctqxj7LnHnJWFlJ6B7GFASkuDoCCOHDqEdOkSktGI7OODHBAAGRmOx8xdEcRiAZsN1enTSCYz9sqVkN3cHNtRq5Hd3MDNDenqVaSUFFTXriHFxyOHhmJt0gT1qdPI/n7Inp6oDx5ESkhEdnfHXqM6qFRoduxEdndH9vYCL29klYRmz15lv9VHj2KvUAEpORlbeDjotEiJSdjLlUN15QpSejqyux7JakNKTs4Zc2YmeHqiunQJgBRfX+LFe1EoAvFdSxBu7X5/nxT7071ixYpUrFiRJk2a4OnpyZgxY7h48SJAvtkf6enpeHl5AY7MklstU1h16tRBXUCDO6PRyIULF3B3d0ev1xdpu0LxyLJMVlYW7u7uora34MJitUGWGbVKhbtWhSkuDgBVXBwqDw9sOMpnacqVwxYfj2w2o/LxQV3Evwml3e2+R1QqFVqtlmrVqom/a8IDyWazceTIkZt+vgtCYXkd/RcS05BkqFWnDpqgoJIe0h0h3if3r/1rNwHgrnenfv36RV5fte1PAPyDgoq1/sPiXr9HLh4+DoBao1GOi8Uuw+9HAdDr9dQVx6vExJYJJhkI9vPDtmEDHo88gueTT5L+44+k/rA5O4PdjMrdHUPTKFK/+w7rteu4RUaiKRtKxi/bAdBVrYo2LAy1lxe29DSw2bHExGCIaoItORnjoUNIbnrM//6LJigIlZcnmsBArPHxWC5dRs4+OUpTtiwA1pgYR/DCakVdpgwqNzcsV66g8vRE5emJNTZWCXaofH0dGWMJCUXad7WfH7bsCh9KkCeb7OaGSq127ceh0yGbzY4rGg1uNWpgvXoVW1JSEZ/1HLolS4u9bm7qlBQAVJcvF23FXGMv37w5PuK9KBSC+K4lCLdWmt8nzrEVRpGCI4mJifz222+0aNGCgIAA5faaNWsCEBsbS3BwMBcuXHBZLyEhgYyMDKpWrQpA5cqVOX78eJ7tX7x4kbp16xZlSKjV6gIPgFqtRpIk5T/h5mT5zjUrFc/5/eNOHvebyf0IssmUc9lmUxqva4KCUKnVqIKD7/p4Slpx3yPO9W72t08QHgTiNS7cGVL2/2XUGs0D95oS75P7j+RspC7LxTp2zvJNau2D93q+G+7Ve0SlyvlO53w8q2R3uV8cr5Kj0jqmPdJ//hnLpUukfr0B/RdLMB47lmdZ4+HDymXT8eOYcs1bmM+exXz2bJ51zKdO5bnNevUqXAXzqdN574uJyXXFEbCwxcbiLABiT01Vei862Z3ZEAVQ+/o6flPZ7ai8vLBnB0QKDGpIEpLJhAyoDAbUAQFY4+OVAI5zbKZ8nqPcnOWPnetJOp2jr4e3FyqdG+bLl7HFx6P288NuNCJnZeFWMxJ7RgYqNz2Wy5eRtFrUQYFoy5XDnpKKbDYjWyyo/fzQhYU5Si4bDOiqVsWWkgxWG2gcGT3WhET0EeFogoKwm83IJjPY7WjLlUUTEoLp+HHMV66g9vHBlpyM7zNPK+UNBaEwxHctQbi1+/19UqTgiNFoZMyYMYwYMYIBAwYot//xxx8AhIeH07x5c3bs2ME777yDLrvR1U8//YRaraZZs2YAPProo3z//fecOXOGatWqAXDmzBnOnj3LoEGD7siOPejGjh3LN998U+D9gYGBynG5ldTUVD744AO6dOlCkyZN7tQQXWzYsIF33nmHX375hfLly9/x7bdp04YrV64o1yVJwsfHhwYNGvDWW28RERFR4LqXL1/m8ccfd7lNq9Xi7e1NvXr1eP3112nYsOEdH/PNxjJlyhQ6dep0Vx/rXhz33FyK6eVTWk8TGIjKmWIuCIIgCHeAcyJZkmXRkF0oFVTOxtDFbMguKwE/8XouXbLLR+e6JffXXZU4XiUr+32XO1CgBEbUanTly+P97LPIZjOm06fxbPkYqT/9ROaevUju7qi9vXGLCMezxWNYsn9zyjYrKjc9bjWqk/HHbsyXL5H15wEMjRsT/O54JRvFfP48Km9vrPHxuNeth1vVKqTv3Imk1+MRFYX54kV0FStiuXoVa3wCpjOnwWZDW64cbjXCSd+5E3t6OpJWi7ZCedTe3ni2aAFqNbbUVOwZGdgzMtHXjHQEZCQJTWgoplOnAAnZmIUmMBDTv/+icndHExiILS0NTeXKHPnlFyLCw3GvXBlJq0WWZYzHjmFLSkZXqRK2xAQs16+j9vJC0mgwnT2He/16qP38ULm7O3pDZv9+k+125KwsJK0WKVfDc9liwRobiyYkBNlmQzYaUXt759x/l0/U82ze/K5tWxAEQXgwFCk4UrZsWV588UXmzZuHRqOhZs2a/Pnnn3z22Wd07tyZatWq0a9fP3744Qf69etH7969+ffff5k5cyYvvfQSZbPTR59++mkWLlzI66+/rjR1nzFjBjVq1KB9+/Z3fi8fUEFBQcydOzff+7TO+qOFcPz4cTZt2sSLL754p4ZWIlq2bMngwYMBR73f2NhYlixZQs+ePdm8ebNLtlN+Bg0aRKtWrQBHX4dr167x5Zdf8uqrrzJnzhzatm17t3fhnirR4579a1FlMKDy9ETSaFA76+gKgiAIwh3iEoq/ST13QbhXnJkjsr3gHow3o6wlgn2litJbM9dtxTvCwt0gZZ/Nak9PV27TlA2l4qJFuFWvnu86Ps8/j/X6dXRhYbfcvs+zzwIgW61Izh4f4eEFLu/XtatyWZs9R+L8lyfbuSzrXqd2gdtR3VBiV1uunHJZf8Pj574PHOVG5LJlcataVXl+JEnCvVatnIXKlyP3qWuGxo0LHIukUiF5eOS9XatVHltSqyFX4MT5mIIgCIJQkorcc2TixIlUqFCBtWvXcuXKFUJDQxk2bBh9+/YFoGrVqixZsoSPP/6YYcOG4efnR69evRg2bJiyDZ1Ox9KlS5k8eTLvvvsuWq2W5s2b884776C5nSZnDxmdTidqDefi7++f5/moU6cObdu25ccff+TVV1+96foVK1bMs3779u157bXXGDduHM2aNcPT0/MOj/rhI+E4QwgAlQptmTIlOh5BEAThweX8vJHyyVgUhJJw+5kjDmI+sXRxHo6CgiOSShywEqV2DY77vtKVkPHjlaBAflR6faECI7lJYi5DEARBEO47RT6FTqfTMWjQIH766SeOHj3K1q1bef3111HlOhuvcePGrF27liNHjvDbb78xcuTIPJkMoaGhzJ07l7///pv9+/cza9YsyohJ0jvu6NGj1KpVi7Fjxyq3JSQkEB0dTe/evdm7dy89evQAoEePHnTv3h2A7t27M2rUKIYNG0b9+vXp3bs3+/btIzw8nH379rk8Rvfu3ZX1AOx2OwsWLKBVq1bUq1ePwYMHk5LdPC23U6dOMWDAABo2bEjDhg0ZMmQIly5dUu53Pt7q1atp3bo1DRs2LHSpMCcfH58iLX8jnU7HG2+8QXJyMlu2bAEcJcLCw8O5fEMjuDZt2rg8z+Hh4cydO5dOnTpRt25dJcvn3LlzDB06lKioKJo0acKAAQM4e0Pt2ri4OIYNG0aDBg2Iiori3XffJSMjQ7n/6NGj9OzZk0aNGtGgQQN69erFwYMHXbaxbt06OnXqRP369albty4dO3ZU9mHfvn15jvvly5cJDw/P9782bdrc1vOYh3OySvyyFwRBEO6i3MERSWSOCKWAMnlazMwR5yy8SnyHKlXyK3Mm5wqPiKNVsiS1a9BCHxF508CIIAiCIAgPj4fy1AZZlpEtxTtb606TtKrbmiC2Zjdwu5GzGX3t2rV5/fXXWbBgAR07diQ6OpoJEyZgt9v56KOP8PDwYMKECbz//vtMmDCBpk2bKtvYsmULzz33HAsWLMBuL/zzNXv2bFatWsWgQYOoV68eW7ZsYcaMGS7LnD9/nq5du1KlShWmTp2K1WplwYIFvPLKK2zatMmlBNbcuXMZP348RqORBg0aFPi4siwrz4fdbic+Pp7Zs2cTGBh4W+XaoqOjUalU/PXXX3Tp0qVI6y5cuJCRI0dSuXJlypUrx/Xr13n55ZcJDg5m4sSJGAwG5syZQ8+ePfn++++V9WbPnk337t2ZP38+f/31F59++imenp6MGTOG9PR0+vXrR7NmzZgzZw5ms5kFCxbQt29fduzYgZeXFytXruSDDz7gjTfeoFGjRqSkpPD5558zatQoGjRoQK1atfIc9zJlyrBmzRqX8e/fv58ZM2bQuXPnYj9/+XKewSt+2AuCIAh3kXKWPYiyWkKp4AyO2G83c0RMt5cuUp4LLqkjIphVspzl7Jw0QUElNBJBEARBEEqbhy44IssycQsPY76QWtJDAUAX5k3QwLrFCpBcuXKFWrlrguYyevRopdTZkCFD2L59O5MmTaJ///5s27aN2bNnExwcDEC1atWUf52XwdG3ZNKkSeiy64LemDGSn9TUVFatWkXv3r0ZOnQoAC1atCA2NpZdu3Ypy82dOxd3d3eWLVumlKqKjo6mbdu2LF68mDFjxijLduvWjaeeeuqWj71x40Y2btzocpskSUybNg1/f/9brl8QjUaDn58fcXFxRV63cePG9O7dW7k+depUzGYzS5cuJSj7S3lERASvvPIKhw4domrVqgA8+eSTvPPOO4Djefnjjz/Yu3cvAGfOnCEpKYkePXoojeKrVKnCmjVryMjIwMvLi0uXLtG3b1+lBwtAuXLl6NSpEwcOHOCZZ57J97jnLit28eJFvvjiC9q1a8egQYOKvO83yqmRjQiOCIIgCPeEUk1LNGQXSgnJGaMrZuKI0pBdvJxLFefxkPOPjSByR0qY6obgiKhYIQiCIAhCtocuOPIgCQoKYsGCBfneFxoaqlzWarVMnTqVLl26MG7cOF544YVCBRuqVKmiBEYK69ChQ1itVlq3bu1ye/v27V2CI3v37iUqKgq9Xq9ke3h6etK4cWN2797tsm5kZGShHrt169YMGTIEcATBEhMT2bJlC6NGjSIrK4uXXnqpSPuSmyzLxQpg3Tj2AwcOUL9+fSUwAhASEsKvv/4KoJTqanxDs7vy5ctz4MABAKpXr46/vz8DBw7kqaeeokWLFjRv3py3335bWd5Z3is1NZVz585x4cIFJbhlNptvOe709HQGDRpEUFAQH330UbGzm+x2M2ZzIjpdAK5n0ongiCAIgnD3OcvaSBTvc1wQ7jRJvr3XoSjUVDo5M3kK7DkiDleJujFzRBsSXEIjEQRBEAShtHnogiOSJBE0sO4DUVZLp9NRp06dQi0bGRlJeHg4R48ezRO4KIiHh0eRx5ScnAyAr6+vy+1BN6QuJycns3nzZjZv3pxnGzdmeRgMhkI9tq+vb57no1WrVsTGxjJt2jRefPFF1MWoLZuVlUVKSgohISFFXvfGsScnJ1O+fPlbrufu7u5yXaVSKXXTPTw8WLlyJQsWLGDLli2sWbMGvV5Px44dGT9+PDqdjosXLzJhwgT27NmDVqulSpUqREREALmaoRfAbrczYsQIYmNjWb9+fbFeB05GYwxWaxpWayq4VVFul0VwRBAEQbgHcnqOID5zhNLBmWFQzNVFQ/bSScrnuLoGR8QBK1G5fgNqQkPRBAaW4GAEQRAEQShNHrrgCDi+nEq6h6sB25o1azh69CgRERFMnjyZ6OhovL29i7QN55f6G/uPZGRkKBPofn5+gKPpu7NEFOQETZy8vLx45JFHXEpOOWk0d/ZlWbt2bXbv3k1SUhKBxfgivH//fmw2G02aNAFu/jzcipeXF4mJiXlu37NnD+XLly/0D6cqVaowbdo0bDYbhw8fZtOmTaxatYqKFSvSp08f+vfvj1arZf369URGRqLRaDhz5gybNm265banTZvGrl27WLRoEWFhYYUaT0FstiwA7HaTcpsEInNEEARBuCecPa8lZNFzRCgVnN/1bjc4InpYlDb5NGTPdUKS+OtTsnI3ZHevV68ERyIIgiAIQmkjvqc9BK5cucLUqVPp3LkzCxcuJC0tjcmTJyv3Fzabwtkb5Nq1a8ptKSkpnD17VrneoEED9Ho9P/74o8u6zrJRTlFRUZw5c4bIyEjq1KlDnTp1qF27NsuWLWPr1q1F3sebOXLkCD4+PkrgpiisVivz588nMDCQJ554Asj/eTh79myeAFB+GjduzKFDh1wCJAkJCfTr14+dO3cWakw//vgjzZo1Iy4uDrVaTYMGDZg4cSLe3t7ExMSQlJTE+fPn6dy5M3Xq1FGCTb/99huQE9TJ77h/8803LFmyhOHDh/PYY48Vajw3o9bkZJ3Iso1cVwBxFp0gCIJwt2VPTsoiOCKUDrf9zUdy9hwR36FKk/yCXq6ZI+LvT4lS5zz/IjgiCIIgCEJuD2XmyIPCbDZz8ODBAu8PDw9Hr9czbtw43N3dGT16ND4+Prz11lt8+OGHPPnkk7Rp0wYvLy8AduzYgY+Pj1J+Kb/thYaGMm/ePDw9PZEkiUWLFrmUgPLw8KBfv37Mnz8fg8FAs2bN2LlzZ57gyODBg+natSsDBgzglVdewc3NjTVr1rBt2zY+/fTTYj0fiYmJLs9HVlYWGzduZM+eANFcOgABAABJREFUPYwYMeKWQaCLFy8q61ssFi5fvszq1av5559/mDdvnrKfTZs2Ra/X89FHH/Hmm2+SkZHBp59+mqeUWH569erFxo0b6devHwMGDECr1bJgwQJCQkJ49tlnSUtLu+U2GjZsiN1uZ8iQIfTv3x8PDw+2bNlCWloa7dq1IyAggHLlyrFy5UpCQkLw9vZm165drFixQnlegDzH3Wg08u677xIVFcWjjz7KoUOHXM54q1mzZpF70EjkPOd22UpOLQmROSIIgiDcfcrHjSw6NAilxA1fhYpKKat1RwYj3CnO45H7sNrzW0AoEVKuhuzu9UVwRBAEQRCEHCI4ch+Li4vj5ZdfLvD+jRs38tdff7Fnzx4++eQTfHx8AOjevTvfffcdEyZMoGHDhlSvXp0OHTqwcuVKdu3axffff5/v9tRqNZ9++ikffvghI0aMIDAwkJ49e3Lu3DnOnz+vLNenTx98fHxYsWIFy5cvp0GDBowZM4aJEycqy0RERLBy5UpmzZrF6NGjkWWZGjVqMG/ePB5//PFiPR87d+50yb4wGAxUrlyZ9957j27dut1y/QULFigN7jUaDf7+/jRu3JgJEyZQq1YtZTlvb2/mzJnDjBkzGDJkCOXKlWPo0KFs3Ljxlo8RGhrKf//7X6ZNm8bYsWPR6XQ0bdqUWbNm4ePjU6jgSJkyZVi8eDGzZ89m3LhxZGVlUb16debMmUOzZs0AmD9/PpMnT1Yeo1q1aixYsIAPP/yQP//8k+7du+c57k8++SQWi4X9+/fzwgsv5HncX375pVD9UlwV8MtfBEcEQRCEe8DZkF0l20XmiFCq3G5ZLfEdqnTJORw5xyV3AEwlDleJslzPyfjX16xZgiMRBEEQBKG0keRbdWcupWw2GwcPHqR+/foFZgQYjUbOnz9P5cqV0ev193iEDydZlsnMzMRgMIh0f4GsrMtYLEkAmHVVuGKW8NSoqJCciC0xEU2ZMmjLlCnhUd5bt/seEX/XhAddYT7fBaGwntt/nP0ZJiZ+NosBXy1BekBeU+J9cv9aOXshI+s2o35WGj8+3aLI61fe+idZGg1bzh+mQZ8ed2GED4Z7/R75efV6egRXIyw9lX3POkrTxput1P7jKAB7v5hBpa++vOvjEPIXN3ce8XPnAhB54ngJj6Z0EJ8jgnBz4j0iCLdWmt8nRRmbyBwRBOGecJy9K4HVis3Zc0UE0ARBEIS7KKeslug5IpQSSv2l28sdES/n0iW/slpyrmuSOGAlyv+1V7FcvoxPx+dKeiiCIAiCIJQyIjgiCMJdlPeHv91oVC6L7CJBEAThbnJOTkqyLD5zhFIhv8bdRSE7p+HF67lUkbjFcRXHq0SpfX0p+9GUkh6GIAiCIAilkDiFRRCEeySfn4vih6IgCIJwF+W0uBKfN0JpIef6fzE4YyOiw3ep4vwTI+c6LMrfH7tdNB0RBEEQBEEopURwRBCEuyjvT38p901iskoQBEG4i+TiT0ELwl1xu0EN5ytafIUqZZToSM5NyrFCBGgFQRAEQRBKKxEcEQThrpELuKwQPxQFQRCEuyhnIll83gilw+2+FJWyWiJzpFRReo7kzhxRLsiI4yUIgiAIglA6ieCIIAh3T+5mo/k1HhWTVYIgCMJdJGd/9qhEBolQStyyN8UtONdTiTJNpUpOADbnuCg9j5BBNGQXBEEQBEEolcS3NEEQ7g0lOJIzHSDO5BUEQRDuplyfOCU4CkHIoWQYFHN9Z2aC+A5VuuRTVSun54iMOCFIEARBEAShlBLBEUEQ7ol8676LH4qCIAjCXZTTkL1kxyEIitt8MTrLaomXdGmT94goZf1kWTRkFwRBEARBKKVEcEQQhLvoFudFitkqQRAE4S7KaYgsPm+EUiKfDIPibUa8pkuT/Mql2ZVLsjhegiAIgiAIpZQIjggKOb+eEMJ9rXQdU5E5IgiCINxbzs9B8XEjlBaq/OovFYEoq1U65V9WK/vvj4zoOSIIgiAIglBKiW9p96mxY8cSHh5e4H/Nmzcv9LZSU1MZPXo0f/75510b74YNGwgPD+fy5ct3Zftt2rRx2f+IiAiaNm3KwIEDOXHixC3Xv9lz2aBBgyKNJTw8nDlz5gCwb98+wsPD2bdvX7H263YcOHCA/v3757m9TZs2jB07Ns/lmyn+fsh5Lrr8lBc/7AVBEIR7QHzaCKVN8RuyO6Mjd2wowh2gBEdyfbd1HmOVbBffeQVBEARBEEopTUkPQCi+oKAg5s6dm+99Wq220Ns5fvw4mzZt4sUXX7xTQysRLVu2ZPDgwQBYrVZiY2NZsmQJPXv2ZPPmzQQEBNx0/c6dO9OlS5c8t6tu40yvWrVqsWbNGqpVq1bsbRTXunXrOHv2bJ7b586di6enZ57Ld5uc54I461EQBEG4u+w4M0fE541QOuRXfqkonJPvKtHDonS5WUaQjOg5IgiCIAiCUEqJ4Mh9TKfTUb9+/ZIeRqnh7++f5/moU6cObdu25ccff+TVV1+96fohISF3/Pn09PQsdceoZs2a+V6+G1zLeomyWoIgCMI9ll/WoiCUoDv11Uf0sChdnAFYOddhyel5JIsArSAIgiAIQiklymo94I4ePUqtWrVcSiclJCQQHR1N79692bt3Lz169ACgR48edO/eHYDu3bszatQohg0bRv369endu3eBpZW6d++urAdgt9tZsGABrVq1ol69egwePJiUlJQ8Yzt16hQDBgygYcOGNGzYkCFDhnDp0iXlfufjrV69mtatW9OwYUP++OOPIu2/j49PkZa/lbS0NKZMmULbtm2pU6cOHTp0YP369QUun99zVtj93rNnD3369KFevXo0b96cadOmYbPZlOX++OMPXnrpJRo0aECTJk0YNGiQkikyduxYvvnmG65cuUJ4eDgbNmwA4PLly4wePZpHH32UWrVqER0dzejRo0lKSlK226ZNGz788EN69uxJ3bp1GTduXJ79MpvN9OnTh6ZNm3L8+HHl9nXr1vHMM89Qu3ZtWrVqxcKFX7mMOQ/xQ1EQBEG4i3ImJwWhtCh+5kjuk05EcKR0ye94OA+XJMsgiZ/dgiAIgiAIpdFDmTkiyzIWi6WkhwE4yl/dzplEVqs139vVajWSJFG7dm1ef/11FixYQMeOHYmOjmbChAnY7XY++ugjPDw8mDBhAu+//z4TJkygadOmyja2bNnCc889x4IFC7Db7YUe0+zZs1m1ahWDBg2iXr16bNmyhRkzZrgsc/78ebp27UqVKlWYOnUqVquVBQsW8Morr7Bp0yaXElhz585l/PjxGI3Gm/b/kGVZeT7sdjvx8fHMnj2bwMBA2rdvf8tx2+32fJ9PjcbxNjEajXTr1o2EhASGDRtGuXLl2LZtG+PGjSM+Pp6BAwfe8jGKst+jRo2iW7duvP766+zYsYPFixdToUIFunbtyqVLlxg8eDAvvvgiI0aMIDU1lZkzZ9K/f3+2bt3K4MGDSUxM5NixY8ydO5eKFSuSlZVFjx49cHd3Z9y4cfj5+XHo0CE+/fRT9Ho977//vvLYK1eupHfv3rz++ut4eHhgNpuV+6xWK8OHD+fo0aMsXbqUyMhIABYtWsSsWbN47bXXeOeddzh+/Dhz5nzKtWtXeffdwcjKNEDuPiSlqWG8IAiC8KCxy6KsllDK3EY/9tzriJd06aJkjuS6zSU4Kw6YIAiCIAhCqfTQBUdkWWbJkiUuZ+qXpAoVKtCnT59i/Wi/cuUKtWrVyve+0aNH07dvXwCGDBnC9u3bmTRpEv3792fbtm3Mnj2b4OBgAKUfRrVq1Vx6Y2i1WiZNmoROpwMoVDPu1NRUVq1aRe/evRk6dCgALVq0IDY2ll27dinLzZ07F3d3d5YtW6b0vIiOjqZt27YsXryYMWPGKMt269aNp5566paPvXHjRjZu3OhymyRJTJs2DX9//1uuP3/+fObPn5/n9t9//52goCA2bNjAqVOnWL16tRKkadGiBVarlfnz59O1a1d8fX1v+hhF2e8uXbowZMgQZZlt27axY8cOunbtyuHDhzEajQwYMEA5jiEhIfzyyy9kZmZSsWJF/P39XUqvnThxgrJlyzJ69Gjq1q0LQLNmzTh48CD79+93GWfZsmUZNWqUct157O12O2PHjmXfvn0sXbpUef2lpaUxf/58Xn75ZcaPHw/Ao48+isFg4f33Z9Ot27OEVC+bvbVcr/Xb6OciCIIgCLciMkeE0uZ2vvm4BFTEi7p0yee3nHJikCyL4IggCIIgCEIp9dAFRx4kQUFBLFiwIN/7QkNDlctarZapU6fSpUsXxo0bxwsvvFCoYEOVKlWUwEhhHTp0CKvVSuvWrV1ub9++vUtwZO/evURFRaHX65VsDU9PTxo3bszu3btd1nVmJtxK69atlWCCLMskJiayZcsWRo0aRVZWFi+99NJN13/ppZfyXcbPzw+A/fv3U65cuTzZK8899xzr16/n0KFDtGzZ8qaPUZT9vvFxQkJCyMzMBKBevXq4ubnRuXNnnnrqKR577DGaNm2qBD3yExERwVdffYXdbufff//lwoULnDlzhrNnz+bJmCnoOZ8+fTpHjx5lyJAh1KlTR7n977//xmg00qZNG5dtPfZYFAD79h2iY3VHVpIEaMuXB6sVlZvbzZ4uQRAEQbgtymSymJgUSovbeC3mTrhViRNMShXnUZVzRa1ceo6IhuyCIAiCIAil0kMXHJEkiT59+jwQZbV0Op3LBPXNREZGEh4eztGjR/MELgri4eFR5DElJycD5MmgCAoKyrPc5s2b2bx5c55t3JjlYTAYCvXYvr6+eZ6PVq1aERsby7Rp03jxxRdRq9UFrl+mTJmbPp8pKSl59gMgMDAQcGTN3EpR9luv17tcV6lUSq3p8uXL89VXX/HZZ5+xfv16VqxYgbe3N926deOtt94q8DW1dOlSFi5cSHJyMoGBgdSuXRt3d3fS0tJclivoOT9//jxNmjRh+fLlvPzyy0rWivO49+/fP9/14uOTyH2+o+YWGTaCIAiCcCc4J5PFNLJQWuRMoheda1ktMdlemjhjHy4N2ZWeIyBSfQRBEARBEEqnhy44Ao4fE0XNiLjfrVmzhqNHjxIREcHkyZOJjo7G29u7SNtw/gi7sf9IRkaGEkhxZlkkJCRQtWpVZRnn5LmTl5cXjzzyCL17987zOM4eH3dK7dq12b17N0lJSUogozh8fHy4cOFCntvj4uKAnH2/mTu533Xr1mXu3LmYzWYOHDjAmjVrWLhwIREREfn2WPnuu+/46KOPePvtt+nUqZMSjHnzzTc5cuRIoR7zP//5D1FRUbRv355JkyYpZcicr6Xp06dTqVIlZXmj8RI2mwl/f99iTQIIgiAIwu1wlrURE8lCaSHdRmNuGdGQvdS6Wc8RWRalZAVBEARBEEop8S3tIXDlyhWmTp1K586dWbhwIWlpaUyePFm5/2bZFLk5e2Rcu3ZNuS0lJYWzZ88q1xs0aIBer+fHH390WffXX391uR4VFcWZM2eIjIykTp061KlTh9q1a7Ns2TK2bt1a5H28mSNHjuDj41Oo4MXNNGnShCtXrvD333+73P7tt9+i1WpvWtLK6U7t97Jly2jdujVmsxmdTkd0dDT/+c9/AIiJiQHylls4cOAA3t7e9OvXTwmMZGRkcODAgTwBr4IEBgYSFBTEiBEj+OWXX9iyZQvgKPOl1Wq5fv26sl916tRBpVYzb95/iYmJLfS+CYIgCMKdIspqCaXOnWrIfifGItw5+fyNsec+YuJvkCAIgiAIQqn0UGaOPCjMZjMH/5+9846Tokgf99Pdk2d3ZyNLFJaMkkUEQVHM+TxRMaFiFs9TVMSf2TsV4wkS9Izo14BnwHDo3ZkQFRFQEAwgsOSwbN7JM931+2Nmenc277KwC9Tz+cBOd1dXV3d1dVW9b73vu2JFncf79OmDw+Hgrrvuwul0MmXKFDweDzfffDMPP/wwJ598MmPHjiU1NRWAr776Co/HQ9++fevMr0OHDsyaNYuUlBQUReG5557D6XSaadxuN1dddRWzZ8/G5XIxYsQIFi5cWEM5csMNNzB+/HiuvfZaLrzwQux2O/PmzeOzzz5jxowZzXoexcXFSc8jEAgwf/58Fi9ezOTJkxutBKqLP//5z7zxxhtMmjSJm266ic6dO/PFF1/w7rvvcuONNzbKEqel7nvEiBE88cQTTJo0iUsuuQRN03jrrbew2Wym27S0tDQKCwtZuHAh/fr1Y+DAgbz55ptMmzaN4447joKCAl588UUKCwvxeDxNehbjx49n/vz5PPTQQxx11FFkZGRw1VVXMX36dLxeL0ceeSS7du3iH/94AkWB3r270jwxgEQikUgkzaeqz3+JpE2wJ8qRJFm7FLa3JWqNOWK61ZIxRyQSiUQikUjaKlI5sh+ze/duLrjggjqPz58/nx9//JHFixfz9NNPmwLwSy+9lI8++oh7772XoUOH0qtXL8444wxef/11Fi1axMcff1xrfpqmMWPGDB5++GEmT55MdnY2l112GRs2bCA/P99MN3HiRDweD6+++ipz585lyJAh3HHHHdx///1mmr59+/L666/zj3/8gylTpiCEoHfv3syaNYvjjz++Wc9j4cKFLFy40Nx2uVzk5eVx3333cdFFFzUrz6o4nU5ee+01nnzySVMJ0L17dx566CHGjRvXqDxa6r779u3Ls88+y6xZs5g8eTK6rtO/f39eeuklunfvDsSUOQsXLjSVOVdffTVbt27l3Xff5Y033iA3N5cxY8Zw0UUXcc8997B+/fokV2j1oaoqDz74IOeeey6PPvooDz/8MDfffDM5OTm88cYbvPDCC3g8Ho44oj833HABKSluvI2+O4lEIpFIWgYZc0TS1lBbyOZDCtvbFvW5OYspZ2V9SSQSiUQikbRFFCHEfrmUTtd1VqxYweDBg+u0CAgGg+Tn55OXl1cjuLVk7yCEwO/343K55Io2CV7vWgwjBECF2p5Cw02q30f39jUD2x8s7Gkbkd81yYFOY/p3iaSxjFi4go0GzJ47iz+/8nxrF6fFkO1k/+XjF1/lqu4D6eEr59szjmnSuQHdIO/rnwFYXrGFTmeduTeKeECwr9vIsgX/4QxnLulBP7+fehQAv3oDjF26hoyyUj5f8l86PvLwXi+HRNJYZD8ikdSPbCMSScO05XbSlLLJhXQSiWQvImr8lCoziUQikewrZEB2SVtDqSVwd2NJijki3+k2RW31keTWT9aXRCKRSCQSSZtEKkckEsk+Yr80UpNIJBLJfkyi55EDXklbYU9E5KLKWKo+N06SfU+l0qtqzJG4clYA0g2aRCKRSCQSSZtEzhUlEsleRCpEJBKJRNJ6iMql2xJJ22APArIjA7K3XWqp16qWI7K+JBKJRCKRSNomUjkikUj2GrVNECUSiUQi2VeYwkkpmJS0Fcx3senvZNWxlCpncW0K05KnSrWa9SVkQHaJRCKRSCSStoocVkskkn2MVJNIJBKJZN8gDUckbY3E5EvGHDnQqGk6Yv60CKnNkkgkEolEImmjyFGaRCLZe4iatiNyKi+RSCSSfYW0HJG0OZTmq0ekcqTtkqgOoVSNORL/myIIuUv3faEkEolEIpFIJA0ilSMSiUQikUgkkgMSMyCyFCRL2gimEL05brWSFp3Id7otURmQvRKjylbIXb6PSySRSCQSiUQiaQxSOSKRSPYi0oWWRCKRSFoP6VZL0tZIxKaQbrUOLNTalF7xClMQ8iMkkUgkEolE0kaRyhGJRLJPkGoSiUQikexrpFstSZtjD17FZOXIHpdE0pKYFVJZS5XKWYGQs26JRCKRSCSSNomltQsgkUgOZGqqRHShsL00gMumIYBI1EBTFRRFwaIqGEKgKLF1lTaLisOq1Z6zEBgi9tei1T/jTLihMIRAVZQWEZLpRixPTa2ZV+J6gpgMRFEUhBBEdIGljqIm7ketEs8zkXN95RVCEIoa2C0qiqIQ1Q02FvnokZMihYESiaQGhiEI60ad39bWIL/Qx/s/beOaY7qTYm+ZoWl5MML364sQugDDoERxEIzolPojLMkvYli3TFIdFn7dXo43GKXEH2ZbacC8fqrDQlgXHNMrm65Z7hYpU1unxBem0BuiZ7uW7T+KfWE0VcEe79OFEEQNQUQ3UFBwWNUmX29XeRCHRcPjshLRDQoqQmS4rBSUh1AVhS6ZToSAtQUVFFaECUZ0nDYNj9OKVVMxhKCDx0E4ahCMGKQ6LOhCsKnIT4bLisdpxWHVKPSGaJfqYP1uLxluG53SnWYZQlGdjYV+OqQ7KPNHsFlUdENgCEFUF2wp8dMjJwWrplIRjFAejKIqsCscGwgIwB+O8v5P29hdEWJUz2yGdc2ov88HEAIlqLNTV/AW+fl9ZznpLhvbSv2Eowb9OqQxsHN6jXMNQxCI6AQiOjaLSiRq4LJZqAhGcNktlAciVASjOKwqgYhOr3ap+MJRygMRvKEoKXYLnTNcbNjtxaqp5KTaKQ9GiOgCq6aQYrfgslnQ43Ub1g0iUYOoIchw2bDFB0C6IVhX4KV9mgO7VSUY0fE4rbXedygaa7NuuwW3TUM3YuM4Qwi2lwZj9WmJjSsjeuzZp9gtWDQVPaqzyxdlW0kAzaJh1RQcVo0yf4RQVCctXsdpDisAZYEIbpvW4JiyyBtiV3kIRYHcNAcVwQiZbhsBHdANDAHzf9rGrvIguZ1TUbwRIuUGr5X2I+M/a+iY7uTPQzs1+xsshMAbipIaL3ddaSpCUdw2C5qqUOoPo6CgqhCI6KQ5rOb1hRDsrghhCHBYY23UblHxhXVKfGE6ZzjNuinyhij0hlGUyngqigJ52W4savL4Ohp/B8Lxd0BTFCyaglVTsaiKOfavCEYo8oZJd1lJdcTac9QQRKIGJf4wmW4bHqeV8kAUh01FVRS+WrObzhlO3DYLoageu3683oQQBCMGDqtKKGoQjOiku2wIISjyhUl1WIjoAlUBbzCKVVNJddTe79T2TtbWOmtrsru9Iexa7BvVGCK6waYiPxCrh1SHlRJfmAyXjTSnhbJAhG2lAYIRHYDsFHtS31TsC2MIQThq0MHjqHTzVsUVnzcUxRCwuyJEXnbs3PKQgT8cJWLo5hykPBBF02JzF0VR0HVBl8zK98AfjhKKGKiKQiiqm3mn2C2UByNkuu24bBoR3cBmUbFbKt91IQQl/ggZLmutZSyK3/PGIh8dPA6smopVUwlFddYVeOmRk1Jr24nE3zWXTWPNrgqsmkrnDCdWVUWtZa4WjOhEDdGoMYcQAiFi318hBILYnBKgzB/h1x3ljO6ZjUVTiegGQkDUiH0LnFat1rlicwlHDQJhvdHv1YFG4tuWarfIea7ERLrQlbQEikh2XrvfoOs6K1asYPDgwWha7YPLYDBIfn4+eXl5OByOfVzC/Y/EAGhP8/D7/bhcLvlx2s9oifqvTkXFLwhhAFCm5FAs0rBWhIiKxl/HbbOQk2onFDXwhaJEDUEoqpuKEYi5qFCUeqxT4gNZwFS8NESNNNV2GAYIqjwzEduuNa/4NQ2z4wZVUZLi1Yv4+Uo8QeLe1Cp1oiigR8Ls3r6Fv39dyG6/QI8LO2KTUojosfNS7BYMIcxr7g1sWkygpRsxwUjVCYw3FE2+v6onVitS9edWvcgum4YQEDGMFi2/pO1iGAaqKpfZ7g10IyaYdtssRKu1qSy3Haum4A/HBKgRvXltTkHhqqPzuPWkPo1Kf+JTC/mjwMu4wzvzxHmDmnXN6lzywhK+WVcYs5Hew0/HpSO60inDyQ/5xeQX+thZFkRQKSwBzD6g+vdLUeJulJSEsrxx12xMT2UYBloDgtzK/BrGF44Juaya0mLCHCEgFK2sAKummP2UWTYFNEWp9Vl5nFZ0IybwC8ffx6p52jQVXQhzwcLeRlMVczHDniCsKlanQrRcT9pvs6io8edgURVyPQ6KfTHljiEENotKmW6ghOt/qQd29nDGwA6sK/Dy3foidleEkuqhuVhUhWg9N68q1PpsEuOeuurJpqmxSq8ylhICdFHZpqrmXdd1moM9rrQJRWOCXFu8TXVMd1AeiFIaCJtjzpjQs+4LV13YUh8Oq8rVR3fn36t2sL000KTyGgaE9cSimDrSiJgQVVXAoqpm20mgxZUTiXe5vvZj01QS3XEwUv87ZFEVVDW2UKcx9WNRlaQ6rqp0qUpif131nrjPxDsU1g3zG6YbArtFRRB7Jvsae5VVUTWKXmVHQpje1LwTz6b6d7Z6+6mt7WqqElc4NtyeLKpizkkihlFrPdWGqmAqRqHy3aytjIlyVC+PzaKa75SigFVTzfZYma8wj1cvm6rE5mJqlT4mkZ/Noib131UVIE2dQtktao3vrKIkvwPm/lq+FLW15+q7wrpBRBf1tv8DmcT7U/W7tNev2cw5SZrDSkUwmrSvLlmBJBmrpsbbjUI4qsflC4Y5DkqxWwjGFf3lwShlgTCOuCIyoUA+GDmqRzYvXjZsn8thGyObby2aUrYmL88zDIN58+bxxhtvsHXrVjIzMzn++OO56aabSElJAWDTpk088sgjLFu2DE3TOOWUU7j99tvN4wA+n48nnniC//73v/j9foYNG8add95J9+7dm1qkg5KpU6fy/vvv13k8Ozubb7/9tlF5lZeX8/e//53zzjuPI444oqWKmMR7773HnXfeyeeff07nzp1bPP+xY8eybds2c1tRFDweD0OGDOHmm2+mb9++9Z7fp0/dAhyXy8VPP/2UlPbGG2/kL3/5S9Lv5nDppZcC8NprrzXr/JZi9uzZ2Gw2rrrqqhbNt8YAQIgkxYimKAhig0aLphKOGlhUxRyYBiI6vnAUX1HywKK26zR2AJsY8DaZOk5qjH65+jUTk4E6L1PlWJJyQ1RamISjBt5QZedffcLjDdX/zFqChibJLUV5cO/fi6QN0kzBvKRx1PaN2NZEIV19bC72Nyrd9tIAfxR4AXhn+Vb+/qf+e2zV8tPmkphiBBpUjFhUBY/TSrrLyqEdPRR5QzitGoYQrNvtZUtxgNe+37RH5dmrjiX1lp8ERnRRQ4HRknlXRwiImn1d8vFgJFRvftWFvk6rZq48bwx1CWMTVFfmNFYJU9uCBZtFpdgXjl03YhCNxNJmum2M7J7Fl2sK8IeT67Mi3jYSBCNGTP6rgBaXxPdsl0IgotMhzYnNovLNukJ+3lrGz1vLGlXWqvfqsGo1hDn2uCCiNBCp8VyrKz3qElwb9Yx7oGY9Vr+GEMl5V7+O3RJbsV3X9ZV6FDNVhZnhqGEK0Nfu8taavjpV35Gq4oh2OS527faDpmBJERynbyKr72jmLdtCMGLwzBfrGpV/XTRG2WWI2p+tbiQrFOt7RmHdgEZ+ZqKGaJLWKvE+OawqwUhNgbvNEpsbJPbXlXVt91n1XlpCMdhcWvraVRUHdeVd23e2tm9iXe3WZlEx4gspks+vvQI0VcFl1aioZVxhiNrnC7WV0aijnqsqtYSoX8lV22fGiGs6Yq9xcoKWVJjVVh+ijvvfG9c6mGjKd6lFaMacpKHxi6RughGDijqO+cMxi1KAQm/Y3B/RpbxgY5EPIaTb1ebSZOXICy+8wNNPP82VV17JyJEjyc/PZ8aMGfzxxx+89NJLVFRUcNlll5Gdnc20adMoLi7m8ccfZ+vWrbz44otmPrfeeisrV640lSYzZ85kwoQJ/Pvf/8bj8bToTR6o5OTkMHPmzFqPWa2NN7X87bff+OCDDzj33HNbqmitwpgxY7jhhhsAiEajFBQU8NJLL3HZZZexYMECsrKy6j1/3LhxnHfeeTX2V18pMG/ePNq3b1/j9/7M9OnTufHGG1s0T5G0pDY+WI2vdLQpCrlalc+PAKKAosbPiX3RA5pCsaHHLRJiLhsMXWDVVBzx1ZWxeZgwV4gJA0L+CJqmYnXGBGwKEIyvMmmwsxC1/qxBYiJpiKqhN2MzeMMQaIlVQgKMuPImsWorFAphs9lqmHknXEUYRsxVWMIVV9UyRyMalNuYdeZAhGLBEJDptOKLr5LwhqOk2S3s9oVJs1tibgOKQzjTbFisLbvExhuKIqgURkXirjTsmorbZkGt51lXX9FQPWnVw7t9YWyaitvWtlYiSPYOuq6bVp9tbfXJgYBCTMjhj+ikVnEnIQTs8sZcxTgtGg6rilVt3spEBejXJ7NRaX/IL07a3ljko2/7tKZftAoJoXAPYWHT8HS8KRb+36cr6NBtKJpuoAgFv2GgWVXSHBb0uFDO2CkQhs3Mx6uk8prTwJVmIyvFTm6Kna7pLoZ08JirNqs+HkVRMHQDf1mYlAx7zKUiya4W66MhXXtVNbuhG2Y7UU13Mo19QrVjURU8DiulwcieZVQNl1XFWxrG4baye5ePtAw7KW6b2c8FogaGIWo8K90QbCsP4rZpOCxaUh+QYrMQDOl4vSGcqTZSVJVQWMdtt1BRHqaoJEAwpNOrTyYuhwV/RMcZV7rphkgSBAsh2Lmtgqhfp1PXVLwVEQIRHdUQGL4oWpYdt8NCBDBUUER8eTrgsVkIRg0sCoR1gRa3ATUMg0BhCK8/Qrd+mdji7sRQ4LNP/sftlna4SwOc2akdI4QNR2GEY9JyiDpUFIuCw2UlraObbWUBOnVKJTPuQmVtqZ/JO3ZiZNqZ/k0JzvbZaEIhbETBF3s26Z40NloN2nsc5KY46J7p4vBOHlxWDadFwwhE8QajGAEdnBqpLgtb15WRlmqjtCBAwQ4vDreVnL4ZREtDiIBBNGoQwsBwaKQ6rZTs9GF3WIiUhqkoC2FxajhzXQgBaR4b7bumoeoCxRD4SkJs2lZBZic3YV+USCBKeqoda7qNYHkYp8tKsS+MxaoSDkbRVAVnio2SHT6sFpV26Q4KC/zoNoVo2MBXGsaeaiXTZSOk67TrmU60OIS/JIQ9xUpUN9AcGiFfhKKinXTs0AmLplK+O4A3pHPoEbnY7BYCuo6qKFSEo+Y7VRqMmGPGXd4QHptGyR/lREM62R3dBMrDRPxRevfLIj3XReFuP05VxRfRWffLWm7KcGLb6eOetBw+dEXZPCgdVMhgB8cu3Y3+YwXj0zz8oIfo2s7NUV0zOaJzepKVcEMoxL7RW7eUY3dZcKRYEQYIw8CIgrckiGZT6doljaKSIEW7fGhhgctjI1QeQegGWoadgC8MQsHhtqB4o7hdVtwZDlSXRnlFmO0bynAoChFbrD503SDijdKlYwppOU42/1qMxaqiqAreUJR2eWmoNhVDxBY/+YuDRH1RnHYL7bqkEgnrVJSHcKTbEIpCOGJQuMNLNKCTpqmEIgb2dg78JSFSMxx4shw4rRq7NlfgFwZZHgdBBIpNJd1mwV8aIjXLiYKgKBBJsiRIsVrwR3VA4LRolPnDhIM6aR47a7aU0TM3FQ2Ihgw82Q685WFURcHuttT6na7P4iO2WfMsh0UjENHrVfxVJ9MV63+icfd8lrhlRyhqEPRGcBiCtCwXJTt8VESiRDWFlMxYX+O2aqaL4rJgBFWJWfFoceVixBCk2S0UFwbwbveR2i2VQCDMjvzN9BrUC7vNAgKiwShui4bNbcUQgoAvjN8bIerQiASiKJpCapodu6aacy8FcFg1fOGYm7Jo3KrdbdXw1mKFmma3UBGKmmVUiY37o4Yw22GG0xr7DisKgfgcJ9Nlo9gfVzDHrfKrWhVZ426tHIrCpjUlpLd34fTYMapYgkQiBkXbvaRnOUhLd5irzKvml7DiV6hi8SnAXxEGHTSLgmZRCBuCVYVeDu/gYfc2L85MO540OxC3ygnolJYFccTrCGr207VZzwcqwvhKQwgBmk3F4bZi6IJARRg9qKMEdXL6pqNHY3NtzaqwbW0ZqRl20nIc7N7iJSXdTiSom/6anSlWygsD2JwWMtq7ACjbHUCPClIz7HhLQ1QUBXF7bLg8duxuCxVFQcKBKDanxXzOaVlOCrd6iUZ0cg5JRdNUlD2YXtY7bqnjWKrNQnm4nnHKHoyFatSPYZC/MZ+8bnkNu7ysgi4Exf4ImS4rWoP2hM0scD2nNZhjc89tRn0lCAUiWGwarjRb/QmBilAUf0THqirxBUtxSzliitWd3qBp5Zlmt5Bqt+ANx1wDOixarXMXIQS+0hBWuwWrQ8NfFqKsIIA7w07xdh8deqZjc2qocZePhiEoLwyY27pu4Ey1EQnpuFJtqNqeayCEEHiLQxhGbK5h3UPXwt27emp1IyhpHE16+oZh8Pzzz3PBBRdw6623AnDUUUeRkZHBLbfcwurVq/nuu+8oLS3lvffeIzMzNjHOzc3lmmuuYfny5Rx++OH89NNPfPnll/zzn/9kzJgxAAwbNozjjz+eN954g+uvv76Fb/PAxGazMXjw4NYuRpshMzOzxvMYMGAAJ5xwAp9++ikXX3xxvee3b9++Uc+zahr5/OvG541Q4M9CFxq60IgaGkp8iYdDh0i04eUeFqAdSmwgqoHuD6MSWyjia+DcKBCqtuSgZcU9e0aUYLPOi0TDhMojrPpoLcGyg3vVkOTAZge/tnYRJHvAlkHZnHb9wAbT7SpKtjDZXujfY+XI5q3lAKSEBYbLCjaVQltPtN9KaqQtaiCvs1FhR5RYr+IjSjFL2bpH5WtJdsp2Ui9rF2xuUvof90IZVny8MWl7a1Y2+gmpqF4X3f9dTEEd5+2I/11j/gKvXcH4UwYAhbsU2FXzDe4X/xcbKfkIUsS3bGlyuTct3920E34rNX+uquVw02qikjV17E/k99v/6m+PW9iQtJ2/eGejr72tyu/dVD6PbUtq1lqJW8U4w46eZiP47m78R7ohMyYoVeISN29xiC7F0AUFtvkJ/+Tn2730PamtDlqCDXXs/30vXW9vsKPhJAcV2xetbu0i7Nd82QrXXP9167/F61u7APuQXfzWrPMOpme0v7NtUePHBm2VnzLsXPrQUVJB0kyapBzxer2cffbZnHrqqUn7E66wtmzZwjfffMPhhx9uKkYARo8ejdvt5uuvv+bwww/nm2++weVyMXr0aDNNZmYmRxxxBAsXLtzryhEhBIbRcq4j9gRVddarhd5TVq9ezQUXXMCZZ57JtGnTACgqKuKMM86gb9++XHvttVx22WUATJgwgeHDh/Paa69x6aWXkpubSzgc5uuvv2bIkCFcd911TJgwgVdffZUjjzzSvEZ111CGYTBnzhzefvttSkpKGDVqVK3uutauXcuTTz7J0qVLARg5ciRTp06lS5cuACxZsoQJEybwwAMP8Nxzz1FWVsYzzzzDqFGjGn3/LW2FVFFRwcyZM/n888/ZtWsXXbt25fLLL2fcuHFmmrFjx3LCCSewZs0afvrpJ84880weeughtm/fzsMPP8zixYtxOBx1urD617/+xSuvvMKmTZvIzs7m3HPP5YYbbjBXUE+dOpUdO3bQrVs3PvroI9q3b89HH33EoYceyl/+8he++OIL1q1bxzXXXMONN97I0qVLefbZZ/n5558JBAK0a9eOc845h0mTJqGqqulSbObMmcycOZM1a2LT0IbqpyG2lgcIC2eN/QqQk+FslD9zPWoQqAhj6CK2OgZiH/sG2oyhGyhqywRebw6KGrNgqQshDJRmLrNRhYKiKrg8Nixq41aa2JwWwgFpairZP4it7AtjtdqkWfD+igIdeqY3KumWLcla7A2byxnbf88sMrfuiLnDcQrFXNHYPpxPTrf+dO2fRUZ7F8IQFG334SsJYXdbyO6cimaJfV8TGLpgy6/F7FhfitEEN1P74pu7v7WTxDOx2jWiYaNRLikbRAGrTSMS0lG1WN2pqoLDbQUFgt4IkVAj/W7EVwYrioLVEbc6VSAtO7ZKVrOoGHrMwkWJx0chESeltu34excORDH0+Ir2+D3btXDSpW0OjaGndCUt24mvNIS/PMy2NSUEKiIoauw9VBLLme3x4gpBx/aQ1a8zdqeFjA7xWH8CNq0uYtvamorABJa4BU5Khp2gL0I4qOPJdsTHWgag4C0JomoK7bqmkZbtQNVUImHdHJM5XFZzpbGqKQQqwqTnulBUhYqiIEIIrHYNVVOx2mOrOMsLAyjx+nGm2ghUhFE1haAviivNStAXxe2xIwyDaMTAmWLDYlMJB6I402yEfFEMQ1C0zYsrzUZGBzdBb4Ti7V5sTgvtuqURCUQJh3QqioI406yUFJSTkZ1GKBCN3UMwiq4LhBH71xisjtiqaVeajaAvEqsPVSHki6BZVFSLijAEVkfChYqgc98M2uc5WSUqF8IclrmV1BP6EwlF+WPpLkp2Ns71YF1olsq4C6qqmG3A7bHhKwsTDkSx2DQ0i4olvvrc7rYg9Ni3L6ujm0jYoKI4iMNtJeSPmG3TYtPI6uQmWBEhEtax2i0EKsK4PDb8ZeHY84y7jjOMWF3rESOpvbnSbCiqQtAXQcQtqi02jUBF5fuvWVVsDgueHCehQJSSHb74KnWFcCCKEAK7y4rFpqJHDbM9QS3fWaXqTyVpn6LE3vtwIIo73U7IH8XQDVRNJRrW0awxy3W9DbsqUlQFzaISjeikZTkJ+WPPp8l9jQJGVKBqsXcmHAmjKVazH1HjVpF63BWUoijm87c5LUTDBtFw2/fpL0RsPlbbanZFVUxXxU3BkWLFYtUwdANdFxi6wIga6LqBqioxC64qeSqqgsWqNr4fiqNqCu50O3aXNWaNWh4mGjZi7cQfxVcWMtsUCggjNk9O1GHMmkxgtVtQVAj6okTDeuybV2n4aM6TFTX2nQv7o9icGoYB0ZCOM9WKzWnBXx42v5eRkA5K5Xxc6KLh59jAGKXeGGv1eiGoP9/6L1r3yVWPCCAcCmGz2+MWRfVl2fwCNfvUBk5sbr71n9dghdaKpimEAlHz29IaWO1arN3qlZbDKLFxltUWs/CtOt63Oizm2E1RFfRI7DtYtR/bU+wuC3aXNd7O9+zb2qlPhlSM7AFNUo6kpaVx991319j/2WefAdCzZ0/Wr1/PaaedlnRc0zQ6d+5Mfn4+AOvXr6dz58413GUccsghfPTRR026Ab0eX8u6rscHjZUfbSEEP/54AWXle2NtWNPxeA5n6JC3mvwxTdxTJFL7WnhNi5nWHnbYYVx99dXMmTOHs88+mxEjRnDvvfei6zqPPPIIbrebe++9lwcffJB7772X4cOHm3l/8sknnHnmmcyePRvDMJKeYdVOsPr+6dOn8+abb3L99dczcOBAPv30U5588smkNPn5+YwfP57u3bszbdo0dF1nzpw5XHjhhcyfP5+srCwz35kzZ3LXXXcRCoUYPHhwrR1wTOFlmM/DMAwKCwuZPn062dnZnHLKKQ123FXPr4rFEmsmwWCQCy+8kOLiYv7yl7/QqVMnPv/8c+666y52797NddddZ5bl9ddf54orruCqq67C7Xbj8/m4+OKLsVqt/O1vf0NRFJ555hk2b96cdE/PPfccTz/9NJdccglTp07l999/55lnnmHHjh089NBDZv7Lli3Dbrczc+ZMAoGA6frr2WefZfLkyeTl5dGpUyd+++03Lr/8ck455RSeeuophBB8/PHHzJw5k7y8PE4//XTeeustxo8fz7hx4xg3blyj66f+ZykIx+8pw16Gpur4SMWvukgL+XGlNN7tmyvNRjRiIAwRMw9vYddQ+xohBIFAAKezeYrRYFCjxG/jrL/2wuFw7IUSSiSti67rrFq1igEDBki3Wvs59Y3REhSXJVvRbd5Z0ajz6qOoIiak7HdYNivi3U2f0HKOvf3SpHQ9Ds9pMK+ewxpO0xrIdrL/snDeuzxLJk4lwHET+tClbyYuT3U3E3m1nrsrFOFvS35HEYLjj1VxH92jRpruQ7P3Qqn3PyrbSP990kbWfh2L9aigc8ZfhvLf37dAQTC+T2C3RMgbHBs/9z4yd6+XRyJpCNmPSCT1I9uIZH9jT+dQe3LN1rh2QzSlTHvm1AxYuXIl//znPznuuOPo3bs3FRUVuN3uGuncbjdeb2wlX0VFRVJw9qppfL6GnOUks2pV/UbDFouFQCCAYcQ0lCLuz7+tYOgGfr+/yUJSXdfZvn07/fv3r/X4zTffzIQJEwC4/PLL+eyzz7j//vvN34899hipqakAZoD0zp0707FjR/x+P4ZhYLFYuOOOO7DZYhO2ZcuWARAKhfD7K1c7JZ6t3++noqKCN998k0suuYQrrrgCgMMPP5wdO3bw3XffEQwG8fv9zJgxA4fDwaxZs8x3YdCgQZx55pk8++yz3HLLLYRCMeHGeeedxzHHHGNer+q1Ewgh+OCDD/jggw+S9iuKwt///nccDket51Vl9uzZzJ49u8b+//73v2RnZ/Ovf/2LP/74g5dffplBgwaZ9xYMBk3lk8fjQQhB+/btkyyg5s2bx44dO/jXv/5lWlr17t2bs88+G8MwzGc3Z84czj33XG655RYAhg4disvl4sEHH2T8+PH06NEDXdeJRqNMnTqV3NzcpGcyZMgQLrjgAvO6H3/8MUceeST333+/qUAZMmQIn3/+Od99953ZbiFmvdW7d+9G1099hCJxDTsCjz3m3iSCEz8KCqLBuqgTnbblG2sPCASaZ70WCoWIRCL8/vv+5MBAImk6DfXvkgODguLkoMf524tZsWLFHuVZ5I0JJFW86IYLNJVgILjH+bZFZDvZ/9i9ezd06I0hDAL2nazNb7wrhyIDQAMhWL9hA0Z8LC+pm33VRnZv2Qxd+oOAFStWUBJQgMSCHkGFt+KA/AZJ9n9kPyKR1I9sIxJJw+zv7WSPlCPLly/nuuuuo3PnzjzyyCNA/eaJlYGoGk7TWOrT4gaDQTZt2oTT6UxaYT1s2Nv7vVstTdPIycmpVZgP0KFDB1wul7n92GOPcf755/Pggw/ypz/9ibPOOss8Zrfbzb+Jc1RVpUePHqSnp9ebLpEWwOVysWzZMqLRKCeeeGJSmtNPP53vvvsOh8OBy+Vi6dKlHHnkkUnu13Jychg2bBhLly7F5XKZ1xswYEBSXrWhKArHHnsskyZNAmLvWHFxMZ988gl33XUXuq5z/vnn15vHeeedV2uajh07YrFYWLFiBZ06dWLkyJFJx8855xzmz5/P2rVrGTNmDIqicOihhyaVedWqVRxyyCFJyqzu3bszaNAgFEXB5XKxfPlygsEgJ510kqmQAjj55JN58MEH+fHHH833PT09nby8mqsKDzvssKTrnn/++Zx//vmEQiE2btzIxo0b+f333zEMIxb4qUpaq9VqbjemfuojVBYColjUmppaBRo8/0BmTy1HVFXFarXSs2dPaTkiOSCRq7QOLgLvfwFArsXCrmiUiqi6R/G8dN3A99b/QIOhA/P4P18hEOt3DqQ4YbKd7L+U/xGL2qCqTX/Xd4QisOR3FAQ9evTEfQC90y3Nvm4jf3j9YACKwuDBg8n4fQsUlAKxsW+qJ4U+sr4kbQjZj0gk9SPbiETSMG25nSTK1hiarRxZsGABU6dOpVu3brzwwgtkZMSCA6akpNRq/eH1es1V7ikpKRQWFtZI4/P5TGuGxqJpWp0VkHAtlfiXQFEUVLWmdcv+hKIo2Gw2Bg5sONgpwKGHHkqfPn1YvXo1Y8eOrfE8En+r/na73Q2mq76/rKwMgIyMjKQ07dq1Szq3tLSUBQsWsGDBghplzczMTLpG9XLU9TwyMjJqPI/jjjuO3bt388QTTzBu3Lh6G2tubm69z7OsrIycnJwaZcnJibncqKioMMtdvcxlZWU1ngnEnkthYaH5TACuueaaWq9fUFBQZ/4Jqu8PBoP87W9/44MPPiAajdK5c2eGDBliugqrXo+J7cbUT32E4z57NbV2P7itFQukLVG9HTX1vPq+fRLJgYB8xw98hBB4wzqo0D3Lza5dZRQHo3tU70FvlED809q5Yyrij9h4U1WVA/J9ku1k/0ONByQR0OS6U+OLThQBqqz7RrGv2oiqaWBU1mvSGBsBB+g3SLL/I/sRiaR+ZBuRSBpmf28nzVKOvPjiizz++OMMHz6cWbNmJSk08vLy2Lx5c1J6XdfZunUrJ510kpnmm2++wTAM0+oAYNOmTfToUdN3rmTPmTdvHqtXr6Zv37489NBDjBw5krS0tCblkRjkJ9xoJfD5fKYrtYSSrKioKKkuE4L/BKmpqRx11FGm662qJAT3LUX//v357rvvKCkpITu7+X6YPR4PmzZtqrF/9+7dQOW910ZGRkat51Z9Lon6eOKJJ+jWrVuNtM0p+0MPPcR//vMfnn76aY466ijTYqO69Ut19rR+EsoRi9L2/A5KJBKJpG0QCekEDANUOLRrOot3lVGu6+hRIxZotBlEIzoBJWahnOm2m7FY6w34KZHsQxIy8z118qvIoJttCjXhIaFKEPBKRIMxbCUSiUQikUgkrUOTZ55vvfUWjz32GKeeeiovvPBCDUuPUaNGsXTpUoqLi81933zzDX6/n1GjRgEwevRofD4fixYtMtMUFxezbNkyM42k5di2bRuPPvoo48aN49lnn6WiosIM7g2NX7WWiD2xc2elb+SysjLWr19vbg8ZMgSHw8Gnn36adO6XX36ZtD18+HDWrVtHv379GDBgAAMGDKB///688sor/O9//2vyPdbHqlWr8Hg89SovGsMRRxzBtm3b+Omnn5L2f/jhh1it1nqtTkaMGMHWrVuTTLqKi5P9qg8aNAir1cquXbvMZzJgwAAsFgtPPfUUW7dubXKZly9fzpFHHskJJ5xgKkZWr15NcXFxkpKrqpIS9rx+ovG4PrW51ZJIJBKJBECPGITiAsP+3dIB8KmC8sLmuz71BaJE43lmuG0IYn1dJDVYz1kSyb5jTxR1prJPiOrSd0lrU091KIBQ2k7MS4lEIpFIJBJJJU1aor97924eeeQROnXqxMUXX8yvv/6adPyQQw7hoosu4v/+7/+44ooruPHGGyktLeXxxx/nmGOOYejQoUBMyDx8+HBuv/12br/9dtLT03nmmWdITU3lwgsvbLm7O8AJh8P1Bvbr06cPDoeDu+66C6fTyZQpU/B4PNx88808/PDDnHzyyYwdO9ZUcH311Vd4PB769u1bZ34dOnQwg3QrisJzzz2H0+k007jdbq666ipmz56Ny+VixIgRLFy4sIZy5IYbbmD8+PFce+21XHjhhdjtdubNm8dnn33GjBkzmvU8qisbAoEA8+fPZ/HixUyePHmPTbz+/Oc/88YbbzBp0iRuuukmOnfuzBdffMG7777LjTfeWK8lztlnn82rr77KjTfeyC233EJKSgpz5sxJUlBkZGRw1VVXMX36dLxeL0ceeSS7du1i+vTpKIpSZ73Ux8CBA/nkk09488036dGjB7///jtz5sxBUZSkgOBpaWn8+OOPLF26lGHDhu1x/RjxuEKqtByRSCQSSR3oUYNQXGDYKzc2FokosH17BRntm+f+tNwfd+cowG3TiBIGLBSNWNsSRZZI9pg9cS0qqv6SypE2hRlbM64lqVo7irQckUgkEolEImmzNEk5snDhQoLBINu2bePiiy+ucfyRRx7hz3/+M6+++ioPP/wwt912G263m1NOOYUpU6YkpZ05cybTpk3jsccewzAMhg4dytNPP43H49mzOzqI2L17NxdccEGdx+fPn8+PP/7I4sWLk57tpZdeykcffcS9997L0KFD6dWrF2eccQavv/46ixYt4uOPP641P03TmDFjBg8//DCTJ08mOzubyy67jA0bNpCfn2+mmzhxIh6Ph1dffZW5c+cyZMgQ7rjjDu6//34zTd++fXn99df5xz/+wZQpUxBC0Lt3b2bNmsXxxx/frOexcOFCFi5caG67XC7y8vK47777uOiii5qVZ1WcTievvfYaTz75pKnA6N69Ow899BDjxo2r91ybzcbcuXN5+OGHeeihh1AUhfPPP58uXbpQVFRkprv55pvJycnhjTfe4IUXXsDj8TBy5EgmT57c5Hg8AFOnTiUSifD0008TDofp3Lkz119/PevWreOLL75A13U0TeO6665j9uzZXH311SxYsGCP6yeh8lEVQTFZ+HHhIL5qV04OJRKJRAJEIrppOdIuzY5DUQgKwcZtFRw2tH2z8gyGY8oRi5IQVsYuEE1pvjWKRNKiJNxqNUO5UWk5UiUjSZtASThkqKVaFAxZXRKJRCKRSCRtFEUIsV/a+Oq6zooVKxg8eHCdFgHBYJD8/Hzy8vJwOBz7uIQHJ0II/H4/LpdLBt0+iFmzrYyQEOQ4iyiwdgBAI4qOhcyAjy65Oa1cwtZjT9uI/K5JDnQa079LDgw2byzlmGe/BeC3B09h7MNfsCMY5s6enbj2qsHNyvP75TsY/68fcaDw+7TT6PrFYkKKk3+IG7hw7HctWPrWRbaT/Zdv35nPuVndaOf38fPpTXMnvDkQYvj3v2EPh/g1XcF95PC9VMr9n33dRvK//4GRARuqYbD9+KHc9Nsm3t5ZAkA3sYEXd3/AYRfM3evlkEgai+xHJJL6kW1EImmYttxOmlK25kW7lEgkknpIqFwNpfITo8TXO0qfyxKJRCIBCIUqXUvaLSrZLhsA24uab+URDMUtR8w9CSW07HskbQQ1OXB3U0iOOdJyRZLsOWq1BS/JsWVEwtxHIpFIJBKJRNLGkMoRiUTS4iSmfzqV2lkj/rkxHFH2U4M1iUQikbQggbgLLBVQVYVcjx2AgvLmB08PhmOxriw1/P/LfkfSNmgRdZ0ARZXTuDaFUrfSS0HIL5BEIpFIJBJJG0WOqiUSSYuTmABGlcq1u0bS58ZAIpFIJAc3wVBMkZFQo3fMigVhLwyEMfTm9ROhGsoRiaRtUd3CoCmYliMyIHvbw1SO1JxeKwCq/Bq1JoYwWLpzKeXh8tYuikQikUgkkjaGVI5IJJIWJzH9S1aIVK6VlIYjEolEIjGtPOL9Q6ccFwBeBN6S0J7lqSb6HGk5Imlj7IHiLjF+irnVksqRtkT1OHLJW6JZbtQkLccH6z5g4n8mcvG/L27tokgkEolEImljSOWIRCJpcUznAXVO3KWQSiKRSA52QqFkK4/cNAcAPkVQstPfvDwj0q2WpG2TiEUhmhE0pNJypPJ/SdtAqac+YpY++7Awkhr8O//fAGws39i6BZFIJBKJRNLmkMoRiUTSogjRGL/KUkglkUgkBzvVFRntUmMxR7yqoGibt5l5xtxxWdXqq/NlvyPZ/zFHWDIge9ujSn3UHltPfoMkEolEIpFI2iJSOSKRSFqUJOVInRN3OUGUSCSSg52gqRyJbbdLiylHfKqgaHvLKEfkGntJm6OFYo7IgOxti6qxZATJ1awgEIoc+7Yq8vFLJBKJRCKpAzmqlkgkLYpRJYZubS4jFOpaUSeRSCSSg4lE8HRrXMibkxpzqxVSYOfWZipHosl5GtKtlqSNkRCaNycGRWXMEWTMkTZG1eqo/rWJudWS36DWpDF27RKJRCKRSA5OpHJEIpG0KIZRZfIhLUckEolEUgfhuJVHwq1WmsOC3RIbmm4t8GLoRp3nNpRndcsR2e9I2gqq0gIxR2RA9jZH9YDs1RFq079nEolEIpFIJJK9j1SOSEzkan5JS5BQjtTvyEROECUSieRgp9LKI95jKIoZlL3cMCgtCDQjz7hyRIsNcYUS+yvFyJI2Qwu41YpntKclkbQkVd1qVQsJE7Nck/MsiUQikUgkkraIVI7sp0ydOpU+ffrU+W/UqFGNzqu8vJwpU6awbNmyvVbe9957jz59+rB169a9kv/YsWOT7r9v374ceeSRXHfddfz++++NLuP48eMZOnQogwYN4vTTT2f69Ol4vU1z7TF27FimTp3anNtoEZ555hn69OmzT641depUxo4dm7RPN6qGv6194t4cRVxt15JIJBLJ/ksoHLccUSv7ik7pTgDKmhmUPRxXuNi05P5HutWStBUUEpYjTSfhGkhajrQ9FJJjjiQfkzFHWhvpVksikUgkEkldWFq7AJLmk5OTw8yZM2s9ZrVaG53Pb7/9xgcffMC5557bUkVrFcaMGcMNN9wAQDQapaCggJdeeonLLruMBQsWkJWVVee5M2fO5Nlnn2XixIlcf/31WK1WVq9ezQsvvMCiRYt48803m/RMD2YMUWk5Uvs0RK6ek0gkEgmEo9VdYMEhmS4WbyiiTBUUb/c1O0+bplZTxMt+R9I2UJQ9UI4kYo4gUFSpHGlLKEkB2avb+AiEIq2mWxPpIUEikUgkEkldSOXIfozNZmPw4MGtXYw2Q2ZmZo3nMWDAAE444QQ+/fRTLr744lrPC4fDPP/881x55ZXccsst5v6jjjqK7t27M2nSJD777DNOPfXUvVn8A4bGudWSExSJRCI52ElYeSRcYAF0yYxZjpRqgt2bK5qep15FOVJlvxQjS9oMLfAyyoDsbRCl7k2pHJFIJBKJRCJpu0i3Wgc4q1ev5rDDDkty81RUVMTIkSO54oor+P7775kwYQIAEyZM4NJLLwXg0ksv5bbbbuOmm25i8ODBXHHFFSxZsoQ+ffqwZMmSpGtceuml5nkAhmEwZ84cjj32WAYNGsQNN9xAWVlZjbKtXbuWa6+9lqFDhzJ06FAmTZrEli1bzOOJ67311lscd9xxDB06lG+//bZJ9+/xeBpM4/V6CQaDGEbNScuYMWO45ZZb6NKlS1L6v/3tbxx99NEMHjyYc889l6+++irpvEgkwmOPPcaoUaMYPHgwEydOZNOmTebxqVOncuWVVzJv3jxOOOEEBg4cyPjx48nPz+fLL7/kzDPPZNCgQZx33nn89ttvSXkvW7aMSy65hEGDBjF8+HDuuOMOiouLa723rVu31ul6raqLqu3btzN58mSGDx/OoEGDuOyyy/j111+T8iorK+POO+9k+PDhHHHEETz++OM1ntmll17KPXffyRP3T+XC08dwz6S/8svyZZx/xBB+WV7ptk0IUeO9EULwyiuvcOqppzJw4EBOPPFEXnzxxTpXev36668MGzaMq6++mnA4XGsaiUQikbRdQnpyfBCALpkuIOZWa1d+eZNX+5rKEUt15YhUykvaBqaFQTOUG+ZbLN1qtTmSLUdqWQYk3Wq1KtKtlkQikUgkkro4KC1HhBD4axGEtwYuVU0aTDeVaDRa635N01AUhf79+3P11VczZ84czj77bEaOHMm9996LYRhMmzYNt9vNvffey4MPPsi9997LkUceaebxySefcNZZZzFnzpxaFQd1MX36dN58802uv/56Bg0axCeffMKTTz6ZlCY/P5/x48fTvXt3Hn30UaLRKHPmzOHCCy/kgw8+SHKBNXPmTO6++26CwSBDhgyp87pCCPN5GIZBYWEh06dPJzs7u16rj8zMTAYNGsSLL75IQUEBJ554IkOHDiUzMxOr1cp1111nptV1nYkTJ7Jx40Zuuukmunfvzvvvv8+kSZOYO3cuw4YNA2DBggWMHj2aadOmUVhYyCOPPMItt9zCe++9Z+b1008/UVBQwNSpUwmFQtx///1cc801KIrCTTfdhNPp5L777uO2227j3//+NwBLly7liiuuYMSIETz99NOUlZUxffp0JkyYwDvvvIPD4Ui6t3bt2jFv3rykfT/88ANPPvkk48aNA6C4uJjx48fjdDq55557cDqdzJ07l4svvph33nmHHj16YBgGV111Fdu2beOOO+4gPT2dF154gVWrVtGuXbuk/D/73384+oSTuevvT2A49HpcayXz2GOPMXfuXK644gpGjRrFqlWreOKJJ4hGo1x77bVJadevX8+VV17JoEGDmDVrFjabrc76lUgkEknbJGK6wKocB1VVjgTLI5QXBvDkuBqdZzga619sFpVkvYoUjEnaBolhf/NijsTzQCpH2hpKtYDsSccQCNrG3FMikUgkEolEksxBpxwRQnDWj+tYWt50P9Z7g+EeNx8M6dksBcm2bds47LDDaj02ZcoUrrzySgAmTZrEF198wQMPPMA111zDZ599xvTp08nNzQWgZ8+e5t/Eb4jFLXnggQdMwXN1i5HaKC8v58033+SKK67gxhtvBODoo4+moKCARYsWmelmzpyJ0+nklVdeISUlBYCRI0dywgkn8MILL3DHHXeYaS+66CJOOeWUBq89f/585s+fn7RPURQef/xxMjMz6z13xowZTJkyxcxDURR69erFiSeeyGWXXWZaoHz99desXLmSWbNmccIJJwAwYsQItmzZwvfff28qR3Jzc5k9e7YZp2TTpk3MmTMHr9dr3q/P5+Ppp5+mR48eQExp8dZbb/HKK68wcuRI87xHH32U8vJy0tLSePLJJ8nLy+O5555D0zQAM3j8u+++W8N1WHXXa5s3b+bFF1/kpJNO4vrrrwdg7ty5lJaW8uabb9KpUycAjjnmGE477TSmT5/OjBkz+Prrr/n55595/vnnOeaYY4BYfdUWIN1qtXDNzVNJcdgwUgQ/LV+VXCdAdZFAeXk5r776Kpdccgm33347EHNrtnv3bpYuXZqkHNmyZQuXX345ffv2Zfbs2VIxIpFIJPspVeODJMjLcgNQoQrCxKxHmqIciRh1WY5IJG2DPVkUZeYhQL7VbYvqliM13WpJBa1EIpFIJBJJW+SgU47AgbPQKicnhzlz5tR6rEOHDuZvq9XKo48+ynnnncddd93FOeec0yhlQ/fu3ZsseF65ciXRaJTjjjsuaf+pp56apBz5/vvvGT58OA6Hw7T2SElJYdiwYXz33XdJ5/br169R1z7uuOOYNGkSEFOCFRcX88knn3DbbbcRCAQ4//zz6zy3ffv2vPrqq6xbt46vv/6aJUuWsHTpUmbNmsXbb7/N//3f/9GtWzeWL1+O1WpNUgqoqspbb72VlN/AgQOTArh37twZiCkBEsoRj8djKkYAsrOzgZiyI0F6erp5ntVqZeXKlVx55ZVJVjJdunShR48efPvtt3XGVYGYO7Drr7+enJwcpk2bZk7iFi9eTL9+/cjNzTXzVFWVY445hg8//BCIufKyWq0cffTRZn4ul4sxY8awdOnSpOt07ZqH1WZDVUCvY+Je3U3KihUriEajnHTSSUn777777qRtn8/H5Zdfzu7du3n99dex2+113q9EIpFI2jYRPdYXVHWrleG2kZ1ip9AbokgT7Mwvp/fw9o3PMx73ym7RktyoSLdakrbDngRkr3LWATKfOVBoUOklY460KjIgu0QikUgkkro46JQjiqLwwZCeB4RbLZvNxoABAxqVtl+/fvTp04fVq1fXUFzUhdvtbnKZSktLgUqhfoKcnJwa6RYsWMCCBQtq5FHdysPlatyK0fT09BrP49hjj6WgoIDHH3+cc88917S2qIuE9czEiROJRCK89957PPjggzz11FPMmDGD0tJS0tPTUdX6w/VUL3MifVX3ZAklSUPnJigvL8cwDJ5//nmef/75GsfrUxQYhsHkyZMpKCjgnXfeSarb0tJSNm3aVKcVUiAQoKysjPT09BrvavV6BXA4Y+XXFAg3MiB74r1pyMKntLSU7t27U15ezuOPP84zzzxTb3qJRCKRtF2qxgepSq92KXHliMGu/PIm5RlJ5GlVq/U0UjAmaRsoCeXIHsQcURAoDYxFJfuaZMuR5CMyILtEIpFIJBJJW+WgU45ATEHibkBIfqAxb948Vq9eTd++fXnooYcYOXIkaWlpTcojIRivHn/E5/OZwvaMjAwgFvS9qlVEQvidIDU1laOOOoorrriixnUslpZ9Lfv37893331HSUmJaZ1Rlblz5zJnzhy+/PJLnE6nud9qtXLBBRewcOFC1q1bZ5a7tLQUIUSSouDXX39FCFGngqElcLvdKIrC5Zdfzumnn17jeNWyV+fxxx9n0aJFPPfcc3Tt2jXpWGpqKsOHD2fKlCm1nmuz2cjIyKCkpARd15MUTNXrFcAQiZXAEKByqmjoepVUIum9SbyLxcXFdO/e3Uy1fft2Nm/ezOGHHw5gxjr58MMPuf/++/nss89M92YSiUQi2b+oSznSOzeFxRuKKFQFhVsqiIZ1LLbGjduSLEfkIntJG2RPLNhlQPa2S9XqEAhTCVa5TypHJBKJRCKRSNoicsnRQcC2bdt49NFHGTduHM8++ywVFRU89NBD5vGGrCkSJCwddu7cae4rKytj/fr15vaQIUNwOBx8+umnSed++eWXSdvDhw9n3bp19OvXjwEDBjBgwAD69+/PK6+8wv/+978m32N9rFq1Co/HYypuqtOzZ09KSkp47bXXahzTdZ0tW7bQu3dvAIYNG0YkEuHrr7820wghuPPOO3nuuedatNzVSUlJ4dBDD2XDhg3mMxswYAC9evXimWeeqTMmzPvvv89LL73ELbfcYsYLqcrw4cPJz88nLy8vKd8PPviAd955B03TGDlyJNFolM8++8w8LxwO8+2339bILzFxt2ggUHDG35uigl1miurvTcINWfX35KWXXmLy5MnmO+p2u3G73VxwwQUMHjyYBx54gIqKisY+QolEIpG0IaJGZfD0qvTKTQWg1A6GLti5oazReZrKEZu0HJG0TRR1D9xqJfIQSOVIG0NN1o4kIVASgWIkrYSQfYBEIpFIJJI6OCgtRw4UwuEwK1asqPN4nz59cDgc3HXXXTidTqZMmYLH4+Hmm2/m4Ycf5uSTT2bs2LGkpsaEEF999RUej4e+ffvWmV+HDh2YNWsWKSkpKIrCc889l2Sx4Ha7ueqqq5g9ezYul4sRI0awcOHCGkLvG264gfHjx3Pttddy4YUXYrfbmTdvHp999hkzZsxo1vMoLi5Oeh6BQID58+ezePHiJAF7dUaNGsUZZ5zBU089xZo1azj55JPJzMxk586dvPXWW+zcuZOnn34aiLnpGjJkCFOnTuXmm2+mS5cufPDBB6xfv56//e1vzSp3U5g8eTLXXHMNt956K2eddRa6rvPSSy+xcuVKbrjhhhrpV6xYwT333MPw4cMZPXo0K1euTPK5e+ihh3L55ZfzwQcfcPnllzNx4kQyMjJYsGABb7/9NnfeeScQC74+evRo7r77boqKiujUqROvvvoqxcXFZGVlJV0zkbstrhzp2rMXWbntefeF53G5U0hVvfzrlbeS3pvMzEwmTJjAK6+8gs1mY/jw4axcuZI333yTKVOm1HBjpqoqDzzwAOeeey6PP/44Dz74YAs9YYlEIpHsKxKWIw5rcv/cq11MqV5sjfUoW9eU0Llv/W4XE0TjfZzdqiWJwqqrSiSSVmMPlBqJIZxSI+S3pLVpyE2ytBxpXWTMEUlbwzAigIKqSpGcRCKRtDbyS7wfs3v3bi644II6j8+fP58ff/yRxYsX8/TTT+PxeAC49NJL+eijj7j33nsZOnQovXr14owzzuD1119n0aJFfPzxx7Xmp2kaM2bM4OGHH2by5MlkZ2dz2WWXsWHDBvLz8810EydOxOPx8OqrrzJ37lyGDBnCHXfcwf3332+m6du3L6+//jr/+Mc/mDJlCkIIevfuzaxZszj++OOb9TwWLlzIwoULzW2Xy0VeXh733XcfF110Ub3nPv744wwfPpwPP/yQu+++G7/fT2ZmJqNGjeKRRx6hS5cu5jN4/vnneeKJJ5g+fTqBQIA+ffrw0ksvMXDgwGaVuymMHj2aF198kZkzZ3LTTTdhtVo57LDDePnllxk8eHCN9IsWLSISifDDDz9wzjnn1Dj++eef07lzZ9566y2efPJJ7r//fkKhEN26deOhhx5i3LhxZtqZM2fyxBNPMGPGDEKhEKeddhrnn38+n3/+uZmm6sTDalFAB1XTuPXRJ5j71ONMv/tOMjLTufSSceRv2snWTVvN9LfffjtZWVm89dZbvPDCC3Tu3Jl77rmH8ePH1/os+vbty4QJE3j55Zc588wzOeKII5rzSCUSiUTSSoSjMWGh3Z6sHOkdtxwpiuqEEWxbU9LoPBOWIw6bVm2lsKjhElMiaQ325B2stBwRUjfS1lCSY45Ur2YZc0QiOfgQQlBe8TN2Ww5lZT+RljYQp7MLu3Z9zG+/34XdnsOggc/jcuW1dlGbhBACIcJEohV4vb/jcnYlEi1j166PsFjS0DQXGRkjCfg34nb3xu3u3nCmkmYRq4soqmpt7aLsVXQ9CBiUli7D799Ahw7jsFhqj+ErkTQHReynyyh0XWfFihUMHjy4TouAYDBougtyOBz7uIQHJ0II/H4/LpdLCiAOQnTd4JcdseC5vXIU/ghpiGre+zIoQtHLKNcVemX0wqbZWqOorcaethH5XZMc6DSmf5ccGJw59b+sIsLNI7tz89n9ko4N+/v/KPSGuaTCTkehcdVTR2NzNLymZ/TUT9mKzuOnHMppow+hx6JVALwkLuSUY1cdMCs0ZTvZf/nli4Ucr3hwRsLknzS8Sef+WO7jtOV/kFu0m6VH9MZWLY6cpJJ93UbK1q+nz+aYq9c/jh7Ag+u389r2IgAOFat4WH+JESd+WV8Wkr3IJQsuYeXulQCsumxVK5embSD7kZYnFNrF7t3/Q9d9eH1/4POupcL7i3ncYklj6NA3+fHHi4hGy+L70snJOZHi4m8IhXbhSRvEoYc+icvV8PddCANFUeO/dcLhYgQ6dlsuiqJQWrqMSLSMrMyjKS//Gb9/I4qioRsBhBFBYKAoGgoaqmpFUTQi0fK6LobX+xvFJd/FriPCjXwqCh07XoDb3bNKVlHKy1eh635SUw8lFCrA6ehMbu4ZuFzdEMJA131oWkqrypT2tI0IYVBSugSv9/dajyuKhqY60XVfLH2VRT1CRCku/pbS0h8wjORnrWlu+h/2NFlZx7J02Z8wjBBHDPsATbM3uYxtkUiklGXLx+FwdKZf34exWDz88MMZBIKbzTRpaUMYdvjb5vsvaT3acl/SlLIdGDNEiUTSJtD1yg7dosZ9LNdCYm/UiB50yhGJRCKRxIgYAlRwOmoOVvu0T6VwXREV6RqiSLD9j1K6DchuMM9ofEV99ZgjCrGJphz6SlqbhJynWavTEidpQrppamNUDcBevW4FiqyvNoS0IpTsDbzeNSxbfj667q0zTTRazurVfzEVI6mph1FR8Qs7dvzLTFNW/hOLvx+L3d4eIQxilq9V/xrmtq77sVozUVUb0WiFeW1FsaAoVgwjsPduOI7Nlk0kUooQUZzOboRCOzCMUJUUgu3b36rz/KKiSqXxhvx/oCgWNM1FNFqO03kI7dqdTlHhF0Si5bHvrKIS++IqcaGCGm/PsX+VwnIlKb2qWDBEGF0PYLGkEo16q5WzdsLhMIu/b568Qtf9Zl23JLrupahoIenpR1BRsRqAwqIvyG13aotfK0FiTf2++Hbu3v0Zfn8+fn8+3353NFZrJpFIcVKa8vKfKChYQG7uGXu9PJKDAzlDlEgkLYYRd2dSX5epIGOISiQSycGOEIKIboAKDnvN4ejAzul8u66I4jQNiqJs/qW4kcoRQInlWT0guxARQFrcSVoXZQ/8YSXeacMjyC96jr5dp7VMoSR7TPWxbfValm612g660LEoUgwiaVnWrL0fXffidBxCmmcQTuchWCyppHsOJyWlH9u2v8Uff/wdv38DAF06X06PHlPYunUuXu/vZGYdg8Penh9/uhiAUGhno64biRRV2Yp9eYSIxheEVGK1ZmK1egiFCkhPH46mOVFQERgIEcXQg0R1Hw57hzon63ZbO7KyxuBy5WGxpKKqDlTViq4HMYwAVmsGhhFCUWwYRhBVdVBS8h27dn2MXk1R43R0RjeCRMLF2B0dKS1dQlnZjwgRJRq3XgkENrNp05xGPYe9SahhHUqdaFoKmZmjUNWaCpZotJxIpBSHo3Py2CD+/G3WLDp2PB+rtTLu3patc9m0aQ66EUxS7hQVflmncqSoaBFlZT/i8QwmM3M0iqIRCu0mGNyCqjpJSemDoqjoehC/Px9VtRPVKxBGhEBgC4YRZPOWFwkEtpHb7jSyssZgt7ev1xrb7uiIw96+qY8rVt7ir5O2qypG8vJuRhghNm6aw7bt86RyRNJiyFGBRCJpMQxRqRxpWE0ikUgkkoMVPWKQmLa7nDWHo0O6pAOw1Yilyv95N0df0KvBFWtRqsYcqURBxIOfSiStixIfJNVlXVsfZswRBNtK/0VfpHKkrVD121Tda7W0HGl9qrqr0YWORYpBJHUQDhfy88/XkZ19At26XVdnuoqKX1j9y2R697oLj2cIpaU/ADB06Os4HB1rpHe7eiRtp6b2R9PsdO16TdL+bl2vZ3fh5+Tl3YTL2RUUNS44T1hFVFpKqKqdSKQ45l5LteJ29UCIKMHgNgQCh709imIhHC7G4eiAoqh7xXJK0xxoWmzxiara4/ucAGRmjiIzc1SDeRhGlE2bniUQ3ErHDuMQIsraPx7C4ehIdtZxpKYeCsTbcsKSBgGi8m+sl6y0tKmaXtcDGCKM3d4ePVqBqjqwWFIbKJPBmrVr6dO7N6raPPdNMUVSy8XGsMUVJUY15Uh5xc810obDRSz/8UL8/vVJ5XE6ulBcsji+aAiczq643b0oLV1CNFpR7/V37prPzl3zGyynomhkZByFRau8d0W1ckiXK0hLqz9Wb1nZcgCGDPk/rJY0hDCwWNJMV3OBwBY2bppDScn3hEIF2O3tGiyPRNIQclQgkUhajITlCCg1JoaVNEccIJFIJJIDiXBQR493Bs5aYokMPiQdgI3lAXSrC29xiKJtXrI71z2RFYZAj/922rSkfqjSrZZE0rooihJTjjRjMFRVOSJpWyjVArJXRaCAtBxpXapUim7o0LbcokvaEJu3vExZ+U+Ulf9Ex47jsNlqt1pd/cvN+P0bWLHyCo4Y9j4ANlu7WhUjAC5XclDylJQ+tabr0eM2evS4rdHldTg6VNtjTYrvAeB0djJ/t1WXcqpqIS/vxqR9Rw7/qJVKE0PXdSxaNK7IahsfDTWuhDL0ZOVIILAJIXQUpbKcBbv/k6QYsVhSTXdVAHZ7e6LRCgKBTQQCm4CYpQuA1ZKGolrQNDcIg5TUQ2nX7lS2b59HKLiTqO6lLgehhhEmFNpJcfGiWo6FGDhgNsXF31FY9AVuVw86dPizqVQTQhAOx6yhXM6utbYnp7MLHs9Qysp+ZPv2eeTl/aXRz08iqQupHJFIJC1GTcuR2olN6tvmwEwikUgke59wMEo43lOk1KIcaZfqoFO6k22lAaJ5brS1XvJXFtarHNGjBnrCcsRuqbZOW0jliKRNsCeCqboXnkham/qUIwAG0nKtraALveFEkoOW3bv/a/7+edX19O59H2mp/WukC4UKzN8JYbPLlVdnvtWFvG53jzpSSiR1o6kx5UjMrVZloHbDCBMMbsPpPMTc5/WuMX+POPI/2GztKC5eRCRahidtEKmph6HrfgoLvyASKcVub0d29vFJCpbq5GQf36hyer1r4m7SYt/b8vKf2bHzXfSoF8OIsGr1jWY8lmBwO506X0xpyQ9kZx9nWrRYLJ468+/c6VLKyn5ky9a55OaeZVqVSCTNRSpHJBJJi2HEJVGNnfaLRq58NIwwuh5E01z1+rbcm0SjPoSIxs1vFXTdj6paa/UfKpFIJJL6iQR1IvHOwl1LzBGIWY9sKw1QlmmhHbBhxW6OOL1uwYMeNZKsUWq61QrXep5Esm9Rqv1tPNJypO1Sn9JLoBCmFMMImatjJfuWJLdahlSOSGonHC40Y4IAlJX9yLJlf6ZP7wfo1OnCpLSqakOPv0qJGAkuV7c681YUle55t7Ah/x94PIfLb4GkWZiWI0aoRkB5n39DknLE51sLwKGHPmlaE+Xmnp50jqa59krcjpSUPknWUQUFn8aUI0aQ0tIfkgLVl5T+wLbtbxKJlHDIIVcDMbdcmuaqM/927U5h8+YXqPD+wuLvx+Jy9SQ7a0ysbWl2VNVOirsPNltmnXnsbbzeNWzZ8gpdu16LP7ARgMyMo6T8qI0ilSMSiaTFMC1HFOr1rJyYPgqhx4N81R3lTCAw9CCJjFXVjgIx3/GKitrsgIoK9ZeyShmEYQrVYr5eFXMVhKo6mrQKVADCMPD51QbFIgl/qVVXb4RCOqFQAT+vehzD2I6q2FA1Bw57BwwRIRopI6r7qN92BzTNTV63Sdhs2UQiZfWmPVgIRwopKPiUiopVqKodmy0bmzWLcKQIw4iQmnoY2VnHkViBbhgRDBEmFNxOJFIaD27orswvXEB5xSpS3L1xOLqgqBop7t6Ew0Xoug+Xqzs+3zrs9vbY7TlxhZsj/o5JJAc2kVCUsBL7Trlsta9QG9IlnX//vION0QjtVYXCLV6Kt/vI7OiuNb0erXSrZbdpVF1krwDrNzyF292L1JS+WCweFEUlJaUfFkvt+UkkewVVAb2hXrp2qitHdD2EpkkBW5ugQbdaAn9gMynuXvu2XBIADFE55peWI5K6qKj4DQCnsxu9e91N/saZlJev4Pc1d1Natpz2uWditWZiGKGkINE7d84H6rccAcjLu5H27c/Bak3fW7cgOcAxLUf0AHo1Gcq2bW/Ej6sYImzGwUlx997XxaxBQhloGEFTmWi3tycU2klZ2TIz3Y4d7wLEx+l1S0tU1cagQS+wbPl5BINb8fvXsdm/Dra8aKbRNBeZmaMRRhTdCOJ29yAa9ZKZMQqnswsoSjyej2rKeBJ/zVg/ioqCiqJoWK3pDZYrga4HWLnyKoKh7Wzf8ba532JJJzX1UNrnno1uBHC7uqMo1iY8ybpxu3u2qjJof0cqRyQSSYuRiDmiQpWZYU0XWgox2YAR2o4hGutmQAFRRVECIHT0femmQFHiCotKDCNYR+L6ae68zDAEQkTw+f7AMLab+5uj3vhpxZLmFeIgwef7I2m7rGwZW7fOrTP9ps3/bNZ1FMVKSkofKip+wWrNxOMZAsTiI3i9v2OzZqFZ3KiKjUikBEXRsDUh8JwwwnG/sCqaakeNB0g0G6moErCw6n4gHNqNqtqw2jIrAx4m/pnbiTwqj1e2E1HHeQYIUet9CCHw+spYtTo++BQG4UhJfLBX+2BUiAh+/0aczkNabCWegoLd0R6HvSNO5yEEgltRFA2XsxtC6LhceQiMmKJU6Aihx5RmQkePeikpWYzN3o6szGPiK4T2titBg4qKX/D788nOHoumueIueKrUjbkNNYJX1nI8FNxBYdGXMeWwMLDaMrBoKYTDRbFvtxBomhO7owMVFaux23JxurqiKjbSM4bXKwQMB3TC8UeSUoflyIjuWQAs21zCuYd1YMuqItYs2cnIc2p3RREJR80g73arWk1AKSgo+Het52VljcHt6omiaPgDm+nY8Tw8aUOIRr1AzBWGEv/+B4PbsNli720kUoTN1i6++k0hGq1AUTQcjk6NmjgJIQgENmKxpGKxpBEK7SL2/BMTtOQArJWTOAXDEBjCRzRagRCWpHSKArruJxIpr3FNh6MTqmqJuQNRVOy1+FKPKWqdbdYv+f6Ouidutaptb906F82SQnbWsdjtuUQiZRhGEJutHYqiSmX7PkSp8o0XAnSj5vjU71uPy9lVrhptBXShc4hN5yh3lFC4GJxZrV0kSRsiENhMWdkKgsGtAKSmHkp29nFkZR3Lxo0z2ZD/NDt3vs/One/Xm09mxlENXqtq/A+JpKlUVTJUXWCqKBYKCz+nsPDzpPRWa2abcOFWWe4w0fj4NDNjFDt2vpuUTtd9AFgsaQ3mabe34/Chb7F+w5MUF3+L09mZ2Bg5RDhcSCi0M8lNXknJdwANtuP6cNg7ollS0DRXLFg8Ij7eio3FY/dqIxjcRjC0PelciyWNaLSUkpLvzLK0JDZbNqNHfVevWzRJ3UjliEQiaTES8diVmAOT+G9RIwS7pkA7iwHCQFEscR+s9a0MsKCqTgwjEBdACnPCX11Z0RiEiCBEFFV1Jq30q09coSg2VNWKrgeAmEAwZs7adB/2oVAIu71hAW5CMFb1HhUljNWq07v3fZSXfY7N3o4Ud2+CoR2oqh2rxYPFklLv3Qihs3XrqxSXfAuA03EIitoyKxb2ZxKryCPhYlzuHrhc3fD51hEMbiMrawwlJUvw+9ejKFZU1Rr7q1iw2jIx9CD+eCC7BJrmxG5vTzCwBUNEiURKklaZJRAiQkXFaiAmbC0s/CzpeCi0s2ZhK1ruvluVeu6jqKjp2QWq1UFbYB2P7PNrbsh/ep9fszrp6Udy+NA36jweDEZMt1ouW+3D0UM7pJHptlHsC6P3SIFVRaz9YScjzu6Ootb8xoXDhvnps2uaGdMEKlfaW62Z2O3tMYwQweAWDCNMUdFCiooWmml37/40Kd/UlMM4pOvV5OdPN/2KK4qlzhgmLldPHPZcLFYPVmvNFVx61Etp2XIMI0A4XFjHE2oc33zbtPR2e3s0zW0G6ExLHYjD2RmrNQNQ8Hp/o6xsOU7nIaS4+2C1ZcUnfI1B4PP9gRBG3JVCyypXdN0by9vdi2Bol7nfFEorlXsSfy2am6hegcPRiRR3HwwjiNvdC1W14/Otx+dbayo/VcWG1erBbm+P03kI5eU/U17xM6Bgs2aSm3sGFksKfv8mystX4nLlEQxux2JJITV1AJrmjI0RFAW/bz0ORyes1nT8gY3Ybe0oLVses/IkCDibF5DdHGfFfqxb/ygAa+pI36P7rSiKhsWShtN5CCWlS8jIGIkwIjGFLwrt2p0Sr//mERuT7d+KNCEEQujNct2acJVV9ZMkgLKKX4Au8e3YwVWrJ+F296ZD+3MIBLfVmp+mObDb2xMIbEEInXC4gLxuN9YIsJzcvtS4MMRIEooYRhRF0cz6MYxIfLt2pdmBUJd1ETWiTM6NCRK3b5pOx4EzMYwQimKRgqSDmHC4iE2b/8nmzS8k7U9NOQyIzcXy8v5Cevpwdux8j7Kyn9B1P4YRxGrNonPni1m79gEAsrKOJTX1sH1+D5KDi8qA7JVutdLSBtM97yY2bnrWDGYOYLNm0rPXnW3ChZumJSxegqacyOHsUiNd4p6sjVCOADgcHTjs0Cdq7BfCoKhoIV7vGqK6l02b5gDgcnUntghJj8lYzEV8hum1o7a/QkTQdV9M4VG305Ma5OX9FVW1k5V5DG53L7bveJs1a+4BwO3ubXojaQkyMkbK/mwPkMoRiUTSYlR1q1XV9YOoYj2iADZTbqHhcnVvtEuI+vxO7iuqul/RNBdaE/sfIQThsIbF4mrWBDQWe8VBVuZoOnU8ocnnJ8jOPp7i4q+xWjNISxvY7HwOJrp0nrBH5xtGmM1bXsbp7ILHM5QNG56mXc7JRCIlBIPbyc4+Dp9vPbruN8/RNCeGiKAqVgwRiQkphajXFV11TJ+tioJhhGPWV0qiPVb3fZ8sZNRUh+k+LLFiPWZqTLVtpcp2pfs5Eivfa5ynIkSUSKS05nMSBlu2bKFLly6ocQGOxZJiruKvC4s1jWi0olJ6uIeEQrsoKf0eqzWTiorV5mpCVXVitXrQdV9caBxz76coGlQROoVCO1EUK4qi7bNA4FZrOroeqOIGsEodVbFEiKEkHVfM+qLKtkpqSj/S0gajqjHLpUi0HJs1E1W1I9ApL1tBVPeSkXEUweA2opFSBAZdOl9eb1m9vspV1XVZjqiqwuie2Xy4cjvrRIR0pwVvSYhtf5TSuU9NYW4wVDnBsFlUwlVcJ3bdcgpdLrgbu729+e3dsXM+69Y9SjgcC6rqcHQimCSwVAGDCu8v/PLLzUnXqq1OE0p8v38dfv+6eu+/NmJKeEuVyVhsslZ1gtYUYn1mZT+j674qytbYvZVXJBQAyQQCmwkENjf5HhJUdZHQ0uxqOMleYe0fD7ZIPsVkgvJ8jYUjjaG6hZ/NlkM4vLtaKsU8vn7DkzXy2LhxVvL2pjn07nU3wbiLyEBwCw57RwLBLaiqg7KyZbjdfUhN6WvalSEMRFwZVlz8DTnZJ+JydWtTk3LDMAgGd7Bp02JUtW4LGsMIs3PXBxhGhI4dziMSLcFua5x1ZjC0kx073iPF3YsM+xHAaeaxqF7ZT1eta59vranUaixVV742hKLY4kIolWi0FFDRNAeKYotvA6ioqiXeR1mw29sRjXqJRitwu3ugaS503Y/P9wfpnmG4XN3jK17L0XU/Dmdnc/wQDhcRDheiak5U1U4wuA0hong8Q1EVC8HgdsrKV5Ca0q9Wl0MCQcC/sZbFUlWjhIgqfbtIbgdJ+6unr7SEPMpW+W0vK/yENWvuZ+eu+QghyMv7C8KIkuxuV6n9d62Lqmo/npyi4fyalraBvOuYY1RNawhBKLyF7TvWoKpaLWnqKkfDZQ2Hd7eo0K+lEAgCgc1Eo+WEQgWUl6+okUbT3OTmnpm0LyPjSDIyjqw1T0WxEImU7PE8QSJpDJoa8wCgG0HTo4aq2snKGkNW1pjWLFq9qGoiVkoQ4gs/tXqUNhZr3cHYG4OiqGRnH0d29nEAZGSMIBTaSYf25zZ7EUAkUkZx8TdoFjfRSFlsbo6GQI97Eoh5R9i27Q283t/o1m0S3fNuSsqjc6eLsNvaEY2W0779OQfsgoT9EakckZgcyKuFJC1DQ++ISChHqNerVmV6zdPmfWUfqO1CUZQ2PYA6EFFVG926XmtuH9pvWo00csUZ6LpOwa4VdOwwGK2p2scW569A7DtgGCHKyn7E7e6JvQluzSS1EzQq+wuHtW7B5dG9YsqRbzYUcfOwdvy6aDtrvt9Rq3IkEKpUWNgsKiJSKeiyhtNwODokpe/Q/k90aP+nuCu3NbhdeZSVrWDnrg/p1PEC3O6ehEK7+e33OyktjbkhHHb4u3i9v+F0HkJa2kBKSpaQnn44mpaCqloJhQrYVfDveFyicFzgVg1Fwe3uhUVLwe3uSUXFaoQwyMk5sUE3SAlXaVE9wsoVKxg0aGBc8GtUUaAIFMVqrtJLEI1WsHPXRwih0z73LKLRcsorVhEK7TJdHFisaWRmjibgzycUKiAcKW6SwnFPrDobIhjcRsHu/+BydSMrawwKGklu4uK/qroKDAa2YLVlUV7+M6HQDlOIZRgRrFYPFosHXfdit+eiKBo+33ozSKjL1YPU1MPQo168vrWmghTAas1C1ytwOrui64G4Ui0x4Kn+vGL7NC0FXfeyJ8HUKxeeQJ/O99K592UIIdi9+7/s3DWfvLy/4nLmsXnz82zI/4d5nt3egUikCMMIo6pOHI6Yu8Cioi8JBrfx86rr672u359fw6KqKjt3zW/2Pe1t8jc2Je2MZl3DVDIqMeWIqFbHShh6229hnTLLVF53PeTaWlfzhiPFeL2/AYKysh+bXBYhwkSj4Sp7jPiiC3/Svlg5YumqBsYtL1+ZlF9R8demf/imUFy8KGk7dk+tx9Bqj3rrttfM3+vW7XsLz7bE2rWtXYLWx2rNJCf7BHQjQIf2f26S66vOnS7aiyWTSJJJKBnC4QI2bpod39f2XTVWVY6YY0RFpesh17Jp83M10lssqS16/azM0Xuch9XqqRHQvjY6dRxPKLSrxrwjQU5O8xe4SvYeUjmynzJ16lTef79uX3nZ2dl8+23jfC2Ul5fz97//nfPOO48jjjiipYqYxHvvvcedd97J559/TufOnVs8/7Fjx7JtW+WKIEVR8Hg8DBkyhJtvvpm+ffs2qoxvv/02a9euRdd1OnfuzEknncSVV15JSkpKk8oyfPhwpk2rKfjcFzzzzDPMnDmTNWvqcrLQPGbPno3NZuOqq66qM03CrZaqKEmuH5JFBbFfIQEW1Ulb5vPPP+c///kPjz32WGsXRSKRtCKKoqBpDjIzG/YlLWkc7fqmw/9iwdjrU0Af0zsHgJ+3ltLumN78umg7637czejzo9idycPYhHJEBTS1UgStCKNepYOiKKSmxMYJ1VeHulxdGTrkNTZs+Ac2ezs8nsF4PIPN49UnOHZ7Ow7pckUDd59MXZOnusoKStxayYKq2hqtRLRYUpOEOFarJxaQshbaatDofmLaXo+jkYgfVF1wresBhIia1nhVF08IoccVD454GktcABCNK0V8aJqbQGATC74fF8twD9ZdKAiUuAsoRVFo1+5k2rU72Tyel3cjAsHOnfMZMGBWzOqjlnbg929i/YYnKSj4NzZbNtnZJ2C1eIhESrDFXarpRgBd9yOEjhKPfQOxQKWJZ2GIcIta7rUEhjAoLiomMyvTtEKsE0WloGAB0WhMSdihw3nVVsTXjaa5CAQ2U1j0RfL1q52foQ5g+LB/88uvkznkkCtpX21lem34fDH3d8HQDtJSB9T57vt866jw/kZ21rFATBEaCG4lNaWf+S7qegCbLQdFUTCMSFyBG0GICF7fWsLhQtzuXoRDBfgDm1BVO6kph1JS8h0CgR71QtyvuqEHzGurmgOnowuGEUI3AtismYQjxYRDcYsmRcHp6EIkWkY0UlZL6WNWj4pqIxwqMM+J/6hh1Vhpia5Uc6VXdX+1beCdte9SHvHi0QRjOo4g3d2FYHAbXu9vpKT0w25rZ1o+JSu3qvwWte+vOz217m9M/rXFgUui1rTV09eftxCCsrIyPJ5EgOGGytL4/K3WDDNwdFvDZs8xLWCzso5tUh8skbQ2VRe+eL2/A7QJt1kNkSijrocq3a8rGj17TsHnX1/DrbTVsmeWI62Joqjyu7IfIpUj+zE5OTnMnDmz1mNWa+PjB/z222988MEHnHvuuS1VtFZhzJgx3HDDDQBEo1EKCgp46aWXuOyyy1iwYAFZWXUH3ps5cybPPvssEydO5Prrr8dqtbJ69WpeeOEFFi1axJtvvtmkZ3ogMn36dG688cZ604g63GolmY8YKjt0hahQ6NDG+/FXXnmltYsgkUgkBySJdc0pjvqHorlpDvp3SmP1tnJ+8vnI7OimeLuP37/bwaDjkwX7wXDMhUdCVZCklFebb4WkKBo9etzW7PMlLce+CDAei8lQ833RNGe1dFXc2iiaeVxRrDXSx2KBxVy3JcZDzXGrpZsWVwK1gVhh3fP+Qve8v1QpY81n53J1ZUD/Gfj9t2C3d6hhbbQ/o+s6K1asoG+fxlkhdut6Ldu3v03nLpdjt2U36VrB4A4Kv/sCRRgIRUWI6vWrgKridndn+BHzG51vIohuQ8F0PZ4heDxDkvbF4v409jrV45lUkpV1dKPzact8tuJzCvyxnufc0ffQK6NtKoD3JYk2MqB/W7DUlUgkjaE2Rcj+oBxJeAsRImy6pk0or/v0vp9IpISysuVm+j11qyWRNBWpHNmPsdlsDB48uLWL0WbIzMys8TwGDBjACSecwKeffsrFF19c63nhcJjnn3+eK6+8kltuucXcf9RRR9G9e3cmTZrEZ599xqmnnro3i39AkGQ5Et9XfepvGAZRkVhp2XZWGEokEolk3+GLW3m46wjGXpWzBnVk9bZyPly5g3uPPYSFb6xh1VdbGXhc56TA7ImYIxZT+FzpumtfCNUlkoZQVSt2JROgWQHZzdhuGPGYRy1DbbEgDjaczkOarQRNKL+SVbJKld+wR6ZCkj1GNyrjX+htMBaGRCKRNIbqCzWg/tgdbQW1iiWZnrA+jI/NHY4OHD50HsuWn0d5+U+AHJdI9j0H5UxRCIE/HG0T//a2cHj16tUcdthhTJ061dxXVFTEyJEjueKKK/j++++ZMCEWPGzChAlceumlAFx66aXcdttt3HTTTQwePJgrrriCJUuW0KdPH5YsWZJ0jUsvvdQ8D2LC7zlz5nDssccyaNAgbrjhBsrKappRr127lmuvvZahQ4cydOhQJk2axJYtW8zjieu99dZbHHfccQwdOrTRrsISeDwNa5y9Xi/BYBDDqOkfe8yYMdxyyy106dIlKf3f/vY3jj76aAYPHsy5557LV199lXReJBLhscceY9SoUQwePJiJEyeyadMm8/jUqVO58sormTdvHieccAIDBw5k/Pjx5Ofn8+WXX3LmmWcyaNAgzjvvPH77LdlH77Jly7jkkksYNGgQw4cP54477qC4uLjWe9u6dSt9+vSp9d/YsWPNdNu3b2fy5MkMHz6cQYMGcdlll/Hrr7+ax/v0ia0+mzlzpvkbatbhff/vVnZt3xa3HBH8snwZ5wwfzqoffuCB667m4tEjuOhPF/O/D/9HcWExUydPZciQIYwZMybJSiNR99988w0XX3wxAwcO5KSTTuKNN95Iuj/DMPjnP//JiSeeSP/+/Tn55JN57bXXktLU9i4nns2UKVMYPXo0hx12GCNHjmTKlCmUlJSY5/3www/88MMP5nv/3nvv0adPH7Zu3Zp0jbFjxya1sT59+jBz5kz+/Oc/M3DgwDotvCQSieRgxR+38nDXEYy9KmcM7AjA0o3FpPXxYHNaKNsdYOOqwqR0oXieFtPVUWy/soeWIxJJS+JQcpp9rh5/qdUWVo5I9gxNcydtV4YCT6CAKpUjrUlVhUhVRYlEIpHsT6i1uKvbHyxHkpUjsRhYCpVjc0VRGHb4PIYd/i+GDH6N9rl/2tdFlBzkHHSjaiEE455dzPJNJa1dFACGdc3gX9eNbHbA52i0lkCfgKbFfHj379+fq6++mjlz5nD22WczcuRI7r33XgzDYNq0abjdbu69914efPBB7r33Xo48stLP9ieffMJZZ53FnDlzalUc1MX06dN58803uf766xk0aBCffPIJTz75ZFKa/Px8xo8fT/fu3Xn00UeJRqPMmTOHCy+8kA8++CDJBdbMmTO5++67CQaDDBkypPrlTIQQ5vMwDIPCwkKmT59OdnZ2vVYfmZmZDBo0iBdffJGCggJOPPFEhg4dSmZmJlarleuuu85Mq+s6EydOZOPGjdx00010796d999/n0mTJjF37lyGDRsGwIIFCxg9ejTTpk2jsLCQRx55hFtuuYX33nvPzOunn36ioKCAqVOnEgqFuP/++7nmmmtQFIWbbroJp9PJfffdx2233ca///1vAJYuXcoVV1zBiBEjePrppykrK2P69OlMmDCBd955B4cjubNs164d8+bNS9r3ww8/8OSTTzJuXMzndXFxMePHj8fpdHLPPffgdDqZO3cuF198Me+88w49evRg3rx5XHDBBYwbN47zzjuvzjqc/sws7v7r1Tz/yjxS0ionik/f/f/402VX8OeJV/HvV57juSeeo0OnDpx48olMnDCRN954g0ceeYShQ4cycOBA87xbbrmFP/3pT1x33XV8/vnnPPDAAwBcdFHMZ/r999/Pe++9x7XXXsuQIUNYunQpDz/8MOXl5UyaNMnMp/q7HAgEmDBhAhkZGdx3332kpqby008/MXPmTBwOBw8++CD33Xcft99+OwD33XcfPXv2TIpr0xDPPvsst956K3l5eXTq1PigfhKJRHIw4I1bjrhsDSstOqY7Gd4tkx82FvPJ77sYeExHfvzPZpb+eyPdBmabY6hgpJpyxMxBoEjliKSNYFU9oINohjVTYjyuYpgxRyStj6KoaKoLxahUjBiiquWI0uy5nqRlkJYjEonkQKA2S2h1P3CJqaqxWHlCRE3lCNX6RUXR8HiGtkLpJJKDUDkCB45R87Zt2zjssMNqPTZlyhSuvPJKACZNmsQXX3zBAw88wDXXXMNnn33G9OnTyc3NBaBnz57m38RviMUteeCBB7DZbAA1LEZqo7y8nDfffJMrrrjCjE9x9NFHU1BQwKJFi8x0M2fOxOl08sorr5jBzkeOHMkJJ5zACy+8wB133GGmveiiizjllFMavPb8+fOZP39+0j5FUXj88cfJzMys99wZM2YwZcoUMw9FUejVqxcnnngil112mWmB8vXXX7Ny5UpmzZrFCSfEgrCOGDGCLVu28P3335vKkdzcXGbPnm3GKdm0aRNz5szB6/Wa9+vz+Xj66afp0SPmR/iHH37grbfe4pVXXmHkyJHmeY8++ijl5eWkpaXx5JNPkpeXx3PPPWf6hh00aBCnn3467777bg3XYdVdr23evJkXX3yRk046ieuvvx6AuXPnUlpayptvvmkK8Y855hhOO+00pk+fzowZM8w82rdvb/6urQ475g3g8ovO5l9vzuWWO24wrzv2zLM44+KYdVGOPcRNV/2Vnof25MobrqSdqx19+/blv//9Lz/++GOScuTEE0/krrvuAirfo9mzZ3PhhReyceNG3n77bSZPnsw111wDwOjRo1EUheeee46LLrqIjIwMoOa7/Ntvv9G+fXseffRR0ypoxIgRrFy5kh9++AGItYfEfTXHfd2wYcNMKxWJRCKRJOMPx5QjKY2wHAH489BO/LCxmDd+2Mwl1x7Fz19tY/fmCjatKqLbwFh8gFA4JjiurhyJudWSyhFJ22BPXLwlYo6oGA3GHJHsWzTVjRKvH1E11h7xb5F6UDpsaDNEReWCQqkckUgkBxL7g+UIxMqp61HTrZYcm0vaEgedckRRFP513UgCkbYxKHJatWavJMrJyWHOnDm1HuvQoYP522q18uijj3Leeedx1113cc455zRK2dC9e3dTmNxYVq5cSTQa5bjjjkvaf+qppyYpR77//nuGDx+Ow+EwrT1SUlIYNmwY3333XdK5/fr1a9S1jzvuONNaQAhBcXExn3zyCbfddhuBQIDzzz+/znPbt2/Pq6++yrp16/j6669ZsmQJS5cuZdasWbz99tv83//9H926dWP58uVYrdYkl1SqqvLWW28l5Tdw4MCkAO6dO3cGYsqjhMDd4/GYihGA7OyYcGfQoEHmvvT0dPM8q9XKypUrufLKK5OsZLp06UKPHj349ttv64yrAjF3YNdffz05OTlMmzbNfO8WL15Mv379yM3NNfNUVZVjjjmGDz/8sM78aqtDp9tFvwGDWb5sCYJK5UifKgqPjMyYwqL3ob1Nf/AJJUZFRUXSNc4555yk7ZNOOonPP/+c/Px8lixZghCCsWPHJllQjR07ljlz5rB8+XJTgVX9Xe7Xrx9vvPEGhmGwceNGNm3axLp169iwYUOd1lhNpbHvrUQikRyMBCMxRUZj3GoBnDW4Iw8t+I1NRX6W7Sxn4LGd+PE/m/nh43y6DshCURRC8bGdVa2uHJGWI5K2g7IHXo11pFuttopFc0M0rhwRkGxzr3DgLM/bP4kaVZQj0q2WRCI5gNi/lCM+DNOtllw0IGk7HJSjakVRcDUiAGhbx2azMWDAgEal7devH3369GH16tU1FBd14Xa7G05UjdLSUqBSqJ8gJyenRroFCxawYMGCGnlUt/JwuVyNunZ6enqN53HsscdSUFDA448/zrnnnmtaW9RFwnpm4sSJRCIR3nvvPR588EGeeuopZsyYQWlpKenp6agNrP6qXuZE+qruyRJKkobOTVBeXo5hGDz//PM8//zzNY7b7XV3ioZhMHnyZAoKCnjnnXeS6ra0tJRNmzbVaYUUCARwOmsG/qqvDtPjyo4EzpSa75LdYa/ukLkGCeumBAl3a2VlZea7dvrpp9d67q5du8zftb3LL7/8Ms8++yylpaVkZ2fTv39/nE5nDQVNc2nseyuRSCQHI2N65zCqZxbnD+vScGLAZbMw7vDOvPztRl79biMzzx1kWo/kryyk++CcypgjCeWIGddNmEEfJZLWpmroCSFEoxdJVVT8ytadXwMj4m61pOVIW6Jq3BEhjGoB2WXMkdYmKeaItByRSCQHEPuLckRTHUQA3UhYjsixuaTtsEcagp07d3LGGWcwa9aspFgVmzZt4pFHHmHZsmVomsYpp5zC7bffniQM9vl8PPHEE/z3v//F7/czbNgw7rzzTrp3774nRZLUwbx581i9ejV9+/bloYceYuTIkaSlpTUpj8TkrXr8EZ/PZwqfExYARUVFSVYRCUF2gtTUVI466qha3Q5ZLC2ruOrfvz/fffcdJSUlpnVGVebOncucOXP48ssvk5QAVquVCy64gIULF7Ju3Tqz3KWlpTUms7/++itCiDoVDC2B2+1GURQuv/zyWhUCtSkwEjz++OMsWrSI5557jq5duyYdS01NZfjw4UyZMqXWc+uyHqpeh0IINu/2EQFy05xJeo+6poOiAe1ISUkJhxxyiLldVFQExJQkifd37ty5tSo/OnbsWGe+H330EdOmTeP222/nz3/+s6mQ++tf/8qqVavqPK++NiCRSCSSxtMt283rV41o0jmXjujKK99t5PPfC9jiDzHwuM78+OkmvntvHV37ZxGKxgRetmqWI6q0HJG0IaqulEx2vlQ/Pyw9k0KOAmVEzBpKkcqRtoSmuc26FIaophxBxhxpRQxhYIjKsbu0HJFIJAcU+4nCNxEbRdcTshOpHJG0HZr9Nu7YsYOJEyfWWGVdXl7OZZddRmFhIdOmTePWW29lwYIF/PWvf01Kd+utt/Lpp59y66238uijj7Jr1y4mTJhAWVlZc4skqYNt27bx6KOPMm7cOJ599lkqKip46KGHzOMNWVMkSCi3du7cae4rKytj/fr15vaQIUNwOBx8+umnSed++eWXSdvDhw9n3bp19OvXjwEDBjBgwAD69+/PK6+8wv/+978m32N9rFq1Co/HYypuqtOzZ09KSkp47bXXahzTdZ0tW7bQu3dvIBZHIhKJ8PXXX5tphBDceeedPPfccy1a7uqkpKRw6KGHsmHDBvOZDRgwgF69evHMM8/UGRPm/fff56WXXuKWW27hmGOOqXF8+PDh5Ofnk5eXl5TvBx98wDvvvGO+H9WtZarX4aF9D6NHr3589M4bfPP1lzWuUxuVq3pr57PPPkva/vTTT+nUqROHHHKIGd+lpKQkqdzFxcVMnz69hkKuKsuXLyctLY2rrrrKVIz4fD6WL1+epPiofs+1tYH169fXey2JRCKRtAzdc1I45bD2AMz6ch2Hn9wVZ5qNsoIAq77ais8XAcCqVR/eSuWIpO2wJysljfjUTcVA1fZ/K/gDCYvqRiHhVitZOQKKjDnSilS3FJGWIxKJZH9m+BEf06/vI+Z2KFzQiqVpPAkLF8MIAwduzJEt5VuSXDlK9g+aPKo2DIP58+fz6KOP1nr8zTffpLS0lPfee88UOubm5nLNNdewfPlyDj/8cH766Se+/PJL/vnPfzJmzBggJnQ+/vjjeeONN8xA0ZL6CYfDrFixos7jffr0weFwcNddd+F0OpkyZQoej4ebb76Zhx9+mJNPPpmxY8eSmpoKwFdffYXH46Fv37515tehQwdmzZpFSkqKGfi6qsWC2+3mqquuYvbs2bhcLkaMGMHChQtrKEduuOEGxo8fz7XXXsuFF16I3W5n3rx5fPbZZ8yYMaNZz6O4uDjpeQQCAebPn8/ixYuZPHlynUqgUaNGccYZZ/DUU0+xZs0aTj75ZDIzM9m5cydvvfUWO3fu5OmnnwZibrqGDBnC1KlTufnmm+nSpQsffPAB69ev529/+1uzyt0UEsHHb731Vs466yx0Xeell15i5cqV3HDDDTXSr1ixgnvuuYfhw4czevRoVq5cmaSQOPTQQ7n88sv54IMPuPzyy5k4cSIZGRksWLCAt99+mzvvvNNMm5aWxo8//sjSpUsZNmxYUh2Ov2A8ekjhjfn/4odvF3LOGf9ItgmpulHl+g1Zjrz88svY7XYGDx7Mf//7X7788kuefPJJIPY+nnXWWdxzzz1s27aN/v37k5+fzz/+8Q86d+5Mt27d6sx34MCBvPnmm0ybNo3jjjuOgoICXnzxRQoLC/F4PEn3/NNPP7F48WIOPfRQjjzySBwOB9OmTeOvf/0rPp+PGTNm1HAjJ5FIJJK9w41je/LJ6p18tHI7N5/QmxFnd+fL135nyYcb2J4S61NcztjwtmpAdulWS9JWqCoMaMC7aA0M4gtWpFutNodmScGsUZE8whUoIC1HWo3qQiqpHJFIJPszqan9SE3tx2+/x2Q1enT/8GKhqo7kHQdgv/jF5i/465d/5cSuJ/LUsU+1dnEkTaDJypE1a9Zw3333cdFFF3HUUUdxzTXXJB3/5ptvOPzww5PiRowePRq3283XX3/N4YcfzjfffIPL5WL06NFmmszMTI444ggWLlwolSONZPfu3VxwwQV1Hp8/fz4//vgjixcv5umnnzaFvpdeeikfffQR9957L0OHDqVXr16cccYZvP766yxatIiPP/641vw0TWPGjBk8/PDDTJ48mezsbC677DI2bNhAfn6+mW7ixIl4PB5effVV5s6dy5AhQ7jjjju4//77zTR9+/bl9ddf5x//+AdTpkxBCEHv3r2ZNWsWxx9/fLOex8KFC1m4cKG57XK5yMvLM9/X+nj88ccZPnw4H374IXfffTd+v5/MzExGjRrFI488QpcuXcxn8Pzzz/PEE08wffp0AoEAffr04aWXXmJglaDje4vRo0fz4osvMnPmTG666SasViuHHXYYL7/8MoMHD66RftGiRUQiEX744Ycawc0BPv/8czp37sxbb73Fk08+yf33308oFKJbt2489NBDjBs3zkx73XXXMXv2bK6++mo+nP8RnXK68fycl5n17DNMueMOhBB07tadKQ8+zumnnkJhsKTB+2nIcuT//b//x/vvv89zzz1H9+7dmTFjBieffLJ5/JFHHuG5554zlVhZWVmcdtpp3HzzzfVaRJ1zzjls3bqVd999lzfeeIPc3FzGjBnDRRddxD333MP69evp0aMHF198MatXr+bqq6/mkUce4cwzz+SZZ57hySefZNKkSXTq1Ikbb7yR+fPnN3ivEolEItlzDuvo4fi+7fj89wJmf7mOx84dyNofdrJtTSlebxRc4HbH3EFWdjFCrrKXtBmqigJEU/xqkWw5IgOyty0sSTFHqlmOCJAB2VuP6m60pFstiURyINCj++1s3vIiXbvuH/JTrVpslAPRcuTF1S8C8L9NLesNR7L3UURD0slqlJaWEgwGad++PUuWLGHChAm8+uqrZsyRo446itNOO42777476byzzjqLbt26MWPGDG6++WbWr1/PRx99lJTm73//Ox999FGd7oGqous6K1asYMCAAXUKQYPBIJs2bSIvLw+Hw1FrGknLIoQwA3hL37oHJkII8rdVEFAEmlBItL6gEvuUOK0aPdulsDtQzI6oE6fuJ6JaicZ9Y2dECqgQsdUNHruHTimdalxjyZIlXHbZZcydOzcpntGBwJ62kWAwSH5+Pl27dpXfNckBia7rrFq1qt7+XXJws2JLKec++z2aqvDJTaNop1l4/Z4lLLdF+cIV4dTDcpl50RA2BEIcvXQtTuHju9JfaXf2xGZfUy8N4f16G9YObiy5LoK/F6M6NGx5HvSyMEZZCGunFNRUG3pxEBHWsbR3YQSiRLZ60VJtKDYVwxdFRA0i233YuqXhGJDVrL5AtpP9l2VL7uLs0HgANh3dH0u1+hdC8O/8f+OP+hnZYSSLdyzm2M7H8uvS0XzF8Tyv3MAQsYx5g4/G7enZGrewX7Cv28j636dxwq4TCCt2vhvcncm/LuL7cGyM2yW4iS9zu+AcNGivl0NSk6JAEce/W7n47vGjH+fErie2YonaBrIfkUjqZ39oI9Vj4bZlfl51NcXFlYuZDzv0GXJyTq7njP2Piz+5mF+KfgFgxSUrWrcw+4i23E4SZRs8eHCDZWvykqOG3MdUVFTUGhzZ7Xbj9XrNNFWDs1dN09TAxvUFT4ZYcO9AIFAjgLJk7xIIBFq7CJK9RCQq8MUVIYYiiFQ77tDA7/cTja/KUupRv0ajUfx+f439oVDI/Fvb8QOB5raRUChEJBLh999/b+ESSSRti4b699ZADRjYt+mE22noafW7aVIDBkJVEPaWm7AoEYFQAW3/mATtTY7oaGfp9hB3zlvK/xudQb/T3Sz7Kha3zu8tY8WKFWzVATQUYPuOnWyvxxVpfSgRQdZ/g2j+pjpBqh//kp2U5dsIdm2+BUBbbCeS+ikpKQNX7PeKFSuxVGvOS8qWMGfLnKR97//yPtdmghG3PlAwWPP7H6h2774o8n7NvmojgYoiM+bIb7/+SihS1ZWTwh9//IHRtDWJkhZie3B70vb6/PXklOS0UmnaHrIfkUjqR7aRlsFbTdabv3ET27ataJ3C7CW8vspxWX0hEA5E9vd20uL22PUZoiQ0mo1J01gaYznidDrlCut9hLQcOfAprQgBsQlf5wwnuhFrzxZVxW5VcVhUFEXBGwjGk9Xd3lVNxeVy1dhvt9vNv7Ud35/Z0zaiqipWq5WePXvK75rkgKQtrj4xQjrBX4uoWLQZoywCRECtr/0KSKzJqDddUwsi6s1TS7PhGJCF5rHXenxfYu+TgSVz732jHuns45Tp37B8R4gSZyeOOy2H5Za1fLlwAx3b5TB48GGk+IOw7A9A0LnLIWTX4n6yMXi/3kaFfxMAapoNoyKMLS8NI6CjFwbQcpwQMdBLQ6CA4rSgqAqGL/ae2LqkonsjoBsIXcTqUVXQi4J4fgjjWVZ9mUHj2J9WC7YEilVFdVswKiKx57if4u+VAt1ivwcOGoitSqBu3dB54OMHapzzm+83yEx2q3Vo/4E43DWtbyUx9nVfkv/Hl7A99l727dcX628lEIs5i0Chd58+OAYM2OvlkNREFAhYV7nd5ZAuDO4+uNXK01Zoi+MtiaQtsb+0EcMfIfhrMSJqVIpeDIEQAtVtRbFqiFAUEW7dBePrDBdVHa93z+tBdvbg1irOXsG+1Q7B2O/a3N4fiLTldpIoW2NoceVISkpKrdYfXq+X3NxcM01hYWGNND6fzwwO3lg0TauzAjRNQ1EU859k3yGf+YFLKBLrVJ2qSqa7cQI4pQ4FiaB2wc6IESNYs2ZN8wu5H9DcNpI4r75vn0RyINAW3nFhCAKrCin7JD8m+E6gUKmoaIjGpmsKdeSpl4bwLdpe67F9jf3XYnKu2XuxuHrmpnH5Ud144Zt8/t/7q/nPzcewtSw2Gzkkyx0bA6qx90eJxxxpzvsULQni/WorABnn9cZ9eG6LlN8IRSl8+RfCG8ub/Y4oUDWwygGPCOnoof0/VoBaRTahaRpaFeXIfzb9h/zyfNJsaXx67qesL13PlK+nsMO3A0hWjlgsjgbf6WhpiMi2ithGba+KRcXazkVgVSFqihXXkHYoLanQbQPsq75EVSun1aqimFY+5r5mfoMke443mmxhJRQh66IKbWG8JZG0ZdpyGxFCUPyvdYTWNhzvtbWJDgxA+8rt8LpywqVlOHqko1jqt8jfXzCoHOS11Xdmb9GW20ljaHHlSF5eHps3b07ap+s6W7du5aSTTjLTfPPNNxiGgVplQrBp0yZ69OjR0kWSSCQtSDga++Bb1fo7sMQcvN611QeRUKctIoRALwuhOiyojv0/sKvQBcp+6G5IGILwlgqMQBTNbW3t4rQ6uq5jKdYJb6lo1QFWtChA+eebie6OucDT0my4Ds8l9ehOoCmIYP1CWsWuxVa3R1tulZZi12KrwmpZNS8EhNaVEvyjpPUF5oqCe2i7vX6ZW0/qw5drCli/28dd81exrSRWV3nZMYvDqv2Q0oSA7OFtXioWbUUvDRHe6oWoge2QVFxDWu6eVLuFdtcNQvdFmvWO6LrOL7/+wmGHHrZfT0QaixDgX7YT3RvBfXguWpqttYvULKKlITZ99rG5Xb2lfrPtGwDO73M+qbZUBrcbzNtnvM3MFTPB/3JyQHa1/nc6tKmc3c//DNHGfw/CWytwD22cAjCyy0/Fwi3YuqTiHtGhzS2Kakpfojg0tDQ7wd+KQFOxpDfe+k7LsKOl2FAUzRzzVh/fChTeW/c+Fw/o39TbkLQAZaGYy8VDQu3ZZS1CF/u/klUikUgAQuvLYooRTcHZLxMUJT7wVVCU2LgDQLVrKDatfuHMXkQvDRHzDVyJb/EulMJfSD3+EOxd0/At3Un62T3QUvbPMR6Q1L8YwkBVasrMdG+Yso83YMl2Yu/uQbG2zDjekuM8IGQ6rUWLP7lRo0bx4osvUlxcTGZmJgDffPMNfr+fUaNGATB69GieffZZFi1axJgxYwAoLi5m2bJlXHvttS1dJIlE0oKE9ZgQx95o7X7yBFEB7JqdkB5C1ONyqzkcbC5G9gRhCEreWYv/xwLQFBw901Hs8Y5ZxFZLi4iBlmaLdbIKGP4oikVF6AYipKM4LKj2aoMsXRDe6QNdYGnnQlHACERjAzWV2IpUVYkN2NT44E2NDd4MfxTFpqFY4++WEDEZb+I1SQgbqm0LASJiEN5cHguWHHflI0I6kZ0+LO1csXsQAkT89Cp56GUhNI8dxabhPqI9jp7pe+GJJyNEzCIh8EsRkR0+ogUHZmyd5pIFFH3eNvyWKk4LqaM7kTKqY/KA0972Bp+WYbm4h7WMZcP+gNOm8Y8LBvPn2d+xYNVOc3+37Fjsu8o+RphWJPVhhHUiO3zsfn5VksJC89jJvLDvXllR32yFqK5jOFU0j/2gUI4ApJ3QtbWLsMeoKbYk4UB1PWbYiPlhynZmm/vSHekc0/kYImtfRiQpR+p+d4xglOI3f4eoQLFrWNvXjAcJMQWHCEbBokLUwLd4B77FO5p0T9HdgdhYog3SlL5ETbFieJvu4k5xWmh/27C4sioxLhFJliNCUXhr7Twu5p4m5y/Zc8pCZRzuPZS/b7mRPxybEbkWIh4/lmznAWcpJZHszwRWFxLeUoGzfzZamg0jGCW8pSL2adUUrLlutJRY3yd0gRGIorosGP4oaoq1Qbm/AAxvBNVlAVUhst1LtCiI7ZBULNVc0hphHT2qo/oNosVBDEUFQ2B4I0TLQtg6p/D/2TvvOLnK8u1/T5neZ3vLZjeb3gOEHnoVFGxgxYKCYkMR9f0pCoqgIIg0CzZQKSoCKiAthE4IpPdsdrO9zezs9JnT3j/OzuzOlmRTIAnsxWfJzDnPec5zyjzlvu77ukXr0PxLT6sgCgXbDFVH6Uqa62mXxYwUFsxrYdBmofWnESRxlNOHYYDSHkMbyGKd4kF0WwABLZoh2xTFeUQpkttK8s1uAFxHlhG4cPq+3PZ3BKlNIXir0H4kGOY9iD3bwmCMK6JTPqSvY0/QjaH1Q0pN4bKMnn+F799CpnHggJ9b9Fqp+O7SyXFtH3HAV/Yf//jH+ctf/sJnP/tZvvKVrxCJRLjppptYtmwZS5YsAeCoo45i6dKlfPvb3+bb3/42fr+f22+/HY/Hw8c+9rED3aRJTGISBxDa4EreMjKD6AgYQzapgomKgECps5TWWCuqrpLKpLBb7XlSI5wOk9EyyIKMois4ZAd+m39M0iOjZkhraeySnZgSozfZi0224ZQP7TwliqIQS8b2XHCsYzMKA5kBfr/+9yTZs0FdN3R2DuzEKlqp8dYgDD6NWU2VLH5rillIM0hvHTsUV+3ed6N9gQzROwSlLY7SViifoA1k93xchykHmW2JUv7to962SUW2NUZqYx/Z9jiZ7ZGhHZKAXOzAyE56MwJkM1mstoPsNSQI2GcE8J09ddIL5xDGgmo/3zxzBj9/ckiKsTY4SI4MylUJeyBH9JRK+MGtpLeE89ts03y4jq5AtMtYa70mETyJSewnBElAsozvXJJbVMtCYZ9jES0oDCVkF9ERpbHJET2t0nPXGrRIBslvo+zKI8Z9f/WMitKRwFrrZeDxJlLre/MGm4l4l2oR07ngYHmi7gkTGUsMRUePK0PEiADSBCNH9ISKkVKJv9yO0CDlZWRNX4xh5AhgCKDoCpbdkFqTeHsQzUb5ctdHAZiengIvQPcLbyL5bLiOqUBPKXBwpfgPCgxDx92bJdrRhDCGd/O7EVokjRS04zunbtKh7hCDnlQIPWCS+rEVbQe7OXmUAL3/fetgN2MU9JSC98yppDaa6QqcE4z6PFgQrBIChf2MrS4AocJyuYj9wxVZbcjuEM/GR5EjSm8yT4yIHguiTcbQDswAZJvmnyRG9gMHfLUfDAa59957+elPf8pVV12Fy+Xi7LPP5uqrry4od8cdd3DjjTfy85//HF3XWbJkCb/85S/x+XwHukmTmMQkDiCG8gHvgRwZ/DeviT6suCAI2HQrFZlipISKwpAx240FN7mFoxkBMHz/cAiAAxHI4sWGl+q9vZzDDmk1SyRp433LFyLHJhp5c9y4e35X+k+221uoyxQmdY1LSSJSjCXOhTh0G62xVgSrhK6qxEigywYVUjmGoqFoCqIg4rV5SSkptgvNIEGZWoxTdBCVE8wKzKTOW8f67vU0RhqRDAEREZ/FR5m9lHg2zq5sK1bdQpmtFJtkoyPRCYLp+21gIItmBEtWy2IMBp3o6BiA2+qmPFCJJ+ukLdqKy+IirsTxOnzM9M0go2XoTnaT0TLYZDszgjOYEZzB5vBm+tQQ0+0NBF8F+jP0P7wdS9mBJ9i0SIb4Kx0FPw73CVXIJQ7sDW9v8urDCZqmsWbNGhYtWvSe8YifxP7hSydNQ1ENbn1mGzPK3DgGvfZykSMCwDgSRGokQ8+da9BjQ4sZS7WbokvmFnj/TWISBwrDydaRo7iqqwBII8i8nEE9J6slYCAOM7IbhkHkXztQ+1KIThm1J4Vglwh+bNZuiT3RJmOrM9de/vPq8Z9Xv8/XdahhomOJnlbpvGElRkZD8lop3wuvy+SaHsIPbCX2XCvR7W0wey5g9j0j46YNAfqSfVS4K/bqOgxVJ7UphNafMeW6dGMw0S5DnzWD9NbwYW9UertwAQtHb5QEtIEM0f81v+PtOZTgAhLb9i5a7N0Aa40X5/ziPRecxDuG5OoeM9rRIYNmmAZjQcBa7UawiBiqjtqdRM/lHhNAtEroGQ3RJg1t3wOGlxWsZmSl0h4zpXCHQRiM7tAVDVEWB1UQBARJwFANDKXwfIIsmqoHI+qR/DaMjIaR0Uz1hMFE6blBwjyOMQ3kolNGLnagtCdMhx/DyEvrajEFIz2UZN1avXe5m99piDYJQS8kR+SgY3TBw1Aiezhi2SEH2ISayEfyakkVe4PflO8EbNP9lHx+/sFq5iTGwH6RI0cfffSYSZNnzJjBn/70p90e6/P5uOGGG7jhhhv2pwmTmMQk3mHkhntxjwtHY8S/Q5A0kUqlBNmYNDwdLBgYPOV/leUVb3FW3VnsiOxgVfcqgvYgx1UexxutbxJTYrzFZhCBwDgVjbTnj2PffzL5AvlAl7E4cMfg3zBIdok5RXNY3zdBeaVcoEpubphrS2LwX3HwD6B/8C+HGHzZfRHn959EclX3xM63j7DPDGCt8WCfGcRac2hPZCcxicMBgiDw9dOnc+qsUoo9Q17iQ6HtBqIo5b9HM1FWdq2kxlNDyX8N9FgW1Qv31T9JKNVHzazpLOjs44iyIwjagwfhikxktAzhVJhyV/mEPFwTSgKn7DysvGF1Q8cwDCRRQtM1epI9lLnKxtRohneHfKYoFUYUDEdOq1oSRpAjUiE5Iho6qqFjGSyX3hgisbKr4Bj/+dOw1XoPZNPflRDtMmVfW0xiVTf2mYG98rp0LCzB3RIzHR8QC3KOGCNktQwBupPdEyZHDN1g4PEmEq93YijvwbCGtxEvet5i1RG7+EroYzi6BKw1HiTP4atxv6/QDYOe7m5Ky8r26PT2bkH8lQ4MRSfyWCOxZ3ehZ3Ucc4rwnTV1SNb3bcR4Y1h7vJ3Hdz5Otaea4yqPw2d77zkMp3dEAPCeUoNn2aHh8HgoOmzFX+0g8mgjRkYdIoRk8ZDPu2nKd4+Q1ZJk7POLSa3vy2/Toln0tIqeVpH9Q4YFVVeRBAlBEAilQnisHqzSodVva7pGXIlzdGw+n+49H/mODjri7fnJntI6RJw45hQdpFZOYjxM6kRMYhKT2CvkyZGJRo6MEdwgRwwwJDKCQpu1G7tsJ6OmkUSJgC1AkWNosEiraULpENnB6ASXxfTolwQJu2zHITsGc0gcHgYTw4BUOoXD7mBfmqulJaSEjZLLp2G37XukgSCLfEJeyqfFb+Y9UuPZOHbZjizKKJpCS6yFNT1r6E31srBkIYqukFbTHF1xNCu7VrKqaxWLSxfjtDgREFjRtoJjKo5hQckCVF0lraWJZWMMZAZ4tuVZmgea0Q2dz877LCdVn0Q4Heal9pfQDA27bOfIsiORBImXO15m58BOzq07l5mBmTy16ykMw2BmcCayIOOwOBAQGMgOoOs6xY5i/r3z37TF2hAEgfnF89ENnZSaYld0F9FsFN3QKXOW8fSup837aGi0xlqp9daS0TJ0Jbr4Xek/6baEqEtX7e7WEbAHqPPWUeIsQRIlwqkwXpsXm7QHGQ5RwNbgx7mo5B17V7NaFlEQzaibSUziXY751YXGBGNQJ0XA4PqVN/Dkru9gFa0IgkBGy1CRLeYPjdeho3NF8fW0ZDrNddu2V2AbLCxZyH3n3PeO/F4Nw6Aj0UFjpJG/b/s763rXEU6bMl9lzjJKnaWjyieTSZydJhkSy8ZojjbjkB3UeGqwy3ZERERBxCJaWFqxlNnB2aS1NKIgErAFRl1Xg78Bj9VDR7yDgcwADYEGLKKFhJIgnA7TlzIXr9FMlKyexWP1IAlSfrE6FuySnZnBmSSVJDbJRlu8jeaBZlZ2rWRD3wa2R7ajGzpTvVPpS/XRm+olaA9S5S7sh1VdZUdkB4quUOOpodpdjTRIeKXVNGWuMjrjnazpXUODv4FiRzF2yc77G96PRbTwVPNTzCuex8WzLqY52kwimzAjDwc1QKcHppNSU3Qnu5kRmEHzQDNV7ipUQ6U11spU71Tskp3+TD9F9qJR15ubq5Q7y2kcaEREpN5fT0JJ4JAd+T54ectyUBKMhzw5sofIERGd0x46jQff9yCOF1LEX2zPl3UuLkUK2nEuLnxn9hZ7S0S1xdp4peMVREHk9CmnE8vGKHWVouoqNslWMA7phk57rJ0yVxlJJYnP5sufS9EVdkZ2Uu+vxyJaSKkprKK14J7ohk7TQBNTvVNRDbVg/NUNndZYKxWOCqJqlHA6jFW2sr1/O0vKlhQQb4qmsCu6i3s23YPkkni/7f0s0hbRHmunxlODLMrctfYuIukIn5rzKURBpDfVW3DdxacUw5xKhCdFxss5wuCea1+9lh8c8wNsso2/b/07Jc4SGvwNpNQUz7c+T1+qD1mUyWpZjgjP4qKNpwCQcWr0lsRw2lz0ZfoIZ8PoGBiCgSGAjsYWZQdv+DZx6rTTaPA30BRtYmdkJ6F0iNZYK1M8U/jC/C8QV+LsiDTisbrpT/fTm+olno1T7anGZ/Oxvnd9XiddEATW9a5jVnA284rnYhiQ1TOk1BQD2QF8Vh8JJcGcojm82f0madX0UvHY3Ez1TCWUClHlqWZTaBNHly9lZnAmm8KbiQye1zAMAvYgm0ObzAhgVzkNvgYq3ZWk1TSre9fQm+zhtNrTCNqL2N6/jYHsAPOL59MY2Uko1UeVu4paby1JJckzLc/QmejCIsoE7AF2RXcRzyboS5vP7OvTriD02i7+XPoY8e4kL8ov8a33fYtSZykzgzPpSfbQGGlkemA6C0oW5Putkb/Hdws0TWPnmn4aFtUeMobftxue06bQffMq0wA7GDEaf6md17a+xH8Xv4FFtHBm7ZmUuYYkimLZGK92vIrL4sJj9fBKxytktSxxJZ5fl2TUjNnXSzLhVJhQKkRvqpejK44mpabQDZ2kkqQ52kyNp4bZwdlEMhEyWgZFVwocwYodxVy24DLcVjciIpIoUeerwyE7WNG6gu5kN6XOUuYXz2db/zZUXaXIUUTAFsDAQNEVVF3FMAxsko2dAzsJp8MYGMwMzGRt71qaBpo4sfpETptyGr3JXlwWFyktxfb+7fmxvTHSSJWnijKneS8csgOn7ETRFSRBYkXbCpwWJ3OCcyhxliCLct6hJJwOk9Wy2CQb4XSYWm8tfak+BjIDuCwuqtxVpNQUnYlO6n319KZ6UTpMw3EkmGJly7NYRAsnVJ1AXImzJbSFSnclL7a/SIO/wZznSHb8dj+t0VacFqepMiDIrGhbgSiILCpdRCQdYXN4M2fXnY1VtJJSUzzT8gxHlB3BW91vcVT5UXitXnqSPWS0DFN9U/FavTQPNJNQE0zzTjPnW0oSj+Qhko4giRJuixtBEEipKRzyGNEPg8iNdaWuUjriHVS5q8Y06Ge0DB3xDqZ6pwLmXGDkmJlzJBEGo0/VlIKSGkq2Phb60/08tPUhar21LC5dXPBeHwiousrqntUklAQ+mw+rZKXB3zDmmliwSaMSsvem+ih73xJC8TaKmky7htqTpPOGlWYewBoRq8UOKY1Nya3cWf0gs6bM5aldTyEgMCMwA0mUqPfVU+IoodxVzhtdb1DjqaHOV4eBwUnVJ+GQHQiCgEW0kNWyWEQLzdFmqj3VSIJELBsjoSTw2/z0Z/rpSfawtmctH57xYbJ6Ft3Q8Vg92CQbA5kBQqkQ9f56IulIXuLdaXHyeufrAFwUOov6TDVkcuFB5twsZyCTy5y4jiw/oM9iEvsPwTCMMUyXhz4mwuKm02mampqoq6vDbp+UK9kTDoQ3Xn6x7jy8PBcnMXFsaBtAx6A+6MLtHF83uT3eR5/uxJONkpWtZETzN1ii9FOc8qIIGq3WTjRBZ27x3Heq+Qcd+/sbmezXDgwMw6An2UOJswTDMPLGgQsfuzBfZlHJIhRdIa7E2RXdhSzKecmTkbBJNqySFa/VS72vHqtkxSbZMDDIalkqXBU0+BtQdZUPz/hwfqEdy8ZwW9w81vgYW/u38v5p76fCVYHH6skTTIqmYJEsRNIRFF0ho2V4peMVdkV3cWLVidR6a3mt8zXskikXti28je2R7Wzo28DW/q2ousp59edhES2E0iEkQWJe8Twq3ZUsLF6IYiis7l5NX8qU+1havhSbZMMu2xEwJ5J7YxjQdI3uZDcbQxvZFNrEhQ0X5hdLdtlOQknwzK5n2BDawKfmfIoqdxV9qT6sohWH7CCcCrNp0ybmzJmDJEkICBQ5igilQhQ5itjYtxGbbKPcWY6iK+iGjl22IwkScSWez2sUzUZJqSn8Nj+KpiAIAl6rl4HMAD6bz/TKT4cRELDJtrzxLWeQFhAod5WP8mBXdIWsliWhJHBZXEQykXHvhSRItMZaWVK6hN5Ub97wCaaBeX3femo8NVgla95AnFJTqIZKR7yDjJZhmm9a3mt8JCLpCBv6NlDrq6XGU0NjpJGuRBeV7kpaoi0klARLK5YWJHcGMzntM7ueodRZSlpNU+4qZ1n1Mp5oeoINoQ28f9r7SSpJNoY2Ms03jY2hjbTF25jimUJGy5DRMswKzmJ7/3a6El3U+ep4X/37CKVCxJU40/3TqfZUH/R5wNqBAc56qwm/0c/RL1/BW1OG3uNiRzHn7Tqei0Jnscq1kR9MuZP3T3s/td5aXul4hXW961B0hXvPuZfFpYvzx2m6ll/EVnuqsct2NF3DwCCajeZJB8MwWNO7ho54BydVn4Tb6iaSjpBQE/kyN7x+A1O8U1havpSb3riJNb1rDsJdGoLL4qLaXc3WfjMq3Ck7mRmcyfq+9eP2fRNB0B7M/64ONlwWF4kxyAlREAuSaI6EQ3bkyRG7ZGd+yXy8VjMqQzd01vSsoT9TmLur1FlKOB3GIlryxpNwOsz/U5Zw5bQrAWhcNh/XsHXMF576Aq91vsYNJ97AefXn5bdv799Oy+qz+Scf4WHhYk5T/0e68UVubf424iBh0uMb4LmTNpMRTaPdEWVHUOmu5L87/0tCSSAgMKdoDuF0mNU9q5kVnEVGyyAKIpqhkdWyeQNMc7SZ5mgzHquHgC1AuauchJLAJtnQDR2bZKM52oyiK9R6a7l0/qX8+NUf05HoKLgHFtGCZmjoho7P5sNv89Maax11r3NjRDQbxSE5iCkxgvYgRY4itvdvxyk7cVqcqLqKpmvEFNOI5rV6iWVjzArOIqWmCKVDBZIWIzE9MB2raCWajSIJEs3R5nHLBmwBip3FbO/fPm6ZHJyand8nPsAH5p5EQvCwYk4VV+xYy4as2feWZzsp3/Bt2osFjojP4YToYt5wb8CtuyhWfGREBadmp9Pah09zU5upYF6ygQqlhCf8L/Gr8r8dsnldDifcd859/PyNn08oInmqd2r+N3Bc5XFYRdOgeVLNSSwuXYxFtGARLbgsLtMQjTGm0TOtprFK1nGj4Q4mDkWv+HcCmaYBGp9Zzb3pv5M00nyr41NISHxt6o1sd7Qc7OYhGAKlSvCg/eT31ThoES3YZftu++AcGvwN9CR7iGajuC1utIzK3U3/B8CHZnyTpJSe0PnKnGW0xQ9cfhKH7KDSVUnjQCNgjgOiLhJSQrgt7vw6w21x58ehOUVzKHWUmuOonkXRzPygrbFW3FZ3wfynzFnGjMAMdg7sxCE7CNqD6IbOjsgOIpkIs4KzCKVCRDIRFpUuwmv1ohs6G/o2oBkaMwMzmdZbwSfWn842xy5+X/EvfrbzG/Ta+kl9sZilFUv5+7a/s7JzJQYGb3a/mV+nyILMh2Z8iEvnX0qJo4TuZLfZb4nW/Doh76SgKWT1LC+0vcCLbS+SVEfnIDUMg8aBRnZFdxVsD9gCLCpdxFTvVLqT3bTGWqnz1bHYu5C6t54hMuXZfNlV24/hL/Z1WHSZj4TO5FN95408TQEeCzzP3eUP7eVTHcJUZy1tiTZUwVyPOWQHKXW0HKVkiGhC4TwlRw6G02EwzIjerD46r6lFl/nn1luwIHNN9V1MD07n4ydeQm3NtH1u96GOQ3ks2Zu2TZIjhym++93v8q9//Wvc/cXFxbz88ssTqisajfKTn/yEj3zkIxx11FH71a7xDL8PP/ww3/ve93j22Weprj7wYZKnnnoq7e1DXnOCIODz+Vi8eDHf+MY3mDVr1h7rePjhh3nooYfYtm0bmqZRXV3NmWeeyec//3ncbvdetWXp0qXceOON+3QtBxoHuj0b2iLoQEOxG+dukhS3xfsIjUGOlCkDBFNuei39RCRzUrGv5Mjrr7/Opz/9ae69916OPvrofaojh9tvv5077rhjTKnAA4lJcuTQxuM7H6c11sql8y8tIARaY62UOktpjDRy3avXsTG0cb/OM803jYyWGXdCLyCMUiw/WLBJNup99ZQ6S9kY2sis4CyC9qBJDKQjdCW7mOKZQlJN8lb3W4TSoXHr8lg9JJVkAUlgFa1jTi73B2MZOgUEfDYfkUwEt8VNUk3u1hiaQ5W7CofsYHpgOuF0OO8VtDd4O67xUMe5defys2U/O6htWDvQz1lv7cJvhFn60ldxB47k67VfRnTKlGQDxF8w5w2954gsXHZsgfHqh6/8kIe3P8wZtWfw8Vkfx2ro3PzmrawObciXkQSJgD1Af7o//057LB4cFgfxbDy/mHTIDqb7p7O+bz0GBg7ZQbGjmNZY66g2+21+zqw9k3PqzqHMVUaRvYgNfRtGLd50XWfnzp3U19cjiiKCYHrQpdU0bfE2MzeTYaAZGjsHdvJsy7MICDhkBxktQ1wpzOXVn+4nmo3mvztl56jFcJW7ClmUTTJYtJLW0uiGXvB7Hg7DMGiJjTY0TfVO5cjyIzmy7Egq3ZV4rV7a4+0ICMwtnkvzQPMo40o4HWZV9ypC6RBLSpdgk2z8r/l/HF91PA7ZwZbQFuaXzGdhyUJ6k71sDm9mdY9J+uqGTneyu+AeVrgqkAQJURCJZWOjiI3hGG/xPB5yHpMZLTPm/v+nLObKad8EoPHE+bjkobHm8//7PCu7VnLTsps4u+7s/PbmgWYa3zyNf3AR/xI+yhnKE5z2pszpA8cA8GDR//hLyX/yC/6DhaA9SFJJktb2bNg6lOCxevLE4HjjgiRIaIZGhasibwRXddV8dw2Bv0Uv4gPzlpnkyOxKvrxjPRsVMxK6LNtJ6aZvU2Rv4Ge7voHExIwHOgZ/OP5/DLiTqLpKc7SZel8959afi1W08mL7i+yK7uK0KadhkUxD4YrWFfSmekmpKUocJUSzUZJqEo/FQ1O0Cb/Nz+zgbJqjzRQ7iqnx1PBi+4tmlIe7CkEQKHYUM69oHhtDG9F0jZSaQjEUyp3l7BzYSaW7kun+6fnfWSgVYlHpIj4151NIgsT2/u00DTSxK7qLpJqk1FlKUknSn+nHb/Mzp2gOm0ObWd+3ngZ/A8uql7ElvIUqdxVre9fmI4r9Nj/xbJy3et6iyF5EQ8B0MulL9WGTbFS6K1F0hbZYG5IgcWzlsRxdfjRxJc6m0CbW961HEiSC9iAXTr+QZdXLUHQFDOhKdPGvHf9idc9qwukwOwd24rK4cFvcdCcnLq9qFU3iw8CgzleHLMhMD0wno2XYEt5Cc7SZgC3AGbVn4LF6cMgOQukQw00wxY5iEkoi388IgtlXJ5Vk/v3MkYM2yUZ/uh8DwzyvYZD7T9M1VF3NO4yUOcs4rfY0REEkkomwJbSFWl8t7bF2kmqSBm8Dy9cvZ9GMRWwf2I7b4ubEqhPzhtKuRBcOi2NI4smASCaCRbSgGioeq4esliWUClHqLC1w5OiKd7G6dzWlzlJkQabUWUqNp4ZQOoRdstOZ6KQ72c2cojl4rB4wYE3vGlZ1rWJR6SKOLDtyXEIwq2XZGt6Ky+Kiwl2Bqqv5OkLpEH6bf5RTT06qMJQKsXNgJ5c/c3me9P9WxyWcPnA0LfX9vLhoM690vFIwrkmCxBTPFEJp00B+YvWJeK1eHm96nApXBTODM/nvzv+yuHQxdtnOKx2vMDs4m/Z4O0eUHUGdtw5ZlPPrvl+s+gWlzlKOqzyOgD2A3+ZHFESOLDuSaDbKlt+uoKG7csLv4LsJnZY+PtdwzTu+DvLb/Ki6WjA32tux/53C/MR0ft5yJa3WLn5d9neub/0qjbZWvjvzduYWzx21TrGK1lEkjUW0mH3hiHKzi2azK7qLWDY27txuJNwWN9WealJqinA6PC45JhoCfzCOI1L7dH7byh1H8zebSVYfUXYEyaZ+5iWnkZDStFm7qcmWYTFkDODy7o+gY/BGxRZ65qapqa/HKTvpTnZz0xs35a/HITsocZTgsXpoibbgTtixGDJnRo7lA+FTUASFTc6dOHQ7PtWNiMCAFCcsD1CbqUQRVWoyZaTEDH2WCAkxicWwkBGyWAwZ2ZCoypbypnsTHs1NRsiSlFIogsrUTBV16SpkJMLSAJ+Y/r18P3ZqzalEMhF6kj0cW3ksX1zwRZwWJ5Ig8VzLc2wMbUQURE6tORW31c1/d/4XQRCQBImZwZls79+Ooik4ZAdxJU6Fq4Lz6s9jV2wXy1uWk1JTLC5bzNlTzx556992vFvIkUmdjcMYJSUl3HHHHWPus1jG9+gfic2bN/Poo4/yoQ996EA17aDgpJNO4stf/jIAqqrS09PDH/7wBy655BIef/xxiorG1/W74447+PWvf83nPvc5vvSlL2GxWNiwYQP33HMPL774Ivfff/9e3dN3K8wJuAlpgprMI0sJg1vcDg+RbGyUdMYkJnEwcW79uWNur/HUAKZ30F/P/SvhdJindj3FopJF1HhreL3zdURBNOW+MgMoukIoZZIEsijzZPOTbOvflq8v55E0HB6rB8MwiCvxMRcETtmJVbIiizJ+m59aby1vdL2BoivMCMwgoSToTHRiES15L6EGfwMt0RaOqTyGoD2IXbLzcsfLlDnL6Ep05YmMqd6p1PnqeLP7TUKpEKox5CWe0TJsDm9mc3gzAC+1vzSqbat7Vo/aZpNso4yDuQlzuasc3dDpSfaQ1bMFk3SraAWDvME3o2XGNVZZRAuSIJHRMhgY2CU7mqGh6AoCgikRN1ivgZG/L7nFj1W0oqMXeMXnIkhyC6L2uGlA3xHZMWYbRsq95KDpWp4QyeqmvNnwMHNRECl3ltMUbSq4PpfFhSSYIfsuq4v2WPu4C0SrZKXSVUlbvA1VVwsWcaXOUmYEZrCmZ82oBY5FtOQN4YtKFtEaayWUDuG1epFFmYyWyZfJtS1oD9Lgb2BD3waSahJREPHb/CSVJOWuctpibTgtToodxXQluvK/mYOJXNsFDI7QjuND7Z+G9iyQJT6YBEl0WVh0/FKEEV69H5r+IR7e/jBP73qa13c+yT/bu7jdMHjY4yIiSsMIuJFG9SGCIUe26EYU+ro5clgZq9FFWhC4z+dBEwQ+MO0DfHnRl6l0jzaILK1YOmqbpml4+7wsqh492Z/qmzqq/OULLx/vNgGmkfeNrjfoSfYwIzCDmcGZbA1vZUvYlLI4qvyoffJ8bo+381rHa8wumo2ma1gkC7OCox1WpvmHPOrGy/Ny4fQLC75fMveScc97Wu1po7a1RltZ1b2KhaULqfcNJR7XdI03u99EEiVmB2fzWudrlDpLyWgZJEFiQckCXmp/iUgmwrEVxxJKh0yia5hxs8xZxoKSBeyI7GBGYAaqrvJC2wt4rV4skgW3xU17vB2nxUn3U/eO2+5cXzTyXo+Vc+So+DwAflFxL8/4XwPgnKnnkNbSVLgqeKzxMeJKnDNqz+Do8qMJpUOs7FpJqbOUOl8d0UwUj9WDbuj0pfoocZZQ4igBzPFoTtEc+lJ99CR7UHUVt8VNNBtFFETSWppSRyn/a/4fT+16ioyWwW/z84uTfpGPZpwemE4kE0E3dFZ2rmR6YDoCAvX+elqiLcwtnsv2/u0E7AFERJKqKWHUk+xhQckCNvRtIJaNcWzlsXlDjSzIiKJIY6SRcmc53cluREEklA5R6iilzFVmeql66+iIdxBpiVBRXwECTPFMYWXXSrJaFlmUsUpW5hfPpz3ezrziefmoou5EN1WeKs78x5l5Y9KLF72IoitEs9GCdxVMSdKB7ABdf/h9fs6rGzCSntdE+GLq4jwxYngkLHYrUsCOHssiui3oCQXJZ0MuciAX27GUOvlR/bJx35dTppwyatuy6vHLj4c99Q97gqIpBcbfk2tO3q/6RqIr0UWZs2yvnIrOn3b+mNtzEnU13hq+tuRrgLm+2Tmwkyp3FXbZzrb+bazpWcOs4Cz+u/O/9KX6aI42s61/GxbRgmEY+XnScMeHLeEtAGwYRqKDabB/YOsDE7/gA4ifvP6TPRc6cI73hywsogUDo2C+t7R8KRdOv5Cliflk/9pCXbicY486f8Lv2QcaPpD/fPGMi83fscvCFYuu2O1xu/uNWhMSDT3mPECwFo7tuqHnJZkFJjIeG+iGMWI8MRhanedFsEfs2z0pMVzOOvfZAHL/Fwb/AyFftaZrSII5r9cNA93QAAFpkFgUEDEEHXlJgH8s+wflrvI8EVrtqcYpO/OR393JbjJahv50P0k1yVTvVLJa1iSBB52wGvwNpLU0bbG2fIR6Uk3mDdPdiW6q3FWs6V2Dy+JiVnAWmq6xqnsVXYkupgemMz0wned2PcfWxq185OiPEFfjZPUsvclerJKVhJLAbXEzkB0gno0jCRIWyYJVtOYVBQayAxxZdiThdBi7ZGd7/3ZC6RC13loGMgP5dVKRo4hyVzmbQpsI2oIUO4vZ2LexwOAPZvS6J2yDFqjSyvhKw+XQCoZVIK7Eeb3zdQQELp1/KeWucnw2H6dOOTUvN/bbdb9lfd96FN3ssy2iJb/OyupZ1vauLXjWXquXD8/48Lj2Gq/Vy7GVx+YJ1KyW5dmWZ+lN9rK8dTnN0WZOm3Ia0UyUt3rewij0y2FB6WL+NrCery/5OpfOv5TXO1/n+dbnmeWr40vlSylxmhEuoWQfmf9o2BpVju6cjdAnUrpwMZYSU279+OwRGO0pAhY/hmaYSe81g2R2AL2x8KSyIXFEYk7BtgqlZNS1uXUn7oxznF8BnBBbMu4+gLfcm3FYHHitXrqT3TzX+lx+39+3/Z2/b/v7mMfdt+m+3dY7HD97o9AJ7dHGRzljyhnvWinItxuT5MhhDKvVyqJFiw52Mw4ZBIPBUfdj/vz5nH766Tz55JN84hOfGPO4bDbL7373Oz7/+c9z5ZVX5rcfd9xx1NfXc8UVV/DMM89wzjnnvJ3NPyxgGMMTsu+h7BifTJizJK/Tx0y3ZzIXwiQOO0iiRImzhE/MHupTzqg9Y7fHfGHBFzAMg6d2PYWiKwRsAaLZKLOCs1jZuZLZRbNZULIAIO+Fl9JSeWmnjJrBb/fvVTtj2Rgeqycvy7U30HSNWDZGXInTGmulNdaa1wd3WVwYGCSUBJIgYZWsecNZsaMYu2xnaflSnIP5gbJali3hLfSl+nBZXHnvOEEQaBpoMr0tvXUMZAaIZWNUuioLPDwyWobmAdNbNpwOY5EsNA80U+erI2APACaBIyJikSzohk4oFcJlceGQHeyI7MhrA7dEW2gINBBKhQjYAxTZi/L6zBktkz+/IAhE0hGebH4yLzWTVJMUD3RSp+kEJCeV7kpSSoqgI5gnfYfDwNQoFgSBcDpMiaNkTA3eHHHSGmvFb/MTsAX26lkVnLN0Frd3rsAu2/nC/C/scYGfezcUXaE32Wt6eA7rk5NKkudan+PEqhMLkoMmlaR5PxzFeUlORVOQBkmDQwXDjddTjYb8Z/vsIJLPhpHRcMwvRpBHt3l+8XwWly5mdc9qPpkRKNdMgumzA3uWi9gbnOapwzf9HOo8U6HxxQkfJ+g6wbZdCOK2PQ/IE4AMHJv7klgHbeuYDcwGSDZDT/M+1VsFfAigbZiMTevacUrvAYYGnWshG4eqI8Ey/qIVgFgHrPs7lMyE0A5qEn3U+KpgoB2GeUxKwHD66dQxqhpuyioBxotHHh5/feGIfblldCodZ62jjj9XXjBqhpQj9PaYc8QAn+ZGcMj88As3cZ08pIGew2fmfoZt/ds4sfrE/O/yy3x5dKM1BdY9BJVLoHR2wa5ab+04V2ripJqT+OmJP6U93o7f5sdlcQGjDdPHVR5X8D1HkO3OQabcVT7mZyBPbs1n/qjjcnXXeetY07OGRaVDBOJYBvNqz1BEu8viot5v1n1C1Qk81vgYQH78LXGOYUCxunFanLQICvmcIyMobUMQuJTPMKPHvF75slrK66aMe+2HG/Z2jrG3GPn8DzQEQSggvWYEZjAjMAMgPzcbCd3QCafDPNH0BLOCs7BKVjb0bcBj9dAeb8ciWpgRmEG9r54VbWaeiJRi5liY6p2KTbbl6+lJ9uCQhyI0VF0llo3ht/kRBIG+VB+hVIjVPaspdhQzv3h+3rApCmZuqVwdXpuXtb1rccgONoc2F0QETvFMGRUpl5s7ZbQMTtlJVssWOMhMBGM5w4AZJd0aa8Vv9+ej+HKwiBZKHCUFUnxO2cnSiqWs6lo1KrpxJHw2H/FsfMLe7SO95Bv8Ddx+6u04LU70rEaH1IoWydD+g1cQ7RIIIHmsIArIATvWajepDSG0aBZD1TB0TBZUFLBWuVG6k/kcJqJThmFOhLapXgIfnI64GznqHFJre8AAa52X0ssWTuja3k0Y6VYzPA9prh/O9Qd7Gp9coouZwZmA2U+D6eQGMMVr9r9HlB2RLy+JEkdXFCpRnD7ldIrDxZQ6S6mQKgYbNfHrySEXjZU773gY7riR64NGQg2l6GIVoiZQsdx8p6aVNjArOIuuRBffPOKbo5xJwCStT6o+ib5UHyk1RaW7ElmUSSgJ/t3477y8ZrmrnFnBWRQ7ivPrtonCKlk5p860mX167qdH7V9939cKvh9ZfhTLz/osRXbzph5dcfSoZ1Dvq6feV4/xWZ3Eqm7ir7Sj9qQYeLyJok/PQQunER/oAd0gOsxBaSQEi4j//GlIARtKRwIpYEO0yyCANpBFT2SRixxo8SyCLCJYJQSrhJ5UEK0SWkIBVQdJRLCIaAMZJJ8NdAM9o2GkVeQiB8gCaSXNvBIbr1V/CwGB/zb9l6aBJnxWH+Wucq5//fqCSJ5KVyWn155OZ6KT5a3LUXWVI8qOyMvPre5ZTYWrgiPKjqA72c3yluX5ftpn87GkdAkN/gYWly6eJEb2A+9Nq6RhgDJaN++gwOJkn7IyTxAbNmzgoosu4vzzz8/LKoVCIc477zxmzZrFZZddxiWXmJ53n/70p1m6dCn33Xcfn/rUpygrKyObzfLCCy+wePFiLr/88jEljD71qU8BcN99Jsup6zp33303Dz30EP39/Rx//PFjynVt27aNX/ziF7zxxhsAHHvssXz3u9+lpsYcFnOSSddeey2/+c1vGBgY4Pbbb+f444+f8PX7fL49lonH46TTaXR9tGfySSedxJVXXplvU678rbfeylNPPUUsFmPatGl89atf5eSTT86XURSFn//85zz66KMkEgmWLFnCD3/4Q2przUH8u9/9Ll1dXZx//vn89re/pb29nWnTpvGtb32LZcuGluAdHR3cfPPNvPTSS2QyGRYtWsR3vvMd5syZs1ftGY5//OMffP/73+erX/0qV1xherZM9Fn8/vd/4La7fs2WDWvxeDxceMEFfPOb30SSJNra2jjttNO44YYb+OAHP5jnRG7+yQ2sW7OWOx57AoCvXfE1GirrqGqYwv333080GmXp0qXccMMNrFixgl//+tf09fWxcOFCfvKTnxTIsD3wwAP88Y9/pLOzkwULFoyKdnr44Yf5/ve/z7XXXssvf/lLFEXhb3/7Gw0NDTzzzDPcddddbN++Ha/XyznnnMM3v/lNnM5C48rzzz/PLbfcQlNTE5WVlXzpS1/iggsuyNc/ljzcSOmymTNncs0117Bp0yaefvppFEVh2bJl/OAHPyiIYPr973/P3/72N3p7e5kzZw5f/OIX+dKXvnRAZMImcehBEATOmnrWqO11vrqC71bJSoW7omDb7pL9jQeP1QPsm9FCEiX8dj9+u59qTzXHDplN9xpWyTqucWH4tefOp2mFC12bZMsvbnKJBEd6lo+MyBhuwJoemJ7/nFtQDT9eQMAm2bBJtnwOgVx7Lp51sfll1R/hP98b1f7dmWYFwDWBcjmF8rrdlJkoBOBrAOXz4ZU9ex5Zhv07loCDEzDVf382anvumnIzmIK3TBDgqEthyeiF0TsJfZAcETCo0Mw+u+izc3HMHDsyYTgEQeD3Z/6eZLwT3+9M8lNb9HEkUYYx5gt7je1PQaKHRS2roWV05NWeIDL4zuz9oe8OrP7LxMv2DJNBjHWMX+4dggP4fPvDY5Ij+YTswh7IkcG1g2NWEL/TP+Z5KgQrFY2vQes6qFjImBo1sS5Y+VtoWwlWN5x9AwQm2BupGdj1Mkw7lari6dCxDvYjN80Bh67j7tsBzfE9E4h2H2RisOJGmHMBFM/ge6UnUhdu48TqZdC0G+IyWI/oqyIr6gi5J2qYy80cDASO7zTXQs5FJQTfRcTIexWiIFLsKOZTcz6V37awZGxj9nBnmv2Bbuh7ZajUdA0dHRERRVewy3YMw0A3dHRDJ6Nm2Lphq+lcKJhzP1VX83mZHLIDVVcLiAWn7EQztLznvCRIeZmXkeSH0+LMO1DEsjGyWtZ02hn0WrdJNmLZWP44p8WZjyRO7sZWIwoiboubrJ41IxJEibRqSvm5LC6SapKRyvGxbCz/zN7sfpN5xfPyTjyiVcJW5yOzIwKqjh4326PHzOtW2uKk1veN257MjkjBdz1Z2A+mNoRQelMEPjANudSJnlTJNA0gB+xIARuCLKInVYyMSuLNHvNeLC4d93yTeG9DGEPW3Oly89B5D2GEGxGjnUNjltUJ5QtBMo8RBGEUye+yuIbWO28zhBFOsYIojsqLOO6xsoj7mApsdV66b1tNenOY7l+8iRbNgG4glzmxTfUiSKJJTkoCgizimFs0GGFiIFjMuZV9+r47ok0ETiA4jOobnkMO4JjKY8ioGbw2L2k1jdfqzTu3qLpKVsvm+6fxkMuBuKdyk5g43nvkiGHAH86C1r3XDH9bUHMMfO7JfSZIVHXsRYgkSQiCwLx58/jCF77A3XffzQc+8AGOPfZYrrnmGnRd58Ybb8TlcnHNNddw3XXXcc011xQYZJ944gne//73c/fdd49JHIyH2267jfvvv58vfelLLFy4kCeeeIJf/OIXBWWampq4+OKLqa+v52c/+xmqqnL33XfzsY99jEcffbTAgHzHHXfw/e9/n3Q6zeLFi0eeLg/DMPL3Q9d1+vr6uO222yguLt5t1EcwGGThwoX8/ve/p6enhzPOOIMlS5YQDAaxWCxcfvlQqLmmaXzuc5+jubmZr33ta9TX1/Ovf/2LK664gj//+c8ceaQpmvH4449zwgkncOONN9LX18cNN9zAlVdeycMPP5yva8OGDfT09PC1r30Nt9vNbbfdxle/+lVeeOEFfD4f4XCYiy++GIfDwQ9+8AMcDgd//vOf+cQnPsE//vEPpk2bNuH25PD444/zgx/8gC9/+ct5YmRvnsXVV3+b08//EBd87BJ2rV/JPffcQ01NDRdfPHpAHRmoOxxPLH+Kub1zuf766+nq6uK6667jk5/8JDabje985zukUqn8e/nb3/4WgL/85S/8+Mc/5pJLLmHZsmW8+uqr/OAHPxhVt6Zp/OEPf+D666+nv7+fadOm8e9//5urrrqK888/n2984xu0t7dz6623smPHDv74xz8WeFpec801fOMb36C0tJTf/e53fPe732XWrFkTylszHLfeeitnnHEGt9xyC62trdxwww1IkpT/Ldx5553ceeedfP7zn+eYY47hxRdf5Bvf+MZenWMS7xIYBigpsDjMscAwxh4TdB3UFFhdo/eNVZ91PyZLmbjZnpz3STYBks38nugFZ5HpvZ0YXCy6ik3jUu78+0P6qxmsiXbQ5oCmm+d2lRTWqeuQiYLDP/p4XTcNYcmw6QjhqzY/2/2Q6gdXkfmvzWtej65BZJfZbv8USEUG6w5A90Z4/kbYNUxKLFhvljvUsuOqaWh51fzctedks287Nj5y0MkRgyFZrTLdJBytFRPIIda7Fd66F0v5Anxb/g3xLlK2MtbrJzDtyGW4AgFsThdKOoWSyaBmM4CA3e1BU7JkkqZhSbJY8RQVkxyIoCkKss2Gw+NF11QysSiOlbcgbPkPeKtgjES+u782g1g0hsfrGTNy6d0HA0I7zeiRhtMgMjpfyyhYnGY/JcpQPB1mvc/87Xsq99yPvl1IR+CP5zAt2YpjjLwcEyVHJMN85vbZuyH6HvwktLwy8bZl4/DYVydePoeXbtn7Y94BSMBMgFf38sCmFwBwA5cCvPXI7svbfPC11ajSsJhpwyiU1RJEJESsUzwEL967+eQkDhC61sN9F0JwGsy9AKadZs4P9meu9A5jbyMzJVHKy7jlvIlzGvYSEuIwaabcflmUCyJFx0oybxl0h8g54Yz8PBy5Ndbw/cPrHOs4i2gpaMN4sEk2cul7hjvJDHd0yWF4fcdWjnb4CV40k+TaXkSHjKXSjdqbJL0phBSwo0Wz6CkV21QvtjofglU056SCSYQonXEkvx17gx89o6HHs/mFsNKbInz/ZtTuJL2/neDcTBJwzpuYwXgS7z2IttGRAZLRivDQ9Qib/z36gNK5cNkLeYLkYEIcIQMrCHvfJkuZC/8HphF5ZAdq32CeJrtE8KMzsVZNPE/wwYTX6s17x41UFZBFeULKKpIo4RQPn/HrcMDB/4UcFLw7FpHt7e3MnTt2Iuurr76az3/+8wBcccUVPPfcc1x77bV88Ytf5JlnnuG2226jrMz0wG1oaMj/m/sMZt6Sa6+9FqvV/OW+/vqeCaVoNMr999/PZz/7Wb7yla8AcOKJJ9LT08OLLw55Xd1xxx04HA7+9Kc/5ZOdH3vssZx++uncc889fOc738mX/fjHP87ZZ+85sdAjjzzCI488UrBNEARuuukmgsHde4n+6le/4uqrr87XIQgC06dP54wzzuCSSy7JR6C88MILrF27ljvvvJPTTz8dgGOOOYbW1lZee+21PBlRVlbGXXfdlc9TsmvXLu6++27i8Xj+emOxGA8//DBTppjeY06nk09+8pO89tprnHXWWfz5z38mEolw//33U1VlhuEvW7aMc889l9tuu41f/epXE24PwPLly7n66qv54he/yNe+NhTSuDfP4oMf/DBnXvw5BOAT55/Os88+y/PPPz8mOVKAEa6Rqqpyxx135O/rU089xYsvvsgzzzyTj1ZZs2YNjz76qHm4YXDXXXdx7rnn8v/+3/8D4IQTTiAej/PAA6M1fC+//PJ85IxhGNx8882ceOKJ3HzzzfkyU6dO5TOf+QwrVqwoiLL5yU9+ko/emTJlCmeccQYrV67ca3JkxowZ3HDDDfnv69at48knnwQglUrxu9/9jk984hNcddVV+etJpVI8+OCDe3We/ULTC6ZhuHIRCNKgIfzd0T/uEwQRLO9gkvu+Haah74mrYefzUDrHNNj3bYXpZxYa7zQFGp8ziYlZ7wPRYn63uWHqiZCLDtEU2P40JHqgbhkEpoKhm3/JMAy0mh5EomTK04gSOIImiRCYan5PR2Hbk+AuhWmnQrQTGp8F2WGSEQOtpjFITUNORkEQzfNFWiHaDvWngLcCkiHo3gTFM8BTZrav5VXzvas9DuQRMlPpAcStTzBfScJzhbvw1Zjez84iaFwOAy1QfZQZJZFDqh+2/c8kL/q2m/fX6oHhyQHdZeZ9DEyFKceZxEd/89B1jJXfRJDMOo+6FI694m2N+NwvbPyX+fxnnHXwjL9g3sfq0RGj7zSMYc9SNCRElwXRM04klZoxibGND8OT32PkwPXo1jLa19zPiw/eD4KA1W4nm9pzok53sIh4OJT/7i+rIBHpR8mkqVt8JOd/cyUW62i5tT1B1zS2H6LJD99W7C/5erBhGGQkOzYtzaxEEwaFvxNNH4ccGezjjVzkyOBYbakY53e+4qYhYqRo+hDRPRKCZJJN8z4ET30f2laZfd1E7rFhmOMVmL95X41Jqh8iMAyDdDqN3W7fvcRgJmaOWzlYnIME+AQQ7YTMANxzGg1ZD0sGprDCf0xBjj4wZbUA7DPeRm9VJT32+HWgEOs05x1F08x/01GT1LXYoeEM8x5GWmDG2eCpgK2Pm+NA/n0SoPlF6FhtbutYbY7HktV8f6afYUbUDbSb75EzaDp6CBLYPOY47Swy5x6ZnHSKYL6/ump6So8h7QSYzh2v321+TvRC62tD++qWQVHD2Me9yyEYBlP6+hA6ig/vfnWiMAwYaANfFbzvloJ+UfJY8ZwwJPNnrXDhXDBaRm8s2GqHyBhJFpFcQ/MMS7kLpbOG2Io2JI8VLZIBUcBS6cLI6qZxVwTRYTHlvEQB15KyCUlwTeK9ibGkYJ19v4X+wUTnvpqhNUDvFjN6tm2lue46yBAkGVEbGh0FPQvZMaLEJMvQ2nYMuI+uwDbVi9KRQHRbsNZ4TImswwUrboKmFabzn9134PrfKcfBgo8cmLregziM3qADBEEwIzXeBbJaJSUl3H333WPuq6gYkmSxWCz87Gc/4yMf+Qj/93//x4UXXjghsqG+vj5PjEwUa9euRVVVTjmlMEHgOeecU0COvPbaayxduhS73Z6P9nC73Rx55JG88kqhl9vs2YX6x+PhlFNOyUdDGIZBOBzmiSee4KqrriKVSvHRj3503GPLy8u599572bFjBy+88AKvv/46b7zxBnfeeScPPfQQf/nLX5g6dSpvvvkmFouFU08dUqQWRXGUgX7BggUFCdxzEkzRaDRPQASDwTwxkmsDmIZzgFdffZXZs2dTVlaWv0eiKLJs2TIee8zUQJ5oezZu3Mjjjz9OaWkpX//61wv27c2zmD/fDBfPvbHl5eUkk2P/lsZL5SYA9VPrTGJE10AQKC4uJhAIFMiX+f1+YjHToLlz505CodCY79VY5Mjwd2bnzp10dXVx2WWXFURaHXXUUbjdbl5++eUCcmQ4oTT8ue0tRua/KS8vzz/bdevWkU6nR/0OzzvvvHeOHNn1Kvx57GSV72kc+xU46/q3r/7tTwOCKe0y0kO3Z9PQ53W7eQ82/mvoc7JvyLA/EjufH3v7eFEFu14u/B5pgTf/NPQ9GxsiGTID5r8Wp+mVnYkWnm/bE4V1hbaPPl9/05jNGHdEHGg1/4aj7Q3zbyR6txS2ezji3ea/4Z3mXw6yw4zMgSFCRbbD4k+RXPA51q1cx4b7nyZ2x4WUTWugetZcLHb7fnnty1armb5S1xFEES2b3eMx48EwDGLhPnatS+Nf/zoV02chW8Yfw0VZNg14I2TMcpAsFnxl5Qz0dCNJEg6fH0+giL62FrKpJKIo4goEifb1YnM4SSfiQzoygkCDNUxxzdiepO8UhkeOgIB9dnBsI2nTi/DPzw+9GwDOYgwMNMHCG73ltKfMa7E6nGRTySFiRBCQrVbQDVQlO1jGAQhkU8k8MSJZLGiKQqS7c+i0q1fRuW0rU+aNLTs3iTFwuBvwBIGotYiSVDtzEjtGSb/kI0dGkBnyoHelPtjf5EwjknfYbzwVMQnovq2wfDAJ85wL4KN/nljbLvl3vo0ThmGY/ai36p11LpgAdE1j00QIxGgH3DrXJBamngif+c/ET7L6r/Dol6G/iVrghm0hjls6SI4UPNrB5+bZu3UVhgGda2DHMyN0ugqoF3P8Hox4OSh45fahz8/fMH653eHp0dHgE8LK3+zbcbLdJFqaXji49+4gQsTMocSug9yQgwFfNSz79jtzqjOn4j2j1kxGnh5c0w8acg3dQBAP83FtEgcdcmJwbXfK/8FJVw/t+OcXYP1D8MdzYPEnwV9rOsdtfwqKZ5rOVPIEx25Xkem81/yyuUacCHTVXHfaveAqZua2+5ivDF+XfQCeGOM42QEffxDqTxq3akuZC0vZQXQE21fEe4fmaAca6/4O8z44vkPMJHaL9x45Auak/2B6VB4gWK1W5s8fnYhwLMyePZuZM2eyYcOGUQbm8eBy7f09ikQigGnYHo6SkpJR5R5//HEef/zxUXWMjPIYmRNiPPj9/lH34+STT6anp4ebbrqJD33oQ3v0rsxFz3zuc59DURQefvhhrrvuOm655RZ+9atfEYlE8Pv9iHvQLh7Z5lz54fJkDkehd13OYJMrE4lE2LVr17jRQalUasLt2bZtGyeffDLPP/88f/3rX/N5YnLnmeizsNrMwTNnDBRFcdTCfiJwO62mN7mWMQ2rSgqn3WZ6meeQIzCTIQa6TYNowCkVlCnJeQBnoub2rJm8zylm8+Ui3ebgfe2113LttdeOaktPZ5tZdvB8TlKQNI1eubtqZBMF9ZOOQHLY89M10+t4WNscslHwXdTS5r1KhhjoNT0Ug06xoEyRSyq8nrGQyZrt2PQYCPtoSFVS8MJN+3bsux2v3gHBuolPFPcGXRuGPBeHI1AHF/7aNOgnQ5iZ4ZTR5Upng5aFUKP53VdlkhN9I4gHRwBqjjKjKwYJSATB/K0J4pCHjjNoasynI6Y3Zv8u034jSOYkVkmZZcVBz+JIC8R7zIl090ZwlULVErPu7k0mIeIuM/86VpvGG4sDvJUmgTPoEY3Db3rKhHaMMPCY0CoWsS7iYqGwBbFqsXmeLf81Iz16N4OaNT1Xa5bC1ifNaJHhsDrNesvmQMVi03va5jUTMPtrTRLHV21eS6LXvF9zLjD/DTeabfZVQyJEqLePtu07efX6G0n0DyXO69y2hc5tWzhUEe3toWXDuoPahpb1a/joD/fRSHaAoOe4GgwEBBzzx5CpaHkN/vbRQqeZqiMwPvc09//oO/nn7Ckq4RM/vQWXP0As1Ed/Zwdl9dOwOpz58XugpwsQ8JWW5b93bNtCWX0DwcpqUrEojatex+Z0sfKxf9C1YxvJaORtvAOTOBSRkc05ol+JjZtzZKR8To4sSWPOPSQDVFlDtA0u5db9HR75ktnP5SRrimfAOT+feMP2hXgSBLM/PpzhrYSP/Mkck4747N4du/Bj5vj26h2gJCnLmHM3U1Zr6H4ag5+lvSFHBtrhd6cUkraHAkTZNHiVzDINZbEu2Pxvc1z315jRobuLYLF6zHlW9ZHmOKyrJkHVtR7K5kL9yea8ZOMjUDZvSHK0dI45t+hcB/M/ZJ6vb5s5P7A4YeHF4BxLisiA9jcxWlcipCPmJmcxfOUN+MfnzPrHy8nzLodu6HR1dVFeXr7Xkl2HLbY9Yb5Hz10Pr95pvs/Tz4Rzb35bJdZy84SR3u2TxMgk9hciYcRsl7nGO+bLhTvnXmCSIzA6V1v3evPvHcSER0A1ZRI4uyFHDmn0bIY1f4OZ55hpFNb/3ZRlrjkqL9WcdZQTqjoHUUlidTrwlZSjZNIkImGUdAbJIhOoqEaURGKhPgRBwB0sQs0qRHu7TKc6RcEwDNzBIpwen1n/JDGyz3hvkiPvQTz44INs2LCBWbNmcf3113Psscfi9Y7W49wdRhrvc0gkEnkiJRAww8VDoRDTpg0tlnKkSQ4ej4fjjjuOz3529CJElg/sazlv3jxeeeUV+vv7KS4ePWn+85//zN13383y5csLCAuLxcJFF13EihUr2LFjR77dkUgkn1wuh02bNmEYxrhExr7A4/GwdOlSrr766jH3W63WCbfnxBNP5De/+Q1XXnklt9xyC6effno+umhvnkWOCBlrGpc7fy6Zcm7BnxohOyIaGoKhDoW+66pplNLVQi+E9KBneqSFgGTqt4dad0BkKG1wpG3QKBzvNo/NEQrRdnCaxmWvYXonXH35J1i6aCiRfQ4+t8s8dtj5RiE9AJEWhKRpHNX7W/P1g/kbIBsf3f4xrkcYaKXGZw5aoV1bqPcP3c1wy9bC6xkLqmHKI718HcQnoLm+O3gq4PKXzHBOXTu0EqkeDPz5fOh4C/77rXfkdNmpp6Geextv/O9/DPzjOeoXH4U7uIBIdxebX1xONp1CEEVc/gDuQBFVOOnc3kKoLUzNnHlUeGci6rJpZMjBMIh0dxHetBI1KzPr+NMJtbfgLS5BsliRLRayqRSaqmCx2U2DQ24tGJgOhkGorYWmF94k1LoLuzvn+X/PsJavGfb5D0MfBQEYigaRZAtVs+bQ2/wmqdhQ9JXF1kPtAiel9SfS+Mbr9DTtQNd1REky+xijhVQ6zVq7HWFYtmnJsnUwwkLAW6IB24j19aKpKsYYebGc/h5ky8vE+nqH+i5BwFtajqF3kEkmyaZTyHInykOr83I9RVU1eIpL2Pbay0R7RxulZp94Ck6fH0PTUPcj0gMg3m/2WbqmoSpZAhVV+xWJEu3rIRbqo2buAgxdw9DHJq+VTJqmtW8RrKymuHps+ZhUPErH1s15WahMIoEoy2iqgre4FKvDQffOHZQ3zEC2WPEUl+TloQRRYNZxB39BUxg5AnJwBOmZTcJDnzbHoGmnwpJLTC/4+R+mcfWqPDFSOXMO51/5XVx+c47jKSrGUzR6PuErLR/1ffg2h8fLvFPM5O6bX36erh3bSEUHDszFTuKwgTEomSUZo6O2comJx9Kb/icf4QXBjBQWDcg4NHPsfuaHhZ77WhYsLtPr0lM2oTZpqkK0r5dMPE46mcDh8VI6tR5BEOjv6sDl8yPbbPS17MIdCDLQ243N6eK5P/4Glz+AkkkjW20c/9FP4CstR1NVpME5pGEYJCL9ODxeQm0tBCqrSEb6cQWK0BQFi81WoEPe39WBJ1iMKEsICAh7SqY+Ako2g2yxomYzY44NYEbrGRiIokQ8HMI56zzEfTEoiCKc+n9w3Ffgxim49DR2LWMmZB9+vsE5suS1gqYCxviSIbpuzkXe/NMQMTL1xNHST8PJLMkG8z8CJTP3/homCkEccj4YSaQNixpEzZgOHlaX+a+hY94Q3TREj5TTzCGXLyyH828v/D5eOTVjOnXsQU//0j+9wQk7buKz8v/g3JtMB5FPP7LHy343w9A0OtesoWzRInivyDOedDU88mVY+7ch55o1fzWdos47NPMnHYpQs1kzahazz01Fo8hWK8mBCK5AEIvVlt9vGAbdO3cgiCKlU+sxdJ2e5p2oSpaKhplIsmz21YJAf2cHSjpFSW0dgihiGPqovjl3PqvdQaithaKaKVhsdvRBhYhULIq3pDChvZJJY+g6Az3dBKtq8uNTMjpANpnEX27aRVLxGL3NOwlUVuEJFhNqayHS0ow6Zw7RSBirw0ksZOZblC0WdF1HUxUyg0oaAmYuGlEU8ZWW09/VQVldA0o2jSRbULNZoj3dOLxelHSadC5HnSRRPn0m/R3tKJkMZfUNdDVuy68z7E4XgiiSipt2DWGRFbnDwOhRsIqm/UAPNtC6dQdl9Q3YB5VKmPU++PRj0P6mGX1vGGbf13CGuW2syP6xYBjmuNS13nTqqzvRHBMmgpxMqKHR2d3Gtpq3MAZJwQVFfyE4d0T09Fv3wv++Vxjdf7jhf//PlJ985VeF24saTEcMYEO7yPK3duR3WR0do+R6bc4QFrs9H4XuDgRJxWNoykgnynaCldXULvRz6rRTmcS+YZIceQ+gvb2dn/3sZ3z4wx/mK1/5Cueeey7XX389P/vZzwAmrFWdk4Pq6urKbxsYGKCxsZEFC8xObfHixdjtdp588kmWLl2aL7d8+fKCupYuXcqOHTuYPXt23gBvGAZXXXUVtbW1E5bSmgjWr1+Pz+fLEzcj0dDQQH9/P/fddx9f/OIXC/ZpmkZrayszZswATMmlP/zhD7zwwgucdNJJ+XZ/73vfo7a2ll/96lej6t9XLF26lH//+9/U1dXl7z2YOTEUReHaa6+dcHtypND3vvc9XnzxRX70ox/xm9/8Jn+eiT4LPW9gHN3eXBu7uwcXcgaoqsLWTVsKFrZCzpvM4jRlGNIDQ3rDtmGEXS45lc3L1GkeKspKePLFVVzw/vPyRZav3DBYl8s8Vh4kt6yefF3102dRFPDT1hPh8wsW54/t6Qtz9XU3c/EF5zKlfkbB+UZBsoHNi9tvJqfvGsgwpd4s19jcSiQaM69hZPvHuB7D5mX6rDl43C6efnUtRy09Ol/kqVfWFF7PWBA0sMSh9njIhscuMxFY3bDsKjORNuxW1/M9g7NvMHMNKCkzcuAAy7cYokyXXkmb80iiHU2seeIVeOJL+f3bXx87cW7v4L8blj+V39a+ZeOEzrn+uaf2XGg3iPfvxzsGdO8ce9LduWPrHo9N7LbeHbvZO4jWsXUieluad3tY34j9VbPmUDN3AUeedyGy1Yokvzt+KyNJ9d1B1zUEQRxVPhWP4cgTaIcecjlHcq0WrSPmO6v+YBof/bVw0V/zXqOGYfDaP28F4IjzLuTkT33+gLfN6TVzbiX3QbZxEoc39EGDgsXQRkWOqINOCmN5cT8sDOV3EwHRth1+cbmZYwrMKLsvPm9KBblKaNy4hRevvxmLzYa3tJxEfwiby42h67j8QQQBMokE8Ug/HVs3jTpfydR6JEmiq3E7VocTh8fDQM/uoxgaV72GO1hMuL2VouopKJk00d6egjKCKGLoOjanCyWTweo0De5WhwO7y0NPcyN2twdNUdB1jYqGmXkj23AYhkHvriasdge+MpOE1BSFzu1bcPj8pOMxRNlCU31DQYS1rml0NW5DEES8pWX0Nu+kpLaOI953AW8+/ihOrw9BHMzqIgiomQyqksXm3E1EvWFwIQIiBgF1AMOoHFNeVhx4C/2PHwVdRahZijBWjpbQ9kIHmfN+CUeaDkzZdGrwXhV6uOu6Rjoex2lzM9DTxdqnnyDS1Ykoy0iyjLekFNlqo2PrJnRNwx0sRs1mSA8aWSwOB4IgoKkq/rJyPEUlFFVPoWXDWtRsBtlqIx2PUTJlKn2tu0gn4lTOmE24vRWb00U6HqO0roHa+QsRRJGB3m42Pv+sSaz5A5RMmUoi0k86HkOSLVgdDjLJZJ5Uc3i8gIGhG6QTcbRBiUKLzU5p3TS6Grdj6BqSbEHTVHRVRbZa8ZVV0N/Rhj5MHlKUJKrnzEfNZIj29eIOBLB7fLStauQWxwd5UDuFJ+d9cPxnOYl3NRQly8uxJcw5/0JKp9RC60p47Cuw6vdmvpwZZx7sJrLttZfYsPzpAnUGXdPo2dWEw+3JR6fuCUomQyzUh7+sfFQibABVyZLoD+P0+U2HqQnAMAwGerqIdHUSqKjCW1JKd1Mj6VjhXMZid1A+bTqCALFwmP6ONgCKp0wlHY8NGXuDJkmeTSWxOJwc7VhDjSvCLt1BLGvBZdGwWgQkWTavwYBMKlnwm+8SzfmpoeumdJmuE3c6TVJFMNucScTzHG5SEmFwTqurqkng2x2IsjxYziAkwIDFhpLNUA90vDwxx7ncE9OBnAbENsXGtlgxMWUvIgcFYczo+pE4veJTuCxmrpG122M899L3cXi8nHLJFyiqqcXucuOpW4ZQfxJKNsP6Z56kaeVb8HpOy8q32/oli5XaBYuonj0PcbGAkIli2L0Ig4SVIIr4SssKCCxd0+jv6qBz2xaa177F9KOPxx0I0vjWSpwWC6plzdBl2txm/szhKDHtboTHll8+5KGpZr8yFgaJEd2A5mw1UxcdAUDX9q2kE3EEQaS4ZgquQJCObZvJJBNkkkMr4uHrcofHS/Hg2BpubyXc0UYqHuPkT1+6b84ek5gkRw5nZLNZ1qxZM+7+mTNnYrfb+b//+z8cDgdXX301Pp+Pb3zjG/z0pz/lrLPO4tRTT8XjMQ0bzz//PD6fb9zE0zNnzqSiooI777wTt9uNIAj85je/KYi2cLlcXHrppdx11104nU6OOeYYVqxYMYoc+fKXv8zFF1/MZZddxsc+9jFsNhsPPvggzzzzzD4TDOFwuOB+pFIpHnnkEV599VW++c1vjksCHX/88Zx33nnccsstbN26lbPOOotgMEhXVxcPPPAAXV1d/PKXvwRMma7Fixfz3e9+l2984xvU1NTw6KOP0tjYyI9//ON9avd4+MxnPsOjjz7KZz7zGT73uc8RCAR4/PHHeeihh/je9763T+0pLS3lyiuv5LrrruM///kP55133l49i7xj2Bjt9fl8LF68mPvuu4/a2lrSFoF/PPBPspkMtmHvSM6DF1+16VVmc5uJFiVLoTSDc1DSq2gaAnDV1d/lW9/6Ft+/9Y+cffbZrFmzhvsfGRzYvZXmse615nf/FCgy84VIwJXfuoprrrkGyRXklFNOIRqNctddd9Hd3c3cY06FopqC842CMwhF0zj69BLs197MjXfdy9e//nUSiQS/+tWvTBk5m2d0+8e6nmA9sj3J5y/9ArfffjuOQDlLly5l5cqV3P8vU9pM9FePL1ORTkNEg/NvA/uhpe992GPKMfDF5XsuNwhNVRAlOT8hz6aHvD0MwyDSZeYW8BQVs/WVF9j88gq6dmwFCokB2WJlyvyFxMKhvIGofslSSuumoSkKfS3NtG/ZSCoeo6JhJjaXm87tW0jHYwWLgxxEScbqcNDb0pxfrAQqqpAtFqKhXgQEAhVV+fwIIyHbbJTVTSMVjVK7cDG+kvEXYQWyeiPm8F07t7Nr3WqKqqcwbclRiIPka6S7i40rnkVJpyiqnsKMY46nY9sWvCWlBCuq0XWNHY2NNDSYBi3z/hqE2luRZJmuxu2o2Qy+0nLKpjVgsdlHSQsahkFfyy76O9uonjMfh9s7qFamEOnuwmK3YbU7ESWR/s4OglU1WGw2UtEBNr24HDWTIVhVTcORx+Qnru82TJQYAcadZB/KxAgMEfrmy6kgNP0HymcOkZ+r7zN3n3AlsViSbKqPTDLJM7+/i97mnUgWC0ed//YY0RyD5Mhk5Mh7EIO/J9lQx5XVyuUYGQ+SAaXKA5AZJB5kB/1H/z92vb4WTclSWjeNx2+/mWzK9GbtapyYd6hss+ErKaO/s53e5iGvTTPPztg55rwlpZQ3zKRrx1aivT2E282o1lDb2BGwuWiO3II/N06l47E8kZKOD2mSt23esNs2JwciBbl8AOKDnr1kMrSsXzPusblr7N3VxJN33brb8+wJ2Vk27EKaoDIwpqyWgUHshVsJaGlzY8vYDhEAqmgnKfqJy6WsfbWXyKPfxmK307VjG6qSpW7REVjsDjRFQVNVenftJNrXi7+sPD/3eLvRuOr1gu+bXhx7/tQLNK95c5/Ps33l+PdpPOx8a3QusjOBAdnLq0u+OPqASbxnsPWVF3nzv4/y5n/hij88gH3Jp0yp2NfvNuXxGk4fO2LpHYKqKDz92zvMPG5jIB2L0t/Zvld1jhUFPRz9nR17Vd/Qce3jtkVJp2jdOFreNeeAZHO6yKZTeZIEoFxt5ciiXH2Jocj2HHID5hic8hitKzxmPG47H8jWP3a5vUwTNR4WBLr2XGif8Fb+U3PCdAROxaI8fscv8tuDldWU1k1j51tvjDuO7w6Nq17b7X5vSRnVs+YgyhYMXWfXurcKjPhbXx3KO3zEMWUwlGKWSGcXD//mdk75zBepX3KUuTEwqIjQ3zx2pOK+4r/fgo41piRkrs6Cuod9zm8fa9setmfj5p/NC2ddj7H+n2yVjyIcjmPR03Q1bqU9pDLzjI9wymfM8SiTTNLTtAN/eWU+Mj2bStK7qxnD0M2IK8Ogt6UZq91BUfWUgujcnuadZBJximtqJ4mR/cAkOXIYo7e3l4suumjc/Y888ghvvfUWr776Kr/85S/NBNjApz71Kf79739zzTXXsGTJEqZPn855553HX//6V1588UX+85+xExFKksSvfvUrfvrTn/LNb36T4uJiLrnkEnbu3ElT0xCz+7nPfQ6fz8e9997Ln//8ZxYvXsx3vvMdfvSjH+XLzJo1i7/+9a/ceuutXH311RiGwYwZM7jzzjs57bTT9ul+rFixghUrVuS/O51O6urq+OEPf8jHP/7x3R570003sXTpUh577DG+//3vk0wmCQaDHH/88dxwww35ROGSJPG73/2Om2++mdtuu41UKsXMmTP5wx/+kI+eOVAoKyvjgQce4Be/+AU/+tGPyGQyTJ06leuvv54Pf/jD+9yej33sYzzyyCNcf/31HH/88Xv1LPKRI+PIvtx44438+Mc/5vvf/z4Op5OTP3Ahi+bN4vF/Pz54nEF+liLt3WzjvPPOQxRF7rrrLh599FFmzJjBddddxze/+c09HvuRj3wEl8vFPffcw4MPPojT6WTJkiXcfPPNBUng9wSv18vtt9/OL37xC6644gqqqqr4yle+wiOPPLJX1wJw2WWXAabk3e9//3sWLlzIVVddxQ033DDhPDuTOPBoXreage5OFpx+zpjGY8Mw2LHyVdY89R9aN27AYrdR3jCTnuado7ymxoOnuISK6bOYe9KpBCqqsNjsuAPBcctPO2LpuPv2hM7tW3EXFeEJFufbvzdG8f1B7YJFHH3BR0ZtnzJvIQtOO6tg2/Slx+U/a5pGWDPLDSe1axcsAmDhGedM7PzzF+19o4FZxx98OahJHBgMjVnglv6D+MjvR5UxJCtPPN/M5ps+U7BdlGRO/Nhn8lJaBxqmp/QkOfJeRE5WSx5LVmuQOBgZOWKMyOFQnO7HlzFlBwfe/1deW9nIhjv/Oao+u9vDEed+wJTSW/MmxTW11MxdQH9XB8lIP5LFgpLJ4C0u4ZgPXowkywiiSCLST9vmjajZDEVVNaiqQrSnm6LqKei6RunUelNiK1iU/40oWZOIyKZSlEyZSn93J/bB6JBARSWZZBKby0W4vQ27y028P4zN4UDTVGSrjUwijqYoeEpKiYf6cHh9ODxeeneN7z1qtTvQdc2UmBDM2and7SGTSiLJFnbs2E51RcUoAt3l82N3e+jv7iSbTPLMPXfm953ymcuwOZ0mwaGpNK56HTWbYeaxy5AsY0cOKuk0ysvr8uQIoxKyg9X6GIFuk0B4qaeWmDq2vFRWk2hOBFCN3Pg3Oln4jjfGNlbliJHyadMpqqmlr2UXuqpgdTrRNY2qWXMpqqoh0t2Jruv4SkrZ8vILSBYLDUcdi6Zk6WrcjpJJ07urGdliIZNKUlpbR7B6CtHeblz+AIau07z2rbz8i8sfoKdpJ7GQGetqsTsQRRGHx0vd4iMJd7QhW61UNMxE01T6O9vpa9mFw+2hbNp0trz0PN7SctyBIDaXi2Cl6eDUu6uJUFsL5dOm4y0tI9bXRzzcR/m0GaQTceLhEP7ySmzD5s3dTTtoemuV+V4MyuwITi/x9mZ8atSUnZvEexbD88c9+MPvMPPYE5k15/34X78bmlbAdQFY9Ek4/5cTj6qPDhKS3or9bl/TmlWkE3FcgSAnfuySgn1WpxMtm82/13uCIAhYnS4y4xAtoizj9PpIDkTGdLYaDxa7ndKp0+jasRVNVQeley1IFivuQBA1myUVHSDc2Y4gioiiSFldA5qq0NW4HYvNRt2iI8kkE7RuWo+3uBQEAf9L/wetYHgqGJjzWYT+JkRXEaotSDoRR9c0BEC0WJAtFmSrDW9JKX0tzWiahiRJaJqGze4wySXDyHMjVocTQ9fwl1cS6epAttlR0ilsTicOr49wW+tgVF0R3pJSor09JKMDODxe+pNJ/A4HDrfHvB5JwtB1dF3D7nSPeY80TSXa24MnWER63WM4+laDYUa2iKKYb1duRWYYZgRgbqwyZYSHIrZN2eHCHDW6btLuQrofzV7M0m/8ivOmz+e1fz3IjpWvkk7EScfjhDvaCA9G7rgCQeafehb+skIZ2PGQiPSzccWzpKIDZpsNI+8ta2CQSSSI9nazaQQBJ1us+MrKcfr8xMN9aKpmRqGPkAx95Z/3Ewn38tRvb+fyX99rbvRPMaW41JSZj8pXteeG5to1fKzXdWh6Hl68xZQQy+UWbF81oWvfXzT323jl4bUYej1djcPHbAmb08uCYetZm9NJzQh5MavDSdWsQkn46lmjJfwFQaCs7jDP/XaIQDD2JZvyIQBN01izZg2LFi0aNyIgnU7T1NREXV0d9kkP63cEhmGQTCZxOp3vmAFuEu8MDMOgsydBn6LiEEWmV+4+Z83OaB8xnPjTIbKynaTswqZnmZVswkBEqFhwwGWLDgcYhkE0GuW5557jmGOOyed+AfjrX//KT37yE15//fVxcwJN9mtvH9LxOL/50iWoWTMfjsU+2jXJGMwNMSHk3m/DwFNUwuwTT8ZbXMqC087aaw319xImMr5PYhITwfPdO7l4U5QKo4OXl9+MUxrtvbwuMY2nW8xcVrLNhprNMuPo4zn1s5cVECMDSYXlW3tIZiduQNgdMttWEf7PH6ieM4+LfnjjXh8/+Ts5fNF0+6nUhd7k19Uf5SOfvpsi65Cx4KQHTyKcDvPP9/+TGYEZ+e26rlC5YkhO8Y87fsQ57cvpEWq4b9PU/HZRkvJGrpIpUznnK9+ipHZYXqr3EPbmN/L8vffQsmEti858HwtOP3uvz2XoOqEfzaJY7Oay2ddw+ZEf4tK2Jtp1sw85Mraef7/1NQR0kqrM6/XXsenll/LRM8Nzo4iSROXM2dTOW2Qax0QRT1ExyYEIFruDoqoaOrZvQRBMqRnJYjEjKCWJWF8vdUuOoqhq4o4/7wX8e207W356OSIG/537BZ675gMHu0mHBN6L48iLf/sTKx/9R8E2yWLh8gXbsSeHcjl21n+avprzERCwJVoI9LyAr+sFRG24571Ayj8LR2QLopYh462nc/H3UJ2V2Fwu4v1h1ExmeHF8peVkBo3WBRAEpi5YzPP33kPnW8s55qQlLPzCje+dtbJhwM3TIdELn3kcph5/sFsEHCa/kVi3GZGak8oehnQizpaXXyCTiOMvr2DG0ccf0DVoOh5nxxuvkorH8jlfApVV1C9ZijyGM8G2/9xCq3PIGaHzoRPp7u9DlCSu/NujQwV/cxJ0rjE/27xmziq7Hz75DygdIb+vZuGPZ5sEiMU5lAtF10yCZThEC5xiKrBQEF0++mP+S4G5fDfbBrcbhsGqp//HmjYbUcW01YiSxPxTz8JTVIwoy8w4+rhReQoPZxzKv5O9adtk5MgkJjGJ3cIwDNJxhWQ0S1bTQARJmvhEbXhJqz5oVJas753J3hiQZZl77rmHe++9ly996UsEAgG2bdvGL3/5Sy644IJxiZFJvL3YuOKZPDECZkj4WLDY7Cw++zzmnXIGyYEBQu0tBCurKatvyGuwAoiSOTnTNd30xn0Pv/PvFjT1JeiIpMiqYyf53WccwFfDLktkNZ2MoiEMelILwuDfgTzR7iDAkbUBPPaDm58l721vGFjFQVmhzz8NFQsBePqeO1m3/DkCFZW8/5v/b9ArXs+HquewrTvGRb95lf7kyASI+46aVC8XAG2bNtC8bjXFNbW7jSCbxLsHxm5ktXIJ2SWhcAE3MnJkbsSUZ1zbMRR9MP/UMznji1/N544Y+R5PYnyc/OlL9+t4QRRRBNMIElSi6AYFslrn9a7IEyMPtR3BJ665lJM+czkYQ4RWPmpo0DN5dxjpTTqJ3aNrIIMiWLAZ2UJj9STec0gNRnkvOut9ODw+drzxKr27mvj7xiLm+jQqHFEqHHGC2/9KZPUTVDqi+KzjvzPu0Or8Z3t0B6GHr2FFT/2+NU4QEBD4eO0myjtWws3/NA3ChxJEGWaeDVVHmJ8FkT1OYg0Nml6ExmcHQyQUyCbAWQyBWjP3m2GYxIjsgOoj35FLedfAM778sd3lZtGZ575tp7a73cw75YyJHyCOnNuY/+qaxqYXnmPGsSdiaBryUV9AeOwKc2dmUJkhG4cnvwvn/8p8b3IVvHq7SYzAUHRIDjYfNJwGjgDYvXD8N8Dh36tr3Fs8vHInO5teR8Rgweln4y+rYPrRx084WmcSBw+Ts+ZJTGIS40LXdKJ9abJpFRVQB4l4Wd4zI2yM8c2Sk5AQ3x0JjfcHv/71r7nlllv40Y9+RDQapbKykksuuSQvtzWJdx6tm9YDUDp1Gsd+5OMUV08Zk8RzB4vy3jCBiqo9GikmtT8PXyiazvbuOBs7Bnhmczf/27h73eZJDOHE6cXc9/mjD2obcoIK9al2ZCGKjsgjf/4XGo+BYdA+mIT6zMu+RvGUqQBIoohhGLSGU8QzKs2hBD94ZAP9SYUqv4M5e4ianChWr+7Lf/7n9T9AlCQqps+kuKaWOctOxV9eicPjRRAE0ok42WQSBPLeprquE+/pondXUz43T6CiikwyQSLSv9ftSUb62bbyFeLhEC5/ACWTIRWN5PeLsoUZxxxPWV3DXtXbsn4Nuzasxe5yI8kydrebGcecmE+w7fB4sTqcRHu7EUSRQEVVPnHzSNhdbiSLZdzrEwQBf0UlqYGBMfXau3Zso3ntW3hLSph+9AlYbCaxoGQybH/9ZdLxGIGKKsLtrchWG4ahE+3rRc1msTmdaKqKoY8fOaQqKhabjUBFJb7ScnatX4OuFhJqdpeH2YJpBB9LVksbrH80OVIoo1KUNu9BR8rLrONP4rTPfQmby5Un4SeJkXceWdEBBmbOEQrnwWUZU1d/dayeYy+7Bou1UFJLlKQ9EiKT2Hd0DqQxRAs2LYuWTR/s5kziICJHjhTX1LLwjHNZfPZ5PHzDD8kkk+yqOJU2Qef01D04pTizfaZMnI5Ip1BHiziDiFBiGnQNg2X6o7gx69vGfGawnnnBEGHfEnrau7C5XBRVTyEuFqELMrqmEtm1hZmOVrTiOYQttfl29Xe2E25vxWtJUu4YHL8SvebfoYbezQemnng39Gws2KQ1nEEkDT6HjiztW4SDqumIgoAoHhiHIN0wyCgazsk+er8hjpjb6MMiMJ648xaeuPMWALzBACeVL8ROkhZhOgo2Ttb/hbDzebhtIZz8PTjq8/DAJ6B1ULLqtGtg3ocKozo8FWCxE+3rZcvLK2i//Vfoqoooy6jZbN4ZwZQwMyXgBElCtlpNSbXODpT0CMJFENE1dZRcZw4bW/rwYBCXnJx+6RWTzpGHESZnzpOYxLsAuqaTjGaRZBGrQx47KGMvJwlKVmOgN0Vc00hIoA1b5rnte9N1DB0nMuj5OGksprq6mltv3b/kn5M4sAi3m5jcdNgAAGaXSURBVHqsyz752X3OV7E3UDSdgZTCs5u7eXNXP6o+MZXLiXr/9yezJDIqU4KmFnc+eNgYNBrvhajmRItOVKlzd6UMw6C/P0Jg21ozGfsET743GqETaacBrGoO0x0t9BisL3Hhso7Tz+4DDqS4qYFBZySNRRKp8NsHnzUFusvvBARB4PyFle/gGcdGbtH1uRYzVL8p5qepbX1BmYqGmVTNmothGESSCrvCSX746AbWthXmAplX5eW+zx1NwHVgsnNeZmi0hKups2dwiDrxcIj2LZto37KJtU8/kS9XWjeN3l1NBbI7wzFcOTknC3ZAX6ph2J+kysPx5n8fHXefbLFOXLrwbWrD/mLXutW73d8wIwkSyLrKyF9mLiH7SHIEhp6/RVdwaWZko3vqPN73tW+Pe66BpEJfIoPPYUEcp9PSDYNXG0Os2NaLqg2dZ/iCvuDIwS8ZReeN5jDFbhszyz30xNKE4llmlHmQxGE66RSOP4XfC/djDJGa+bL5Y4xx6hh28LDvumEwMBDFu/bNfJvHO7/HLnP+gkoeWtWK32nN9+/VASeprEooPvROFohpDPutfcEwCY+AOjrnSEnWJLMqFi+j/tgT0HRTemN3xj9V04mmJ5ZboDWc5H8buyj12JhS5OSFbX1EU8qY1zr6Gsa/LgPoGkhT7rVjk98+OdC3e3xatSvMSaIMGhhKBl03DpjhdBKHF3LkiN1tOjo4PF4+8dMR67HQp+HZ60wpn3kfQqw9nip3CaMyH2w+Cx78JJxxHTOWXga3zsGeDHGm/G/I8R5pTAPtpx+Founw25OgawMYTXDZm/k8JYau09/VgfzGb2D1KnqsNfyl9Cp6YxmcNglZFEkrGt3RNKIgUOS2EnBaWdc2QKXfTqnHzvDuJCcB2tyXIJ5RqQ44sEgSGVWjrT/J9FIPHrtMWtGRRAGXTSIUz5JRNSr9DhTNIJzI4ndYsFmGKq5Ob2dO9CVSqQQSOrbBoUrVDWRRQDfAJouIgoBuGPmxoFMo5eH0EUieUpZZNuPXwrRa6nFluom4pzEtuQ5VtPK9bRfQ8pNnAJhV7qHEY0MUBDoiKfxOC019CYpcNpw2CZdVJuiyouo6qmag6gZZVWdtWwSfw8JRU4PEMypNfQkMw8BttxBOZBAFgWK3DcMwqA44EQXMaL/Bfls3jME/0DSddS0hEo8+xwnTi7HJYkEZ87PZZ+aO0Q0DTTeIphX6EwoOq8ScCi/yXihvjIe0opFVdbx2C50DaayySJHbSpHLyiePqaXMayedixwXQBQEJEHAKovohoFFEsmqOsmsakoziuZ+myySUrT8PCGjakxwWTou3DYZUYBERsMiC1gkcUjyahCGDkdfeBG9u3ay86038tuj4X7+HfYCXiAGxIh6ZrOsPkUg2wzP/9T8A7C44PivmVEhIyNTdJ1XHryP1x5+cP8uZi/gGfy3zT5+rpQtXVEeWNlKNH3gItIBjq4LctFRUw5one8lTJIjk5jEYQ7DMOjtTpJVdazG7gNbbU4ZaQKLG001iKSyJARQRXNkFAcHUI9NJuDcc+SHkV+JDm0Tc7IQwmS+hUkcWtBUhUi3mVAxWFW9V8cag5NjI/eZIQJiuHFH1XUeX9/Jjp44vbEM/9vYTUo5MPkLdofXm8J7LnQooqXzYLcAAFGA+dV+Tmgo4uy5Fcyv9h3sJk1igjAGDcpH928A4I1QNcd99BP4SsvNhaMoUjV7Pr97cSd/ea2FlvCQd5gsCgRcVuwWkdNmlfHts2bish24aXO538WfK87nyydP4+tnzmDXhrVEOjvY+MKzhNtbyaZM43dPU2P+GEEUcXp9ZkSbYaAoKhaLDIKAkk7lj3F4vIj7EDlQNWM2tQsW0751E6G2FmafcDLOwbwr/R3t7Fj5CslBw9JEYbHamHncMmSrlWhvNwO9PYTbWweTixokBiJgGDg8XpRMBjWbQRBEnD5fQeSerqp5o5bV4RgzJ5SSTpNNJRElGccY8pQWq42GpccS6eqka8fWgoSo5Q0zcAeLad+6iYqGGQii6Tloczhp27yeqQuPwOUPmFrdI0kGw6CrcTv9ne3ULz6KnavfYKCnm2lHLKVk6pC8iqHrvPbPB9AMs0+WDW0UUZsnR0ZJTwwZyYsV09CuGyKlMxfnt8czKqmsRonHNNLf92ozP/nvZjIHWgZwBHpiGTZ1Dr0XW7pGR/0cVHRNzPP68fVd+3Waj1lkkMCqK6ahbNiMPPfMRG8Ff319F7c9Y8r8nTW3HLddLpi7D6QUNnZE2dwZfduf3XsJimCuXSy6QkbVcVgnHbXei0hFzb7K4dlNFGjRNPjon/dc2ezz4fs9IA9Ggn387/DfK80cEDnEuyDWCfd/DKYcC13rzO1KAv54Dsw+DzJxhI0PEyybjx7eCcAvE2fytx0l45873+WWQBfm3+4wIvD55faJly1EMXDsHk62G6Tgb0wt3NYDMH9U0S1dsTHHk774nh0oYmmVtv7xL3JXyJzvvdUS2WNdOTy9ad+jx3f0jI5mPdD43YtN4+4TBZBEAa/dQiytktXe/rHFOsjWDT/X/6sIM23Yow5OmUrblGM5+/0XcUosTNOaVdTMnke4sx1NHZr3JMIhVvzlD+xYC4sCEqeWN5rjpn8Kxsf/wbOPPUv8jZ9SVFWDK1CEw+sl3NZCV+N2mte+BYC3pJR5p5yBO1iEmslgd7nRdR1D19F1zfxXM//tbmqke+cOph99/Ci7QKSzg1B7K7ULFiGPiAIFuGv5DjZ1J2h1VNMbz1DqGcoRaxgGPbEMH7rrFRIHKIfhcDy1sZsPH1GTJyUnsXeYJEcmMYnDHJqq06ur6IN8w3hdoWQI6JksYtosoQmFC/LhxxmAMVifKAiUee0EXdb97mjz5Mhk5MghASWTNg1AYyRMy+1vXreaRHjIuC7bbNQvPpLWTRtIxaK4AgHsThfxSL+Zo8MAi92O3e1hoGdwEjnM+CPKEvVLluIOBDEMg+a1bxEL9WLoOhXTZ1E6tZ7unTvobtpBcc1UKmfMIt4fpnXTenRVRc1m8ZdVEKisQs1mMXSNouoptG5aT6i1BQCby4XV4chPqtKxWD5B7fBrC7W1EKisxuZwkopHzUmRZGXhz18zdX+H52vA9IAaSYAcCFT5HXxoSdW4EVm7O8/umiCLArIokBwkYASEvE1v+HVNFAc6Kni8MGND12lvb6eqqiqfNHCip96bNk6kqN0icfa8cvzOAxMtMIl3DrqeJZlsBUpwDnrZq7Yijr7gowXyNd/95zoeeGMoCasowJlzyrn2A3Mp89pHVnvAUOYz6+6KphFEkakLFsOCxSw6630YhsHOt1YSD4dQs1lKauuomDELQ9exDpICIxMM6rpG984dWGx2imtqd3fqPWK8hNTHfeTj+1XvWEhGzQgdp9eHqiiE2lrwFBWbJNAItG/dTDoepX7J0jH7D13XGOjpxlNUMu64tm/42B5LzD7h5PznRWe9b9xyZXUNtN99MTAoqzUOOSKO9K4cJsGVi0LIaA6qZptWhs6BFB+++1XaIykskoCiDdUri8IeIxPLvXYuWFxFsbuwrxs5/uRI/9zmoMuKLAqE4llkyXSiySjmXG/kI8o9MyH/nRHfhTG3M+5xY5fPfTd0g5bWFmqnTDElM8ZpQ1bV+fF/NuUNFd84fTpOq8Qdz+3IR25cftI0ikZEjQ2/vrSikX3eHMOtumJGjmg6DD7G3DN7sk3ixpUb8sfd99ouDgQkUWBxjZ/OgTQDKYVz55fTUOoedY8K2z86Mmh4udxHp00mmlLeroC0Ye15e+vPPBog3dqHxVBJZtVJcuQ9ivjgmHPhH9YwpSHG++ZXML/Kx9wqH+59cYAYJEY03SBetADfZS+QymoMpBTKvDaEZAjuOhbCjeYfwLTToOU16G+CV24fqmvXS4hAxHCx2n8m3z9mBtUBB6GESQZYRBGvw0IkmSWZ1VB1HYdVRhIEMqqGMswIbRuUwJZEgYDTiihALKOSzJjvvqKZ0Q3WQafJWFqhyGXDAMKJDLIoUuyx0RvLoI0RuVrldxJwWvJOJVOCTtr6U3gdMr2xDIpmYJGEvBOYKAjMKveSyKq0hpPohoEoCNgtEvGMakZd6AYum8zSuiCGATv7EqiajqqbdSUyGkGXlXAii90ikcio+X2yKJrrHknAaZXojWdJZVVEQcBhlXBaJUTBHBs9dplwIovLJtMTTefbJwogiub6TxyMusAw6O5opapmCtG0lt8uCgyWGyqbi9QQRfIRGIYBdotY4HyzP5BFMX9fHRYJTTfIqDr3r2yhOTT+OXQDdM3Iv0vvBMYiYDSjcG6zomgZ/3xsK2s6k9z4oQUsPus8gLzU7XD0d3aw7tknWdNfSUvCz8wZ5bgaPkDXw0+yYfnTADSuen3UcZLFwhlf+ApzTzrtAFzVnrFjlUhj3GQvl17/LJefNI3/beyirT9ZMDer8Nn5zHFTD+jYt3hKYJIY2Q+8J8iRicp8TGIShyMSaZXhQ894b7s6SIbowtglRm6VRZESj42gy4I0jqbixGDko0iGZLWG6tN1nWzSDHdVsxlkqx27251ftOmaRnacxNiyxYJksb6ntBwPVH8W7mjjgR9+ByWdxhUIjFkmOTAwblLy/YEoybiDQbLJ5ChdeJvTRSaZyH93B4IkBiLjSsrAoJTMAUqw2Sv5TW/PUezHgbnvAaeF8xZUUuy2MbfSy7IZJVgk4T31Du8JptE3wqJFU5Em9X0nsY9Yt/5LNIfCCPwAp272D0XT5hQQI0+s7+SBN1oRBbjmvDl85MgaHBbpHZFbKR8kXrqjo/XvBUFg2hF7l69FFCUqGmYekLa9kxhOgsgWC2V108YtWzVz9m7rEkWJQPnBl3PbHTxFReiDxgGLoY6OHBnMOSKLhUu04QnZS7Omw0Jad/CvHSlWrXqNN3f15w0RucW3IMAXT6znu+fM2qNhWxDGJ6wPV2iaxhq5j0WLqvc4lhzfUMxLO/o4bloRtUUuAGqLXFx235t89MhqvnvOrN0ebxgGzz8/7Lnm49ZM+bSAYhpK/tUsgRUuW1ZPuc9OTyxDekQEqd0iMbvCy/wqHzUBx7hyaCMxKRO1ezz8so+mVjNy5J2I2p3EoQdd18gOrjtSop3VLRFWD0YOCAJU+hx7NCwmMiqxjIrXbsFhFfME5EBKYSClUOGz05/MklZ0ilxWXDaZ2dpVnCs/S0y30ZTxsrz1HKqF8zhNegkZFQyDftWKoavYUHhGW8JXzlrI+xZUvK3340DguLex7nlVBz9S21yT9LFoUdUhvSa56KgantvSw+IpAaYEnQUSYRlFpytqSnAlMibhP7Pcky+T1XTeaApTW+Skwmc64FhlEWk/5gQG0Ngbx2GRKPfZUTWDrKbT9UI3bcPKre2IARYeeKOV02eXcey0onGjtE//whXMPuFkelubee6Pv+HVDVHYcF9+v8XuYO5Jp9KxdQu6rlHRMAOLzc6ck07b7dzyQKNzoNB28usVjaPK2GSR2y5ezNK64DvVrElMAO9qcsQy6DWWTCZxOEaH309iEu8GJDPmBN8hikwtc4+5ADYMg3hGRdNN7U+LLGKTRFOZAxiefiDnGW+X989AVNCMXBTKMFktwzBQ0ilioT6UEYbteFhGlGQMDLRsdreEgGy1Hl5JLA3T+JGJSBN3hx+GWNK8Z4/9468Y2sS0qEc1wTDobtqRJxQGusePxfYUl1A+bToCAgYGHVs3k4j0Y3U4qJ4zn/6OdgZ6unD6AwQrKrHYHXQ1bifRH6Zq1hycPj8w5OEZam8l1NZCtLcHMJOQ1sxdgJLJ0LF1E5lkAkmWKZlaT9eObcT7TSOQt6QMm9OJzemia+f2AjJEzWRAEJgydz42l5tEfz99rbvIppI4PF4qZ85GkkYPdyP1+V9qDPOyPJ3vnD2LDy2pGiWPlfMKEswLykdi5LxRC6IxBMbcZ5OlSY+OSUziHUAo9DwGC3BqQ+TDS4kAf/7tq3nP982DUkBfOnkanzm+7h1tX44cWd82wFst/fmxe1xvegr3a5rOjrCC2DaAtI9JSw83WCQRSRRGGZMPN+iDUSGSoRUQ/7qh5/Nq7C4he8kgOZLSHNzzRjcpycwrNaPMza0XLSI4GOFgl6V8jpx3Ge9xwFETdPKxpYU63WfNLWf5VSdT5d/zGlIQBHRxULZpkPTKOQYVKRFEDHRDoFso4sTpxXzn7FmTZMY7jJwUn0XPHpA+JPfbHejtQVOyOLw+4uHQqEjlgwk1kyHc2U6wssqUfxmUAOxq3E5RzRQ6d2xFKiqj0+1Ali2kYlE6tm4i1NaKrutoShZdU1EyGSw2Gwgisb5enD4/hq4jShJ2t4dARSXdO7fjLSln9gknEQ+HiPeHKatvwDpog2nZsI6W9avxFJXgCgSJ9fUy64STMHSDDcufJh2P4SkuYeaxJ9C46nX6OzvGvS5RkqiZM5+B3m40RaF2wWKivT1kknFURSU1ECEVjyFbrQSraghWVtG0+k12vPEqABoiacnOJ4+ZQk80w/r2AToH0rRHJu4Q1hcf2ymrc2BozhFKZAklsrRQzf+4ZKhQFHZSxgt8aMw65lf5OGde+YTbMolJ+J1WPrhkSP5JGmZksFskfLuRRLdbJE6bXXbA2zSjzJP/bJHAgUT/iKhezRhq56X3rqLMa+PUWWVouk53NMOMMjfRlEpS0Sjz2LjilBlUz5lHef10drzxKv2dHQiiSO2CRcw/5cy84sBYiCSzPL+1ly1dMY6oDSCJZuRoOKEwtdiJVRJJZjUiKYWugRQVPgfFbhvtkRRumxl5lMvlIwgwtciFMJhTZUdvHJsssrY1QiRp5hEpdltZWhfELkvMr/Zx+uwy7BYJqyRit4r56K5JHDp4V5MjkiTh9/vp6TGNcE6n813nFXWowTAMMpkMoihO3ut3CPFkCkPXkGUJbTdJTF35X7sBaBiaNqYvvDD4l83um+E9By2bxUBEVVQ0Q8WQM2iKSlozIKuT6OkmORDJl7fY7AiiiJJODZIlQ5NOUZIRRw4gBmhKFiV14CMbDlUomkYo3M+uVa/SumHNftdXPGUqx338s7hcpmFF1QySWQ27RcQqi4iiRMnUOkRJJppW8dplwrEkye52yqqrsdgdJLIaDlkkqxk4rBKxtIKmKgjJGILXDIt22+Q8IRBNZelpb0dNxvHYZfyl5Th9fjKqTjY2QCoSxlNcgsXloWVXK9ZsEo/fj6ukDE03hqRBEgPINhtKViEZ6cfl9yM4vVgkkbSq4bHJDPRHcLldSLLF9JxRdZxWif6kgmswuWE4kSXgtKDqBl/9wZMAnLeggtK3UU5nEpOYxDsDAyGfuNowBJZHfHRkCnPwzCzz8LXTpr/jbasJmv1uNK3ywbte2feKnn31ALVoEu8U/jYoLWox1IIIRW2YdNZIWS1dH9qXjxzR7KQlO0unBvnmmTM4ojZgJjydxAFDXbFrwmUNyQIaWHS1QHqsNE9mWfng8XO45oOLx69kEm8brHZzXmc1FFLZwSirdJpEpJ9wRxudO7aiaxp2twdPsIj2rZuRZAld1/N5KnLQNY32rZtI9B+mOd1GYMt/H96r8v2d4+VyWMeG5U9NuJ7NLz0/etuLyyd0bMuGtbutJ4em1atGbVvnnce3zp7DFac05Lf1xTO0hJN7jLKzySI+h5m3YXgEkiQKlHhsdA2ksVtE6opdNPYkCqSFBAECTiv9yeyQNNOgo5UkmknCW/uTzCzzTJKnk3hXQhAlhkueqAZ8YFEl4USWrV0xuqMZ7l/Zkt+/YlthzrA3msPc8fEllNVNp2L6xKKll2/p4U+vNPPSjj60/c0yP0EsrPbxyBXHT9pDDzO8q8kRgPJyk3XPESSTeHthGAaKomCxWCY7g3cIXZEUqgEZm0w6diA1tvcPPekEGSwksjGyog1FtmJk+wnpGXBqxOIZDE1DkmUcXh/ioAFL1zQzV0TeA19AknWE7Nj64pqicqAkj94pqKqKvA/Jcg3DwG6RWXDMcSw8Zv8CmXuSGj9bq/HDB82oEa9dRtWNvEeExyZT6Xcgid3s7IuTVnSskpif5Ff6WomlzdDyHKyySHYweejwsrn6FF0nregF5Us9nQykFGKDut42WaSuuIP2SIpYWsVjl7Fb2gknVhdMaIrdVkKJLIZhap6Xee1s6YqOuajJacFqhkGx29TPhXxOYwJOC/2DXh5FLivVgclIw0lM4t0BAbdmajCrhpWU5OQ7Z8+ielCqRhLhqKnBg+K9VRN0ctWZM7j1me0UuazYLVI+aiDXj43K9zBsgwEo2SwWq3VfghAPSyQVDW1QL/xwRVYzyKTM9lt0leHrdH2YdNZIWa2sPjTWBhVTMz+h2zluVgl/vfSYt7HFk5goBNEKmml8N4ZJyvpV07Ce0qzMrS0+mE18TyOXr+nY/pU8+73P85xhJuE9EJEeVoeTbCqJzenC6nDud30HDIJ53dlhjmROnx9NyZJJJUlG+rF6/ciiKSUgiCKVM2ZRPGUqoiiSikWx2h10N+1Aki2UT5vOtpWvUFRVg2EYuHx+NE0j3N5K+bTphNpa6W7agc3pwh0IEmprMaOoDANvcSn1S45i1/o1RLo6SMdjuPxBEASqZ82hevY8mtetpnPbFlyBAHOWnYrd5R7zsnp3NbFj1et4i0sQRIme5ka8xSXY3WaS9dKp9QQqq0hFo2x56XnSiRgzj1uGzeni7y9s5HXmcsEIJ6hit41i9+jkynuL4ZFm86vHloWqY3zStcSz/22YxCQOVQiiXECOLJpSzI0fXIDDKjGQVLjuP5voT2ZZXONHlkReaewjkVGxWyReaQyxtm2AE3++HL/TwvffN4cLFlWSyGj86rntbGgfoNhto8htRRbFfPTJkxuHFDJmlnko89lpCyex/P/27jw+qur+//jrzpoVEkgIQcKO7IICVUAW2aVS6dcFFZe6i+Da+rVW+3Nfqq1W0Na69FutVasIFREQLYpb0bLvm8gaAoQkZM8s9/7+mMwkgWQImGSSzPv5ePAgmbkzc+7MfHLOPZ+z2G1YWLRtGcuu7CJs5Ss8GAa0iHFS4g3sHZQc78LnNyn1+mmd4MZpNygo9ZEd6k8wSI534nYE9rYp9fr5/SX91RfaBDXdq4taMgyD9PR02rRpg9frjXRxmj2/38+WLVvo1q1bo16XsTm54/fLKDIsnhjbkzM6N54puI9/8xFb7F04f8s8NicMZlf7Hsze/AK9CrdwdODdfP7uF9hcLv7d9VIyCwsiXdyGZXHSS2pZFuSWmpT66iYRdOzAieCmo0EFZT62Hqz6uVROdmQePX6d/GBi5Nhjg88XFExKeHwm+3Krzvwp85lsyap43YJSXyhxUll2YcUsqZwiDzlhNpgzy9dcBUKJEajoeAwmRgAm9m2rxoxIM2FBaOaI33KR1CKWm0d0aTQjMmeO7s6NI7qcUnLm2A3ZpWk4cLSEdU9XWlar8swRs+aZI6WVlt+KNQP1b6nl5KYRDbeOtoRnOdzgrTxzJPAZtvSV73FgOuiSWn1nr9S/4LJaAH5PRRvWZnfgigskNBLL97mLTWxBh379MTDwlJbQun0HbMcs19KqXXvsLhexCYmkdOiEp7QEp8sddlmXxuZU6pFBk//nR73mTy68uMb7eo8YXavnOP2ccxk29apaHXvW+ZOr/H7ftlZ4DxWGlrYUkYZjs1X9O/Pc5WfhcAVuaxnn5A+X9q9y//RRFW2cdfvyuPOfa9h9pJi8Yi+/em8tv3pvLbUx7ewO3DC8y0nNBpXo0+yTI0F2u10Xjw3AXz76JiYmRu93Aygs9LCtIDDLonuXFGJiGk9DL8v0sc8wKCjOIcvtY59pEFe4m5jCvWQX5VOck83huHS+y6ybjbTl5I3u2Ya7xp7O0ZLA5pRtW8TQMz2R3GIP+3NL2Hk4MB18cKdk2raM5WB+Ka3jXXy8MYucIi/tkmIYkJHE94cLOT0tkW0HC+jfPolYl50jhR4SYxw47Dbyij1kHS2l2OOnXVIs3dokcLigjFV7cin2+MhIjqN7WiJ+0yKv2MOOQ4W0aRFD7/QWbD9UwPKdOWQkxzKwYzIOuw2/aZGZVxLapLTI4yO/xEv3Nom0iHVQ5jP59occerZNZOXuXHqntyAl0Y1pWhzML6V7m0SWbMoiNdHNOV1a8/WObIo8fs7p0orUOhg1JiKNQ+VltXymgyG90htNYiRIaw5Hl7TEGHxG+cwRy19lepCv0r4iDqPqJVpZpT3GYszAYIBSnHTRhX6jYdmDe454y/ccCbjG+38AFFkOuqQoORIpzkrXSK5Wafxk/AS6DjqblIyOdfL8rhjNOm4KsvIDibG0FmrvizS4Y/ZTM06iv/CM9kks/eUofH6TP3/+PX9e9n1oxYvW8S6mDs4gNdFNdmEZfhOcdgOn3cawbikM7Jhcp6chzVPUJEdEmorC3DJiE53YHVVHHnlKfeRkFlW5bcvOXDDAbkF6I7tANoO7sFsmZvkoqmAnVe7hXAAKLRexTjuL7xxOrCs6OohMv8mGjRvo26cvtgiuDe6y20iKc1V7X5vEGNokxnBmh6oNiYTyEY9TB1fdtLRj68B3r31yxVICca0qqpcEt6PKfRCYNj6hz/EznVrFu6qMrOzTriV92h0/LT244WxNfta/HVB1MzioWOf/kkEZodvqYxM6EYk8P/ZKy2o5ObdnuwiXSKKdzWZglXcOHDtzxKw0O+TYmSNVkyOBQSUlOE9YF0rDCW7I7jJ9lJp+zPIBHLG+8s/LcobdFFcajnPkVBLPPofdXj+7d+dGujgRY5p+th/xYO7JPW5Ed3OUX1qxjG9ay8YzoFAkWlT5O2MZgbWvT5LDbuO2Md2ZPqor+aU+Yp12Ypza71h+PCVHROqZZVkU53sw/Ram32THykMc3l0QGFFmVawhbllQVuzlwI6jdB+cxvjr+1R5nn8+8V/2Ha6aHDngsCAekh32RlchmMF1owwT06iaHFnz2TIggVK7m0n90kOd69HA7/eTHGMnNdGt2VUiIvWolNhQveM1nXRpmxTZAokARmiGga9KcqTyzJFjkyOHFy+G08YA4A7OHDGcxEXJwJImwR7obHVaXoosf2hZLbcvMLK1xNJldyQ53RWd4U+t9vDUmi8jWJpGZum3kS5BgxrePYUWMUpUijS4KskRG8aPmM3tsNs0QETqlFppIvXAsiyy9xWydXkW21cdpCg3MGrMB2x1+jnkMGt8rBFjkN7q+NB83VnMjpbVL0GV0boRbf5XLjhzxKpm5ojHDFSMZTY3t43uFpkCiohIs1ZCHHHB5IjlID5GF1ESeabNCSY4LF+VZbWCG7LbjYoBL5YnkAjJXrwYrg8kR2L8gbZgmc3d6AbGRDPLEeh8d5m+8g3ZA1zls35K0N+fSOo3ehzL12zh74dTcDpstIx1EuuyY5zsJoDNikVZmQe328VJb4bYRCXHu3ji5/0iXQyRqGSzV/RxGZYBTWeLJokCSo6I1LFDu/P56O+b+OpQPuvcPg7bLUg6uedoeTSfY7e7692tFXs2Zh13rMtu47LhnU61uPUmNHMEK5AcsayKEbxWIDkyrHcGnRrZcmAiItL0ZZHOu1zOFf7FAHhMB/FOjbKXyDPtweSIv3JuBL/p5+wtJlcs87N7yVV4DxzAe+gQeL14z67YFyE4c8Tr0LIwjYqzYuYIlolV3g52lc8cKUZ7HESS0x3Drffdw41+E4fNUGKRU9uQXUTkVBlVlu/7cTNHROqakiMidcQ0Lb5ZsJPnP9vBapcPXw2TOVIT3Zzfty0J7urDz24zuOCM49dFn3X5mXVZ3HoX2nMEE8swcFse7ARGRXrNwH0d0lMjVDoREWnO7udpSo24ihmLlpMULUEkjYBVvtm6w/JhWRUzif2Wn5HrLdJzLIpzVlR5jMdVsQRMjBVIjviVHGlUDGeg4e8yfZiVltUKJkdKlBxpFJwR3O9PRCSaGbZK/V+WDZSklkZEyRGRk1RS4CH3YDHBTUOO7C/ih3XZrD1wlA98RWS7A8MAT09NYNqQjkzo0xZ3+ebqhgEtY51RMVopOHPEIrDnSLCDCsBbvqxWx3atI1I2ERFp3kqNQEdlsO4ps+zEONQpJpFnOtzgAafpr7Kslt/yk1wY+N3dqxdp996Ls/1p2OLj2bbh08DarECMP5AcMZ1KjjQmNlcsEJg5YlpmxbJaweSITckRERGJXpVnjhinuCG7SH1RckSkFo4cKOKzJbtYeSCPldkFmJXXQQAKDYsshwV2SHY7+P1lAxjds01UJEFqEkqOlO85Euyg8lvO0FIDsQmJESufiIg0T5U3uU7wFwNQarlwaMSwNAb2wN4TDsuP/5jkSFJR4Of0Rx4htl/f0H2+xETIDfwcXFbLFqNlSRsTe0wCAE7Th1Vp5ojTF5gdVGIomSUiItFLy2pJY6bkiEgYO7/P49n3N/DV4Xzy7OUXsDVEjd2An53Rjvsu6EWbRF0AhTZkx6oyc8RvVWxI6Y5PiEjZRESk+bIsf+jnYN1TaqnJK42DZQ/MIHBYPnyVkiM+n5ekwsDPjjZVlx0tq7SJaYwZ2JAddw3rt0pEOMqTVU7Lh9/vwypfRsvtCXQGldljI1Y2ERGRSDOO3ZBdpBHRlaJIDd6cu4Unln9PsQ0oT3Knx7kY3ak1fXq0Pm5WyMCOyZyeppkQQWbo/TEx7A7i/PkA+K2KZQVsNo3iFRGRumWV78kAEKfkiDQ2rmByxI9pVuw5YubmYbfANMDRuuqyo15Hxfe3YuaIBpg0Js7YYHLEi9/0H7che5ldySwREYletsp7jqB+IGlcdKUoUo0//mMdf1y3F2zQ1uHginM6ctV5XUiOd534wQJUnjliYhiO0Ohdn+UivXsPCnKOkNqxcySLKCIizZBpViRHgstqlaD6WxoHW/lG6g7Lh0mlDdkPH8IGFMbbMBxVL9EKqFiKIjhzxIhrUf+FlVpzxwUGSLlMH36/FxMbhmViJ/B5lTmUHBERkShmr7SslmaOSCOj5Egz9MPefNZsP4JhlOdjDTAwQvsd2QwjMJbJqPjZAIzgzwa0TXCT4Kr56+Fw2EjtmIjDaa/xmKbIsiyee3MdszfsAwNGprbgpduGEBvmvZDqVU6OmIZBnL8UAB9uLnvkaUy/icPpjGQRRUSkGfL7K5IjoWW10GbI0jgY5cthOS0flukL3W4ezsEGHE08vm2dY5Y34i2LmPLknzu+Zb2XVWovmByxY+LzBT6j4N8fAK9TyREREYletsobsmvmiDQy6vFtZoqKPUx44Us8PzIR67Ag3W/DYUFNT2WzG7hiHVWWl7IsC5/Ph2P+Z01yM/Iir59dfi8YMCqlBX+9a5iWfjpFlZMj1jF7jths9iqVo4iISF0pqSY5UmJo5og0Do7y5Ij9mGW1rKOB5UeL449vH+X6ypdosrzYCOxT4o5Lru+iykmITayYyWN6SrDsttDMNdOyYTm1H6GIiESvKhuyW+pjk8ZFyZFmJjbGwU+SE9l+tJjgFo9W6J9V5feafvZYFmWGxV5HxQVbjcr81d/ur+H2JsCwYGqXNjx2w0AlRn4EszytZmHix6hYVsvQ6F0REak/Bb6y0M/BzskymzompXFwBjfuNv2Yfm/odrM48F31uGtOjrgrLRkX20IzRxqT2PhKyRFvMdiplByJwe5SglZERKKX4ahYNUQzR6SxUXKkmbHZbLz56xE/6jksy+K/u3LJyi+lzFt9ksOy4Mj+QrylVe83LZOcnBxatWqFzWiaf/CG9mvDoD5tIl2MJi84c8RtxGLY3KHkiFdLm4iISD0q8FZ0ILfwFwBQao+NVHFEqnCVJ0cc+DG9FckRq7R8+VFXNctqlSdHgvuNWBjEJWjPkcYkNj4+9LPXWwYxkOgLJEd8lhu7U8kRERGJXlVmjtS4Po1IZCg5IscxDIOfdG514gMHH3+T3+9nzZo1DBjQB7tdyyZFM7M8OTaei/nUERNKjhQb+l6IiEj9KfAFOpyTrFwS/UUAeOzx4R4i0mCcsRVJDU955zmAVVI+iMR1/OCinPKtSYL7jfhxkJio73RjEut24TPsOCw/RT4PbrOMZ7b9HgDTcuN0aXCQiIhEL5u9ovtZM0eksdE3UkTqhVU+GiDGcuE3KtZ9z7b7KSgoiGTRRESkGSvyB3qS461C7OUrhPpdCREskUiFyssvHSqtWAKOkvKZI+7jx67l+gOXbDH+wPF+w0lCvDb4bkxcdhve8gFAB00bN+57n75FOwDItzlwujVzREREopdh19h8abwi+u386quveO6559ixYwetW7dm2rRpXHfddU1yI28RqSq4rJZh2Rif6SW+fN1lD062b9/OWWedFcniiYhIM1Xo8wEukv15ods8Ds1alMYhvtLG3dtKyphY/rNVUsq2jE68N+TnrNm8G4dhYMfAbkCeP7BOd3DPkTLDhq+0CEhq2MJLjQzDwGc4AA/b7PFMKU+MABTa7Jgl3pofLCIi0swZWllGGrGIJUfWrFnDLbfcwvnnn88dd9zBypUreeaZZ/D7/dx0002RKpaI1JFgciTOd4Tf//BrXLaDAHhx8uUXy2jTpg1xcXF4PB5SU1O1DJuISBQ6VObll1v38t3RItq6nczu1YEzEn/ciPhQcsSXB4AfG36/B4/Hg0ubIkuEJSRWbKS+otTPx9lHKTMtNqZ15c1bp3IkKZk1WbnHPMqG0yoLLavlNWzMe+9t7vnVrxqw5M1Xfn4+Ho+HlJSUH/U8lg3wwx5HXGgGNYAfO1aZ70eWUuqCZVmUlZVhGAZut5Y6ExFpKIatovu50CrhX//6F0OGDCEtLS2CpRIJiFhyZPbs2fTq1YtnnnkGgBEjRuDz+XjppZe4+uqriYmJiVTRRORH+jq3gBIjjhRPLn1Ln8ZlOxS6z4OT3LyjvPrqq6HbkpOT6dOnDzZb81/pzzRNDh48SG5ublScb1Nis9no27fvj+4cEZHau2/7Pj45kg/AUZ+fC1ft4KFu7biyXWvspziTeHlh4G9razMHAI/hAAvmvPcul18xrVHMULYsi40bN5KdnY1lWaF/tRWsS3Jycmp1Pifz3PV9fGMqy8keXxfP7fF4aV/+82qvwZL1PwR+OWscAMlFh5l5Rn/8loXfsvBZFnsy36dz2Wd09QbaUyY2igoL2blzJ126dDmpMklVXq+XV155hYKCAkaNGsWoUaNO+bkMAp+3y/IR5y8N3R5DGcVHjvL999/TtWvXH1tkOUVlZWW89dZb7N69G4BWrVrRo0cPHI7GsdTLyf59+bFM0+TQoUMcOXKk2muShi7PidRVeTp37kz37t3r5LlEpPZslf7W+gnsVbx16xbuvPMuJasl4iLSEvB4PHz77bfcfvvtVW6fMGECr776KitXrmTYsGGRKJqI1IFHvz9AK28en6y6jgRyqtx3emkW+fEp7OY0PCZgWeTm5vLVV19FprARsn379kgXQaqx/OsvGNEtEae9cSSuGtuFaUOwLMjNy2XF919SU59vo3pXGlFhckt8bMkqochjRrooDBnUn9HnTwl7zPfFpXx0+Ch2A97o14VX9h7m89wC7t22j1m7D3Jlu9ZclJZMh9jaXzDtLinjvZzAAJtR/mUAeMubu9u27+DxRx9uFMkRr79uvjg7duw48UHS6PzcAJsFXcwS2ifG4bYZFGzdRGLWAfpt+IKCN4urHD/xsq3EtPDQuaAIgMO2JPDDu2+9wbndWuKyB77ThmHg81tsOVjC4ULvSX3PTkty0TMtFrst8vFRF6zy9uXKnV+GjfnsAg8F5e/rF8s+I+bQamzB411xYPox/J5avWYfIzA7pKW/kK5lu0O3xxBIlCx89/84p3vqqZyO/FjeYtZk+dl/tGIGT05ODv/5z38iWKjGIdrqkW+++YahXRJpGWvHsgLtTguLUq/F9sMlnNbSRVoLzTJtcIYBrbtDfOMZpGZZFnv37sXv9zeKtmN1LMti56Y17Nx3gMZ+2Wi3+Rk8pOptJSWlPPnkkyTG2PGbFi1j7ZT5LApK/ZEpZB2wGxY92sbTvlUsmA13Hqd1P4N2/c9rsNdrbiKSHNm7dy9er5dOnTpVub1jx44A/PDDD0qO/AjfvP8Uvr3fROjVLfymn6+/tAONswJp7CzDoP3oX9K134gqt6/65K8UbvxXZAp1Eva5W5OaOpirs78kvSyHAiseJxBjBC48W1j7uMK/CQj0KRYQzwrOoBSNFpDI2kAPir1xLNl8NNJFEQzIzIt0IeRHyN3wbwiTHPGUlvLO+79jXGwq7UsO41yzkluAkQkdWNGiOx6bk1WZsNqyaOPJ5fTiA7T15NDSWxS2dbE6sQvjWp7OuUXLueyHlYHXMpz81Po3HzEGnwmNJaPlwEdPdhBLaWjEec3nVn2Zqz/+ZI49leOrO/bk3tP6PP7kn7sm9fOeHyEZywg85LebnqbIHlhmy55XgNdwQIfjH5NcUkT7PcWk5AT2rcixWtKaHI74WvHplrqps3YdKWPXkbITH9iknFxdYloGizdVfj9r/1iAnuULHzx28AlOL8wO3R5rleE0yjhS5uajDccumSYNyU0pV/ABqRxhDX04SosTP6hW6qZeqburZ5WnOjvpQBZt+GZnQY3HHDjqBYp+5CvJqfk20gWo1vr16yNdhGbBe0yeYAqLmc84TOyhZEhxIxjg9WN5gbX7ili7r2H/jrg3fMK9fc7F5nA26Os2FxFJjhQUBCqjhISEKrfHx8cDUFhYWOvn8vubbkaxPhzYvYmzNz6F3WwcF/5yajI/XoO/984qt/VccQ9xTWS94isOzg39/IlzCJ18WZzFBgBib/835p5PIXMN+MtISOrIqKP7MHzN7YK8epZlcTT/KC1btGy0I1Ci1Sj/AZbnJnGozN3IUruN5+95Q7wvFoFlTpxOZ/jXa0QfUmMpig2LnolFpLtLa5x10xAMwyBh4CVh22j/fuVy7j+y9LjbR+X9uNcekbcM9la97QipDOzdjR7GYfwlJ2pj1iLe6mhoXqzNh9tmQdjBAdW/lmVZ5Bfk0yKxRe0+61qVua7OvbE9T21equHKU+ItwH/QwI7FoIK1tXuJrVV/zfI4uHJoO77aepgSb+WHBb4MiQ4vZ7QoIM7ur9X3w2Pa2FSQwMGy5jRQxcLr9eF0OjjRX+kYl53+3dqzause/H6z/NGA3wOGDWwVz1HTR+Txm/jL99vrkZVd5b6txf05v3se2/PjG1GNHmUMGy2MQn7CWlqn9QObkyEnflSzFo3XJKZ1lBV5sLs4FghMVjAAmxHYJajEb8NuWFX2DJIG4iuD0pwTHyfHSaKQfqfFEedu3DOethfGEkxLGqaTfje+RJdvXqWgsBi/FQjG0vIYTHZ6y6OwhlgM8zerxvg1wtXjJ/86Ncksi2dLDvgtW/njG+bvScf2bbEMW4P3kQdfrzH2zZ9MmSKSHDHN8NnAk1mHX1ncqrylpZQmtKW1T6OSmioLg++N7hxYs6bK7U5XJzLs+yJTqEq8OMm1JePBhWnYMLEdVwHFWKW0MQ+Rb4tnbUFbcgtakZRQyuHtqTiyCsB1NnQ6u+IBrRv4JERq0LL8n0hTVQr8EOlCAJQAx9RjlW1IHo7LKiLdc4i2ZlaNo/29OCky4igx4igz3LXqsLBhkuHbi9vv47CzFZu2pFN0/s1YNo2kksbBueQWMtx7TuoxeUYS+A0sh8XRw7HsTB1HuzArNB08yTK1KP8XrXKATmlDf9RzbF/wX7rGbcfAwms4WZ89kMKjfmwZ0O7Ma+lms9dNYeWU7S7/J9HLAWjnn8bJWZqNqzgr0sVocsri0smKafwdKjbLomibDyNuM86ycazJ8kOXa6s9Nq9hi1bnIrUb3Jow1171ran3zUckOZKYmAhAUVHVaUbBGSPHzigJp1+/ftjtamhWcc7GiL203+9n/fr1+lx+pGovzQZ8V+Px6wtKWF1QTIHfxG9ZmOVrp1oE11EFEyv0c7ALyrSs0O/BY6FiFE3l2ywsPKbFW1m5eI8ZGek2DGLtNpyGQb/EWJ7pfho548fhz87m3vdv5cif/0zhu8UkJRyh24ABp/q2NAuKEZHwFCPRoV///qzIv4FOibG4TzAoJni55zUtthaXsjK/mNUFxawvKKHIb+KxLAr9JkXlI75vcMO06y4PPX7U726kxVmD6+tUIkJx0rRtv9Ekq7gVnRYuxNUxsI7WzjFj8WVl0eGdt4np1++4xzjXrGHvtCsBg0te+TXxUd6eOpFIxMiuh5I5uD2Z9q+9StbNt9DOt43OS/+NMy2tQV5f5GSoHhEJTzFS98488y+RLoLUscYcJ8Gy1UZEkiMdOnTAbreze3fVcRt79gRGUHXtWvtcvt1ub3QfQKR9mVPAxsISnDYDu1F1LKZR6Z/NMDhukpdBldsMjOM6yoOd7MHjbeXH2Qzwmya7PQbbDufjtNuwl79GkHXM/1Cx4XC194WOCf5uHX/fsTdUc1/wcVbYY2pXpopjai6T34ISv0mpaeIPvmdUvH+mBSYVyQnTssp/D7zXl6e34uykqknC97Jy+Dq3EJsBHtPCY1mUmSaHPT5W5VfduLO+dY518dTpGaS4HCTabbRzu3Acs4HntvIpbHaXi/RHHyWnWzeSfv5zxWs5/e0SCU8x0rzZgaGtTm6cut0O/Vsm0L9l9YNo9pZ6OFzmpcumDRyodHvCT37SbL9LipOmyShvn9ttRujzs0pKAHAkJFT7mTqTW4V+dsTH63OvpYaMkeDSRAaAL7AUrsPt1mcljZrqEZHwFCMiJ9bU4yQiyRG3282gQYP45JNPuP7660MNyY8//pjExETOOOOMSBSrWSjxm1yxbudxI/sblg22RX75p6Ysx+s7LjnyxM4DHCjzVnu8DRjZKpFUlwNHeULKVikxZTOM8iRWxawQg/L7jcpJs0AyzCqfeVI5OVZ+KHF2G5ent6atO/zyJFZwfT+HA0dyMm3uvPNk3wYREZFay4hxkRHjoqAkMGDAiI2l49/+D2d6eoRLJnKM8tlSpscTusksT47YYmOrfYg9OSn0s+GIyCWcnEj5NW3Z9u2BX10ubOUrJoiIiIhI4xSxlvX06dO59tprueOOO7joootYvXo1r732Gr/85S+JreGiQE4s1m7jt61jWZGdh9ewYdpsGIZRnoCyymcuAOUzFUIplPIRbKEllywLyyrfjMwwsIzyTnLLwmZVrPhd8RgD0wg8rri4mLj4+MBtgFU+7SRwt1Wloz34LMfeVnkOwrGri1e5z6p6W2i/JOv4VcnDPXfV/60Tv24Nr1/+4hhALODGwk4geRHcmcMGoUSFDeuY38FuGFzYvhXH+lvH1nyyfRemZeGwLGIsE5fdjtduZ6jhp1tRMRSXn4Fpln8wZuCzNM3AtBQsMM3AGxXM6poWmP6KYywz8DNgOJ0YNlv5RbwBNiPw+x7IO3w49BxmQfkGt7bA98W7bz9m+bJ5uoAXEZFj+QuLyP37G/iyj1C6cSNmUSG2hEQcqamBusfhAKcDwyivg2zGMT8b5Rsl2wJJfpsNW3w8toRESjdsACC2f39i+/eP8JmKHC+mRw+KV6zgwAO/xZmejlVailWeKDHi4qp9jL1SJ7vlrX6wjERWsM176KnfARA/dCg2V+PeIFdEREQk2kWs13LIkCHMnj2bWbNmMWPGDNLS0vjf//1frrvuukgVqVkwS0sZcfVlDC8ffSZNk3P8eJj1fJXb4q+9hp/uqXnzzpPb1rNhONu1w9GmTaSLISIijUzh559z+PlZ9foamjEijVWr666jeMUKSteto3TdutDttpYtsdew96LhcOBs3x7vgQO4e/RsqKLKSWh50f9QurFi78fECRMiWBoRERERqY2IDukeN24c48aNi2QRmh3D7SbpoosoWbMGy+fD8nqxfF7w+QOj/A0CIy8No2KahVlpdsExIzOxCDzeG1g3F4ej0kwCytdfCvyzymcllJWW4XK5MIK3WeVr8NoqjfCsPNfCqMXPYW+r4b2o7tjqDq7t69RFmWrz+nY7iePHH3dYy5/9jIKl/8aw2cFuw8DA9HqgfAPaik1QrKrvdWhkbXCkbWDdLMvvD/wc/Mzt9qqfk2UFRiYGvx/Bz9rvx7IsHCkp5ZuomNhatKjYmMbvx57UkvihQ0k47zxsbnf1b4aIiEStxNHnkfrLu/FnZxPTtx+O1BR8h7Px5x+FUPvFV3X2Y3BmY/nPgfvKZ0RaJv6iIszCIgybQezAgbS8cEqkT1OkWgnnjeK0557Fe/BgYKaU04nN7SZ2wAAMZ83LlnZd+BGmx4M9Ib7hCiu11uqKK3BldKBs2zbsrVvRcvLkSBdJRERERE5A6900M4Zh0PaB+yP2+n6/nzVr1tBzwIAmvRlPY5Q6cwapM2dEuhgiIiI/mi0ujpQbb4x0MUQiwjAMWpx//sk/zuXCrmWaGrWE4eeSMPzcSBdDRERERGrJFukCiIiIiIiIiIiIiIiINCQlR0REREREREREREREJKooOSIiIiIiIiIiIiIiIlFFyREREREREREREREREYkqSo6IiIiIiIiIiIiIiEhUUXJERERERERERERERESiipIjIiIiIiIiIiIiIiISVZQcERERERERERERERGRqKLkiIiIiIiIiIiIiIiIRBUlR0REREREREREREREJKooOSIiIiIiIiIiIiIiIlFFyREREREREREREREREYkqjkgX4FRZlgWA3++PcEmksuDnoc9FpHqKEZHwFCMiJ6Y4EQlPMSISnmJEJDzFiMiJNeY4CZYpmD8Ix7Bqc1Qj5PF4WL9+faSLISIiIiIiIiIiIiIijUi/fv1wuVxhj2myyRHTNPH5fNhsNgzDiHRxREREREREREREREQkgizLwjRNHA4HNlv4XUWabHJERERERERERERERETkVGhDdhERERERERERERERiSpKjoiIiIiIiIiIiIiISFRRckRERERERERERERERKKKkiMiIiIiIiIiIiIiIhJVlBwREREREREREREREZGoouSIiIiIiIiIiIiIiIhEFSVHREREREREREREREQkqig5IiIiIiIiIiIiIiIiUUXJERGROuTz+SJdBJFGLScnBwDTNCNcEhERERERiTYrVqxg1apVkS6GiDQSSo7ICamzV+TEdu3axd13383KlSsjXRSRRmnfvn3MnDmT66+/HsuysNnUBBGpzOPxRLoIIk2CZVmRLoJIo5SVlcWSJUvYvHmzruFFqrF3716mTZvGjTfeyKFDhyJdHJFG68svv+SFF14AomNQoyPSBZDGyzRNZs+ezf79+0lPT+fcc89l8ODBkS6WSKNimiYPP/ww7733HoMGDSIjIyPSRRJpVIIx8s9//hOAc845J3S7EiQigVh47rnn2LVrFwkJCZx99tmMGTOGxMRExYlIuaNHj/L3v/+dmTNnYhhGpIsj0qhYlsWzzz7Lm2++SWpqKvv27WPo0KHcfffd9O7dO9LFE4m44PXIu+++y8iRI3n99dfp169fpIsl0ujs2bOHhx56iG+++YbY2Fiuu+464uLiIl2seqerLanWunXrmDBhAp9//jk+n48FCxYwffp01q9fH+miiTQaH3zwAWeddRZr167l+eef5y9/+Qvt2rWLdLFEGo1//OMfDBo0iBUrVvD8888zYcIEEhMTMQxDHb4iwMaNG5kwYQJffvklXbp0obi4mCeffJL777+f7OxsbDabRsmLAMuWLeOFF17gs88+A6JjFKNIbc2dO5eFCxfy9NNP8+KLL/Lyyy+zfPly5syZo1mJEvVycnK45pprmDNnDq+//jrPPPMMZ5xxhhLtIpWYpsmDDz7I+PHjSUpKYtiwYQwYMIC4uLioaHOpZ0KqtWjRIlJTU3nhhRf43e9+x+uvv05KSgp///vfAV2QiOzYsYPnnnuOM844g7/+9a+MGzeO2NjYSBdLJOIsyyIzM5OxY8cya9YsZsyYwZtvvsmECRPYtWsXrVu3BsDv90e4pCKR9+GHHwLw9NNPc9ttt/H8889z1113sWrVKv7whz9EuHQikRe85ti+fTsAf/rTnzSjSqQSr9fL/Pnz6d69O8OHD6d79+6ce+65jB49mg0bNuByuSJdRJGIstlstG3bljPOOIOePXuSmJgIQGZmJocPH1bflkS9d955hwEDBrBq1SpmzZrFE088QVxcHA6HA7/fHxVtruZ/hlJrwZGJHo+HDz/8kD59+nDaaafhdDpJSkoiNTWV/Px8gKgIDpHqBBtP3bp1o2fPnjgcjlA8HD58mH/84x8sXbqUPXv2hB6jUb8STQzDICUlhSlTpvDuu+9y/fXXk5ycTGlpKbGxsZSWlgJgt9sjXFKRyAjWCYWFhSxYsIDBgwdz+umnh0YwXnbZZbRq1Yp58+bx3XffYRiGLtwlatlsNg4ePMj8+fPp3Lkz+/fv55VXXgE0WEui07EzQZxOJzt37iQjI4OYmBgASkpKyMrKolevXuTk5IQeo2sSiQbHxkhSUhK9e/emoKCAXbt2kZ+fz/Tp05k6dSqXXXYZd955J//5z38AxYhEny1btrBkyRJuu+023njjDcaPH09MTAxFRUW43W7sdntUtLfUwy18/fXX/PDDD+Tm5gJw5MgRUlNTSU1NDR1z5MgRvv/+e9LT0/n000/Jy8uLUGlFIiMYJ5W/+5deeinfffcdWVlZvPrqq4waNYo///nPzJw5k6lTp4Y2sNKUXYkGwRg5fPgwLpeLmTNn0rFjx9D9MTExFBcXh5Iimjki0ebY9lZBQUEoLiCQMPR6vRQWFtKiRQuSkpJ4/PHHAQ1KkehQ0/I/y5YtIyYmhkcffZRBgwbxzjvvcPDgQS07J1Fn1qxZTJ8+nfvuu48FCxYA4PP5uO+++5g4cSIQ6Nz94x//yLp16/j222+ZNGkSTz75JB6PB8MwFDPSrB0bI8Hv+6RJkzhw4ADz58/n1ltvxefzcfPNNzNhwgQ2bdrEzJkz2bt3r67bJSpUbm/17NmT5557jhtvvJHk5OTQ/TabDa/XGzUzR7QhexRbuHAhzz77LIZhkJ2dTZs2bXj88ccZNGgQL774IsnJyViWxb59+5g4cSKnnXYaK1as4O2332bAgAHcddddnH322ViWpUpEmq3q4uSxxx7jrLPOYtiwYWRkZPDwww+TmprKM888Q8+ePXE6nTz22GO88cYbdOnShUmTJilOpNmqKUYGDhxYpePKsiw6derE7t27Ac0ckegRrr3Vu3dv1q1bx5IlSxg/fjxOp5PXXnsttAHiyy+/zOLFi5k4caLqEWnWZs2axdq1a2nTpg3Dhg3jggsuAAIX6Nu2bWPSpEkMGjSIrKws1q5dywsvvMCjjz6qmJCokJWVxd13301ubi5nnXUW27dvZ968eWzevJnrr7+eSZMmhUb27t+/n927d/P444/Tpk0bPvnkE+bMmYNhGPy///f/VJdIs1RTjGzZsoWrrrqKtLQ0xo0bx5tvvsnIkSN5/PHHSUlJAWD48OH85je/4U9/+hNPPvmklm6UZq1ye2vo0KFMnjyZli1bhu43TROXy4XD4aC0tDQ0c6S5x0TzPjup0SeffMLvfvc7LrjgAn7/+9/z7LPPEh8fz0MPPcTmzZtp164dsbGxGIZBRkYGDzzwAH/729949913eeWVVzhw4ADvvvsufr9fjStptmqKk0ceeYTt27fjdDo599xzWb16NW3atGHcuHF07tyZjIwM7rzzTjp27MjixYsVJ9JshYuRrVu3AoGZU5U3YPf5fBQUFESy2CINpqYY+e1vf8uBAweYMWMGSUlJ3H777Vx22WWMHDmSv/zlL9xyyy1MnDiRpKQkVqxYoc4sabaysrK44oorWLRoEW3btuX777/nV7/6Fc888wyHDh3C5XJx9dVXc9tttwEwevRohg8fzpIlS1i1ahWg5bWk+VuzZg379u3jqaee4pFHHuHdd99l+vTpLFiwgJdeegkIzDD0+Xy0b9+exx57jAsvvJDhw4dz7733csEFF/Dxxx9z5MiRZt/BJdGpphj58MMPQ0sxjhkzho4dOzJlypRQYgSgX79+nHfeeXzzzTfk5OQoRqRZqq69dc899/D73/+eI0eOAFTpt+ratSuZmZkUFhZGRUxo5kgUsiyLOXPm0K1bN2655ZbQ2qRJSUlcfvnl7N69m169emGaJqZp4nA4uOyyy0KjfIcPH86gQYPYsWMHJSUlxMfH64Jdmp0TxckPP/xAz549OfPMM3E4HFx++eU4nU4gcJHeq1cvOnXqxA8//IDP56vSOSzSHNS2LrEsC9M0sdvtDB8+nCeeeCLUkaUOX2nOThQjq1evZtKkSfzud79j+fLlbNmyhZEjR/I///M/pKWlAZCamsr+/ftD+46oHpHmJtihNXv2bPr27Yvdbuf5559n7ty5eDwe7r//fjp06AAELtrj4uKYPHkya9as4YUXXuCvf/2r4kKavaVLl9K2bVv69+8fmpE7Y8YMsrKyWLhwIcOHD2f48OGhNlWw49eyLOLj40lPTw+tHS/SHIWLkUWLFjFmzBjGjRtHp06d6NKlCxC4ZjcMg4SEBJKSkoiJiQm11USamxO1t37zm99gt9tD8RMbG0txcTFZWVl069YtwqWvf2pJRqHc3Fz2799P586dQ3/8TdOkRYsWuN1udu3aBQRGnzgcVfNnPp8PCARKQUEBbrdbHVvSLJ0oTnbu3AnAhAkTuOeee+jQocNxa2XHxMSQn5+Py+XShbs0O7WtSwzDCF2Mt23blri4OD755JNIFVukwdQ2Rrp3785VV13F448/zvTp00lLS8OyLCzLwuFwhPYkUT0izVHlDq3gd3zGjBkMGTKERYsW8eWXXwJUWfP67LPPZuzYsWzYsIEPPvgA0OwRaZ6C3+v09HQyMzND+4Z4vV4cDgdTp06lTZs2vPzyy0BgydLDhw9z4MABIJAcyc/PZ+3atXTr1o24uLiInYtIfahNjKSlpfGnP/0JCLS5jhw5wr59+0J1SjBGOnXqFBrsKNLc1NTeGjp0KAsXLgy1t7xeLwDjx48nOzubo0ePAs1/v1BdZUWhVq1akZKSQnx8PIWFhUDFBXdZWRk9e/YMHev1elmzZk0oUAC2bNnCihUrmDZtmioPabZOFCd9+vQBAhchPp+P5cuXs3TpUiDQSNuyZQvfffcdl156qRKI0iydTF0SHIHSuXNnkpOT2bx5M6WlpYoNadZOFCO9e/cGAvHh9XrZsGEDb7zxBhC4ANm4cSN79uzh5z//eWROQKQenahD67LLLjuu0zc4gwrgggsuoGfPnrz66quUlZUpeSjNUvB73b59exwOBx999BFQsW/bgAEDGD16NLt372bhwoUALFiwgBtvvJFFixbx7bff8uqrr7Jp0yYuvfRSXC5XZE5EpJ6cSoy8//77XHfddcydO5cvv/ySV155hY0bN3LxxRerf0uanZNNsgfriYSEBLp37867774LNP/9QtWKjDLBbN9vf/tbLr74YhISEkL3/ec//wGgY8eOoY4sp9PJnDlzuPXWW7nqqqt48sknuf3220lISGDMmDENfwIiDeBk48TlcvHBBx9w5513cumll/Lwww+H4mTs2LENfwIi9ay2MRJkGAaWZdG+fXv69+/Pxx9/zJYtWxq20CIN6GTqEcMwcDqd/PDDDzzxxBP87Gc/46GHHuKXv/wlnTt35pxzzonIOYjUp5Pp0AreZ1lW6HHdunVj9OjR7Nixg8WLF0fgDETqX/Ba45xzziEuLo6lS5eSm5uLzWYLje49//zziY+PD+3Bc/rpp+N2u3nwwQe5//77+fjjj3n00Ud1TSLN0qnESP/+/WnRogWPPfYYDz74IJ988gmPPfYY48aNi9h5iNSXU2lvQWB5xgEDBrB9+3a2b98OVMRbc6Q9R5oZv98fNqMXvK9r166h24IX5suWLaN79+5kZGSEOrIMw+Dmm2+mV69erFixgl27dnHJJZdw44031vu5iNSX+oiTW2+9lb59+/Ldd9+RmZnJpZdeyg033FDv5yJSH+oqRioL3n/dddexcOHCZj81V5q3uqxHgiZMmEBKSgpLlixh3759XHTRRdx00031dxIiERSMh3POOYf/+7//Y+nSpYwaNYrk5GS8Xi9Op5Pzzz+fhQsXsnr1asaPH19lbzebzcaUKVP4yU9+EpqFJdKUlJWV8c9//hOXy0WHDh3o06cPLVu2rFK/BGdLZWRkMH78eN5//30WLVrEFVdcgdPpxO/307VrV7p3787GjRsBGDZsGIMHD+bo0aNkZmbSv3//SJ6myCmrrxgZOnQogwYNIjc3l6ysLMWINGun2t5yuVwMHDiQ9957j+XLl9O9e/dmveqDkiPNyObNm5k9ezZXXXUVQ4YMqXHjzmNvNwyDgwcP8u233zJz5szQPiOGYZCXl0d6ejrTpk3jiiuuwOfzaaqhNGn1HSeXX345fr9fcSJNVl3HCEBeXh4JCQnYbDa6du3K8uXLcbvdDXI+InWtvmIkLi6OIUOGMGTIEHw+33H7vok0FXXZodWtWzc2b96M0+kMxVQwrpKSkkhKSorgmYqcmvnz5/Pkk0+SlJSE1+vlyJEjDBs2jBdeeCG0IW4wRoLf+xtuuIEvvviCefPmMWDAAHr37h3qqGrdujVbt24NxZjL5SI1NZXU1NQIn6nIqWmIGElLSyMtLS3CZypy6uqrvRW8DpkyZQoxMTFMnDgxwmda/7SsVjMQXENu165dfPHFFyxduhSv14vNZqsy7cmyrCqbGZaVlYXu+/TTT/F4PKHptkeOHGHWrFncdtttbNiwASC07INIU9RQcWKz2RQn0iTVd4wER2sBSoxIk1TfMbJ58+bQcUqMSFM1f/58Ro0axdtvv82rr77KjBkzuP/++wFCHVpAqEML4IYbbiA1NZV58+axadMmgCodWtnZ2VViSqQp27x5My+99BJXX301r732GvPmzWPGjBksW7aMV155BQjMTgx+5x0OB59++ik//PAD9913HwUFBTz++ON4PB5KS0vZs2cP3333HZMmTWr2a8JLdFCMiJxYfba3HA5H6PHRkBgBJUeaheCFwtq1a4/bGDrINE0Mw8But7Nr1y5uvfVWXn/99dD933zzDR07diQ1NZV//etfXHHFFbz++utMmTKFAQMGNOTpiNQLxYlIePUdI5qyLk2dYkQkPHVoiZzYwoULKSoq4oILLqBdu3YkJiYyefJkOnXqxPLly0Ob5AbrkWuuuYaZM2eycuVKBg8ezIwZM9izZw/Dhw/n5ptv5sYbb8QwDCZMmBDpUxOpE4oRkfAaor3VnJfQqo6SI82AZVns3buXTz/9lClTpnD06FE++ugjcnJyQlOobDYbPp+Phx56iIkTJ+L1ehk/fjwAhYWF7N+/H5fLxb333st9993Heeedx8qVK7nooosifHYidUNxIhKeYkQkPMWISHjq0BKpWXCvtaNHj5KYmBjam83r9ZKWlobD4cDlcoWWkHvrrbc4//zziYmJYd68eVx99dUATJ48mX/84x9cffXVtG/fnilTprBgwQJ69uwZsXMTqQuKEZHaUXur7mnOfhNiWRa7du3i2Wef5cILL2Ts2LGhC/FPP/2U0tJSHn74Yf7yl7/w9ttvs2jRIqZNmxYa6XjvvfeyYcMGXnzxRUaMGBFa+ichIYHMzEzy8/Pp0KEDn332GW3bto3kqYqcMsWJSHiKEZHwFCMiJye4tnVtO7TeeecdHn30UUaMGMG8efPo0aMHEOjQ6t+/Px9++CH79u1j6NChTJ8+PZKnJnJKqqtHgqNwhw0bRl5eHtnZ2aSkpOB0OiksLOTo0aP069cPCMxUTE9P57XXXmPQoEG4XK4qz92hQwdmzJhR455XIo2dYkTk5Km9VX+UHGkightOHThwgM8++4yWLVsyZMgQ4uPjAUhLS+NXv/oVbrebm266iSVLlrBo0SLOOeccunbtCsD999+Py+UiISEh9LymaeL3+7nrrrvo0aMHZ511VkTOT6QuKE5EwlOMiISnGBEJTx1aIuGdqB4ZNmwYgwcPplWrVqHHbNu2jcOHDzNo0KDQbeedd161z195qRPFiDRFihGRE1N7q2HpHWgigkGwZs0afD4fK1asYMmSJaH7J02axIUXXojX6yUmJoYbbriBbdu28eGHH4aOadWqVZULdajYPPryyy/Xhbo0eYoTkfAUIyLhKUZEanZsh9bnn39OUVFR6KJ62LBhPPTQQ6SkpIQeU1OH1tChQ6tcqIM6tKR5OFE9kpCQEOr09Xq9AHzyySfExMRwxhlnHPd8wU1xRZoLxYhIeGpvNTy9C02EZVlkZ2ezYMECfvrTn+LxeFi0aBGZmZlAxQagwaUbpkyZQp8+ffj000/59ttvQ8eINGeKE5HwFCMi4SlGRGqmDi2RE6tNPRIUrEuWLl3KkCFD6NSpU+i+rVu3AtG3Ka40f4oRkfDU3mp4So40EYZhsGzZMo4cOcIDDzzANddcw9q1a1m4cCFQNdsX3Mhq5syZHD58mI8++ojS0lJsNpuCQpo1xYlIeIoRkfAUIyI1U4eWyImdTD0CsHr1anbv3s3EiRMByMzM5Omnn+bCCy/kvffea/Dyi9Q3xYhIeGpvNTwlR5qQlJQU7rrrLpKTk7nmmmvo0KEDixcvZsOGDUBFgNjtdgAGDhzIhAkT+Pjjj/noo48ABYU0f4oTkfAUIyLhKUZEqqcOLZHaqW09AvD111/jcrlo06YN7733HpdccgmLFy/mhRde4JJLLonUKYjUK8WISM3U3mp4So40ISNHjuTSSy/F5/MBcMstt7Bnzx4WLFgQ2kQnOFIxWJncfPPNtG7duspadCLNmeJEJDzFiEh4ihGRmqlDS+TEalOPBO8rLi7Gsizuv/9+Hn74Ya666iqWLl3K2LFjI3kKIvVKMSISntpbDcsR6QJEu0WLFhEXF0dqaiq9e/cGKjbfqY7NZgtlCceMGcPZZ5/N0qVLOeeccxg1alSV4/x+P6eddhrz58/H4dBHLU2X4kQkPMWISHiKEZG6MXLkSEzTxOfz4XA4uOWWW7jvvvtYsGABvXv3DnVoORyOKh1a+/fvZ+bMmdxyyy2RPgWRU1LX9YjdbseyLHJzc/F6vZx99tnce++9xMfHN9g5idQlxYhI3VF7q2HpCi5CVq5cyX333UdpaSl2u52srCzuuOMOrrzyShISEkKViGVZWJZVZdqUaZpYloXdbue2227j2muv5aOPPuKss86iRYsWoUx7cLkHXahLU6U4EQlPMSISnmJE5MTUoSVSs/qsRwCmTp3KzTffXGWdeJGmRDEiUjtqbzVeWlYrAoqKivjjH//IwIEDefPNN3n77be55JJLeOWVV1i1ahUQWGPO7/djGAY2m42tW7fy1ltvAYQuxC3L4vTTT2fKlCl89tln/Otf/wrdL9LUKU5EwlOMiISnGBEJb+XKlYwfP54nn3yShx56iIsuuoiXXnqJwsLCUEcWBC7cKy/fAIEOLb/fD8Btt91GUVERH330Efn5+RiGgWEYTJ06lcWLF/PII4/oQl2apPquRwAGDBigTl9pshQjIiem9lbjp6u6CFi1ahXr169n3LhxdOjQgbZt2zJjxgyKioqqZAztdjslJSX85je/4cILL2TdunV4PJ7Q/cEAuummm0hLSyM1NbXBz0WkvihORMJTjIiEpxgRqZk6tEROTPWISHiKEZHw1N5qGpQcaQB5eXmhzaSCv5eWloamUUEgk5iSkoJhGOTk5ACwfPlyRo4cyYYNG3j55Zd57LHHcLlcoccE17lu2bIl8+bN4/zzz2+4kxKpY4oTkfAUIyLhKUZEak8dWiLHUz0iEp5iROTkqL3VNBhW8B2WOrd8+XJmz54dWkfxmmuuYcyYMRQXF/PVV19x3nnnER8fz7p167jxxhvx+Xy43W4SEhK4++676devH2vXrmXEiBEkJCRE+nRE6oXiRCQ8xYhIeIoRkRPLy8sjISEhtDfOhx9+yD333MPnn39O27ZtAVi4cCFPPPEETz31FL1796ZVq1YsX76c22+/nbZt2/KrX/2KoUOHHre/jt/vx2634/F4qnR2iTQVqkdEwlOMiNSO2ltNk5Ij9cA0TV555RXeeOMNxowZQ3p6Op999hl79+5l2rRpzJw5s8qmO2+++SaZmZmMGTOGzMxM/vWvf/Htt9+yaNEiMjIywm7QI9JUKU5EwlOMiISnGBE5MXVoidRM9YhIeIoRkdpRe6tpU3KkHuTk5HDNNdcwevRobr31VtxuNyUlJdxxxx3s3r2bN954g7S0tFDWr6SkBNM0QxvnrFq1iunTpzN16lTuvvvuCJ+NSP1QnIiEpxgRCU8xIlIzdWiJnJjqEZHwFCMi4am91Txoz5E6FMwzbdy4ke3bt3PxxRfjdrsxTZPY2Fj69u1Lbm4upmkCgTXlAGJjY4mPj8fv9wPQo0cPnE4nsbGxkTkRkXqkOBEJTzEiEp5iROTE8vLyWLBgARdffDH3338/06dP5/XXX6dfv358+OGHHDx4MLQBKMBFF13EjBkzGDhwIJMnT2bGjBnEx8fz3nvvAehCXZoV1SMi4SlGRGpH7a3mQcmRH2nnzp2sW7eOnJyc0JfY4XCQnJzMrl27AEIVRjAYbLaKt93j8bBlyxYgUKFYlsW3336L1+ulb9++DXgmIvVHcSISnmJEJDzFiEjtqENLpHqqR0TCU4yI1J7aW82L48SHSHUKCwt58MEH+fLLL7Hb7TgcDq699lquvfZaevXqxS9/+UtOO+00oCJoVq9eTZcuXUhOTg5Nldq/fz8PPfQQGRkZDB06lJiYGGbPns2ZZ55J//79I3mKIj+a4kQkPMWISHiKEZET27lzJ4WFhbRv355WrVoBVTu0MjIyME0Tm81WY4fWzp076dmzpzq0pNlRPSISnmJEpHbU3mq+lBw5RbNnz2bbtm08++yzlJaWsn79ep5++mmOHj3KzTffzMUXXxw61ul0kpeXx5YtW7jqqqtwuVyhSqV169ZMnDiR1157jdWrV2OaJsOHD+eBBx7A6XRG6vRE6oTiRCQ8xYhIeIoRkZqpQ0vkxFSPiISnGBEJT+2t5k/JkZNkmiaHDx9m0aJFTJs2jaFDh2Kz2Rg7diz5+fnMmzeP5ORkfvGLX4QyhgBLly6loKCAc889F6hYRy4hIYFf/OIXTJkyhfz8fOLi4khJSYnY+YnUBcWJSHiKEZHwFCMiJ6YOLZGaqR4RCU8xIlI7am81f0qO1EJmZibZ2dm0a9eOlJQU4uLiyM/Pp3PnzthsNsrKynC73dx5551s27aN+fPnM2zYMLp3747H48HlcrFgwQI6d+5M9+7dgUDm8cMPPyQ7O5vbbruNpKQkkpKSInuiIj+C4kQkPMWISHiKEZHaUYeWSPVUj4iEpxgRqT21t6KHNmQPo7S0lPvvv58rrriC22+/ncmTJ/PSSy+RmZlJ165d+eKLLwBwu934fD5atmzJ5ZdfTm5uLnPnzgXA5XJRUFDAmjVrGDt2LAkJCXz11Vdcd911PP7449p0R5o8xYlIeIoRkfAUIyInlpmZybp168jOzsZms1XboQVw5513kpGRwfz589m+fTs2mw2PxwNQbYfW22+/zYsvvghAUlISHTp00IW6NDmqR0TCU4yI1I7aW9FJM0dqkJubywMPPMCBAwd44IEH8Pv9rF+/nj/+8Y+0bduW1NRUNm3axIYNG+jbt28oE3jBBRfwwQcfsHLlSnbu3EmXLl3YunUrxcXFJCYmcs899/Dhhx/y05/+lL/+9a8kJCRE+ExFTp3iRCQ8xYhIeIoRkfBKS0t59NFH+frrrwEoKyvjmmuu4bzzzgt1aI0fP/64Dq0//OEPzJ07l3vvvbdKh9aVV14Z6tCaNWsWmzZt4s4774zsSYr8CKpHRMJTjIicmNpb0U3JkRps2bKF//73vzzzzDOMHDkSgHHjxrFq1SqWLl3KDTfcwJVXXsnnn3/O6aefjsvlwuv14nQ6mTp1KnfccQe5ubkA5OXlAfCHP/yB3r17M2fOHPr27RupUxOpM4oTkfAUIyLhKUZEaqYOLZETUz0iEp5iRCQ8tbdEy2rVYP369bhcrtA0KADLsujYsSOrVq1i0KBBjBkzhvfff59vv/0WILS+3IgRI0hISGDnzp0AtGzZkg4dOvDiiy8yd+5cVR7SbChORMJTjIiEpxgRqVmwQ+uOO+5g7NixTJgwgbvvvpuzzjor1KG1adMmPv/8czweD3a7Ha/XC8DUqVPZuHFjtR1a33//PXPmzOEPf/iDLtSlyVM9IhKeYkQkPLW3RMmRGmRkZNCyZUvy8/OxLAvLsrDb7RQUFNC6dWsA7rvvPkpLS3nrrbfYs2cPdrsdgP/85z+UlJTQpk0bAAYPHsySJUsYM2ZMxM5HpD4oTkTCU4yIhKcYEamZOrRETkz1iEh4ihGR8NTeEi2rVYOJEyeSlJRE165dQ1Om8vLyWLNmDaNHjwagffv23HHHHbz22mv8+te/5pZbbiEmJoZ58+bRq1cv+vXrF8lTEKl3ihOR8BQjIuEpRkRqVrlDKz09HaDaDq1LLrmEt956i44dO9KhQweg5g4tkeZG9YhIeIoRkfDU3hIlR2pgGAZDhgwBAhlDwzDYvn072dnZnHvuuaHjLrvsMtLS0njqqae46667sNlstGvXjqeeeopWrVpFqvgiDUJxIhKeYkQkPMWISM3UoSVyYqpHRMJTjIiEp/aWKDlSC8HgWLp0KS1atAhtYgXg8/kYOXIkP/nJT8jKyiIvL4+BAwdGqqgiEaM4EQlPMSISnmJEpCp1aImcHNUjIuEpRkSOp/aWKDlSS4cPH+bjjz9m+PDhuN1uPB4PW7ZsYfbs2QwZMoQrr7ySrl27RrqYIhGlOBEJTzEiEp5iRKR66tASqR3VIyLhKUZEaqb2VnTShuy1tGvXLjIzMxk5ciSHDh3iqaee4sorryQ/P5+f/vSnuFyuSBdRJOIUJyLhKUZEwlOMiNSsug6tdevWMX36dP72t7/hdDrp2rWrLtQlqqkeEQlPMSISntpb0UczR2rp+++/xzAMVq1axbPPPovdbueFF15gxIgRkS6aSKOhOBEJTzEiEp5iRKRmwQ6tO++8k0OHDvHSSy8xZ84cevXqpQ4tkXKqR0TCU4yIhKf2VvRRcqSW7HY7lmWxcOFCbrrpJm644YZIF0mk0VGciISnGBEJTzEiUjN1aImcmOoRkfAUIyLhqb0VfZQcqaXOnTtz991384tf/EJZQpEaKE5EwlOMiISnGBGpmTq0RE5M9YhIeIoRkfDU3oo+hmVZVqQL0RRYlhXamEdEqqc4EQlPMSISnmJEpGYrVqxg1apV6tASCUP1iEh4ihGR8NTeij5KjoiIiIiIiDRy6tASERERqV9qb0UfJUdERERERERERERERCSq2CJdABERERERERERERERkYak5IiIiIiIiIiIiIiIiEQVJUdERERERERERERERCSqKDkiIiIiIiIiIiIiIiJRRckRERERERERERERERGJKkqOiIiIiIiIiIiIiIhIVFFyREREREREIq6srIysrCwA5s6dS48ePXjvvfciXCoREREREWmulBwREREREZGIWr9+PRMmTOCbb74BYPDgwTz99NOcffbZES6ZiIiIiIg0V45IF0BERERERKLbli1bOHDgQOj3jIwMMjIyIlgiERERERFp7jRzREREREREREREREREooqSIyIiIiIiEjG//vWveeCBBwC477776NGjR7V7jvTo0YNHHnmEBQsWcOGFF9KvXz9GjhzJn/70JyzLYvHixUyZMoUzzjiDcePG8dprrx33WgcPHuS3v/0tI0aMoG/fvpx33nk89thj5ObmNtj5ioiIiIhI46BltUREREREJGKmTp2K3W5nzpw5TJ06lYEDB+L3+6s9dunSpSxYsIArr7ySSy+9lHfeeYfnn3+edevWsXr16tDtb7/9Nk8//TTp6elMmjQJgL1793L55Zfj8XiYOnUqp512Glu2bOGdd97hiy++4J133qFVq1YNeeoiIiIiIhJBSo6IiIiIiEjEnHnmmezYsYM5c+YwYMAALrzwQubOnVvtsQcOHOCdd97hzDPPDD325z//OcuWLWPOnDn06dMHgEGDBjF58mQ+++yzUHLkkUceoaSkhHnz5tGhQ4fQc44fP55rr72WWbNm8dBDD9XvyYqIiIiISKOhZbVERERERKRJaNeuXSgxAtC1a1cAOnToEEqMAHTq1AkILKMFcPToUb766isGDRpEQkICOTk5oX89e/YkIyODTz75pOFOREREREREIk4zR0REREREpElISUmp8rvdbgcgNTW1yu02W2AMmGVZAOzevRvTNPn8888ZMmRIjc9fVlaG2+2uyyKLiIiIiEgjpeSIiIiIiIg0CQ5H9ZcvhmGEfZxpmgCMHTuWadOm1XhcMNkiIiIiIiLNn5IjIiIiIiLSrLVv3x6A0tJShg4detz9n376KUlJSTUmX0REREREpPnRniMiIiIiIhJRwWWwgjM86lpKSgoDBw7k66+/5r///W+V+5YtW8aMGTN4+eWX6+W1RURERESkcdLQKBERERERiajgXiLz58/HsqzQXiF16cEHH+TKK6/k2muvZerUqZx++uns3LmTd955h6SkJO699946f00REREREWm8NHNEREREREQiaujQoVxwwQWsW7eOJ554ol6SIz169GDu3LlMnjyZjz/+mEcffZQlS5YwceJE3n33Xbp27VrnrykiIiIiIo2XYdXHlYeIiIiIiIiIiIiIiEgjpZkjIiIiIiIiIiIiIiISVZQcERERERERERERERGRqKLkiIiIiIiIiIiIiIiIRBUlR0REREREREREREREJKooOSIiIiIiIiIiIiIiIlFFyREREREREREREREREYkqSo6IiIiIiIiIiIiIiEhUUXJERERERERERERERESiipIjIiIiIiIiIiIiIiISVZQcERERERERERERERGRqKLkiIiIiIiIiIiIiIiIRBUlR0REREREREREREREJKr8fxqqtXnxGFW3AAAAAElFTkSuQmCC", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "fig, ax = plt.subplots(1, 1, figsize=(20,5))\n", + "data_resampled.plot(x=\"time\", ax=ax);" + ] + }, + { + "cell_type": "markdown", + "id": "809202b5-b86b-4f2e-8695-af1167e33c82", + "metadata": {}, + "source": [ + "#### Subplots" + ] + }, + { + "cell_type": "code", + "execution_count": 38, + "id": "dcc24cf2-2d0f-4846-a2df-786c478eb342", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "y_values = [col for col in data_resampled.columns if \"Extruder A\" in col]\n", + "\n", + "rows = len(y_values) \n", + "cols = 1\n", + "\n", + "fig, axes = plt.subplots(rows, cols, figsize=(20, 2*rows), sharex=True)\n", + "\n", + "for i, col in enumerate(y_values):\n", + " data_resampled.plot(x=\"time\", y=col, ax=axes[i])" + ] + }, + { + "cell_type": "markdown", + "id": "5f237b06-8e03-4f46-8f4e-a99486d8aff5", + "metadata": {}, + "source": [ + "#### Pairplots" + ] + }, + { + "cell_type": "code", + "execution_count": 39, + "id": "780a9b10-2ca0-40d3-8d4a-3b7db7715bab", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "sns.set_theme(style=\"ticks\")\n", + "\n", + "columns = [col for col in data_resampled.columns if \"Durchsatz\" in col]\n", + "\n", + "sns.pairplot(data_resampled[columns]);" + ] + }, + { + "cell_type": "markdown", + "id": "40ec502a-9ce0-4fdc-be44-b114a1564580", + "metadata": {}, + "source": [ + "## Production runs\n", + "If your production runs are integrated to the ENLYZE Platform, you can also use the `EnlyzeClient` to query them, alongside the calculated OEE metrics per run.\n", + "You can fetch all runs at once or use filters to fetch runs specific product, orders, machines or within a certain time range." + ] + }, + { + "cell_type": "markdown", + "id": "1b25c5f8-82b6-40b3-aef4-ee5058d61842", + "metadata": {}, + "source": [ + "We will use our previously selected machine from above to filter the runs. Since this is a demo account, the KPIs are not meaningful, but this should give you a good idea of the data and what is possible." + ] + }, + { + "cell_type": "code", + "execution_count": 40, + "id": "1da46a78-75d2-41f2-9447-56c4486d0e4f", + "metadata": {}, + "outputs": [], + "source": [ + "runs = enlyze.get_production_runs(machine=machine)" + ] + }, + { + "cell_type": "markdown", + "id": "8086ac26-d0b9-4ec9-bf9b-6b0e26697725", + "metadata": {}, + "source": [ + "Unlike the time series data or the machines, the runs are returned as a `enlyze.models.ProductionRuns` model. This model as a `to_dataframe` method, which conveniently converts the runs into a pandas DataFrame for us." + ] + }, + { + "cell_type": "code", + "execution_count": 41, + "id": "7bc79a88-14ed-4ee0-8704-0d3b79ce5673", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "enlyze.models.ProductionRuns" + ] + }, + "execution_count": 41, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "type(runs)" + ] + }, + { + "cell_type": "code", + "execution_count": 42, + "id": "ae8eba18-ed77-4594-b722-c295947c9121", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
uuidaverage_throughputproduction_orderstartendquantity_totalquantity_yieldavailabilityperformancequality...quantity_yield.valueavailability.scoreavailability.time_lossperformance.scoreperformance.time_lossquality.scorequality.time_lossproductivity.scoreproductivity.time_lossquantity_scrap
099245861-840e-42f6-9888-494339e88eb4NaNFA-0A5A-12022-03-18 00:13:08+00:002022-03-18 01:16:09+00:00NaNNaNNaNNaNNaN...NaNNaNNaTNaNNaTNaNNaTNaNNaTNaN
13756a10c-b564-42bb-83f1-5f5d6aba6d30230.552952FA-0A5A-22022-03-18 01:16:19+00:002022-03-18 18:41:55+00:00NaNNaNNaNNaNNaN...3236.3194441.0000000 days 00:00:000.3842550 days 08:38:360.8055020 days 03:23:220.3095180 days 12:01:58NaN
21366390c-a1c5-4fc3-a4f5-74ade57e118d459.256323FA-0A5A-32022-03-18 18:42:01+00:002022-03-19 06:38:57+00:00NaNNaNNaNNaNNaN...5107.4527781.0000000 days 00:00:000.9076210 days 01:01:380.9307260 days 00:49:390.8447460 days 01:51:17NaN
398de1da6-69c5-4afa-bd0f-1c1f52440a8b486.231561FA-0A5A-42022-03-19 06:39:04+00:002022-03-20 22:34:24+00:00NaNNaNNaNNaNNaN...18144.6944441.0000000 days 00:00:000.8090380 days 07:07:340.9347420 days 02:36:180.7562410 days 09:43:52NaN
4b9efc595-ceb7-424f-99b0-f47d2d588cfe499.188133FA-0A5A-52022-03-20 22:34:30+00:002022-03-22 15:33:14+00:00NaNNaNNaNNaNNaN...19749.2694440.9991870 days 00:01:590.8183410 days 07:10:520.9654430 days 01:24:530.7894190 days 08:37:44NaN
\n", + "

5 rows × 34 columns

\n", + "
" + ], + "text/plain": [ + " uuid average_throughput production_order \\\n", + "0 99245861-840e-42f6-9888-494339e88eb4 NaN FA-0A5A-1 \n", + "1 3756a10c-b564-42bb-83f1-5f5d6aba6d30 230.552952 FA-0A5A-2 \n", + "2 1366390c-a1c5-4fc3-a4f5-74ade57e118d 459.256323 FA-0A5A-3 \n", + "3 98de1da6-69c5-4afa-bd0f-1c1f52440a8b 486.231561 FA-0A5A-4 \n", + "4 b9efc595-ceb7-424f-99b0-f47d2d588cfe 499.188133 FA-0A5A-5 \n", + "\n", + " start end quantity_total \\\n", + "0 2022-03-18 00:13:08+00:00 2022-03-18 01:16:09+00:00 NaN \n", + "1 2022-03-18 01:16:19+00:00 2022-03-18 18:41:55+00:00 NaN \n", + "2 2022-03-18 18:42:01+00:00 2022-03-19 06:38:57+00:00 NaN \n", + "3 2022-03-19 06:39:04+00:00 2022-03-20 22:34:24+00:00 NaN \n", + "4 2022-03-20 22:34:30+00:00 2022-03-22 15:33:14+00:00 NaN \n", + "\n", + " quantity_yield availability performance quality ... \\\n", + "0 NaN NaN NaN NaN ... \n", + "1 NaN NaN NaN NaN ... \n", + "2 NaN NaN NaN NaN ... \n", + "3 NaN NaN NaN NaN ... \n", + "4 NaN NaN NaN NaN ... \n", + "\n", + " quantity_yield.value availability.score availability.time_loss \\\n", + "0 NaN NaN NaT \n", + "1 3236.319444 1.000000 0 days 00:00:00 \n", + "2 5107.452778 1.000000 0 days 00:00:00 \n", + "3 18144.694444 1.000000 0 days 00:00:00 \n", + "4 19749.269444 0.999187 0 days 00:01:59 \n", + "\n", + " performance.score performance.time_loss quality.score quality.time_loss \\\n", + "0 NaN NaT NaN NaT \n", + "1 0.384255 0 days 08:38:36 0.805502 0 days 03:23:22 \n", + "2 0.907621 0 days 01:01:38 0.930726 0 days 00:49:39 \n", + "3 0.809038 0 days 07:07:34 0.934742 0 days 02:36:18 \n", + "4 0.818341 0 days 07:10:52 0.965443 0 days 01:24:53 \n", + "\n", + " productivity.score productivity.time_loss quantity_scrap \n", + "0 NaN NaT NaN \n", + "1 0.309518 0 days 12:01:58 NaN \n", + "2 0.844746 0 days 01:51:17 NaN \n", + "3 0.756241 0 days 09:43:52 NaN \n", + "4 0.789419 0 days 08:37:44 NaN \n", + "\n", + "[5 rows x 34 columns]" + ] + }, + "execution_count": 42, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "runs_df = runs.to_dataframe()\n", + "runs_df.head()" + ] + }, + { + "cell_type": "markdown", + "id": "bdab41a4-3558-49a9-8a0b-613d828ddb7f", + "metadata": {}, + "source": [ + "## Exporting Data" + ] + }, + { + "cell_type": "markdown", + "id": "352d7f7d-e41e-4a3d-8b47-33bcde4bcd99", + "metadata": {}, + "source": [ + "Pandas DataFrames can be exported to many different data formats. For a full list of supported methods, you can check out the [IO tools](https://pandas.pydata.org/docs/user_guide/io.html#io) section of the pandas documentation or the [Importing and exporting data](https://pandas.pydata.org/docs/user_guide/10min.html#importing-and-exporting-data) part of the **10 minutes to pandas** guide." + ] + }, + { + "cell_type": "markdown", + "id": "b7648dcc-ec46-4f1f-80c8-43c6b5ea2ae4", + "metadata": {}, + "source": [ + "### To CSV\n", + "\n", + "Probably the most used method and quite straightforward:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9afb2b53-0528-4f42-bba7-dfd2b44a357b", + "metadata": {}, + "outputs": [], + "source": [ + "# path to directory with filename\n", + "runs_df.to_csv(\"./filename.csv\")" + ] + }, + { + "cell_type": "markdown", + "id": "1c772885-39dc-4be7-83b5-628357069ed3", + "metadata": {}, + "source": [ + "If a different separator or decimal character is required, they can be specified in the export method.\n", + "\n", + "A full, detailed list of all the options can be found [here](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.to_csv.html). Specifying the separator and the decimal character is especially useful when opening the file in German Excel." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "bce5ebd8-9e1b-4cdf-8771-741ecfb8a32e", + "metadata": {}, + "outputs": [], + "source": [ + "runs_df.to_csv(\"./filename.csv\", sep=\";\", decimal=\",\", index=False)" + ] + }, + { + "cell_type": "markdown", + "id": "505dbb46-cf71-4126-9ea7-2e2f7ef056a6", + "metadata": {}, + "source": [ + "### Excel and Power BI" + ] + }, + { + "cell_type": "markdown", + "id": "f614084c-99c0-42d7-b1d1-b2a622306f66", + "metadata": {}, + "source": [ + "
\n", + "Note: Excel and Power BI don't support all Python data types.\n", + "
" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "84ec8124-28ef-4488-aef5-620c1e38868a", + "metadata": {}, + "source": [ + "There is an option to directly save the data as an Excel file. This requires the additional `openpyxl` package, which can be easily installed via pip:\n", + "\n", + " pip install openpyxl\n", + "\n", + "If you use the [Python Scripts](https://learn.microsoft.com/de-de/power-bi/connect-data/desktop-python-scripts) data connector in Power BI to load data into Power BI, you can directly use the API but need to convert the unsupported data types before returning the DataFrame.\n", + "\n", + "To avoid errors or false data representations, some Python data types need to be converted to supported types. Excel and Power BI don't support timestamps with time zone information and timedeltas. You can use the `DataFrame.info()` method, to see which data types your DataFrame contains." + ] + }, + { + "cell_type": "code", + "execution_count": 43, + "id": "71534943-20b0-42bd-9d9a-ded3d4e129f4", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 227 entries, 0 to 226\n", + "Data columns (total 34 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 uuid 227 non-null object \n", + " 1 average_throughput 217 non-null float64 \n", + " 2 production_order 227 non-null object \n", + " 3 start 227 non-null datetime64[ns, UTC]\n", + " 4 end 226 non-null datetime64[ns, UTC]\n", + " 5 quantity_total 0 non-null float64 \n", + " 6 quantity_yield 0 non-null float64 \n", + " 7 availability 0 non-null float64 \n", + " 8 performance 0 non-null float64 \n", + " 9 quality 0 non-null float64 \n", + " 10 productivity 0 non-null float64 \n", + " 11 machine.uuid 227 non-null object \n", + " 12 machine.display_name 227 non-null object \n", + " 13 machine.genesis_date 227 non-null object \n", + " 14 machine.site._id 227 non-null int64 \n", + " 15 machine.site.display_name 227 non-null object \n", + " 16 machine.site.address 227 non-null object \n", + " 17 product.code 227 non-null object \n", + " 18 product.name 0 non-null object \n", + " 19 quantity_scrap.unit 213 non-null object \n", + " 20 quantity_scrap.value 213 non-null float64 \n", + " 21 quantity_total.unit 221 non-null object \n", + " 22 quantity_total.value 221 non-null float64 \n", + " 23 quantity_yield.unit 208 non-null object \n", + " 24 quantity_yield.value 208 non-null float64 \n", + " 25 availability.score 217 non-null float64 \n", + " 26 availability.time_loss 217 non-null timedelta64[ns] \n", + " 27 performance.score 217 non-null float64 \n", + " 28 performance.time_loss 217 non-null timedelta64[ns] \n", + " 29 quality.score 204 non-null float64 \n", + " 30 quality.time_loss 204 non-null timedelta64[ns] \n", + " 31 productivity.score 204 non-null float64 \n", + " 32 productivity.time_loss 204 non-null timedelta64[ns] \n", + " 33 quantity_scrap 0 non-null float64 \n", + "dtypes: datetime64[ns, UTC](2), float64(15), int64(1), object(12), timedelta64[ns](4)\n", + "memory usage: 60.4+ KB\n" + ] + } + ], + "source": [ + "runs_df.info()" + ] + }, + { + "cell_type": "markdown", + "id": "0806290e-36d6-49d9-867d-c7ee6a9ed87f", + "metadata": {}, + "source": [ + "**Removing timezone information:**\n", + "\n", + "To remove the time zone information, we will first convert the timestamp to the desired time zone (if it's UTC, than this can be skipped) and then remove the timezone information, making them timezone naive." + ] + }, + { + "cell_type": "code", + "execution_count": 44, + "id": "514652f5-283e-46ef-9150-f159ebdee339", + "metadata": {}, + "outputs": [], + "source": [ + "# convert timestamps to local time zone\n", + "runs_df[\"start\"] = runs_df[\"start\"].dt.tz_convert(\"Europe/Berlin\")\n", + "runs_df[\"end\"] = runs_df[\"end\"].dt.tz_convert(\"Europe/Berlin\")\n", + "\n", + "# remove time zone information\n", + "runs_df[\"start\"] = runs_df[\"start\"].dt.tz_localize(None)\n", + "runs_df[\"end\"] = runs_df[\"end\"].dt.tz_localize(None)" + ] + }, + { + "cell_type": "markdown", + "id": "add2d1cf-6f83-432c-9b76-9bedea7b97c3", + "metadata": {}, + "source": [ + "**Convert timedeltas into total_seconds:**\n", + "\n", + "The timedelta datatype is best transformed into the total seconds equivalent of the duration. From there, it can be converted to other proprietary data types." + ] + }, + { + "cell_type": "code", + "execution_count": 45, + "id": "dceb8113-5985-4861-9560-24c34e1bf6c6", + "metadata": {}, + "outputs": [], + "source": [ + "runs_df[\"availability.time_loss\"] = runs_df[\"availability.time_loss\"].dt.total_seconds()\n", + "runs_df[\"performance.time_loss\"] = runs_df[\"performance.time_loss\"].dt.total_seconds()\n", + "runs_df[\"quality.time_loss\"] = runs_df[\"quality.time_loss\"].dt.total_seconds()\n", + "runs_df[\"productivity.time_loss\"] = runs_df[\"productivity.time_loss\"].dt.total_seconds()" + ] + }, + { + "cell_type": "code", + "execution_count": 46, + "id": "9051d479-3fbb-4337-964e-91cc5e49833b", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 227 entries, 0 to 226\n", + "Data columns (total 34 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 uuid 227 non-null object \n", + " 1 average_throughput 217 non-null float64 \n", + " 2 production_order 227 non-null object \n", + " 3 start 227 non-null datetime64[ns]\n", + " 4 end 226 non-null datetime64[ns]\n", + " 5 quantity_total 0 non-null float64 \n", + " 6 quantity_yield 0 non-null float64 \n", + " 7 availability 0 non-null float64 \n", + " 8 performance 0 non-null float64 \n", + " 9 quality 0 non-null float64 \n", + " 10 productivity 0 non-null float64 \n", + " 11 machine.uuid 227 non-null object \n", + " 12 machine.display_name 227 non-null object \n", + " 13 machine.genesis_date 227 non-null object \n", + " 14 machine.site._id 227 non-null int64 \n", + " 15 machine.site.display_name 227 non-null object \n", + " 16 machine.site.address 227 non-null object \n", + " 17 product.code 227 non-null object \n", + " 18 product.name 0 non-null object \n", + " 19 quantity_scrap.unit 213 non-null object \n", + " 20 quantity_scrap.value 213 non-null float64 \n", + " 21 quantity_total.unit 221 non-null object \n", + " 22 quantity_total.value 221 non-null float64 \n", + " 23 quantity_yield.unit 208 non-null object \n", + " 24 quantity_yield.value 208 non-null float64 \n", + " 25 availability.score 217 non-null float64 \n", + " 26 availability.time_loss 217 non-null float64 \n", + " 27 performance.score 217 non-null float64 \n", + " 28 performance.time_loss 217 non-null float64 \n", + " 29 quality.score 204 non-null float64 \n", + " 30 quality.time_loss 204 non-null float64 \n", + " 31 productivity.score 204 non-null float64 \n", + " 32 productivity.time_loss 204 non-null float64 \n", + " 33 quantity_scrap 0 non-null float64 \n", + "dtypes: datetime64[ns](2), float64(19), int64(1), object(12)\n", + "memory usage: 60.4+ KB\n" + ] + } + ], + "source": [ + "runs_df.info()" + ] + }, + { + "cell_type": "markdown", + "id": "51d8d167-3bc4-44f5-acf5-576f85bc4d0d", + "metadata": {}, + "source": [ + "With the necessary data transformations done, the DataFrame can now be saved as an Excel file or is ready to be returned back to Power BI" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9525437a-28a3-4811-9163-f7e4c075d72e", + "metadata": {}, + "outputs": [], + "source": [ + "runs_df.to_excel(\"./filename.xlsx\")" + ] + }, + { + "cell_type": "markdown", + "id": "030dc29e-ba51-4413-bd34-e9301e87049d", + "metadata": {}, + "source": [ + "🎉 This is it for now. Thank you for reading, and please [reach out to us]() if you have questions or would like us to cover a certain topic. \n", + "Best \n", + "Clemens" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.6" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/pyproject.toml b/pyproject.toml index e3a81d6..9aafb9d 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -4,15 +4,12 @@ requires = [ "setuptools", ] -[tool.setuptools.dynamic] -version = {attr = "enlyze._version.VERSION"} - [project] name = "enlyze" description = "Python SDK for interacting with the ENLYZE platform https://www.enlyze.com" readme = "README.rst" -license = {text = "MIT"} -authors = [{name = "ENLYZE GmbH", email = "hello@enlyze.com"},] +license = { text = "MIT" } +authors = [ { name = "ENLYZE GmbH", email = "hello@enlyze.com" } ] requires-python = ">=3.10" classifiers = [ "Programming Language :: Python :: 3 :: Only", @@ -21,21 +18,20 @@ classifiers = [ "Programming Language :: Python :: 3.12", ] dynamic = [ - 'version', + "version", ] dependencies = [ "httpx", "pandas>=2", "pydantic>=2", ] -[project.optional-dependencies] -docs = [ +optional-dependencies.docs = [ "sphinx", "sphinx-rtd-theme", "sphinx-tabs", "sphinxcontrib-spelling", ] -lint = [ +optional-dependencies.lint = [ "bandit", "black", "flake8", @@ -45,7 +41,7 @@ lint = [ "safety", "tox-ini-fmt", ] -test = [ +optional-dependencies.test = [ "coverage", "hypothesis", "pandas-stubs", @@ -57,8 +53,11 @@ test = [ "respx", ] -[tool.mypy] -exclude="tests" +[tool.setuptools.dynamic] +version = { attr = "enlyze._version.VERSION" } [tool.isort] profile = "black" + +[tool.mypy] +exclude = "tests" diff --git a/src/enlyze/__init__.py b/src/enlyze/__init__.py index 547a6ec..dd25043 100644 --- a/src/enlyze/__init__.py +++ b/src/enlyze/__init__.py @@ -1,9 +1,9 @@ from .client import EnlyzeClient from .errors import EnlyzeError, InvalidTokenError -from .models import Appliance, ResamplingMethod, Site, Variable +from .models import Machine, ResamplingMethod, Site, Variable __all__ = [ - "Appliance", + "Machine", "EnlyzeClient", "EnlyzeError", "InvalidTokenError", diff --git a/src/enlyze/api_clients/base.py b/src/enlyze/api_clients/base.py index 557deb3..1eec15c 100644 --- a/src/enlyze/api_clients/base.py +++ b/src/enlyze/api_clients/base.py @@ -8,10 +8,20 @@ import httpx from pydantic import BaseModel, ValidationError +from enlyze._version import VERSION from enlyze.auth import TokenAuth -from enlyze.constants import HTTPX_TIMEOUT +from enlyze.constants import HTTPX_TIMEOUT, USER_AGENT from enlyze.errors import EnlyzeError, InvalidTokenError +USER_AGENT_NAME_VERSION_SEPARATOR = "/" + + +@cache +def _construct_user_agent( + *, user_agent: str = USER_AGENT, version: str = VERSION +) -> str: + return f"{user_agent}{USER_AGENT_NAME_VERSION_SEPARATOR}{version}" + class ApiBaseModel(BaseModel): """Base class for ENLYZE platform API object models using pydantic @@ -60,6 +70,7 @@ def __init__( auth=TokenAuth(token), base_url=httpx.URL(base_url), timeout=timeout, + headers={"user-agent": _construct_user_agent()}, ) @cache diff --git a/src/enlyze/api_clients/production_runs/models.py b/src/enlyze/api_clients/production_runs/models.py index 192c2d3..d52fe06 100644 --- a/src/enlyze/api_clients/production_runs/models.py +++ b/src/enlyze/api_clients/production_runs/models.py @@ -3,6 +3,8 @@ from typing import Any, Optional from uuid import UUID +from pydantic import Field + import enlyze.models as user_models from enlyze.api_clients.base import ApiBaseModel @@ -59,14 +61,14 @@ def to_user_model(self) -> user_models.Quantity: ) -class Appliance(ApiBaseModel): +class Machine(ApiBaseModel): name: str uuid: UUID class ProductionRun(ProductionRunsApiModel): uuid: UUID - appliance: Appliance + machine: Machine = Field(alias="appliance") average_throughput: Optional[float] production_order: str product: Product @@ -81,7 +83,7 @@ class ProductionRun(ProductionRunsApiModel): productivity: Optional[OEEComponent] def to_user_model( - self, appliances_by_uuid: dict[UUID, user_models.Appliance] + self, machines_by_uuid: dict[UUID, user_models.Machine] ) -> user_models.ProductionRun: """Convert into a :ref:`user model `""" @@ -101,7 +103,7 @@ def to_user_model( return user_models.ProductionRun( uuid=self.uuid, - appliance=appliances_by_uuid[self.appliance.uuid], + machine=machines_by_uuid[self.machine.uuid], average_throughput=self.average_throughput, production_order=self.production_order, product=self.product.to_user_model(), diff --git a/src/enlyze/api_clients/timeseries/models.py b/src/enlyze/api_clients/timeseries/models.py index 5f84820..632657f 100644 --- a/src/enlyze/api_clients/timeseries/models.py +++ b/src/enlyze/api_clients/timeseries/models.py @@ -32,16 +32,16 @@ def to_user_model(self) -> user_models.Site: ) -class Appliance(TimeseriesApiModel): +class Machine(TimeseriesApiModel): uuid: UUID name: str genesis_date: date site: int - def to_user_model(self, site: user_models.Site) -> user_models.Appliance: + def to_user_model(self, site: user_models.Site) -> user_models.Machine: """Convert into a :ref:`user model `""" - return user_models.Appliance( + return user_models.Machine( uuid=self.uuid, display_name=self.name, genesis_date=self.genesis_date, @@ -55,7 +55,7 @@ class Variable(TimeseriesApiModel): unit: Optional[str] data_type: user_models.VariableDataType - def to_user_model(self, appliance: user_models.Appliance) -> user_models.Variable: + def to_user_model(self, machine: user_models.Machine) -> user_models.Variable: """Convert into a :ref:`user model `.""" return user_models.Variable( @@ -63,7 +63,7 @@ def to_user_model(self, appliance: user_models.Appliance) -> user_models.Variabl display_name=self.display_name, unit=self.unit, data_type=self.data_type, - appliance=appliance, + machine=machine, ) @@ -75,6 +75,31 @@ def extend(self, other: "TimeseriesData") -> None: """Add records from ``other`` after the existing records.""" self.records.extend(other.records) + def merge(self, other: "TimeseriesData") -> "TimeseriesData": + """Merge records from ``other`` into the existing records.""" + slen, olen = len(self.records), len(other.records) + if olen < slen: + raise ValueError( + "Cannot merge. Attempted to merge" + f" an instance with {olen} records into an instance with {slen}" + " records. The instance to merge must have a number" + " of records greater than or equal to the number of records of" + " the instance you're trying to merge into." + ) + + self.columns.extend(other.columns[1:]) + + for s, o in zip(self.records, other.records[:slen]): + if s[0] != o[0]: + raise ValueError( + "Cannot merge. Attempted to merge records " + f"with mismatched timestamps {s[0]}, {o[0]}" + ) + + s.extend(o[1:]) + + return self + def to_user_model( self, start: datetime, diff --git a/src/enlyze/client.py b/src/enlyze/client.py index ec4aa11..a29f6bf 100644 --- a/src/enlyze/client.py +++ b/src/enlyze/client.py @@ -1,6 +1,7 @@ +from collections import abc from datetime import datetime -from functools import cache -from typing import Iterator, Mapping, Optional, Sequence +from functools import cache, reduce +from typing import Any, Iterator, Mapping, Optional, Sequence, Tuple, Union from uuid import UUID import enlyze.api_clients.timeseries.models as timeseries_api_models @@ -10,9 +11,11 @@ from enlyze.api_clients.timeseries.client import TimeseriesApiClient from enlyze.constants import ( ENLYZE_BASE_URL, + MAXIMUM_NUMBER_OF_VARIABLES_PER_TIMESERIES_REQUEST, VARIABLE_UUID_AND_RESAMPLING_METHOD_SEPARATOR, ) -from enlyze.errors import EnlyzeError +from enlyze.errors import EnlyzeError, ResamplingValidationError +from enlyze.iterable_tools import chunk from enlyze.validators import ( validate_datetime, validate_resampling_interval, @@ -21,6 +24,8 @@ validate_timeseries_arguments, ) +FETCHING_TIMESERIES_DATA_ERROR_MSG = "Error occurred when fetching timeseries data." + def _get_timeseries_data_from_pages( pages: Iterator[timeseries_api_models.TimeseriesData], @@ -42,6 +47,38 @@ def _get_timeseries_data_from_pages( return timeseries_data +def _get_variables_sequence_and_query_parameter_list( + variables: Union[ + Sequence[user_models.Variable], + Mapping[user_models.Variable, user_models.ResamplingMethod], + ], + resampling_interval: Optional[int], +) -> Tuple[Sequence[user_models.Variable], Sequence[str]]: + if isinstance(variables, abc.Sequence) and resampling_interval is not None: + raise ResamplingValidationError( + "`variables` must be a mapping {variable: ResamplingMethod}" + ) + + if resampling_interval: + validate_resampling_interval(resampling_interval) + variables_sequence = [] + variables_query_parameter_list = [] + for variable, resampling_method in variables.items(): # type: ignore + variables_sequence.append(variable) + variables_query_parameter_list.append( + f"{variable.uuid}" + f"{VARIABLE_UUID_AND_RESAMPLING_METHOD_SEPARATOR}" + f"{resampling_method.value}" + ) + + validate_resampling_method_for_data_type( + resampling_method, variable.data_type + ) + return variables_sequence, variables_query_parameter_list + + return variables, [str(v.uuid) for v in variables] # type: ignore + + class EnlyzeClient: """Main entrypoint for interacting with the ENLYZE platform. @@ -82,28 +119,27 @@ def get_sites(self) -> list[user_models.Site]: """ return [site.to_user_model() for site in self._get_sites()] - def _get_appliances(self) -> Iterator[timeseries_api_models.Appliance]: - """Get all appliances from the API""" + def _get_machines(self) -> Iterator[timeseries_api_models.Machine]: + """Get all machines from the API""" return self._timeseries_api_client.get_paginated( - "appliances", timeseries_api_models.Appliance + "appliances", timeseries_api_models.Machine ) @cache - def get_appliances( + def get_machines( self, site: Optional[user_models.Site] = None - ) -> list[user_models.Appliance]: - """Retrieve all :ref:`appliances `, optionally filtered by site. + ) -> list[user_models.Machine]: + """Retrieve all :ref:`machines `, optionally filtered by site. - :param site: Only get appliances of this site. Gets all appliances of the + :param site: Only get machines of this site. Gets all machines of the organization if None. - :type site: :class:`~enlyze.models.Site` or None :raises: |token-error| :raises: |generic-error| - :returns: Appliances - :rtype: list[:class:`~enlyze.models.Appliance`] + :returns: Machines + :rtype: list[:class:`~enlyze.models.Machine`] """ @@ -112,46 +148,134 @@ def get_appliances( else: sites_by_id = {site._id: site for site in self.get_sites()} - appliances = [] - for appliance_api in self._get_appliances(): - site_ = sites_by_id.get(appliance_api.site) + machines = [] + for machine_api in self._get_machines(): + site_ = sites_by_id.get(machine_api.site) if not site_: continue - appliances.append(appliance_api.to_user_model(site_)) + machines.append(machine_api.to_user_model(site_)) - return appliances + return machines def _get_variables( - self, appliance_uuid: UUID + self, machine_uuid: UUID ) -> Iterator[timeseries_api_models.Variable]: - """Get variables for an appliance from the API.""" + """Get variables for a machine from the API.""" return self._timeseries_api_client.get_paginated( "variables", timeseries_api_models.Variable, - params={"appliance": str(appliance_uuid)}, + params={"appliance": str(machine_uuid)}, ) def get_variables( - self, appliance: user_models.Appliance + self, machine: user_models.Machine ) -> Sequence[user_models.Variable]: - """Retrieve all variables of an :ref:`appliance `. + """Retrieve all variables of a :ref:`machine `. - :param appliance: The appliance for which to get all variables. - :type appliance: :class:`~enlyze.models.Appliance` + :param machine: The machine for which to get all variables. :raises: |token-error| :raises: |generic-error| - :returns: Variables of ``appliance`` + :returns: Variables of ``machine`` """ return [ - variable.to_user_model(appliance) - for variable in self._get_variables(appliance.uuid) + variable.to_user_model(machine) + for variable in self._get_variables(machine.uuid) + ] + + def _get_paginated_timeseries( + self, + *, + machine_uuid: str, + start: datetime, + end: datetime, + variables: Sequence[str], + resampling_interval: Optional[int], + ) -> Iterator[timeseries_api_models.TimeseriesData]: + params: dict[str, Any] = { + "appliance": machine_uuid, + "start_datetime": start.isoformat(), + "end_datetime": end.isoformat(), + "variables": ",".join(variables), + } + + if resampling_interval: + params["resampling_interval"] = resampling_interval + + return self._timeseries_api_client.get_paginated( + "timeseries", timeseries_api_models.TimeseriesData, params=params + ) + + def _get_timeseries( + self, + start: datetime, + end: datetime, + variables: Union[ + Sequence[user_models.Variable], + Mapping[user_models.Variable, user_models.ResamplingMethod], + ], + resampling_interval: Optional[int] = None, + ) -> Optional[user_models.TimeseriesData]: + variables_sequence, variables_query_parameter_list = ( + _get_variables_sequence_and_query_parameter_list( + variables, resampling_interval + ) + ) + + start, end, machine_uuid = validate_timeseries_arguments( + start, end, variables_sequence + ) + + try: + chunks = chunk( + variables_query_parameter_list, + MAXIMUM_NUMBER_OF_VARIABLES_PER_TIMESERIES_REQUEST, + ) + except ValueError as e: + raise EnlyzeError(FETCHING_TIMESERIES_DATA_ERROR_MSG) from e + + chunks_pages = ( + self._get_paginated_timeseries( + machine_uuid=machine_uuid, + start=start, + end=end, + variables=chunk, + resampling_interval=resampling_interval, + ) + for chunk in chunks + ) + + timeseries_data_chunked = [ + _get_timeseries_data_from_pages(pages) for pages in chunks_pages ] + if not timeseries_data_chunked or all( + data is None for data in timeseries_data_chunked + ): + return None + + if any(data is None for data in timeseries_data_chunked) and any( + data is not None for data in timeseries_data_chunked + ): + raise EnlyzeError( + "The timeseries API didn't return data for some of the variables." + ) + + try: + timeseries_data = reduce(lambda x, y: x.merge(y), timeseries_data_chunked) # type: ignore # noqa + except ValueError as e: + raise EnlyzeError(FETCHING_TIMESERIES_DATA_ERROR_MSG) from e + + return timeseries_data.to_user_model( # type: ignore + start=start, + end=end, + variables=variables_sequence, + ) + def get_timeseries( self, start: datetime, @@ -161,7 +285,7 @@ def get_timeseries( """Get timeseries data of :ref:`variables ` for a given time frame. Timeseries data for multiple variables can be requested at once. However, all - variables must belong to the same appliance. + variables must belong to the same machine. You should always pass :ref:`timezone-aware datetime ` objects to this method! If you don't, naive @@ -182,30 +306,7 @@ def get_timeseries( """ - start, end, appliance_uuid = validate_timeseries_arguments( - start, end, variables - ) - - pages = self._timeseries_api_client.get_paginated( - "timeseries", - timeseries_api_models.TimeseriesData, - params={ - "appliance": appliance_uuid, - "start_datetime": start.isoformat(), - "end_datetime": end.isoformat(), - "variables": ",".join(str(v.uuid) for v in variables), - }, - ) - - timeseries_data = _get_timeseries_data_from_pages(pages) - if timeseries_data is None: - return None - - return timeseries_data.to_user_model( - start=start, - end=end, - variables=variables, - ) + return self._get_timeseries(start, end, variables) def get_timeseries_with_resampling( self, @@ -217,7 +318,7 @@ def get_timeseries_with_resampling( """Get resampled timeseries data of :ref:`variables ` for a given time frame. Timeseries data for multiple variables can be requested at once. However, all - variables must belong to the same appliance. + variables must belong to the same machine. You should always pass :ref:`timezone-aware datetime ` objects to this method! If you don't, naive @@ -245,55 +346,14 @@ def get_timeseries_with_resampling( request """ # noqa: E501 - variables_sequence = [] - variables_query_parameter_list = [] - for variable, resampling_method in variables.items(): - variables_sequence.append(variable) - variables_query_parameter_list.append( - f"{variable.uuid}" - f"{VARIABLE_UUID_AND_RESAMPLING_METHOD_SEPARATOR}" - f"{resampling_method.value}" - ) - - validate_resampling_method_for_data_type( - resampling_method, variable.data_type - ) - - start, end, appliance_uuid = validate_timeseries_arguments( - start, - end, - variables_sequence, - ) - validate_resampling_interval(resampling_interval) - - pages = self._timeseries_api_client.get_paginated( - "timeseries", - timeseries_api_models.TimeseriesData, - params={ - "appliance": appliance_uuid, - "start_datetime": start.isoformat(), - "end_datetime": end.isoformat(), - "variables": ",".join(variables_query_parameter_list), - "resampling_interval": resampling_interval, - }, - ) - - timeseries_data = _get_timeseries_data_from_pages(pages) - if timeseries_data is None: - return None - - return timeseries_data.to_user_model( - start=start, - end=end, - variables=variables_sequence, - ) + return self._get_timeseries(start, end, variables, resampling_interval) def _get_production_runs( self, *, production_order: Optional[str] = None, product: Optional[str] = None, - appliance: Optional[UUID] = None, + machine: Optional[UUID] = None, start: Optional[datetime] = None, end: Optional[datetime] = None, ) -> Iterator[ProductionRun]: @@ -302,7 +362,7 @@ def _get_production_runs( filters = { "production_order": production_order, "product": product, - "appliance": appliance, + "appliance": machine, "start": start.isoformat() if start else None, "end": end.isoformat() if end else None, } @@ -316,13 +376,13 @@ def get_production_runs( *, production_order: Optional[str] = None, product: Optional[user_models.Product | str] = None, - appliance: Optional[user_models.Appliance] = None, + machine: Optional[user_models.Machine] = None, start: Optional[datetime] = None, end: Optional[datetime] = None, ) -> user_models.ProductionRuns: """Retrieve optionally filtered list of :ref:`production runs `. - :param appliance: The appliance for which to get all production runs. + :param machine: The machine for which to get all production runs. :param product: Filter production runs by product. :param production_order: Filter production runs by production order. @@ -344,12 +404,12 @@ def get_production_runs( product_filter = ( product.code if isinstance(product, user_models.Product) else product ) - appliances_by_uuid = {a.uuid: a for a in self.get_appliances()} + machines_by_uuid = {a.uuid: a for a in self.get_machines()} return user_models.ProductionRuns( [ - production_run.to_user_model(appliances_by_uuid) + production_run.to_user_model(machines_by_uuid) for production_run in self._get_production_runs( - appliance=appliance.uuid if appliance else None, + machine=machine.uuid if machine else None, production_order=production_order, product=product_filter, start=start, diff --git a/src/enlyze/constants.py b/src/enlyze/constants.py index 90c8eaf..1989dbe 100644 --- a/src/enlyze/constants.py +++ b/src/enlyze/constants.py @@ -9,7 +9,7 @@ #: HTTP timeout for requests to the Timeseries API. #: -#: Reference: https://www.python-httpx.org/advanced/#timeout-configuration +#: Reference: https://www.python-httpx.org/advanced/timeouts/ HTTPX_TIMEOUT = 30.0 #: The separator to use when to separate the variable UUID and the resampling method @@ -18,3 +18,10 @@ #: The minimum allowed resampling interval when resampling timeseries data. MINIMUM_RESAMPLING_INTERVAL = 10 + +#: The maximum number of variables that can be used in a single request when querying +#: timeseries data. +MAXIMUM_NUMBER_OF_VARIABLES_PER_TIMESERIES_REQUEST = 100 + +#: The user agent that the SDK identifies itself as when making HTTP requests +USER_AGENT = "enlyze-python" diff --git a/src/enlyze/iterable_tools.py b/src/enlyze/iterable_tools.py new file mode 100644 index 0000000..64ca17c --- /dev/null +++ b/src/enlyze/iterable_tools.py @@ -0,0 +1,12 @@ +from typing import Iterable, Sequence, TypeVar + +MINIMUM_CHUNK_SIZE = 1 + +T = TypeVar("T") + + +def chunk(seq: Sequence[T], chunk_size: int) -> Iterable[Sequence[T]]: + if chunk_size < MINIMUM_CHUNK_SIZE: + raise ValueError(f"{chunk_size=} is less than {MINIMUM_CHUNK_SIZE=}") + + return (seq[i : i + chunk_size] for i in range(0, len(seq), chunk_size)) diff --git a/src/enlyze/models.py b/src/enlyze/models.py index a47d6a5..637b294 100644 --- a/src/enlyze/models.py +++ b/src/enlyze/models.py @@ -7,6 +7,8 @@ import pandas +from enlyze.schema import dataframe_ensure_schema + @dataclass(frozen=True) class Site: @@ -26,23 +28,23 @@ class Site: @dataclass(frozen=True) -class Appliance: - """Representation of an :ref:`appliance ` in the ENLYZE platform. +class Machine: + """Representation of a :ref:`machine ` in the ENLYZE platform. - Contains details about the appliance. + Contains details about the machine. """ - #: Stable identifier of the appliance. + #: Stable identifier of the machine. uuid: UUID - #: Display name of the appliance. + #: Display name of the machine. display_name: str - #: The date when the appliance has been connected to the ENLYZE platform. + #: The date when the machine has been connected to the ENLYZE platform. genesis_date: date - #: The site where the appliance is located. + #: The site where the machine is located. site: Site @@ -99,8 +101,8 @@ class Variable: #: The underlying data type of the variable. data_type: VariableDataType - #: The appliance on which this variable is read out. - appliance: Appliance + #: The machine on which this variable is read out. + machine: Machine @dataclass(frozen=True) @@ -233,7 +235,7 @@ class Quantity: @dataclass(frozen=True) class Product: - """Representation of a product that is produced on an appliance""" + """Representation of a product that is produced on a machine""" #: The identifier of the product code: str @@ -253,8 +255,8 @@ class ProductionRun: #: The UUID of the production run uuid: UUID - #: The appliance the production run was executed on. - appliance: Appliance + #: The machine the production run was executed on. + machine: Machine #: The average throughput of the production run excluding downtimes. average_throughput: Optional[float] @@ -280,10 +282,10 @@ class ProductionRun: #: The amount of product produced that can be sold. quantity_yield: Optional[Quantity] - #: OEE component that reflects when the appliance did not produce. + #: OEE component that reflects when the machine did not produce. availability: Optional[OEEComponent] - #: OEE component that reflects how fast the appliance has run. + #: OEE component that reflects how fast the machine has run. performance: Optional[OEEComponent] #: OEE component that reflects how much defects have been produced. @@ -303,13 +305,15 @@ def to_dataframe(self) -> pandas.DataFrame: ``end`` of every production run will be represented as :ref:`timezone-aware ` :py:class:`datetime.datetime` localized in UTC. - :returns: DataFrame with production runs - + :returns: DataFrame with production runs. """ if not self: return pandas.DataFrame() - df = pandas.json_normalize([asdict(run) for run in self]) + path_separator = "." + + df = pandas.json_normalize([asdict(run) for run in self], sep=path_separator) df.start = pandas.to_datetime(df.start, utc=True, format="ISO8601") df.end = pandas.to_datetime(df.end, utc=True, format="ISO8601") - return df + + return dataframe_ensure_schema(df, ProductionRun, path_separator=path_separator) diff --git a/src/enlyze/schema.py b/src/enlyze/schema.py new file mode 100644 index 0000000..ebd31a9 --- /dev/null +++ b/src/enlyze/schema.py @@ -0,0 +1,62 @@ +import typing +from dataclasses import is_dataclass +from types import UnionType +from typing import Any, Protocol + +import pandas + + +class DataclassTypeOrInstance(Protocol): + __dataclass_fields__: dict[str, Any] + + +def _flat_dataclass_schema( + dataclass_obj_or_type: DataclassTypeOrInstance, + path_separator: str, + _parent_path: list[str] = [], +) -> list[str]: + """Derive flat schema of potentially nested dataclass ``dataclass_obj_or_type``""" + + flat: list[str] = [] + + for field, typ in typing.get_type_hints(dataclass_obj_or_type).items(): + current_path = _parent_path + [field] + field_types = (typ,) + + # expand union types (includes typing.Optional) + origin_type = typing.get_origin(typ) + if origin_type is UnionType or origin_type is typing.Union: + field_types = typing.get_args(typ) + + for field_type in field_types: + if is_dataclass(field_type): + flat.extend( + _flat_dataclass_schema(field_type, path_separator, current_path) + ) + elif field_type is not type(None): + flat.append(path_separator.join(current_path)) + + # dedupe while preserving order + return list(dict.fromkeys(flat)) + + +def dataframe_ensure_schema( + df: pandas.DataFrame, + dataclass_obj_or_type: DataclassTypeOrInstance, + path_separator: str = ".", +) -> pandas.DataFrame: + """Add missing columns to ``df`` based on flattened dataclass schema""" + + flat_schema = _flat_dataclass_schema( + dataclass_obj_or_type, + path_separator=path_separator, + ) + + add_colums = set(flat_schema) - set(df.columns) + remove_columns = set(df.columns) - set(flat_schema) + + return df.assign( + **{col: None for col in add_colums}, + ).drop( + columns=list(remove_columns), + ) diff --git a/src/enlyze/validators.py b/src/enlyze/validators.py index dd95fde..50bb992 100644 --- a/src/enlyze/validators.py +++ b/src/enlyze/validators.py @@ -56,14 +56,14 @@ def validate_timeseries_arguments( start, end = validate_start_and_end(start, end) - appliance_uuids = frozenset(v.appliance.uuid for v in variables) + machine_uuids = frozenset(v.machine.uuid for v in variables) - if len(appliance_uuids) != 1: + if len(machine_uuids) != 1: raise EnlyzeError( - "Cannot request timeseries data for more than one appliance per request." + "Cannot request timeseries data for more than one machine per request." ) - return start, end, str(next(iter(appliance_uuids))) + return start, end, str(next(iter(machine_uuids))) def validate_resampling_interval( diff --git a/tests/conftest.py b/tests/conftest.py index 3cb10a2..2129782 100644 --- a/tests/conftest.py +++ b/tests/conftest.py @@ -8,6 +8,22 @@ hypothesis.settings.register_profile("ci", deadline=None) hypothesis.settings.load_profile(os.getenv("HYPOTHESIS_PROFILE", "default")) +# https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#timestamp-limitations +PANDAS_MIN_DATETIME = datetime(1677, 9, 21, 0, 12, 44) +PANDAS_MAX_DATETIME = datetime(2262, 4, 11, 23, 47, 16) + +# https://github.com/python/cpython/issues/94414 +WINDOWS_MIN_DATETIME = datetime(1970, 1, 2, 1, 0, 0) +WINDOWS_MAX_DATETIME = datetime(3001, 1, 19, 7, 59, 59) + +st.register_type_strategy( + datetime, + st.datetimes( + min_value=max(PANDAS_MIN_DATETIME, WINDOWS_MIN_DATETIME), + max_value=min(PANDAS_MAX_DATETIME, WINDOWS_MAX_DATETIME), + ), +) + datetime_today_until_now_strategy = st.datetimes( min_value=datetime.now().replace(hour=0), max_value=datetime.now(), diff --git a/tests/enlyze/api_clients/test_base.py b/tests/enlyze/api_clients/test_base.py index 922db32..897cd81 100644 --- a/tests/enlyze/api_clients/test_base.py +++ b/tests/enlyze/api_clients/test_base.py @@ -7,11 +7,15 @@ from hypothesis import HealthCheck, given, settings from hypothesis import strategies as st +from enlyze._version import VERSION from enlyze.api_clients.base import ( + USER_AGENT_NAME_VERSION_SEPARATOR, ApiBaseClient, ApiBaseModel, PaginatedResponseBaseModel, + _construct_user_agent, ) +from enlyze.constants import USER_AGENT from enlyze.errors import EnlyzeError, InvalidTokenError @@ -83,6 +87,46 @@ def base_client(auth_token, string_model, base_url): yield client +@pytest.fixture +def custom_user_agent(): + return "custom-user-agent" + + +@pytest.fixture +def custom_user_agent_version(): + return "3.4.5" + + +class TestConstructUserAgent: + def test__construct_user_agent_with_defaults(self): + ua, version = _construct_user_agent().split(USER_AGENT_NAME_VERSION_SEPARATOR) + assert ua == USER_AGENT + assert version == VERSION + + def test__construct_user_agent_custom_agent(self, custom_user_agent): + ua, version = _construct_user_agent(user_agent=custom_user_agent).split( + USER_AGENT_NAME_VERSION_SEPARATOR + ) + assert ua == custom_user_agent + assert version == VERSION + + def test__construct_user_agent_custom_version(self, custom_user_agent_version): + ua, version = _construct_user_agent(version=custom_user_agent_version).split( + USER_AGENT_NAME_VERSION_SEPARATOR + ) + assert ua == USER_AGENT + assert version == custom_user_agent_version + + def test__construct_user_agent_custom_agent_and_version( + self, custom_user_agent, custom_user_agent_version + ): + ua, version = _construct_user_agent( + user_agent=custom_user_agent, version=custom_user_agent_version + ).split(USER_AGENT_NAME_VERSION_SEPARATOR) + assert ua == custom_user_agent + assert version == custom_user_agent_version + + @settings(suppress_health_check=[HealthCheck.function_scoped_fixture]) @given( token=st.text(string.printable, min_size=1), diff --git a/tests/enlyze/api_clients/timeseries/test_models.py b/tests/enlyze/api_clients/timeseries/test_models.py new file mode 100644 index 0000000..7e6683c --- /dev/null +++ b/tests/enlyze/api_clients/timeseries/test_models.py @@ -0,0 +1,113 @@ +import itertools +import random +from datetime import datetime, timedelta, timezone + +import pytest + +from enlyze.api_clients.timeseries.models import TimeseriesData + +# We use this to skip columns that contain the timestamp assuming +# it starts at the beginning of the sequence. We also use it +# when computing lengths to account for a timestamp column. +TIMESTAMP_OFFSET = 1 +NOW = datetime.now(tz=timezone.utc) + + +def _generate_timeseries_data(*, columns, number_of_records): + timeseries_columns = ["time"] + timeseries_columns.extend(columns) + + counter = itertools.count(start=10) + + return TimeseriesData( + columns=timeseries_columns, + records=[ + [ + (NOW - timedelta(minutes=next(counter))).isoformat(), + *[random.randint(1, 100) for _ in range(len(columns))], + ] + for _ in range(number_of_records) + ], + ) + + +class TestTimeseriesData: + @pytest.mark.parametrize( + "data_parameters,data_to_merge_parameters", + [ + ( + {"columns": ["var1", "var2"], "number_of_records": 1}, + {"columns": ["var3"], "number_of_records": 1}, + ), + ( + {"columns": ["var1", "var2"], "number_of_records": 1}, + {"columns": ["var3"], "number_of_records": 3}, + ), + ], + ) + def test_merge(self, data_parameters, data_to_merge_parameters): + data = _generate_timeseries_data(**data_parameters) + data_to_merge = _generate_timeseries_data(**data_to_merge_parameters) + data_records_len = len(data.records) + data_columns_len = len(data.columns) + data_to_merge_columns_len = len(data_to_merge.columns) + expected_merged_record_len = len(data.records[0]) + len( + data_to_merge.records[0][TIMESTAMP_OFFSET:] + ) + + merged = data.merge(data_to_merge) + + assert merged is data + assert len(merged.records) == data_records_len + assert ( + len(merged.columns) + == data_columns_len + data_to_merge_columns_len - TIMESTAMP_OFFSET + ) + + for r in merged.records: + assert len(r) == expected_merged_record_len == len(merged.columns) + + @pytest.mark.parametrize( + "data_parameters,data_to_merge_parameters", + [ + ( + {"columns": ["var1", "var2"], "number_of_records": 2}, + {"columns": ["var3"], "number_of_records": 1}, + ), + ], + ) + def test_merge_raises_number_of_records_to_merge_less_than_existing( + self, data_parameters, data_to_merge_parameters + ): + data = _generate_timeseries_data(**data_parameters) + data_to_merge = _generate_timeseries_data(**data_to_merge_parameters) + + with pytest.raises( + ValueError, + match=( + "The instance to merge must have a number of" + " records greater than or equal to the number" + " of records of the instance you're trying to merge into." + ), + ): + data.merge(data_to_merge) + + @pytest.mark.parametrize( + "data_parameters,data_to_merge_parameters", + [ + ( + {"columns": ["var1", "var2"], "number_of_records": 1}, + {"columns": ["var3"], "number_of_records": 1}, + ), + ], + ) + def test_merge_raises_mismatched_timestamps( + self, data_parameters, data_to_merge_parameters + ): + data = _generate_timeseries_data(**data_parameters) + data_to_merge = _generate_timeseries_data(**data_to_merge_parameters) + + data_to_merge.records[0][0] = (NOW - timedelta(days=1)).isoformat() + + with pytest.raises(ValueError, match="mismatched timestamps"): + data.merge(data_to_merge) diff --git a/tests/enlyze/test_client.py b/tests/enlyze/test_client.py index 4c3e1aa..9bfe547 100644 --- a/tests/enlyze/test_client.py +++ b/tests/enlyze/test_client.py @@ -23,10 +23,11 @@ from enlyze.client import EnlyzeClient from enlyze.constants import ( ENLYZE_BASE_URL, + MAXIMUM_NUMBER_OF_VARIABLES_PER_TIMESERIES_REQUEST, PRODUCTION_RUNS_API_SUB_PATH, TIMESERIES_API_SUB_PATH, ) -from enlyze.errors import EnlyzeError +from enlyze.errors import EnlyzeError, ResamplingValidationError from tests.conftest import ( datetime_before_today_strategy, datetime_today_until_now_strategy, @@ -34,7 +35,7 @@ MOCK_RESPONSE_HEADERS = {"Content-Type": "application/json"} -APPLIANCE_UUID = "ebef7e5a-5921-4cf3-9a52-7ff0e98e8306" +MACHINE_UUID = "ebef7e5a-5921-4cf3-9a52-7ff0e98e8306" PRODUCT_CODE = "product-code" PRODUCTION_ORDER = "production-order" SITE_ID = 1 @@ -62,7 +63,7 @@ start=datetime_before_today_strategy, end=datetime_today_until_now_strategy, appliance=st.builds( - production_runs_api_models.Appliance, uuid=st.just(APPLIANCE_UUID) + production_runs_api_models.Machine, uuid=st.just(MACHINE_UUID) ), product=st.builds( production_runs_api_models.Product, @@ -142,58 +143,58 @@ def test_get_sites(site1, site2): @given( site1=st.builds(timeseries_api_models.Site, id=st.just(1)), site2=st.builds(timeseries_api_models.Site, id=st.just(2)), - appliance1=st.builds(timeseries_api_models.Appliance, site=st.just(1)), - appliance2=st.builds(timeseries_api_models.Appliance, site=st.just(2)), + machine1=st.builds(timeseries_api_models.Machine, site=st.just(1)), + machine2=st.builds(timeseries_api_models.Machine, site=st.just(2)), ) -def test_get_appliances(site1, site2, appliance1, appliance2): +def test_get_machines(site1, site2, machine1, machine2): client = make_client() with respx_mock_with_base_url(TIMESERIES_API_SUB_PATH) as mock: mock.get("appliances").mock( - PaginatedTimeseriesApiResponse(data=[appliance1, appliance2]) + PaginatedTimeseriesApiResponse(data=[machine1, machine2]) ) mock.get("sites").mock(PaginatedTimeseriesApiResponse(data=[site1, site2])) - all_appliances = client.get_appliances() - assert all_appliances == [ - appliance1.to_user_model(site1.to_user_model()), - appliance2.to_user_model(site2.to_user_model()), + all_machines = client.get_machines() + assert all_machines == [ + machine1.to_user_model(site1.to_user_model()), + machine2.to_user_model(site2.to_user_model()), ] - appliances_site2 = client.get_appliances(site2.to_user_model()) - assert appliances_site2 == [ - appliance2.to_user_model(site2.to_user_model()), + machines_site2 = client.get_machines(site2.to_user_model()) + assert machines_site2 == [ + machine2.to_user_model(site2.to_user_model()), ] @given( - appliance=st.builds(timeseries_api_models.Appliance), + machine=st.builds(timeseries_api_models.Machine), ) -def test_get_appliances_site_not_found(appliance): +def test_get_machines_site_not_found(machine): client = make_client() with respx_mock_with_base_url(TIMESERIES_API_SUB_PATH) as mock: mock.get("sites").mock(PaginatedTimeseriesApiResponse(data=[])) - mock.get("appliances").mock(PaginatedTimeseriesApiResponse(data=[appliance])) + mock.get("appliances").mock(PaginatedTimeseriesApiResponse(data=[machine])) - assert client.get_appliances() == [] + assert client.get_machines() == [] @given( - appliance=st.builds(user_models.Appliance), + machine=st.builds(user_models.Machine), var1=st.builds(timeseries_api_models.Variable), var2=st.builds(timeseries_api_models.Variable), ) -def test_get_variables(appliance, var1, var2): +def test_get_variables(machine, var1, var2): client = make_client() with respx_mock_with_base_url(TIMESERIES_API_SUB_PATH) as mock: mock.get("variables").mock(PaginatedTimeseriesApiResponse(data=[var1, var2])) - variables = client.get_variables(appliance) + variables = client.get_variables(machine) assert variables == [ - var1.to_user_model(appliance), - var2.to_user_model(appliance), + var1.to_user_model(machine), + var2.to_user_model(machine), ] @@ -335,6 +336,74 @@ def test_get_timeseries_returns_none_on_empty_response( ) +@given( + data_strategy=st.data(), + records=st.lists( + st.tuples( + datetime_today_until_now_strategy.map(datetime.isoformat), + st.integers(), + ), + min_size=2, + max_size=5, + ), + machine=st.builds(timeseries_api_models.Machine, uuid=st.just(MACHINE_UUID)), +) +@settings(suppress_health_check=[HealthCheck.function_scoped_fixture]) +def test__get_timeseries_raises_on_mixed_response( + data_strategy, + start_datetime, + end_datetime, + records, + machine, +): + """ + Tests that an `EnlyzeError` is raised if the timeseries API returns + data for some of the variables but not all of them. + """ + client = make_client() + variables = data_strategy.draw( + st.lists( + st.builds( + user_models.Variable, + data_type=st.just("INTEGER"), + machine=st.just(machine), + ), + min_size=MAXIMUM_NUMBER_OF_VARIABLES_PER_TIMESERIES_REQUEST + 1, + max_size=MAXIMUM_NUMBER_OF_VARIABLES_PER_TIMESERIES_REQUEST + 5, + ) + ) + + with respx_mock_with_base_url(TIMESERIES_API_SUB_PATH) as mock: + mock.get("timeseries").mock( + side_effect=[ + PaginatedTimeseriesApiResponse( + data=timeseries_api_models.TimeseriesData( + columns=[ + "time", + *[ + str(variable.uuid) + for variable in variables[ + :MAXIMUM_NUMBER_OF_VARIABLES_PER_TIMESERIES_REQUEST + ] + ], + ], + records=records, + ).model_dump(), + ), + PaginatedTimeseriesApiResponse( + data=timeseries_api_models.TimeseriesData( + columns=[], + records=[], + ).model_dump(), + ), + ] + ) + with pytest.raises( + EnlyzeError, match="didn't return data for some of the variables" + ): + client._get_timeseries(start_datetime, end_datetime, variables) + + def test_get_timeseries_raises_no_variables(start_datetime, end_datetime): client = make_client() with pytest.raises(EnlyzeError, match="at least one variable"): @@ -354,17 +423,17 @@ def test_get_timeseries_raises_invalid_time_bounds(variable): @given( - # we rely on variable{1,2}.appliance.uuid to be different because they are random + # we rely on variable{1,2}.machine.uuid to be different because they are random variable1=st.builds(user_models.Variable), variable2=st.builds(user_models.Variable), ) @settings(suppress_health_check=[HealthCheck.function_scoped_fixture]) -def test_get_timeseries_raises_variables_of_different_appliances( +def test_get_timeseries_raises_variables_of_different_machines( variable1, variable2, start_datetime, end_datetime ): client = make_client() - with pytest.raises(EnlyzeError, match="for more than one appliance"): + with pytest.raises(EnlyzeError, match="for more than one machine"): client.get_timeseries(start_datetime, end_datetime, [variable1, variable2]) @@ -390,16 +459,89 @@ def test_get_timeseries_raises_api_returned_no_timestamps( client.get_timeseries(start_datetime, end_datetime, [variable]) +@given( + variable=st.builds(user_models.Variable), +) +@settings(suppress_health_check=[HealthCheck.function_scoped_fixture]) +def test__get_timeseries_raises_variables_without_resampling_method( + start_datetime, end_datetime, variable +): + """ + Test that `get_timeseries` will raise an `EnlyzeError` when a + `resampling_interval` is specified but variables don't have + resampling methods. + """ + client = make_client() + with pytest.raises(ResamplingValidationError): + client._get_timeseries(start_datetime, end_datetime, [variable], 30) + + +@given( + variable=st.builds(user_models.Variable), +) +@settings(suppress_health_check=[HealthCheck.function_scoped_fixture]) +def test__get_timeseries_raises_on_chunk_value_error( + start_datetime, end_datetime, variable, monkeypatch +): + monkeypatch.setattr( + "enlyze.client.MAXIMUM_NUMBER_OF_VARIABLES_PER_TIMESERIES_REQUEST", 0 + ) + client = make_client() + with pytest.raises(EnlyzeError) as exc_info: + client._get_timeseries(start_datetime, end_datetime, [variable]) + assert isinstance(exc_info.value.__cause__, ValueError) + + +@given( + start=datetime_before_today_strategy, + end=datetime_today_until_now_strategy, + variable=st.builds( + user_models.Variable, + data_type=st.just("INTEGER"), + machine=st.builds(timeseries_api_models.Machine), + ), + records=st.lists( + st.tuples( + datetime_today_until_now_strategy.map(datetime.isoformat), + st.integers(), + ), + min_size=2, + ), +) +@settings(suppress_health_check=[HealthCheck.function_scoped_fixture]) +def test__get_timeseries_raises_on_merge_value_error( + start, end, variable, records, monkeypatch +): + client = make_client() + + def f(*args, **kwargs): + raise ValueError + + monkeypatch.setattr("enlyze.client.reduce", f) + + with respx_mock_with_base_url(TIMESERIES_API_SUB_PATH) as mock: + mock.get("timeseries").mock( + PaginatedTimeseriesApiResponse( + data=timeseries_api_models.TimeseriesData( + columns=["time", str(variable.uuid)], + records=records, + ).model_dump() + ) + ) + with pytest.raises(EnlyzeError): + client._get_timeseries(start, end, [variable]) + + @given( production_order=st.just(PRODUCTION_ORDER), product=st.one_of( st.builds(user_models.Product, code=st.just(PRODUCT_CODE)), st.text(), ), - appliance=st.builds( - timeseries_api_models.Appliance, + machine=st.builds( + timeseries_api_models.Machine, site=st.just(SITE_ID), - uuid=st.just(APPLIANCE_UUID), + uuid=st.just(MACHINE_UUID), ), site=st.builds(timeseries_api_models.Site, id=st.just(SITE_ID)), start=st.one_of(datetime_before_today_strategy, st.none()), @@ -409,7 +551,7 @@ def test_get_timeseries_raises_api_returned_no_timestamps( def test_get_production_runs( production_order, product, - appliance, + machine, site, start, end, @@ -418,37 +560,38 @@ def test_get_production_runs( client = make_client() site_user_model = site.to_user_model() - appliance_user_model = appliance.to_user_model(site_user_model) - appliances_by_uuid = {appliance.uuid: appliance_user_model} - - with respx_mock_with_base_url( - TIMESERIES_API_SUB_PATH - ) as timeseries_api_mock, respx_mock_with_base_url( - PRODUCTION_RUNS_API_SUB_PATH - ) as production_runs_api_mock: + machine_user_model = machine.to_user_model(site_user_model) + machines_by_uuid = {machine.uuid: machine_user_model} + + with ( + respx_mock_with_base_url(TIMESERIES_API_SUB_PATH) as timeseries_api_mock, + respx_mock_with_base_url( + PRODUCTION_RUNS_API_SUB_PATH + ) as production_runs_api_mock, + ): timeseries_api_mock.get("appliances").mock( - PaginatedTimeseriesApiResponse(data=[appliance]) + PaginatedTimeseriesApiResponse(data=[machine]) ) timeseries_api_mock.get("sites").mock( PaginatedTimeseriesApiResponse(data=[site]) ) production_runs_api_mock.get("production-runs").mock( PaginatedProductionRunsApiResponse( - data=[p.model_dump() for p in production_runs] + data=[p.model_dump(by_alias=True) for p in production_runs] ) ) result = client.get_production_runs( production_order=production_order, product=product, - appliance=appliance_user_model, + machine=machine_user_model, start=start, end=end, ) assert ( user_models.ProductionRuns( - [pr.to_user_model(appliances_by_uuid) for pr in production_runs] + [pr.to_user_model(machines_by_uuid) for pr in production_runs] ) == result ) diff --git a/tests/enlyze/test_iterable_tools.py b/tests/enlyze/test_iterable_tools.py new file mode 100644 index 0000000..eb9f034 --- /dev/null +++ b/tests/enlyze/test_iterable_tools.py @@ -0,0 +1,26 @@ +from typing import Sequence + +import pytest +from hypothesis import given +from hypothesis.strategies import integers, lists + +from enlyze.iterable_tools import MINIMUM_CHUNK_SIZE, chunk + + +@given( + seq=lists(integers()), + chunk_size=integers(min_value=MINIMUM_CHUNK_SIZE), +) +def test_chunk(seq: Sequence[int], chunk_size: int): + result = list(chunk(seq, chunk_size)) + assert sum(len(sublist) for sublist in result) == len(seq) + assert all(len(sublist) <= chunk_size for sublist in result) + + +@given( + seq=lists(integers()), + chunk_size=integers(max_value=MINIMUM_CHUNK_SIZE - 1), +) +def test_chunk_raises_invalid_chunk_size(seq: Sequence[int], chunk_size: int): + with pytest.raises(ValueError): + chunk(seq, chunk_size) diff --git a/tests/enlyze/test_models.py b/tests/enlyze/test_models.py new file mode 100644 index 0000000..3bba305 --- /dev/null +++ b/tests/enlyze/test_models.py @@ -0,0 +1,30 @@ +from dataclasses import replace + +import hypothesis.strategies as st +from hypothesis import given + +from enlyze.models import ProductionRun, ProductionRuns + + +@given(runs=st.lists(st.from_type(ProductionRun), max_size=10)) +def test_production_runs_to_dataframe(runs: list[ProductionRun]): + runs = ProductionRuns(runs) + runs.to_dataframe() + + +@given(run=st.from_type(ProductionRun)) +def test_production_runs_to_dataframe_no_empty_columns_for_optional_dataclasses( + run: ProductionRun, +): + df = ProductionRuns( + [ + replace( + run, + average_throughput=None, + quantity_total=None, + ) + ] + ).to_dataframe() + + assert "quantity_total" not in df.columns + assert "average_throughput" in df.columns diff --git a/tests/enlyze/test_schema.py b/tests/enlyze/test_schema.py new file mode 100644 index 0000000..4137d31 --- /dev/null +++ b/tests/enlyze/test_schema.py @@ -0,0 +1,32 @@ +from dataclasses import dataclass + +import pandas + +from enlyze.schema import dataframe_ensure_schema + + +@dataclass +class Some: + a: int + + +@dataclass +class Thing: + number: int + maybe_string: str | None + maybe_some: Some | None + multiple_but_required: float | str | Some + + +def test_dataframe_ensure_schema(): + df = pandas.DataFrame() + + added_columns = set(dataframe_ensure_schema(df, Thing, path_separator="|").columns) + + assert added_columns == { + "number", + "maybe_string", + "maybe_some|a", + "multiple_but_required", + "multiple_but_required|a", + } diff --git a/tests/enlyze/test_validators.py b/tests/enlyze/test_validators.py index 132d9ac..6bc7859 100644 --- a/tests/enlyze/test_validators.py +++ b/tests/enlyze/test_validators.py @@ -69,12 +69,10 @@ class TestValidateTimeseriesArguments: variable=VARIABLE_STRATEGY, ) def test_validate_timeseries_arguments(self, start, end, variable): - start, end, appliance_uuid = validate_timeseries_arguments( - start, end, [variable] - ) + start, end, machine_uuid = validate_timeseries_arguments(start, end, [variable]) assert start assert end - assert UUID(appliance_uuid) + assert UUID(machine_uuid) @given(variable=VARIABLE_STRATEGY) def test_validate_start_must_be_earlier_than_end(self, variable): @@ -97,7 +95,7 @@ def test_empty_variables(self, start, end): variable1=VARIABLE_STRATEGY, variable2=VARIABLE_STRATEGY, ) - def test_variables_with_different_appliance(self, start, end, variable1, variable2): + def test_variables_with_different_machine(self, start, end, variable1, variable2): with pytest.raises(EnlyzeError): validate_timeseries_arguments(start, end, [variable1, variable2])