World of Tanks Attendance Tracker
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
docker/postgres
scripts
vagrant
whyattend
.gitignore
Dockerfile
LICENSE
README.md
alembic.ini.template
alembic_docker.ini
docker_config.py
initdb.py
requirements.txt
runtornado.py
server.py
wsgi_app.wsgi

README.md

World of Tanks CW Attendance Tracker

This web application helps a clan keep track of the clan war attendance of its players and with gold payouts.

Dependencies

  • Python 2.7 (Python 2.6 probably works too)
  • Various Python libraries. See below for installation instructions.
  • A relational database - MySQL/MariaDB (tested) / PostgreSQL
  • A somewhat recent Linux distribution. Windows and Mac OS X most likely work too but are not considered in the installation examples below.

Installation

Start off by configuring the application:

  • Copy whyattend/config.py to local_config.py and adjust the settings.
  • Put your clan's logo into whyattend/static/img/clanicons.
  • Replace whyattend/static/img/header.jpg with an image of your choice.
  • Copy alembic.ini.template to alembic.ini and set sqlalchemy.url to the same value as DATABASE_URI in local_config.py

Now the initial database schema should be generated by running python initdb.py from the shell. It is important that this script runs without errors or it might leave the database in a broken state.

MySQL specifics

Make sure the replay_blob column in the generated replay table is large enough to hold WoT replay files. In MySQL, for example, the default BLOB type can only store around 64 kB. This can be changed by running the SQL statement ALTER TABLE replay MODIFY replay_blob LONGBLOB and ALTER TABLE replay MODIFY replay_pickle LONGBLOB, for example. Additionally, MySQL defines a max_allowed_packet size in my.cnf, which might have to be increased.

Python libraries

The application requires several Python libraries listed in requirements.txt These should be installed into a virtual Python environment to isolate them from the globally installed Python packages that come with your Linux distribution, for example.

In bash, run

> virtualenv ./myenv
> source ./myenv/bin/activate
> pip install -r requirements.txt

These commands will create a environment in the "myenv" folder and then install the required libraries. Some of these libraries might need to be compiled which requires further build dependencies. For such libraries it might be easier to use a binary package provided by your Linux distribution.

Deployment Examples

As a Python WSGI application, this web application can be deployed in various ways. See the Flask documentation (http://flask.pocoo.org/docs/deploying/) for further information. The following examples show tested and popular options.

Apache + mod_wsgi

Follow the instructions from example 2 to install a virtual Python environment and use the provided wsgi_app.wsgi file as example.

The apache2 virtual host configuration could look like this:

WSGIDaemonProcess clanwars processes=1 threads=5
WSGIProcessGroup clanwars
WSGIScriptAlias /clanwars /var/www/clanwars/wsgi_app.wsgi

Alias /clanwars/static/ /var/www/clanwars/whyattend/static/
<Directory /var/www/clanwars/whyattend/static>
    Order allow,deny
    Allow from all
</Directory>

Tornado (+ Reverse Proxy)

First, install Tornado into your virtual Python environment:

> source ./myenv/bin/activate
> pip install tornado

To start the server you can use the provided runtornado.py script:

> source ./myenv/bin/activate
> python runtornado.py

This will start a Tornado server listening on port 5001. It is recommended to let Tornado listen only on localhost and put it behind a web server such as Nginx or Apache with mod_proxy that runs on port 80.

Docker

Docker is a program that automates deployment of applications inside containers, which are similar to virtual machines but uses OS level isolation features. The provided Dockerfiles can also serve as guide on how to install the tracker on Ubuntu.

First, adjust the settings in docker_config.py.

To build docker containers with a PostgreSQL database and another one with the whyattend web application simply run the following commands:

docker build -t whyattend_postgresql docker/postgresql
docker build -t webapp .

Then start the database container with:

docker run -d --name db whyattend_postgresql
docker run -d -P --name web --link db:db webapp

docker ps -a will then show you on which port the web application runs on the host system.

Clan member synchronisation

The clan member roles can be synchronized with Wargaming's server by opening /sync-players/<clanid>/?API_KEY=<configured API KEY>. The tracker will query the Wargaming API for all clan members which should show up in the list of players right away.

config.API_TOKEN is used to authenticate your application instance with the Wargaming API. Tokens can be generated on the Wargaming Developer Partner program website, e.g. https://eu.wargaming.net/developers/

config.API_KEY should be set to something random and secret, so only you can trigger the synchronization.

This can be automated by a cron script, e.g:

#!/bin/bash
# Synchronize WHY members
curl "http://myserver.com/sync-players/500014725?API_KEY=<configured API KEY>"

Updating

Whenever you update your installation to the latest source code you should check if a database migration is required.

Database migration is required, when the data model changes and table structures have to be changed accordingly. For database migration we use Alembic. Whenever updating to the latest source code, simply run alembic upgrade head, which will issue the necessary SQL statements and bring the database up to date.

Note for developers: When making changes to the data model in model.py, you should create an Alembic migration file and check it into version control along the code changes. See the Alembic documentation on how to create migration files.

Development Hints

For developing the builtin development HTTP server is sufficient. To install the required dependencies simply run the following commands (assuming a bash shell and virtualenv using Python 2.7)

cd topleveldir
virtualenv devenv
source devenv/bin/activate
pip install -r requirements.txt
python server.py

which will start a web server listening on port 5000. The development server will automatically restart when it detects changes to the code.