The-elite Notification System (TNS)

Overview

• add more stuff here

Project structure

• startup.py: starts everything (we still need to adapt some things when the job framework is added)
• tns
  • cfg: config layer
  • db: database layer
    • model: one file for every table in DB
  • job: schedule/job layer
• config: yml files
  • tns.yml: db and app configuration
  • logging.yml: logging config
• logs: default folder where logs are written (more details in config/logging.yml)

Setup

Python Virtual Env

• Install Miniconda
  • https://conda.io/miniconda.html
• Create the virtual env using the environment.yml file in the root of the project
  • conda env create -f environment.yml -n tns
• In PyCharm Preferences > Project > select the project and set the 'tns' env you just created

Database

• Install latest PostgreSQL (e.g 9.6.5)
• Create a user and a database, ex:

# create the user, fill in the info
createuser -P --interactive

# create user with the owner as whatever user you created above
createdb -O <user>

# test it out
psql -h localhost <user> -U <pass>

• Configure it in config/tns.yml

• Easier to use a docker container for it though, e.g:

docker run -d -p 5432:5432 --name tns_postgres -e POSTGRES_PASSWORD=postgres postgres:9.6.12

Execution

• Open a terminal with the tns virtual env activated
  • It's easier to use PyCharm's terminal option as it starts a terminal with the configured env activated
• Inside startup.py file there' a call to tsn_startup.main() with some optional env variables thatn be used. Examples:
  • If you want the db to be recreated and some test models to be inserted, just use TNS_RECREATE_TABLES=true TNS_TEST_DB_MODELS=true python startup.py
  • To change the log level to INFO use TNS_LOG_LEVEL='INFO'
• Run it
  • python startup.py

Extra

• To generate the conda environments.yml file again, with the tns source activated:

conda env export -n tns > environment.yml