Spark Networks DevOps coding challenge

https://gitlab.com/askainet/spark-networks-code-challenge

Requirements

• Python >= 3.4

Install

From public Pypi repository:

pip install sparkd

Demo

docker build . -t sparkd

docker run -it -e RETRIES=5 sparkd demo.sh

Usage

usage: sparkd [-h] [--version] [-l LOGFILE]
              [--command-logfile COMMAND_LOGFILE] [-n NAME] [-v] [-d]
              [-r RETRIES] [-i RETRY_INTERVAL] [-c CHECK_INTERVAL]
              [command] [arguments [arguments ...]]

Run any command as a daemon and supervise it

positional arguments:
  command               The command to run as a daemon to supervise
  arguments             Arguments to the command

optional arguments:
  -h, --help            show this help message and exit
  --version             Show version
  -l LOGFILE, --logfile LOGFILE
                        Set the logfile for the Sparkd supervisor
  --command-logfile COMMAND_LOGFILE
                        Set the logfile for the command to supervise
  -n NAME, --name NAME  Set the name of this Sparkd instance
  -v, --verbose         Enable verbose logging
  -d, --debug           Enable debug logging
  -r RETRIES, --retries RETRIES
                        Number of retries to restart the process
  -i RETRY_INTERVAL, --retry-interval RETRY_INTERVAL
                        Seconds to wait between retries to restart the process
  -c CHECK_INTERVAL, --check-interval CHECK_INTERVAL
                        Seconds to wait between checking process status

Developing

Tests

Some basic tests have been added to have coverage only for the Process class, just to demonstrate how to use unittest with Python.

make test

Linting

Linting with pylint to keep a nice and healthy code.

make lint

Publish to Pypi

Package and publish to Pypi repository.

make dist

Install from source

make install

Documentation

Documentation is built using sphinx based on docstrings present in the code.

Readthedocs integration is configured to automatically generate and upload docs to https://sparkd.readthedocs.io/en/latest/ on updates to master branch.

Manual local docs generation is possible running make docs