A HTTP commanded, YAML configured task runner

Yet another HTTP task runner?

Yes.

Why?

Small scale deployments on single machines rarely need sophisticated build systems.

Some times, simple scripts can be all it takes to deploy software.

This aims to provide exactly that.

How does it work?

The example configuration can serve as starting point to define a YAML of the configuration. By default, the program would look for config.yaml in the working directory (that can be overriden using --config option). To auto reload the program when the configuration file change, use the --watch flag.

This program works by:

• HTTP listen on LISTEN port defined in configuration (using format :xxxx)
• add prefix to endpoints
• for each task specified in the configuration:
  • RunnerExecutable that will be ran
  • Args that will be passed to the RunnerExecutable
  • MaxRunSeconds that each task will be allowed to run
• The program would create the directory logs inside the working directory, that would be the log of builds

Each taskKey (buildBackend in the example config) would result in these endpoints:

• {{routePrefix}}/tasks/{{task.WebhookRoute}} would trigger the task.
• {{routePrefix}}/logs/{{task.TaskKey}} would render a HTML page listing task execution history (as timestamp entries), where each should contain the task executions logs that default to run.log if SeparateRunLogs is false in the configuration, or out.log (stdout) and err.log (stderr) of the task execution.