short for workflow, which this aims to improve
flow is just a concept for now and won't be in use for quite some time, but the idea is to essentially create a parent program (i.e. flow) that holds references to and configurations for other programs / scripts. flow would then use the configs to execute these programs. The goal is to have an all inclusive piece of software that holds all of an institution's / individual's procedural programs in one place to run whenever needed.
For example, let's take the datamatrix software written for FLMNH. I would boot up flow, 'create' a new program, enter in the path to the python script, enter in any variables or flags to be passed to the python environment, generate a run config and execute whenever needed. So if my file was called dm_reader.py, the run config would look something like this:
runConfig = {
commandPrefix: "python3",
arguments: [{ argType, name, flag, value, description }, { ... }],
defaultArgs: ["-u"],
command: "{commandPrefix} {defaults} {program_location} {arguments}",
};and then translate to a string and look something like this:
python3 -u /path/to/dm_reader.py --arg1 arg1_valueUpon execution, flow would recieve output as it is displayed from the child process (i.e. the program you are now executing) and display it to you live.
Here's a demo of simple python program that loops 3 times, and sleeps in between iterations:
The installation and development usage after cloning is as follows:
$ yarn
$ yarn dev