erikiado/pyscheduler

pyscheduler

Pyscheduler is a python-like interpreter which incorporates time and processes as a native type. This allows for automation and scheduling of tasks in an easy and familiar way.

How to run for development:

If you want you can create a virtual environment just to avoid conflicts with dependencies.

virtualenv --python=`which python3` ~/.myenvs/scheduler

Then activate the environment running:

source ~/.myenvs/scheduler/bin/activate

Now for the interpreter, first you need to install all the dependencies

pip install -r requirements.txt

Then you just need to run the interpreter by executing

python pyscheduler.py

Or you can feed a file to the interpreter by passing the path as an argument

python pyscheduler.py [file]

Some tests files are provided under the tests/ directory. Remember to use local paths.

Some basic syntax:

a = 5
t = 4m2s
py_proc = p'test_script.py'
shell_proc = p'test_script.sh'

start py_proc
kill py_proc

every t:
  start py_proc

How does it work:

At the moment pyscheduler will take any valid input and compile it to python code. For example, the next valid file with extension .pys:

x = 3s
cont = 0
every x:
	if cont % 2 == 0:
		start p'test_script.py'
	if cont % 2 == 1:
		start p'test_script2.sh'
	cont = cont + 1

Is then interpreted as the following python code:

import time
from pyscheduler_worker import SchedulerWorker

def seconds_passed(last,seconds):
    return ((time.time() - last) >= seconds)
x = 3
cont = 0
last_pyscheduler_second = time.time()
while True:
    time.sleep(1)
    if seconds_passed(last_pyscheduler_second,x):
        last_pyscheduler_second = time.time()
        if cont % 2 == 0:
            SchedulerWorker('test_script.py').run()

        if cont % 2 == 1:
            SchedulerWorker('test_script2.sh').run()

        cont = cont + 1