How do I write tests for a snakemake pipeline?

You can run CI on a snakemake workflow when provoding a minimal example dataset. Maybe an example from the snakemake-workflows repo will be helpful: https://github.com/snakemake-workflows/dna-seq-varlociraptor

This is what I do when I want to check whether a program runs from start to finish and produces the expected output on test data. Here I used it to test a C program but the idea can be applied also to a snakemake pipeline as in your case.

The idea is pretty simple really: use python's subprocess module to run the pipeline then check the results using the unittest module (or your favourite unit test package).

For example, save this in test.py:

#!/usr/bin/env python3

import unittest
import os
import subprocess as sp

class Tester(unittest.TestCase):

    def testPipelineMakesStuff(self):
        # Run the pipeline
        p= sp.Popen('snakemake [some parameters]', shell=True, stdout= sp.PIPE, stderr= sp.PIPE)
        
        # Optionally, get stdout and stderr
        stdout, stderr= p.communicate()

        # Check exits code and other expected output            
        self.assertEqual(0, p.returncode)
        self.assertTrue(os.path.isfile('some.pipeline.output.txt'))

        ...Maybe read some.pipeline.output.txt and check it looks ok

    def testPipelineReturnsWithError(self):
        p= sp.Popen('snakemake [some parameters]', shell=True, stdout= sp.PIPE, stderr= sp.PIPE)
        stdout, stderr= p.communicate()
        self.assertTrue(p.returncode != 0)
        self.assertTrue('Something went wrong' in stderr.decode())
    
    def ...more test cases...

if __name__ == '__main__':
    unittest.main()

Run this script as:

 python test.py 

and check all tests pass ok. Of course, you can embed python test.py in travis or other continuous integration system as I've done here.