How to specify params path using nrunner?

[marioaag]

I have being trying to get the parameters specifying a path without success and looking at the documentation it talks about the --mux-yaml option to make it available but it requires the mux plugin

self.params.get('param_name', 'path_to_yaml_file')

Is it a way to specify the path where the yaml file whit the params values is located?
What is the default directory where abocado is going to look for the files with params?

[willianrampazzo]

Hi @marioaag, the path, in this case, is the path of the parameter in the params tree. I gave an example about the use of mux to another user here: #4920 (comment) . If you are using a single parameter, -p command-line option is what you are looking for.

Another example, much more complex, is our test on supported images for the vmimage module: selftests/pre_release/tests/vmimage.py (https://github.com/avocado-framework/avocado/blob/master/selftests/pre_release/tests/vmimage.py), the yml file is at selftests/pre_release/tests/vmimage.py.data/variants.yml (https://github.com/avocado-framework/avocado/blob/master/selftests/pre_release/tests/vmimage.py.data/variants.yml).

[marioaag]

So if I don't specify -m switch it should take the parameters form 'test_name.py.data/test_name.yaml' by default?

Using something like:

self.params.get('param_name', '*')

[willianrampazzo]

So if I don't specify -m switch it should take the parameters form 'test_name.py.data/test_name.yaml' by default?

Using something like:

self.params.get('param_name', '*')

No, you still need to set the correct file to load. Following is a complete example.

sleep.py

#!/usr/bin/env python3

import sys
import time

from avocado import Test
from avocado.core.job import Job

config = {'run.references': ['sleep_test.py:PassTest.test'],
          'yaml_to_mux.files': ['sleep_test.py.data/sleep_test.yml']}


with Job.from_config(job_config=config) as j:
    sys.exit(j.run())

sleep_test.py

import time
from avocado import Test


class PassTest(Test):
    def test(self):
        length = float(self.params.get('length', default=3))
        time.sleep(length)

sleep_test.py.data/sleep_test.yml

!mux
small:
  length: 0.5
medium:
  length: 1.0
large:
  length: 2.0

When I run sleep.py, I get:

$ python sleep.py 
JOB ID     : fcbfb189f73efb397967526a1a51ff2eed0a7609
JOB LOG    : /home/wrampazz/avocado/job-results/job-2021-09-24T15.45-fcbfb18/job.log
 (1-1/3) sleep_test.py:PassTest.test;run-small-74d3: PASS (0.50 s)
 (1-2/3) sleep_test.py:PassTest.test;run-medium-7fac: PASS (1.00 s)
 (1-3/3) sleep_test.py:PassTest.test;run-large-8fc9: PASS (2.00 s)
RESULTS    : PASS 3 | ERROR 0 | FAIL 0 | SKIP 0 | WARN 0 | INTERRUPT 0 | CANCEL 0
JOB HTML   : /home/wrampazz/avocado/job-results/job-2021-09-24T15.45-fcbfb18/results.html
JOB TIME   : 4.02 s

[willianrampazzo]

I edited the previous comment with the correct information as the bug I mentioned was because of the spell difference in the YAML file.

[marioaag]

Thanks so much will give ti a try!!!

The yaml_to_mux.files needs to be specified at Job level or could be specified at runnable level?

[willianrampazzo]

Thanks so much will give ti a try!!!

The yaml_to_mux.files needs to be specified at Job level or could be specified at runnable level?

You need to specify it on the Job level because it automatically creates new tasks.

[marioaag]

oooooo ic, it is not possible to use is as default value instead of hardcode the value in the TC.