Skip to content

Latest commit

 

History

History
77 lines (57 loc) · 5.2 KB

README.md

File metadata and controls

77 lines (57 loc) · 5.2 KB
Raw

CompletenessLevel markers

CompletenessLevel marker enables testcases to be executed in different meaningful levels. Each level is a representation of a scope of execution of a testcase. This document describes the usage of CompletenessLevel marker. Defined levels (in increasing order) -

    debug

    basic

    confident

    thorough

An unordered level diagnose is also supported: diagnose is meant to analyze an existing defect or a feature. This level is different from the ordered levels that are primarily meant to validate SONiC. Diagnose encompasses a special set of test scenarios that are known to fail because of an existing image issue. This level is intended for a manual test targetted at executing unhealthy scenarios only as defined in the test. If diagnose level is specified and a test doesn’t support it, then it shall fall back to basic mode.

To use CompletenessLevel:

  • Mark the testcase with marker supported_completeness_level. This marker is a list of all the completeness levels supported by a testcase.
  • During Pytest execution, use command line option --completeness_level to specify the test completeness level.
  • Automatic normalization between specified --completeness_level and defined supported_completeness_level will be performed and the test will be executed at the resultant normalized level of completeness.
  • If module/session/testcase have different supported levels of completeness, the inner most level will supersede any defined level. For eg., if the module and testcase have supported levels "debug, basic, thorough" and "confident" respectively, the resultant defined level for this testcase will be "confident".
  • Within a testcase - Class method CompletenessLevel.get_normalized_level() can be called with the test's request object to get the normalized level.

Different cases for CompletenessLevel

To handle any discrepancy between specified and defined completeness levels, normalization will be performed during testcase setup. Normalization accounts for below mentioned cases: 1. Completeness level not specified during test execution - set the specified level to "basic".

2. Test does not define any completeness level - run the full testcase by default.

3. Specified completeness level do not match any defined level in a test case:
    3.1 Specified level is higher than any defined level - go to highest level defined
    3.2 Specified level is lower than any defined level - go to lowest level defined
    3.3 Specified level is in between two defined levels - go to next lower level

4. Specified level matches one of the defined levels - run testcase at the specified level.

CompletenessLevel usage example

import pytest
from tests.common.plugins.test_completeness import CompletenessLevel

pytestmark = [pytest.mark.supported_completeness_level(CompletenessLevel.debug, CompletenessLevel.thorough)]

def test_test_completeness_default(request):
    normalized_level = CompletenessLevel.get_normalized_level(request)
    logger.info("Completeness level set to: {}".format(str(normalized_level)))

    ## Continue execution of the testecase until the completeness level specified.
    # debug - Do something - end the test if the specified level is debug
    ...
    ...
    # basic - Do something more - extra tests/verifications - end the test now if the level is basic
    ...
    ...
    # thorough - Run entire test - if the set level is thorough

CompletenessLevel execution snippets

Case - Specified level higher than any defined level - set to highest defined level

 __init__.py:check_test_completeness:139: Setting test completeness level. Specified: CompletenessLevel.basic. Defined: (<CompletenessLevel.debug: 0>,)
 __init__.py:check_test_completeness:153: Specified level (CompletenessLevel.basic) not found in defined levels. Setting level to CompletenessLevel.debug

Case - Specified level lesser than any defined level - set to lowest defined level

 __init__.py:check_test_completeness:139: Setting test completeness level. Specified: CompletenessLevel.basic. Defined: (<CompletenessLevel.confident: 2>, <CompletenessLevel.thorough: 3>)
 __init__.py:check_test_completeness:153: Specified level (CompletenessLevel.basic) not found in defined levels. Setting level to CompletenessLevel.confident

Case - Specified level present in the defined levels

 __init__.py:check_test_completeness:139: Setting test completeness level. Specified: CompletenessLevel.basic. Defined: (<CompletenessLevel.debug: 0>, <CompletenessLevel.basic: 1>, <CompletenessLevel.confident: 2>, <CompletenessLevel.thorough: 3>)
 __init__.py:check_test_completeness:156: Setting the completeness level to CompletenessLevel.basic

Case - Specified level between two defined levels - set to lesser defined level

 __init__.py:check_test_completeness:139: Setting test completeness level. Specified: CompletenessLevel.basic. Defined: (<CompletenessLevel.debug: 0>, <CompletenessLevel.confident: 2>, <CompletenessLevel.thorough: 3>)
 __init__.py:check_test_completeness:153: Specified level (CompletenessLevel.basic) not found in defined levels. Setting level to CompletenessLevel.debug