THIS REPO IS DEPRECATED. IT WAS MOVED HERE

Qase TMS Pytest Plugin

Installation

pip install qase-pytest

Usage

Configuration

Configuration could be provided both by pytest.ini/tox.ini params
and using command-line arguments:

• Command-line args:

  --qase                Use Qase TMS
  --qase-api-token=QS_API_TOKEN
                        Api token for Qase TMS
  --qase-project=QS_PROJECT_CODE
                        Project code in Qase TMS
  --qase-testrun=QS_TESTRUN_ID
                        Testrun ID in Qase TMS
  --qase-testplan=QS_TESTPLAN_ID
                        Testplan ID in Qase TMS
  --qase-new-run        Create new testrun, if no testrun id provided
  --qase-debug          Prints additional output of plugin

• INI file parameters:

  qs_enabled (bool):    default value for --qase
  qs_api_token (string):
                        default value for --qase-api-token
  qs_project_code (string):
                        default value for --qase-project
  qs_testrun_id (string):
                        default value for --qase-testrun
  qs_testplan_id (string):
                        default value for --qase-testplan
  qs_new_run (bool):    default value for --qase-new-run
  qs_debug (bool):      default value for --qase-debug

Link tests with test-cases

To link tests with test-cases in Qase TMS you should use predefined decorator:

from qaseio.pytest import qase
@qase.id(13)
def test_example_1():
    pass
@qase.id(12, 156)
def test_example_2():
    pass

You could pass as much IDs as you need.

Possible cases statuses

• PASSED - when test passed
• FAILED - when test failed with AssertionError
• BLOCKED - when test failed with any other exception
• SKIPPED - when test has been skipped

Add attachments to test results

When you need to push some additional information to server you could use
attachments:

import pytest
from qaseio.client.models import MimeTypes
from qaseio.pytest import qase
@pytest.fixture(scope="session")
def driver():
    driver = webdriver.Chrome()
    yield driver
    logs = "\n".join(str(row) for row in driver.get_log('browser')).encode('utf-8')
    qase.attach((logs, MimeTypes.TXT, "browser.log"))
    driver.quit()
@qase.id(13)
def test_example_1():
    qase.attach("/path/to/file", "/path/to/file/2")
    qase.attach(
        ("/path/to/file/1", "application/json"),
        ("/path/to/file/3", MimeTypes.XML),
    )
@qase.id(12, 156)
def test_example_2(driver):
    qase.attach((driver.get_screenshot_as_png(), MimeTypes.PNG, "result.png"))

You could pass as much files as you need.

Also you should know, that if no case id is associated with current test in
pytest - attachment would not be uploaded:

import pytest
from qaseio.client.models import MimeTypes
from qaseio.pytest import qase
@pytest.fixture(scope="session")
def driver():
    driver = webdriver.Chrome()
    yield driver
    logs = "\n".join(str(row) for row in driver.get_log('browser')).encode('utf-8')
    # This would do nothing, because last test does not have case id link
    qase.attach((logs, MimeTypes.TXT, "browser.log"))
    driver.quit()
def test_example_2(driver):
    # This would do nothing
    qase.attach((driver.get_screenshot_as_png(), MimeTypes.PNG, "result.png"))

Linking code with steps

It is possible to link test step with function, or using context.
There is 3 variants to link code with step:

• position in case
• step name
• step uniq hash

from qaseio.pytest import qase
@qase.step(1)  # position
def some_step():
    sleep(5)
@qase.step("Second step")  # test step name
def another_step():
    sleep(3)
...
def test_example():
    some_step()
    another_step()
    # test step hash
    with qase.step("2898ba7f3b4d857cec8bee4a852cdc85f8b33132"):
        sleep(1)

Sending tests to existing testrun

Testrun in TMS will contain only those test results, which are presented in testrun,
but every test would be executed.

pytest \
    --qase \
    --qase-api-token=<your api token here> \
    --qase-project=PRJCODE \ # project, where your testrun exists in
    --qase-testrun=3 # testrun id

Creating testrun base on testplan

Create new testrun base on testplan. Testrun in TMS will contain only those
test results, which are presented in testrun, but every test would be executed.

pytest \
    --qase \
    --qase-api-token=<your api token here> \
    --qase-project=PRJCODE \ # project, where your testrun exists in
    --qase-testplan=3 # testplan id

Creating new testrun according to current pytest run

Testrun in TMS will contain only those test results, which has correct case ids,
but every test would be executed.

pytest \
    --qase \
    --qase-api-token=<your api token here> \
    --qase-project=PRJCODE \ # project, where your testrun would be created
    --qase-new-run

Debug information

If you specify --qase-debug parameter you would get additional output:

=================================== Qase TMS ===================================
This tests does not have test case ids:
test_no_deco
For test test_complex_run.py::test_multiple_ids_fail could not find test cases in run: 3
=========================== Qase TMS setup finished ============================

Execution logic

1. Check project exists
2. Check testrun exists
3. Load all ids for each test-case
4. Check which tests does not have ids (debug: will list them all)
5. Check every id exists in project (debug: will show which missing)
6. Check every id present in testrun (debug: will show which missing)
7. Execute tests and publish results in a runtime,
   not waiting all run to finish