runbot

mirror of https://github.com/odoo/runbot.git synced 2025-03-19 09:25:46 +07:00

History

Xavier Morel a564723fd8 [ADD] runbot_merge: indication of status states in the mergebot Currently it can be difficult to know why the mergebot refuses to merge a PR (not that people care, they generally just keep sending new commands without checking what the 'bot is telling them, oh well...). Anyway knowing the CI state is the most complicated bit because the CI tag only provides a global pass/fail for statuses but not a view of specific statuses, and sometimes either the runbot or github fails to notify the mergebot, leading to inconsistent internal states & shit. By adding a tag per status context per PR, we can more clearly indicate what's what. Fixes #389		2020-07-27 12:56:41 +02:00
..
conftest.py	[IMP] runbot_merge: split staging cron in two	2020-02-11 08:07:57 +01:00
README.rst	[IMP] runbot_merge: test readme & remove leftover pass	2018-09-03 13:16:36 +02:00
test_basic.py	[ADD] runbot_merge: indication of status states in the mergebot	2020-07-27 12:56:41 +02:00
test_by_branch.py	[IMP] runbot_merge: provide for statuses being only on PRs or stagings	2020-07-14 13:34:05 +02:00
test_multirepo.py	[IMP] runbot_merge: don't inject `Related` if related PR already referenced	2020-03-04 11:56:01 +01:00
test_oddities.py	[ADD] runbot_merge: allow overriding statuses	2020-07-14 13:34:05 +02:00
test_utils.py	[IMP] *: unify gh test API between runbot and fw-bot	2019-10-10 10:11:48 +02:00

README.rst

Execute this test suite using pytest.

The default mode is to run tests locally using a mock github.com.

See the docstring of remote.py for instructions to run against github "actual"
(including remote-specific options) and the end of this file for a sample.

Shared properties running tests, regardless of the github implementation:

* test should be run from the root of the runbot repository providing the
  name of this module aka ``pytest runbot_merge`` or
  ``python -mpytest runbot_merge``
* a database name to use must be provided using ``--db``, the database should
  not exist beforehand
* the addons path must be specified using ``--addons-path``, both "runbot" and
  the standard addons (odoo/addons) must be provided explicitly

See pytest's documentation for other options, I would recommend ``-rXs``,
``-v`` and ``--showlocals``.

When running "remote" tests as they take a very long time (hours) ``-x``
(aka ``--maxfail=1``) and ``--ff`` (run previously failed first) is also
recommended unless e.g. you run the tests overnight.

``pytest.ini`` sample
---------------------

.. code:: ini

    [github]
    owner = test-org
    token = aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa

    [role_reviewer]
    name = Dick Bong
    user = loginb
    token = bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb

    [role_self_reviewer]
    name = Fanny Chmelar
    user = loginc
    token = cccccccccccccccccccccccccccccccccccccccc

    [role_other]
    name = Harry Baals
    user = logind
    token = dddddddddddddddddddddddddddddddddddddddd