https://github.com/dcramer/kleenex
A Nose plugin designed to detect coverage and only run the tests that matter.
https://github.com/dcramer/kleenex
Last synced: 4 months ago
JSON representation
A Nose plugin designed to detect coverage and only run the tests that matter.
- Host: GitHub
- URL: https://github.com/dcramer/kleenex
- Owner: dcramer
- License: apache-2.0
- Created: 2011-12-01T07:30:16.000Z (over 13 years ago)
- Default Branch: master
- Last Pushed: 2012-01-05T02:06:50.000Z (over 13 years ago)
- Last Synced: 2025-02-06T11:18:56.369Z (4 months ago)
- Language: Python
- Homepage: http://code.disqus.com
- Size: 199 KB
- Stars: 18
- Watchers: 4
- Forks: 0
- Open Issues: 2
-
Metadata Files:
- Readme: README.rst
- License: LICENSE
Awesome Lists containing this project
README
kleenex
=======
**Under Development!**
kleenex attempts to stop the pain associated with a large test suite by
only running the tests applicable to the changes in your [git] branch.
About
-----
Assuming ``--with-kleenex``, kleenex is fully active on all test runs. What this means, is it will attempt
to connect to a coverage database (sqlite:///coverage.db by default) and determine which tests it needs to run
based on the code changes. If data for a test is missing in the database (e.g. the test is new), it will include
that test in the suite automatically. This is fine for prototyping but not generally useful in the real world.
Kleenex also checks for missing coverage along the way. By default it will raise an error when it hits a code
change that has no test coverage. To disable this functionality, you should set ``skip_missing`` in your config.
Real World Usage
----------------
Kleenex is designed to work in CI environments. Generally, you would setup your CI server to run it in record
mode (with ``record``), and your clients (yourself, other developers) would simply connect to this
database to discover coverage. This ensures that the installation stays aware of your parent branch (e.g. master)
and doesn't record data from children.
Configuration
-------------
All configuration for Kleenex is read from your ``setup.cfg``. By default it reads from the [kleenex] section,
which can be changed by passing ``--kleenex-config-section=foo``.
Example configuration for a master postgres CI server::
# nosetests --with-kleenex --kleenex-config-section=kleenex:master
# setup.cfg
[kleenex:master]
record = true
skip_missing = true
parent = origin/master
db = postgres://postgres@localhost:5432/kleenex
test_missing = true
report = false
discover = false
You can also change the file which is read (setup.cfg by default) using ``--kleenex-config``.
Options
=======
The following options are recognized by Kleenex::
db
SQLAlchemy compatible DSN to connect to the database.
test_missing
Run tests which are missing coverage (generally they are new).
discover
Discover tests to run using database.
skip_missing
Allow missing coverage in discovery mode.
parent
Parent commit that your branch was based from.
report
Generate a coverage report against your diff.
report_output
Location to output report. If provided will record as JSON. For stdout/stderr you can use stream://stderr.
record
Record test coverage to database.
max_distance
Maximum distance from plugin integration of test for it to be recorded
max_revisions
Maximum number of revisions (ordered by commit_date) to maintain coverage for. Defaults to 100.