I saw a couple of questions about how the Tinderbox scripts run tests so I figured it may be time to refresh the information.

The full-build tinderboxes (alanui, ahukini, hilo and imac1) run unit tests using rt.py after each build:

        python ./tools/rt.py -Ti -m <mode> -u

This runs for debug and release the localization check and the unit tests.

The quick-build tinderboxes (koa, mana, molokini and imac2) run rt.py 3 times after each build:

        python ./tools/rt.py -Ti -m <mode> -u
        python ./tools/rt.py -Ti -m <mode> -f
        python ./tools/rt.py -Ti -m <mode> -F

So for debug and release the localization check, unit tests (-u), functional test suite (-f), recorded tests and each functional test run in it's own process (-F) are run after each build.

The -Ti parameter tells rt.py to run the tests in tbox mode (-T) and to clear out the chandler environment variables (-i) before each run.

Both the full and quick build tinderboxes install chandler using the "distrib" target.

The performance tinderboxes (p_*) start with the same steps that the quick-build tinderboxes run except that instead of the final -F parameter they use -p which runs the performance test suite.

---
Bear

Build and Release Engineer
Open Source Applications Foundation (OSAF)
[EMAIL PROTECTED]
http://www.osafoundation.org

[EMAIL PROTECTED]
http://code-bear.com

PGP Fingerprint = 9996 719F 973D B11B E111  D770 9331 E822 40B3 CD29

Attachment: PGP.sig
Description: This is a digitally signed message part

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Open Source Applications Foundation "chandler-dev" mailing list
http://lists.osafoundation.org/mailman/listinfo/chandler-dev

Reply via email to