Will we still be able to run testng tests w/o osgi? Or can osgi load
times be kept very very low?
Currently many unit tests don't configure osgi because it is the
difference between a test taking <1s to run and >10s to run. To
encourage test-driven dev I'd like to keep that low round-trip time.
--A
On 12/10/2015 17:40, Ciprian Ciubotariu wrote:
Oh, I see.
The pax-exam runner has a testng driver, which means that the general aspect
of tests will remain the same. However, pax-exam will start/stop the osgi
framework before running tests and inject various needed services into the
test class. One such service might be the framework itself.
Cipi
On Monday 12 October 2015 16:31:22 Aled Sage wrote:
Thanks Ciprian,
Yes, I think that answers my question.
I assumed you would say pax-exam runner :-) I'm not familiar enough with
it to picture what the tests would look like and how it would change the
way we run our tests (both in the IDE and from jenkins), but that
doesn't need described/addressed right now.
Aled
On 12/10/2015 16:25, Ciprian Ciubotariu wrote:
On Monday 12 October 2015 15:35:24 Aled Sage wrote:
+1; sounds sensible.
Would brooklyn-core have a dependency on brooklyn-rt-felix?
Yes, during build time and with the current launcher(s). However, the
brooklyn-core OSGi bundle won't import the packages of brooklyn-rt-felix,
and the osgi container services will be used instead.
Presumably the unit tests in core would continue to run outside of Karaf
(e.g. org.apache.brooklyn.core.mgmt.osgi.OsgiVersionMoreEntityTest)? How
would you imagine those tests be run/rewritten once the OSGi'fication is
complete?
At the moment they will continue to be executed by the testng runner, but
they would be migrated to the pax-exam runner. Was this the target of
your question?
Aled
On 12/10/2015 14:53, Ciprian Ciubotariu wrote:
While working on running Brooklyn inside Karaf I ran into the code that
manages an embedded Apache Felix runtime.
Since the roles are now reversed, I propose to split brooklyn-core into
the
following:
* brooklyn-core - new implementation using the Karaf-provided OSGi
runtime
* brooklyn-rt-felix - existing code using felix as an embedded OSGI
runtime
(which will no longer be needed when OSGification is complete)
This change is much needed since the Apache Felix packages that are used
to
start and stop the embedded framework are private within the felix
bundle
My intention is to move current code to brooklyn-rt-felix, and defer to
it
when the Karaf container is not present, while using the Karaf-provided
felix/equinox framework by default.
I have already started to code in that direction, but since this is a
larger change I would like to hear your thoughts on the matter.
--
Cloudsoft Corporation Limited, Registered in Scotland No: SC349230.
Registered Office: 13 Dryden Place, Edinburgh, EH9 1RP
This e-mail message is confidential and for use by the addressee only. If
the message is received by anyone other than the addressee, please return
the message to the sender by replying to it and then delete the message
from your computer. Internet e-mails are not necessarily secure. Cloudsoft
Corporation Limited does not accept responsibility for changes made to this
message after it was sent.
Whilst all reasonable care has been taken to avoid the transmission of
viruses, it is the responsibility of the recipient to ensure that the
onward transmission, opening or use of this message and any attachments
will not adversely affect its systems or data. No responsibility is
accepted by Cloudsoft Corporation Limited in this regard and the recipient
should carry out such virus and other checks as it considers appropriate.
