Yep, there will be an IBM build server, that is on my list.
But first we have to succeed with an "mvn -Dall clean install" on 2.5.x and
2.6.x otherwise the ibm hudson will start crying around with errors :-(
Andrea Aime writes:
> Christian Müller ha scritto:
>> The mehtod System.getenv() was depricated and afterwards reinstated in
>> htm l
>> Sun and IBM sdk have a different behavior.
>> Concerning following properties
>> Sun sdk returns a value if you query with System.getenv(..) and
>> System.getPropertiy whilst IBM sdk returns NULL when querying a Java
>> property with System.getenv(...).
>> So please, as a rule, use System.getProperty(..) when querying a Java
>> property and System.getenv(..) when querying an OS environment variable.
>> Otherwise I would always have to create a JIRA, which should not be
>> necessary. Thanks
> Christian, a way to help avoiding
> other jiras is to stand up a continous build server that can
> point a finger to whoever makes a change that does not work with
> IBM JDK.
> Time passes, people working on the project change, this mail
> will eventually be forgotten, but a continous build server
> that operates on IBM JDK will point out any misbehaviour as
> it happens.
> And then there is a need for a person or a group of persons
> that do care about fixing the issues, which in this specific
> case, it's you.
> I don't want to sound rude,
> but this IBM JDK "requirement" is extra work that at the
> moment benefits only you, so it's also natural that your
> is the burden to keep the build going on in that environment.
> Hopefully people will try to help out (as I'm doing),
> but as you know, time and people is something we have
> in short supply, so try to be patient, and point out IBM failures with
> explanations and the like as they happen (especially cases
> like this where the two JDK are taking two different
> positions and behaviors for certain methods).
> Andrea Aime
> OpenGeo - http://opengeo.org
> Expert service straight from the developers.