Ken Gentle wrote:
Guys, if you're willing to introduce a JNI lib for VMS, all the operations you wish to perform are available through VMS System Services, including getting/setting the Logical Names, and, IIRC, launching a DCL interpreter directly, avoiding the need for the DCL wrapper.
It's been a while (like 1997!) since I worked on VMS, and I don't have access to a VMS box right now, but I'd be willing to chip in on the JNI library if that is acceptable.
Downside is that you'll have to ship the VMS shared library, or the ant installation process would need to include the source with instructions on how to compile and link it (or a DCL script to do so).
Just another option...
if we were going to add JNI support to ant core tasks, I'd leave VMS alone and add the ability to get & retain execute permissions in unix :)
Regarding overall VMS support, given my day job PC is definitely somewhere with HP's class A domain, it is probably my moral responsibility to help get this working. Maybe I could grab the itanium 1box down the corridor that we are all scared of because it sounds like
a stress test of a rolls royce aero engine (I know, I used to work down the road from RR Aero engines), put an early build of vms for ia64 on it and gump it up. But I am not sure I have the time, and I lack enough morals to be morally responsible.
Maybe we could start with a run of the full ant unit test suite on VMS to see what happens.
-steve
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
