Re: Proposal: Remove the "user-command-packages" system property

2017-05-25 Thread Darrel Schneider
The only concern I would have is that this might cause gfsh to load other things that implement the spring CommandMarker that have nothing to do with gfsh simply because they are in one of the jars. I really don't know how likely that would be to happen but it seems more likely since the scan happe

Re: Proposal: Remove the "user-command-packages" system property

2017-05-25 Thread Jinmei Liao
both sides. On Thu, May 25, 2017 at 2:16 PM, Darrel Schneider wrote: > What JVM does this scanning? Is it just the top-level shell JVM or does it > happen on the jmx-manager side? > > On Thu, May 25, 2017 at 1:01 PM, Jared Stewart > wrote: > > > It sounds like we're on the same page. > > > > Th

Re: Proposal: Remove the "user-command-packages" system property

2017-05-25 Thread Darrel Schneider
What JVM does this scanning? Is it just the top-level shell JVM or does it happen on the jmx-manager side? On Thu, May 25, 2017 at 1:01 PM, Jared Stewart wrote: > It sounds like we're on the same page. > > The current behavior requires two pieces from a user who wants to add > commands: > 1) imp

Re: Proposal: Remove the "user-command-packages" system property

2017-05-25 Thread Jared Stewart
It sounds like we're on the same page. The current behavior requires two pieces from a user who wants to add commands: 1) implement CommandMarker 2) Add a system property pointing to the package of your class, or provide a META-INF.services file pointing to your class The proposed behavior is to

Re: Proposal: Remove the "user-command-packages" system property

2017-05-25 Thread Udo Kohlmeyer
What I meant was, given our hammer of choice for GFSH is Spring Shell, all command definitions should be of type CommandMarker. Not sure how it loads the correct CommandMarker classes (for different commands). But my gut feel is, have single way of doing something... KISS On 5/25/17 12:22, J

Re: Proposal: Remove the "user-command-packages" system property

2017-05-25 Thread Jinmei Liao
+1 for scanning the whole classpath without eager loading. On Thu, May 25, 2017 at 11:52 AM, Jared Stewart wrote: > I would like to propose that we eliminate the “user-command-packages” > system property, in favor of scanning the entire classpath to find commands. > > To give more detail, Geode

Re: Proposal: Remove the "user-command-packages" system property

2017-05-25 Thread Jared Stewart
Can you clarify - I'm proposing we use *implements CommandMarker* alone as the way to specify commands. What do we gain be *also* requiring a META-INF.services file? On May 25, 2017 12:18 PM, "Udo Kohlmeyer" wrote: imo, I think that GFSH is Spring Shell, it should really only be commands that ar

Re: Proposal: Remove the "user-command-packages" system property

2017-05-25 Thread Udo Kohlmeyer
imo, I think that GFSH is Spring Shell, it should really only be commands that are registered inside of META-INF.services .. aka implements CommandMarker. This way we has standard simple way to specify commands. --Udo On 5/25/17 11:52, Jared Stewart wrote: I would like to propose that we el