I think that for issue 325 we don't need to change anything. All we need is to upgrade the command to work with wildchars.
For the rest issues, I feel that we need to keep the existing way of working (use of bundle ids) and possibly provide alternative ways with the use of some flags. Now regarding the flags I would prefer to the bundle location (mvn:/groupId/arifactId/version or file:/folderA/folderB). I am not sure how we could use autocompletion with flags. If we are to keep the existing way "identifying bundles" then I vote +1 on adding it to 2.2.x, if we are to change it then it would be better to move it 3.x. On Mon, Mar 28, 2011 at 11:46 AM, Jean-Baptiste Onofré <[email protected]>wrote: > Hi all, > > I'm working on KARAF-452, KARAF-33 and KARAF-325. > > These three jira are related the same topic: identification of bundles and > features. > > I propose the following convention to identify a bundles or features. > The identifier could be: > > - a id (long) (as it is currently) > - a name (for instance bundle_A), the name could contain regexp (for > instance bundle_*) > - a name and a version (for instance bundle_A/1.0.0), the name and version > could contain regexp (for instance bundle_*/1.0.0 or bundle_A/1*) > - a range of id (for instance 1-4 7-9 will install bundles with id 1, 2, 3, > 4 and 7, 8, 9) > > WDYT ? > > I made good progress on the trunk. > I wonder if I merge this to karaf-2.2.x branch. I think it could be > interesting to provide this in the next 2.2.2 version. > > WDYT ? > > Thanks, > Regards > JB > -- *Ioannis Canellos* * http://iocanel.blogspot.com Apache Karaf <http://karaf.apache.org/> Committer & PMC Apache ServiceMix <http://servicemix.apache.org/> Committer *
