Community
Participate
Working Groups
Aggregator apparently (mis)uses the p2 profile of the Eclipse IDE (or the headless aggregator) installation to load the repositories referenced from an aggregation definition file. As soon as I open an aggregation definition file, the "available software sites" list is polluted with all the repositories referenced from that file. This makes it impossible to update/install new software into the IDE without careful deletion of the added software sites/p2 repositories because of conflicting components found in those repositories. Agregator should not pollute the p2 profile used by the IDE to manage itself, instead it should create its own private p2 profile, maybe even multiple ones - a separate one for each aggregation definition file it processes.
The aggregator now uses its own agent at location "${user.home}/.b3/p2agent" and will no longer interfere with the IDE installer. The location is currently hard-coded and should be settable using preferences. See bug 308872. Fixed in revision 1514.
I can confirm that the pollution of the p2 profile has now ceased! Excellent! This makes is much easier to maintain my Aggregation Eclipse install.
[Bookkeeping change only. Moving bugs to the new "home" of aggregator, CBI. No change to assignee for resolved and verified bugs.]