Hi all, https://issues.jboss.org/browse/JBIDE-19005 : We'd like to move the master builds to Mars M4, that means concretely that we want to update the 4.50.0.Alpha1-SNAPSHOT target-platfroms. That should automatically fix at least one bug in THyM. Jira contain the p2diff details of what is added/removed with this change. This change also *removes org.apache.jasper*

and its dependencies (the dependency chain for jasper got huge on Orbit, and it was simpler to remove it than to cascade inclusion of all dependencies).

I made a quick check and did not find any reference to it in our sources so I guess it's fine. However, I'd like you to double check that you don't have dependency on one of the bundles removed (see p2diff on jira) before we merge this change.

On 01/08/2015 11:43 PM, Max Rydahl Andersen wrote: >> and its dependencies (the dependency chain for jasper got huge on >> Orbit, and it was simpler to remove it than to cascade inclusion of >> all dependencies). > is that your comment or some decision on Eclipse.org side ? This is my comment.

> I wonder though if parts of WTP uses/needs it ? Target-platform-validation is expected to fail because of a missing dependency in that case. However, I can see that org.apache.jasper.glassfish bundle still remains (it's actually included in JBDS 8).

On 01/09/2015 09:58 AM, Max Rydahl Andersen wrote: > On 9 Jan 2015, at 9:40, Mickael Istria wrote: > >> On 01/08/2015 11:43 PM, Max Rydahl Andersen wrote: >>>> and its dependencies (the dependency chain for jasper got huge on >>>> Orbit, and it was simpler to remove it than to cascade inclusion >>>> of all dependencies). >>> is that your comment or some decision on Eclipse.org side ? >> This is my comment. > so i'm a bit puzzled...if apache.jasper is part of eclipse release > train why do we even need to chase it on orbit ? > why is this dependency in here in the first place ? > if not part of release train there should be a reason.. Actually, it's not part of the Eclipse release train. Nothing on the release train depends on it. I don't know why it used to be in our target definition, and I imagine that it might have been there for some reason in the past, and that it has become useless. Let's give a few days (until Friday) for developers to raise their hand if they actually need it, if not, let's remove it; and in case something is actually failing after that, we'll re-add jasper and its family, and document on the target files why it needs to be here.