As discussed on the last Mylyn call, I have gone ahead with the removal of the OSLC connector from master. The connector was always intended as a framework for building OSLC integrations but was never shipped and hasn't been updated in the past few release cycles. More discussion and background can be found on the task:
421071: remove OSLC connector from Mylyn https://bugs.eclipse.org/bugs/show_bug.cgi?id=421071 This removes the dependency on JDOM. Steffen On Fri, Nov 8, 2013 at 4:31 PM, Steffen Pingel <[email protected]>wrote: > Let's just use this bug to discuss further :) > > 421071: Remove jdom > https://bugs.eclipse.org/bugs/show_bug.cgi?id=421071 > > Steffen > > > > On Fri, Nov 8, 2013 at 1:54 PM, Steffen Pingel <[email protected] > > wrote: > >> Could you open a bug about the OSLC connector? It's only distributed as >> part of the SDK at the moment as it's only intended as a framework for >> concrete connectors. It's not entirely clear to me at what level it has >> been adopted and we should discuss what the right place is for the >> connector moving forward. >> >> Steffen >> >> >> On Mon, Nov 4, 2013 at 3:45 PM, Krzysztof Daniel <[email protected]>wrote: >> >>> Hey Mylyn masters, >>> >>> My attention was brought recently to the fact that rome has been >>> revived, and that version 2.0 is going to be shipped soon (it looks like >>> there is only 6 more bugs planned for that release, although the date >>> has not been scheduled yet). However, the rome 2.0 requires jdom2, which >>> unfortunately is not compatible with "old" jdom - the namespace has been >>> changed. >>> >>> As I'd like Mylyn dependencies to be up to date, I thought about further >>> removal of rome and jdom (partially done in 412953). >>> >>> JDom itself is used in o.e.m.oslc.core, which has all classes internal, >>> and the OSLC spec that is parsed also doesn't seem to be very >>> complicated. Maybe it would be possible to remove jdom from this bundle >>> entirely? The main reason I'm asking about this is that I know OSLC is >>> important for adopters, and changing critical parts of code may have >>> unpredictable consequences. Yet, this part seems to be perfectly doable. >>> >>> The same case with o.e.m.web.tasks. The ROME+JDOM is used to get author, >>> date, link, and title, so I believe this could be easily parsed using >>> just pure java. I'd have to dig and compare various RSS standards to >>> estimate what will be the code added, but I hope it will be one >>> implementation for Atom and one for RSS - nothing really fancy, as mylyn >>> doesn't seem to use advanced capabilities of either protocol. >>> >>> I'm willing to work on that (although have no idea when). Thoughts? >>> -- >>> Krzysztof Daniel <[email protected]> >>> Red Hat >>> >>> _______________________________________________ >>> mylyn-dev mailing list >>> [email protected] >>> https://dev.eclipse.org/mailman/listinfo/mylyn-dev >>> >> >> >> >> -- >> Steffen Pingel >> Principal Software Engineer, Eclipse Mylyn >> Mylyn Tasks Lead >> http://tasktop.com >> > > > > -- > Steffen Pingel > Principal Software Engineer, Eclipse Mylyn > Mylyn Tasks Lead > http://tasktop.com > -- Steffen Pingel Principal Software Engineer, Eclipse Mylyn Mylyn Tasks Lead http://tasktop.com
_______________________________________________ mylyn-integrators mailing list [email protected] https://dev.eclipse.org/mailman/listinfo/mylyn-integrators
