|
Re: Recommendations for provider id when using the equinox extension registry [message #53995 is a reply to message #53907] |
Fri, 02 December 2005 01:59 ![Go to previous message Go to previous message](theme/Solstice/images/up.png) ![Go to next message Go to next message](theme/Solstice/images/down.png) |
Eclipse User![Friend of Eclipse Friend](/donate/web-api/friends_decorator.php?email=) |
|
|
|
Originally posted by: jeff_nospam_mcaffer.ca.ibm.com
Hey Jim
That long id will change very shortly to be a String. You are right in
identifying its OSGi heritage. Expect to see some new code in the repo real
soon now.
BTW, the framework assigns the long bundle ids according to the order in
which the bundles are installed. They are used to uniquely identify the
bundle without having to talk in terms of symbolic id and version. So there
is no overt corelation until you actually install the bundle. For fun,
start Eclipse with -console. Then at the osgi> prompt, type ss. you will
see a number list of bundles and symbolic names (and versions). That is the
relationship.
Jeff
"Jim D'Anjou" <danjou@us.ibm.com> wrote in message
news:8b86990febddec3587abfb51c5ff0542$1@www.eclipse.org...
> I am exploring the new extension registry that can be used independent of
> eclipse. The method IExtensionRegistry.addXMLContribution takes as input a
> provider id defined as a long. I presume this has its heritage in OSGI
> bundles.
> I would appreciate any recommendations on how to best derive a id in this
> form from a string since that is what is referenced by extension points
> and their extensions. I looked for some correlation between
> Bundle-SymbolicName (a string) and BundleId (a long) but could not find
> something helpful. I know how to derive a long from a string by obtaining
> it's hashcode but I am not sure that this approach is sufficiently robust
> to guarantee uniqueness.
>
> Thanks
> Jim D'Anjou
>
>
|
|
|
|
|
|
|
|
Powered by
FUDForum. Page generated in 0.04789 seconds