Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [orbit-dev] How to name bundles from different implementers?

Just to be sure everyone is on the same page can we capture the current thinking on the wiki and circulate for ageement? (apologies if it is already on the wiki. I am on an airplane and cannot check)

Jeff

On Jul 21, 2008, at 3:48 PM, Thomas Watson <tjwatson@xxxxxxxxxx> wrote:

My first thought was exactly what Chris mentions with using export-package attributes. But the bundle symbolic name really should be distinct so the different bundles can be distinguished from each other.

My vote is to use the name space of the provider first followed by the package (e.g. org.glassfish.javax.mail)

Tom



<graycol.gif>David M Williams---07/20/2008 10:42:55 PM---That's what we've done so far, but, in bug 226813 it's been argued that bundle-name + version-numbe

<ecblank.gif>
From:
<ecblank.gif>
David M Williams/Raleigh/IBM@IBMUS
<ecblank.gif>
To:
<ecblank.gif>
Orbit Developer discussion <orbit-dev@xxxxxxxxxxx>
<ecblank.gif>
Date:
<ecblank.gif>
07/20/2008 10:42 PM
<ecblank.gif>
Subject:
<ecblank.gif>
Re: [orbit-dev] How to name bundles from different implementers?





That's what we've done so far, but, in bug 226813 it's been argued that bundle-name + version-number has always been presented as "one set of bits".


From: "Chris Aniszczyk" <caniszczyk@xxxxxxxxx>
To: "Orbit Developer discussion" <orbit-dev@xxxxxxxxxxx>
Date: 07/20/2008 09:06 PM
Subject: Re: [orbit-dev] How to name bundles from different implementers?






The suggestion has been made, to include "vendor" name in the bundle ID (as well as name an attribute on the
export package statements).

Can't we make the BSN to just be the package space and include the vendor info on export package statement?

In CVS, we can store things like javax.mail.{vendor}

This way at least, people are kind of forced to use Import-Package with attributes if they want a specific implementation. That is one of the strengths of attributes.

In the end, we should come up with a policy on this issue soon as I can see it coming up more in the future.

--
Cheers,

~ Chris Aniszczyk
_______________________________________________
orbit-dev mailing list
orbit-dev@xxxxxxxxxxx

https://dev.eclipse.org/mailman/listinfo/orbit-dev

_______________________________________________
orbit-dev mailing list
orbit-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/orbit-dev


_______________________________________________
orbit-dev mailing list
orbit-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/orbit-dev

Back to the top