[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [p2-dev] FW: P2 Profile Questions
|
>Maybe you could publish a p2 metadata spec ?
The externalized format of an IU (p2 metadata) is left unspecified on purpose. The externalized format is the responsibility of the metadata repository.
>Maybe the people could generate the metadata they need by themselves if they feel that the standard tooling is not what they need?
This is what the p2 publisher allows.
![Inactive hide details for Alessandro Mottadelli ---21/11/2008 07:52:15 AM---Pascal Rapicault ha scritto: >]()
Alessandro Mottadelli ---21/11/2008 07:52:15 AM---Pascal Rapicault ha scritto: >
![]()
From: | ![]()
Alessandro Mottadelli <amottadelli@xxxxxxxx> |
![]()
To: | ![]()
P2 developer discussions <p2-dev@xxxxxxxxxxx> |
![]()
Date: | ![]()
21/11/2008 07:52 AM |
![]()
Subject: | ![]()
Re: [p2-dev] FW: P2 Profile Questions |
Pascal Rapicault ha scritto:
>
> The overall idea of creating what you need (metadata and profile) as
> part of the build is good and I believe this is the path to follow.
> What I like about this is that by the same token that the eclipse you
> download today result from a provisioning operation, you can then see
> the WAR you ship resulting itself from a provisioning operation and it
> makes things consistent.
> I think that what made this whole experience weird for you is
> deficiencies in our tooling / metadata generation tooling that is
> still very driven of complete products in a standard eclipse shape.
> This is something that is being improved as we speak and we are
> looking at ways to generate metadata from source information and not
> require the product or whatever you build to be in its final shape.
> We are mostly focused on RCP product builds, so please make sure to
> follow what is going there to make sure your scenario is addressed if
> possible.
>
> One last thing, please try to not use the "dropins" folder to
> provision... this is the worst way. Try invoking the director
> directly, this will be much cleaner.
>
> PaScaL
>
Maybe you could publish a p2 metadata spec ?
Maybe the people could generate the metadata they need by themselves if
they feel that the standard tooling is not what they need?
_______________________________________________
p2-dev mailing list
p2-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/p2-dev
