Re: [egit-dev] Fwd: [cross-project-issues-dev] Juno SR2 RC4 staging repo

[Matthias Sohn <matthias.sohn@xxxxxxxxx>]

2013/2/14 Robin Stocker <robin@xxxxxxxxx>

Matthias Sohn wrote:
> 2013/2/14 Daniel Megert < daniel_megert@xxxxxxxxxx >
>
>
> Agreed. The duplicate change-Id probably just looks ugly but won't
> block the interaction with Gerrit.
>
> no, it's not just ugly but it really does the wrong thing since if
> you already pushed a change for review,
> then amend it e.g. to react on comments. This will add another new
> change-id at the end. Then when
> you push this amended commit in order to create patchset 2 for your
> change you will endup with a
> new duplicate change instead of the second patchset for the already
> existing change.

I don't think it's critical because it only happens on commit message
footers that are not formatted correctly, see my comment on the bug:

https://bugs.eclipse.org/bugs/show_bug.cgi?id=400818#c1

the native commit message hook is more tolerant and we should

fix jgit/egit to behave in the same way (see comment on the bug).

I didn't spot this when reviewing Stefan's changes.

--

Matthias