[Insight-developers] Fwd: fix backport into 3.20, 3.21?

Alexandre GOUAILLARD agouaillard at gmail.com
Thu Sep 23 07:45:52 EDT 2010


hi brad,

just a few questions regarding the releases.

before, for each release, we tagged the repository and made a branch.
Only the gatekeeper could commit in branch later.

now I guess with git things have changed a little bit.

My understanding is that the Release branch right now correspond to
itk 3.21 and is where we should commit (critical) bug fixes for 3.20.
Is it correct? Then those bugs are merged into master.

Everybody with write access to the repository would have write access
to release branch as well.
- Do we still have a gatekeeper?
- Who should / should not commit to the release branch, and what is the process?

what happen if a fix was commited first to master (like
fd39d4138c579a3526511f21a12d3b03a611fa41)
and then we realized it should also apply to another branch?

thanks in advance.

alex.
PS: luis, yes I will create a bug and assign it to you, along with the
patch. However I want to be sure of the process as I am going to teach
a few guys about it by the end of the year.

---------- Forwarded message ----------
From: kent williams <norman-k-williams at uiowa.edu>
Date: Thu, Sep 16, 2010 at 10:15 PM
Subject: Re: [Insight-developers] Problem with ITK 3.20
itkQuadEdgeMeshDiscretePrincipalCurvaturesEstimator.h
To: Luis Ibanez <luis.ibanez at kitware.com>, Alexandre GOUAILLARD
<agouaillard at gmail.com>
Cc: Hans Johnson <hans-johnson at uiowa.edu>, ITK <insight-developers at itk.org>


I'm still grabbing the ITK-3-20 tag for our production code.  I think if
enough critical bug fixes accumulate, a new minor release might make sense.
I don't think you all would just slipstream fixes in under that tag, right?

The QuadMeshDiscretePrincipalCurvaturesEstimator fix isn't a showstopper;
the fix I suggested just helps the compiler to disambiguate overloaded
functions.  The code still works -- you just have to use double instead of
float for the output mesh pixel type.

So it isn't a showstopper; I should have logged it as a bug, but initially
was more concerned with what the process to getting it fixed both places
would be.


On 9/15/10 5:08 PM, "Luis Ibanez" <luis.ibanez at kitware.com> wrote:

> Hi Alex,
>
>
> Yes, we maintain ITK 3.20.
>
> That's still the official release of ITK.
>
> Please create a bug in  MANTIS describing the issue.
>
> Assign it to me,
> and if possible indicate what code patch
> from ITKv4 must be applied to ITK 3.20.
>
>
>      Thanks
>
>
>            Luis
>


More information about the Insight-developers mailing list