[petsc-dev] preparing for a PETSc release

[Blaise Bourdin]

On May 9, 2011, at 4:03 PM, Jed Brown wrote:

> On Mon, May 9, 2011 at 22:53, Blaise Bourdin <bourdin at lsu.edu> wrote:
> How feasible would it be to mark all non-namespace names as deprecated until the next release but offer preprocessor macros in the meantime? This would buy everybody some time.
> 
> Is the problem that you want to delay running "sed" on your project, or that you want to support older versions of PETSc for a little longer? If the latter, what about a header that went the other way (update your code now, include the header when building against petsc-3.1)?

Is this -really- going to be just a few lines of sed? Were you able to sed your way out of the DA -> DMDA / Mesh -> DMMesh change? I wasn't. The technically less gifted of us are probably going to spend a lot of time going through their source code and making changes by hand. 

What you suggest amounts to modifying the current release to make it forward compatible insead of modifying the next release in order to make it backward compatible. What I am suggesting is giving one release cycle before obsoleting legacy data types. I am just trying to find a way to minimize discontent while allowing to move forward with namespacing all petsc data types, and to make it easier for people to stay up to date with the latest release.

Blaise




-- 
Department of Mathematics and Center for Computation & Technology
Louisiana State University, Baton Rouge, LA 70803, USA
Tel. +1 (225) 578 1612, Fax  +1 (225) 578 4276 http://www.math.lsu.edu/~bourdin







-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20110509/1d297bad/attachment.html>