Totally my fault man... I hadn't updated mesh.C yet... so when I did
"cvs diff" it was still diffing against the version I checked out.
Updating mesh.C I now see that the new constructor got commited.
On 6/27/06, Roy Stogner <roystgnr@...> wrote:
> On Tue, 27 Jun 2006, Derek Gaston wrote:
> > I told Roy yesterday that we missed a couple of things with our new
> > mesh copy constructor...
> > Firstly we weren't copying over the refinement flag to the newly
> > created elements... leaving the entire tree active!
> > Secondly prepare_for_use() needs to get called to fix up the neighbor
> > relations and whatnot.
> > Attached is my modified copy constructor that does these two things...
> > I have been using it with my code so it atleast partially works.
> > I didn't provide it in patch form because I don't know what everyone
> > has... it doesn't look like the last patch was commited to CVS so I
> > can't patch against that...
> I'm confused - when you say "it doesn't look like the last patch was
> commited to CVS" do you mean that something didn't get committed at
> all? I think I reedited the last patch of yours, so I'm afraid I
> probably scrambled up the white space and comments that way... but I
> don't remember seeing a patch that never got committed at all.
> > so to avoid confusion I just copied out the entire copy constructor.
> This looks good. Unless gcc catches some problem I missed, I'll
> commit it in a few minutes.