[Users] [Commits] [svn:einsteintoolkit] TOVSolver/trunk/ (Rev. 137)

[Frank Loeffler]

Hi,

On Sun, Jul 07, 2013 at 11:06:26AM -0700, Roland Haas wrote:
> This commit (and the following ones) make the tests fail, at least in
> subversion (Zelmani is ok since the author *does* provide updated tests,
> just not in GRHydr/svn, so I could port over the change).

Regardless of what happens here: the tests should be updated accordingly
if this is necessary.

> I am personally generally unhappy about commits that require everybody
> to change/update their paramater files unless the change is required to
> prevent a bug from manifesting. Removing "[0]" from ones parameter file
> is not hard, but will break each and every single one of them. One of
> the nice things about Cactus has been that (some parts of it) are fairly
> stable and developers take care to not introduce changes that require
> each user to update their parameter files (unless the user wants/needs
> new features).

In general, I agree, but sometimes I see that changes are necessary,
even without bugs being involved. A code cleanup from time to time is
good and might involve some changes - in moderation of course.

> I am also not sure I agree with the statement that one cannot have
> different Gama or K for different stars.

I guess the argument is that there is only _one_ EOS in physics in
theory. Of course, what we have here is an approximation, and different
stars might use different approximations, resulting in the same
functional form of a polytropic EOS, but with different parameters.
Something like a piecewise polytropic EOS might be another solution for
this problem, but why should I dictate someone else's choice.
Also, someone might want to perturb one of the stars but not another by
changing its EOS initially.

> I would like to revert parts of this patch (keeping the atmosphere
> treatment that prevents asymetries in the data setup and seems to be
> currently buggy) and restore the possibility to have different K and
> Gamma (indep of whether this is physical I don't see a point in removing
> this functionality).

I agree. Could you provide a patch? Having said that, I would like to
use the opportunity to thank Christian for working on this. The
TOVSolver is indeed in dire need for cleanup and I very much welcome him
working on it. I hope the correct "translation" of "Wo gehobelt wird,
fallen Späne" is really "You can't make an omelet without breaking
eggs", but it is certainly the same spirit.

Frank

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
Url : http://lists.einsteintoolkit.org/pipermail/users/attachments/20130707/6158cd9a/attachment.bin