← Back to team overview

dolfin team mailing list archive

Re: [Bug 424411] [NEW] Inconsistency in form_data upon multiple runs

 



Anders Logg wrote:
On Fri, Sep 04, 2009 at 07:35:12PM +0200, Johan Hake wrote:
On Friday 04 September 2009 18:17:11 Garth N. Wells wrote:
Kristian Oelgaard wrote:
I got this e-mail twice, from Launchpad and DOLFIN. If I go to:

https://bugs.launchpad.net/dolfin/+bug/424411

I see the Marie is in the subscribers section and:

Andy, DOLFIN mailing list, FEniCS team and Johan Hake are in the also
notified section. I'm a member of both the DOLFIN mailing list and the
FEniCS team, is that why I got two, and did Johan Hake get three?
I assigned the bug to Johan which probably accounts for one copy.
Thanks! ;)

Yes I got three, but that is because as Garth said, I was assigned the bug.
But I have made the smart move of registering with my personal email instead
of the simula email I am registered with on the dolfin list. This might give
me the email twice. I know that email list use to figure out when I user with
the same email gets an email twice, and only deliver one.

Yes, that's the case when cross-posting to foo-dev@xxxxxxxxxx and bar-dev@xxxxxxxxxx.

However, when launchpad sends email, one is sent to each member of
FEniCS Team directly and then one mail is sent to dolfin-dev, which
then sends out to each member. I don't think there is any way for
mailman on fenics.org to figure out that I don't need to get a second
email because launchpad has sent me a copy before... :-)


We might be able to change Launchpad so it doesn't send to the FEniCS Team.

Garth

--
Anders



Johan

Garth

Kristian

Quoting Marie Rognes <meg@xxxxxxxxx>:
Public bug reported:


The representation of a form changes somewhat erratically when the same
code is run multiple times

Cf:

{ v_0 * <Function in <Function space of dimension 9 (<CG1 on a <triangle
of degree 1>>)>> } * dx0
{ v_0 * w_0 } * dx0

as output from:

from dolfin import *

for i in range(2):

    mesh = UnitSquare(2, 2)
    V_h = FunctionSpace(mesh, "CG", 1)
    f = Function(V_h, "1.0")
    u = TrialFunction(V_h)
    v = TestFunction(V_h)
    a = inner(grad(v), grad(u))*dx
    l = dot(v, f)*dx
    pde = VariationalProblem(a, l)
    print pde.L.form_data.form

** Affects: dolfin
     Importance: Undecided
         Status: New


Status in DOLFIN: New

Bug description:

The representation of a form changes somewhat erratically when the same
code is run multiple times

Cf:

{ v_0 * <Function in <Function space of dimension 9 (<CG1 on a <triangle
of degree 1>>)>> } * dx0
{ v_0 * w_0 } * dx0

as output from:

from dolfin import *

for i in range(2):

    mesh = UnitSquare(2, 2)
    V_h = FunctionSpace(mesh, "CG", 1)
    f = Function(V_h, "1.0")
    u = TrialFunction(V_h)
    v = TestFunction(V_h)
    a = inner(grad(v), grad(u))*dx
    l = dot(v, f)*dx
    pde = VariationalProblem(a, l)
    print pde.L.form_data.form
_______________________________________________
DOLFIN-dev mailing list
DOLFIN-dev@xxxxxxxxxx
http://www.fenics.org/mailman/listinfo/dolfin-dev
_______________________________________________
DOLFIN-dev mailing list
DOLFIN-dev@xxxxxxxxxx
http://www.fenics.org/mailman/listinfo/dolfin-dev
_______________________________________________
DOLFIN-dev mailing list
DOLFIN-dev@xxxxxxxxxx
http://www.fenics.org/mailman/listinfo/dolfin-dev

_______________________________________________
DOLFIN-dev mailing list
DOLFIN-dev@xxxxxxxxxx
http://www.fenics.org/mailman/listinfo/dolfin-dev

------------------------------------------------------------------------

_______________________________________________
DOLFIN-dev mailing list
DOLFIN-dev@xxxxxxxxxx
http://www.fenics.org/mailman/listinfo/dolfin-dev




References