Open GoogleCodeExporter opened 9 years ago
is the mesh regenerated between saving and loading?
if that is the case then it is normal that you don't get the same number of
degrees of freedom, from one run to another of gmsh you may not get the same
mesh and hence the same number of degrees of freedom and degrees of freedom
table for that matter.
Original comment by christop...@feelpp.org
on 30 Jun 2012 at 2:35
The mesh was regenerated but when loading from database the element was not
linked to the function space and that was the mistake.
I will fix that on the test I have introduced in testsuite
Thanks
Original comment by stephane...@gmail.com
on 30 Jun 2012 at 4:05
it is now possible to save the mesh in binary format (see mymesh.cpp)
if you use this functionality you should never this problem again.
It is probablement better to do that anyway to make sure that all the data are
consistent
mesh -> functionspace -> function ->...
should this be be closed as "invalid" now ?
Original comment by christop...@feelpp.org
on 30 Jun 2012 at 4:19
This issue was closed by revision 23804690f49a.
Original comment by christop...@feelpp.org
on 2 Jul 2012 at 9:28
As shown by test_element_serialize there is no problem to save/load one
element_type in the database.
On the other hand I have a problem to load a vector of element_type with still
the same error message :
std::logic_error: load function: invalid number of degrees of freedom
I have updated test_element_serialize to show that
Concerning mymesh.cpp it seems that there is a problem when running in
sequential ( it crash )
Original comment by stephane...@gmail.com
on 2 Jul 2012 at 9:41
Original comment by stephane...@gmail.com
on 2 Jul 2012 at 1:51
As I said, there is a problem during the load of a vector of elements.
As a workaround it's possible to save separately each elements and then when
loading, to rebuild the vector.
I have updated the file test_element_serialize to show this
Original comment by stephane...@gmail.com
on 5 Jul 2012 at 9:14
Original issue reported on code.google.com by
stephane...@gmail.com
on 30 Jun 2012 at 2:18