elmergrid regressions

Clearly defined bug reports and their fixes
Post Reply
NJank
Posts: 99
Joined: 05 Dec 2009, 00:05
Location: Baltimore, MD, USA

elmergrid regressions

Post by NJank »

running simple 2D grid files from tests and examples here now produce spurious warnings.

Running Windows 10, v1703. Elmer 8.4. Elmergrid does not report a version, but the elmersolver version bundled with it is
"Version: 8.3 (Rev: 432ee0dd, Compiled: 2018-12-18)". Downloaded as the "elmerfem-nightly_Windows-AMD64.zip, 2018-12-19" package

running the following build grd script:

Code: Select all

***** ElmerGrid input file for structured grid generation *****
Version = 210903
Load MATC = True

$variable1 = 1$

Coordinate System = Cartesian 2D
Subcell Divisions in 2D = 3 3 

Subcell Sizes 1 = 1e-3 1e-3 1e-3
Subcell Sizes 2 = 1e-3 1e-3 1e-3

Material Structure in 2D
  2 1 2
  2 2 2
  2 2 2
End

Materials Interval = 1 2

! Boundary table format:
Boundary Definitions
! type     out      int      double   of the boundaries
  1        -1       2       1
  !2       -2       2       1
  !2       -4       2       1
  2        -3       1       1
End

Surface Elements = 100
Triangles = logical True  !needed for RGB adaptive meshing
Numbering = Horizontal
Element Degree = 1
produces this output:

Code: Select all

elmergrid 1 2 adapt.grd

Starting program Elmergrid
Elmergrid reading in-line arguments
Output will be saved to file adapt.

Elmergrid loading data:
-----------------------
Loading the geometry from file 'adapt.grd'.
Loading ElmerGrid file version: 210903
Defining the coordinate system (2-DIM).
Loading 3 subcell sizes in X-direction
Loading 3 subcell sizes in Y-direction
Loading material structure
Loading boundary conditions
Found 2 boundaries
There is a risk that somethings was missed in line:
Element Degree = 1
The program encountered a minor error...
Check your output line length!

...we'll try to continue...
Reached the end of command file
Found 1 divisions for grid

Loading ElmerGrid commands from file 'adapt.grd'.
There is a risk that somethings was missed in line:
Element Degree = 1
The program encountered a minor error...
Check your output line length!

...we'll try to continue...
Reached the end of command file
Read commands from a file

Elmergrid creating and manipulating meshes:
-------------------------------------------
9 cells were created.
Numbered 130 knots in 108 4-node elements.
Numbering order was <x><y> and max levelwidth 12.
9 element sides between materials -1 and 2 were located to type 1.
3 element sides between materials -3 and 1 were located to type 2.
Dimension 3 not active but higher dimensions are?
Coordinates defined in 2 dimensions
There are 216 elements after triangularization (was 108)

Elmergrid saving data with method 2:
-------------------------------------
Saving mesh in ElmerSolver format to directory adapt.
Reusing an existing directory
Saving 130 coordinates to mesh.nodes.
Saving 216 element topologies to mesh.elements.
Saving boundary elements to mesh.boundary.
Saving header info to mesh.header.

Thank you for using Elmergrid!
Send bug reports and feature wishes to elmeradm@csc.fi
The mesh appears to be generated without problem. it is unclear what is generating these warnings/errors.

specific errors:

1.
Unable to activate matc as it is not even compiled.
the packaged elmergrid is now compiled without MATC support, and LUA support has not been added, so variable scripting within geometry definition has been significantly curtailed.

2.
There is a risk that somethings was missed in line:
Element Degree = 1
The program encountered a minor error...
Check your output line length!

...we'll try to continue...
It is unclear what's generating this error. it seems independent of actual line length

3.
Dimension 3 not active but higher dimensions are?
It is unclear what's generating this error.

4.
No boundary between materials 0 and 0 exists.
No boundary between materials 0 and 0 exists.
This appears to be generated by commented out lines within the Boundary definition array. deleting those comments removes this warning. note that the initial comment specifying "!type out int double" does not generate this warning, just the ones that separate the numerical part of the array. I've never had a problem with commented out lines in arrays in Elmer, and I frequently use it in elmergrid to toggle different boundary definitions.
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: elmergrid regressions

Post by raback »

Hi,

Yes, this seems to have been dropped at some stage.

In $ELMERSRC/elmergrid/src/CMakeLists.txt there is a line that does this:
ADD_DEFINITIONS(-DDISABLE_MATC)

I guess this was added to circumvent some other compilation problem. It seems that this features has been nearly forgotten and there is no ctest for that so it was dropped without anybody noticing - except you. Thanx for reporting!

We will see whether there is a quick remedy.

-Peter
NJank
Posts: 99
Joined: 05 Dec 2009, 00:05
Location: Baltimore, MD, USA

Re: elmergrid regressions

Post by NJank »

I guess that's what happens when you take a 4 year hiatus :) I wasn't sure if it was part of a shift to LUA
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: elmergrid regressions

Post by raback »

Hi

I looked at some of the errors. For example, there was new check for very long lines and this failed if the line ended not to end-of-line (\n) but to end-of-file (\0). This has now been fixed. Basically I don't think there is any other real problem than the missing of MATC which occured when ElmerGrid was moved to cmake. Should not be too much work for a cmake wizard.

-Peter
NJank
Posts: 99
Joined: 05 Dec 2009, 00:05
Location: Baltimore, MD, USA

Re: elmergrid regressions

Post by NJank »

Thanks for the quick attention. Any idea about the #3 and #4 warnings above? I think I understand what's causing 4 (commented lines mid array), but I see the "Dimension 3 not active..." warning come up in lots of grd files and I haven't been able to identify the trigger.
Post Reply