Pickup dpg-dev and particleset/particlevector.hpp wrapping. by sshiraiwa · Pull Request #308 · mfem/PyMFEM

sshiraiwa · 2026-01-01T22:29:18Z

This PR includes

particleset.hpp, particlevector.hpp wrapper
fixed GetDataArray of Vector/DenseMatrix/DenseTensor/SparseMatrix so that
destructore is not pre-maturely called when return value is not stored to
a variable.
miniapp/dpg/utils/*.hpp is wrapped.
Python version of pmaxwell.py in miniapp/dpg directory

merge 48

…can pass list/tuple

…d5dd)

…r/SparseMatrix/DenseMatrix/DenseTensor returns numpy array via GetDataArray

Copilot

Pull request overview

This PR adds Python wrappers for MFEM's Discontinuous Petrov-Galerkin (DPG) miniapp utilities and includes a complete Python implementation of the parallel Maxwell DPG example. The key changes include:

New Python DPG Maxwell example (pmaxwell.py) with support for various problem types (plane wave, Fichera oven, PML)
Python wrappers for DPG utility classes (weakform, complexweakform, blockstaticcond, etc.)
Wrappers for particleset and particlevector modules
Fixes to GetDataArray methods to prevent premature object destruction
Build system updates to enable miniapp compilation
Support for non-square matrix coefficients

Reviewed changes

Copilot reviewed 51 out of 51 changed files in this pull request and generated 23 comments.

Show a summary per file

File	Description
miniapps/dpg/pmaxwell.py	Complete Python implementation of parallel Maxwell DPG example with multiple problem types
miniapps/dpg/meshes/*.mesh	Mesh files for DPG examples (scatter.mesh, fichera-waveguide.mesh)
mfem/ser.py, mfem/par.py	Added imports for new dpg, particleset, particlevector modules
mfem/_ser/dpg.i, mfem/_par/dpg.i	SWIG interface files for DPG utilities wrapping
mfem/_ser/vector.i, mfem/_par/vector.i	Fixed GetDataArray to prevent premature destructor calls
mfem/_ser/densemat.i, mfem/_par/densemat.i	Fixed GetDataArray for DenseMatrix and DenseTensor
mfem/_ser/sparsemat.i, mfem/_par/sparsemat.i	Fixed GetDataArray, GetJArray, GetIArray for SparseMatrix
mfem/common/numba_coefficient.	Removed square matrix requirement for MatrixCoefficient
mfem/_ser/setup.py, mfem/_par/setup.py	Added build_miniapps configuration and library dependencies
_build_system/*.py	Build system updates for miniapp support
docs/changelog.txt	Documentation of changes

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-01-01T23:56:39Z

mfem/common/numba_coefficient.i

-        warning.warn("Rectangular matrix coefficient is experimental", UserWarning)
+    #if shape[0] != shape[1]:
+    #    import warnings
+    #    warnings.warn("Rectangular matrix coefficient is experimental", UserWarning)


Typo in variable name: "warning" should be "warnings". The code should use "warnings.warn(...)" instead of "warning.warn(...)".

Copilot · 2026-01-01T23:56:39Z

mfem/par.py

+def load_module(module_name, module_code):
+    spec = importlib.util.spec_from_loader(module_name, loader=None, origin="dynamic")
+    module = types.ModuleType(module_name)
+    spec.loader.exec_module(module) if spec.loader else exec(module_code, module.__dict__)
+    sys.modules[module_name] = module
+    return module


The load_module function is defined but never used in this file. If this is intended for future use or as a utility, it should be documented. Otherwise, it should be removed to avoid dead code.

Copilot · 2026-01-01T23:56:39Z

mfem/_ser/setup.py

+    if build_miniapps ==  '0':
+        libraries.append("mfem-common")


The logic here appears inverted. When build_miniapps == '0' (False), the code adds "mfem-common" to libraries. This seems backwards - shouldn't "mfem-common" be added when miniapps ARE being built (build_miniapps == '1')? Please verify this logic is correct.

Copilot · 2026-01-01T23:56:39Z

mfem/_par/setup.py

        mpiinc = ''

    libraries = ['mfem',]
+    if build_miniapps ==  '0':


The logic here appears inverted. When build_miniapps == '0' (False), the code adds "mfem-common" to libraries. This seems backwards - shouldn't "mfem-common" be added when miniapps ARE being built (build_miniapps == '1')? Please verify this logic is correct.

Suggested change

if build_miniapps == '0':

if build_miniapps == '1':

Copilot · 2026-01-01T23:56:40Z

_build_system/build_config.py

    command_obj.mfem_branch = ''
    command_obj.mfem_debug = False
-    command_obj.mfem_build_miniapps = False
+    command_obj.mfem_miniapps = True


The default value for mfem_miniapps is changed from False to True. This is a significant behavior change that could affect existing workflows. Consider whether this should default to False for backward compatibility, or ensure this change is well-documented in the PR description.

Suggested change

command_obj.mfem_miniapps = True

command_obj.mfem_miniapps = False

Copilot · 2026-01-01T23:56:43Z

mfem/par.py



+try:
+    import mfem._par.dpg as dpg


Import of 'dpg' is not used.

Copilot · 2026-01-01T23:56:43Z

miniapps/dpg/pmaxwell.py

+from numba import njit, void, int32, int64, float64, complex128, types
+from mfem.common.bessel import yv as yn
+from mfem.common.bessel import jv as jn
+import os


Import of 'os' is not used.

Copilot · 2026-01-01T23:56:43Z

mfem/ser.py

 try:
-    import mfem._ser.complexweakform as complexweakform
-    import mfem._ser.commlexstaticcond as complexstaticcond
+    import mfem._ser.dpg as dpg


Import of 'dpg' is not used.

Copilot · 2026-01-01T23:56:44Z

mfem/par.py


+try:
+    import mfem._par.dpg as dpg
+except:


Except block directly handles BaseException.

Copilot · 2026-01-01T23:56:44Z

mfem/par.py


+try:
+    import mfem._par.dpg as dpg
+except:


'except' clause does nothing but pass and there is no explanatory comment.

sshiraiwa and others added 30 commits April 23, 2025 08:14

Merge pull request #269 from mfem/mfem_48_dev

446f080

merge 48

added wrappers for miniapps/dpg/utils

ccac1f7

Merge branch 'mfem_48_dev' into dpg_dev

0383446

minor edit to setup.py

3b56b38

Merge branch 'mfem_48_dev' into dpg_dev

6a13251

pmaxwell.py

2201fb9

pmaxwell.py

4d3b446

(WIP) work in progress

0e11b17

made ser/par.dpg module to load all dpg related routines

bb3f56e

(WIP)...pmaxwell.py

26657dc

(WIP) pmaxwell

e234fca

(WIP...)

e1914ba

merged

ceeaca7

merged

1782435

(WIP) pmaxwell.py seems okay

96a3d26

merged new build-install system

8ae8d8e

Merge remote-tracking branch 'origin/mfem_48_dev' into dpg_dev

b05be34

merged

3da4f4f

removed rectangular matrix coefficient warnings

68f5506

Merge branch 'dpg_dev' of github.com:mfem/PyMFEM into dpg_dev

f8a77e5

preparig for merge

993774c

Merge branch 'dpg_dev' of github.com:mfem/PyMFEM into dpg_dev

cf44e73

improved DPGweakform constructor. Instead of passing mfem::Array, we …

3cca3b7

…can pass list/tuple

merged perlmutter_fix

9b68fe8

(WIP)

8e69e62

added partcileset/particlevector wrapping

d75793c

merged dpg

c7ebcef

addressed change in hypre.hpp to fix const correctness (mfem/mfem@613…

2f29cfd

…d5dd)

Merge branch 'fix_hypre_typemap' into dgp_merge

69b975e

picked-up dpg-dev branch, added link to the orignal object when Vecto…

4e9df3a

…r/SparseMatrix/DenseMatrix/DenseTensor returns numpy array via GetDataArray

sshiraiwa added 2 commits January 1, 2026 17:17

merged particleset/particlevector branch

81e3435

update change log

96c2d21

sshiraiwa requested a review from Copilot January 1, 2026 23:51

Copilot started reviewing on behalf of sshiraiwa January 1, 2026 23:52 View session

Copilot AI reviewed Jan 1, 2026

View reviewed changes

fixed DenseMatrix::GetDataArray()

7ba9465

sshiraiwa mentioned this pull request Jan 5, 2026

PyMFEM 4.9 ToDo #305

Open

12 tasks

sshiraiwa added 2 commits January 6, 2026 11:34

minor fix...response to copilot...

49d6112

update shadowed method in fespace.i and pfespace.i

99ac290

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Pickup dpg-dev and particleset/particlevector.hpp wrapping.#308

Pickup dpg-dev and particleset/particlevector.hpp wrapping.#308
sshiraiwa wants to merge 35 commits intomasterfrom
dpg-merge

sshiraiwa commented Jan 1, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Jan 1, 2026

Uh oh!

Copilot AI Jan 1, 2026

Uh oh!

Copilot AI Jan 1, 2026

Uh oh!

Copilot AI Jan 1, 2026

Uh oh!

Copilot AI Jan 1, 2026

Uh oh!

Copilot AI Jan 1, 2026

Uh oh!

Copilot AI Jan 1, 2026

Uh oh!

Copilot AI Jan 1, 2026

Uh oh!

Copilot AI Jan 1, 2026

Uh oh!

Copilot AI Jan 1, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

	command_obj.mfem_miniapps = True
	command_obj.mfem_miniapps = False

Conversation

sshiraiwa commented Jan 1, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Jan 1, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 1, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 1, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 1, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 1, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 1, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 1, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 1, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 1, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 1, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant