Weakly enforced Dirichlet BCs by RemDelaporteMathurin · Pull Request #1158 · festim-dev/FESTIM

RemDelaporteMathurin · 2026-05-18T14:02:33Z

Description

Summary

This PR adds the possibility to weakly enforce dirichlet BCs.

Related Issues

Motivation and Context

Sometimes, it is necessary to weakly impose a Dirichlet BC (eg. flux conservation). So far FESTIM only allowed strongly enforced BCs but it wasn't too complicated to weakly enforce it in the variational formulation thanks to @jorgensd tutorials.

Type of Change

🐛 Bug fix (non-breaking change which fixes an issue)
✨ New feature (non-breaking change which adds functionality)
💥 Breaking change (fix or feature that would cause existing functionality to not work as expected)
🔨 Code refactoring (no functional changes, no API changes)
📝 Documentation update
✅ Test update (adding missing tests or correcting existing tests)
🔧 Build/CI configuration change

Testing

All existing tests pass locally (pytest)
I have added new tests that prove my fix is effective or that my feature works

Code Quality Checklist

My code follows the code style of this project (Ruff formatted: ruff format .)
My code passes linting checks (ruff check .)
I have performed a self-review of my own code
I have commented my code, particularly in hard-to-understand areas

Documentation

I have updated the documentation accordingly (if applicable)
I have added docstrings to new functions/classes following the project conventions

Breaking Changes

Screenshots/Examples

Additional Notes

codecov · 2026-05-18T14:04:50Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 94.76%. Comparing base (0605dd2) to head (ad8d19e).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1158      +/-   ##
==========================================
+ Coverage   94.72%   94.76%   +0.03%     
==========================================
  Files          46       46              
  Lines        3508     3534      +26     
==========================================
+ Hits         3323     3349      +26     
  Misses        185      185

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

…into weak_dirichlet

…into weak_dirichlet Co-authored-by: hhy2022 <huihuay15@gmail.com>

…into weak_dirichlet

RemDelaporteMathurin · 2026-05-18T14:43:14Z

@jorgensd @jhdark The Docker CI for dolfinx 0.9 doesn't go pas the install and I'm confused as to why. It seems related to some pip PEP8 rules about the license in scifem being "MIT" instead of { text = "MIT" } ???? but for some reason that's not required in the 0.10.... I'm keen on ignoring this for now but I thought I'd bring it up with you

jhdark

Looks great, that was a quick turn around, are we going to add this to the discontinuous class too?

jhdark · 2026-05-18T14:37:49Z

      - name: Install local package and dependencies
        run: |
-          pip install .[test]
+          python -m pip install --upgrade pip


Whats this for?

Oh never mind, saw the final comment on the PR

I'll revert this cause i don't think it's related to this. See this PR scientificcomputing/scifem#213

jhdark · 2026-05-18T14:39:57Z

+        assert alpha is not None, (
+            "Penalty parameter must be given for weakly enforced Dirichlet BCs"
+        )
+        assert self.value_fenics is not None, (
+            "value_fenics must be defined for weakly enforced Dirichlet BCs"
+        )


Should this not be done in a setter instead? Only if the enforce_weakly is True?

No not in a setter because then you always have to set say enforce_weakly before alpha or the opposite because there is no way to set them both at the same time.

This method is only called if enforce_weakly is True anyway.

jhdark · 2026-05-18T14:41:29Z

+        """
+        mesh = ds.ufl_domain()
+        n = ufl.FacetNormal(mesh)
+        h = ufl.Circumradius(mesh)  # FIXME this doesn't work for rectangles


Will this break for hex meshes or only give the wrong value? Because if it breaks, maybe we could have a mini test in here and give a more descriptive error?

It will break. But because there is an open issue for this elswhere I'm hoping we can fix this pretty soon anyway

RemDelaporteMathurin

are we going to add this to the discontinuous class too?

It has already been added to the discontinuous case.

RemDelaporteMathurin · 2026-05-18T15:13:19Z

+        """
+        mesh = ds.ufl_domain()
+        n = ufl.FacetNormal(mesh)
+        h = ufl.Circumradius(mesh)  # FIXME this doesn't work for rectangles


It will break. But because there is an open issue for this elswhere I'm hoping we can fix this pretty soon anyway

RemDelaporteMathurin · 2026-05-18T15:14:34Z

+        assert alpha is not None, (
+            "Penalty parameter must be given for weakly enforced Dirichlet BCs"
+        )
+        assert self.value_fenics is not None, (
+            "value_fenics must be defined for weakly enforced Dirichlet BCs"
+        )


No not in a setter because then you always have to set say enforce_weakly before alpha or the opposite because there is no way to set them both at the same time.

This method is only called if enforce_weakly is True anyway.

jorgensd · 2026-05-18T15:22:40Z

@jorgensd @jhdark The Docker CI for dolfinx 0.9 doesn't go pas the install and I'm confused as to why. It seems related to some pip PEP8 rules about the license in scifem being "MIT" instead of { text = "MIT" } ???? but for some reason that's not required in the 0.10.... I'm keen on ignoring this for now but I thought I'd bring it up with you

Try setting a bound on setuptools, like: https://github.com/FEniCS/ufl/blob/main/pyproject.toml#L3

RemDelaporteMathurin added 7 commits May 18, 2026 09:17

added a enforce_weakly argument to DirichletBCBase

6424040

calculate the form everytime

dec3499

add formulation in HydrogenTransportProblem

2fbb912

fixed u, v for loop

6f5b24f

propagated change to Discontinuous class

b8570e0

refactoring + documentation

e9add2b

added test

8265cd2

RemDelaporteMathurin and others added 7 commits May 18, 2026 10:05

ruff

bdd9637

test for discontinuous

90c8bff

test with multiple species

a137992

test with multiple species

48d775f

Merge branch 'weak_dirichlet' of https://github.com/festim-dev/FESTIM …

fc716a9

…into weak_dirichlet

Merge branch 'weak_dirichlet' of https://github.com/festim-dev/FESTIM …

39ebee7

…into weak_dirichlet Co-authored-by: hhy2022 <huihuay15@gmail.com>

Merge branch 'weak_dirichlet' of https://github.com/festim-dev/FESTIM …

3c4b7c1

…into weak_dirichlet

RemDelaporteMathurin marked this pull request as ready for review May 18, 2026 14:22

RemDelaporteMathurin requested review from jhdark and jorgensd May 18, 2026 14:22

RemDelaporteMathurin added 4 commits May 18, 2026 10:25

add no-build-isolation to pip dependencies

41d10dc

python -m

5dd570d

try changing the license in pyproject

cf30494

upgrade pip

ac946f0

jhdark requested changes May 18, 2026

View reviewed changes

revert

b80166d

RemDelaporteMathurin commented May 18, 2026

View reviewed changes

RemDelaporteMathurin added 2 commits May 18, 2026 11:32

lower bound on setuptools

0294cb1

docstrings

ad8d19e

RemDelaporteMathurin requested a review from jhdark May 18, 2026 16:26

Conversation

RemDelaporteMathurin commented May 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Summary

Related Issues

Motivation and Context

Type of Change

Testing

Code Quality Checklist

Documentation

Breaking Changes

Screenshots/Examples

Additional Notes

Uh oh!

codecov Bot commented May 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

RemDelaporteMathurin commented May 18, 2026

Uh oh!

jhdark left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

RemDelaporteMathurin left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jorgensd commented May 18, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

RemDelaporteMathurin commented May 18, 2026 •

edited

Loading

codecov Bot commented May 18, 2026 •

edited

Loading