Point source merge by rwalkerlewis · Pull Request #866 · geodynamics/pylith

rwalkerlewis · 2026-01-21T21:18:31Z

Adds moment tensor point sources. Includes source class and documentation describing class and components.

…at time delay is implemented across sources.

baagaard-usgs · 2026-03-17T17:07:47Z

.vscode/settings.json

The changes to this file are not relevant. Please remove.

baagaard-usgs · 2026-03-17T17:09:51Z

examples/point-source/mesh_quad.jou

Gmsh files should be the default for all examples. Use Python scripts to generate Cubit meshes.

baagaard-usgs · 2026-03-17T17:11:14Z

examples/point-source/pylithapp.cfg

Rename directory to something more descriptive. Is the point source for a dynamic simulation or quasistatic? Can it fit within an existing example as an additional step?

baagaard-usgs · 2026-03-17T17:12:03Z

examples/point-source/step02_time_history.cfg

The difference between step01 and step02 should be clear by the filename.

baagaard-usgs · 2026-03-17T17:12:59Z

libsrc/pylith/feassemble/DSLabelAccess.icc

PyLith coding style includes comments for all closing braces. Please revert this change.

baagaard-usgs · 2026-03-17T17:14:16Z

libsrc/pylith/fekernels/Poroelasticity.hh

Why is this line being commented out? This PR is related to point sources, not poroelasticity.

baagaard-usgs · 2026-03-17T17:16:29Z

libsrc/pylith/fekernels/SquareWavelet.hh

Isn't this a step function? If so, rename to StepFunction.

Remove unused auxiliary fields (center frequency) and debugging code that is commented out.

baagaard-usgs · 2026-03-17T17:34:45Z

libsrc/pylith/meshio/MeshIOPetsc.cc

This change does not appear to be related to point sources.

baagaard-usgs · 2026-03-17T17:35:58Z

libsrc/pylith/sources/AuxiliaryFactoryMomentTensorForce.cc

Why is this called ...MomentTensorForce and not ...MomentTensor?

There is a lot of duplicate code here among the AuxiliaryFactor classes. Refactor to remove duplicate code.

baagaard-usgs · 2026-03-17T17:41:17Z

libsrc/pylith/sources/Source.cc

+    err = VecCreateSeqWithArray(PETSC_COMM_SELF, dim, _pointCoords.size(),
+                                &_pointCoords[0], &vecPoints);PYLITH_CHECK_ERROR(err);
+
+    // Debug


Remove debugging code

baagaard-usgs · 2026-03-17T17:43:59Z

pylith/meshio/MeshIOPetsc.py

These changes do not appear to be related to point sources.

baagaard-usgs · 2026-03-17T17:46:40Z

portinfo

This file is user-specific and should not be checked into the repository.

baagaard-usgs

High-level comments:

The commit history is extremely messy. There should be no rebase commits. The workflow is not clear to other developers. Separate changes into several simple sets of commits (changes to libsrc, implementation of libtests, implementation of MMS tests, changes to modules, changes to Python, changes to examples, changes to docs).
Remove changes not related to point sources (these may all be coming from rebasing, which should never introduce commits in your branch).
The directory name used in the example is inconsistent with the other examples. Can these examples fit within the context of any existing example? For example, put a point source in the middle of strikeslip-2d to show a similar deformation field to the corresponding Green's function.
All examples must use Gmsh files as the default. Cubit meshes can be provided as alternatives (see existing examples). Convert any Cubit journal scripts to Python scripts for consistency with the rest of the repo.
The C++ implementations of the point source time functions (including kernels) should be in their own directory. Given them appropriate names, like those for the kinematic source time functions used in FaultCohesiveKin. Why are the sources all called wavelets? The step function, Gaussian, and time history do not seem to be associated with wavelets. Do you really need separate kernels for 2D and 3D?
Remove debugging code that is commented out, especially if it is PetscPrintf() (journal debugging code can be acceptable if it is likely to be used in debugging MMS or full-scale tests).

Recommended path forward:

Create an issue with the proposed layout of the files, including source implementation, tests, examples, and documentation. List names of classes and files. This will permit converging on the naming scheme and code layout.
I would create a new branch and a new, clean commit history rather than trying to clean up the existing one.

Note: Additional comments are available on some files. I didn't closely examine the implementation details because the commit history is messy and the PR includes changes unrelated to point sources. Once a PR contains a clean commit history with appropriate code layout and names, it will be easier to review and suggest incremental changes.

rwalkerlewis and others added 30 commits December 31, 2025 05:00

Initial commit. Adding in source class.

c1861f1

Rebased on new upstream

559e175

Fixed STF nomenclature and added files to makefile

a0584a9

Issue with identifing point for source in parallel

8327dfd

Added Square Wavelet for testing

458b224

Added listing for squarewavelet classes

ab046c0

Added lumped LHS mass matrix for MT source

6766e72

removed local source label functions

67f3626

Removed other label function from source

0057d4d

Moved coordinate to label assignment to function in Source.cc

7976afa

First work on user selectable subfields for moment tensor sources

5ab9a4c

Fixed variable issue

323823e

Parallel works now

2596703

Removed comments

b6d84c5

Field is now selectable for MT

a2f4368

Changed square pulse to user selected subfield

1d5c8fb

Updated to new location function

61da8ed

Array source input works for parallel

d37a8e5

Passes libtest for source

529f041

Intermediate update for unit tests

a08a9e3

Minor changes

c793863

Cleanup of testing loose ends, still not passing source check. See th…

236ba38

…at time delay is implemented across sources.

added point force header to makefile

77db4ad

Point Force libtest fails on factory assertion

ef6d3cf

Libtests pass, moving to implement timehistorydb

1d8421d

Added change to allow moment tensor force STFs to use time history db

7699459

Added fekernel for timehistorywavelet

ff96de9

need to find out how to pass _dbtimehistory

a819d5d

Kicked dbtimehistory down a level

db18b1d

first pass for libsrc finished

a233ff9

knepley and others added 17 commits December 31, 2025 05:03

MeshIOPetsc: Allow creation from command line

f015c2c

Added break to switch statement

3b8d922

Restricted branch to point source only.

2a10bbe

Adjustments to permit compilation

598e695

Fixed install files so that it runs

ee33790

Removed unrelated tests

caf6d6a

Added makefile back

875856d

Removed vestigal parts of test makefile

57fdecc

just removing source tests

6568888

Just removing source tests

4493475

Update configure.ac

a3ce73d

Added functioning MMS tests to moment tensor source

0daf65a

Updated docs to include source components

b9330e9

Fixed old include

3e047e1

fixes to ensure CI tests work

26203b4

source mms test passes

cb7867c

Believed to pass with no failures

1e8d4e1

baagaard-usgs reviewed Mar 17, 2026

View reviewed changes

.vscode/settings.json

Copy link
Copy Markdown

Contributor

baagaard-usgs Mar 17, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The changes to this file are not relevant. Please remove.

baagaard-usgs reviewed Mar 17, 2026

View reviewed changes

libsrc/pylith/meshio/MeshIOPetsc.cc

Copy link
Copy Markdown

Contributor

baagaard-usgs Mar 17, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This change does not appear to be related to point sources.

baagaard-usgs reviewed Mar 17, 2026

View reviewed changes

pylith/meshio/MeshIOPetsc.py

Copy link
Copy Markdown

Contributor

baagaard-usgs Mar 17, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

These changes do not appear to be related to point sources.

baagaard-usgs reviewed Mar 17, 2026

View reviewed changes

portinfo

Copy link
Copy Markdown

Contributor

baagaard-usgs Mar 17, 2026 •

edited

Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This file is user-specific and should not be checked into the repository.

baagaard-usgs requested changes Mar 17, 2026

View reviewed changes

Conversation

rwalkerlewis commented Jan 21, 2026

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!

baagaard-usgs Mar 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

baagaard-usgs Mar 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

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!

baagaard-usgs Mar 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

baagaard-usgs left a comment

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

baagaard-usgs Mar 17, 2026 •

edited

Loading

baagaard-usgs Mar 17, 2026 •

edited

Loading

baagaard-usgs Mar 17, 2026 •

edited

Loading