Repository navigation
feat: boundary data on the walls of the simulation domain - #147
Merged
Merged
Conversation
Boundary data on walls of the domain that are not part of an obstruction, e.g. the ground of an outdoor case, are attached to the meshes. FDS may split such a wall into several patches with the same orientation, e.g. where a vent begins, but Boundary.data maps each orientation to one patch, so all but the last were dropped; vmin/vmax per orientation saw only that patch, and orientations listed duplicates. - Boundary.patches lists all patches; orientations are unique and sorted; vmin/vmax cover all patches of an orientation. Boundary.data stays as it is but warns when it drops patches. - MeshCollection.get_global_boundary_data_arrays combines the domain walls over all meshes, with the same grid code as the obstruction faces. - Mesh.get_boundary_data has a docstring, the API page and the boundary notebook say where these data live. The new case bndf_domain (FDS 6.11.1) has two meshes with 0.1 m and 0.2 m cells and a burner vent splitting the floor of each into four patches. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Boundary.vmin/vmax for an orientation without patches return NaN, and Obstruction.vmin/vmax skip parts without that face instead of failing; an obstruction across meshes has faces only some of its parts write. - export_obst_raw writes the orientation metadata in the sorted order the data follow. - Mesh and SubObstruction share one quantity lookup, which the global arrays of meshes and obstructions use as well. - The warning of Boundary.data points at the first caller outside fdsreader, with the helper moved from smoke3d to utils.misc. - The Mesh.get_boundary_data docstring mentions faces between meshes. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- LAZY_LOAD=False keeps its previous effect on boundary data (none): reading every patch at load time would have been a silent change. - get_boundary_data(quantity, orientation) and SubObstruction.get_coordinates no longer go through Boundary.data, which may warn. - The global arrays take the public lookups of meshes and subobstructions instead of their internal dicts; the quantity lookup uses Quantity.__eq__. - Obstruction.vmin/vmax reduce without the NaN values in one expression. - The domain tests no longer depend on how FDS splits the floor into patches, since they also run on the output of new FDS builds; the warning test picks the matching warning. - The notebook states what the ambient-only data in the example are. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Codecov Report❌ Patch coverage is
📢 Thoughts on this report? Let us know! |
m-fehr
added a commit
that referenced
this pull request
Oct 15, 2026
refactor: address the review of #147
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Boundary data on walls of the simulation domain that are not part of an obstruction, e.g. the burning rate on the ground of an outdoor case, are attached to the meshes (
Mesh.get_boundary_data). FDS may split such a wall into several patches with the same orientation, e.g. where a vent begins, butBoundary.datamaps each orientation to one patch, so all but the last patch were dropped. This is why the data in #81 could not be read.Boundary.patcheslists all patches;Boundary.orientationsis unique and sorted;vmin/vmaxper orientation cover all patches of that orientation and return NaN if there is none.Boundary.datais unchanged but warns when it drops patches.sim.meshes.get_global_boundary_data_arrays(quantity)combines the domain walls over all meshes into one array per orientation, with the same grid code asObstruction.get_global_boundary_data_arrays(fix: global boundary arrays keep the time axis first and all face positions #146).Obstruction.vmin/vmaxper orientation skip parts of the obstruction without that face instead of failing.Mesh.get_boundary_datahas a docstring; the API page and the boundary notebook say where these data live.Closes #81
Changes in behaviour
Boundary.orientationsreturns each orientation once, sorted (before: one entry per patch, in file order).Tests
New case
bndf_domain(FDS 6.11.1, input intests/cases/fds_inputs/): two meshes with 0.1 m and 0.2 m cells and a burner vent that splits the floor of each mesh into several patches. The tests check that the patches together cover the floor and that each lands at its place in the global array, without relying on how FDS splits the floor, as they also run on the output of new FDS builds.🤖 Generated with Claude Code