[Zarr-Support] Route nwbExport through HDF5 backend writer#801
Merged
ehennestad merged 104 commits intomainfrom Apr 21, 2026
Merged
[Zarr-Support] Route nwbExport through HDF5 backend writer#801ehennestad merged 104 commits intomainfrom
nwbExport through HDF5 backend writer#801ehennestad merged 104 commits intomainfrom
Conversation
Extend the test helper to treat types.untyped.DataPipe like DataStub by calling load() on actualValue before comparing.
…ject-checks-in-tutorialtests
Replace the old SKIP_PYNWB_COMPATIBILITY_TEST_FOR_TUTORIALS env var with SKIP_PYNWB_TESTS across tests and nwbtest.m, and update +tests/nwbtest.default.env.
Update CI workflow matrix keys and usage in prepare_release.yml, run_tests.yml, and configurations/matlab_release_matrix_strategy.yml to use the new key (matrix.skip-pynwb-tests) and wire the MATLAB setenv accordingly.
Move conditional installation of pynwb into the workflows (install pynwb only when skip-pynwb-tests == '0') and remove the direct git+ dependency from +tests/requirements.txt (update comment to reflect conditional installs).
Also add TestTags = {'UsesPython'} to PyNWBIOTest.
These changes centralize the CI control of pynwb installation and standardize the skip variable name.
…nore-nwbinspector-subject-checks-in-tutorialtests
checking if this class attribute trips up docs build
Move TestTags to methods blocks. Check if that fixes issue with sphinx build of docs
…nore-nwbinspector-subject-checks-in-tutorialtests
Install pynwb and nwbinspector from PyPI stable releases in the main test workflow so CI only fails due to matnwb regressions. Add a separate weekly workflow that tests against the dev branches to catch upstream incompatibilities early without blocking PRs. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Added more details in comments Removed unused vars NWB_TEST_DEBUG, GITHUB_TOKEN Added PYNWB_REPO_DIR
Replace custom venv/download/GitHub API infrastructure with a simple
approach: read tutorial files from PYNWB_REPO_DIR env var pointing to
a pre-cloned pynwb repo, and run them against the system Python. The
CI workflows are now responsible for cloning the repo and setting the
env var. TestTags = {'UsesPython'} added so the tag selector in the
dev workflow picks up these tests, and SKIP_PYNWB_TESTS correctly
excludes them on older MATLAB releases.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Covers prerequisites, running tests via nwbtest() and the MATLAB unit testing framework, Python dependency setup, environment variable configuration, and test authoring conventions. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
actions/checkout requires path to be within the workspace directory. Use path: pynwb-repo (inside workspace) instead of ../pynwb-repo. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Suppress output
Suppress warning that happens due to wrong values in the file's version attribute in the source schemas for v 2.2.0 and 2.6.0
Added section about setting up dynamically loaded filters
Move test tags to methods block, as testtags in classdef breaks the sphinx/docs build
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Add note on implementation detail of not using abstract class explicitly
nwbExport through backend writer
nwbExport through backend writernwbExport through HDF5 backend writer
…azy-array-generalization
Allow creating instance without inputs
…-support/3-backend-writer
bendichter
approved these changes
Apr 13, 2026
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #801 +/- ##
==========================================
- Coverage 95.44% 95.38% -0.06%
==========================================
Files 203 205 +2
Lines 7261 7323 +62
==========================================
+ Hits 6930 6985 +55
- Misses 331 338 +7 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
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
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.
Motivation
This change decouples generated export code from raw HDF5 file identifiers by routing export operations through a backend writer object in order to easier facilitate export to other formats, i.e Zarr.
The generated
exportmethods now accept awriterand call writer methods instead of callingio.writeGroup,io.writeDataset, andio.writeAttributedirectly. The handwritten export helpers on the same path were updated to normalize either a writer object or legacy file id input, and the HDF5 path now uses a thinHDF5Writeradapter so behavior stays functionally identical for existing HDF5 export.This keeps the scope limited to the export path while removing another HDF5-hardcoded seam from the architecture.
How to test the behavior?
Checklist
fix #XXwhereXXis the issue number?