[stevevanhooser]

MATLAB half of the fix for VH-Lab/NDR-python#34. The NDR-python half is VH-Lab/NDR-python#35, on the same branch name, claude/clever-bardeen-hf5jgx, so its symmetry workflow pairs with this one. Merge this one first.

The problem

When the other suite's artifact directory, or a JSON file inside it, was missing, the +readArtifacts tests ended with disp(...) + return. A return makes matlab.unittest report the test as Passed, so a comparison that never ran looked exactly like one that ran and matched. That is one of the three ways VH-Lab/NDR-python#33's first symmetry run went green with every video comparison unrun.

Changes

• All 7 read tests: every silent path is now an assumption, which matlab.unittest records as Incomplete:

  • the directory guard (7 places) is now assumeTrue(isfolder(artifactDir), ...);
  • the "metadata.json / readFrames.json / readData.json not found" else branches (10 places) are now assumeFail(...);
  • SpikeGadgets' early return when readData.json is missing is now assumeTrue(isfile(...)).

  On a machine with only one language port, nothing fails. The result simply isn't Passed.

• .github/workflows/test-symmetry.yml:

  • Python never runs in this workflow, so the pythonArtifacts cases are deselected by name, and the log says how many.
  • Then the workflow fails on any Incomplete result, so an Incomplete here always means a comparison that should have run.

• +readArtifacts/INSTRUCTIONS.md and the readOmeZarr.m header: both said to "skip silently". They now give the rule above, and explain why a return is wrong.

NDR-python's workflow applies the same rule to its four stages. It also declares one known gap: readOmeZarr has no Python half, so its pythonArtifacts case is deselected there and logged.

Verification

I wrote this without access to MATLAB. The edits are mechanical, and no disp/return path remains in +readArtifacts. Two CI runs check them:

• NDR-matlab's own symmetry workflow: MATLAB make plus the MATLAB-artifacts read cases, now failing on Incomplete.
• NDR-python's cross-language symmetry run on symmetry: a comparison that could not run fails CI instead of passing NDR-python#35: both directions, strict.

If an Incomplete shows up in either run, the log lists it as INCOMPLETE: <test name>.

🤖 Generated with Claude Code

https://claude.ai/code/session_01QjkvVXsdCVw6NNe5SnGNyX

[github-actions]

Test Results

415 tests  ±0   405 ✅  - 7   34s ⏱️ +8s
 54 suites ±0    10 💤 +7 
  1 files   ±0     0 ❌ ±0 

Results for commit 48c7625. ± Comparison against base commit a45a4e6.

This pull request skips 7 tests.

ndr.symmetry.readArtifacts.format.readOmeZarr ‑ testReadOmeZarrArtifacts(SourceType=pythonArtifacts)
ndr.symmetry.readArtifacts.format.readSpikeGadgets ‑ testReadSpikeGadgetsArtifacts(SourceType=pythonArtifacts)
ndr.symmetry.readArtifacts.reader.readData ‑ testReadDataArtifacts(SourceType=pythonArtifacts)
ndr.symmetry.readArtifacts.reader.readPrairieview ‑ testReadPrairieviewArtifacts(SourceType=pythonArtifacts)
ndr.symmetry.readArtifacts.reader.readTiffstack ‑ testReadTiffstackArtifacts(SourceType=pythonArtifacts)
ndr.symmetry.readArtifacts.reader.readVideo ‑ testReadVideoArtifacts(SourceType=pythonArtifacts)
ndr.symmetry.readArtifacts.reader.readVld ‑ testReadVldArtifacts(SourceType=pythonArtifacts)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 54.94%. Comparing base (a45a4e6) to head (48c7625).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #145   +/-   ##
=======================================
  Coverage   54.94%   54.94%           
=======================================
  Files         164      164           
  Lines        9598     9598           
=======================================
  Hits         5274     5274           
  Misses       4324     4324           

☔ View full report in Codecov by Harness.
📢 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.