fix(manifests): reject non-string metadata instead of crashing on it

[Noor-ul-ain001]

Problem

ExtensionManifest._validate() and PresetManifest._validate() check only key presence for id/name/version/description, then feed the values straight to re.match() and packaging.Version(). Both raise a bare TypeError on a non-string — and TypeError is neither ValidationError nor PresetValidationError, so it escapes every caller that already handles a malformed manifest.

YAML makes this an easy authoring slip rather than a contrived one:

extension:
  id: myext
  version: 1.0      # unquoted -> parses as a float, not a string

Symptom

This is exactly the failure mode the in-tree comment above the section guards was written to prevent (the preset side landed as #3898). list_installed() degrades a bad manifest to ⚠️ Corrupted extension, but its except catches only the domain error:

except ValidationError:
    # Corrupted extension

So a single bad manifest makes specify extension list exit 1 with a raw TypeError traceback and no output at all — hiding every healthy extension too, not just the broken one. Same for specify preset list.

Before this change:

$ specify extension list
TypeError: 'float' object is not iterable      # exit 1, nothing listed

After:

$ specify extension list

Installed Extensions:

    good-ext (v1.0.0)
     ...
  ✗ bad-ext (v1.0)
     ⚠️ Corrupted extension

Also unguarded on the same path

• extension provides.commands[].name → TypeError from the command-name pattern match. The sibling file field was already safe, since relative_extension_path_violation() rejects a non-string — line 404 even special-cases one for its error label, so non-string values were known to arrive here.
• preset provides.templates[].name / .file → TypeError from re.match and os.path.normpath respectively.

Why this shape of fix

The third manifest twin, IntegrationDescriptor (integrations/catalog.py), is already hardened — it type-checks the same four fields and catches TypeError alongside InvalidVersion. This PR brings the other two in line with that existing reference implementation, so the three manifest validators stay consistent.

Testing

68 tests added across tests/test_extensions.py and tests/test_presets.py:

• each of the four metadata fields × float, int, None, list, dict, bool
• extension commands[].name and preset templates[].name/.file, same matrix
• one end-to-end guard per manifest type asserting a healthy entry still lists normally while only the bad one degrades to "Corrupted"

All 68 fail with the source change reverted and pass with it. Full tests/test_extensions.py + tests/test_presets.py run shows no new failures: the 7 failures / 66 errors present are identical before and after (pre-existing Windows symlink-elevation and pytest tmpdir WinError 5 env issues, unrelated to this change).

Assisted-by: Claude Opus 5 (1M context)

[Copilot]

Pull request overview

Hardens extension and preset manifest validation against non-string YAML values.

Changes:

• Adds string-type validation for manifest metadata and entries.
• Adds 68 regression cases, including end-to-end listing behavior.
• One preset template-type validation gap remains.

Show a summary per file

File	Description
src/specify_cli/extensions/__init__.py	Validates extension metadata and command names.
src/specify_cli/presets/__init__.py	Validates preset metadata and template fields.
tests/test_extensions.py	Adds extension validation regressions.
tests/test_presets.py	Adds preset validation regressions.

Review details

Tip

Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Balanced

[Copilot]

Review details

Suppressed comments (1)

tests/test_presets.py:253

• The validator now explicitly hardens all three required template fields, but this matrix omits type. In particular, list/dict type values previously leaked a raw TypeError from set membership, so that newly fixed path has no regression coverage. Include type in this parameterization.

    @pytest.mark.parametrize("field", ["name", "file"])

• Files reviewed: 4/4 changed files
• Comments generated: 0 new
• Review effort level: Balanced

[mnriem]

Thank you!