Decode byte-string fill values in get_json_serializable_value (fixes #61)

[arpitjain099]

Description

Fixes #61.

As @owenlittlejohns confirmed in the issue thread, the crash comes from a byte value in umm_var_dict (a _FillValue of b'\x00' on a MISR variable). It passes through get_json_serializable_value untouched, and then publish_umm_var's requests.put(..., json=umm_var_dict) blows up with "Object of type bytes is not JSON serializable".

That function already special-cases numpy ints and floats, so I added a bytes branch right alongside them. Byte strings are decoded with errors='replace' so an undecodable fill value can't itself raise. If you'd rather coerce these upstream (e.g. where the FillValues list is built), I'm glad to move it.

Jira Issue ID

N/A (GitHub issue #61)

Local Test Steps

git clone https://github.com/arpitjain099/earthdata-varinfo.git
cd earthdata-varinfo
git checkout chore/bytes-fillvalue-json
make install
make test

Added a byte-string subtest to test_get_json_serializable_value; test_umm_var.py passes (21), ruff and black clean on the changed files.

PR Acceptance Checklist

• Jira ticket acceptance criteria met.
• CHANGELOG.md updated to include high level summary of PR changes.
• VERSION updated if publishing a release.
• Tests added/updated and passing.
• Documentation updated (if needed).

Summary by CodeRabbit

• Bug Fixes

  • Fixed an issue where publishing JSON could fail when certain metadata values were stored as byte strings.
  • Byte-string values are now converted to text so they can be safely serialized.
  • Improved handling of non-JSON-friendly values to reduce crashes during publishing.

• Tests

  • Added coverage for byte-string conversion to confirm the result can be encoded and decoded as JSON.

• Chores

  • Updated the release version to 5.1.2 and added release notes.

[coderabbitai]

📝 Walkthrough

Walkthrough

get_json_serializable_value now decodes bytes inputs to text for JSON serialization, the unit test covers the round-trip, and the release files advance to v5.1.2.

Changes

JSON serialization fix

Layer / File(s)	Summary
Decode bytes for JSON values varinfo/umm_var.py, tests/unit/test_umm_var.py, CHANGELOG.md, VERSION	get_json_serializable_value decodes bytes to UTF-8 text with replacement, the unit test checks JSON round-trip behavior, and the changelog and version file advance to v5.1.2.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~4 minutes

Poem

🐇 I nibbled bytes and found them light,
turned fuzzy bits to text just right.
JSON now hops with a merry thump,
v5.1.2 makes the burrow jump! ✨

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Linked Issues check	⚠️ Warning	The PR fixes the bytes serialization crash, but it does not add the CMR response validation and exception handling requested in #61.	Add checks for CMR response type/content before calling .json() and improve exception handling/debugging in the publish path, with tests.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly matches the main change: decoding byte-string fill values to make JSON serialization safe.
Out of Scope Changes check	✅ Passed	The changelog, version bump, and tests are all directly related to the byte-string serialization fix.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
Description check	✅ Passed	The PR description follows the required template and includes the summary, issue reference, test steps, and checklist with only minor omissions.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}