fix(partition): support map keys in put_newer

[SAY-5]

Closes #18.

put_newer/5 raised ArgumentError ("not a valid match specification") whenever the key contained a map, because ETS rejects a raw map at the key position of a select_replace/2 match head. This binds the key to a match variable and compares it in the guard instead, which works for any key shape. Added a regression test alongside the existing tuple/list-key cases and updated the changelog.

Summary by CodeRabbit

• Bug Fixes

  • Fixed an issue where put_newer/5 and put_all_newer/3 could raise an ArgumentError when entry keys contained maps.
  • Map-containing keys (including maps nested inside list/tuple structures) are now correctly compared and updated, preserving “latest version wins” behavior.

• Tests

  • Added coverage to verify correct stored values, versioning, and update/telemetry results for map-based keys in put_newer/5.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 47f3a219-f8a4-42aa-8915-887c82265235

📥 Commits

Reviewing files that changed from the base of the PR and between f56e892 and 84891ed.

📒 Files selected for processing (2)

• lib/partitioned_buffer/partition.ex
• test/partitioned_buffer/map_test.exs

🚧 Files skipped from review as they are similar to previous changes (1)

• lib/partitioned_buffer/partition.ex

---

📝 Walkthrough

Walkthrough

replace_match_spec/3 now selects a literal or guarded ETS match specification based on whether the key contains a map. Guarded specifications bind the stored key to a match variable and compare it with ms_literal(key). Tests cover direct and nested map keys with increasing versions, and the changelog documents the fix with a reference to issue #18.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the main change: adding map-key support to put_newer.
Linked Issues check	✅ Passed	The PR addresses issue #18 by preventing the match-spec crash for map-containing keys and adds regression coverage.
Out of Scope Changes check	✅ Passed	The changes are limited to the bug fix, regression tests, and changelog update described by the issue.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✨ 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.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@lib/partitioned_buffer/partition.ex`:
- Around line 520-538: The current implementation of replace_match_spec uses a
guard-based key equality check with key: :"$3", which prevents ETS from using
its O(1) keyed lookup optimization on :set tables and forces a full table scan
instead. To fix this, implement a conditional match spec that uses the original
literal-key approach in the match head for non-map keys (which enables the fast
lookup), and only falls back to the guard-based approach with ms_literal(key)
for keys that contain maps (which would otherwise cause ETS to reject the match
spec). This way, the fast path is preserved for the common case while still
handling map-containing keys correctly.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 6d511c27-ffd9-4e7b-89b2-7cca28c7c652

📥 Commits

Reviewing files that changed from the base of the PR and between fdcd7b0 and f56e892.

📒 Files selected for processing (3)

• CHANGELOG.md
• lib/partitioned_buffer/partition.ex
• test/partitioned_buffer/map_test.exs

[SAY-5]

Split the match spec so non-map keys keep the literal-key head (single keyed lookup on the set table), and only keys that actually contain a map fall back to the $3 guard form that avoids the ETS ArgumentError. Added a regression test for a map nested inside a list key; full suite is green.