test(message): deflake HLC timestamp serde_json::Value round-trip test

[phil-opp]

Summary

Fixes #3104.

The test a_freshly_generated_hlc_id_uses_all_sixteen_bytes in libraries/message/tests/uhlc_wire_format.rs was non-deterministic and failed on roughly 1 in 256 cargo test runs. Because dora-message is exercised by cargo test --all in both the PR gate and nightly, this introduced an intermittent, hard-to-reproduce CI failure.

Root cause

Under uhlc 0.9, HLC::default() seeds its id via ID::rand() — a uniformly random 128-bit value. ID::size() returns the number of significant little-endian bytes, so size() == 16 only holds when the most-significant byte is non-zero. That byte is zero with probability 1/256, so the assertion assert_eq!(timestamp.get_id().size(), 16) failed ~0.39% of the time. Under uhlc 0.5 the id was seeded from a UUID whose top bits were fixed version/variant nibbles, so it was effectively always 16 bytes — the assertion was implicitly safe then and became flaky after the 0.5 → 0.9 bump (#3016).

No production path depends on size() == 16; the JSON round-trip works for any non-zero id. This is a test-quality defect, not a runtime bug.

Fix

The full-width (> u64) property is already pinned deterministically by timestamp_survives_a_serde_json_value_round_trip, which uses golden_timestamp (top byte 0x10). This test's remaining value is confirming that a real, randomly-generated HLC::default() timestamp survives the serde_json::Value round trip — which is lossless for an id of any width.

So the probabilistic size() assertion is dropped and the real-HLC round-trip check is kept (and the test renamed accordingly). This removes the flakiness while preserving the real-HLC coverage the original test added, on top of the deterministic full-width coverage that already exists.

Testing

• cargo test -p dora-message --test uhlc_wire_format — all 7 tests pass.
• cargo fmt -p dora-message -- --check — clean.

🤖 Generated with Claude Code

https://claude.ai/code/session_01EZXbbi2V3Dw4174Xn3AgMX

---

Generated by Claude Code

[trunk-io]

Merging to main in this repository is managed by Trunk.

• To merge this pull request, check the box to the left or comment /trunk merge below.

After your PR is submitted to the merge queue, this comment will be automatically updated with its status. If the PR fails, failure details will also be posted here

[phil-opp]

🤖 Automated review by Claude Code — fully automated review, not vetted by a human.

No issues found.

This is a clean test-quality fix. The dropped assertion (timestamp.get_id().size() == 16) was genuinely probabilistic — under uhlc 0.9, HLC::default() seeds its id via ID::rand(), and ID::size() counts significant little-endian bytes, so a zero most-significant byte (~1/256) yields size() == 15.

The retained round-trip check still exercises meaningful behavior: it builds a real HLC::default() timestamp, runs it through serde_json::to_value → from_value, and asserts decoded.timestamp == timestamp, so it is not vacuous. Deterministic full-width (> u64) coverage is still pinned separately by timestamp_survives_a_serde_json_value_round_trip (golden timestamp), so nothing is lost.

---

Generated by Claude Code