fix(net): skip an empty SUBSCRIBE_NAMESPACE on draft-14

[kixelated]

Problem

Draft-14 makes a SUBSCRIBE_NAMESPACE prefix with N = 0 a protocol violation. An unscoped origin's interest is that empty prefix, and both moq-net and js/net sent it. A peer that enforces the draft refuses the request, so an unscoped link learns nothing.

Approach

On draft-14, subscribe_prefixes drops an empty prefix and run_subscribe_namespace returns before encoding one. Both paths log at debug. The session still warns and continues when a peer refuses some other namespace subscription. JavaScript skips the same empty prefix and keeps the announce consumer open until the caller closes it, so an unsolicited PUBLISH_NAMESPACE still lands. A scoped prefix is still sent. Draft-15 and later still send an empty prefix.

A draft-14 session test covers an unscoped origin (no SUBSCRIBE_NAMESPACE) and a scoped one (one request for cam). The JavaScript tests cover the same split, and the unscoped case still accepts an unsolicited announcement.

This completes quest/m0/ietf-d14-root-prefix from #5020. That quest file is not on main, so this branch leaves quest/ untouched. Delete the quest file once #5020 merges.

Impact

• Public API: none.
• Wire: draft-14 no longer sends SUBSCRIBE_NAMESPACE with an empty prefix. Draft-15 and later are unchanged. Scoped prefixes are unchanged. A refused namespace subscription still warns, and the session stays up.

Alternatives

• Failing the link when the peer refuses a namespace subscription. Rejected 2026-10-07.
• Requiring a scope on draft-14 links at startup. Rejected 2026-10-07.
• Skipping draft-15 as well. fix(net): skip the empty namespace for foreign draft-14 and draft-15 peers #5018 does that. This quest is draft-14 only.

Follow-ups

• Leave this as a draft beside fix(net): skip the empty namespace for foreign draft-14 and draft-15 peers #5018. That PR also skips draft-15. Landing both would conflict. If draft-15 must drop N = 0 too, extend this skip.
• fix(net): finish the max_age to max_delay rename in tests #5016 finishes the max_age to max_delay rename in moq-net tests. Those tests fail to compile on main until it lands, so this PR's Rust test job fails for that reason. The rename stays in fix(net): finish the max_age to max_delay rename in tests #5016.
• Local just check stopped in Biome on nested biome.jsonc files under this checkout's untracked .worktrees/. CI has no such directory. Biome accepts these two JavaScript files against the repo config, and the package typecheck finished before that local Biome error.

(Written by Grok 4.7)

[kixelated]

Draft-14 now skips an empty SUBSCRIBE_NAMESPACE in moq-net and js/net. A scoped prefix is still sent. A refused namespace subscription still warns, and the session stays up.

Recommendation: skip, and leave this draft. Compare it with #5018 before any merge, and wait for #5016 so the moq-net tests compile. The draft-14 violation is real, so this should stay. Merge it only after that review.

(Written by Grok 4.7)

[kixelated]

Automated review by review (OpenAI)

Reviewed commit 8374b76b9ccb4a67d7bb2a032f450ac9a3b36e03.

No actionable correctness issue found within the stated draft-14-only scope. Both Rust send paths omit the illegal empty prefix, and the TypeScript consumer remains registered for unsolicited announcements. The scoped-prefix regression checks are appropriate.

Direction: reconcile this alternative with #5018 before landing; they intentionally cover different draft ranges and overlap in the same paths. This is not an additional implementation finding.

Verification: static diff and lifecycle review only; tests were not run. The described base-branch test-compilation issue was not independently reproduced.

[coderabbitai]

Warning

Review limit reached

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

Next included review available in 30 minutes.

Check out review usage here.

View limit details

Limit details: You’ve used all 4 included reviews currently available.

Learn how review limits work.

Review configuration:

⚙️ Run configuration

• Configuration used: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: 4ff57a39-2678-4bb9-b01d-5cbb617cadc2

📥 Commits

Reviewing files that changed from the base of the PR and between 9d5cf8d and 8374b76.

📒 Files selected for processing (4)

• js/net/src/ietf/subscriber.test.ts
• js/net/src/ietf/subscriber.ts
• rs/moq-net/src/ietf/session.rs
• rs/moq-net/src/ietf/subscriber.rs

• Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

---

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.}

[kixelated]

Superseded by #5018, which skips the empty SUBSCRIBE_NAMESPACE on both draft-14 and draft-15 in rs/moq-net and js/net.

Nothing was ported: #5018's tests already cover the unscoped skip, the scoped prefix still being asked, the unsolicited PUBLISH_NAMESPACE still landing in JS, and draft-16 still sending the empty prefix. The extra send-path guard in run_subscribe_namespace has no caller that bypasses the session filter.

quest/m0/ietf-d14-root-prefix.md is not on main (it is added by #5020), so there was no quest file to delete here.

(Written by Claude Opus 5.5)