moq-mux: harden the scte35_inject fixture generator

[kixelated]

Summary

Hardens the scte35_inject example (the SCTE-35 fixture generator) against the issues CodeRabbit flagged. The SCTE-35 feature itself merged via #1685; this is a follow-up to the example tool only.

• main returns anyhow::Result<()>; every unwrap/expect/panic becomes a ? with .context(), so routine failures print a clean error instead of a backtrace.
• Allocate the SCTE-35 PID by walking up from 0x0021 to the first PID the input program doesn't already use, instead of hardcoding 0x0021 and risking a collision with an existing ES. Fresh clips that leave 0x0021 free still land on it, so existing fixtures regenerate byte-identically.
• Iterate full packets via chunks_exact(188) and reject inputs whose length isn't a multiple of 188, so pid_of never indexes a short trailing slice.
• Force the cue insertion positions strictly increasing so two cues can't collapse onto one packet and get silently dropped by position().

Test plan

• scte35_inject compiles.
• Smoke-tested end-to-end: clean Error: output on bad args / unknown set; picks 0x0021 for fresh inputs (byte-compatible with prior behavior).
• All 7 moq-mux SCTE-35 tests pass.
• cargo clippy -D warnings and cargo fmt --check clean via nix.

(Written by Claude)

[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

Run ID: b5ae91a4-0292-4cc0-9634-5c9732cd7c60

📥 Commits

Reviewing files that changed from the base of the PR and between d568672 and 6467728.

📒 Files selected for processing (1)

• rs/moq-mux/examples/scte35_inject.rs

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

• rs/moq-mux/examples/scte35_inject.rs

---

Walkthrough

Refactors the SCTE-35 injector to use anyhow for contextual errors and makes main return anyhow::Result. Replaces a fixed SCTE-35 PID with a scan-from-constant allocator that avoids collisions, validates TS packetization (188-byte chunks), decodes cues with context, enforces strictly increasing insertion indices (erroring if out-of-range), updates cue_packet to accept scte_pid and return Result, and writes augmented PMT and output TS with contextual I/O errors.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title 'moq-mux: harden the scte35_inject fixture generator' clearly and concisely summarizes the main change—hardening the SCTE-35 fixture injector example.
Description check	✅ Passed	The description provides detailed context about hardening the scte35_inject example, including specific changes (fallible control flow, PID allocation, packet validation), testing done, and commits clarifying the work.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

✨ Simplify code

• Create PR with simplified code
• Commit simplified code in branch claude/scte35-inject-harden

---

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 and usage tips.}

[coderabbitai]

Actionable comments posted: 1

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

rs/moq-mux/examples/scte35_inject.rs (1)

89-100: ⚠️ Potential issue | 🟡 Minor | ⚡ Quick win

Continue scanning until both PAT and PMT are resolved.

Line 94 breaks on the first PMT payload, so PAT-late inputs can fail at Line 99 with a false “no PAT/PMT PID” error even though PMT is present. Capture PMT PID from the packet header when needed and only break once both values are known.

Suggested patch

 while let Some(pkt) = reader.read_ts_packet().context("reading TS packet")? {
 	match pkt.payload {
 		Some(TsPayload::Pat(pat)) => pmt_pid = pat.table.first().map(|p| p.program_map_pid),
 		Some(TsPayload::Pmt(pmt)) => {
+			pmt_pid.get_or_insert(pkt.header.pid);
 			orig = Some(pmt);
-			break;
 		}
 		_ => {}
 	}
+	if pmt_pid.is_some() && orig.is_some() {
+		break;
+	}
 }

🤖 Prompt for 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.

In `@rs/moq-mux/examples/scte35_inject.rs` around lines 89 - 100, The loop breaks
as soon as a PMT is seen, which can cause pmt_pid to still be None for inputs
where PAT arrives later; update the loop that iterates reader.read_ts_packet()
so it continues scanning until both pmt_pid and orig (the PMT) are known: when
encountering TsPayload::Pat set pmt_pid = pat.table.first().map(...), when
encountering TsPayload::Pmt set orig = Some(pmt) but do not break immediately —
instead also attempt to extract the PMT PID from the packet header if pmt_pid is
None (or simply continue looping until pmt_pid.is_some() && orig.is_some()),
then after the loop use pmt_pid.context(...) and orig.context(...).

🤖 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 `@rs/moq-mux/examples/scte35_inject.rs`:
- Around line 180-188: The current clamping logic using `let last =
packets.len().saturating_sub(1);` and setting `cue_at[i] = (cue_at[i - 1] +
1).min(last);` can produce duplicate `last` entries and cause later cues to be
dropped by `position()`; fix by ensuring all remaining cues get unique
decreasing positions when you hit the end: compute how many cues remain (e.g.
remaining = cue_at.len() - i) and, if `(cue_at[i - 1] + remaining) >= last`,
instead backfill the tail so cues occupy distinct indices ending at `last` (for
example set cue_at[j] = last - (remaining - (j - i)) for j in i..cue_at.len()),
otherwise continue the current strictly-increasing increment logic; update
references to `cue_at`, `last`, and the loop that enforces uniqueness so
`position()` will not drop cues.

---

Outside diff comments:
In `@rs/moq-mux/examples/scte35_inject.rs`:
- Around line 89-100: The loop breaks as soon as a PMT is seen, which can cause
pmt_pid to still be None for inputs where PAT arrives later; update the loop
that iterates reader.read_ts_packet() so it continues scanning until both
pmt_pid and orig (the PMT) are known: when encountering TsPayload::Pat set
pmt_pid = pat.table.first().map(...), when encountering TsPayload::Pmt set orig
= Some(pmt) but do not break immediately — instead also attempt to extract the
PMT PID from the packet header if pmt_pid is None (or simply continue looping
until pmt_pid.is_some() && orig.is_some()), then after the loop use
pmt_pid.context(...) and orig.context(...).

🪄 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

Run ID: 62903042-61f5-4291-82ae-be80f8513c0d

📥 Commits

Reviewing files that changed from the base of the PR and between 4d09d42 and d568672.

📒 Files selected for processing (1)

• rs/moq-mux/examples/scte35_inject.rs