Make audio processing nodes handle None -inputs

[kijai]

GetVideoComponents returns None if video doesn't have audio, the CreateVideo handles it graciously, while none of the audio processing nodes do. This PR allows None input to flow through.

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR adds consistent None-input handling across audio processing nodes in comfy_extras/nodes_audio.py. Nodes that require valid audio input (VAEEncodeAudio, SaveAudio variants, PreviewAudio) now validate and reject None with descriptive errors. Transformation and effect nodes (TrimAudioDuration, SplitAudioChannels, AudioAdjustVolume, AudioEqualizer3Band) return None passthrough when input is None. Combination nodes (JoinAudioChannels, AudioConcat, AudioMerge) return non-None inputs or None when both inputs are None. TrimAudioDuration adjusts frame clamping bounds, and several nodes add early return paths for zero-length waveforms.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately summarizes the main change: making audio processing nodes handle None inputs.
Description check	✅ Passed	The description clearly relates to the changeset by explaining the context (GetVideoComponents returning None) and the solution (allowing None input to flow through audio processing nodes).
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
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.}

---

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

🤖 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 `@comfy_extras/nodes_audio.py`:
- Line 418: The change allowing start_frame to be clamped to audio_length is
causing invalid ranges (start_frame can equal audio_length and thus start_frame
>= end_frame); restore the previous upper bound by clamping start_frame to
audio_length - 1 instead of audio_length (keep the lower bound at 0) so
start_frame remains a valid last index, and ensure end_frame logic (the code
that raises ValueError on lines ~423-424) will no longer be triggered for an
intended final-segment trim; update the min(...) call that sets start_frame
accordingly.

🪄 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: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 685548f9-6887-451d-ba4d-5a011802e62b

📥 Commits

Reviewing files that changed from the base of the PR and between 74c17a2 and bfc5d50.

📒 Files selected for processing (1)

• comfy_extras/nodes_audio.py

[alexisrolland]

Tested and it's working fine.