Remove input frames allocation

[OpenSauce]

No description provided.

[Copilot]

Pull Request Overview

This PR refactors the audio processor to reuse pre-allocated input buffers instead of allocating a new Vec each callback, reducing per-frame allocations.

• Introduces input_frames field to hold reusable per-channel input buffers
• Initializes input_frames in the constructor
• Updates the processing loop to clear and refill input_frames rather than creating a new Vec

Comments suppressed due to low confidence (1)

src/io/processor.rs:20

• [nitpick] The field name input_frames is ambiguous about its role as per-channel buffers; consider renaming to input_buffers or channel_input_buffers for clarity.

input_frames: Vec<Vec<f32>>,

[Copilot]

Add a doc comment explaining that this field holds reusable per-channel input buffers to avoid per-frame allocations.

Suggested change

	downsampler: SincFixedIn<f32>,
	downsampler: SincFixedIn<f32>,
	/// Reusable per-channel input buffers to avoid per-frame allocations.

[Copilot]

Only one buffer is pushed below, so input_frames will be empty for other channels and panic on access; initialize a Vec for each channel (e.g., using resize_with(channels, || Vec::with_capacity(buffer_size))).

Suggested change

	input_frames.push(Vec::with_capacity(client.buffer_size() as usize));
	input_frames.resize_with(channels, || Vec::with_capacity(client.buffer_size() as usize));

[Copilot]

Use a structured logging macro (e.g., log::info!) instead of println! to integrate with the application's logging and control verbosity.

[Copilot]

Pull Request Overview

Refactors the audio processing pipeline to eliminate per-callback frame allocations, standardize logging, and update dependency version specifications.

• Processor now reuses an input buffer (input_frames) instead of allocating on each process call.
• Switched from eprintln! to the log crate for error reporting and improved send error handling.
• Updated Cargo.toml to use broader version ranges and removed some dependencies.

Reviewed Changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
src/sim/chain.rs	Changed default amplifier channel index
src/main.rs	Moved set_amp_chain call to after start()
src/io/processor.rs	Introduced reusable input_frames, replaced vec allocs, and switched to log
src/io/manager.rs	Improved error handling in set_amp_chain using unwrap_or_else
Cargo.toml	Broadened dependency version constraints and added log

Comments suppressed due to low confidence (6)

src/sim/chain.rs:213

• The default channel is labeled ‘Clean’ but channel 1 is ‘Rhythm’ (per define_channel). It should probably be set_channel(0) for the clean channel.

chain.set_channel(1);

src/main.rs:52

• Setting the amplifier chain after start() may cause the audio thread to run one callback with the default chain. Consider moving this call before processor_manager.start() to ensure the new chain is applied immediately.

processor_manager.set_amp_chain(chain);

src/io/processor.rs:5

• [nitpick] Importing the entire log module is unnecessary; you can call the logging macros (log::error!, etc.) without a use log; statement.

use log;

src/io/processor.rs:13

• [nitpick] The field doc comment was shortened and is now less descriptive. Consider restoring a more explanatory comment such as 'The current mutable amplifier chain used by the audio thread.'

/// Amplifier.

src/io/manager.rs:41

• [nitpick] Using eprintln! inside unwrap_or_else is inconsistent with the rest of the code’s logging strategy. Consider using log::error! for consistency and better output control.

tx.try_send(Box::new(new_chain)).unwrap_or_else(|e| {

Cargo.toml:7

• You removed serde_json, serde_derive, and other dependencies without updating code references; this may break serialization or functionality elsewhere. Verify that all removed crates are truly unused.

jack = "0.13"