Overview

Trendline compiles the functionality inside twinleaf-rust into a graphical interface. It is intended to be a software oscilloscope and companion tool for use with Twinleaf devices. It targets desktop platforms with support on macOS, Linux, and Windows.

Implemented Features

• Time series data visualization of device streams
• Periodogram data visualization of device streams
• CSV export of plotted data
• RPC control

Developer Setup

npm run tauri dev will launch the program on the root directory. Make sure to run npm install first.

cargo test export_bindings will export the Rust bindings to be used in the front end Typescript when in the src-tauri directory (run to update structures inside shared.rs)

Application Architectural Overview

Trendline uses Tauri to create a desktop application by combining a Rust backend with frontend frameworks (in our case, Svelte). The corresponding src-tauri and src contain README.md detailing their architecture.

Known Issues

• Multi series plots are notably slower than individual plots of the same series
• Typing in the RPC filtering slows down over time
• Changing sample rates from high sample rate to low sample rate truncates front-end buffers
• Stream monitor does not handle Time epoch type correctly and window statistics may fill zeros on non-time aligned data
• Seems to randomly drop sample packets during acquisition. Could be caused by IPC back pressure? Known issue on macOS with process priority.