[xellanix]

Overview

This pull request optimizes the file import workflow and upgrades the server persistence layer to leverage asynchronous operations. It improves user feedback with a loading toast, offloads ZIP decompression to prevent UI blocking, and refactors filesystem interactions for better efficiency.

Key Changes

🚀 Features

• Add loading toast for file import: Introduces a persistent loading toast notification at the start of the file import process and dismisses it upon completion to provide visual feedback.
  • Create a persistent loading toast before iterating through files.
  • Dismiss the toast notification immediately prior to calling the success callback.

🐛 Bug Fixes

• Align array indices with visual group order: Resolves an issue where items grouped visually in the queue had disjointed, non-sequential underlying indices.
  • Sort the master's contents chronologically during the import phase by the first appearance of each group to ensure accurate rendering sequence.
  • Centralize group name fallback parsing using the new groupName utility and implement sortMasterContents.
  • Standardize formatting and indentation for errors in the import processor.

🧹 Chore/Refactoring

• Migrate server persistence to async Bun file APIs: Transitions server persistence from blocking, synchronous Node.js fs methods to asynchronous, native Bun file operations to improve concurrency and prevent event loop blockages.
  • Save data to a temporary file before overwriting the target destination to ensure atomic file writes and prevent corruption.
  • Update the build configuration, server socket listeners, and tsconfig to support ESM, top-level await, and Bun types.
• Remove redundant directory creation: Refactors file importing by removing explicit directory checks and manual creation logic.
  • Rely on Bun.write to automatically create missing parent directories recursively, simplifying the pipeline and reducing synchronous filesystem checks.
• Offload zip decompression to worker thread using async unzip: Optimizes UI performance during .zip file processing by transitioning from the blocking unzipSync function to the asynchronous, non-blocking unzip API provided by fflate.
  • Streamline the iteration model into a single pass over the unzipped object entries instead of running two separate full-array loops sequentially.
• Centralize mime type mapping and optimize unzipped entries loop: Consolidates MIME type resolution during the asset import process to eliminate duplicate execution paths and brittle extension checks.
  • Introduce a dedicated MIME_MAP dictionary and a getMime helper function to safely evaluate file extensions.
  • Cache zip file entry extraction to avoid repeated descriptor processing and unnecessary structural recalculations.
  • Simplify error handling via ternary variable declarations and remove redundant response constraint checks.
• Centralize error logging: Extracts duplicate toast notification and console error logging routines into a unified local helper function to improve maintainability.
• Deduplicate projection json parsing: Extracts the JSON parsing and projection store hydration logic into a reusable helper function to eliminate code duplication across file and ZIP workflows.

Version Bump

0.1.0