$ show chn-lee-yumi/MaterialSearch
MaterialSearch is a fully offline, local media search engine that uses CLIP (Chinese-CLIP) to search photos and videos by natural language or image similarity. The repository contains the frontend web application and a desktop GUI launcher; the core search logic (scanning, embedding, matching) lives in a separate, obfuscated pip package (`materialsearch-core`). The frontend is a Vue 3 single-page application built with Element Plus, providing five search modes (text-to-image, image-to-image, text-to-video, image-to-video, text-image similarity), along with file path copying, clipboard image paste, and video clip download. Deployment options include a Docker container (via docker-compose.yml) or a Windows executable bundle. The system is designed to run entirely on the user's machine with no data uploaded to the cloud, making it suitable for privacy-sensitive or air-gapped environments.
The LLM identified 1 files worth highlighting for reuse:
The main Vue 3 single-page application implementing the entire user interface for MaterialSearch. It provides five distinct search modes (text-to-image, image-to-image, text-to-video, image-to-video, text-image similarity) with a clean tabbed layout, file upload, date range filtering, and result display. Notable reusable patterns include: (1) a `copyAllPaths()` function with clipboard fallback for copying file paths, (2) a `pasteImg()` method using the async Clipboard API to paste images from the clipboard, (3) integration with a video clip download endpoint (`/api/download_video_clip/{hash}/{start}/{end}`), and (4) full internationalization support via vue-i18n with Chinese and English locales. The component is self-contained (all UI logic in one file) and could be extracted and adapted for other local search or media browsing applications with minimal modification.