# Project: fumen-analyze

## Technical Stack
- **Runtime**: [Bun](https://bun.sh/)
- **Language**: TypeScript
- **Library**: [tja](https://www.npmjs.com/package/tja) (TJA parser)

## Library Usage (tja)
- **Import**: `import { TJAParser } from "tja";`
- **Parsing**: `const parsed = TJAParser.parse(content);`
- **Structure**:
    - `parsed.title`, `parsed.bpm`, `parsed.offset`
    - `parsed.courses` (Array of `Course` objects)
    - `course.difficulty` (e.g., "Oni")
    - `course.stars` (Level/Stars)
    - `course.activeCourse.getCommands()` returns an array of commands and note sequences.

## Purpose
Analyzing or processing TJA (Taiko Jiro) file formats.

## Key References
- `tja-format.mediawiki`: Detailed specification of the TJA format.

## TJA Format Overview
- **Encoding**: UTF-8 with BOM or Shift-JIS.
- **Extension**: `.tja`.
- **Comments**: Start with `//`.
- **Metadata**: Key-value pairs (e.g., `TITLE:`, `BPM:`, `OFFSET:`).
- **Course Metadata**: Specific to difficulties (e.g., `COURSE:`, `LEVEL:`, `BALLOON:`).
- **Notation**: Commands prefixed with `#` (e.g., `#START`, `#END`, `#MEASURE`, `#BPMCHANGE`).
- **Notes**: `0-9`, `A`, `B`, `F`.