Character primitives used by Yozora tokenizers: code-point constants, character predicates,
source-position tracking, entity references, and Unicode case folding.
## Install
```bash
npm install --save @yozora/character
```
## Usage
```typescript
import {
createNodePointGenerator,
foldCase,
isPunctuationCharacter,
} from '@yozora/character'
const points = [...createNodePointGenerator('Hello, 世界')]
const identifier = foldCase('Yozora')
const commaIsPunctuation = isPunctuationCharacter(','.codePointAt(0)!)
```
Key exports include:
- `AsciiCodePoint`, Unicode category enums, and `VirtualCodePoint`
- ASCII/Unicode whitespace, punctuation, control, digit, and letter predicates
- `createNodePointGenerator` and helpers for reading text from node-point intervals
- named and numeric entity-reference parsing
- `foldCase`, `stripChineseCharacters`, and `tightenChineseCharacters`
## Related
- [homepage][]
[homepage]: https://github.com/yozorajs/yozora/tree/v2.4.1/packages/character#readme
