mirror of
https://github.com/hanzoai/font.git
synced 2026-11-02 23:42:52 +00:00
The presets carried what they were fitted against — a licensed text face, a licensed display one, and the residuals of each search. That is our own working metadata, not the licence, and it is not what a preset IS: book is the reading weight, medium the step up before display, wide is widened and weighted to hold the widening. Each now says that and nothing else, and the numbers a search produced stay out of the shipped data, the stylesheet comments, the types and the crate. A preset also stopped naming another brand's surface. One typeface serves several brands, and a page for one of them should read as though the others do not exist — so the five are described by what the setting does, and the page says which of them that brand uses. The prose was still the upstream project's: it credited another company's design team, pointed downloads at their releases, told you to `npm install zen`, and listed seven other foundries' typefaces under Inspiration. Every import path in font.d.ts declared module "zen/font", which is not this package's name, so the types matched nothing a consumer could write. LICENSE.txt is untouched. The copyright notice travels with each copy, and that is where a licence belongs.
10 lines
324 B
TypeScript
10 lines
324 B
TypeScript
export interface ZenPreset {
|
|
wght: number
|
|
scaleX: number
|
|
track: number
|
|
note: string
|
|
}
|
|
export declare const PRESETS: Record<'air' | 'book' | 'medium' | 'wide' | 'round', ZenPreset>
|
|
export declare const XHEIGHT_FACTOR: number
|
|
export declare function preset(name: keyof typeof PRESETS): Record<string, string | number>
|