truncate

Shorten a string to a maximum length, marking the cut with an ellipsis. It is Unicode-safe, and aware that which end you keep changes what the truncation means.

Usage

import { truncate } from 'ranuts/utils';

truncate('the quick brown fox', 12); // 'the quick b…'

truncate('/Users/me/code/app/src/index.ts', { length: 20, position: 'start' });
// '…de/app/src/index.ts'

truncate('0xabcdef0123456789', { length: 11, position: 'middle' });
// '0xabc…56789'

API

truncate(value, options)

Parameters

TruncateOptions

position decides which end survives, and that choice carries real information:

• 'end' (default) keeps the beginning — right for prose and titles.
• 'start' keeps the tail, which is what a file path wants: /Users/someone/work/… are the bytes a reader already knows; …/src/utils/str.ts is the part they need.
• 'middle' keeps both ends, for identifiers whose head and tail are meaningful, such as a hash or an account number.

Return

string: never longer than length. If length is shorter than the ellipsis itself, the result is a truncated ellipsis rather than an overflow.

Notes

1. Slices by Unicode code point, not UTF-16 code unit. A naive value.slice(i) can land inside a surrogate pair: any character outside the Basic Multilingual Plane (emoji, some CJK extension characters) is 2 UTF-16 units, producing an unpaired surrogate next to the ellipsis that renders as mojibake. truncate iterates by code point instead, so multi-unit characters are never split.
2. A value shorter than length is returned unchanged — no ellipsis is added.
3. Pass a custom ellipsis (e.g. '...' or '[cut]') if the default '…' character isn't available in the font you're rendering with.