Skip to main content
Pretext supports two whitespace modes, selected via the whiteSpace option passed to prepare() or prepareWithSegments(). The mode controls how spaces, tabs, and newlines are treated before line breaking.
The default mode targets white-space: normal behavior, matching how browsers render ordinary paragraph text:
  • Consecutive spaces are collapsed to a single space.
  • Leading and trailing spaces on a line are trimmed.
  • Newlines in the source string are treated as collapsible whitespace (not hard breaks).
import { prepare, layout } from '@chenglou/pretext'

const prepared = prepare('Hello   world\n  foo', '16px Inter')
// Equivalent to the browser rendering "Hello world foo"
const { height } = layout(prepared, containerWidth, 20)

​
When to use pre-wrap

Use { whiteSpace: 'pre-wrap' } when the text comes from a user-editable source where whitespace is meaningful:
  • A <textarea> field where the user controls indentation or blank lines.
  • A code or diff editor displaying source text.
  • Any content that was serialized from a pre-formatted block.
For prose, comments, labels, and most UI copy, the default normal mode is appropriate.

​
Special whitespace characters

Beyond the two modes above, certain Unicode characters have fixed behavior regardless of mode:
  • Non-breaking spaces (NBSP, U+00A0). Survive prepare() as visible content and act as glue — they prevent a word-boundary break at that position, similar to CSS white-space: nowrap applied to a single gap.
  • Zero-width spaces (ZWSP, U+200B). Survive as zero-width break opportunities. They add no visible width but allow the line breaker to split there when needed.
  • Soft hyphens (U+00AD). Invisible until broken. If the line breaker chooses a soft hyphen as the break point, layoutWithLines adds a visible trailing - to the line’s text string. The character takes up no width when unbroken.

Build docs developers (and LLMs) love

Get started for freeTalk to us