Rehype API

@tanstack/highlight/rehype transforms existing HAST <pre><code> pairs. It has no dependency on Rehype itself.

RehypeHighlightOptions

type RehypeHighlightOptions = {
  highlighter: Highlighter
  lineNumbers?: boolean
  getDecorations?: (node: HastElement) =>
    ReadonlyArray<HighlightDecoration> | undefined
  getTitle?: (node: HastElement) => string | undefined
}

Callbacks receive the original pre node. Language names are read from a language-* class on the child code element.

Fence titles, annotations, and line numbers are read from the child code node's data.meta, as emitted by mdast-util-to-hast. A callback title or explicit lineNumbers option overrides metadata.

rehypeHighlightCodeBlocks

function rehypeHighlightCodeBlocks(
  options: RehypeHighlightOptions,
): (tree: unknown) => void

Returns a synchronous Rehype transformer. It recursively replaces supported <pre><code> nodes and skips any pre that already has the th-code class.

import rehypeStringify from 'rehype-stringify'
import remarkParse from 'remark-parse'
import remarkRehype from 'remark-rehype'
import { unified } from 'unified'
import { rehypeHighlightCodeBlocks } from '@tanstack/highlight/rehype'

const file = await unified()
  .use(remarkParse)
  .use(remarkRehype)
  .use(rehypeHighlightCodeBlocks, { highlighter })
  .use(rehypeStringify)
  .process(markdown)

rehypePreCodeToHast

function rehypePreCodeToHast(
  node: HastElement,
  options: RehypeHighlightOptions,
): HastElement | undefined

Converts one eligible pre node. It returns undefined when the node is already highlighted or does not contain a direct code child. Descendant text is concatenated and trailing whitespace is removed before highlighting.

Existing attributes, classes, and plugin data on the pre and code elements are preserved.

See Markdown Pipelines for choosing between the Remark and Rehype integrations.