\n
![](\"https://avatars1.githubusercontent.com/u/12551863?s=256&v=4\")
\n\n
![](\"https://avatars1.githubusercontent.com/u/14985020?s=256&v=4\")
\n\n \n
![](\"https://images.opencollective.com/netlify/4087de2/logo/256.png\"){kind=logo}
\n\n
![](\"https://avatars1.githubusercontent.com/u/35904294?s=128&v=4\")
\n\n
![](\"https://avatars1.githubusercontent.com/u/58979018?s=128&v=4\")
\n\n
![](\"https://avatars1.githubusercontent.com/u/13612118?s=128&v=4\")
\n\n
![](\"https://avatars1.githubusercontent.com/u/12504344?s=128&v=4\")
\n\n You?\n
\n
| \n Gatsby π₯ \n \n | \n\n Vercel π₯ \n \n | \n\n Netlify \n \n \n | \n\n Holloway \n \n | \n\n ThemeIsle \n \n | \n\n BoostIO \n \n | \n\n Expo \n \n | \n
| \n \n You?\n \n | \n
Emphasis and stress, you guys!
\n```\n\n###### Processing between syntaxes\n\n[*Processors*][processors] can be combined in two modes.\n\n**Bridge** mode transforms the [*syntax tree*][syntax-tree] from one format\n(*origin*) to another (*destination*).\nAnother processor runs on the destination tree.\nFinally, the original processor continues transforming the origin tree.\n\n**Mutate** mode also transforms the syntax tree from one format to another.\nBut the original processor continues transforming the destination tree.\n\nIn the previous example (βProgramming interfaceβ), `remark-retext` is used in\n*bridge* mode: the origin syntax tree is kept after [**retext**][retext] is\ndone; whereas `remark-rehype` is used in *mutate* mode: it sets a new syntax\ntree and discards the origin tree.\n\n* [`remark-retext`][remark-retext]\n* [`remark-rehype`][remark-rehype]\n* [`rehype-retext`][rehype-retext]\n* [`rehype-remark`][rehype-remark]\n\n## API\n\n### `processor()`\n\n[*Processor*][processors] describing how to *process* text.\n\n###### Returns\n\n`Function` β New [*unfrozen*][freeze] processor that is configured to work the\nsame as its ancestor.\nWhen the descendant processor is configured in the future it does not affect the\nancestral processor.\n\n###### Example\n\nThe following example shows how a new processor can be created (from the remark\nprocessor) and linked to **stdin**(4) and **stdout**(4).\n\n```js\nvar remark = require('remark')\nvar concat = require('concat-stream')\n\nprocess.stdin.pipe(concat(onconcat))\n\nfunction onconcat(buf) {\n var doc = remark().processSync(buf).toString()\n\n process.stdout.write(doc)\n}\n```\n\n### `processor.use(plugin[, options])`\n\n[*Configure*][configuration] the processor to use a [*plugin*][plugin] and\noptionally configure that plugin with options.\n\nIf the processor is already using this plugin, the previous plugin configuration\nis changed based on the options that are passed in.\nThe plugin is not added a second time.\n\n###### Signatures\n\n* `processor.use(plugin[, options])`\n* `processor.use(preset)`\n* `processor.use(list)`\n\n###### Parameters\n\n* `plugin` ([`Attacher`][plugin])\n* `options` (`*`, optional) β Configuration for `plugin`\n* `preset` (`Object`) β Object with an optional `plugins` (set to `list`),\n and/or an optional `settings` object\n* `list` (`Array`) β List of plugins, presets, and pairs (`plugin` and\n `options` in an array)\n\n###### Returns\n\n`processor` β The processor that `use` was called on.\n\n###### Note\n\n`use` cannot be called on [*frozen*][freeze] processors.\nCall the processor first to create a new unfrozen processor.\n\n###### Example\n\nThere are many ways to pass plugins to `.use()`.\nThe below example gives an overview.\n\n```js\nvar unified = require('unified')\n\nunified()\n // Plugin with options:\n .use(pluginA, {x: true, y: true})\n // Passing the same plugin again merges configuration (to `{x: true, y: false, z: true}`):\n .use(pluginA, {y: false, z: true})\n // Plugins:\n .use([pluginB, pluginC])\n // Two plugins, the second with options:\n .use([pluginD, [pluginE, {}]])\n // Preset with plugins and settings:\n .use({plugins: [pluginF, [pluginG, {}]], settings: {position: false}})\n // Settings only:\n .use({settings: {position: false}})\n```\n\n### `processor.parse(file)`\n\nParse text to a [*syntax tree*][syntax-tree].\n\n###### Parameters\n\n* `file` ([`VFile`][vfile]) β [*File*][file], any value accepted by `vfile()`\n\n###### Returns\n\n[`Node`][node] β Parsed [*syntax tree*][syntax-tree] representing `file`.\n\n###### Note\n\n`parse` freezes the processor if not already [*frozen*][freeze].\n\n`parse` performs the [*parse phase*][description], not the *run phase* or other\nphases.\n\n###### Example\n\nThe below example shows how `parse` can be used to create a syntax tree from a\nfile.\n\n```js\nvar unified = require('unified')\nvar markdown = require('remark-parse')\n\nvar tree = unified().use(markdown).parse('# Hello world!')\n\nconsole.log(tree)\n```\n\nYields:\n\n```js\n{\n type: 'root',\n children: [\n {type: 'heading', depth: 1, children: [Array], position: [Position]}\n ],\n position: {\n start: {line: 1, column: 1, offset: 0},\n end: {line: 1, column: 15, offset: 14}\n }\n}\n```\n\n#### `processor.Parser`\n\nA **parser** handles the parsing of text to a [*syntax tree*][syntax-tree].\nUsed in the [*parse phase*][description] and called with a `string` and\n[`VFile`][vfile] representation of the text to parse.\n\n`Parser` can be a function, in which case it must return a [`Node`][node]: the\nsyntax tree representation of the given file.\n\n`Parser` can also be a constructor function (a function with a `parse` field, or\nother fields, in its `prototype`), in which case itβs constructed with `new`.\nInstances must have a `parse` method that is called without arguments and must\nreturn a [`Node`][node].\n\n### `processor.stringify(node[, file])`\n\nCompile a [*syntax tree*][syntax-tree].\n\n###### Parameters\n\n* `node` ([`Node`][node]) β [*Syntax tree*][syntax-tree] to compile\n* `file` ([`VFile`][vfile], optional) β [*File*][file], any value accepted by\n `vfile()`\n\n###### Returns\n\n`string` or `Buffer` (see notes) β Textual representation of the [*syntax\ntree*][syntax-tree]\n\n###### Note\n\n`stringify` freezes the processor if not already [*frozen*][freeze].\n\n`stringify` performs the [*stringify phase*][description], not the *run phase*\nor other phases.\n\nunified typically compiles by serializing: most [*compiler*][compiler]s return\n`string` (or `Buffer`).\nSome compilers, such as the one configured with [`rehype-react`][rehype-react],\nreturn other values (in this case, a React tree).\nIf youβre using a compiler doesnβt serialize, expect different result values.\nWhen using TypeScript, cast the type on your side.\n\n###### Example\n\nThe below example shows how `stringify` can be used to serialize a syntax tree.\n\n```js\nvar unified = require('unified')\nvar html = require('rehype-stringify')\nvar h = require('hastscript')\n\nvar tree = h('h1', 'Hello world!')\n\nvar doc = unified().use(html).stringify(tree)\n\nconsole.log(doc)\n```\n\nYields:\n\n```html\n