Reference manual

This is the reference manual for the ProseMirror rich text editor. It lists and describes the full public API exported by the library. For more introductory material, please see the guide.

ProseMirror is structured as a number of separate modules. This reference manual describes the exported API per module. If you want to use something from the prosemirror-state module, for example, you can import it like this:

var EditorState = require("prosemirror-state").EditorState
var state = EditorState.create({schema: mySchema})

Or, using ES6 syntax:

import {EditorState} from "prosemirror-state"
let state = EditorState.create({schema: mySchema})

prosemirror-state module

This module implements the state object of a ProseMirror editor, along with the representation of the selection and the plugin abstraction.

Editor State

ProseMirror keeps all editor state (the things, basically, that would be required to create an editor just like the current one) in a single object. That object is updated (creating a new state) by applying transactions to it.

EditorState class

The state of a ProseMirror editor is represented by an object of this type. A state is a persistent data structure—it isn't updated, but rather a new state value is computed from an old one using the apply method.

A state holds a number of built-in fields, and plugins can define additional fields.

doc: Node

The current document.

selection: Selection

The selection.

storedMarks: ?⁠[Mark]

A set of marks to apply to the next input. Will be null when no explicit marks have been set.

schema: Schema

The schema of the state's document.

plugins: [Plugin]

The plugins that are active in this state.

apply(tr: Transaction) → EditorState

Apply the given transaction to produce a new state.

applyTransaction(rootTr: Transaction) → {state: EditorState, transactions: [Transaction]}

Verbose variant of apply that returns the precise transactions that were applied (which might be influenced by the transaction hooks of plugins) along with the new state.

tr: Transaction

Start a transaction from this state.

reconfigure(config: Object) → EditorState

Create a new state based on this one, but with an adjusted set of active plugins. State fields that exist in both sets of plugins are kept unchanged. Those that no longer exist are dropped, and those that are new are initialized using their init method, passing in the new configuration object..

config: Object

configuration options

plugins: [Plugin]

New set of active plugins.

toJSON(pluginFields: ?⁠Object<Plugin> | string | number) → Object

Serialize this state to JSON. If you want to serialize the state of plugins, pass an object mapping property names to use in the resulting JSON object to plugin objects. The argument may also be a string or number, in which case it is ignored, to support the way JSON.stringify calls toString methods.

static create(config: Object) → EditorState

Create a new state.

config: Object

Configuration options. Must contain schema or doc (or both).

schema: ?⁠Schema

The schema to use (only relevant if no doc is specified).

doc: ?⁠Node

The starting document.

selection: ?⁠Selection

A valid selection in the document.

storedMarks: ?⁠[Mark]

The initial set of stored marks.

plugins: ?⁠[Plugin]

The plugins that should be active in this state.

static fromJSON(config: Object, json: Object, pluginFields: ?⁠Object<Plugin>) → EditorState

Deserialize a JSON representation of a state. config should have at least a schema field, and should contain array of plugins to initialize the state with. pluginFields can be used to deserialize the state of plugins, by associating plugin instances with the property names they use in the JSON object.

config: Object

configuration options

schema: Schema

The schema to use.

plugins: ?⁠[Plugin]

The set of active plugins.

Transaction class extends Transform

An editor state transaction, which can be applied to a state to create an updated state. Use to create an instance.

Transactions track changes to the document (they are a subclass of Transform), but also other state changes, like selection updates and adjustments of the set of stored marks. In addition, you can store metadata properties in a transaction, which are extra pieces of information that client code or plugins can use to describe what a transacion represents, so that they can update their own state accordingly.

The editor view uses a few metadata properties: it will attach a property "pointer" with the value true to selection transactions directly caused by mouse or touch input, and a "uiEvent" property of that may be "paste", "cut", or "drop".

time: number

The timestamp associated with this transaction, in the same format as

storedMarks: ?⁠[Mark]

The stored marks set by this transaction, if any.

selection: Selection

The transaction's current selection. This defaults to the editor selection mapped through the steps in the transaction, but can be overwritten with setSelection.

setSelection(selection: Selection) → Transaction

Update the transaction's current selection. Will determine the selection that the editor gets when the transaction is applied.

selectionSet: bool

Whether the selection was explicitly updated by this transaction.

setStoredMarks(marks: ?⁠[Mark]) → Transaction

Set the current stored marks.

ensureMarks(marks: [Mark]) → Transaction

Make sure the current stored marks or, if that is null, the marks at the selection, match the given set of marks. Does nothing if this is already the case.

addStoredMark(mark: Mark) → Transaction

Add a mark to the set of stored marks.

removeStoredMark(mark: Mark | MarkType) → Transaction

Remove a mark or mark type from the set of stored marks.

storedMarksSet: bool

Whether the stored marks were explicitly set for this transaction.

setTime(time: number) → Transaction

Update the timestamp for the transaction.

replaceSelection(slice: Slice) → Transaction

Replace the current selection with the given slice.

replaceSelectionWith(node: Node, inheritMarks: ?⁠bool) → Transaction

Replace the selection with the given node. When inheritMarks is true and the content is inline, it inherits the marks from the place where it is inserted.

deleteSelection() → Transaction

Delete the selection.

insertText(text: string, from: ?⁠number, to: ?⁠number = from) → Transaction

Replace the given range, or the selection if no range is given, with a text node containing the given string.

setMeta(key: string | Plugin | PluginKey, value: any) → Transaction

Store a metadata property in this transaction, keyed either by name or by plugin.

getMeta(key: string | Plugin | PluginKey) → any

Retrieve a metadata property for a given name or plugin.

isGeneric: bool

Returns true if this transaction doesn't contain any metadata, and can thus safely be extended.

scrollIntoView() → Transaction

Indicate that the editor should scroll the selection into view when updated to the state produced by this transaction.


A ProseMirror selection can be one of several types. This module defines types for classical text selections (of which cursors are a special case) and node selections, where a specific document node is selected. It is possible to extend the editor with custom selection types.

Selection class

Superclass for editor selections. Every selection type should extend this. Should not be instantiated directly.

new Selection($anchor: ResolvedPos, $head: ResolvedPos, ranges: ?⁠[SelectionRange])

Initialize a selection with the head and anchor and ranges. If no ranges are given, constructs a single range across $anchor and $head.

ranges: [SelectionRange]

The ranges covered by the selection.

$anchor: ResolvedPos

The resolved anchor of the selection (the side that stays in place when the selection is modified).

$head: ResolvedPos

The resolved head of the selection (the side that moves when the selection is modified).

anchor: number

The selection's anchor, as an unresolved position.

head: number

The selection's head.

from: number

The lower bound of the selection's main range.

to: number

The upper bound of the selection's main range.

$from: ResolvedPos

The resolved lower bound of the selection's main range.

$to: ResolvedPos

The resolved upper bound of the selection's main range.

empty: bool

Indicates whether the selection contains any content.

eq(Selection) → bool

Test whether the selection is the same as another selection.

map(doc: Node, mapping: Mappable) → Selection

Map this selection through a mappable thing. doc should be the new document to which we are mapping.

content() → Slice

Get the content of this selection as a slice.

replace(tr: Transaction, content: ?⁠Slice = Slice.empty)

Replace the selection with a slice or, if no slice is given, delete the selection. Will append to the given transaction.

replaceWith(tr: Transaction, node: Node)

Replace the selection with the given node, appending the changes to the given transaction.

toJSON() → Object

Convert the selection to a JSON representation. When implementing this for a custom selection class, make sure to give the object a type property whose value matches the ID under which you registered your class.

getBookmark() → SelectionBookmark

Get a bookmark for this selection, which is a value that can be mapped without having access to a current document, and later resolved to a real selection for a given document again. (This is used mostly by the history to track and restore old selections.) The default implementation of this method just converts the selection to a text selection and returns the bookmark for that.

visible: bool

Controls whether, when a selection of this type is active in the browser, the selected range should be visible to the user. Defaults to true.

static findFrom($pos: ResolvedPos, dir: number, textOnly: ?⁠bool) → ?⁠Selection

Find a valid cursor or leaf node selection starting at the given position and searching back if dir is negative, and forward if positive. When textOnly is true, only consider cursor selections. Will return null when no valid selection position is found.

static near($pos: ResolvedPos, bias: ?⁠number = 1) → Selection

Find a valid cursor or leaf node selection near the given position. Searches forward first by default, but if bias is negative, it will search backwards first.

static atStart(doc: Node) → Selection

Find the cursor or leaf node selection closest to the start of the given document. Will return an AllSelection if no valid position exists.

static atEnd(doc: Node) → Selection

Find the cursor or leaf node selection closest to the end of the given document.

static fromJSON(doc: Node, json: Object) → Selection

Deserialize the JSON representation of a selection. Must be implemented for custom classes (as a static class method).

static jsonID(id: string, selectionClass: constructor<Selection>)

To be able to deserialize selections from JSON, custom selection classes must register themselves with an ID string, so that they can be disambiguated. Try to pick something that's unlikely to clash with classes from other modules.

TextSelection class extends Selection

A text selection represents a classical editor selection, with a head (the moving side) and anchor (immobile side), both of which point into textblock nodes. It can be empty (a regular cursor position).

new TextSelection($anchor: ResolvedPos, $head: ?⁠ResolvedPos = $anchor)

Construct a text selection between the given points.

$cursor: ?⁠ResolvedPos

Returns a resolved position if this is a cursor selection (an empty text selection), and null otherwise.

static create(doc: Node, anchor: number, head: ?⁠number = anchor) → TextSelection

Create a text selection from non-resolved positions.

static between($anchor: ResolvedPos, $head: ResolvedPos, bias: ?⁠number) → Selection

Return a text selection that spans the given positions or, if they aren't text positions, find a text selection near them. bias determines whether the method searches forward (default) or backwards (negative number) first. Will fall back to calling Selection.near when the document doesn't contain a valid text position.

NodeSelection class extends Selection

A node selection is a selection that points at a single node. All nodes marked selectable can be the target of a node selection. In such a selection, from and to point directly before and after the selected node, anchor equals from, and head equals to..

new NodeSelection($pos: ResolvedPos)

Create a node selection. Does not verify the validity of its argument.

node: Node

The selected node.

static create(doc: Node, from: number) → NodeSelection

Create a node selection from non-resolved positions.

static isSelectable(node: Node) → bool

Determines whether the given node may be selected as a node selection.

AllSelection class extends Selection

A selection type that represents selecting the whole document (which can not necessarily be expressed with a text selection, when there are for example leaf block nodes at the start or end of the document).

new AllSelection(doc: Node)

Create an all-selection over the given document.

SelectionRange class

Represents a selected range in a document.

new SelectionRange($from: ResolvedPos, $to: ResolvedPos)
$from: ResolvedPos

The lower bound of the range.

$to: ResolvedPos

The upper bound of the range.

SelectionBookmark interface

A lightweight, document-independent representation of a selection. You can define a custom bookmark type for a custom selection class to make the history handle it well.

map(mapping: Mapping) → SelectionBookmark

Map the bookmark through a set of changes.

resolve(doc: Node) → Selection

Resolve the bookmark to a real selection again. This may need to do some error checking and may fall back to a default (usually TextSelection.between) if mapping made the bookmark invalid.

Plugin System

To make it easy to package and enable extra editor functionality, ProseMirror has a plugin system.

PluginSpec interface

This is the type passed to the Plugin constructor. It provides a definition for a plugin.

props: ?⁠EditorProps

The view props added by this plugin. Props that are functions will be bound to have the plugin instance as their this binding.

state: ?⁠StateField<any>

Allows a plugin to define a state field, an extra slot in the state object in which it can keep its own data.

key: ?⁠PluginKey

Can be used to make this a keyed plugin. You can have only one plugin with a given key in a given state, but it is possible to access the plugin's configuration and state through the key, without having access to the plugin instance object.

view: ?⁠fn(EditorView) → Object

When the plugin needs to interact with the editor view, or set something up in the DOM, use this field. The function will be called when the plugin's state is associated with an editor view.

returns: Object

Should return an object with the following optional properties:

update: ?⁠fn(view: EditorView, prevState: EditorState)

Called whenever the view's state is updated.

destroy: ?⁠fn()

Called when the view is destroyed or receives a state with different plugins.

filterTransaction: ?⁠fn(Transaction, EditorState) → bool

When present, this will be called before a transaction is applied by the state, allowing the plugin to cancel it (by returning false).

appendTransaction: ?⁠fn(transactions: [Transaction], oldState: EditorState, newState: EditorState) → ?⁠Transaction

Allows the plugin to append another transaction to be applied after the given array of transactions. When another plugin appends a transaction after this was called, it is called again with the new state and new transactions—but only the new transactions, i.e. it won't be passed transactions that it already saw.

StateField interface

A plugin spec may provide a state field (under its state property) of this type, which describes the state it wants to keep. Functions provided here are always called with the plugin instance as their this binding.

init(config: Object, instance: EditorState) → T

Initialize the value of the field. config will be the object passed to EditorState.create. Note that instance is a half-initialized state instance, and will not have values for plugin fields initialized after this one.

apply(tr: Transaction, value: T, oldState: EditorState, newState: EditorState) → T

Apply the given transaction to this state field, producing a new field value. Note that the newState argument is again a partially constructed state does not yet contain the state from plugins coming after this one.

toJSON: ?⁠fn(value: T) → any

Convert this field to JSON. Optional, can be left off to disable JSON serialization for the field.

fromJSON: ?⁠fn(config: Object, value: any, state: EditorState) → T

Deserialize the JSON representation of this field. Note that the state argument is again a half-initialized state.

Plugin class

Plugins bundle functionality that can be added to an editor. They are part of the editor state and may influence that state and the view that contains it.

new Plugin(spec: PluginSpec)

Create a plugin.

props: EditorProps

The props exported by this plugin.

spec: Object

The plugin's spec object.

getState(state: EditorState) → any

Extract the plugin's state field from an editor state.

PluginKey class

A key is used to tag plugins in a way that makes it possible to find them, given an editor state. Assigning a key does mean only one plugin of that type can be active in a state.

new PluginKey(name: ?⁠string = "key")

Create a plugin key.

get(state: EditorState) → ?⁠Plugin

Get the active plugin with this key, if any, from an editor state.

getState(state: EditorState) → ?⁠any

Get the plugin's state from an editor state.

prosemirror-view module

ProseMirror's view module displays a given editor state in the DOM, and handles user events.

Make sure you load style/prosemirror.css as a stylesheet when using this module.

EditorView class

An editor view manages the DOM structure that represents an editable document. Its state and behavior are determined by its props.

new EditorView(place: ?⁠dom.Node | fn(dom.Node) | {mount: dom.Node}, props: DirectEditorProps)

Create a view. place may be a DOM node that the editor should be appended to, a function that will place it into the document, or an object whose mount property holds the node to use as the document container. If it is null, the editor will not be added to the document.

state: EditorState

The view's current state.

dom: dom.Element

An editable DOM node containing the document. (You probably should not directly interfere with its content.)

editable: bool

Indicates whether the editor is currently editable.

dragging: ?⁠{slice: Slice, move: bool}

When editor content is being dragged, this object contains information about the dragged slice and whether it is being copied or moved. At any other time, it is null.

composing: boolean

Holds true when a composition is active.

props: DirectEditorProps

The view's current props.

update(props: DirectEditorProps)

Update the view's props. Will immediately cause an update to the DOM.

setProps(props: DirectEditorProps)

Update the view by updating existing props object with the object given as argument. Equivalent to view.update(Object.assign({}, view.props, props)).

updateState(state: EditorState)

Update the editor's state prop, without touching any of the other props.

someProp(propName: string, f: ?⁠fn(prop: any) → any) → any

Goes over the values of a prop, first those provided directly, then those from plugins (in order), and calls f every time a non-undefined value is found. When f returns a truthy value, that is immediately returned. When f isn't provided, it is treated as the identity function (the prop value is returned directly).

hasFocus() → bool

Query whether the view has focus.


Focus the editor.

root: dom.Document | dom.DocumentFragment

Get the document root in which the editor exists. This will usually be the top-level document, but might be a shadow DOM root if the editor is inside one.

posAtCoords(coords: {left: number, top: number}) → ?⁠{pos: number, inside: number}

Given a pair of viewport coordinates, return the document position that corresponds to them. May return null if the given coordinates aren't inside of the editor. When an object is returned, its pos property is the position nearest to the coordinates, and its inside property holds the position of the inner node that the position falls inside of, or -1 if it is at the top level, not in any node.

coordsAtPos(pos: number, side: ?⁠number = 1) → {left: number, right: number, top: number, bottom: number}

Returns the viewport rectangle at a given document position. left and right will be the same number, as this returns a flat cursor-ish rectangle. If the position is between two things that aren't directly adjacent, side determines which element is used. When < 0, the element before the position is used, otherwise the element after.

domAtPos(pos: number, side: ?⁠number = 0) → {node: dom.Node, offset: number}

Find the DOM position that corresponds to the given document position. When side is negative, find the position as close as possible to the content before the position. When positive, prefer positions close to the content after the position. When zero, prefer as shallow a position as possible.

Note that you should not mutate the editor's internal DOM, only inspect it (and even that is usually not necessary).

nodeDOM(pos: number) → ?⁠dom.Node

Find the DOM node that represents the document node after the given position. May return null when the position doesn't point in front of a node or if the node is inside an opaque node view.

This is intended to be able to call things like getBoundingClientRect on that DOM node. Do not mutate the editor DOM directly, or add styling this way, since that will be immediately overriden by the editor as it redraws the node.

posAtDOM(node: dom.Node, offset: number, bias: ?⁠number = -1) → number

Find the document position that corresponds to a given DOM position. (Whenever possible, it is preferable to inspect the document structure directly, rather than poking around in the DOM, but sometimes—for example when interpreting an event target—you don't have a choice.)

The bias parameter can be used to influence which side of a DOM node to use when the position is inside a leaf node.

endOfTextblock(dir: "up" | "down" | "left" | "right" | "forward" | "backward", state: ?⁠EditorState) → bool

Find out whether the selection is at the end of a textblock when moving in a given direction. When, for example, given "left", it will return true if moving left from the current cursor position would leave that position's parent textblock. Will apply to the view's current state by default, but it is possible to pass a different state.


Removes the editor from the DOM and destroys all node views.

dispatch(tr: Transaction)

Dispatch a transaction. Will call dispatchTransaction when given, and otherwise defaults to applying the transaction to the current state and calling updateState with the result. This method is bound to the view instance, so that it can be easily passed around.


EditorProps interface

Props are configuration values that can be passed to an editor view or included in a plugin. This interface lists the supported props.

The various event-handling functions may all return true to indicate that they handled the given event. The view will then take care to call preventDefault on the event, except with handleDOMEvents, where the handler itself is responsible for that.

How a prop is resolved depends on the prop. Handler functions are called one at a time, starting with the base props and then searching through the plugins (in order of appearance) until one of them returns true. For some props, the first plugin that yields a value gets precedence.

handleDOMEvents: ?⁠Object<fn(view: EditorView, event: dom.Event) → bool>

Can be an object mapping DOM event type names to functions that handle them. Such functions will be called before any handling ProseMirror does of events fired on the editable DOM element. Contrary to the other event handling props, when returning true from such a function, you are responsible for calling preventDefault yourself (or not, if you want to allow the default behavior).

handleKeyDown: ?⁠fn(view: EditorView, event: dom.KeyboardEvent) → bool

Called when the editor receives a keydown event.

handleKeyPress: ?⁠fn(view: EditorView, event: dom.KeyboardEvent) → bool

Handler for keypress events.

handleTextInput: ?⁠fn(view: EditorView, from: number, to: number, text: string) → bool

Whenever the user directly input text, this handler is called before the input is applied. If it returns true, the default behavior of actually inserting the text is suppressed.

handleClickOn: ?⁠fn(view: EditorView, pos: number, node: Node, nodePos: number, event: dom.MouseEvent, direct: bool) → bool

Called for each node around a click, from the inside out. The direct flag will be true for the inner node.

handleClick: ?⁠fn(view: EditorView, pos: number, event: dom.MouseEvent) → bool

Called when the editor is clicked, after handleClickOn handlers have been called.

handleDoubleClickOn: ?⁠fn(view: EditorView, pos: number, node: Node, nodePos: number, event: dom.MouseEvent, direct: bool) → bool

Called for each node around a double click.

handleDoubleClick: ?⁠fn(view: EditorView, pos: number, event: dom.MouseEvent) → bool

Called when the editor is double-clicked, after handleDoubleClickOn.

handleTripleClickOn: ?⁠fn(view: EditorView, pos: number, node: Node, nodePos: number, event: dom.MouseEvent, direct: bool) → bool

Called for each node around a triple click.

handleTripleClick: ?⁠fn(view: EditorView, pos: number, event: dom.MouseEvent) → bool

Called when the editor is triple-clicked, after handleTripleClickOn.

handlePaste: ?⁠fn(view: EditorView, event: dom.ClipboardEvent, slice: Slice) → bool

Can be used to override the behavior of pasting. slice is the pasted content parsed by the editor, but you can directly access the event to get at the raw content.

handleDrop: ?⁠fn(view: EditorView, event: dom.Event, slice: Slice, moved: bool) → bool

Called when something is dropped on the editor. moved will be true if this drop moves from the current selection (which should thus be deleted).

handleScrollToSelection: ?⁠fn(view: EditorView) → bool

Called when the view, after updating its state, tries to scroll the selection into view. A handler function may return false to indicate that it did not handle the scrolling and further handlers or the default behavior should be tried.

createSelectionBetween: ?⁠fn(view: EditorView, anchor: ResolvedPos, head: ResolvedPos) → ?⁠Selection

Can be used to override the way a selection is created when reading a DOM selection between the given anchor and head.

domParser: ?⁠DOMParser

The parser to use when reading editor changes from the DOM. Defaults to calling DOMParser.fromSchema on the editor's schema.

transformPastedHTML: ?⁠fn(html: string) → string

Can be used to transform pasted HTML text, before it is parsed, for example to clean it up.

clipboardParser: ?⁠DOMParser

The parser to use when reading content from the clipboard. When not given, the value of the domParser prop is used.

transformPastedText: ?⁠fn(text: string, plain: bool) → string

Transform pasted plain text. The plain flag will be true when the text is pasted as plain text.

clipboardTextParser: ?⁠fn(text: string, $context: ResolvedPos, plain: bool) → Slice

A function to parse text from the clipboard into a document slice. Called after transformPastedText. The default behavior is to split the text into lines, wrap them in <p> tags, and call clipboardParser on it. The plain flag will be true when the text is pasted as plain text.

transformPasted: ?⁠fn(Slice) → Slice

Can be used to transform pasted content before it is applied to the document.

nodeViews: ?⁠Object<fn(node: Node, view: EditorView, getPos: fn() → number, decorations: [Decoration], innerDecorations: DecorationSource) → NodeView>

Allows you to pass custom rendering and behavior logic for nodes and marks. Should map node and mark names to constructor functions that produce a NodeView object implementing the node's display behavior. For nodes, the third argument getPos is a function that can be called to get the node's current position, which can be useful when creating transactions to update it. For marks, the third argument is a boolean that indicates whether the mark's content is inline.

decorations is an array of node or inline decorations that are active around the node. They are automatically drawn in the normal way, and you will usually just want to ignore this, but they can also be used as a way to provide context information to the node view without adding it to the document itself.

innerDecorations holds the decorations for the node's content. You can safely ignore this if your view has no content or a contentDOM property, since the editor will draw the decorations on the content. But if you, for example, want to create a nested editor with the content, it may make sense to provide it with the inner decorations.

clipboardSerializer: ?⁠DOMSerializer

The DOM serializer to use when putting content onto the clipboard. If not given, the result of DOMSerializer.fromSchema will be used.

clipboardTextSerializer: ?⁠fn(Slice) → string

A function that will be called to get the text for the current selection when copying text to the clipboard. By default, the editor will use textBetween on the selected range.

decorations: ?⁠fn(state: EditorState) → ?⁠DecorationSource

A set of document decorations to show in the view.

editable: ?⁠fn(state: EditorState) → bool

When this returns false, the content of the view is not directly editable.

attributes: ?⁠Object<string> | fn(EditorState) → ?⁠Object<string>

Control the DOM attributes of the editable element. May be either an object or a function going from an editor state to an object. By default, the element will get a class "ProseMirror", and will have its contentEditable attribute determined by the editable prop. Additional classes provided here will be added to the class. For other attributes, the value provided first (as in someProp) will be used.

scrollThreshold: ?⁠number | {top: number, right: number, bottom: number, left: number}

Determines the distance (in pixels) between the cursor and the end of the visible viewport at which point, when scrolling the cursor into view, scrolling takes place. Defaults to 0.

scrollMargin: ?⁠number | {top: number, right: number, bottom: number, left: number}

Determines the extra space (in pixels) that is left above or below the cursor when it is scrolled into view. Defaults to 5.

DirectEditorProps interface extends EditorProps

The props object given directly to the editor view supports two fields that can't be used in plugins:

state: EditorState

The current state of the editor.

dispatchTransaction: ?⁠fn(tr: Transaction)

The callback over which to send transactions (state updates) produced by the view. If you specify this, you probably want to make sure this ends up calling the view's updateState method with a new state that has the transaction applied. The callback will be bound to have the view instance as its this binding.

NodeView interface

By default, document nodes are rendered using the result of the toDOM method of their spec, and managed entirely by the editor. For some use cases, such as embedded node-specific editing interfaces, you want more control over the behavior of a node's in-editor representation, and need to define a custom node view.

Mark views only support dom and contentDOM, and don't support any of the node view methods.

Objects returned as node views must conform to this interface.

dom: ?⁠dom.Node

The outer DOM node that represents the document node. When not given, the default strategy is used to create a DOM node.

contentDOM: ?⁠dom.Node

The DOM node that should hold the node's content. Only meaningful if the node view also defines a dom property and if its node type is not a leaf node type. When this is present, ProseMirror will take care of rendering the node's children into it. When it is not present, the node view itself is responsible for rendering (or deciding not to render) its child nodes.

update: ?⁠fn(node: Node, decorations: [Decoration], innerDecorations: DecorationSource) → bool

When given, this will be called when the view is updating itself. It will be given a node (possibly of a different type), an array of active decorations around the node (which are automatically drawn, and the node view may ignore if it isn't interested in them), and a decoration source that represents any decorations that apply to the content of the node (which again may be ignored). It should return true if it was able to update to that node, and false otherwise. If the node view has a contentDOM property (or no dom property), updating its child nodes will be handled by ProseMirror.

selectNode: ?⁠fn()

Can be used to override the way the node's selected status (as a node selection) is displayed.

deselectNode: ?⁠fn()

When defining a selectNode method, you should also provide a deselectNode method to remove the effect again.

setSelection: ?⁠fn(anchor: number, head: number, root: dom.Document)

This will be called to handle setting the selection inside the node. The anchor and head positions are relative to the start of the node. By default, a DOM selection will be created between the DOM positions corresponding to those positions, but if you override it you can do something else.

stopEvent: ?⁠fn(event: dom.Event) → bool

Can be used to prevent the editor view from trying to handle some or all DOM events that bubble up from the node view. Events for which this returns true are not handled by the editor.

ignoreMutation: ?⁠fn(dom.MutationRecord) → bool

Called when a DOM mutation or a selection change happens within the view. When the change is a selection change, the record will have a type property of "selection" (which doesn't occur for native mutation records). Return false if the editor should re-read the selection or re-parse the range around the mutation, true if it can safely be ignored.

destroy: ?⁠fn()

Called when the node view is removed from the editor or the whole editor is destroyed. (Not available for marks.)


Decorations make it possible to influence the way the document is drawn, without actually changing the document.

Decoration class

Decoration objects can be provided to the view through the decorations prop. They come in several variants—see the static members of this class for details.

from: number

The start position of the decoration.

to: number

The end position. Will be the same as from for widget decorations.

spec: Object

The spec provided when creating this decoration. Can be useful if you've stored extra information in that object.

static widget(pos: number, toDOM: fn(view: EditorView, getPos: fn() → number) → dom.Node | dom.Node, spec: ?⁠Object) → Decoration

Creates a widget decoration, which is a DOM node that's shown in the document at the given position. It is recommended that you delay rendering the widget by passing a function that will be called when the widget is actually drawn in a view, but you can also directly pass a DOM node. getPos can be used to find the widget's current document position.

spec: ?⁠Object

These options are supported:

side: ?⁠number

Controls which side of the document position this widget is associated with. When negative, it is drawn before a cursor at its position, and content inserted at that position ends up after the widget. When zero (the default) or positive, the widget is drawn after the cursor and content inserted there ends up before the widget.

When there are multiple widgets at a given position, their side values determine the order in which they appear. Those with lower values appear first. The ordering of widgets with the same side value is unspecified.

When marks is null, side also determines the marks that the widget is wrapped in—those of the node before when negative, those of the node after when positive.

marks: ?⁠[Mark]

The precise set of marks to draw around the widget.

stopEvent: ?⁠fn(event: dom.Event) → bool

Can be used to control which DOM events, when they bubble out of this widget, the editor view should ignore.

ignoreSelection: ?⁠bool

When set (defaults to false), selection changes inside the widget are ignored, and don't cause ProseMirror to try and re-sync the selection with its selection state.

key: ?⁠string

When comparing decorations of this type (in order to decide whether it needs to be redrawn), ProseMirror will by default compare the widget DOM node by identity. If you pass a key, that key will be compared instead, which can be useful when you generate decorations on the fly and don't want to store and reuse DOM nodes. Make sure that any widgets with the same key are interchangeable—if widgets differ in, for example, the behavior of some event handler, they should get different keys.

static inline(from: number, to: number, attrs: DecorationAttrs, spec: ?⁠Object) → Decoration

Creates an inline decoration, which adds the given attributes to each inline node between from and to.

spec: ?⁠Object

These options are recognized:

inclusiveStart: ?⁠bool

Determines how the left side of the decoration is mapped when content is inserted directly at that position. By default, the decoration won't include the new content, but you can set this to true to make it inclusive.

inclusiveEnd: ?⁠bool

Determines how the right side of the decoration is mapped. See inclusiveStart.

static node(from: number, to: number, attrs: DecorationAttrs, spec: ?⁠Object) → Decoration

Creates a node decoration. from and to should point precisely before and after a node in the document. That node, and only that node, will receive the given attributes.

spec: ?⁠Object

Optional information to store with the decoration. It is also used when comparing decorators for equality.

DecorationAttrs interface

A set of attributes to add to a decorated node. Most properties simply directly correspond to DOM attributes of the same name, which will be set to the property's value. These are exceptions:

class: ?⁠string

A CSS class name or a space-separated set of class names to be added to the classes that the node already had.

style: ?⁠string

A string of CSS to be added to the node's existing style property.

nodeName: ?⁠string

When non-null, the target node is wrapped in a DOM element of this type (and the other attributes are applied to this element).

DecorationSet class extends DecorationSource

A collection of decorations, organized in such a way that the drawing algorithm can efficiently use and compare them. This is a persistent data structure—it is not modified, updates create a new value.

find(start: ?⁠number, end: ?⁠number, predicate: ?⁠fn(spec: Object) → bool) → [Decoration]

Find all decorations in this set which touch the given range (including decorations that start or end directly at the boundaries) and match the given predicate on their spec. When start and end are omitted, all decorations in the set are considered. When predicate isn't given, all decorations are assumed to match.

map(mapping: Mapping, doc: Node, options: ?⁠Object) → DecorationSet

Map the set of decorations in response to a change in the document.

options: ?⁠Object

An optional set of options.

onRemove: ?⁠fn(decorationSpec: Object)

When given, this function will be called for each decoration that gets dropped as a result of the mapping, passing the spec of that decoration.

add(doc: Node, decorations: [Decoration]) → DecorationSet

Add the given array of decorations to the ones in the set, producing a new set. Needs access to the current document to create the appropriate tree structure.

remove(decorations: [Decoration]) → DecorationSet

Create a new set that contains the decorations in this set, minus the ones in the given array.

static create(doc: Node, decorations: [Decoration]) → DecorationSet

Create a set of decorations, using the structure of the given document.

static empty: DecorationSet

The empty set of decorations.

DecorationSource interface

An object that can provide decorations. Implemented by DecorationSet, and passed to node views.

prosemirror-model module

This module defines ProseMirror's content model, the data structures used to represent and work with documents.

Document Structure

A ProseMirror document is a tree. At each level, a node describes the type of the content, and holds a fragment containing its children.

Node class

This class represents a node in the tree that makes up a ProseMirror document. So a document is an instance of Node, with children that are also instances of Node.

Nodes are persistent data structures. Instead of changing them, you create new ones with the content you want. Old ones keep pointing at the old document shape. This is made cheaper by sharing structure between the old and new data as much as possible, which a tree shape like this (without back pointers) makes easy.

Do not directly mutate the properties of a Node object. See the guide for more information.

type: NodeType

The type of node that this is.

attrs: Object

An object mapping attribute names to values. The kind of attributes allowed and required are determined by the node type.

content: Fragment

A container holding the node's children.

marks: [Mark]

The marks (things like whether it is emphasized or part of a link) applied to this node.

text: ?⁠string

For text nodes, this contains the node's text content.

nodeSize: number

The size of this node, as defined by the integer-based indexing scheme. For text nodes, this is the amount of characters. For other leaf nodes, it is one. For non-leaf nodes, it is the size of the content plus two (the start and end token).

childCount: number

The number of children that the node has.

child(index: number) → Node

Get the child node at the given index. Raises an error when the index is out of range.

maybeChild(index: number) → ?⁠Node

Get the child node at the given index, if it exists.

forEach(f: fn(node: Node, offset: number, index: number))

Call f for every child node, passing the node, its offset into this parent node, and its index.

nodesBetween(from: number, to: number, f: fn(node: Node, pos: number, parent: Node, index: number) → ?⁠bool, startPos: ?⁠number = 0)

Invoke a callback for all descendant nodes recursively between the given two positions that are relative to start of this node's content. The callback is invoked with the node, its parent-relative position, its parent node, and its child index. When the callback returns false for a given node, that node's children will not be recursed over. The last parameter can be used to specify a starting position to count from.

descendants(f: fn(node: Node, pos: number, parent: Node) → ?⁠bool)

Call the given callback for every descendant node. Doesn't descend into a node when the callback returns false.

textContent: string

Concatenates all the text nodes found in this fragment and its children.

textBetween(from: number, to: number, blockSeparator: ?⁠string, leafText: ?⁠string) → string

Get all text between positions from and to. When blockSeparator is given, it will be inserted whenever a new block node is started. When leafText is given, it'll be inserted for every non-text leaf node encountered.

firstChild: ?⁠Node

Returns this node's first child, or null if there are no children.

lastChild: ?⁠Node

Returns this node's last child, or null if there are no children.

eq(other: Node) → bool

Test whether two nodes represent the same piece of document.

sameMarkup(other: Node) → bool

Compare the markup (type, attributes, and marks) of this node to those of another. Returns true if both have the same markup.

hasMarkup(type: NodeType, attrs: ?⁠Object, marks: ?⁠[Mark]) → bool

Check whether this node's markup correspond to the given type, attributes, and marks.

copy(content: ?⁠Fragment = null) → Node

Create a new node with the same markup as this node, containing the given content (or empty, if no content is given).

mark(marks: [Mark]) → Node

Create a copy of this node, with the given set of marks instead of the node's own marks.

cut(from: number, to: ?⁠number) → Node

Create a copy of this node with only the content between the given positions. If to is not given, it defaults to the end of the node.

slice(from: number, to: ?⁠number = this.content.size) → Slice

Cut out the part of the document between the given positions, and return it as a Slice object.

replace(from: number, to: number, slice: Slice) → Node

Replace the part of the document between the given positions with the given slice. The slice must 'fit', meaning its open sides must be able to connect to the surrounding content, and its content nodes must be valid children for the node they are placed into. If any of this is violated, an error of type ReplaceError is thrown.

nodeAt(pos: number) → ?⁠Node

Find the node directly after the given position.

childAfter(pos: number) → {node: ?⁠Node, index: number, offset: number}

Find the (direct) child node after the given offset, if any, and return it along with its index and offset relative to this node.

childBefore(pos: number) → {node: ?⁠Node, index: number, offset: number}

Find the (direct) child node before the given offset, if any, and return it along with its index and offset relative to this node.

resolve(pos: number) → ResolvedPos

Resolve the given position in the document, returning an object with information about its context.

rangeHasMark(from: number, to: number, type: Mark | MarkType) → bool

Test whether a given mark or mark type occurs in this document between the two given positions.

isBlock: bool

True when this is a block (non-inline node)

isTextblock: bool

True when this is a textblock node, a block node with inline content.

inlineContent: bool

True when this node allows inline content.

isInline: bool

True when this is an inline node (a text node or a node that can appear among text).

isText: bool

True when this is a text node.

isLeaf: bool

True when this is a leaf node.

isAtom: bool

True when this is an atom, i.e. when it does not have directly editable content. This is usually the same as isLeaf, but can be configured with the atom property on a node's spec (typically used when the node is displayed as an uneditable node view).

toString() → string

Return a string representation of this node for debugging purposes.

contentMatchAt(index: number) → ContentMatch

Get the content match in this node at the given index.

canReplace(from: number, to: number, replacement: ?⁠Fragment = Fragment.empty, start: ?⁠number = 0, end: ?⁠number) → bool

Test whether replacing the range between from and to (by child index) with the given replacement fragment (which defaults to the empty fragment) would leave the node's content valid. You can optionally pass start and end indices into the replacement fragment.

canReplaceWith(from: number, to: number, type: NodeType, marks: ?⁠[Mark]) → bool

Test whether replacing the range from to to (by index) with a node of the given type would leave the node's content valid.

canAppend(other: Node) → bool

Test whether the given node's content could be appended to this node. If that node is empty, this will only return true if there is at least one node type that can appear in both nodes (to avoid merging completely incompatible nodes).


Check whether this node and its descendants conform to the schema, and raise error when they do not.

toJSON() → Object

Return a JSON-serializeable representation of this node.

static fromJSON(schema: Schema, json: Object) → Node

Deserialize a node from its JSON representation.

Fragment class

A fragment represents a node's collection of child nodes.

Like nodes, fragments are persistent data structures, and you should not mutate them or their content. Rather, you create new instances whenever needed. The API tries to make this easy.

size: number

The size of the fragment, which is the total of the size of its content nodes.

nodesBetween(from: number, to: number, f: fn(node: Node, start: number, parent: Node, index: number) → ?⁠bool, nodeStart: ?⁠number = 0)

Invoke a callback for all descendant nodes between the given two positions (relative to start of this fragment). Doesn't descend into a node when the callback returns false.

descendants(f: fn(node: Node, pos: number, parent: Node) → ?⁠bool)

Call the given callback for every descendant node. The callback may return false to prevent traversal of a given node's children.

textBetween(from: number, to: number, blockSeparator: ?⁠string, leafText: ?⁠string) → string

Extract the text between from and to. See the same method on Node.

append(other: Fragment) → Fragment

Create a new fragment containing the combined content of this fragment and the other.

cut(from: number, to: ?⁠number) → Fragment

Cut out the sub-fragment between the two given positions.

replaceChild(index: number, node: Node) → Fragment

Create a new fragment in which the node at the given index is replaced by the given node.

eq(other: Fragment) → bool

Compare this fragment to another one.

firstChild: ?⁠Node

The first child of the fragment, or null if it is empty.

lastChild: ?⁠Node

The last child of the fragment, or null if it is empty.

childCount: number

The number of child nodes in this fragment.

child(index: number) → Node

Get the child node at the given index. Raise an error when the index is out of range.

maybeChild(index: number) → ?⁠Node

Get the child node at the given index, if it exists.

forEach(f: fn(node: Node, offset: number, index: number))

Call f for every child node, passing the node, its offset into this parent node, and its index.

findDiffStart(other: Fragment) → ?⁠number

Find the first position at which this fragment and another fragment differ, or null if they are the same.

findDiffEnd(other: Fragment) → ?⁠{a: number, b: number}

Find the first position, searching from the end, at which this fragment and the given fragment differ, or null if they are the same. Since this position will not be the same in both nodes, an object with two separate positions is returned.

toString() → string

Return a debugging string that describes this fragment.

toJSON() → ?⁠Object

Create a JSON-serializeable representation of this fragment.

static fromJSON(schema: Schema, value: ?⁠Object) → Fragment

Deserialize a fragment from its JSON representation.

static fromArray(array: [Node]) → Fragment

Build a fragment from an array of nodes. Ensures that adjacent text nodes with the same marks are joined together.

static from(nodes: ?⁠Fragment | Node | [Node]) → Fragment

Create a fragment from something that can be interpreted as a set of nodes. For null, it returns the empty fragment. For a fragment, the fragment itself. For a node or array of nodes, a fragment containing those nodes.

static empty: Fragment

An empty fragment. Intended to be reused whenever a node doesn't contain anything (rather than allocating a new empty fragment for each leaf node).

Mark class

A mark is a piece of information that can be attached to a node, such as it being emphasized, in code font, or a link. It has a type and optionally a set of attributes that provide further information (such as the target of the link). Marks are created through a Schema, which controls which types exist and which attributes they have.

type: MarkType

The type of this mark.

attrs: Object

The attributes associated with this mark.

addToSet(set: [Mark]) → [Mark]

Given a set of marks, create a new set which contains this one as well, in the right position. If this mark is already in the set, the set itself is returned. If any marks that are set to be exclusive with this mark are present, those are replaced by this one.

removeFromSet(set: [Mark]) → [Mark]

Remove this mark from the given set, returning a new set. If this mark is not in the set, the set itself is returned.

isInSet(set: [Mark]) → bool

Test whether this mark is in the given set of marks.

eq(other: Mark) → bool

Test whether this mark has the same type and attributes as another mark.

toJSON() → Object

Convert this mark to a JSON-serializeable representation.

static fromJSON(schema: Schema, json: Object) → Mark
static sameSet(a: [Mark], b: [Mark]) → bool

Test whether two sets of marks are identical.

static setFrom(marks: ?⁠Mark | [Mark]) → [Mark]

Create a properly sorted mark set from null, a single mark, or an unsorted array of marks.

static none: [Mark]

The empty set of marks.

Slice class

A slice represents a piece cut out of a larger document. It stores not only a fragment, but also the depth up to which nodes on both side are ‘open’ (cut through).

new Slice(content: Fragment, openStart: number, openEnd: number)

Create a slice. When specifying a non-zero open depth, you must make sure that there are nodes of at least that depth at the appropriate side of the fragment—i.e. if the fragment is an empty paragraph node, openStart and openEnd can't be greater than 1.

It is not necessary for the content of open nodes to conform to the schema's content constraints, though it should be a valid start/end/middle for such a node, depending on which sides are open.

content: Fragment

The slice's content.

openStart: number

The open depth at the start.

openEnd: number

The open depth at the end.

size: number

The size this slice would add when inserted into a document.

eq(other: Slice) → bool

Tests whether this slice is equal to another slice.

toJSON() → ?⁠Object

Convert a slice to a JSON-serializable representation.

static fromJSON(schema: Schema, json: ?⁠Object) → Slice

Deserialize a slice from its JSON representation.

static maxOpen(fragment: Fragment, openIsolating: ?⁠bool = true) → Slice

Create a slice from a fragment by taking the maximum possible open value on both side of the fragment.

static empty: Slice

The empty slice.

ReplaceError class extends Error

Error type raised by Node.replace when given an invalid replacement.

Resolved Positions

Positions in a document can be represented as integer offsets. But you'll often want to use a more convenient representation.

ResolvedPos class

You can resolve a position to get more information about it. Objects of this class represent such a resolved position, providing various pieces of context information, and some helper methods.

Throughout this interface, methods that take an optional depth parameter will interpret undefined as this.depth and negative numbers as this.depth + value.

pos: number

The position that was resolved.

depth: number

The number of levels the parent node is from the root. If this position points directly into the root node, it is 0. If it points into a top-level paragraph, 1, and so on.

parentOffset: number

The offset this position has into its parent node.

parent: Node

The parent node that the position points into. Note that even if a position points into a text node, that node is not considered the parent—text nodes are ‘flat’ in this model, and have no content.

doc: Node

The root node in which the position was resolved.

node(depth: ?⁠number) → Node

The ancestor node at the given level. p.node(p.depth) is the same as p.parent.

index(depth: ?⁠number) → number

The index into the ancestor at the given level. If this points at the 3rd node in the 2nd paragraph on the top level, for example, p.index(0) is 1 and p.index(1) is 2.

indexAfter(depth: ?⁠number) → number

The index pointing after this position into the ancestor at the given level.

start(depth: ?⁠number) → number

The (absolute) position at the start of the node at the given level.

end(depth: ?⁠number) → number

The (absolute) position at the end of the node at the given level.

before(depth: ?⁠number) → number

The (absolute) position directly before the wrapping node at the given level, or, when depth is this.depth + 1, the original position.

after(depth: ?⁠number) → number

The (absolute) position directly after the wrapping node at the given level, or the original position when depth is this.depth + 1.

textOffset: number

When this position points into a text node, this returns the distance between the position and the start of the text node. Will be zero for positions that point between nodes.

nodeAfter: ?⁠Node

Get the node directly after the position, if any. If the position points into a text node, only the part of that node after the position is returned.

nodeBefore: ?⁠Node

Get the node directly before the position, if any. If the position points into a text node, only the part of that node before the position is returned.

posAtIndex(index: number, depth: ?⁠number) → number

Get the position at the given index in the parent node at the given depth (which defaults to this.depth).

marks() → [Mark]

Get the marks at this position, factoring in the surrounding marks' inclusive property. If the position is at the start of a non-empty node, the marks of the node after it (if any) are returned.

marksAcross($end: ResolvedPos) → ?⁠[Mark]

Get the marks after the current position, if any, except those that are non-inclusive and not present at position $end. This is mostly useful for getting the set of marks to preserve after a deletion. Will return null if this position is at the end of its parent node or its parent node isn't a textblock (in which case no marks should be preserved).

sharedDepth(pos: number) → number

The depth up to which this position and the given (non-resolved) position share the same parent nodes.

blockRange(other: ?⁠ResolvedPos = this, pred: ?⁠fn(Node) → bool) → ?⁠NodeRange

Returns a range based on the place where this position and the given position diverge around block content. If both point into the same textblock, for example, a range around that textblock will be returned. If they point into different blocks, the range around those blocks in their shared ancestor is returned. You can pass in an optional predicate that will be called with a parent node to see if a range into that parent is acceptable.

sameParent(other: ResolvedPos) → bool

Query whether the given position shares the same parent node.

max(other: ResolvedPos) → ResolvedPos

Return the greater of this and the given position.

min(other: ResolvedPos) → ResolvedPos

Return the smaller of this and the given position.

NodeRange class

Represents a flat range of content, i.e. one that starts and ends in the same node.

new NodeRange($from: ResolvedPos, $to: ResolvedPos, depth: number)

Construct a node range. $from and $to should point into the same node until at least the given depth, since a node range denotes an adjacent set of nodes in a single parent node.

$from: ResolvedPos

A resolved position along the start of the content. May have a depth greater than this object's depth property, since these are the positions that were used to compute the range, not re-resolved positions directly at its boundaries.

$to: ResolvedPos

A position along the end of the content. See caveat for $from.

depth: number

The depth of the node that this range points into.

start: number

The position at the start of the range.

end: number

The position at the end of the range.

parent: Node

The parent node that the range points into.

startIndex: number

The start index of the range in the parent node.

endIndex: number

The end index of the range in the parent node.

Document Schema

Every ProseMirror document conforms to a schema, which describes the set of nodes and marks that it is made out of, along with the relations between those, such as which node may occur as a child node of which other nodes.

Schema class

A document schema. Holds node and mark type objects for the nodes and marks that may occur in conforming documents, and provides functionality for creating and deserializing such documents.

new Schema(spec: SchemaSpec)

Construct a schema from a schema specification.

spec: SchemaSpec

The spec on which the schema is based, with the added guarantee that its nodes and marks properties are OrderedMap instances (not raw objects).

nodes: Object<NodeType>

An object mapping the schema's node names to node type objects.

marks: Object<MarkType>

A map from mark names to mark type objects.

topNodeType: NodeType

The type of the default top node for this schema.

cached: Object

An object for storing whatever values modules may want to compute and cache per schema. (If you want to store something in it, try to use property names unlikely to clash.)

node(type: string | NodeType, attrs: ?⁠Object, content: ?⁠Fragment | Node | [Node], marks: ?⁠[Mark]) → Node

Create a node in this schema. The type may be a string or a NodeType instance. Attributes will be extended with defaults, content may be a Fragment, null, a Node, or an array of nodes.

text(text: string, marks: ?⁠[Mark]) → Node

Create a text node in the schema. Empty text nodes are not allowed.

mark(type: string | MarkType, attrs: ?⁠Object) → Mark

Create a mark with the given type and attributes.

nodeFromJSON(json: Object) → Node

Deserialize a node from its JSON representation. This method is bound.

markFromJSON(json: Object) → Mark

Deserialize a mark from its JSON representation. This method is bound.

SchemaSpec interface

An object describing a schema, as passed to the Schema constructor.

nodes: Object<NodeSpec> | OrderedMap<NodeSpec>

The node types in this schema. Maps names to NodeSpec objects that describe the node type associated with that name. Their order is significant—it determines which parse rules take precedence by default, and which nodes come first in a given group.

marks: ?⁠Object<MarkSpec> | OrderedMap<MarkSpec>

The mark types that exist in this schema. The order in which they are provided determines the order in which mark sets are sorted and in which parse rules are tried.

topNode: ?⁠string

The name of the default top-level node for the schema. Defaults to "doc".

NodeSpec interface

content: ?⁠string

The content expression for this node, as described in the schema guide. When not given, the node does not allow any content.

marks: ?⁠string

The marks that are allowed inside of this node. May be a space-separated string referring to mark names or groups, "_" to explicitly allow all marks, or "" to disallow marks. When not given, nodes with inline content default to allowing all marks, other nodes default to not allowing marks.

group: ?⁠string

The group or space-separated groups to which this node belongs, which can be referred to in the content expressions for the schema.

inline: ?⁠bool

Should be set to true for inline nodes. (Implied for text nodes.)

atom: ?⁠bool

Can be set to true to indicate that, though this isn't a leaf node, it doesn't have directly editable content and should be treated as a single unit in the view.

attrs: ?⁠Object<AttributeSpec>

The attributes that nodes of this type get.

selectable: ?⁠bool

Controls whether nodes of this type can be selected as a node selection. Defaults to true for non-text nodes.

draggable: ?⁠bool

Determines whether nodes of this type can be dragged without being selected. Defaults to false.

code: ?⁠bool

Can be used to indicate that this node contains code, which causes some commands to behave differently.

defining: ?⁠bool

Determines whether this node is considered an important parent node during replace operations (such as paste). Non-defining (the default) nodes get dropped when their entire content is replaced, whereas defining nodes persist and wrap the inserted content. Likewise, in inserted content the defining parents of the content are preserved when possible. Typically, non-default-paragraph textblock types, and possibly list items, are marked as defining.

isolating: ?⁠bool

When enabled (default is false), the sides of nodes of this type count as boundaries that regular editing operations, like backspacing or lifting, won't cross. An example of a node that should probably have this enabled is a table cell.

toDOM: ?⁠fn(node: Node) → DOMOutputSpec

Defines the default way a node of this type should be serialized to DOM/HTML (as used by DOMSerializer.fromSchema). Should return a DOM node or an array structure that describes one, with an optional number zero (“hole”) in it to indicate where the node's content should be inserted.

For text nodes, the default is to create a text DOM node. Though it is possible to create a serializer where text is rendered differently, this is not supported inside the editor, so you shouldn't override that in your text node spec.

parseDOM: ?⁠[ParseRule]

Associates DOM parser information with this node, which can be used by DOMParser.fromSchema to automatically derive a parser. The node field in the rules is implied (the name of this node will be filled in automatically). If you supply your own parser, you do not need to also specify parsing rules in your schema.

toDebugString: ?⁠fn(node: Node) → string

Defines the default way a node of this type should be serialized to a string representation for debugging (e.g. in error messages).

MarkSpec interface

attrs: ?⁠Object<AttributeSpec>

The attributes that marks of this type get.

inclusive: ?⁠bool

Whether this mark should be active when the cursor is positioned at its end (or at its start when that is also the start of the parent node). Defaults to true.

excludes: ?⁠string

Determines which other marks this mark can coexist with. Should be a space-separated strings naming other marks or groups of marks. When a mark is added to a set, all marks that it excludes are removed in the process. If the set contains any mark that excludes the new mark but is not, itself, excluded by the new mark, the mark can not be added an the set. You can use the value "_" to indicate that the mark excludes all marks in the schema.

Defaults to only being exclusive with marks of the same type. You can set it to an empty string (or any string not containing the mark's own name) to allow multiple marks of a given type to coexist (as long as they have different attributes).

group: ?⁠string

The group or space-separated groups to which this mark belongs.

spanning: ?⁠bool

Determines whether marks of this type can span multiple adjacent nodes when serialized to DOM/HTML. Defaults to true.

toDOM: ?⁠fn(mark: Mark, inline: bool) → DOMOutputSpec

Defines the default way marks of this type should be serialized to DOM/HTML. When the resulting spec contains a hole, that is where the marked content is placed. Otherwise, it is appended to the top node.

parseDOM: ?⁠[ParseRule]

Associates DOM parser information with this mark (see the corresponding node spec field). The mark field in the rules is implied.

AttributeSpec interface

Used to define attributes on nodes or marks.

default: ?⁠any

The default value for this attribute, to use when no explicit value is provided. Attributes that have no default must be provided whenever a node or mark of a type that has them is created.

NodeType class

Node types are objects allocated once per Schema and used to tag Node instances. They contain information about the node type, such as its name and what kind of node it represents.

name: string

The name the node type has in this schema.

schema: Schema

A link back to the Schema the node type belongs to.

spec: NodeSpec

The spec that this type is based on

contentMatch: ContentMatch

The starting match of the node type's content expression.

inlineContent: bool

True if this node type has inline content.

isBlock: bool

True if this is a block type

isText: bool

True if this is the text node type.

isInline: bool

True if this is an inline type.

isTextblock: bool

True if this is a textblock type, a block that contains inline content.

isLeaf: bool

True for node types that allow no content.

isAtom: bool

True when this node is an atom, i.e. when it does not have directly editable content.

hasRequiredAttrs() → bool

Tells you whether this node type has any required attributes.

create(attrs: ?⁠Object, content: ?⁠Fragment | Node | [Node], marks: ?⁠[Mark]) → Node

Create a Node of this type. The given attributes are checked and defaulted (you can pass null to use the type's defaults entirely, if no required attributes exist). content may be a Fragment, a node, an array of nodes, or null. Similarly marks may be null to default to the empty set of marks.

createChecked(attrs: ?⁠Object, content: ?⁠Fragment | Node | [Node], marks: ?⁠[Mark]) → Node

Like create, but check the given content against the node type's content restrictions, and throw an error if it doesn't match.

createAndFill(attrs: ?⁠Object, content: ?⁠Fragment | Node | [Node], marks: ?⁠[Mark]) → ?⁠Node

Like create, but see if it is necessary to add nodes to the start or end of the given fragment to make it fit the node. If no fitting wrapping can be found, return null. Note that, due to the fact that required nodes can always be created, this will always succeed if you pass null or Fragment.empty as content.

validContent(content: Fragment) → bool

Returns true if the given fragment is valid content for this node type with the given attributes.

allowsMarkType(markType: MarkType) → bool

Check whether the given mark type is allowed in this node.

allowsMarks(marks: [Mark]) → bool

Test whether the given set of marks are allowed in this node.

allowedMarks(marks: [Mark]) → [Mark]

Removes the marks that are not allowed in this node from the given set.

MarkType class

Like nodes, marks (which are associated with nodes to signify things like emphasis or being part of a link) are tagged with type objects, which are instantiated once per Schema.

name: string

The name of the mark type.

schema: Schema

The schema that this mark type instance is part of.

spec: MarkSpec

The spec on which the type is based.

create(attrs: ?⁠Object) → Mark

Create a mark of this type. attrs may be null or an object containing only some of the mark's attributes. The others, if they have defaults, will be added.

removeFromSet(set: [Mark]) → [Mark]

When there is a mark of this type in the given set, a new set without it is returned. Otherwise, the input set is returned.

isInSet(set: [Mark]) → ?⁠Mark

Tests whether there is a mark of this type in the given set.

excludes(other: MarkType) → bool

Queries whether a given mark type is excluded by this one.

ContentMatch class

Instances of this class represent a match state of a node type's content expression, and can be used to find out whether further content matches here, and whether a given position is a valid end of the node.

validEnd: bool

True when this match state represents a valid end of the node.

matchType(type: NodeType) → ?⁠ContentMatch

Match a node type, returning a match after that node if successful.

matchFragment(frag: Fragment, start: ?⁠number = 0, end: ?⁠number = frag.childCount) → ?⁠ContentMatch

Try to match a fragment. Returns the resulting match when successful.

defaultType: ?⁠NodeType

Get the first matching node type at this match position that can be generated.

fillBefore(after: Fragment, toEnd: ?⁠bool = false, startIndex: ?⁠number = 0) → ?⁠Fragment

Try to match the given fragment, and if that fails, see if it can be made to match by inserting nodes in front of it. When successful, return a fragment of inserted nodes (which may be empty if nothing had to be inserted). When toEnd is true, only return a fragment if the resulting match goes to the end of the content expression.

findWrapping(target: NodeType) → ?⁠[NodeType]

Find a set of wrapping node types that would allow a node of the given type to appear at this position. The result may be empty (when it fits directly) and will be null when no such wrapping exists.

edgeCount: number

The number of outgoing edges this node has in the finite automaton that describes the content expression.

edge(n: number) → {type: NodeType, next: ContentMatch}

Get the _n_​th outgoing edge from this node in the finite automaton that describes the content expression.

DOM Representation

Because representing a document as a tree of DOM nodes is central to the way ProseMirror operates, DOM parsing and serializing is integrated with the model.

(But note that you do not need to have a DOM implementation loaded to use this module.)

DOMParser class

A DOM parser represents a strategy for parsing DOM content into a ProseMirror document conforming to a given schema. Its behavior is defined by an array of rules.

new DOMParser(schema: Schema, rules: [ParseRule])

Create a parser that targets the given schema, using the given parsing rules.

schema: Schema

The schema into which the parser parses.

rules: [ParseRule]

The set of parse rules that the parser uses, in order of precedence.

parse(dom: dom.Node, options: ?⁠ParseOptions = {}) → Node

Parse a document from the content of a DOM node.

parseSlice(dom: dom.Node, options: ?⁠ParseOptions = {}) → Slice

Parses the content of the given DOM node, like parse, and takes the same set of options. But unlike that method, which produces a whole node, this one returns a slice that is open at the sides, meaning that the schema constraints aren't applied to the start of nodes to the left of the input and the end of nodes at the end.

static fromSchema(schema: Schema) → DOMParser

Construct a DOM parser using the parsing rules listed in a schema's node specs, reordered by priority.

ParseOptions interface

These are the options recognized by the parse and parseSlice methods.

preserveWhitespace: ?⁠bool | "full"

By default, whitespace is collapsed as per HTML's rules. Pass true to preserve whitespace, but normalize newlines to spaces, and "full" to preserve whitespace entirely.

findPositions: ?⁠[{node: dom.Node, offset: number}]

When given, the parser will, beside parsing the content, record the document positions of the given DOM positions. It will do so by writing to the objects, adding a pos property that holds the document position. DOM positions that are not in the parsed content will not be written to.

from: ?⁠number

The child node index to start parsing from.

to: ?⁠number

The child node index to stop parsing at.

topNode: ?⁠Node

By default, the content is parsed into the schema's default top node type. You can pass this option to use the type and attributes from a different node as the top container.

topMatch: ?⁠ContentMatch

Provide the starting content match that content parsed into the top node is matched against.

context: ?⁠ResolvedPos

A set of additional nodes to count as context when parsing, above the given top node.

ParseRule interface

A value that describes how to parse a given DOM node or inline style as a ProseMirror node or mark.

tag: ?⁠string

A CSS selector describing the kind of DOM elements to match. A single rule should have either a tag or a style property.

namespace: ?⁠string

The namespace to match. This should be used with tag. Nodes are only matched when the namespace matches or this property is null.

style: ?⁠string

A CSS property name to match. When given, this rule matches inline styles that list that property. May also have the form "property=value", in which case the rule only matches if the property's value exactly matches the given value. (For more complicated filters, use getAttrs and return false to indicate that the match failed.) Rules matching styles may only produce marks, not nodes.

priority: ?⁠number

Can be used to change the order in which the parse rules in a schema are tried. Those with higher priority come first. Rules without a priority are counted as having priority 50. This property is only meaningful in a schema—when directly constructing a parser, the order of the rule array is used.

consuming: ?⁠boolean

By default, when a rule matches an element or style, no further rules get a chance to match it. By setting this to false, you indicate that even when this rule matches, other rules that come after it should also run.

context: ?⁠string

When given, restricts this rule to only match when the current context—the parent nodes into which the content is being parsed—matches this expression. Should contain one or more node names or node group names followed by single or double slashes. For example "paragraph/" means the rule only matches when the parent node is a paragraph, "blockquote/paragraph/" restricts it to be in a paragraph that is inside a blockquote, and "section//" matches any position inside a section—a double slash matches any sequence of ancestor nodes. To allow multiple different contexts, they can be separated by a pipe (|) character, as in "blockquote/|list_item/".

node: ?⁠string

The name of the node type to create when this rule matches. Only valid for rules with a tag property, not for style rules. Each rule should have one of a node, mark, or ignore property (except when it appears in a node or mark spec, in which case the node or mark property will be derived from its position).

mark: ?⁠string

The name of the mark type to wrap the matched content in.

ignore: ?⁠bool

When true, ignore content that matches this rule.

closeParent: ?⁠bool

When true, finding an element that matches this rule will close the current node.

skip: ?⁠bool

When true, ignore the node that matches this rule, but do parse its content.

attrs: ?⁠Object

Attributes for the node or mark created by this rule. When getAttrs is provided, it takes precedence.

getAttrs: ?⁠fn(dom.Node | string) → ?⁠Object | false

A function used to compute the attributes for the node or mark created by this rule. Can also be used to describe further conditions the DOM element or style must match. When it returns false, the rule won't match. When it returns null or undefined, that is interpreted as an empty/default set of attributes.

Called with a DOM Element for tag rules, and with a string (the style's value) for style rules.

contentElement: ?⁠string | fn(dom.Node) → dom.Node

For tag rules that produce non-leaf nodes or marks, by default the content of the DOM element is parsed as content of the mark or node. If the child nodes are in a descendent node, this may be a CSS selector string that the parser must use to find the actual content element, or a function that returns the actual content element to the parser.

getContent: ?⁠fn(dom.Node, schema: Schema) → Fragment

Can be used to override the content of a matched node. When present, instead of parsing the node's child nodes, the result of this function is used.

preserveWhitespace: ?⁠bool | "full"

Controls whether whitespace should be preserved when parsing the content inside the matched element. false means whitespace may be collapsed, true means that whitespace should be preserved but newlines normalized to spaces, and "full" means that newlines should also be preserved.

DOMSerializer class

A DOM serializer knows how to convert ProseMirror nodes and marks of various types to DOM nodes.

new DOMSerializer(nodes: Object<fn(node: Node) → DOMOutputSpec>, marks: Object<?⁠fn(mark: Mark, inline: bool) → DOMOutputSpec>)

Create a serializer. nodes should map node names to functions that take a node and return a description of the corresponding DOM. marks does the same for mark names, but also gets an argument that tells it whether the mark's content is block or inline content (for typical use, it'll always be inline). A mark serializer may be null to indicate that marks of that type should not be serialized.

nodes: Object<fn(node: Node) → DOMOutputSpec>

The node serialization functions.

marks: Object<?⁠fn(mark: Mark, inline: bool) → DOMOutputSpec>

The mark serialization functions.

serializeFragment(fragment: Fragment, options: ?⁠Object = {}) → dom.DocumentFragment

Serialize the content of this fragment to a DOM fragment. When not in the browser, the document option, containing a DOM document, should be passed so that the serializer can create nodes.

serializeNode(node: Node, options: ?⁠Object = {}) → dom.Node

Serialize this node to a DOM node. This can be useful when you need to serialize a part of a document, as opposed to the whole document. To serialize a whole document, use serializeFragment on its content.

static renderSpec(doc: dom.Document, structure: DOMOutputSpec) → {dom: dom.Node, contentDOM: ?⁠dom.Node}

Render an output spec to a DOM node. If the spec has a hole (zero) in it, contentDOM will point at the node with the hole.

static fromSchema(schema: Schema) → DOMSerializer

Build a serializer using the toDOM properties in a schema's node and mark specs.

DOMOutputSpec interface

A description of a DOM structure. Can be either a string, which is interpreted as a text node, a DOM node, which is interpreted as itself, a {dom: Node, contentDOM: ?Node} object, or an array.

An array describes a DOM element. The first value in the array should be a string—the name of the DOM element, optionally prefixed by a namespace URL and a space. If the second element is plain object, it is interpreted as a set of attributes for the element. Any elements after that (including the 2nd if it's not an attribute object) are interpreted as children of the DOM elements, and must either be valid DOMOutputSpec values, or the number zero.

The number zero (pronounced “hole”) is used to indicate the place where a node's child nodes should be inserted. If it occurs in an output spec, it should be the only child element in its parent node.

prosemirror-transform module

This module defines a way of modifying documents that allows changes to be recorded, replayed, and reordered. You can read more about transformations in the guide.


Transforming happens in Steps, which are atomic, well-defined modifications to a document. Applying a step produces a new document.

Each step provides a change map that maps positions in the old document to position in the transformed document. Steps can be inverted to create a step that undoes their effect, and chained together in a convenience object called a Transform.

Step class

A step object represents an atomic change. It generally applies only to the document it was created for, since the positions stored in it will only make sense for that document.

New steps are defined by creating classes that extend Step, overriding the apply, invert, map, getMap and fromJSON methods, and registering your class with a unique JSON-serialization identifier using Step.jsonID.

apply(doc: Node) → StepResult

Applies this step to the given document, returning a result object that either indicates failure, if the step can not be applied to this document, or indicates success by containing a transformed document.

getMap() → StepMap

Get the step map that represents the changes made by this step, and which can be used to transform between positions in the old and the new document.

invert(doc: Node) → Step

Create an inverted version of this step. Needs the document as it was before the step as argument.

map(mapping: Mappable) → ?⁠Step

Map this step through a mappable thing, returning either a version of that step with its positions adjusted, or null if the step was entirely deleted by the mapping.

merge(other: Step) → ?⁠Step

Try to merge this step with another one, to be applied directly after it. Returns the merged step when possible, null if the steps can't be merged.

toJSON() → Object

Create a JSON-serializeable representation of this step. When defining this for a custom subclass, make sure the result object includes the step type's JSON id under the stepType property.

static fromJSON(schema: Schema, json: Object) → Step

Deserialize a step from its JSON representation. Will call through to the step class' own implementation of this method.

static jsonID(id: string, stepClass: constructor<Step>)

To be able to serialize steps to JSON, each step needs a string ID to attach to its JSON representation. Use this method to register an ID for your step classes. Try to pick something that's unlikely to clash with steps from other modules.

StepResult class

The result of applying a step. Contains either a new document or a failure value.

doc: ?⁠Node

The transformed document.

failed: ?⁠string

Text providing information about a failed step.

static ok(doc: Node) → StepResult

Create a successful step result.

static fail(message: string) → StepResult

Create a failed step result.

static fromReplace(doc: Node, from: number, to: number, slice: Slice) → StepResult

Call Node.replace with the given arguments. Create a successful result if it succeeds, and a failed one if it throws a ReplaceError.

ReplaceStep class extends Step

Replace a part of the document with a slice of new content.

new ReplaceStep(from: number, to: number, slice: Slice, structure: ?⁠bool)

The given slice should fit the 'gap' between from and to—the depths must line up, and the surrounding nodes must be able to be joined with the open sides of the slice. When structure is true, the step will fail if the content between from and to is not just a sequence of closing and then opening tokens (this is to guard against rebased replace steps overwriting something they weren't supposed to).

from: number

The start position of the replaced range.

to: number

The end position of the replaced range.

slice: Slice

The slice to insert.

ReplaceAroundStep class extends Step

Replace a part of the document with a slice of content, but preserve a range of the replaced content by moving it into the slice.

new ReplaceAroundStep(from: number, to: number, gapFrom: number, gapTo: number, slice: Slice, insert: number, structure: ?⁠bool)

Create a replace-around step with the given range and gap. insert should be the point in the slice into which the content of the gap should be moved. structure has the same meaning as it has in the ReplaceStep class.

from: number

The start position of the replaced range.

to: number

The end position of the replaced range.

gapFrom: number

The start of preserved range.

gapTo: number

The end of preserved range.

slice: Slice

The slice to insert.

insert: number

The position in the slice where the preserved range should be inserted.

AddMarkStep class extends Step

Add a mark to all inline content between two positions.

new AddMarkStep(from: number, to: number, mark: Mark)
from: number

The start of the marked range.

to: number

The end of the marked range.

mark: Mark

The mark to add.

RemoveMarkStep class extends Step

Remove a mark from all inline content between two positions.

new RemoveMarkStep(from: number, to: number, mark: Mark)
from: number

The start of the unmarked range.

to: number

The end of the unmarked range.

mark: Mark

The mark to remove.

Position Mapping

Mapping positions from one document to another by running through the step maps produced by steps is an important operation in ProseMirror. It is used, for example, for updating the selection when the document changes.

Mappable interface

There are several things that positions can be mapped through. Such objects conform to this interface.

map(pos: number, assoc: ?⁠number) → number

Map a position through this object. When given, assoc (should be -1 or 1, defaults to 1) determines with which side the position is associated, which determines in which direction to move when a chunk of content is inserted at the mapped position.

mapResult(pos: number, assoc: ?⁠number) → MapResult

Map a position, and return an object containing additional information about the mapping. The result's deleted field tells you whether the position was deleted (completely enclosed in a replaced range) during the mapping. When content on only one side is deleted, the position itself is only considered deleted when assoc points in the direction of the deleted content.

MapResult class

An object representing a mapped position with extra information.

pos: number

The mapped version of the position.

deleted: bool

Tells you whether the position was deleted, that is, whether the step removed its surroundings from the document.

StepMap class extends Mappable

A map describing the deletions and insertions made by a step, which can be used to find the correspondence between positions in the pre-step version of a document and the same position in the post-step version.

new StepMap(ranges: [number])

Create a position map. The modifications to the document are represented as an array of numbers, in which each group of three represents a modified chunk as [start, oldSize, newSize].

forEach(f: fn(oldStart: number, oldEnd: number, newStart: number, newEnd: number))

Calls the given function on each of the changed ranges included in this map.

invert() → StepMap

Create an inverted version of this map. The result can be used to map positions in the post-step document to the pre-step document.

static offset(n: number) → StepMap

Create a map that moves all positions by offset n (which may be negative). This can be useful when applying steps meant for a sub-document to a larger document, or vice-versa.

Mapping class extends Mappable

A mapping represents a pipeline of zero or more step maps. It has special provisions for losslessly handling mapping positions through a series of steps in which some steps are inverted versions of earlier steps. (This comes up when ‘rebasing’ steps for collaboration or history management.)

new Mapping(maps: ?⁠[StepMap])

Create a new mapping with the given position maps.

maps: [StepMap]

The step maps in this mapping.

from: number

The starting position in the maps array, used when map or mapResult is called.

to: number

The end position in the maps array.

slice(from: ?⁠number = 0, to: ?⁠number = this.maps.length) → Mapping

Create a mapping that maps only through a part of this one.

appendMap(map: StepMap, mirrors: ?⁠number)

Add a step map to the end of this mapping. If mirrors is given, it should be the index of the step map that is the mirror image of this one.

appendMapping(mapping: Mapping)

Add all the step maps in a given mapping to this one (preserving mirroring information).

getMirror(n: number) → ?⁠number

Finds the offset of the step map that mirrors the map at the given offset, in this mapping (as per the second argument to appendMap).

appendMappingInverted(mapping: Mapping)

Append the inverse of the given mapping to this one.

invert() → Mapping

Create an inverted version of this mapping.

Document transforms

Because you often need to collect a number of steps together to effect a composite change, ProseMirror provides an abstraction to make this easy. State transactions are a subclass of transforms.

Transform class

Abstraction to build up and track an array of steps representing a document transformation.

Most transforming methods return the Transform object itself, so that they can be chained.

new Transform(doc: Node)

Create a transform that starts with the given document.

doc: Node

The current document (the result of applying the steps in the transform).

steps: [Step]

The steps in this transform.

docs: [Node]

The documents before each of the steps.

mapping: Mapping

A mapping with the maps for each of the steps in this transform.

before: Node

The starting document.

step(step: Step) → this

Apply a new step in this transform, saving the result. Throws an error when the step fails.

maybeStep(step: Step) → StepResult

Try to apply a step in this transformation, ignoring it if it fails. Returns the step result.

docChanged: bool

True when the document has been changed (when there are any steps).

addMark(from: number, to: number, mark: Mark) → this

Add the given mark to the inline content between from and to.

removeMark(from: number, to: number, mark: ?⁠Mark | MarkType = null) → this

Remove marks from inline nodes between from and to. When mark is a single mark, remove precisely that mark. When it is a mark type, remove all marks of that type. When it is null, remove all marks of any type.

clearIncompatible(pos: number, parentType: NodeType, match: ?⁠ContentMatch) → this

Removes all marks and nodes from the content of the node at pos that don't match the given new parent node type. Accepts an optional starting content match as third argument.

replace(from: number, to: ?⁠number = from, slice: ?⁠Slice = Slice.empty) → this

Replace the part of the document between from and to with the given slice.

replaceWith(from: number, to: number, content: Fragment | Node | [Node]) → this

Replace the given range with the given content, which may be a fragment, node, or array of nodes.

delete(from: number, to: number) → this

Delete the content between the given positions.

insert(pos: number, content: Fragment | Node | [Node]) → this

Insert the given content at the given position.

replaceRange(from: number, to: number, slice: Slice) → this

Replace a range of the document with a given slice, using from, to, and the slice's openStart property as hints, rather than fixed start and end points. This method may grow the replaced area or close open nodes in the slice in order to get a fit that is more in line with WYSIWYG expectations, by dropping fully covered parent nodes of the replaced region when they are marked non-defining, or including an open parent node from the slice that is marked as defining.

This is the method, for example, to handle paste. The similar replace method is a more primitive tool which will not move the start and end of its given range, and is useful in situations where you need more precise control over what happens.

replaceRangeWith(from: number, to: number, node: Node) → this

Replace the given range with a node, but use from and to as hints, rather than precise positions. When from and to are the same and are at the start or end of a parent node in which the given node doesn't fit, this method may move them out towards a parent that does allow the given node to be placed. When the given range completely covers a parent node, this method may completely replace that parent node.

deleteRange(from: number, to: number) → this

Delete the given range, expanding it to cover fully covered parent nodes until a valid replace is found.

lift(range: NodeRange, target: number) → this

Split the content in the given range off from its parent, if there is sibling content before or after it, and move it up the tree to the depth specified by target. You'll probably want to use liftTarget to compute target, to make sure the lift is valid.

wrap(range: NodeRange, wrappers: [{type: NodeType, attrs: ?⁠Object}]) → this

Wrap the given range in the given set of wrappers. The wrappers are assumed to be valid in this position, and should probably be computed with findWrapping.

setBlockType(from: number, to: ?⁠number = from, type: NodeType, attrs: ?⁠Object) → this

Set the type of all textblocks (partly) between from and to to the given node type with the given attributes.

setNodeMarkup(pos: number, type: ?⁠NodeType, attrs: ?⁠Object, marks: ?⁠[Mark]) → this

Change the type, attributes, and/or marks of the node at pos. When type isn't given, the existing node type is preserved,

split(pos: number, depth: ?⁠number = 1, typesAfter: ?⁠[?⁠{type: NodeType, attrs: ?⁠Object}]) → this

Split the node at the given position, and optionally, if depth is greater than one, any number of nodes above that. By default, the parts split off will inherit the node type of the original node. This can be changed by passing an array of types and attributes to use after the split.

join(pos: number, depth: ?⁠number = 1) → this

Join the blocks around the given position. If depth is 2, their last and first siblings are also joined, and so on.

The following helper functions can be useful when creating transformations or determining whether they are even possible.

replaceStep(doc: Node, from: number, to: ?⁠number = from, slice: ?⁠Slice = Slice.empty) → ?⁠Step

‘Fit’ a slice into a given position in the document, producing a step that inserts it. Will return null if there's no meaningful way to insert the slice here, or inserting it would be a no-op (an empty slice over an empty range).

liftTarget(range: NodeRange) → ?⁠number

Try to find a target depth to which the content in the given range can be lifted. Will not go across isolating parent nodes.

findWrapping(range: NodeRange, nodeType: NodeType, attrs: ?⁠Object, innerRange: ?⁠NodeRange = range) → ?⁠[{type: NodeType, attrs: ?⁠Object}]

Try to find a valid way to wrap the content in the given range in a node of the given type. May introduce extra nodes around and inside the wrapper node, if necessary. Returns null if no valid wrapping could be found. When innerRange is given, that range's content is used as the content to fit into the wrapping, instead of the content of range.

canSplit(doc: Node, pos: number, depth: ?⁠number = 1, typesAfter: ?⁠[?⁠{type: NodeType, attrs: ?⁠Object}]) → bool

Check whether splitting at the given position is allowed.

canJoin(doc: Node, pos: number) → bool

Test whether the blocks before and after a given position can be joined.

joinPoint(doc: Node, pos: number, dir: ?⁠number = -1) → ?⁠number

Find an ancestor of the given position that can be joined to the block before (or after if dir is positive). Returns the joinable point, if any.

insertPoint(doc: Node, pos: number, nodeType: NodeType) → ?⁠number

Try to find a point where a node of the given type can be inserted near pos, by searching up the node hierarchy when pos itself isn't a valid place but is at the start or end of a node. Return null if no position was found.

dropPoint(doc: Node, pos: number, slice: Slice) → ?⁠number

Finds a position at or around the given position where the given slice can be inserted. Will look at parent nodes' nearest boundary and try there, even if the original position wasn't directly at the start or end of that node. Returns null when no position was found.

prosemirror-commands module

This module exports a number of commands, which are building block functions that encapsulate an editing action. A command function takes an editor state, optionally a dispatch function that it can use to dispatch a transaction and optionally an EditorView instance. It should return a boolean that indicates whether it could perform any action. When no dispatch callback is passed, the command should do a 'dry run', determining whether it is applicable, but not actually doing anything.

These are mostly used to bind keys and define menu items.

chainCommands(...commands: [fn(EditorState, ?⁠fn(tr: Transaction), ?⁠EditorView) → bool]) → fn(EditorState, ?⁠fn(tr: Transaction), ?⁠EditorView) → bool

Combine a number of command functions into a single function (which calls them one by one until one returns true).

deleteSelection(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Delete the selection, if there is one.

joinBackward(state: EditorState, dispatch: ?⁠fn(tr: Transaction), view: ?⁠EditorView) → bool

If the selection is empty and at the start of a textblock, try to reduce the distance between that block and the one before it—if there's a block directly before it that can be joined, join them. If not, try to move the selected block closer to the next one in the document structure by lifting it out of its parent or moving it into a parent of the previous block. Will use the view for accurate (bidi-aware) start-of-textblock detection if given.

selectNodeBackward(state: EditorState, dispatch: ?⁠fn(tr: Transaction), view: ?⁠EditorView) → bool

When the selection is empty and at the start of a textblock, select the node before that textblock, if possible. This is intended to be bound to keys like backspace, after joinBackward or other deleting commands, as a fall-back behavior when the schema doesn't allow deletion at the selected point.

joinForward(state: EditorState, dispatch: ?⁠fn(tr: Transaction), view: ?⁠EditorView) → bool

If the selection is empty and the cursor is at the end of a textblock, try to reduce or remove the boundary between that block and the one after it, either by joining them or by moving the other block closer to this one in the tree structure. Will use the view for accurate start-of-textblock detection if given.

selectNodeForward(state: EditorState, dispatch: ?⁠fn(tr: Transaction), view: ?⁠EditorView) → bool

When the selection is empty and at the end of a textblock, select the node coming after that textblock, if possible. This is intended to be bound to keys like delete, after joinForward and similar deleting commands, to provide a fall-back behavior when the schema doesn't allow deletion at the selected point.

joinUp(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Join the selected block or, if there is a text selection, the closest ancestor block of the selection that can be joined, with the sibling above it.

joinDown(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Join the selected block, or the closest ancestor of the selection that can be joined, with the sibling after it.

lift(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Lift the selected block, or the closest ancestor block of the selection that can be lifted, out of its parent node.

newlineInCode(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

If the selection is in a node whose type has a truthy code property in its spec, replace the selection with a newline character.

exitCode(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

When the selection is in a node with a truthy code property in its spec, create a default block after the code block, and move the cursor there.

createParagraphNear(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

If a block node is selected, create an empty paragraph before (if it is its parent's first child) or after it.

liftEmptyBlock(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

If the cursor is in an empty textblock that can be lifted, lift the block.

splitBlock(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Split the parent block of the selection. If the selection is a text selection, also delete its content.

splitBlockKeepMarks(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Acts like splitBlock, but without resetting the set of active marks at the cursor.

selectParentNode(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Move the selection to the node wrapping the current selection, if any. (Will not select the document node.)

selectAll(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Select the whole document.

wrapIn(nodeType: NodeType, attrs: ?⁠Object) → fn(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Wrap the selection in a node of the given type with the given attributes.

setBlockType(nodeType: NodeType, attrs: ?⁠Object) → fn(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Returns a command that tries to set the selected textblocks to the given node type with the given attributes.

toggleMark(markType: MarkType, attrs: ?⁠Object) → fn(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Create a command function that toggles the given mark with the given attributes. Will return false when the current selection doesn't support that mark. This will remove the mark if any marks of that type exist in the selection, or add it otherwise. If the selection is empty, this applies to the stored marks instead of a range of the document.

autoJoin(command: fn(state: EditorState, ?⁠fn(tr: Transaction)) → bool, isJoinable: fn(before: Node, after: Node) → bool | [string]) → fn(state: EditorState, ?⁠fn(tr: Transaction)) → bool

Wrap a command so that, when it produces a transform that causes two joinable nodes to end up next to each other, those are joined. Nodes are considered joinable when they are of the same type and when the isJoinable predicate returns true for them or, if an array of strings was passed, if their node type name is in that array.

baseKeymap: Object

Depending on the detected platform, this will hold pcBasekeymap or macBaseKeymap.

pcBaseKeymap: Object

A basic keymap containing bindings not specific to any schema. Binds the following keys (when multiple commands are listed, they are chained with chainCommands):

  • Enter to newlineInCode, createParagraphNear, liftEmptyBlock, splitBlock
  • Mod-Enter to exitCode
  • Backspace and Mod-Backspace to deleteSelection, joinBackward, selectNodeBackward
  • Delete and Mod-Delete to deleteSelection, joinForward, selectNodeForward
  • Mod-Delete to deleteSelection, joinForward, selectNodeForward
  • Mod-a to selectAll
macBaseKeymap: Object

A copy of pcBaseKeymap that also binds Ctrl-h like Backspace, Ctrl-d like Delete, Alt-Backspace like Ctrl-Backspace, and Ctrl-Alt-Backspace, Alt-Delete, and Alt-d like Ctrl-Delete.

prosemirror-history module

An implementation of an undo/redo history for ProseMirror. This history is selective, meaning it does not just roll back to a previous state but can undo some changes while keeping other, later changes intact. (This is necessary for collaborative editing, and comes up in other situations as well.)

history(config: ?⁠Object) → Plugin

Returns a plugin that enables the undo history for an editor. The plugin will track undo and redo stacks, which can be used with the undo and redo commands.

You can set an "addToHistory" metadata property of false on a transaction to prevent it from being rolled back by undo.

config: ?⁠Object

Supports the following configuration options:

depth: ?⁠number

The amount of history events that are collected before the oldest events are discarded. Defaults to 100.

newGroupDelay: ?⁠number

The delay between changes after which a new group should be started. Defaults to 500 (milliseconds). Note that when changes aren't adjacent, a new group is always started.

undo(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

A command function that undoes the last change, if any.

redo(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

A command function that redoes the last undone change, if any.

undoDepth(state: EditorState) → number

The amount of undoable events available in a given state.

redoDepth(state: EditorState) → number

The amount of redoable events available in a given editor state.

closeHistory(tr: Transaction) → Transaction

Set a flag on the given transaction that will prevent further steps from being appended to an existing history event (so that they require a separate undo command to undo).

prosemirror-collab module

This module implements an API into which a communication channel for collaborative editing can be hooked. See the guide for more details and an example.

collab(config: ?⁠Object = {}) → Plugin

Creates a plugin that enables the collaborative editing framework for the editor.

config: ?⁠Object

An optional set of options

version: ?⁠number

The starting version number of the collaborative editing. Defaults to 0.

clientID: ?⁠number | string

This client's ID, used to distinguish its changes from those of other clients. Defaults to a random 32-bit number.

getVersion(state: EditorState) → number

Get the version up to which the collab plugin has synced with the central authority.

receiveTransaction(state: EditorState, steps: [Step], clientIDs: [number | string], options: ?⁠Object) → Transaction

Create a transaction that represents a set of new steps received from the authority. Applying this transaction moves the state forward to adjust to the authority's view of the document.

options: ?⁠Object

Additional options.

mapSelectionBackward: ?⁠boolean

When enabled (the default is false), if the current selection is a text selection, its sides are mapped with a negative bias for this transaction, so that content inserted at the cursor ends up after the cursor. Users usually prefer this, but it isn't done by default for reasons of backwards compatibility.

sendableSteps(state: EditorState) → ?⁠{version: number, steps: [Step], clientID: number | string, origins: [Transaction]}

Provides data describing the editor's unconfirmed steps, which need to be sent to the central authority. Returns null when there is nothing to send.

origins holds the original transactions that produced each steps. This can be useful for looking up time stamps and other metadata for the steps, but note that the steps may have been rebased, whereas the origin transactions are still the old, unchanged objects.

prosemirror-keymap module

A plugin for conveniently defining key bindings.

keymap(bindings: Object) → Plugin

Create a keymap plugin for the given set of bindings.

Bindings should map key names to command-style functions, which will be called with (EditorState, dispatch, EditorView) arguments, and should return true when they've handled the key. Note that the view argument isn't part of the command protocol, but can be used as an escape hatch if a binding needs to directly interact with the UI.

Key names may be strings like "Shift-Ctrl-Enter"—a key identifier prefixed with zero or more modifiers. Key identifiers are based on the strings that can appear in KeyEvent.key. Use lowercase letters to refer to letter keys (or uppercase letters if you want shift to be held). You may use "Space" as an alias for the " " name.

Modifiers can be given in any order. Shift- (or s-), Alt- (or a-), Ctrl- (or c- or Control-) and Cmd- (or m- or Meta-) are recognized. For characters that are created by holding shift, the Shift- prefix is implied, and should not be added explicitly.

You can use Mod- as a shorthand for Cmd- on Mac and Ctrl- on other platforms.

You can add multiple keymap plugins to an editor. The order in which they appear determines their precedence (the ones early in the array get to dispatch first).

keydownHandler(bindings: Object) → fn(view: EditorView, event: dom.Event) → bool

Given a set of bindings (using the same format as keymap, return a keydown handler that handles them.

prosemirror-inputrules module

This module defines a plugin for attaching input rules to an editor, which can react to or transform text typed by the user. It also comes with a bunch of default rules that can be enabled in this plugin.

InputRule class

Input rules are regular expressions describing a piece of text that, when typed, causes something to happen. This might be changing two dashes into an emdash, wrapping a paragraph starting with "> " into a blockquote, or something entirely different.

new InputRule(match: RegExp, handler: string | fn(state: EditorState, match: [string], start: number, end: number) → ?⁠Transaction)

Create an input rule. The rule applies when the user typed something and the text directly in front of the cursor matches match, which should end with $.

The handler can be a string, in which case the matched text, or the first matched group in the regexp, is replaced by that string.

Or a it can be a function, which will be called with the match array produced by RegExp.exec, as well as the start and end of the matched range, and which can return a transaction that describes the rule's effect, or null to indicate the input was not handled.

inputRules(config: {rules: [InputRule]}) → Plugin

Create an input rules plugin. When enabled, it will cause text input that matches any of the given rules to trigger the rule's action.

undoInputRule(state: EditorState, dispatch: ?⁠fn(Transaction)) → bool

This is a command that will undo an input rule, if applying such a rule was the last thing that the user did.

The module comes with a number of predefined rules:

emDash: InputRule

Converts double dashes to an emdash.

ellipsis: InputRule

Converts three dots to an ellipsis character.

openDoubleQuote: InputRule

“Smart” opening double quotes.

closeDoubleQuote: InputRule

“Smart” closing double quotes.

openSingleQuote: InputRule

“Smart” opening single quotes.

closeSingleQuote: InputRule

“Smart” closing single quotes.

smartQuotes: [InputRule]

Smart-quote related input rules.

These utility functions take schema-specific parameters and create input rules specific to that schema.

wrappingInputRule(regexp: RegExp, nodeType: NodeType, getAttrs: ?⁠Object | fn([string]) → ?⁠Object, joinPredicate: ?⁠fn([string], Node) → bool) → InputRule

Build an input rule for automatically wrapping a textblock when a given string is typed. The regexp argument is directly passed through to the InputRule constructor. You'll probably want the regexp to start with ^, so that the pattern can only occur at the start of a textblock.

nodeType is the type of node to wrap in. If it needs attributes, you can either pass them directly, or pass a function that will compute them from the regular expression match.

By default, if there's a node with the same type above the newly wrapped node, the rule will try to join those two nodes. You can pass a join predicate, which takes a regular expression match and the node before the wrapped node, and can return a boolean to indicate whether a join should happen.

textblockTypeInputRule(regexp: RegExp, nodeType: NodeType, getAttrs: ?⁠Object | fn([string]) → ?⁠Object) → InputRule

Build an input rule that changes the type of a textblock when the matched text is typed into it. You'll usually want to start your regexp with ^ to that it is only matched at the start of a textblock. The optional getAttrs parameter can be used to compute the new node's attributes, and works the same as in the wrappingInputRule function.

prosemirror-gapcursor module

This is a plugin that adds a type of selection for focusing places that don't allow regular selection (such as positions that have a leaf block node, table, or the end of the document both before and after them).

You'll probably want to load style/gapcursor.css, which contains basic styling for the simulated cursor (as a short, blinking horizontal stripe).

By default, gap cursor are only allowed in places where the default content node (in the schema content constraints) is a textblock node. You can customize this by adding an allowGapCursor property to your node specs—if it's true, gap cursor are allowed everywhere in that node, if it's false they are never allowed.

gapCursor() → Plugin

Create a gap cursor plugin. When enabled, this will capture clicks near and arrow-key-motion past places that don't have a normally selectable position nearby, and create a gap cursor selection for them. The cursor is drawn as an element with class ProseMirror-gapcursor. You can either include style/gapcursor.css from the package's directory or add your own styles to make it visible.

GapCursor class extends Selection

Gap cursor selections are represented using this class. Its $anchor and $head properties both point at the cursor position.

prosemirror-schema-basic module

This module defines a simple schema. You can use it directly, extend it, or just pick out a few node and mark specs to use in a new schema.

schema: Schema

This schema roughly corresponds to the document schema used by CommonMark, minus the list elements, which are defined in the prosemirror-schema-list module.

To reuse elements from this schema, extend or read from its spec.nodes and spec.marks properties.

nodes: Object

Specs for the nodes defined in this schema.

doc: NodeSpec

The top level document node.

paragraph: NodeSpec

A plain paragraph textblock. Represented in the DOM as a <p> element.

blockquote: NodeSpec

A blockquote (<blockquote>) wrapping one or more blocks.

horizontal_rule: NodeSpec

A horizontal rule (<hr>).

heading: NodeSpec

A heading textblock, with a level attribute that should hold the number 1 to 6. Parsed and serialized as <h1> to <h6> elements.

code_block: NodeSpec

A code listing. Disallows marks or non-text inline nodes by default. Represented as a <pre> element with a <code> element inside of it.

text: NodeSpec

The text node.

image: NodeSpec

An inline image (<img>) node. Supports src, alt, and href attributes. The latter two default to the empty string.

hard_break: NodeSpec

A hard line break, represented in the DOM as <br>.

marks: Object

Specs for the marks in the schema.

A link. Has href and title attributes. title defaults to the empty string. Rendered and parsed as an <a> element.

em: MarkSpec

An emphasis mark. Rendered as an <em> element. Has parse rules that also match <i> and font-style: italic.

strong: MarkSpec

A strong mark. Rendered as <strong>, parse rules also match <b> and font-weight: bold.

code: MarkSpec

Code font mark. Represented as a <code> element.

prosemirror-schema-list module

This module exports list-related schema elements and commands. The commands assume lists to be nestable, with the restriction that the first child of a list item is a plain paragraph.

These are the node specs:

orderedList: NodeSpec

An ordered list node spec. Has a single attribute, order, which determines the number at which the list starts counting, and defaults to 1. Represented as an <ol> element.

bulletList: NodeSpec

A bullet list node spec, represented in the DOM as <ul>.

listItem: NodeSpec

A list item (<li>) spec.

addListNodes(nodes: OrderedMap<NodeSpec>, itemContent: string, listGroup: ?⁠string) → OrderedMap<NodeSpec>

Convenience function for adding list-related node types to a map specifying the nodes for a schema. Adds orderedList as "ordered_list", bulletList as "bullet_list", and listItem as "list_item".

itemContent determines the content expression for the list items. If you want the commands defined in this module to apply to your list structure, it should have a shape like "paragraph block*" or "paragraph (ordered_list | bullet_list)*". listGroup can be given to assign a group name to the list node types, for example "block".

Using this would look something like this:

const mySchema = new Schema({
  nodes: addListNodes(baseSchema.spec.nodes, "paragraph block*", "block"),
  marks: baseSchema.spec.marks

The following functions are commands:

wrapInList(listType: NodeType, attrs: ?⁠Object) → fn(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Returns a command function that wraps the selection in a list with the given type an attributes. If dispatch is null, only return a value to indicate whether this is possible, but don't actually perform the change.

splitListItem(itemType: NodeType) → fn(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Build a command that splits a non-empty textblock at the top level of a list item by also splitting that list item.

liftListItem(itemType: NodeType) → fn(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Create a command to lift the list item around the selection up into a wrapping list.

sinkListItem(itemType: NodeType) → fn(state: EditorState, dispatch: ?⁠fn(tr: Transaction)) → bool

Create a command to sink the list item around the selection down into an inner list.