The features in this specification extend or modify those found in Pointer Events, a W3C Recommendation that describes events and related interfaces for handling hardware agnostic pointer input from devices including a mouse, pen, touchscreen, etc. For compatibility with existing mouse based content, this specification also describes a mapping to fire Mouse Events for other pointer device types.

Pointer Events and interfaces

`PointerEvent` interface

dictionary PointerEventInit : MouseEventInit {
    long        pointerId = 0;
    double      width = 1;
    double      height = 1;
    float       pressure = 0;
    float       tangentialPressure = 0;
    long        tiltX;
    long        tiltY;
    long        twist = 0;
    double      altitudeAngle;
    double      azimuthAngle;
    DOMString   pointerType = "";
    boolean     isPrimary = false;
    long        persistentDeviceId = 0;
    sequence<PointerEvent> coalescedEvents = [];
    sequence<PointerEvent> predictedEvents = [];
};

[Exposed=Window]
interface PointerEvent : MouseEvent {
    constructor(DOMString type, optional PointerEventInit eventInitDict = {});
    readonly        attribute long        pointerId;
    readonly        attribute double      width;
    readonly        attribute double      height;
    readonly        attribute float       pressure;
    readonly        attribute float       tangentialPressure;
    readonly        attribute long        tiltX;
    readonly        attribute long        tiltY;
    readonly        attribute long        twist;
    readonly        attribute double      altitudeAngle;
    readonly        attribute double      azimuthAngle;
    readonly        attribute DOMString   pointerType;
    readonly        attribute boolean     isPrimary;
    readonly        attribute long        persistentDeviceId;
    [SecureContext] sequence<PointerEvent> getCoalescedEvents();
    sequence<PointerEvent> getPredictedEvents();
};

pointerId

A unique identifier for the pointer causing the event. User agents MAY reserve a generic pointerId value of 0 or 1 for the primary mouse pointer. The pointerId value of -1 MUST be reserved and used to indicate events that were generated by something other than a pointing device. For any other pointers, user agents are free to implement different strategies and approaches in how they assign a pointerId value. However, all active pointers in the [=top-level browsing context=] (as defined by [[HTML]]) must be unique, and the identifier MUST NOT be influenced by any other top-level browsing context (i.e. one top-level browsing context cannot assume that the pointerId of a pointer will be the same when the pointer moves outside of the browsing context and into another top-level browsing context). The user agent MAY recycle previously retired values for pointerId from previous active pointers, or it MAY always reuse the same pointerId for a particular pointing device (for instance, to uniquely identify particular pen/stylus inputs from a specific user in a multi-user collaborative application). However, in the latter case, to minimize the chance of fingerprinting and tracking across different pages or domains, the pointerId MUST only be associated explicitly with that particular pointing device for the lifetime of the page / session, and a new randomized pointerId MUST be chosen the next time that particular pointing device is used again in a new session.

The pointerId selection algorithm is implementation specific. Therefore authors cannot assume values convey any particular meaning other than an identifier for the pointer that is unique from all other active pointers. As an example, user agents may simply assign a number, starting from 0, to any active pointers, in the order that they become active — but these values are not guaranteed to be monotonically increasing.

width

The width (magnitude on the X axis), in CSS pixels (see [[CSS21]]), of the contact geometry of the pointer. This value MAY be updated on each event for a given pointer. For inputs that typically lack contact geometry (such as a traditional mouse), and in cases where the actual geometry of the input is not detected by the hardware, the user agent MUST return a default value of 1.

height

The height (magnitude on the Y axis), in CSS pixels (see [[CSS21]]), of the contact geometry of the pointer. This value MAY be updated on each event for a given pointer. For inputs that typically lack contact geometry (such as a traditional mouse), and in cases where the actual geometry of the input is not detected by the hardware, the user agent MUST return a default value of 1.

pressure

The normalized pressure of the pointer input in the range of [0,1], where 0 and 1 represent the minimum and maximum pressure the hardware is capable of detecting, respectively. For hardware and platforms that do not support pressure, the value MUST be 0.5 when in the active buttons state and 0 otherwise.

tangentialPressure

The normalized tangential pressure (also known as barrel pressure), typically set by an additional control (e.g. a finger wheel on an airbrush stylus), of the pointer input in the range of [-1,1], where 0 is the neutral position of the control. Note that some hardware may only support positive values in the range of [0,1]. For hardware and platforms that do not support tangential pressure, the value MUST be 0.

Despite the property's name, in practice the hardware controls/sensors that generate the values for this property may not necessarily be pressure sensitive. As an example, in most cases the finger wheel on most airbrush/painting stylus implementations can be freely set, rather than requiring the user to apply a constant pressure on the wheel to prevent it from returning to the zero position.

tiltX

The plane angle (in degrees, in the range of [-90,90]) between the Y-Z plane and the plane containing both the transducer (e.g. pen/stylus) axis and the Y axis. A positive tiltX is to the right, in the direction of increasing X values. tiltX can be used along with tiltY to represent the tilt away from the normal of a transducer with the digitizer. For hardware and platforms that do not report tilt or angle, the value MUST be 0.

tiltY

The plane angle (in degrees, in the range of [-90,90]) between the X-Z plane and the plane containing both the transducer (e.g. pen/stylus) axis and the X axis. A positive tiltY is towards the user, in the direction of increasing Y values. tiltY can be used along with tiltX to represent the tilt away from the normal of a transducer with the digitizer. For hardware and platforms that do not report tilt or angle, the value MUST be 0.

twist

The clockwise rotation (in degrees, in the range of [0,359]) of a transducer (e.g. pen/stylus) around its own major axis. For hardware and platforms that do not report twist, the value MUST be 0.

altitudeAngle

The altitude (in radians) of the transducer (e.g. pen/stylus), in the range [0,π/2] — where 0 is parallel to the surface (X-Y plane), and π/2 is perpendicular to the surface. For hardware and platforms that do not report tilt or angle, the value MUST be π/2.

The default value defined here for altitudeAngle is π/2, which positions the transducer as being perpendicular to the surface. This differs from the Touch Events - Level 2 specification's definition for the altitudeAngle property, which has a default value of 0.

altitudeAngle explanation diagram — Example `altitudeAngle` of `π/4` (45 degrees from the X-Y plane).

azimuthAngle

The azimuth angle (in radians) of the transducer (e.g. pen/stylus), in the range [0, 2π] — where 0 represents a transducer whose cap is pointing in the direction of increasing X values (point to "3 o'clock" if looking straight down) on the X-Y plane, and the values progressively increase when going clockwise (π/2 at "6 o'clock", π at "9 o'clock", 3π/2 at "12 o'clock"). When the transducer is perfectly perpendicular to the surface (altitudeAngle of π/2), the value MUST be 0. For hardware and platforms that do not report tilt or angle, the value MUST be 0.

pointerType

Indicates the device type that caused the event (mouse, pen, touch, etc.). If the user agent is to fire a pointer event for a mouse, pen/stylus, or touch input device, then the value of pointerType MUST be according to the following table:

Pointer Device Type	`pointerType` Value
Mouse	`mouse`
Pen / stylus	`pen`
Touch contact	`touch`

If the device type cannot be detected by the user agent, then the value MUST be an empty string. If the user agent supports pointer device types other than those listed above, the value of pointerType SHOULD be vendor prefixed to avoid conflicting names for different types of devices. Future specifications MAY provide additional normative values for other device types.

See Example 2 for a basic demonstration of how the pointerType can be used. Also note that developers should include some form of default handling to cover user agents that may have implemented their own custom pointerType values and for situations where pointerType is simply an empty string.

isPrimary

Indicates if the pointer represents the primary pointer of this pointer type.

persistentDeviceId

A unique identifier for the pointing device. If the hardware supports multiple pointers, pointer events generated from pointing devices MUST only get a persistentDeviceId if those pointers are uniquely identifiable over the session. If the pointer is uniquely identifiable, the assigned persistentDeviceId to that pointing device will remain constant for the remainder of the session. The persistentDeviceId value of 0 MUST be reserved and used to indicate events whose generating device could not be identified. Like pointerId, to minimize the chance of fingerprinting and tracking across different pages or domains, the persistentDeviceId MUST only be associated explicitly with that particular pointing device for the lifetime of the page / session, and a new randomized persistentDeviceId MUST be chosen the next time that particular pointing device is used again in a new session.

Due to digitizer and pointing device hardware constraints, a persistentDeviceId is not guaranteed to be available for all pointer events from a pointing device. For example, the device may not report its hardware id to the digitizer in time for pointerdown to have a persistentDeviceId. In such a case, the persistentDeviceId may initially be 0 and change to a valid value.

getCoalescedEvents()

A method that returns the list of coalesced events.

getPredictedEvents()

A method that returns the list of predicted events.

The PointerEventInit dictionary is used by the {{PointerEvent}} interface's constructor to provide a mechanism by which to construct untrusted (synthetic) pointer events. It inherits from the {{MouseEventInit}} dictionary defined in [[UIEVENTS]]. See the examples for sample code demonstrating how to fire an untrusted pointer event.

The [=event constructing steps=] for PointerEvent clones PointerEventInit's coalescedEvents to coalesced events list and clones PointerEventInit's predictedEvents to predicted events list.

The PointerEvent interface inherits from {{MouseEvent}}, defined in [[[UIEVENTS]]]. Also note the proposed extension in [[[CSSOM-VIEW]]], which changes the various coordinate properties from long to double to allow for fractional coordinates. For user agents that already implement this proposed extension for {{PointerEvent}}, but not for regular {{MouseEvent}}, there are additional requirements when it comes to the click, auxclick, and contextmenu events.

Button states

Chorded button interactions

Some pointer devices, such as mouse or pen, support multiple buttons. In the [[UIEVENTS]] Mouse Event model, each button press produces a mousedown and mouseup event. To better abstract this hardware difference and simplify cross-device input authoring, Pointer Events do not fire overlapping {{GlobalEventHandlers/pointerdown}} and {{GlobalEventHandlers/pointerup}} events for chorded button presses (depressing an additional button while another button on the pointer device is already depressed).

Instead, chorded button presses can be detected by inspecting changes to the button and buttons properties. The button and buttons properties are inherited from the {{MouseEvent}} interface, but with a change in semantics and values, as outlined in the following sections.

The modifications to the button and buttons properties apply only to pointer events. For any compatibility mouse events the value of button and buttons MUST follow [[UIEVENTS]].

The `button` property

To identify button state transitions in any pointer event (and not just {{GlobalEventHandlers/pointerdown}} and {{GlobalEventHandlers/pointerup}}), the button property indicates the device button whose state change fired the event.

Device Button Changes	`button`
Neither buttons nor touch/pen contact changed since last event	-1
Left Mouse, Touch contact, Pen contact	0
Middle Mouse	1
Right Mouse, Pen barrel button	2
X1 (back) Mouse	3
X2 (forward) Mouse	4
Pen eraser button	5

During a mouse drag, the value of the button property in a {{GlobalEventHandlers/pointermove}} event will be different from that in a mousemove event. For example, while moving the mouse with the right button pressed, the {{GlobalEventHandlers/pointermove}} events will have the button value -1, but the mousemove events will have the button value 2.

The `buttons` property

The buttons property gives the current state of the device buttons as a bitmask (same as in MouseEvent, but with an expanded set of possible values).

Current state of device buttons	`buttons`
Mouse moved with no buttons pressed, Pen moved while hovering with no buttons pressed	0
Left Mouse, Touch contact, Pen contact	1
Middle Mouse	4
Right Mouse, Pen barrel button	2
X1 (back) Mouse	8
X2 (forward) Mouse	16
Pen eraser button	32

The primary pointer

In a multi-pointer (e.g. multi-touch) scenario, the isPrimary property is used to identify a master pointer amongst the set of active pointers for each pointer type.

At any given time, there can only ever be at most one primary pointer for each pointer type.
The first pointer to become active for a particular pointer type (e.g. the first finger to touch the screen in a multi-touch interaction) becomes the primary pointer for that pointer type.
Only a primary pointer will produce compatibility mouse events. In the case where there are multiple primary pointers, these pointers will all produce compatibility mouse events.

Authors who desire single-pointer interaction can achieve this by ignoring non-primary pointers (however, see the note below on multiple primary pointers).

When two or more pointer device types are being used concurrently, multiple pointers (one for each pointerType) are considered primary. For example, a touch contact and a mouse cursor moved simultaneously will produce pointers that are both considered primary.

Some devices, operating systems and user agents may ignore the concurrent use of more than one type of pointer input to avoid accidental interactions. For instance, devices that support both touch and pen interactions may ignore touch inputs while the pen is actively being used, to allow users to rest their hand on the touchscreen while using the pen (a feature commonly referred to as "palm rejection"). Currently, it is not possible for authors to suppress this behavior.

In some cases, it is possible for the user agent to fire pointer events in which no pointer is marked as a primary pointer. For instance, when there are multiple active pointers of a particular type, like a multi-touch interaction, and the primary pointer is removed (e.g. it leaves the screen), there may end up being no primary pointers. Also on platforms where the primary pointer is determined using all active pointers of the same type on the device (including those targeted at an application other than the user agent), if the first (primary) pointer is outside of the user agent and other (non-primary) pointers targeted inside the user agent, then the user agent may fire pointer events for the other pointers with a value of false for isPrimary.

Current operating systems and user agents don't usually have a concept of multiple mouse inputs. When more than one mouse device is present (for instance, on a laptop with both a trackpad and an external mouse), all mouse devices are generally treated as a single device — movements on any of the devices are translated to movement of a single mouse pointer, and there is no distinction between button presses on different mouse devices. For this reason, there will usually only be a single mouse pointer, and that pointer will be primary.

Firing events using the `PointerEvent` interface

To fire a pointer event named |e| means to [=fire an event=] named |e| using PointerEvent whose attributes are set as defined in {{PointerEvent}} Interface and Attributes and Default Actions.

If the event is not a {{GlobalEventHandlers/gotpointercapture}}, {{GlobalEventHandlers/lostpointercapture}}, click, auxclick or contextmenu event, run the process pending pointer capture steps for this PointerEvent.

The target object at which the event is fired is determined as follows:

If the pointer capture target override has been set for the pointer, set the target to pointer capture target override object.
Otherwise, set the target to the object returned by normal hit test mechanisms (out of scope for this specification).

Let |targetDocument| be target's [=Node/node document=] [[DOM]].

If the event is {{GlobalEventHandlers/pointerdown}}, {{GlobalEventHandlers/pointermove}}, or {{GlobalEventHandlers/pointerup}} set active document for the event's pointerId to |targetDocument|.

If the event is {{GlobalEventHandlers/pointerdown}}, the associated device is a direct manipulation device, and the target is an {{Element}}, then set pointer capture for this pointerId to the target element as described in implicit pointer capture.

Before firing this event, the user agent SHOULD treat the target as if the pointing device has moved over it from the |previousTarget| for the purpose of ensuring event ordering [[UIEVENTS]]. If the |needsOverEvent| flag is set, a {{GlobalEventHandlers/pointerover}} event is needed even if the target element is the same.

Fire the event to the determined target.

Using the pointer capture target override as the target instead of the normal hit-test result may fire some boundary events, as defined by [[UIEVENTS]]. This is the same as the pointer leaving its previous target and entering this new capturing target. When the capture is released, the same scenario may happen, as the pointer is leaving the capturing target and entering the hit-test target.

Attributes and default actions

The bubbles and cancelable properties and the default actions for the event types defined in this specification appear in the following table. Details of each of these event types are provided in Pointer Event types.

Event Type	Bubbles	Cancelable	Default Action
{{GlobalEventHandlers/pointerover}}	Yes	Yes	None
{{GlobalEventHandlers/pointerenter}}	No	No	None
{{GlobalEventHandlers/pointerdown}}	Yes	Yes	Varies: when the pointer is primary, all default actions of the `mousedown` event Canceling this event also prevents subsequent firing of compatibility mouse events.
{{GlobalEventHandlers/pointermove}}	Yes	Yes	Varies: when the pointer is primary, all default actions of `mousemove`
{{GlobalEventHandlers/pointerrawupdate}}	Yes	No	None
{{GlobalEventHandlers/pointerup}}	Yes	Yes	Varies: when the pointer is primary, all default actions of `mouseup`
{{GlobalEventHandlers/pointercancel}}	Yes	No	None
{{GlobalEventHandlers/pointerout}}	Yes	Yes	None
{{GlobalEventHandlers/pointerleave}}	No	No	None
{{GlobalEventHandlers/gotpointercapture}}	Yes	No	None
{{GlobalEventHandlers/lostpointercapture}}	Yes	No	None

Viewport manipulations (panning and zooming) — generally, as a result of a direct manipulation interaction — are intentionally NOT a default action of pointer events, meaning that these behaviors (e.g. panning a page as a result of moving a finger on a touchscreen) cannot be suppressed by canceling a pointer event. Authors must instead use touch-action to explicitly declare the direct manipulation behavior for a region of the document. Removing this dependency on the cancelation of events facilitates performance optimizations by the user agent.

For {{GlobalEventHandlers/pointerenter}} and {{GlobalEventHandlers/pointerleave}} events, the {{EventInit/composed}} [[DOM]] attribute SHOULD be false; for all other pointer events in the table above, the attribute SHOULD be true.

For all pointer events in the table above, the {{UIEvent/detail}} [[UIEVENTS]] attribute SHOULD be 0.

Many user agents expose non-standard attributes fromElement and toElement in MouseEvents to support legacy content. We encourage those user agents to set the values of those (inherited) attributes in PointerEvents to null to transition authors to the use of standardized alternates (i.e. target and relatedTarget).

Similar to MouseEvent {{MouseEventInit/relatedTarget}}, the relatedTarget should be initialized to the element whose bounds the pointer just left (in the case of a {{GlobalEventHandlers/pointerover}} or pointerenter event) or the element whose bounds the pointer is entering (in the case of a {{GlobalEventHandlers/pointerout}} or {{GlobalEventHandlers/pointerleave}}). For other pointer events, this value will default to null. Note that when an element receives the pointer capture all the following events for that pointer are considered to be inside the boundary of the capturing element.

For {{GlobalEventHandlers/gotpointercapture}} and {{GlobalEventHandlers/lostpointercapture}} events, all the attributes except the ones defined in the table above should be the same as the Pointer Event that caused the user agent to run the process pending pointer capture steps and fire the {{GlobalEventHandlers/gotpointercapture}} and {{GlobalEventHandlers/lostpointercapture}} events.

Process pending pointer capture

The user agent MUST run the following steps when implicitly releasing pointer capture as well as when firing Pointer Events that are not {{GlobalEventHandlers/gotpointercapture}} or {{GlobalEventHandlers/lostpointercapture}}.

If the pointer capture target override for this pointer is set and is not equal to the pending pointer capture target override, then fire a pointer event named {{GlobalEventHandlers/lostpointercapture}} at the pointer capture target override node.
If the pending pointer capture target override for this pointer is set and is not equal to the pointer capture target override, then fire a pointer event named {{GlobalEventHandlers/gotpointercapture}} at the pending pointer capture target override.
Set the pointer capture target override to the pending pointer capture target override, if set. Otherwise, clear the pointer capture target override.

As defined in the section for click, auxclick, and contextmenu events, even after the {{GlobalEventHandlers/lostpointercapture}} event has been dispatched, the corresponding click, auxclick or contextmenu event, if any, would still be dispatched to the capturing target.

Suppressing a pointer event stream

The user agent MUST suppress a pointer event stream when it detects that a pointer is unlikely to continue to produce events. Any of the following scenarios satisfy this condition (there MAY be additional scenarios):

The user agent has opened a modal dialog or menu.
A pointer input device is physically disconnected, or a hoverable pointer input device (e.g. a hoverable pen/stylus) has left the hover range detectable by the digitizer.
The pointer is subsequently used by the user agent to manipulate the page viewport (e.g. panning or zooming). See the section on touch-action CSS property for details.
User agents can trigger panning or zooming through multiple pointer types (such as touch and pen), and therefore the start of a pan or zoom action may result in the suppression of various pointers, including pointers with different pointer types.
As part of the drag operation initiation algorithm as defined in the drag and drop processing model [[HTML]], for the pointer that caused the drag operation.

Other scenarios in which the user agent MAY suppress a pointer event stream include:

A device's screen orientation is changed while a pointer is active.
The user attempts to interact using more simultaneous pointer inputs than the device supports.
The user agent interprets the input as accidental (for example, the hardware supports palm rejection).

Methods for detecting any of these scenarios are out of scope for this specification.

The user agent MUST run the following steps to suppress a pointer event stream:

Fire a {{GlobalEventHandlers/pointercancel}} event.
Fire a {{GlobalEventHandlers/pointerout}} event.
Fire a {{GlobalEventHandlers/pointerleave}} event.
Implicitly release the pointer capture if the pointer is currently captured.

Boundary events caused by layout changes

A pointing device that moved relative to the screen surface or underwent some change in any of its properties fires various events as defined in Pointer Event types. For a stationary pointing device (that neither moved relative to the screen surface nor underwent any change in any properties), the user agent MUST fire certain boundary events after a layout change that affected the hit test target for the pointer, see {{GlobalEventHandlers/pointerover}}, {{GlobalEventHandlers/pointerenter}}, {{GlobalEventHandlers/pointerout}} and {{GlobalEventHandlers/pointerleave}} for details. The user agent MAY delay the firing of these boundary events because of performance reasons (e.g. to avoid too many hit-tests or layout changes caused by boundary event listeners).

A stationary pointing device (that neither moved relative to the screen surface nor underwent any change in any properties) never fires a {{GlobalEventHandlers/pointermove}} event.

Converting between `tiltX` / `tiltY` and `altitudeAngle` / `azimuthAngle`

Pointer Events include two complementary sets of attributes to express the orientation of a transducer relative to the X-Y plane: tiltX / tiltY (introduced in the original Pointer Events specification), and azimuthAngle / altitudeAngle (adopted from the Touch Events - Level 2 specification).

Depending on the specific hardware and platform, user agents will likely only receive one set of values for the transducer orientation relative to the screen plane — either tiltX / tiltY or altitudeAngle / azimuthAngle. User agents MUST use the following algorithm for converting these values.

When the user agent calculates tiltX / tiltY from azimuthAngle / altitudeAngle it SHOULD round the final integer values using Math.round [[ECMASCRIPT]] rules.

/* Converting between tiltX/tiltY and altitudeAngle/azimuthAngle */

function spherical2tilt(altitudeAngle, azimuthAngle) {
  const radToDeg = 180/Math.PI;

  let tiltXrad = 0;
  let tiltYrad = 0;

  if (altitudeAngle == 0) {
    // the pen is in the X-Y plane
    if (azimuthAngle == 0 || azimuthAngle == 2*Math.PI) {
      // pen is on positive X axis
      tiltXrad = Math.PI/2;
    }
    if (azimuthAngle == Math.PI/2) {
      // pen is on positive Y axis
      tiltYrad = Math.PI/2;
    }
    if (azimuthAngle == Math.PI) {
      // pen is on negative X axis
      tiltXrad = -Math.PI/2;
    }
    if (azimuthAngle == 3*Math.PI/2) {
      // pen is on negative Y axis
      tiltYrad = -Math.PI/2;
    }
    if (azimuthAngle>0 && azimuthAngle<Math.PI/2) {
      tiltXrad = Math.PI/2;
      tiltYrad = Math.PI/2;
    }
    if (azimuthAngle>Math.PI/2 && azimuthAngle<Math.PI) {
      tiltXrad = -Math.PI/2;
      tiltYrad = Math.PI/2;
    }
    if (azimuthAngle>Math.PI && azimuthAngle<3*Math.PI/2) {
      tiltXrad = -Math.PI/2;
      tiltYrad = -Math.PI/2;
    }
    if (azimuthAngle>3*Math.PI/2 && azimuthAngle<2*Math.PI) {
      tiltXrad = Math.PI/2;
      tiltYrad = -Math.PI/2;
    }
  }

  if (altitudeAngle != 0) {
    const tanAlt = Math.tan(altitudeAngle);

    tiltXrad = Math.atan(Math.cos(azimuthAngle) / tanAlt);
    tiltYrad = Math.atan(Math.sin(azimuthAngle) / tanAlt);
  }

  return {"tiltX":tiltXrad*radToDeg, "tiltY":tiltYrad*radToDeg};
}

function tilt2spherical(tiltX, tiltY) {
  const tiltXrad = tiltX * Math.PI/180;
  const tiltYrad = tiltY * Math.PI/180;

  // calculate azimuth angle
  let azimuthAngle = 0;

  if (tiltX == 0) {
    if (tiltY > 0) {
      azimuthAngle = Math.PI/2;
    }
    else if (tiltY < 0) {
      azimuthAngle = 3*Math.PI/2;
    }
  } else if (tiltY == 0) {
    if (tiltX < 0) {
      azimuthAngle = Math.PI;
    }
  } else if (Math.abs(tiltX) == 90 || Math.abs(tiltY) == 90) {
    // not enough information to calculate azimuth
    azimuthAngle = 0;
  } else {
    // Non-boundary case: neither tiltX nor tiltY is equal to 0 or +-90
    const tanX = Math.tan(tiltXrad);
    const tanY = Math.tan(tiltYrad);

    azimuthAngle = Math.atan2(tanY, tanX);
    if (azimuthAngle < 0) {
      azimuthAngle += 2*Math.PI;
    }
  }

  // calculate altitude angle
  let altitudeAngle = 0;

  if (Math.abs(tiltX) == 90 || Math.abs(tiltY) == 90) {
      altitudeAngle = 0
  } else if (tiltX == 0) {
    altitudeAngle = Math.PI/2 - Math.abs(tiltYrad);
  } else if (tiltY == 0) {
    altitudeAngle = Math.PI/2 - Math.abs(tiltXrad);
  } else {
    // Non-boundary case: neither tiltX nor tiltY is equal to 0 or +-90
    altitudeAngle =  Math.atan(1.0/Math.sqrt(Math.pow(Math.tan(tiltXrad),2) + Math.pow(Math.tan(tiltYrad),2)));
  }

  return {"altitudeAngle":altitudeAngle, "azimuthAngle":azimuthAngle};
}

Pointer Event types

Below are the event types defined in this specification.

In the case of the primary pointer, these events (with the exception of {{GlobalEventHandlers/gotpointercapture}} and {{GlobalEventHandlers/lostpointercapture}}) may also fire compatibility mouse events.

The pointerover event

The user agent MUST fire a pointer event named {{GlobalEventHandlers/pointerover}} when any of the following occurs:

A pointing device is moved into the hit test boundaries of an element. Note that setPointerCapture() or releasePointerCapture() might have changed the hit test target. Also note that while a pointer is captured it is considered to be always inside the boundaries of the capturing element for the purpose of firing boundary events.
A layout change causes the hit test boundaries of an element to move underneath an uncaptured primary pointer of a stationary pointing device.
Before the user agent fires a {{GlobalEventHandlers/pointerdown}} event for a device that does not support hover (see {{GlobalEventHandlers/pointerdown}}).

The pointerenter event

The user agent MUST fire a pointer event named {{GlobalEventHandlers/pointerenter}} when any of the following occurs:

A pointing device is moved into the hit test boundaries of an element or one of its descendants. Note that setPointerCapture() or releasePointerCapture() might have changed the hit test target. Also note that while a pointer is captured it is considered to be always inside the boundaries of the capturing element for the purpose of firing boundary events.
A layout change causes the hit test boundaries of an element or one of its descendants to move underneath an uncaptured primary pointer of a stationary pointing device.
Before the user agent fires a {{GlobalEventHandlers/pointerdown}} event for a device that does not support hover (see {{GlobalEventHandlers/pointerdown}}).

This event type is similar to {{GlobalEventHandlers/pointerover}} but with two differences: {{GlobalEventHandlers/pointerenter}} does not bubble, and its dispatch considers the hit test boundaries of even descendant elements.

There are similarities between this event type, the mouseenter event described in [[UIEVENTS]], and the CSS :hover pseudo-class described in [[CSS21]]. See also the {{GlobalEventHandlers/pointerleave}} event.

The pointerdown event

The user agent MUST fire a pointer event named {{GlobalEventHandlers/pointerdown}} when a pointer enters the active buttons state. For mouse, this is when the device transitions from no buttons depressed to at least one button depressed. For touch, this is when physical contact is made with the digitizer. For pen, this is when the pen either makes physical contact with the digitizer without any button depressed, or transitions from no buttons depressed to at least one button depressed while hovering.

For mouse (or other multi-button pointer devices), this means {{GlobalEventHandlers/pointerdown}} and {{GlobalEventHandlers/pointerup}} are not fired for all of the same circumstances as mousedown and mouseup. See chorded buttons for more information.

For input devices that do not support hover, the user agent MUST also fire a pointer event named {{GlobalEventHandlers/pointerover}} followed by a pointer event named {{GlobalEventHandlers/pointerenter}} prior to dispatching the {{GlobalEventHandlers/pointerdown}} event.

Authors can prevent the firing of certain compatibility mouse events by canceling the {{GlobalEventHandlers/pointerdown}} event (if the isPrimary property is true). This sets the PREVENT MOUSE EVENT flag on the pointer. Note, however, that this does not prevent the mouseover, mouseenter, mouseout, or mouseleave events from firing.

The pointermove event

The user agent MUST fire a pointer event named {{GlobalEventHandlers/pointermove}} when a pointer changes any properties that don't fire {{GlobalEventHandlers/pointerdown}} or {{GlobalEventHandlers/pointerup}} events. This includes any changes to coordinates, pressure, tangential pressure, tilt, twist, contact geometry (i.e. width and height) or chorded buttons.

User agents MAY delay dispatch of the {{GlobalEventHandlers/pointermove}} event (for instance, for performance reasons). The coalesced events information will be exposed via the getCoalescedEvents() method for the single dispatched {{GlobalEventHandlers/pointermove}} event. The final coordinates of such events should be used for finding the target of the event.

The pointerrawupdate event

The user agent MUST fire a pointer event named {{GlobalEventHandlers/pointerrawupdate}}, and only do so within a [=secure context=], when a pointer changes any properties that don't fire pointerdown or pointerup events. See pointermove event for a list of such properties.

In contrast with {{GlobalEventHandlers/pointermove}}, user agents SHOULD dispatch {{GlobalEventHandlers/pointerrawupdate}} events as soon as possible and as frequently as the JavaScript can handle the events.

The target of {{GlobalEventHandlers/pointerrawupdate}} events might be different from the {{GlobalEventHandlers/pointermove}} events due to the fact that {{GlobalEventHandlers/pointermove}} events might get delayed or coalesced, and the final position of the event which is used for finding the target could be different from its coalesced events.

Note that if there is already another {{GlobalEventHandlers/pointerrawupdate}} with the same pointerId that hasn't been dispatched in the [=event loop=], the user agent MAY coalesce the new {{GlobalEventHandlers/pointerrawupdate}} with that event instead of creating a new [=task=]. This may cause {{GlobalEventHandlers/pointerrawupdate}} to have coalesced events, and they will all be delivered as coalesced events of one {{GlobalEventHandlers/pointerrawupdate}} event as soon as the event is processed in the [=event loop=]. See getCoalescedEvents() for more information.

In terms of ordering of {{GlobalEventHandlers/pointerrawupdate}} and {{GlobalEventHandlers/pointermove}}, if the user agent received an update from the platform that causes both {{GlobalEventHandlers/pointerrawupdate}} and {{GlobalEventHandlers/pointermove}} events, then the user agent MUST dispatch the {{GlobalEventHandlers/pointerrawupdate}} event before the corresponding {{GlobalEventHandlers/pointermove}}.

Other than the target, the concatenation of coalesced events lists of all dispatched {{GlobalEventHandlers/pointerrawupdate}} events since the last {{GlobalEventHandlers/pointermove}} event is the same as the coalesced events of the next {{GlobalEventHandlers/pointermove}} event in terms of the other event attributes. The attributes of {{GlobalEventHandlers/pointerrawupdate}} are mostly the same as {{GlobalEventHandlers/pointermove}}, with the exception of cancelable which MUST be false for {{GlobalEventHandlers/pointerrawupdate}}.

User agents SHOULD not fire compatibility mouse events for {{GlobalEventHandlers/pointerrawupdate}}.

Adding listeners for the {{GlobalEventHandlers/pointerrawupdate}} event might negatively impact the performance of the web page, depending on the implementation of the user agent. For most use cases the other pointerevent types should suffice. A {{GlobalEventHandlers/pointerrawupdate}} listener should only be added if JavaScript needs high frequency events and can handle them just as fast. In these cases, there is probably no need to listen to other types of pointer events.

The pointerup event

The user agent MUST fire a pointer event named {{GlobalEventHandlers/pointerup}} when a pointer leaves the active buttons state. For mouse, this is when the device transitions from at least one button depressed to no buttons depressed. For touch, this is when physical contact is removed from the digitizer. For pen, this is when the pen is removed from the physical contact with the digitizer while no button is depressed, or transitions from at least one button depressed to no buttons depressed while hovering.

For input devices that do not support hover, the user agent MUST also fire a pointer event named {{GlobalEventHandlers/pointerout}} followed by a pointer event named {{GlobalEventHandlers/pointerleave}} after dispatching the {{GlobalEventHandlers/pointerup}} event.

All {{GlobalEventHandlers/pointerup}} events have a pressure value of 0.

The user agent MUST also implicitly release the pointer capture if the pointer is currently captured.

The pointercancel event

The user agent MUST fire a pointer event named {{GlobalEventHandlers/pointercancel}} when it detects a scenario to suppress a pointer event stream.

The values of the following properties of the {{GlobalEventHandlers/pointercancel}} event MUST match the values of the last dispatched pointer event with the same pointerId: width, height, pressure, tangentialPressure, tiltX, tiltY, twist, altitudeAngle, azimuthAngle, pointerType, isPrimary, and the coordinates inherited from [[UIEVENTS]]. The coalescedEvents and predictedEvents lists in the {{GlobalEventHandlers/pointercancel}} event MUST be empty, and the event's {{Event/cancelable}} attribute MUST be false.

The pointerout event

The user agent MUST fire a pointer event named {{GlobalEventHandlers/pointerout}} when any of the following occurs:

A pointing device is moved out of the hit test boundaries of an element. Note that setPointerCapture() or releasePointerCapture() might have changed the hit test target. Also note that while a pointer is captured it is considered to be always inside the boundaries of the capturing element for the purpose of firing boundary events.
A layout change causes the hit test boundaries of an element to move away from underneath an uncaptured primary pointer of a stationary pointing device.
After the user agent fires a {{GlobalEventHandlers/pointerup}} event for a device that does not support hover (see {{GlobalEventHandlers/pointerup}}).
The user agent has detected a scenario to suppress a pointer event stream.

The pointerleave event

The user agent MUST fire a pointer event named {{GlobalEventHandlers/pointerleave}} when any of the following occurs:

A pointing device is moved out of the hit test boundaries of an element and all of its descendants. Note that setPointerCapture() or releasePointerCapture() might have changed the hit test target. Also note that while a pointer is captured it is considered to be always inside the boundaries of the capturing element for the purpose of firing boundary events.
A layout change causes the hit test boundaries of an element and all of its descendants to move away from underneath an uncaptured primary pointer of a stationary pointing device.
After the user agent fires the {{GlobalEventHandlers/pointerup}} event for a device that does not support hover (see {{GlobalEventHandlers/pointerup}}).
The user agent has detected a scenario to suppress a pointer event stream.

This event type is similar to {{GlobalEventHandlers/pointerout}} but with two differences: {{GlobalEventHandlers/pointerleave}} does not bubble, and its dispatch considers the hit test boundaries of even descendant elements.

There are similarities between this event type, the mouseleave event described in [[UIEVENTS]], and the CSS :hover pseudo-class described in [[CSS21]]. See also the {{GlobalEventHandlers/pointerenter}} event.

The gotpointercapture event

The user agent MUST fire a pointer event named {{GlobalEventHandlers/gotpointercapture}} when an element receives pointer capture. This event is fired at the element that is receiving pointer capture. Subsequent events for that pointer will be fired at this element. See the setting pointer capture and process pending pointer capture sections.

The lostpointercapture event

The user agent MUST fire a pointer event named {{GlobalEventHandlers/lostpointercapture}} after pointer capture is released for a pointer. This event MUST be fired prior to any subsequent events for the pointer after capture was released. This event is fired at the element from which pointer capture was removed. All subsequent events for the pointer except click, auxclick, and contextmenu events follow normal hit testing mechanisms (out of scope for this specification) for determining the event target. See the releasing pointer capture, implicit release of pointer capture, and process pending pointer capture sections.

The `click`, `auxclick`, and `contextmenu` events

This section is an addition to click, auxclick and contextmenu events defined in [[UIEVENTS]]. These events are typically tied to user interface activation, and are fired even from non-pointer input devices, such as keyboards.

These events MUST be of type PointerEvent, and are subject to the additional requirements mentioned in the rest of this section.

Event attributes

For these events, all PointerEvent specific attributes (defined in this spec) other than pointerId and pointerType MUST have their default values. In addition:

If the events are generated by a pointing device, their pointerId and pointerType MUST be the same as the PointerEvents that caused these events.
If the events are generated by a non-pointing device (such as voice recognition software or a keyboard interaction), pointerId MUST be -1 and pointerType MUST be an empty string.

Event coordinates

As noted in {{PointerEvent}}, [[[CSSOM-VIEW]]] proposes to redefine the various coordinate properties (screenX, screenY, pageX, pageY, clientX, clientY, x, y, offsetX, offsetY) as double, to allow for fractional coordinates. However, this change — when applied only to {{PointerEvent}}, but not to regular {{MouseEvent}} — has proven to lead to web compatibility issues with legacy code in the case of click, auxclick, and contextmenu. For this reason, user agents that have implemented the proposed change in [[[CSSOM-VIEW]]] only for {{PointerEvent}} MUST convert the various coordinate properties for the click, auxclick, and contextmenu to long values (as defined in the original [[[UIEVENTS]]]) using Math.floor [[ECMASCRIPT]].

Event dispatch

A click, auxclick or contextmenu event MUST follow the dispatch process defined in the [[UIEVENTS]] spec except that the event target is overridden using the algorithm below:

Let |event| be the click, auxclick or contextmenu event being dispatched, and |userEvent| be the user interaction event that caused the firing of |event|.

Event |userEvent| could be a non-PointerEvent; for example, it is a KeyboardEvent when a click event dispatch is caused by hitting the spacebar on a checkbox element.

When |userEvent| is a PointerEvent, |userEvent| is a {{GlobalEventHandlers/pointerup}} for a click or auxclick event, and either a {{GlobalEventHandlers/pointerdown}} or a {{GlobalEventHandlers/pointerup}} event (depending on native platform convention) for a contextmenu event.
If |userEvent| is not a PointerEvent, dispatch |event| following the [[UIEVENTS]] spec without overriding |event| target and skip the remaining steps below.
Define |target| as follows:

If |event| is a contextmenu event, or |userEvent| was dispatched while the corresponding pointer was captured, then let |target| be the target of |userEvent|.

Otherwise (|event| is a click or auxclick event for which |userEvent| is a pointerup event that was dispatched uncaptured) let |target| be the nearest common inclusive ancestor of the corresponding pointerdown and pointerup targets in the DOM at the moment |event| is being dispatched.
Dispatch |event| to |target| following the [[UIEVENTS]] spec.

If |userEvent| was captured, |event| is dispatched to the capturing target of |userEvent| even though the {{GlobalEventHandlers/lostpointercapture}} event with the same pointerId has been dispatched already.

Declaring direct manipulation behavior

As noted in Attributes and Default Actions, viewport manipulations (panning and zooming) cannot be suppressed by canceling a pointer event. Instead, authors must declaratively define which of these behaviors they want to allow, and which they want to suppress, using the touch-action CSS property.

While the issue of pointers used to manipulate the viewport is generally limited to touch input (where a user's finger can both interact with content and pan/zoom the page), certain user agents may also allow the same types of (direct or indirect) manipulation for other pointer types. For instance, on mobile/tablet devices, users may also be able to scroll using a stylus. While, for historical reasons, the touch-action CSS property defined in this specification appears to refer only to touch inputs, it does in fact apply to all forms of pointer inputs that allow direct manipulation for panning and zooming.

The `touch-action` CSS property

Name:	`touch-action`
Value:	`auto` \| `none` \| [ [ `pan-x` \| `pan-left` \| `pan-right` ] \|\| [ `pan-y` \| `pan-up` \| `pan-down` ] ] \| `manipulation`
Initial:	`auto`
Applies to:	all elements except: non-replaced inline elements, table rows, row groups, table columns, and column groups.
Inherited:	no
Percentages:	N/A
Media:	visual
Computed value:	Same as specified value.

The touch-action CSS property determines whether direct manipulation interactions (which are not limited to touch, despite the property's name) MAY trigger the user agent's panning and zooming behavior. See the section on touch-action values.

Right before starting to pan or zoom, the user agent MUST suppress a pointer event stream if all of the following conditions are true:

The user agent has determined (via methods out of scope for this specification) that a direct manipulation interaction is to be consumed for panning or zooming,
a {{GlobalEventHandlers/pointerdown}} event has been sent for the pointer, and
a {{GlobalEventHandlers/pointerup}} or {{GlobalEventHandlers/pointercancel}} event (following the above mentioned {{GlobalEventHandlers/pointerdown}}) has not yet been sent for the pointer.

Some user agents implement complex gestures for behaviors that involve a series of separate discrete gestures, but which are all treated as part of a single continuous gesture. For example, consider a "fling to scroll" gesture on a touchscreen: a user starts panning the document with a rapid finger movement, lifts the finger from the touchscreen, and the document continues panning with simulated inertia. While the document is still moving, the user may place their finger on the touchscreen and execute another "fling" to provide further momentum for the panning, or counteract the current panning to slow it down, stop panning altogether, or reverse the direction of the panning. As this specification does not normatively define how gestures and behaviors are implemented, it is left up to the user agent to decide whether or not the second touch (before it is interpreted as a second "fling" or counteraction of the current panning) fires pointer events or not.

touch-action does not apply/cascade through to embedded browsing contexts. For instance, even applying touch-action to an <iframe> won't have any effect on the behavior of direct manipulation interactions for panning and zooming within the <iframe> itself.

Determining supported direct manipulation behavior

When a user interacts with an element using a direct manipulation pointer (such as touch or stylus on a touchscreen), the effect of that input is determined by the value of the touch-action property, and the default direct manipulation behaviors of the element and its ancestors, as follows:

A direct manipulation interaction for panning and zooming conforms to an element's touch-action if the behavior is allowed in the coordinate space of the element. Note that if CSS transforms have been applied, the element's coordinate space may differ from the screen coordinate in a way that affects the conformity here; for example, the X axis of an element rotated by 90 degrees with respect to the screen will be parallel to the Y-axis of the screen coordinate.
A direct manipulation interaction for panning is supported if it conforms to the touch-action property of each element between the hit tested element and its nearest inclusive ancestor that is a [=scroll container=] (as defined in [[CSS-OVERFLOW-3]]).
A direct manipulation interaction for zooming is supported if it conforms to the touch-action property of each element between the hit tested element and the document element of the [=top-level browsing context=] (as defined in [[HTML]]).
Once panning or zooming has been started, and the user agent has already determined whether or not the gesture should be handled as a user agent direct manipulation behavior, any changes to the relevant touch-action value will be ignored for the duration of the action. For instance, programmatically changing the touch-action value for an element from auto to none as part of a {{GlobalEventHandlers/pointerdown}} handler script will not result in the user agent aborting or suppressing any of the pan or zoom behavior for that input for as long as that pointer is active.
Similarly, in the case of the various touch-action values of pan-*, once the user agent has determined whether to handle a gesture directly or not at the start of the gesture, a subsequent change in the direction of the same gesture SHOULD be ignored by the user agent for as long as that pointer is active. For instance, if an element has been set to touch-action: pan-y (meaning that only vertical panning is handled by the user agent), and a touch gesture starts off horizontally, no vertical panning should occur if the user changes the direction of their gesture to be vertical while their finger is still touching the screen.

Some user agents support panning and zooming interactions involving multiple concurrent pointers (e.g. multi-touch). Methods for processing or associating the touch-action values of multiple concurrent pointers is out of scope for this specification.

Details of `touch-action` values

The touch-action property covers direct manipulation behaviors related to viewport panning and zooming. Any additional user agent behaviors, such as text selection/highlighting, or activating links and form controls, MUST NOT be affected by this CSS property.

The terms "panning" and "scrolling" are considered synonymous (or, more aptly, "panning" is "scrolling" using a direct manipulation input). Defining an interaction or gesture for triggering panning/scrolling, or for triggering behavior for the auto or none values, are out of scope for this specification.

auto: The user agent MAY consider any permitted direct manipulation behaviors related to panning and zooming of the viewport that begin on the element.
none: Direct manipulation interactions that begin on the element MUST NOT trigger behaviors related to viewport panning and zooming.
pan-x pan-left pan-right pan-y pan-up pan-down: The user agent MAY consider direct manipulation interactions that begin on the element only for the purposes of panning that starts in any of the directions specified by all of the listed values. Once panning has started, the direction may be reversed by the user even if panning that starts in the reversed direction is disallowed. In contrast, when panning is restricted to a single axis (eg. pan-y), the axis cannot be changed during panning.
manipulation: The user agent MAY consider direct manipulation interactions that begin on the element only for the purposes of panning and continuous zooming (such as pinch-zoom), but MUST NOT trigger other related behaviors that rely on multiple activations that must happen within a set period of time (such as double-tap to zoom, or double-tap and hold for single-finger zoom).

Additional touch-action values common in implementations are defined in [[COMPAT]].

The touch-action property only applies to elements that support both the CSS width and height properties (see [[CSS21]]). This restriction is designed to facilitate user agent optimizations for low-latency direct manipulation panning and zooming. For elements not supported by default, such as <span> which is a non-replaced inline element, authors can set the display CSS property to a value, such as block, that supports width and height. Future specifications could extend this API to all elements.

The direction-specific pan values are useful for customizing some overscroll behaviors. For example, to implement a simple pull-to-refresh effect the document's touch-action can be set to pan-x pan-down whenever the scroll position is 0 and pan-x pan-y otherwise. This allows pointer event handlers to define the behavior for upward panning/scrolling that start from the top of the document.

The direction-specific pan values can also be used for composing a component that implements custom panning with pointer event handling within an element that scrolls natively (or vice-versa). For example, an image carousel may use pan-y to ensure it receives pointer events for any horizontal pan operations without interfering with vertical panning of the document. When the carousel reaches its right-most extent, it may change its touch-action to pan-y pan-right so that a subsequent scroll operation beyond its extent can scroll the document within the viewport if possible. It's not possible to change the behavior of a panning/scrolling operation while it is taking place.

Disabling some default direct manipulation behaviors for panning and zooming may allow user agents to respond to other behaviors more quickly. For example, with auto user agents typically add 300ms of delay before click to allow for double-tap gestures to be handled. In these cases, explicitly setting touch-action: none or touch-action: manipulation will remove this delay. Note that the methods for determining a tap or double-tap gesture are out of scope for this specification.

<div style="touch-action: none;">
    This element receives pointer events for all direct manipulation interactions that otherwise lead to panning or zooming.
</div>

<div style="touch-action: pan-x;">
    This element receives pointer events when not panning in the horizontal direction.
</div>

<div style="overflow: auto;">
    <div style="touch-action: none;">
        This element receives pointer events for all direct manipulation interactions that otherwise lead to panning or zooming.
    </div>
    <div>
        Direct manipulation interactions on this element MAY be consumed for manipulating the parent.
    </div>
</div>

<div style="overflow: auto;">
    <div style="touch-action: pan-y;">
        <div style="touch-action: pan-x;">
            This element receives pointer events for all direct manipulation interactions because
            it allows only horizontal panning yet an intermediate ancestor
            (between it and the scrollable element) only allows vertical panning.
            Therefore, no direct manipulation behaviors for panning/zooming are
            handled by the user agent.
        </div>
    </div>
</div>

<div style="overflow: auto;">
    <div style="touch-action: pan-y pan-left;">
        <div style="touch-action: pan-x;">
            This element receives pointer events when not panning to the left.
        </div>
    </div>
</div>

Coalesced and predicted events

This specification does not define how user agents should coalesce or predict pointer movement data. It only specifies the API for accessing this information.

Coalesced events

For performance reasons, user agents may choose not to send a {{GlobalEventHandlers/pointermove}} event every time a measurable property (such as coordinates, pressure, tangential pressure, tilt, twist, or contact geometry) of a pointer is updated. Instead, they may coalesce (combine/merge) multiple changes into a single {{GlobalEventHandlers/pointermove}} or {{GlobalEventHandlers/pointerrawupdate}} event. While this approach helps in reducing the amount of event handling the user agent must perform, it will naturally reduce the granularity and fidelity when tracking a pointer position, particularly for fast and large movements. Using the getCoalescedEvents() method it is possible for applications to access the raw, un-coalesced position changes. These allow for a more precise handling of pointer movement data. In the case of drawing applications, for instance, the un-coalesced events can be used to draw smoother curves that more closely match the actual movement of a pointer.

Close-up view of a curve, showing coalesced and un-coalesced points — Example of a curve in a drawing application — using only the coalesced coordinates from {{GlobalEventHandlers/pointermove}} events (the grey dots), the curve is noticeably angular and jagged; the same line drawn using the more granular points provided by `getCoalescedEvents()` (the red circles) results in a smoother approximation of the pointer movement.

A PointerEvent has an associated coalesced events list (a list of zero or more PointerEvents). For trusted {{GlobalEventHandlers/pointermove}} and {{GlobalEventHandlers/pointerrawupdate}} events, the list is a sequence of all PointerEvents that were coalesced into this event. The "parent" trusted {{GlobalEventHandlers/pointermove}} and {{GlobalEventHandlers/pointerrawupdate}} event represents an accumulation of these coalesced events, but may have additional processing (for example to align with the display refresh rate). As a result, the coalesced events lists for these events always contain at least one event. For all other trusted event types, it is an empty list. Untrusted events have their coalesced events list initialized to the value passed to the constructor.

Since a trusted parent event is a summary or aggregation of the coalesced events, developers should only need to process either the parent events or all of the coalesced events, but not both.

When a trusted event containing a coalesced events list is re-dispatched from JavaScript, the event dispatch algorithm sets the {{Event/isTrusted}} bit of the event to false but the same bits in the coalesced events list remain unchaged from their original true values.

The events in the coalesced events list of a trusted event will have:

Monotonically increasing {{Event/timeStamp}} values [[DOM]] — all coalesced events have a {{Event/timeStamp}} that is smaller than or equal to the {{Event/timeStamp}} of the dispatched pointer event that the getCoalescedEvents() method was called on. The coalesced events list MUST be chronologically sorted by {{Event/timeStamp}}, so the first event will have the smallest {{Event/timeStamp}}.
The same pointerId, pointerType, and isPrimary as the dispatched "parent" pointer event.
Empty coalesced events list and predicted events list of their own.

<style>
    /* Disable intrinsic user agent direct manipulation behaviors (such as panning or zooming)
    so that all events on the canvas element are given to the application instead. */

    canvas { touch-action: none; }
</style>

<canvas id="drawSurface" width="500px" height="500px" style="border:1px solid black;"></canvas>

<script>
    const canvas = document.getElementById("drawSurface"),
    context = canvas.getContext("2d");

    canvas.addEventListener("pointermove", (e)=> {

        if (e.getCoalescedEvents) {
            for (let coalesced_event of e.getCoalescedEvents()) {
                paint(coalesced_event); // Paint all raw/non-coalesced points
            }
        } else {
            paint(e); // Paint the final coalesced point
        }
    });

    function paint(event) {
        if (event.buttons>0) {
            context.fillRect(event.clientX, event.clientY, 5, 5);
        }
    }

</script>

The PointerEvent's attributes will be initialized in a way that best represents the events in the coalesced events list. The specific method by which user agents should do this is not covered by this specification.

The order of all these dispatched events MUST match the actual order of the original events. For example if a {{GlobalEventHandlers/pointerdown}} event causes the dispatch for the coalesced {{GlobalEventHandlers/pointermove}} events the user agent MUST first dispatch one {{GlobalEventHandlers/pointermove}} event with all those coalesced events of a pointerId followed by the {{GlobalEventHandlers/pointerdown}} event.

Here is an example of the actual events happening with increasing {{Event/timeStamp}} values and the events dispatched by the user agent:

Actual events	Dispatched events
pointer (`pointerId`=2) coordinate change	{{GlobalEventHandlers/pointerrawupdate}} (`pointerId`=2) w/ one coalesced event
pointer (`pointerId`=1) coordinate change	{{GlobalEventHandlers/pointerrawupdate}} (`pointerId`=1) w/ one coalesced event
pointer (`pointerId`=2) coordinate change	{{GlobalEventHandlers/pointerrawupdate}} (`pointerId`=2) w/ one coalesced event
pointer (`pointerId`=2) coordinate change	{{GlobalEventHandlers/pointerrawupdate}} (`pointerId`=2) w/ one coalesced event
pointer (`pointerId`=1) coordinate change	{{GlobalEventHandlers/pointerrawupdate}} (`pointerId`=1) w/ one coalesced event
pointer (`pointerId`=2) coordinate change	{{GlobalEventHandlers/pointerrawupdate}} (`pointerId`=2) w/ one coalesced event
pointer (`pointerId`=1) button press	{{GlobalEventHandlers/pointermove}} (`pointerId`=1) w/ two coalesced events {{GlobalEventHandlers/pointermove}} (`pointerId`=2) w/ four coalesced events {{GlobalEventHandlers/pointerdown}} (`pointerId`=1) w/ zero coalesced events
pointer (`pointerId`=2) coordinate change	{{GlobalEventHandlers/pointerrawupdate}} (`pointerId`=2) w/ one coalesced event
pointer (`pointerId`=2) coordinate change	{{GlobalEventHandlers/pointerrawupdate}} (`pointerId`=2) w/ one coalesced event
pointer (`pointerId`=1) button release	{{GlobalEventHandlers/pointermove}} (`pointerId`=2) w/ two coalesced events {{GlobalEventHandlers/pointerup}} (`pointerId`=1) w/ zero coalesced events

Predicted events

Some user agents have built-in algorithms which, after a series of confirmed pointer movements, can make a prediction (based on the preceding events for the current gesture, and the speed/trajectory of the movement) what the position of future pointer movements may be. Applications can use this information with the getPredictedEvents() method to speculatively "draw ahead" to a predicted position to reduce perceived latency, and then discarding these predicted points once the actual points are received.

A line drawn using coalesced points, showing predicted future points — Example of a line in a drawing application (the result of a drawing gesture from the bottom left to the top right), using the coalesced coordinates from {{GlobalEventHandlers/pointermove}} events, showing the user agent's predicted future points (the grey circles).

A PointerEvent has an associated predicted events list (a list of zero or more PointerEvents). For trusted {{GlobalEventHandlers/pointermove}} events, it is a sequence of PointerEvents that the user agent predicts will follow the event in the future. For all other trusted event types, it is an empty list. Untrusted events have their predicted events list initialized to the value passed to the constructor.

While pointerrawupdate events may have a non-empty coalesced events list, their predicted events list will, for performance reasons, usually be an empty list.

When a trusted event containing a predicted events list is re-dispatched from JavaScript, the event dispatch algorithm sets the {{Event/isTrusted}} bit of the event to false but the same bits in the predicted events list remain unchaged from their original true values.

The number of events in the list and how far they are from the current timestamp are determined by the user agent and the prediction algorithm it uses.

The events in the predicted events list of a trusted event will have:

Monotonically increasing {{Event/timeStamp}} values [[DOM]] — all predicted events have a {{Event/timeStamp}} that is greater than or equal to the {{Event/timeStamp}} of the dispatched pointer event that the getPredictedEvents() method was called on. The predicted events list MUST be chronologically sorted by {{Event/timeStamp}}, so the first event will have the smallest {{Event/timeStamp}}.
The same pointerId, pointerType, and isPrimary as the dispatched "parent" pointer event.
Empty coalesced events list and predicted events list of their own.

Note that authors should only consider predicted events as valid predictions until the next pointer event is dispatched. It is possible, depending on how far into the future the user agent predicts events, that regular pointer events are dispatched earlier than the timestamp of one or more of the predicted events.


let predicted_points = [];
window.addEventListener("pointermove", function(event) {
    // Clear the previously drawn predicted points.
    for (let e of predicted_points.reverse()) {
        clearPoint(e.pageX, e.pageY);
    }

    // Draw the actual movements that happened since the last received event.
    for (let e of event.getCoalescedEvents()) {
        drawPoint(e.pageX, e.pageY);
    }

    // Draw the current predicted points to reduce the perception of latency.
    predicted_points = event.getPredictedEvents();
    for (let e of predicted_points) {
        drawPoint(e.pageX, e.pageY);
    }
});

Populating and maintaining the coalesced and predicted events lists

When a trusted PointerEvent is created, user agents SHOULD run the following steps for each event in the coalesced events list and predicted events list:

Set the event's pointerId, pointerType, isPrimary and {{Event/isTrusted}} to match the respective properties of the "parent" pointer event.
Set the event's {{Event/cancelable}} and {{Event/bubbles}} to false (as these events will never be dispatched in isolation).
Set the event's coalesced events list and predicted events list to an empty list.
Initialize all other attributes to default {{PointerEvent}} values.

When a trusted PointerEvent's {{Event/target}} is changed, user agents SHOULD, for each event in the coalesced events list and predicted events list:

Set the event's {{Event/target}} to match the {{Event/target}} of the "parent" pointer event.

Compatibility mapping with mouse events

The vast majority of web content existing today codes only to Mouse Events. The following describes an algorithm for how the user agent MAY map generic pointer input to mouse events for compatibility with this content.

The compatibility mapping with mouse events is an OPTIONAL feature of this specification. User agents are encouraged to support the feature for best compatibility with existing legacy content.

At a high level, compatibility mouse events are intended to be "interleaved" with their respective pointer events. However, this specific order is not mandatory, and user agents that implement compatibility mouse events MAY decide to delay or group the dispatch of mouse events, as long as their relative order is consistent.

Particularly in the case of touchscreen inputs, user agents MAY apply additional heuristics for gesture recognition (unless explicitly suppressed by authors through touch-action). During a sequence of events between a {{GlobalEventHandlers/pointerdown}} event and a {{GlobalEventHandlers/pointerup}} event, the gesture recognition may have to wait until the {{GlobalEventHandlers/pointerup}} event to detect or ignore a gesture. As a result the compatibility mouse events for the whole sequence may be dispatched together after the last {{GlobalEventHandlers/pointerup}} event, if the user agent determined that an interaction was not intended as a particular gesture. These specifics of user agent gesture recognition are not defined in this specification, and they may differ between implementations.

Regardless of their support for compatibility mouse events, the user agents MUST always support the click, auxclick and contextmenu events because these events are of type PointerEvent and are therefore not compatibility mouse events. Calling preventDefault during a pointer event MUST NOT have an effect on whether click, auxclick, or contextmenu are fired or not.

The relative order of some of these high-level events (contextmenu, focus, blur, etc.) with pointer events is undefined and varies between user agents. For example, in some user agents contextmenu will often follow a {{GlobalEventHandlers/pointerup}}, while in others it'll often precede a {{GlobalEventHandlers/pointerup}} or {{GlobalEventHandlers/pointercancel}}, and in some situations it may be fired without any corresponding pointer event (for instance, as a result of a keyboard interaction).

In addition, user agents may apply their own heuristics to determine whether or not a click, auxclick, or contextmenu event should be fired. Some user agents may choose not to fire these events if there are other (non-primary) pointers of the same type, or other primary pointers of a different type. User agents may determine that a particular action was not a "clean" tap, click, or long-press (for instance, if an interaction with a finger on a touch screen includes too much movement while the finger is in contact with the screen) and decide not to fire a click, auxclick, or contextmenu event. These aspects of user agent behavior are not defined in this specification, and they may differ between implementations.

Unless otherwise noted, the target of any mapped mouse event SHOULD be the same target as the respective pointer event unless the target is no longer participating in its ownerDocument's tree. In this case, the mouse event should be fired at the original target's nearest ancestor node (at the time it was removed from the tree) that still participates in its ownerDocument's tree, meaning that a new event path (based on the new target node) is built for the mouse event.

Authors can prevent the production of certain compatibility mouse events by canceling the {{GlobalEventHandlers/pointerdown}} event.

Mouse events can only be prevented when the pointer is down. Hovering pointers (e.g. a mouse with no buttons pressed) cannot have their mouse events prevented.

The mouseover, mouseout, mouseenter, and mouseleave events are never prevented (even if the pointer is down).

Compatibility mouse events can't be prevented when a pointer event {{EventListener}} is set to be {{AddEventListenerOptions/passive}} [[DOM]].

Tracking the effective position of the legacy mouse pointer

While only primary pointers can produce compatibility mouse events, multiple primary pointers can be active simultaneously, each producing its own compatibility mouse events. For compatibility with scripts relying on MouseEvents, the mouse transition events (mouseover, mouseout, mouseenter and mouseleave) SHOULD simulate the movement of a single legacy mouse input. This means that the entry/exit state for every event target is valid, in accordance with [[UIEVENTS]]. Users agents SHOULD guarantee this by maintaining the effective position of the legacy mouse pointer in the document as follows.

Right before firing a {{GlobalEventHandlers/pointerdown}}, {{GlobalEventHandlers/pointerup}} or {{GlobalEventHandlers/pointermove}} event, or a {{GlobalEventHandlers/pointerleave}} event at the window, the user agent SHOULD run the following steps:

Let |T| be the target of the {{GlobalEventHandlers/pointerdown}}, {{GlobalEventHandlers/pointerup}} or {{GlobalEventHandlers/pointermove}} event being dispatched. For the {{GlobalEventHandlers/pointerleave}} event, unset |T|.
If |T| and current effective legacy mouse pointer position are both unset or they are equal, terminate these steps.
Dispatch mouseover, mouseout, mouseenter and mouseleave events as per [[UIEVENTS]] for a mouse moving from the current effective legacy mouse pointer position to |T|. Consider an unset value of either current effective legacy mouse pointer position or |T| as an out-of-window mouse position.
Set effective legacy mouse pointer position to |T|.

The effective position of the legacy mouse pointer models the fact that we cannot always have a direct mapping from pointer transition events (i.e., pointerover, pointerout, pointerenter and pointerleave) to corresponding legacy mouse transition events (i.e., mouseover, mouseout, mouseenter and mouseleave). The following animation illustrates a case where a user agent needs to dispatch more legacy mouse transition events than pointer transition events to be able to reconcile two primary pointers using a single legacy mouse input.

Simultaneous mouse pointer (white cursor) and touch pointer (white "hand" cursor) causing the single legacy mouse input (orange cursor) to move between the two pointers.

In this animation, note the time period between the mouse click and the touch tap. Button 1 receives no pointerout event (because the "real" mouse pointer didn't leave the button rectangle within this period), but Button 1 receives a mouseout event when the effective position of the legacy mouse pointer moves to Button 2 on touch tap. Similarly, in the time period between the touch tap and the moment before the mouse leaves Button 1, Button 1 receives no pointerover event for the same reason, but Button 1 receives a mouseover event when the effective position of the legacy mouse pointer moves back inside Button 1.

Mapping for devices that support hover

Whenever the user agent is to dispatch a pointer event for a device that supports hover, it SHOULD run the following steps:

If the isPrimary property for the pointer event to be dispatched is false then dispatch the pointer event and terminate these steps.
If the pointer event to be dispatched is a {{GlobalEventHandlers/pointerdown}}, {{GlobalEventHandlers/pointerup}} or {{GlobalEventHandlers/pointermove}} event, or a {{GlobalEventHandlers/pointerleave}} event at the window, dispatch compatibility mouse transition events as described in Tracking the effective position of the legacy mouse pointer.
Dispatch the pointer event.
If the pointer event dispatched was {{GlobalEventHandlers/pointerdown}} and the event was canceled, then set the PREVENT MOUSE EVENT flag for this pointerType.
If the PREVENT MOUSE EVENT flag is not set for this pointerType and the pointer event dispatched was:
- {{GlobalEventHandlers/pointerdown}}, then fire a mousedown event.
- {{GlobalEventHandlers/pointermove}}, then fire a mousemove event.
- {{GlobalEventHandlers/pointerup}}, then fire a mouseup event.
- {{GlobalEventHandlers/pointercancel}}, then fire a mouseup event at the window.
If the pointer event dispatched was {{GlobalEventHandlers/pointerup}} or {{GlobalEventHandlers/pointercancel}}, clear the PREVENT MOUSE EVENT flag for this pointerType.

Mapping for devices that do not support hover

Some devices, such as most touchscreens, do not support hovering a coordinate (or set of coordinates) while not in the active state. Much existing content coded to mouse events assumes that a mouse is producing the events and thus certain qualities are generally true:

The input can hover independently of activation (e.g. moving a mouse cursor without any buttons pressed).
The input will likely produce the mousemove event on an element before clicking it.

Hover is sometimes used to toggle the visibility of UI elements in content designed for mouse (e.g. "hover menus"). This content is often incompatible with devices that do not support hover. This specification does not define a mapping or behavior for compatibility with this scenario. It will be considered in a future version of the specification.

This requires that user agents provide a different mapping for these types of input devices. Whenever the user agent is to dispatch a pointer event for a device that does not support hover, it SHOULD run the following steps:

If the isPrimary property for the pointer event to be dispatched is false then dispatch the pointer event and terminate these steps.
If the pointer event to be dispatched is {{GlobalEventHandlers/pointerover}} and the {{GlobalEventHandlers/pointerdown}} event has not yet been dispatched for this pointer, then fire a mousemove event (for compatibility with legacy mouse-specific code).
If the pointer event to be dispatched is a {{GlobalEventHandlers/pointerdown}}, {{GlobalEventHandlers/pointerup}} or {{GlobalEventHandlers/pointermove}} event, or a {{GlobalEventHandlers/pointerleave}} event at the window, dispatch compatibility mouse transition events as described in Tracking the effective position of the legacy mouse pointer.
Dispatch the pointer event.
If the pointer event dispatched was {{GlobalEventHandlers/pointerdown}} and the event was canceled, then set the PREVENT MOUSE EVENT flag for this pointerType.
If the PREVENT MOUSE EVENT flag is not set for this pointerType and the pointer event dispatched was:
- {{GlobalEventHandlers/pointerdown}}, then fire a mousedown event.
- {{GlobalEventHandlers/pointermove}}, then fire a mousemove event.
- {{GlobalEventHandlers/pointerup}}, then fire a mouseup event.
- {{GlobalEventHandlers/pointercancel}}, then fire a mouseup event at the window.
If the pointer event dispatched was {{GlobalEventHandlers/pointerup}} or {{GlobalEventHandlers/pointercancel}}, clear the PREVENT MOUSE EVENT flag for this pointerType.

If the user agent supports both Touch Events (as defined in [[TOUCH-EVENTS]]) and Pointer Events, the user agent MUST NOT generate both the compatibility mouse events as described in this section, and the fallback mouse events outlined in [[TOUCH-EVENTS]].

The activation of an element (click) with a primary pointer that does not support hover (e.g. single finger on a touchscreen) would typically produce the following event sequence:

mousemove
{{GlobalEventHandlers/pointerover}}
pointerenter
mouseover
mouseenter
{{GlobalEventHandlers/pointerdown}}
mousedown
Zero or more {{GlobalEventHandlers/pointermove}} and mousemove events, depending on movement of the pointer
{{GlobalEventHandlers/pointerup}}
mouseup
{{GlobalEventHandlers/pointerout}}
{{GlobalEventHandlers/pointerleave}}
mouseout
mouseleave
click

If, however, the {{GlobalEventHandlers/pointerdown}} event is canceled during this interaction then the sequence of events would be:

mousemove
{{GlobalEventHandlers/pointerover}}
pointerenter
mouseover
mouseenter
{{GlobalEventHandlers/pointerdown}}
Zero or more {{GlobalEventHandlers/pointermove}} events, depending on movement of the pointer
{{GlobalEventHandlers/pointerup}}
{{GlobalEventHandlers/pointerout}}
{{GlobalEventHandlers/pointerleave}}
mouseout
mouseleave
click

Introduction

Examples

Pointer Events and interfaces

PointerEvent interface

Button states

Chorded button interactions

The button property

The buttons property

The primary pointer

Firing events using the PointerEvent interface