ApproachLayoutModifierNode

interface ApproachLayoutModifierNode : LayoutModifierNode

ApproachLayoutModifierNode is designed to support gradually approaching the destination layout calculated in the lookahead pass. This can be particularly helpful when the destination layout is anticipated to change drastically and would consequently result in visual disruptions.

In order to create a smooth approach, an interpolation (often through animations) can be used in approachMeasure to interpolate the measurement or placement from a previously recorded size and/or position to the destination/target size and/or position. The destination size is available in ApproachMeasureScope as ApproachMeasureScope.lookaheadSize. And the target position can also be acquired in ApproachMeasureScope during placement by using LookaheadScope.localLookaheadPositionOf with the layout's Placeable.PlacementScope.coordinates. The sample code below illustrates how that can be achieved.

During the lookahead pass, measure will be invoked. By default measure simply passes the incoming constraints to its child, and returns the child measure result to parent without any modification. The default behavior for measure is simply a pass through of constraints and measure results without modification. This can be overridden as needed. approachMeasure will be invoked during the approach pass after lookahead.

isMeasurementApproachInProgress signals whether the measurement is in progress of approaching destination size. It will be queried after the destination has been determined by the lookahead pass, before approachMeasure is invoked. The lookahead size is provided to isMeasurementApproachInProgress for convenience in deciding whether the destination size has been reached.

isPlacementApproachInProgress indicates whether the position is actively approaching destination defined by the lookahead, hence it's a signal to the system for whether additional approach placements are necessary. isPlacementApproachInProgress will be invoked after the destination position has been determined by lookahead pass, and before the placement phase in approachMeasure.

IMPORTANT: When both isMeasurementApproachInProgress and isPlacementApproachInProgress become false, the approach is considered complete. Approach pass will subsequently snap the measurement and placement to lookahead measurement and placement. Once approach is complete, approachMeasure may never be invoked until either isMeasurementApproachInProgress or isPlacementApproachInProgress becomes true again. Therefore it is important to ensure approachMeasure and measure result in the same measurement and placement when the approach is complete. Otherwise, there may be visual discontinuity when we snap the measurement and placement to lookahead.

It is important to be accurate in isPlacementApproachInProgress and isMeasurementApproachInProgress. A prolonged indication of incomplete approach will prevent the system from potentially skipping approach pass when possible.

Functions

fun isMeasurementApproachInProgress(lookaheadSize: IntSize): Boolean

isMeasurementApproachInProgress signals whether the measurement is currently approaching destination size. It will be queried after the destination has been determined by the lookahead pass, before approachMeasure is invoked. The lookahead size is provided to isMeasurementApproachInProgress for convenience in deciding whether the destination size has been reached.

Note: It is important to be accurate in isPlacementApproachInProgress and isMeasurementApproachInProgress. A prolonged indication of incomplete approach will prevent the system from potentially skipping approach pass when possible.

fun Placeable.PlacementScope.isPlacementApproachInProgress(
        lookaheadCoordinates: LayoutCoordinates
    ): Boolean

isPlacementApproachInProgress indicates whether the position is approaching destination defined by the lookahead, hence it's a signal to the system for whether additional approach placements are necessary. isPlacementApproachInProgress will be invoked after the destination position has been determined by lookahead pass, and before the placement phase in approachMeasure.

Note: It is important to be accurate in isPlacementApproachInProgress and isMeasurementApproachInProgress. A prolonged indication of incomplete approach will prevent the system from potentially skipping approach pass when possible.

By default, isPlacementApproachInProgress returns false.

override fun MeasureScope.measure(
        measurable: Measurable,
        constraints: Constraints,
    ): MeasureResult

fun ApproachMeasureScope.approachMeasure(
        measurable: Measurable,
        constraints: Constraints,
    ): MeasureResult

approachMeasure defines how the measurement and placement of the layout approach the destination size and position. In order to achieve a smooth approach from the current size and position to the destination, an interpolation (often through animations) can be used in approachMeasure to interpolate the measurement or placement from a previously recorded size and position to the destination/target size and position. The destination size is available in ApproachMeasureScope as ApproachMeasureScope.lookaheadSize. And the target position can also be acquired in ApproachMeasureScope during placement by using LookaheadScope.localLookaheadPositionOf with the layout's Placeable.PlacementScope.coordinates. Please see sample code below for how that can be achieved.

Note: approachMeasure is only guaranteed to be invoked when either isMeasurementApproachInProgress or isMeasurementApproachInProgress is true. Otherwise, the system will consider the approach complete (i.e. destination reached) and may skip the approach pass when possible.

fun ApproachIntrinsicMeasureScope.minApproachIntrinsicWidth(
        measurable: IntrinsicMeasurable,
        height: Int,
    ): Int

The function used to calculate minIntrinsicWidth for the approach pass changes.

fun ApproachIntrinsicMeasureScope.minApproachIntrinsicHeight(
        measurable: IntrinsicMeasurable,
        width: Int,
    ): Int

The function used to calculate minIntrinsicHeight for the approach pass changes.

fun ApproachIntrinsicMeasureScope.maxApproachIntrinsicWidth(
        measurable: IntrinsicMeasurable,
        height: Int,
    ): Int

The function used to calculate maxIntrinsicWidth for the approach pass changes.

fun ApproachIntrinsicMeasureScope.maxApproachIntrinsicHeight(
        measurable: IntrinsicMeasurable,
        width: Int,
    ): Int

The function used to calculate maxIntrinsicHeight for the approach pass changes.

Code Examples

LookaheadLayoutCoordinatesSample

@OptIn(ExperimentalAnimatableApi::class)
@Composable
fun LookaheadLayoutCoordinatesSample() {
    /**
     * Creates a custom implementation of ApproachLayoutModifierNode to approach the placement of
     * the layout using an animation.
     */
    class AnimatedPlacementModifierNode(var lookaheadScope: LookaheadScope) :
        ApproachLayoutModifierNode, Modifier.Node() {
        // Creates an offset animation, the target of which will be known during placement.
        val offsetAnimation: DeferredTargetAnimation<IntOffset, AnimationVector2D> =
            DeferredTargetAnimation(IntOffset.VectorConverter)
        override fun isMeasurementApproachInProgress(lookaheadSize: IntSize): Boolean {
            // Since we only animate the placement here, we can consider measurement approach
            // complete.
            return false
        }
        // Returns true when the offset animation is in progress, false otherwise.
        override fun Placeable.PlacementScope.isPlacementApproachInProgress(
            lookaheadCoordinates: LayoutCoordinates
        ): Boolean {
            val target =
                with(lookaheadScope) {
                    lookaheadScopeCoordinates.localLookaheadPositionOf(lookaheadCoordinates).round()
                }
            offsetAnimation.updateTarget(target, coroutineScope)
            return !offsetAnimation.isIdle
        }
        override fun ApproachMeasureScope.approachMeasure(
            measurable: Measurable,
            constraints: Constraints,
        ): MeasureResult {
            val placeable = measurable.measure(constraints)
            return layout(placeable.width, placeable.height) {
                val coordinates = coordinates
                if (coordinates != null) {
                    // Calculates the target offset within the lookaheadScope
                    val target =
                        with(lookaheadScope) {
                            lookaheadScopeCoordinates.localLookaheadPositionOf(coordinates).round()
                        }
                    // Uses the target offset to start an offset animation
                    val animatedOffset = offsetAnimation.updateTarget(target, coroutineScope)
                    // Calculates the *current* offset within the given LookaheadScope
                    val placementOffset =
                        with(lookaheadScope) {
                            lookaheadScopeCoordinates
                                .localPositionOf(coordinates, Offset.Zero)
                                .round()
                        }
                    // Calculates the delta between animated position in scope and current
                    // position in scope, and places the child at the delta offset. This puts
                    // the child layout at the animated position.
                    val (x, y) = animatedOffset - placementOffset
                    placeable.place(x, y)
                } else {
                    placeable.place(0, 0)
                }
            }
        }
    }
    // Creates a custom node element for the AnimatedPlacementModifierNode above.
    data class AnimatePlacementNodeElement(val lookaheadScope: LookaheadScope) :
        ModifierNodeElement<AnimatedPlacementModifierNode>() {
        override fun update(node: AnimatedPlacementModifierNode) {
            node.lookaheadScope = lookaheadScope
        }
        override fun create(): AnimatedPlacementModifierNode {
            return AnimatedPlacementModifierNode(lookaheadScope)
        }
    }
    val colors = listOf(Color(0xffff6f69), Color(0xffffcc5c), Color(0xff264653), Color(0xff2a9d84))
    var isInColumn by remember { mutableStateOf(true) }
    LookaheadScope {
        // Creates movable content containing 4 boxes. They will be put either in a [Row] or in a
        // [Column] depending on the state
        val items = remember {
            movableContentOf {
                colors.forEach { color ->
                    Box(
                        Modifier.padding(15.dp)
                            .size(100.dp, 80.dp)
                            .then(AnimatePlacementNodeElement(this))
                            .background(color, RoundedCornerShape(20))
                    )
                }
            }
        }
        Box(modifier = Modifier.fillMaxSize().clickable { isInColumn = !isInColumn }) {
            // As the items get moved between Column and Row, their positions in LookaheadScope
            // will change. The `animatePlacementInScope` modifier created above will
            // observe that final position change via `localLookaheadPositionOf`, and create
            // a position animation.
            if (isInColumn) {
                Column(Modifier.fillMaxSize()) { items() }
            } else {
                Row { items() }
            }
        }
    }
}