Use the After Effects expression elements along with standard JavaScript elements to write your expressions. You can use the Expression Language menu at any time to insert methods and attributes into an expression, and you can use the pick whip at any time to insert properties.

If an argument description contains an equal sign (=) and a value (such as t=time or width=.2), then the argument uses the included default value if you don’t specify a different value.

Some argument descriptions include a number in square brackets—this number indicates the dimension of the expected property or Array.

Some return-value descriptions include a number in square brackets—this number specifies the dimension of the returned property or Array. If a specific dimension is not included, the dimension of the returned Array depends on the dimension of the input.

The W3Schools JavaScript reference website provides information for the standard JavaScript language, including pages for the JavaScript Math and String objects.

Global objects, attributes, and methods (expression reference)

comp(name)

Return type: Comp.

Argument type: name is a String.

Retrieves another composition by name.

footage(name)

Return type: Footage.

Argument type: name is a String.

Retrieves a footage item by name.

thisComp

Return type: Comp.

Represents the composition containing the expression.

thisLayer

Return type: Layer, Light, or Camera.

Represents the layer containing the expression. Because thisLayer is the default object, its use is optional. For example, you can start an expression with thisLayer.width or width and get the same result.

thisProperty

Return type: Property.

Represents the property containing the expression. For example, if you write an expression on the Rotation property, you can start an expression with thisProperty to refer to the Rotation property.

time

Return type: Number.

Represents the composition time, in seconds, at which the expression is being evaluated.

colorDepth

Return type: Number.

Returns the project color depth value. For example, colorDepth returns 16 when the project color depth is 16 bits per channel.

posterizeTime(framesPerSecond)

Return type: Number.

Argument type: framesPerSecond is a Number.

The framesPerSecond value becomes the frame rate from which the rest of the expression operates. This expression allows you to set the frame rate for a property to be lower than the frame rate of the composition. For example, the following expression updates the property value with a random value once per second:

  posterizeTime(1);   random()

value

Return type: Number, Array, or String.

Represents the value at the current time for the property containing the expression.

Time conversion methods (expression reference)

timeToFrames(t = time + thisComp.displayStartTime, fps = 1.0 / thisComp.frameDuration, isDuration = false)

Return type: Number.

Argument type: t and fps are Numbers; isDuration is a Boolean.

Converts the value of t, which defaults to the current composition time, to an integer number of frames. The number of frames per second is specified in the fps argument, which defaults to the frame rate of the current composition (1.0 / thisComp.frameDuration). The isDuration argument, which defaults to false, should be true if the t value represents a difference between two times instead of an absolute time. Absolute times are rounded down toward negative infinity; durations are rounded away from zero (up for positive values).

framesToTime(frames, fps = 1.0 / thisComp.frameDuration)

Return type: Number.

Argument type: frames and fps are Numbers.

The inverse of timeToFrames. Returns the time corresponding to the frames argument, which is required. It doesn’t have to be an integer. See timeToFrames for explanation of the fps argument.

timeToTimecode(t = time + thisComp.displayStartTime, timecodeBase = 30, isDuration = false)

Return type: String.

Argument type: t and timecodeBase are Numbers; isDuration is a Boolean.

Converts the value of t to a String representing timecode. See timeToFrames for an explanation of the t and isDuration arguments. The timecodeBase value, which defaults to 30, specifies the number of frames in one second.

timeToNTSCTimecode(t = time + thisComp.displayStartTime, ntscDropFrame = false, isDuration = false)

Return type: String.

Argument type: t is a Number, ntscDropFrame and isDuration are Booleans.

Converts t to a String representing NTSC timecode. See timeToFrames for an explanation of the t and isDuration arguments. If ntscDropFrame is false (the default), the result String is NTSC non-drop-frame timecode. If ntscDropFrame is true, the result String is NTSC drop-frame timecode.

timeToFeetAndFrames(t = time + thisComp.displayStartTime, fps = 1.0 / thisComp.frameDuration, framesPerFoot = 16, isDuration = false)

Return type: String.

Argument type: t, fps, and framesPerFoot are Numbers; isDuration is a Boolean.

Converts the value of t to a String representing feet of film and frames. See timeToFrames for an explanation of the t, fps, and isDuration arguments. The framesPerFoot argument specifies the number of frames in one foot of film. It defaults to 16, which is the most common rate for 35mm footage.

timeToCurrentFormat(t = time + thisComp.displayStartTime, fps = 1.0 / thisComp.frameDuration, isDuration = false)

Return type: String.

Argument type: t and fps are Numbers; isDuration is a Boolean.

Converts the value of t to a String representing time in the current Project Settings display format. See timeToFrames for a definition of all the arguments.

An optional ntscDropFrame argument was added to the timeToCurrentFormat() function in After Effects CS5.5 and later. Default: ntscDropFrame = thisComp.ntscDropFrame.

Note:

If you want more control over the look of timecode in your footage, use the timeToCurrentFormat method or other timeTo methods to generate the timecode instead of using the Timecode or Numbers effect. Create a text layer, add an expression to the Source Text property, and enter timeToCurrentFormat() in the expression field. With this method, you can format and animate the timecode text. In addition, the timecode uses the same display style defined by the current project settings.

Vector Math methods (expression reference)

Vector Math functions are global methods that perform operations on arrays, treating them as mathematical vectors. Unlike built-in JavaScript methods, such as Math.sin, these methods are not used with the Math prefix. Unless otherwise specified, Vector Math methods are lenient about dimensions and return a value that is the dimension of the largest input Array object, filling in missing elements with zeros. For example, the expression add([10, 20], [1, 2, 3]) returns [11, 22, 3].

JJ Gifford’s website provides explanations and examples that show how to use simple geometry and trigonometry with expressions.

add(vec1, vec2)

Return type: Array.

Argument type: vec1 and vec2 are Arrays.

Adds two vectors.

sub(vec1, vec2)

Return type: Array.

Argument type: vec1 and vec2 are Arrays.

Subtracts two vectors.

mul(vec, amount)

Return type: Array.

Argument type: vec is an Array, amount is a Number.

Multiplies every element of the vector by the amount.

div(vec, amount)

Return type: Array.

Argument type: vec is an Array, amount is a Number.

Divides every element of the vector by the amount.

clamp(value, limit1, limit2)

Return type: Number or Array.

Argument type: value, limit1, and limit2 are Numbers or Arrays.

The value of each component of value is constrained to fall between the values of the corresponding values of limit1 and limit2.

dot(vec1, vec2)

Return type: Number.

Argument type: vec1 and vec2 are Arrays.

Returns the dot (inner) product of the vector arguments.

cross(vec1, vec2)

Return type: Array [2 or 3].

Argument type: vec1 and vec2 are Arrays [2 or 3].

Returns the vector cross product of vec1 and vec2. Refer to a math reference or JavaScript guide for more information.

normalize(vec)

Return type: Array.

Argument type: vec is an Array.

Normalizes the vector so that its length is 1.0. Using the normalize method is a short way of performing the operation div(vec, length(vec)).

length(vec)

Return type: Number.

Argument type: vec is an Array.

Returns the length of vector vec.

length(point1, point2)

Return type: Number.

Argument type: point1 and point2 are Arrays.

Returns the distance between two points. The point2 argument is optional. For example, length(point1, point2) is the same as length(sub(point1, point2)).

For example, add this expression to the Focus Distance property of a camera to lock the focal plane to the camera's point of interest so that the point of interest is in focus:

  length(position, pointOfInterest)

lookAt(fromPoint, atPoint)

Return type: Array [3].

Argument type: fromPoint and atPoint are Arrays [3].

The argument fromPoint is the location in world space of the layer you want to orient. The argument atPoint is the point in world space you want to point the layer at. The return value can be used as an expression for the Orientation property, making the z-axis of the layer point at atPoint. This method is especially useful for cameras and lights. If you use this expression on a camera, turn off auto-orientation. For example, this expression on the Orientation property of a spot light makes the light point at the anchor point of layer number 1 in the same composition: lookAt(position, thisComp.layer(1).position)

Random Numbers methods (expression reference)

Note:

The wiggle method—which is used to randomly vary a property value—is in the Property attributes and methods category. (See Property attributes and methods (expression reference).)

seedRandom(offset, timeless=false)

Return type: none.

Argument type: offset is a Number, timeless is a Boolean.

The random and gaussRandom methods use a seed value that controls the sequence of numbers. By default, the seed is computed as a function of a unique layer identifier, the property within the layer, the current time, and an offset value of 0. Call seedRandom to set the offset to something other than 0 to create a different random sequence.

Use true for the timeless argument to not use the current time as input to the random seed. Using true for the timeless argument allows you to generate a random number that doesn’t vary depending on the time of evaluation.

The offset value, but not the timeless value, is also used to control the initial value of the wiggle function.

For example, this expression on the Opacity property sets the Opacity value to a random value that does not vary with time:

  seedRandom(123456, true);   random()*100

The multiplication by 100 in this example converts the value in the range 0–1 returned by the random method into a number in the range 0–100; this range is more typically useful for the Opacity property, which has values from 0% to 100%.

random()

Return type: Number.

Returns a random number in the range 0–1.

In After Effects CC and CS6, the behavior of random() is changed to be more random when layer IDs are close together. The wiggle() expression is not affected.

random(maxValOrArray)

Return type: Number or Array.

Argument type: maxValOrArray is a Number or Array.

If maxValOrArray is a Number, this method returns a number in the range from 0 to maxValOrArray. If maxValOrArray is an Array, this method returns an Array with the same dimension as maxValOrArray, with each component ranging from 0 to the corresponding component of maxValOrArray.

random(minValOrArray, maxValOrArray)

Return type: Number or Array.

Argument type: minValOrArray and maxValOrArray are Numbers or Arrays.

If minValOrArray and maxValOrArray are Numbers, this method returns a number in the range from minValOrArray to maxValOrArray. If the arguments are Arrays, this method returns an Array with the same dimension as the argument with the greater dimension, with each component in the range from the corresponding component of minValOrArray to the corresponding component of maxValOrArray. For example, the expression random([100, 200], [300, 400]) returns an Array whose first value is in the range 100–300 and whose second value is in the range 200–400. If the dimensions of the two input Arrays don’t match, higher-dimension values of the shorter Array are filled out with zeros.

gaussRandom()

Return type: Number.

Returns a random number. The results have a Gaussian (bell-shaped) distribution. Approximately 90% of the results are in the range 0–1, and the remaining 10% are outside this range.

gaussRandom(maxValOrArray)

Return type: Number or Array.

Argument type: maxValOrArray is a Number or Array.

When maxValOrArray is a Number, this method returns a random number. Approximately 90% of the results are in the 0 to maxValOrArray range, and the remaining 10% are outside this range. When maxValOrArray is an Array, this method returns an Array of random values, with the same dimension as maxValOrArray. 90% of the values are in the range from 0 to maxValOrArray, and the remaining 10% are outside this range. The results have a Gaussian (bell-shaped) distribution.

gaussRandom(minValOrArray, maxValOrArray)

Return type: Number or Array.

Argument type: minValOrArray and maxValOrArray are Numbers or Arrays.

If minValOrArray and maxValOrArray are Numbers, this method returns a random number. Approximately 90% of the results are in the range from minValOrArray to maxValOrArray, and the remaining 10% are outside this range. If the arguments are Arrays, this method returns an Array of random numbers with the same dimension as the argument with the greater dimension. For each component, approximately 90% of the results are in the range from the corresponding component of minValOrArray to the corresponding component of maxValOrArray, and the remaining 10% are outside this range. The results have a Gaussian (bell-shaped) distribution.

noise(valOrArray)

Return type: Number.

Argument type: valOrArray is a Number or an Array [2 or 3].

Returns a number in the range from -1 to 1. The noise is not actually random; it is based on Perlin noise, which means that the return values for two input values that are near one another tend to be near one another. This type of noise is useful when you want a sequence of seemingly random numbers that don’t vary wildly from one to the other—as is usually the case when animating any apparently random natural motion. Example: rotation + 360*noise(time)

Interpolation methods (expression reference)

For all the Interpolation methods, the argument t is often time or value, though it can have other values, instead. If t is time, the interpolation between values happens over a duration. If t is value, then the expression maps one range of values to a new range of values.

For additional explanations and examples of the Interpolation methods, see JJ Gifford’s website.

Chris and Trish Meyer provide additional information and examples for these methods in an article on the ProVideo Coalition website.

Ian Haigh provides a script on After Effects Scripts website that you can use to easily apply advanced interpolation method expressions—such as bounces—to properties.

Andrew Devis provides a pair of video tutorials on the Creative COW website that show in detail how to use the linear expression method along with the Convert Audio To Keyframes command.

linear(t, tMin, tMax, value1, value2)

Return type: Number or Array.

Argument type: t, tMin, and tMax are Numbers, and value1 and value2 are Numbers or Arrays.

Returns value1 when t <= tMin. Returns value2 when t >= tMax. Returns a linear interpolation between value1 and value2 when tMin < t < tMax.

For example, this expression on the Opacity property causes Opacity values to ramp linearly from 20% to 80% over the time from 0 seconds to 6 seconds:

  linear(time, 0, 6, 20, 80)

This method—like all the Interpolation methods—can also be used to convert from one range of values to another. For example, this expression on the Opacity property converts the Opacity values from the range 0%-100% to the range 20%-80%:

  linear(value, 0, 100, 20, 80)

linear(t, value1, value2)

Return type: Number or Array.

Argument type: t is a Number, and value1 and value2 are Numbers or Arrays.

Returns a value that linearly interpolates from value1 to value2 as t ranges from 0 to 1. Returns value1 when t <= 0. Returns value2 when t >= 1.

ease(t, value1, value2)

Return type: Number or Array.

Argument type: t is a Number, and value1 and value2 are Numbers or Arrays.

Similar to linear with the same arguments, except that the interpolation eases in and out so that the velocity is 0 at the start and end points. This method results in a smooth animation.

ease(t, tMin, tMax, value1, value2)

Return type: Number or Array.

Argument type: t, tMin, and tMax are Numbers, and value1 and value2 are Numbers or Arrays.

Similar to linear with the same arguments, except that the interpolation eases in and out so that the velocity is 0 at the start and end points. This method results in a smooth animation.

easeIn(t, value1, value2)

Return type: Number or Array.

Argument type: t is a Number, and value1 and value2 are Numbers or Arrays.

Similar to ease, except that the tangent is 0 only on the value1 side and interpolation is linear on the value2 side.

easeIn(t, tMin, tMax, value1, value2)

Return type: Number or Array.

Argument type: t, tMin, and tMax are Numbers, and value1 and value2 are Numbers or Arrays.

Similar to ease, except that the tangent is 0 only on the tMin side and interpolation is linear on the tMax side.

easeOut(t, value1, value2)

Return type: Number or Array.

Argument type: t is a Number, and value1 and value2 are Numbers or Arrays.

Similar to ease, except that the tangent is 0 only on the value2 side and interpolation is linear on the value1 side.

easeOut(t, tMin, tMax, value1, value2)

Return type: Number or Array.

Argument type: t, tMin, and tMax are Numbers, and value1 and value2 are Numbers or Arrays.

Similar to ease, except that the tangent is 0 only on the tMax side and interpolation is linear on the tMin side.

Color Conversion methods (expression reference)

Harry Frank provides a video tutorial on his graymachine website that shows how to use these color conversion methods to change the color of the waves produced by the Radio Waves effect.

rgbToHsl(rgbaArray)

Return type: Array [4].

Argument type: rgbaArray is an Array [4].

Converts a color in RGBA space to HSLA space. The input is an Array of normalized red, green, blue, and alpha channel values, all in the range of 0.0 to 1.0. The resulting value is an Array of hue, saturation, lightness, and alpha channel values, also in the range of 0.0 to 1.0. Example:

  rgbToHsl.effect("Change Color")("Color To Change")

hslToRgb(hslaArray)

Return type: Array [4].

Argument type: hslaArray is an Array [4].

Converts a color in HSLA space to RGBA space. This conversion is the opposite of the conversion performed by the rgbToHsl method.

Other Math methods (expression reference)

degreesToRadians(degrees)

Return type: Number.

Argument type: degrees is a Number.

Converts degrees to radians.

radiansToDegrees(radians)

Return type: Number.

Argument type: radians is a Number.

Converts radians to degrees.

Comp attributes and methods (expression reference)

layer(index)

Return type: Layer, Light, or Camera.

Argument type: index is a Number.

Retrieves the layer by number (order in the Timeline panel). Example: thisComp.layer(3)

layer(name)

Return type: Layer, Light, or Camera.

Argument type: name is a String.

Retrieves the layer by name. Names are matched according to layer name, or source name if there is no layer name. If duplicate names exist, After Effects uses the first (topmost) one in the Timeline panel. Example:

  thisComp.layer("Solid 1")

layer(otherLayer, relIndex)

Return type: Layer, Light, or Camera.

Argument type: otherLayer is a Layer object, and relIndex is a Number.

Retrieves the layer that is relIndex layers above or below otherLayer. For example, thisComp.layer(thisLayer, 1).active returns true if the next layer down in the Timeline panel is active.

marker

Return type: MarkerProperty.

Note:

You cannot access a composition marker by marker number. If you have a project created in a previous version of After Effects that uses composition marker numbers in expressions, you must change those calls to use marker.key(name) instead. Because the default name of a composition marker is a number, converting the reference to use the name is often just a matter of surrounding the number with quotation marks.

marker.key(index)

Return type: MarkerKey.

Argument type: index is a Number.

Returns the MarkerKey object of the marker with the specified index. The index refers to the order of the marker in composition time, not to the name of the marker. For example, this expression returns the time of the first composition marker:

  thisComp.marker.key(1).time

marker.key(name)

Return type: MarkerKey.

Argument type: name is a String.

Returns the MarkerKey object of the marker with the specified name. The name value is the name of the marker, as typed in the comment field in the marker dialog box, for example, marker.key("1"). For a composition marker, the default name is a number. If more than one marker in the composition has the same name, this method returns the marker that occurs first in time (in composition time). The value for a marker key is a String, not a Number. For example, this expression returns the time of the composition marker with the name "0":

  thisComp.marker.key("0").time

marker.nearestKey(t)

Return type: MarkerKey.

Argument type: t is a Number.

Returns the marker that is nearest in time to t. For example, this expression returns the time of the composition marker nearest to the time of 1 second:

  thisComp.marker.nearestKey(1).time

This expression returns the time of the composition marker nearest to the current time:

  thisComp.marker.nearestKey(time).time

marker.numKeys

Return type: Number.

Returns the total number of composition markers in the composition.

numLayers

Return type: Number.

Returns the number of layers in the composition.

activeCamera

Return type: Camera.

Returns the Camera object for the camera through which the composition is rendered at the current frame. This camera is not necessarily the camera through which you are looking in the Composition panel.

width

Return type: Number.

Returns the composition width, in pixels.

Apply the following expression to the Position property of a layer to center the layer in the composition frame:

  [thisComp.width/2, thisComp.height/2]

height

Return type: Number.

Returns the composition height, in pixels.

duration

Return type: Number.

Returns the composition duration, in seconds.

ntscDropFrame

Return type: Boolean.

Returns true if the timecode is in drop-frame format. (After Effects CS5.5and later.)

displayStartTime

Return type: Number.

Returns the composition start time, in seconds.

frameDuration

Return type: Number.

Returns the duration of a frame, in seconds.

shutterAngle

Return type: Number.

Returns the shutter-angle value of the composition, in degrees.

shutterPhase

Return type: Number.

Returns the shutter phase of the composition, in degrees.

bgColor

Return type: Array [4].

Returns the background color of the composition.

pixelAspect

Return type: Number.

Returns the pixel aspect ratio of the composition.

name

Return type: String.

Returns the name of the composition.

Footage attributes and methods (expression reference)

To use a footage item from the Project panel as an object in an expression, use the global footage method, as in footage("file_name"). You can also access a footage object using the source attribute on a layer whose source is a footage item.

width

Return type: Number.

Returns the width of the footage item, in pixels.

height

Return type: Number.

Returns the height of the footage item, in pixels.

duration

Return type: Number.

Returns the duration of the footage item, in seconds.

frameDuration

Return type: Number.

Returns the duration of a frame in the footage item, in seconds.

ntscDropFrame

Return type: Boolean.

Returns true if the timecode is in drop-frame format. (After Effects CS5.5 and later.)

pixelAspect

Return type: Number.

Returns the pixel aspect ratio of the footage item.

name

Return type: String.

Returns the name of the footage item as shown in the Project panel.

Layer Sub-objects attributes and methods (expression reference)

Note:

For After Effects CC and CS6, the Expression language menu, the "Layer Sub-objects", "Layer General", "Layer Properties", "Layer 3D", and "Layer Space Transforms" have been arranged into a "Layer" submenu.

source

Return type: Comp or Footage.

Returns the source Comp or source Footage object for the layer. Default time is adjusted to the time in the source. Example: source.layer(1).position

sourceTime(t = time)

Return type: Number.

Returns the layer source corresponding to time t. (After Effects CS5.5 and later.)

effect(name)

Return type: Effect.

Argument type: name is a String.

After Effects finds the effect by its name in the Effect Controls panel. The name can be the default name or a user-defined name. If multiple effects have the same name, the effect closest to the top of the Effect Controls panel is used. Example:

  effect("Fast Blur")("Blurriness")

effect(index)

Return type: Effect.

Argument type: index is a Number.

After Effects finds the effect by its index in the Effect Controls panel, starting at 1 and counting from the top.

mask(name)

Return type: Mask.

Argument type: name is a String.

The name can be the default name or a user-defined name. If multiple masks have the same name, the first (topmost) mask is used. Example:

  mask("Mask 1")

mask(index)

Return type: Mask.

Argument type: index is a Number.

After Effects finds the mask by its index in the Timeline panel, starting at 1 and counting from the top.

Layer General attributes and methods (expression reference)

width

Return type: Number.

Returns the width of the layer, in pixels. It is the same as source.width.

height

Return type: Number.

Returns the height of the layer, in pixels. It is the same as source.height.

index

Return type: Number.

Returns the index number of the layer in the composition.

parent

Return type: Layer, Light, or Camera.

Returns the parent Layer object of the layer, if it has one. Example: position[0] + parent.width

hasParent

Return type: Boolean.

Returns true if the layer has a parent or false if it doesn’t.

Use the hasParent attribute to determine if a layer has a parent layer. You can use this attribute even if the layer has no parent layer at present. For example, the following expression indicates that the layer to which you apply it wiggles based on the position of the parent. If the layer has no parent, then it wiggles based on its own position. If the layer is given a parent later, then the behavior of the layer changes accordingly:

  idx = index;   if (hasParent) {     idx = parent.index;   }   thisComp.layer(idx).position.wiggle(5,20)

inPoint

Return type: Number.

Returns the In point of the layer, in seconds.

Note:

In general, the value of outPoint is greater than the value of inPoint. However, if a layer is reversed in time, the value of inPoint is greater than the value of outPoint. Similarly, the value of startTime can be greater than the value of inPoint.

outPoint

Return type: Number.

Returns the Out point of the layer, in seconds.

startTime

Return type: Number.

Returns the start time of the layer, in seconds.

hasVideo

Return type: Boolean.

Returns true if the layer has video, or false if it doesn’t.

hasAudio

Return type: Boolean.

Returns true if the layer has audio or false if it doesn’t.

active

Return type: Boolean.

Returns true if the Video switch is on for the layer and the current time is in the range from the In point of the layer to the Out point of the layer; false otherwise.

enabled

Return type: Boolean.

Returns true if the Video switch is on for the layer; false otherwise.

audioActive

Return type: Boolean.

Returns true if the Audio switch is on for the layer and the current time is in the range from the In point of the layer to the Out point of the layer; false otherwise.

sampleImage(point, radius = [.5, .5], postEffect=true, t=time)

Return type: Array [4].

Argument type: point is an Array [2], radius is an Array [2], postEffect is a Boolean, and t is a Number.

Samples the color and alpha channel values of a layer and returns the average alpha-weighted value of the pixels within the specified distance of the point as an array: [red, green, blue, alpha]. If postEffect is true, the sampled values are for the layer after masks and effects on that layer have been rendered; if postEffect is false, the sampled values are for the layer before masks and effects have been rendered. The input value point is in layer space; the point [0,0] is the center of the upper-left pixel in the layer. The input value radius specifies the horizontal and vertical distance from the sample center to the edges of the sampled rectangle. The default value samples one pixel.

Note:

The postEffect parameter refers to effects applied directly to the layer, not to effects applied indirectly, such as with an adjustment layer.

Note:

Using sampleImage in an expression no longer disables multiprocessing.

This example samples a rectangle 4 pixels wide and 3 pixels high, centered around a point 100 pixels down and to the right of the upper-left corner of the layer:

  thisComp.layer(1).sampleImage([100, 100], [2, 1.5])

Dan Ebberts provides an example of how to use the sampleImage method on his MotionScript website.

Todd Kopriva provides instructions for using the sampleImage method and the Point Control effect to monitor colors for a specified point during color correction on his After Effects Region of Interest blog.

Layer Properties attributes and methods (expression reference)

When you add masks, effects, paint, or text to a layer, After Effects adds new properties to the Timeline panel. There are too many of these properties to list here, so use the pick whip to learn the syntax for referring to them in your expressions.

anchorPoint

Return type: Property [2 or 3].

Returns the anchor point value of the layer in the coordinate system of the layer (layer space).

position

Return type: Property [2 or 3].

Returns the position value of the layer, in world space if the layer has no parent. If the layer has a parent, it returns the position value of the layer in the coordinate system of the parent layer (in the layer space of the parent layer).

scale

Return type: Property [2 or 3].

Returns the scale value of the layer, expressed as a percentage.

rotation

Return type: Property.

Returns the rotation value of the layer in degrees. For a 3D layer, it returns the z rotation value in degrees.

opacity

Return type: Property.

Returns the opacity value for the layer, expressed as a percentage.

audioLevels

Return type: Property [2].

Returns the value of the Audio Levels property of the layer, in decibels. This value is a 2D value; the first value represents the left audio channel, and the second value represents the right. The value is not the amplitude of the audio track of the source material. Instead, it is the value of the Audio Levels property, which may be affected by keyframes.

timeRemap

Return type: Property.

Returns the value of the Time Remap property, in seconds, if Time Remap is enabled.

marker.key(index)

Return type: MarkerKey.

Argument type: index is a Number.

Returns the MarkerKey object of the layer marker with the specified index.

marker.key(name)

Return type: MarkerKey.

Argument type: name is a String.

Returns the MarkerKey object of the layer marker with the specified name. The name value is the name of the marker, as typed in the comment field in the marker dialog box, for example, marker.key("ch1"). If more than one marker on the layer has the same name, this method returns the marker that occurs first in time (in layer time). The value for a marker key is a String, not a Number.

This expression on a property ramps the value of the property from 0 to 100 between two markers identified by name:

  m1 = marker.key("Start").time;   m2 = marker.key("End").time;   linear(time, m1, m2, 0, 100);

marker.nearestKey(t)

Return type: MarkerKey.

Argument type: t is a Number.

Returns the layer marker that is nearest in time to t. For example, this expression returns the time of the marker on the layer nearest to the time of 1 second:

  marker.nearestKey(1).time

This expression returns the time of the marker on the layer nearest to the current time:

  marker.nearestKey(time).time

marker.numKeys

Return type: Number.

Returns the total number of markers on the layer.

name

Return type: String.

Returns the name of the layer.

Layer 3D attributes and methods (expression reference)

orientation

Return type: Property [3].

Returns the 3D orientation value, in degrees, for a 3D layer.

rotationX

Return type: Property.

Returns the x rotation value, in degrees, for a 3D layer.

rotationY

Return type: Property.

Returns the y rotation value, in degrees, for a 3D layer.

rotationZ

Return type: Property.

Returns the z rotation value, in degrees, for a 3D layer.

lightTransmission

Return type: Property.

Returns the value of the Light Transmission property for a 3D layer.

castsShadows

Return type: Property.

Returns a value of 1.0 if the layer casts shadows.

acceptsShadows

Return type: Property.

Returns a value of 1.0 if the layer accepts shadows.

acceptsLights

Return type: Property.

Returns a value of 1.0 if the layer accepts lights.

ambient

Return type: Property.

Returns the ambient component value as a percentage.

diffuse

Return type: Property.

Returns the diffuse component value as a percentage.

specular

Return type: Property.

Returns the specular component value as a percentage.

shininess

Return type: Property.

Returns the shininess component value as a percentage.

metal

Return type: Property.

Returns the metal component value as a percentage.

Layer Space Transforms methods (expression reference)

Use layer space transform methods to transform values from one space to another, such as from layer space to world space. The “from” methods transform values from the named space (composition or world) to the layer space. The “to” methods transform values from the layer space to the named space (composition or world). Each transform method takes an optional argument to determine the time at which the transform is computed; however, you can almost always use the current (default) time.

Use “Vec” transform methods when transforming a direction vector, such as the difference between two position values. Use the plain (non-”Vec”) transform methods when transforming a point, such as position. Composition (comp) and world space are the same for 2D layers. For 3D layers, however, composition space is relative to the active camera, and world space is independent of the camera.

toComp(point, t=time)

Return type: Array [2 or 3].

Argument type: point is an Array [2 or 3], and t is a Number.

Transforms a point from layer space to composition space.

fromComp(point, t=time)

Return type: Array [2 or 3].

Argument type: point is an Array [2 or 3], and t is a Number.

Transforms a point from composition space to layer space. The resulting point in a 3D layer may have a nonzero value even though it is in layer space. Example: fromComp(thisComp.layer(2).position)

toWorld(point, t=time)

Return type: Array [2 or 3].

Argument type: point is an Array [2 or 3], and t is a Number.

Transforms a point from layer space to view-independent world space. Example:

  toWorld.effect("Bulge")("Bulge Center")

Dan Ebberts provides an expression on his MotionScript website that uses the toWorld method to auto-orient a layer along only one axis. This is useful, for example, for having characters turn from side to side to follow the camera while remaining upright.

 

Rich Young provides a set of expressions on his AE Portal website that use the toWorld method link a camera and light to a layer with the CC Sphere effect.

fromWorld(point, t=time)

Return type: Array [2 or 3].

Argument type: point is an Array [2 or 3], and t is a Number.

Transforms a point from world space to layer space. Example: fromWorld(thisComp.layer(2).position)

See Expression example: Create a bulge between two layers for an example of how this method can be used.

toCompVec(vec, t=time)

Return type: Array [2 or 3].

Argument type: vec is an Array [2 or 3], and t is a Number.

Transforms a vector from layer space to composition space. Example: toCompVec([1,0])

fromCompVec(vec, t=time)

Return type: Array [2 or 3].

Argument type: vec is an Array [2 or 3], and t is a Number.

Transforms a vector from composition space to layer space. Example (2D layer):

  dir=sub(position, thisComp.layer(2).position);   fromCompVec(dir)

toWorldVec(vec, t=time)

Return type: Array [2 or 3].

Argument type: vec is an Array [2 or 3], and t is a Number.

Transforms a vector from layer space to world space. Example:   p1 = effect("Eye Bulge 1")("Bulge Center");    p2 = effect("Eye Bulge 2")("Bulge Center");     toWorld(sub(p1, p2))

fromWorldVec(vec, t=time)

Return type: Array [2 or 3].

Argument type: vec is an Array [2 or 3], and t is a Number.

Transforms a vector from world space to layer space. Example: fromWorld(thisComp.layer(2).position)

fromCompToSurface(point, t=time)

Return type: Array [2].

Argument type: point is an Array [2 or 3], and t is a Number.

Projects a point located in composition space to a point on the surface of the layer (zero z-value) at the location where it appears when viewed from the active camera. This method is useful for setting effect control points. Use with 3D layers only.

Camera attributes and methods (expression reference)

Camera objects have the same attributes and methods as Layer objects, except for source, effect, mask, width, height, anchorPoint, scale, opacity, audioLevels, timeRemap, and all the material properties.

pointOfInterest

Return type: Property [3].

Returns the point of interest values of a camera in world space.

zoom

Return type: Property.

Returns the zoom values of a camera in pixels.

Here’s an expression for the Scale property of a layer that maintains the relative size of the layer in frame while changing the z position (depth) of a layer or the Zoom value of a camera:

  cam = thisComp.activeCamera;    distance = length(sub(position, cam.position));    scale * distance / cam.zoom;

depthOfField

Return type: Property.

Returns 1 if the Depth Of Field property of a camera is on, or returns 0 if the Depth Of Field property is off.

focusDistance

Return type: Property.

Returns the focus distance value of a camera, in pixels.

aperture

Return type: Property.

Returns the aperture value of a camera, in pixels.

blurLevel

Return type: Property.

Returns the blur level value of a camera as a percentage.

active

Return type: Boolean.

Returns true if the camera is the active camera for the composition at the current time: the Video switch for the camera layer is on, the current time is in the range from the In point of the camera layer to the Out point of the camera layer, and it is the first (topmost) such camera layer listed in the Timeline panel. Returns false otherwise.

Light attributes and methods (expression reference)

Light objects have the same attributes and methods as Layer objects, except for source, effect, mask, width, height, anchorPoint, scale, opacity, audioLevels, timeRemap, and all the material properties.

pointOfInterest

Return type: Property [3].

Returns the point of interest values for a light in world space.

intensity

Return type: Property.

Returns the intensity values of a light as a percentage.

color

Return type: Property [4].

Returns the color value of a light.

coneAngle

Return type: Property.

Returns the cone angle of a light, in degrees.

coneFeather

Return type: Property.

Returns the cone feather value of a light as a percentage.

shadowDarkness

Return type: Property.

Returns the shadow darkness value of a light as a percentage.

shadowDiffusion

Return type: Property.

Returns the shadow diffusion value of a light, in pixels.

David Van Brink provides an instructional article and sample project on his omino pixel blog that show how to use expressions with lights.

Effect attributes and methods (expression reference)

active

Return type: Boolean.

Returns true if the effect is turned on (the Effect switch is selected).

param(name)

Return type: Property.

Argument type: name is a String.

Returns a property within an effect. Effect control points are always in layer space. Example:

  effect("Bulge").param("Bulge Height")

param(index)

Return type: Property.

Argument type: index is a Number.

Returns a property within an effect. Effect control points are always in layer space. For example, effect("Bulge").param(4) returns the Bulge Height property.

Mask attributes and methods (expression reference)

Note:

You can link Mask Path properties to other path properties (paths in a shape layer and brush strokes), but the properties are not accessible for direct numerical manipulation through expressions.

maskOpacity

Return type: Property.

Returns the opacity value of a mask as a percentage.

maskFeather

Return type: Property.

Returns the feather value of a mask, in pixels.

maskExpansion

Return type: Property.

Returns the expansion value of a mask, in pixels.

invert

Return type: Boolean.

Returns true if the mask is inverted or false if it is not.

Property attributes and methods (expression reference)

value

Return type: Number, Array, or String.

Returns the value of a property at the current time.

valueAtTime(t)

Return type: Number or Array.

Argument type: t is a Number.

Returns the value of a property at the specified time, in seconds.

For example, to have a property value for each frame chosen randomly from a set of four values, set your four values as keyframes at 0, 1, 2, and 3 seconds, and then apply the following expression to the property:

  valueAtTime(random(4))

Note:

Dan Ebberts provides more examples and techniques for using the valueAtTime and velocityAtTime methods on his MotionScript website.

velocity

Return type: Number or Array.

Returns the temporal velocity value at the current time. For spatial properties, such as Position, it returns the tangent vector value. The result is the same dimension as the property.

velocityAtTime(t)

Return type: Number or Array.

Argument type: t is a Number.

Returns the temporal velocity value at the specified time.

speed

Return type: Number.

Returns a 1D, positive speed value equal to the speed at which the property is changing at the default time. This element can be used only for spatial properties.

speedAtTime(t)

Return type: Number.

Argument type: t is a Number.

Returns the spatial speed value at the specified time.

wiggle(freq, amp, octaves=1, amp_mult=.5, t=time)

Return type: Number or Array.

Argument type: freq, amp, octaves, amp_mult, and t are Numbers.

Randomly shakes (wiggles) the value of the property.

freq value is the frequency in wiggles per second.

amp value is the amplitude in units of the property to which it is applied.

octaves is the number of octaves of noise to add together. This value controls how much detail is in the wiggle. Make this value higher than the default of 1 to include higher frequencies or lower to include amplitude harmonics in the wiggle.

amp_mult is the amount that amp is multiplied by for each octave. This value controls how fast the harmonics drop off. The default is 0.5; make it closer to 1 to have the harmonics added at the same amplitude as the base frequency, or closer to 0 to add in less detail.

t is the base start time. This value defaults to the current time. Use this parameter if you want the output to be a wiggle of the property value sampled at a different time.

Example: position.wiggle(5, 20, 3, .5) produces about 5 wiggles per second with an average size of about 20 pixels. In addition to the main wiggle, two more levels of detailed wiggles occur with a frequency of 10 and 20 wiggles per second, and sizes of 10 and 5 pixels, respectively.

This example, on a two-dimensional property such as Scale, wiggles both dimensions by the same amount:

  v = wiggle(5, 10);   [v[0], v[0]]

This example, on a two-dimensional property, wiggles only along the y-axis:

  freq = 3;   amp = 50;   w = wiggle(freq,amp);   [value[0],w[1]];

Dan Ebberts provides an example expression and a detailed explanation on his MotionScript website that shows how to use the time parameter of the wiggle method to create a looping animation.

temporalWiggle(freq, amp, octaves=1, amp_mult=.5, t=time)

Return type: Number or Array.

Argument type: freq, amp, octaves, amp_mult, and t are Numbers.

Samples the property at a wiggled time. The freq value is the frequency in wiggles per second, amp is the amplitude in units of the property to which it is applied, octaves is the number of octaves of noise to add together, amp_mult is the amount that amp is multiplied by for each octave, and t is the base start time. For this function to be meaningful, the property it samples must be animated, because the function alters only the time of sampling, not the value. Example: scale.temporalWiggle(5, .2)

smooth(width=.2, samples=5, t=time)

Return type: Number or Array.

Argument type: width, samples, and t are Numbers.

Smooths the property values over time, converting large, brief deviations in the value to smaller, more evenly distributed deviations. This smoothing is accomplished by applying a box filter to the value of the property at the specified time. The width value is the range of time (in seconds) over which the filter is averaged. The samples value is the number of discrete samples evenly spaced over time; use a larger value for greater smoothness (but decreased performance). Generally, you’ll want samples to be an odd number so that the value at the current time is included in the average. Example: position.smooth(.1, 5)

loopIn(type="cycle", numKeyframes=0)

Return type: Number or Array.

Loops a segment of time that is measured from the first keyframe on the layer forward toward the Out point of the layer. The loop plays from the In point of the layer. The numKeyframes value determines what segment is looped: The segment looped is the portion of the layer from the first keyframe to the numKeyframes+1 keyframe. For example, loopIn("cycle", 3) loops the segment bounded by the first and fourth keyframes. The default value of 0 means that all keyframes loop.

You can use keyframe-looping methods to repeat a series of keyframes. You can use these methods on most properties. Exceptions include properties that can’t be expressed by simple numeric values in the Timeline panel, such as the Source Text property, path shape properties, and the Histogram property for the Levels effect. Keyframes or duration values that are too large are clipped to the maximum allowable value. Values that are too small result in a constant loop.

 

loop type

result

cycle

(default) Repeats the specified segment.

pingpong

Repeats the specified segment, alternating between forward and backward.

offset

Repeats the specified segment, but offsets each cycle by the difference in the value of the property at the start and end of the segment, multiplied by the number of times the segment has looped.

continue

Does not repeat the specified segment, but continues to animate a property based on the velocity at the first or last keyframe. For example, if the last keyframe of a Scale property of a layer is 100%, the layer continues to scale from 100% to the Out point, instead of looping directly back to the Out point. This type does not accept a keyframes or duration argument.

loopOut(type="cycle", numKeyframes=0)

Return type: Number or Array.

Loops a segment of time that is measured from the last keyframe on the layer back toward the In point of the layer. The loop plays until the Out point of the layer. The specified number of keyframes determines the segment to loop. The numKeyframes value sets the number of keyframe segments to loop; the specified range is measured backward from the last keyframe. For example, loopOut("cycle", 1) loops the segment bounded by the last keyframe and second-to-last keyframe. The default value of 0 means that all keyframes loop. See the entry for loopIn for more information.

David Van Brink provides an instructional article and sample project on his omino pixel blog that show how to use the Echo effect, the Particle Playground effect, and the loopOut method to animate a swarm of stylized swimming bacteria.

loopInDuration(type="cycle", duration=0)

Return type: Number or Array.

Loops a segment of time that is measured from the first keyframe on the layer forward toward the Out point of the layer. The loop plays from the In point of the layer. Specified duration determines the segment to loop. The duration value sets the number of composition seconds in a segment to loop; the specified range is measured from the first keyframe. For example, loopInDuration("cycle",1) loops the first second of the entire animation. The default of 0 means that the segment to loop begins at the layer Out point. See the entry for loopIn for more information.

loopOutDuration(type="cycle", duration=0)

Return type: Number or Array.

Loops a segment of time that is measured from the last keyframe on the layer back toward the In point of the layer. The loop plays until the Out point of the layer. Specified duration determines the segmetn to loop. The duration value sets the number of composition seconds in a segment to loop; the specified range is measured backward from the last keyframe. For example, loopOutDuration("cycle", 1) loops the last second of the entire animation. The default of 0 means that the segment to loop begins at the layer In point. See the entry for loopIn for more information.

key(index)

Return type: Key or MarkerKey.

Argument type: index is a Number.

Returns the Key or MarkerKey object by number. For example, key(1) returns the first keyframe.

key(markerName)

Return type: MarkerKey.

Argument type: markerName is a String.

Returns the MarkerKey object with this name. Use only on marker properties.

nearestKey(t)

Return type: Key or MarkerKey.

Returns the Key or MarkerKey object nearest to a designated time.

numKeys

Return type: Number.

Returns the number of keyframes on a property. Returns the number of markers on a marker property.

Note:

If you use the Separate Dimensions command to separate the dimensions of the Position property into individual components, the number of keyframes changes, so the value returned by this method changes.

propertyGroup(countUp = 1)

Return type: Group.

Returns a group of properties relative to the property on which the expression is written. For example, if you add the propertyGroup(1) expression to the Rotation property of a brush stroke, the expression targets the Transform property group, which contains the Rotation property. If you add propertyGroup(2) instead, the expression targets the Brush property group. This method lets you establish name-independent relationships in the property hierarchy. This method is especially useful when duplicating properties that contain expressions.

The numProperties method for propertyGroup returns the number of properties in the property group.

This example returns the number of properties in the group that contains the property on which the expression is written:

  thisProperty.propertyGroup(1).numProperties

propertyIndex

Return type: Number.

Returns the index of a property relative to other properties in its property group, including property groups within masks, effects, text animators, selectors, shapes, trackers, and track points.

name

Return type: String.

Returns the name of the property or property group.

Example: Animating with the propertyGroup method and propertyIndex attribute

Values of propertyGroup relative to Position property of a brush stroke
Values of propertyGroup relative to Position property of a brush stroke

A. propertyGroup(4) B. propertyGroup(3) C. propertyGroup(2) D. propertyGroup(1) E. Position propertyIndex value is 2; Rotation propertyIndex value is 4. 

In this example, the propertyGroup method for each brush stroke targets the Brush property group because that group is two property groups up from the Rotation property. The propertyIndex attribute in each Brush stroke then returns a unique value for each Brush stroke. The resulting value is then multiplied by the time and 200 and applied to each rotation value, rotating each brush stroke differently, creating swirling paint strokes:  propertyGroup(2).propertyIndex * time * 200

  propertyGroup(2).propertyIndex * time * 200
Animating a brush stroke with an expression
Animating a brush stroke with an expression

Key attributes and methods (expression reference)

When you access a Key object, you can get time, index, and value properties from it. For example, the following expression gives you the value of the third Position keyframe: position.key(3).value.

The following expression, when written on an Opacity property with keyframes, ignores the keyframe values and uses only the placement of the keyframes in time to determine where a flash should occur:   d = Math.abs(time - nearestKey(time).time);   easeOut(d, 0, .1, 100, 0)

  d = Math.abs(time - nearestKey(time).time); 
  easeOut(d, 0, .1, 100, 0)

value

Return type: Number or Array.

Returns the value of the keyframe.

time

Return type: Number.

Returns the time of the keyframe.

index

Return type: Number.

Returns the index of the keyframe.

MarkerKey attributes (expression reference)

You can access values for composition markers and layer markers using the same methods. Access layer markers through the thisLayer.marker object; access composition markers through the thisComp.marker object.

For the purpose of expressions, markers are a special type of Key object, so you can use methods such as nearestKey(time) to access markers, and markers also have time and index attributes. The index attribute is not the number (name) of the marker; it is the keyframe index number, representing the order of the marker in the time ruler.

Expressions have access to all the values for a marker that you can set in the Composition Marker or Layer Marker dialog box. This expression on the Source Text property of a text layer displays the time, duration, index, comment (name), chapter, URL, frame target, and cue point name for the layer marker nearest the current time, and whether the marker is for an event cue point:

  m = thisLayer.marker.nearestKey(time);  
  s = "time:" + timeToCurrentFormat(m.time) + "\r" +  
      "duration: " + m.duration + "\r" +  
      "key index: " + m.index + "\r" +  
      "comment:" + m.comment + "\r" +  
      "chapter:" + m.chapter + "\r" +  
      "URL:" + m.url + "\r" +  
      "frame target: " + m.frameTarget + "\r" +  
      "cue point name: " + m.cuePointName + "\r" +  
      "Event cue point? " + m.eventCuePoint + "\r";  
  for (param in m.parameters){  
       s += "parameter: " + param + " value: " + m.parameters[param] + "\r";  
  } 
  s

Because the XMP metadata in a footage item can be converted into layer markers for a layer based on that item, expressions can interact with XMP metadata. For information, see XMP metadata in After Effects.

Dan Ebberts provides a tutorial on the After Effects Developer Center that includes an example of using XMP metadata with expressions.

duration

Return type: Number.

Duration, in seconds, of marker.

comment

Return type: String.

Contents of Comment field in marker dialog box.

chapter

Return type: String.

Contents of Chapter field in marker dialog box.

url

Return type: String.

Contents of URL field in marker dialog box.

frameTarget

Return type: String.

Contents of Frame Target field in marker dialog box.

eventCuePoint

Return type: Boolean.

Setting for cue point type in marker dialog box. True for Event; false for Navigation.

cuePointName

Return type: String.

Contents of cue point Name field in marker dialog box.

parameters

Return type: associative array of String values.

Contents of Parameter Name and Parameter Value fields in marker dialog box.

For example, if you have a parameter named “background color”, then you can use the following expression to access its value at the nearest marker:

  thisComp.marker.nearestKey(time).parameters["background color"]

This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License  Twitter™ and Facebook posts are not covered under the terms of Creative Commons.

Legal Notices   |   Online Privacy Policy