Snapshot

class Snapshot(**properties: Any)

Superclasses: Snapshot, Object

GtkSnapshot assists in creating RenderNode’s for widgets.

It functions in a similar way to a cairo context, and maintains a stack of render nodes and their associated transformations.

The node at the top of the stack is the one that `gtk_snapshot_append_…()` functions operate on. Use the `gtk_snapshot_push_…()` functions and pop to change the current node.

The typical way to obtain a GtkSnapshot object is as an argument to the snapshot vfunc. If you need to create your own GtkSnapshot, use new.

Constructors

class Snapshot
classmethod new() → Snapshot

Creates a new GtkSnapshot.

Methods

class Snapshot
append_border(outline: RoundedRect, border_width: Sequence[float], border_color: Sequence[RGBA]) → None

Appends a stroked border rectangle inside the given outline.

The four sides of the border can have different widths and colors.

Parameters:

  • outline – the outline of the border

  • border_width – the stroke width of the border on the top, right, bottom and left side respectively.

  • border_color – the color used on the top, right, bottom and left side.

append_cairo(bounds: Rect) → Context

Creates a new CairoNode and appends it to the current render node of snapshot, without changing the current node.

Parameters:

bounds – the bounds for the new node

append_color(color: RGBA, bounds: Rect) → None

Creates a new render node drawing the color into the given bounds and appends it to the current render node of snapshot.

You should try to avoid calling this function if color is transparent.

Parameters:

  • color – the color to draw

  • bounds – the bounds for the new node

append_conic_gradient(bounds: Rect, center: Point, rotation: float, stops: Sequence[ColorStop]) → None

Appends a conic gradient node with the given stops to snapshot.

Parameters:

  • bounds – the rectangle to render the gradient into

  • center – the center point of the conic gradient

  • rotation – the clockwise rotation in degrees of the starting angle. 0 means the starting angle is the top.

  • stops – the color stops defining the gradient

append_fill(path: Path, fill_rule: FillRule, color: RGBA) → None

A convenience method to fill a path with a color.

See push_fill if you need to fill a path with more complex content than a color.

Added in version 4.14.

Parameters:

  • path – The path describing the area to fill

  • fill_rule – The fill rule to use

  • color – the color to fill the path with

append_inset_shadow(outline: RoundedRect, color: RGBA, dx: float, dy: float, spread: float, blur_radius: float) → None

Appends an inset shadow into the box given by outline.

Parameters:

  • outline – outline of the region surrounded by shadow

  • color – color of the shadow

  • dx – horizontal offset of shadow

  • dy – vertical offset of shadow

  • spread – how far the shadow spreads towards the inside

  • blur_radius – how much blur to apply to the shadow

append_layout(layout: Layout, color: RGBA) → None
Parameters:

  • layout

  • color

append_linear_gradient(bounds: Rect, start_point: Point, end_point: Point, stops: Sequence[ColorStop]) → None

Appends a linear gradient node with the given stops to snapshot.

Parameters:

  • bounds – the rectangle to render the linear gradient into

  • start_point – the point at which the linear gradient will begin

  • end_point – the point at which the linear gradient will finish

  • stops – the color stops defining the gradient

append_node(node: RenderNode) → None

Appends node to the current render node of snapshot, without changing the current node.

If snapshot does not have a current node yet, node will become the initial node.

Parameters:

node – a GskRenderNode

append_outset_shadow(outline: RoundedRect, color: RGBA, dx: float, dy: float, spread: float, blur_radius: float) → None

Appends an outset shadow node around the box given by outline.

Parameters:

  • outline – outline of the region surrounded by shadow

  • color – color of the shadow

  • dx – horizontal offset of shadow

  • dy – vertical offset of shadow

  • spread – how far the shadow spreads towards the outside

  • blur_radius – how much blur to apply to the shadow

append_radial_gradient(bounds: Rect, center: Point, hradius: float, vradius: float, start: float, end: float, stops: Sequence[ColorStop]) → None

Appends a radial gradient node with the given stops to snapshot.

Parameters:

  • bounds – the rectangle to render the readial gradient into

  • center – the center point for the radial gradient

  • hradius – the horizontal radius

  • vradius – the vertical radius

  • start – the start position (on the horizontal axis)

  • end – the end position (on the horizontal axis)

  • stops – the color stops defining the gradient

append_repeating_linear_gradient(bounds: Rect, start_point: Point, end_point: Point, stops: Sequence[ColorStop]) → None

Appends a repeating linear gradient node with the given stops to snapshot.

Parameters:

  • bounds – the rectangle to render the linear gradient into

  • start_point – the point at which the linear gradient will begin

  • end_point – the point at which the linear gradient will finish

  • stops – the color stops defining the gradient

append_repeating_radial_gradient(bounds: Rect, center: Point, hradius: float, vradius: float, start: float, end: float, stops: Sequence[ColorStop]) → None

Appends a repeating radial gradient node with the given stops to snapshot.

Parameters:

  • bounds – the rectangle to render the readial gradient into

  • center – the center point for the radial gradient

  • hradius – the horizontal radius

  • vradius – the vertical radius

  • start – the start position (on the horizontal axis)

  • end – the end position (on the horizontal axis)

  • stops – the color stops defining the gradient

append_scaled_texture(texture: Texture, filter: ScalingFilter, bounds: Rect) → None

Creates a new render node drawing the texture into the given bounds and appends it to the current render node of snapshot.

In contrast to append_texture, this function provides control about how the filter that is used when scaling.

Added in version 4.10.

Parameters:

  • texture – the texture to render

  • filter – the filter to use

  • bounds – the bounds for the new node

append_stroke(path: Path, stroke: Stroke, color: RGBA) → None

A convenience method to stroke a path with a color.

See push_stroke if you need to stroke a path with more complex content than a color.

Added in version 4.14.

Parameters:

  • path – The path describing the area to fill

  • stroke – The stroke attributes

  • color – the color to fill the path with

append_texture(texture: Texture, bounds: Rect) → None

Creates a new render node drawing the texture into the given bounds and appends it to the current render node of snapshot.

If the texture needs to be scaled to fill bounds, linear filtering is used. See append_scaled_texture if you need other filtering, such as nearest-neighbour.

Parameters:

  • texture – the texture to render

  • bounds – the bounds for the new node

gl_shader_pop_texture() → None

Removes the top element from the stack of render nodes and adds it to the nearest GLShaderNode below it.

This must be called the same number of times as the number of textures is needed for the shader in push_gl_shader.

perspective(depth: float) → None

Applies a perspective projection transform.

See perspective for a discussion on the details.

Parameters:

depth – distance of the z=0 plane

pop() → None

Removes the top element from the stack of render nodes, and appends it to the node underneath it.

push_blend(blend_mode: BlendMode) → None

Blends together two images with the given blend mode.

Until the first call to pop, the bottom image for the blend operation will be recorded. After that call, the top image to be blended will be recorded until the second call to pop.

Calling this function requires two subsequent calls to pop.

Parameters:

blend_mode – blend mode to use

push_blur(radius: float) → None

Blurs an image.

The image is recorded until the next call to pop.

Parameters:

radius – the blur radius to use. Must be positive

push_clip(bounds: Rect) → None

Clips an image to a rectangle.

The image is recorded until the next call to pop.

Parameters:

bounds – the rectangle to clip to

push_color_matrix(color_matrix: Matrix, color_offset: Vec4) → None

Modifies the colors of an image by applying an affine transformation in RGB space.

In particular, the colors will be transformed by applying

pixel = transpose(color_matrix) * pixel + color_offset

for every pixel. The transformation operates on unpremultiplied colors, with color components ordered R, G, B, A.

The image is recorded until the next call to pop.

Parameters:

  • color_matrix – the color matrix to use

  • color_offset – the color offset to use

push_cross_fade(progress: float) → None

Snapshots a cross-fade operation between two images with the given progress.

Until the first call to pop, the start image will be snapshot. After that call, the end image will be recorded until the second call to pop.

Calling this function requires two subsequent calls to pop.

Parameters:

progress – progress between 0.0 and 1.0

push_fill(path: Path, fill_rule: FillRule) → None

Fills the area given by path and fill_rule with an image and discards everything outside of it.

The image is recorded until the next call to pop.

If you want to fill the path with a color, append_fill may be more convenient.

Added in version 4.14.

Parameters:

  • path – The path describing the area to fill

  • fill_rule – The fill rule to use

push_gl_shader(shader: GLShader, bounds: Rect, take_args: Bytes) → None

Push a GLShaderNode.

The node uses the given GLShader and uniform values Additionally this takes a list of n_children other nodes which will be passed to the GLShaderNode.

The take_args argument is a block of data to use for uniform arguments, as per types and offsets defined by the shader. Normally this is generated by format_args or ShaderArgsBuilder.

The snapshotter takes ownership of take_args, so the caller should not free it after this.

If the renderer doesn’t support GL shaders, or if there is any problem when compiling the shader, then the node will draw pink. You should use compile to ensure the shader will work for the renderer before using it.

If the shader requires textures (see get_n_textures), then it is expected that you call gl_shader_pop_texture the number of times that are required. Each of these calls will generate a node that is added as a child to the GskGLShaderNode, which in turn will render these offscreen and pass as a texture to the shader.

Once all textures (if any) are pop:ed, you must call the regular pop.

If you want to use pre-existing textures as input to the shader rather than rendering new ones, use append_texture to push a texture node. These will be used directly rather than being re-rendered.

For details on how to write shaders, see GLShader.

Parameters:

  • shader – The code to run

  • bounds – the rectangle to render into

  • take_args – Data block with arguments for the shader.

push_mask(mask_mode: MaskMode) → None

Until the first call to pop, the mask image for the mask operation will be recorded.

After that call, the source image will be recorded until the second call to pop.

Calling this function requires 2 subsequent calls to pop().

Added in version 4.10.

Parameters:

mask_mode – mask mode to use

push_opacity(opacity: float) → None

Modifies the opacity of an image.

The image is recorded until the next call to pop.

Parameters:

opacity – the opacity to use

push_repeat(bounds: Rect, child_bounds: Rect | None = None) → None

Creates a node that repeats the child node.

The child is recorded until the next call to pop.

Parameters:

  • bounds – the bounds within which to repeat

  • child_bounds – the bounds of the child or None to use the full size of the collected child node

push_rounded_clip(bounds: RoundedRect) → None

Clips an image to a rounded rectangle.

The image is recorded until the next call to pop.

Parameters:

bounds – the rounded rectangle to clip to

push_shadow(shadow: Sequence[Shadow]) → None

Applies a shadow to an image.

The image is recorded until the next call to pop.

Parameters:

shadow – the first shadow specification

push_stroke(path: Path, stroke: Stroke) → None

Strokes the given path with the attributes given by stroke and an image.

The image is recorded until the next call to pop.

Note that the strokes are subject to the same transformation as everything else, so uneven scaling will cause horizontal and vertical strokes to have different widths.

If you want to stroke the path with a color, append_stroke may be more convenient.

Added in version 4.14.

Parameters:

  • path – The path to stroke

  • stroke – The stroke attributes

render_background(context: StyleContext, x: float, y: float, width: float, height: float) → None

Creates a render node for the CSS background according to context, and appends it to the current node of snapshot, without changing the current node.

Deprecated since version 4.10: Please do not use it in newly written code

Parameters:

  • context – the style context that defines the background

  • x – X origin of the rectangle

  • y – Y origin of the rectangle

  • width – rectangle width

  • height – rectangle height

render_focus(context: StyleContext, x: float, y: float, width: float, height: float) → None

Creates a render node for the focus outline according to context, and appends it to the current node of snapshot, without changing the current node.

Deprecated since version 4.10: Please do not use it in newly written code

Parameters:

  • context – the style context that defines the focus ring

  • x – X origin of the rectangle

  • y – Y origin of the rectangle

  • width – rectangle width

  • height – rectangle height

render_frame(context: StyleContext, x: float, y: float, width: float, height: float) → None

Creates a render node for the CSS border according to context, and appends it to the current node of snapshot, without changing the current node.

Deprecated since version 4.10: Please do not use it in newly written code

Parameters:

  • context – the style context that defines the frame

  • x – X origin of the rectangle

  • y – Y origin of the rectangle

  • width – rectangle width

  • height – rectangle height

render_insertion_cursor(context: StyleContext, x: float, y: float, layout: Layout, index: int, direction: Direction) → None

Draws a text caret using snapshot at the specified index of layout.

Deprecated since version 4.10: Please do not use it in newly written code

Parameters:

  • context – a GtkStyleContext

  • x – X origin

  • y – Y origin

  • layout – the PangoLayout of the text

  • index – the index in the PangoLayout

  • direction – the PangoDirection of the text

render_layout(context: StyleContext, x: float, y: float, layout: Layout) → None

Creates a render node for rendering layout according to the style information in context, and appends it to the current node of snapshot, without changing the current node.

Deprecated since version 4.10: Please do not use it in newly written code

Parameters:

  • context – the style context that defines the text

  • x – X origin of the rectangle

  • y – Y origin of the rectangle

  • layout – the PangoLayout to render

restore() → None

Restores snapshot to the state saved by a preceding call to save and removes that state from the stack of saved states.

rotate(angle: float) → None

Rotates @``snapshot``’s coordinate system by angle degrees in 2D space - or in 3D speak, rotates around the Z axis. The rotation happens around the origin point of (0, 0) in the snapshot’s current coordinate system.

To rotate around axes other than the Z axis, use rotate_3d.

Parameters:

angle – the rotation angle, in degrees (clockwise)

rotate_3d(angle: float, axis: Vec3) → None

Rotates snapshot’s coordinate system by angle degrees around axis.

For a rotation in 2D space, use rotate.

Parameters:

  • angle – the rotation angle, in degrees (clockwise)

  • axis – The rotation axis

save() → None

Makes a copy of the current state of snapshot and saves it on an internal stack.

When restore is called, snapshot will be restored to the saved state.

Multiple calls to save and restore can be nested; each call to :func:`~gi.repository.Gtk.Snapshot.restore` restores the state from the matching paired :func:`~gi.repository.Gtk.Snapshot.save`.

It is necessary to clear all saved states with corresponding calls to :func:`~gi.repository.Gtk.Snapshot.restore`.

scale(factor_x: float, factor_y: float) → None

Scales snapshot’s coordinate system in 2-dimensional space by the given factors.

Use scale_3d to scale in all 3 dimensions.

Parameters:

  • factor_x – scaling factor on the X axis

  • factor_y – scaling factor on the Y axis

scale_3d(factor_x: float, factor_y: float, factor_z: float) → None

Scales snapshot’s coordinate system by the given factors.

Parameters:

  • factor_x – scaling factor on the X axis

  • factor_y – scaling factor on the Y axis

  • factor_z – scaling factor on the Z axis

to_node() → RenderNode | None

Returns the render node that was constructed by snapshot.

Note that this function may return None if nothing has been added to the snapshot or if its content does not produce pixels to be rendered.

After calling this function, it is no longer possible to add more nodes to snapshot. The only function that should be called after this is unref.

to_paintable(size: Size | None = None) → Paintable | None

Returns a paintable encapsulating the render node that was constructed by snapshot.

After calling this function, it is no longer possible to add more nodes to snapshot. The only function that should be called after this is unref.

Parameters:

size – The size of the resulting paintable or None to use the bounds of the snapshot

transform(transform: Transform | None = None) → None

Transforms snapshot’s coordinate system with the given transform.

Parameters:

transform – the transform to apply

transform_matrix(matrix: Matrix) → None

Transforms snapshot’s coordinate system with the given matrix.

Parameters:

matrix – the matrix to multiply the transform with

translate(point: Point) → None

Translates snapshot’s coordinate system by point in 2-dimensional space.

Parameters:

point – the point to translate the snapshot by

translate_3d(point: Point3D) → None

Translates snapshot’s coordinate system by point.

Parameters:

point – the point to translate the snapshot by