Module image_cache

Structs§

AllPendingLoads 🔒
CompletedLoad 🔒: Represents an image that has completed loading. Images that fail to load (due to network or decode failure) are still stored here, so that they aren’t fetched again.
DecoderMsg 🔒: Message that the decoder worker threads send to the image cache.
ImageCacheImpl
ImageCacheStore 🔒: Image cache implementation.
KeyCache 🔒: As getting new keys takes a round trip over the constellation, we keep a small cache of them. Additionally, this cache will store image resources that do not have a key yet because those are needed to complete the load.
LoadKeyGenerator 🔒
PendingLoad 🔒: Represents an image that is either being loaded by the resource thread, or decoded by a worker thread.
RasterizationTask 🔒
VectorImageData 🔒

CacheResult 🔒: Result of accessing a cache.
DecodedImage 🔒
ImageBytes 🔒
KeyCacheState 🔒: The state of the `WebRenderImageKey`` cache
LoadResult 🔒
PendingKey 🔒: Used for storing images that do not have a WebRenderImageKey yet.

FALLBACK_RIPPY 🔒
MAX_SVG_PIXMAP_DIMENSION 🔒: The current SVG stack relies on resvg to provide the natural dimensions of the SVG, which it automatically infers from the width/height/viewBox properties of the SVG. Since these can be arbitrarily large, this can cause us to allocate a pixmap with very large dimensions leading to the process being killed due to memory exhaustion. For example, the /css/css-transforms/perspective-svg-001.html test uses very large values for viewBox. Hence, we just clamp the maximum width/height of the pixmap allocated for rasterization.

decode_bytes_sync 🔒
get_placeholder_image 🔒: This will block on getting an ImageKey but that is ok because it is done once upon start-up of a script-thread.
parse_svg_document_in_memory 🔒
set_webrender_image_key 🔒

ImageKey 🔒: https://html.spec.whatwg.org/multipage/#list-of-available-images
LoadKey 🔒