kitepdf-compose-viewer/io.github.yuroyami.kitepdf.compose/ImageDecoder

ImageDecoder

Bitmaps for the Compose canvas. decodeRaw wraps pixels that KitePDF's core decoded with ImageKodec. decode reads an encoded file with the decoder the host platform provides: Skia on the JVM, iOS, macOS and the web, and BitmapFactory on Android. KitePDF does not call decode for page images: an image that the shared decoders refuse draws as a placeholder on every canvas (#184).

The actual implementations live in :kitepdf-compose-viewer's platform source sets. If a platform can't decode the bytes (corrupt file, unsupported format), decode returns null.

Functions

decode

expect fun decode(bytes: ByteArray): ImageBitmap?

decodeRaw

expect fun decodeRaw(rgba: ByteArray, width: Int, height: Int): ImageBitmap?

Build a bitmap from already-decoded RGBA8888 pixels (R,G,B,A per pixel, row-major, width*height*4 bytes). Unlike decode this is fully synchronous on every platform (the samples are already decoded), so it works on JS too. Used for RAW (FlateDecode) PDF images.