Showing 4 of ~4 matching filters (D · 10+ stars)
ListDowncasting out of range should clamp for float => 8b/16b
The task requires modifying Gamut's image processing functions that convert floating-point pixel data (float) to 8-bit or 16-bit integer representations (ubyte/ushort). Currently, out-of-range values (less than 0 or greater than 1) cause wraparound, leading to bugs. The fix involves clamping these values to the valid 0-1 range before the conversion, eliminating the wraparound issue and ensuring predictable behavior.
Also browsing: Python , TypeScript , JavaScript