+ Agent.md

Author: metatronsw

Agent.md

@@ -0,0 +1,55 @@
+# Kolor Agent
+
+## Overview
+
+Kolor is a Swift library for working with various color spaces and performing color operations. It allows easy conversion between color spaces, color distance calculations, palette generation, image processing, and color grouping. It includes a large collection of predefined colors useful for design and scientific tasks.
+
+## Features
+
+* **Color Spaces:** Supports sRGB, Display P3, Rec. 2020, ProPhoto RGB, linear RGB, HSV, HSL, HWB, HSI, Okhsv/Okhsl, OkLCh, HPLuv, HSLuv, XYZ, Lab-like, LCh-like, ACES, CMYK, and more.
+* **Conversions:** Convert colors between different color spaces if possible.
+* **Color Distance:** Use `distance(from:)` or deltaE functions for Uniform Color Spaces (UCS) and sRGB.
+* **Palette Generation:** For cylindrical color spaces (Hue channel), generate palettes via `paletteGen(bySteps: Int, saturation: Double, value: Double)`.
+* **Image Processing:** Convert, analyze, and group image colors quickly and easily.
+* **Predefined Color Collections:** A large collection of artistic and scientific colors.
+
+## Design Principles
+
+* Simple and flexible logic for color manipulation.
+* sRGB stores values per channel as `UInt8` (0…255), other color spaces use normalized `Double` (0…1), except cylindrical color spaces where Hue is 0…360.
+* Built-in extensions for `CGImage` and `Image` classes to simplify image operations.
+
+## Usage Examples
+
+```swift
+// Converting a color from sRGB to HSL
+let hslColor = sRGBColor.red.toHSL()
+
+// Generating a palette in HSV
+let palette = hsvColor.paletteGen(bySteps: 12, saturation: 0.8, value: 0.9)
+
+// Calculating distance between two colors
+let distance = color1.distance(from: color2)
+
+// Convert NSImage to array of pixels
+let image = NSImage(named: "cat").toCGImage()
+let pixels = image.toSRGB()
+
+// Quantanize image	
+let counts = pixels.makeQuantizedCounts(levels: 4)
+let hist = Histogram(from: counts, levels: 4, totalPixels: pixels.count)
+
+// Identifies the most significant colors	 
+let colors = extractColorsWithRatio(from: image, maxDepth: 5)
+	
+// Posterize image
+let quantized = pixels.map { $0.toQuantized(by: 4) }
+let posterized = CGImage.initFrom(rgb: quantized, width: image.width, height: image.height)
+
+```
+
+## Notes
+
+* Components and formulas are adapted from multiple sources and rewritten in Swift.
+* Other components translated from C, C++, C#, Python, and JavaScript libraries.
+* Library targets Swift 5.6+.

README.md

@@ -2,7 +2,7 @@
 
 A library for color spaces and conversions in Swift 5.6+ 
 
-With this library, image conversion, analysis, color similarity, quantization, indexing, and grouping by tones can be implemented easily and quickly.
+With this library, image conversion, analysis, color similarity, quantization, indexing, histogram, and grouping by tones can be implemented easily and quickly.
 
 In addition, the library includes a large collection of colors. These artistic and scientific values can be useful for design work or tasks involving color groups.
 
@@ -17,7 +17,14 @@ LUV › LCh › HSLuv / HPLuv
 
 ```Swift
 	let srgb = sRGB(123, 223, 68)
-	let lab = srgb.toLAB()
+	let oklab = srgb.toOKLAB()
+	
+	let rgb = RGB(r: 1, g: 0.5, b: 0.2)
+	let p3 = rgb.toDisplayP3()
+	
+	let xyz = rgb.toXYZ()
+	let adapted = xyz.to_XYZ(fromWhiteRef: .D65, toWhiteRef: .D50)
+		
 ```
 
 In color spaces that are Uniform Color Spaces (UCS) and sRGB, the *distance(from:)* method or special deltaE functions can be applied.
@@ -26,6 +33,20 @@ In color spaces that include a cylindrical (Hue) channel, palette generation can
 
 There are excellent and useful extensions for the CGImage and Image classes.
 
+```Swift
+
+	let image = NSImage(named: "cat").toCGImage()
+	let pixels = image.toSRGB()
+	
+	let counts = pixels.makeQuantizedCounts(levels: 4)
+	let hist = Histogram(from: counts, levels: 4, totalPixels: pixels.count)
+	
+	let colors = extractColorsWithRatio(from: image, maxDepth: 5)
+	
+	let quantized = pixels.map { $0.toQuantized(by: 4) }
+	let posterized = CGImage.initFrom(rgb: quantized, width: image.width, height: image.height)
+	
+```
 
 ## Supported:
 

ToDo.md

@@ -1,9 +1 @@
-kesselni a már átalakított cam16 képpontokat ... ( ha már megvan ne kelljen átalakítani az rgb-t)
 
-channels set -et átnézni, most direkt értéket adunk vagy a struct skálázza magának ? 
-
-https://culorijs.org/color-spaces/
-
-HSI
-– I = (R+G+B)/3 (átlag)
-– A legjobban közelíti az emberi fényesség-érzetet.