colorful

package module
v1.0.4-0...-238d372 Latest Latest
Warning

This package is not in the latest version of its module.

Go to latest
Published: Jun 8, 2019 License: MIT Imports: 6 Imported by: 0

README

go-colorful

A library for playing with colors in go (golang).

Build Status Coverage Status

Why?

I love games. I make games. I love detail and I get lost in detail. One such detail popped up during the development of Memory Which Does Not Suck, when we wanted the server to assign the players random colors. Sometimes two players got very similar colors, which bugged me. The very same evening, I want hue was the top post on HackerNews' frontpage and showed me how to Do It Right™. Last but not least, there was no library for handling color spaces available in go. Colorful does just that and implements Go's color.Color interface.

What?

Go-Colorful stores colors in RGB and provides methods from converting these to various color-spaces. Currently supported colorspaces are:

  • RGB: All three of Red, Green and Blue in [0..1].
  • HSL: Hue in [0..360], Saturation and Luminance in [0..1]. For legacy reasons; please forget that it exists.
  • HSV: Hue in [0..360], Saturation and Value in [0..1]. You're better off using HCL, see below.
  • Hex RGB: The "internet" color format, as in #FF00FF.
  • Linear RGB: See gamma correct rendering.
  • CIE-XYZ: CIE's standard color space, almost in [0..1].
  • CIE-xyY: encodes chromacity in x and y and luminance in Y, all in [0..1]
  • CIE-L*a*b*: A perceptually uniform color space, i.e. distances are meaningful. L* in [0..1] and a*, b* almost in [-1..1].
  • CIE-L*u*v*: Very similar to CIE-L*a*b*, there is no consensus on which one is "better".
  • CIE-L*C*h° (HCL): This is generally the most useful one; CIE-L*a*b* space in polar coordinates, i.e. a better HSV. H° is in [0..360], C* almost in [-1..1] and L* as in CIE-L*a*b*.

For the colorspaces where it makes sense (XYZ, Lab, Luv, HCl), the D65 is used as reference white by default but methods for using your own reference white are provided.

A coordinate being almost in a range means that generally it is, but for very bright colors and depending on the reference white, it might overflow this range slightly. For example, C* of #0000ff is 1.338.

Unit-tests are provided.

Nice, but what's it useful for?

  • Converting color spaces. Some people like to do that.
  • Blending (interpolating) between colors in a "natural" look by using the right colorspace.
  • Generating random colors under some constraints (e.g. colors of the same shade, or shades of one color.)
  • Generating gorgeous random palettes with distinct colors of a same temperature.

What not (yet)?

There are a few features which are currently missing and might be useful. I just haven't implemented them yet because I didn't have the need for it. Pull requests welcome.

  • Sorting colors (potentially using above mentioned distances)

So which colorspace should I use?

It depends on what you want to do. I think the folks from I want hue are on-spot when they say that RGB fits to how screens produce color, CIE L*a*b* fits how humans perceive color and HCL fits how humans think colors.

Whenever you'd use HSV, rather go for CIE-L*C*h°. for fixed lightness L* and chroma C* values, the hue angle h° rotates through colors of the same perceived brightness and intensity.

How?

Installing

Installing the library is as easy as

$ go get github.com/lucasb-eyer/go-colorful

The package can then be used through an

import "github.com/lucasb-eyer/go-colorful"
Basic usage

Create a beautiful blue color using different source space:

// Any of the following should be the same
c := colorful.Color{0.313725, 0.478431, 0.721569}
c, err := colorful.Hex("#517AB8")
if err != nil {
    log.Fatal(err)
}
c = colorful.Hsv(216.0, 0.56, 0.722)
c = colorful.Xyz(0.189165, 0.190837, 0.480248)
c = colorful.Xyy(0.219895, 0.221839, 0.190837)
c = colorful.Lab(0.507850, 0.040585,-0.370945)
c = colorful.Luv(0.507849,-0.194172,-0.567924)
c = colorful.Hcl(276.2440, 0.373160, 0.507849)
fmt.Printf("RGB values: %v, %v, %v", c.R, c.G, c.B)

And then converting this color back into various color spaces:

hex := c.Hex()
h, s, v := c.Hsv()
x, y, z := c.Xyz()
x, y, Y := c.Xyy()
l, a, b := c.Lab()
l, u, v := c.Luv()
h, c, l := c.Hcl()

Note that, because of Go's unfortunate choice of requiring an initial uppercase, the name of the functions relating to the xyY space are just off. If you have any good suggestion, please open an issue. (I don't consider XyY good.)

The color.Color interface

Because a colorful.Color implements Go's color.Color interface (found in the image/color package), it can be used anywhere that expects a color.Color.

Furthermore, you can convert anything that implements the color.Color interface into a colorful.Color using the MakeColor function:

c, ok := colorful.MakeColor(color.Gray16{12345})

Caveat: Be aware that this latter conversion (using MakeColor) hits a corner-case when alpha is exactly zero. Because color.Color uses pre-multiplied alpha colors, this means the RGB values are lost (set to 0) and it's impossible to recover them. In such a case MakeColor will return false as its second value.

Comparing colors

In the RGB color space, the Euclidian distance between colors doesn't correspond to visual/perceptual distance. This means that two pairs of colors which have the same distance in RGB space can look much further apart. This is fixed by the CIE-L*a*b*, CIE-L*u*v* and CIE-L*C*h° color spaces. Thus you should only compare colors in any of these space. (Note that the distance in CIE-L*a*b* and CIE-L*C*h° are the same, since it's the same space but in cylindrical coordinates)

Color distance comparison

The two colors shown on the top look much more different than the two shown on the bottom. Still, in RGB space, their distance is the same. Here is a little example program which shows the distances between the top two and bottom two colors in RGB, CIE-L*a*b* and CIE-L*u*v* space. You can find it in doc/colordist/colordist.go.

package main

import "fmt"
import "github.com/lucasb-eyer/go-colorful"

func main() {
	c1a := colorful.Color{150.0 / 255.0, 10.0 / 255.0, 150.0 / 255.0}
	c1b := colorful.Color{53.0 / 255.0, 10.0 / 255.0, 150.0 / 255.0}
	c2a := colorful.Color{10.0 / 255.0, 150.0 / 255.0, 50.0 / 255.0}
	c2b := colorful.Color{99.9 / 255.0, 150.0 / 255.0, 10.0 / 255.0}

	fmt.Printf("DistanceRgb:       c1: %v\tand c2: %v\n", c1a.DistanceRgb(c1b), c2a.DistanceRgb(c2b))
	fmt.Printf("DistanceLab:       c1: %v\tand c2: %v\n", c1a.DistanceLab(c1b), c2a.DistanceLab(c2b))
	fmt.Printf("DistanceLuv:       c1: %v\tand c2: %v\n", c1a.DistanceLuv(c1b), c2a.DistanceLuv(c2b))
	fmt.Printf("DistanceCIE76:     c1: %v\tand c2: %v\n", c1a.DistanceCIE76(c1b), c2a.DistanceCIE76(c2b))
	fmt.Printf("DistanceCIE94:     c1: %v\tand c2: %v\n", c1a.DistanceCIE94(c1b), c2a.DistanceCIE94(c2b))
	fmt.Printf("DistanceCIEDE2000: c1: %v\tand c2: %v\n", c1a.DistanceCIEDE2000(c1b), c2a.DistanceCIEDE2000(c2b))
}

Running the above program shows that you should always prefer any of the CIE distances:

$ go run colordist.go
DistanceRgb:       c1: 0.3803921568627451	and c2: 0.3858713931171159
DistanceLab:       c1: 0.32048458312798056	and c2: 0.24397151758565272
DistanceLuv:       c1: 0.5134369614199698	and c2: 0.2568692839860636
DistanceCIE76:     c1: 0.32048458312798056	and c2: 0.24397151758565272
DistanceCIE94:     c1: 0.19799168128511324	and c2: 0.12207136371167401
DistanceCIEDE2000: c1: 0.17274551120971166	and c2: 0.10665210031428465

It also shows that DistanceLab is more formally known as DistanceCIE76 and has been superseded by the slightly more accurate, but much more expensive DistanceCIE94 and DistanceCIEDE2000.

Note that AlmostEqualRgb is provided mainly for (unit-)testing purposes. Use it only if you really know what you're doing. It will eat your cat.

Blending colors

Blending is highly connected to distance, since it basically "walks through" the colorspace thus, if the colorspace maps distances well, the walk is "smooth".

Colorful comes with blending functions in RGB, HSV and any of the LAB spaces. Of course, you'd rather want to use the blending functions of the LAB spaces since these spaces map distances well but, just in case, here is an example showing you how the blendings (#fdffcc to #242a42) are done in the various spaces:

Blending colors in different spaces.

What you see is that HSV is really bad: it adds some green, which is not present in the original colors at all! RGB is much better, but it stays light a little too long. LUV and LAB both hit the right lightness but LAB has a little more color. HCL works in the same vein as HSV (both cylindrical interpolations) but it does it right in that there is no green appearing and the lighthness changes in a linear manner.

While this seems all good, you need to know one thing: When interpolating in any of the CIE color spaces, you might get invalid RGB colors! This is important if the starting and ending colors are user-input or random. An example of where this happens is when blending between #eeef61 and #1e3140:

Invalid RGB colors may crop up when blending in CIE spaces.

You can test whether a color is a valid RGB color by calling the IsValid method and indeed, calling IsValid will return false for the redish colors on the bottom. One way to "fix" this is to get a valid color close to the invalid one by calling Clamped, which always returns a nearby valid color. Doing this, we get the following result, which is satisfactory:

Fixing invalid RGB colors by clamping them to the valid range.

The following is the code creating the above three images; it can be found in doc/colorblend/colorblend.go

package main

import "fmt"
import "github.com/lucasb-eyer/go-colorful"
import "image"
import "image/draw"
import "image/png"
import "os"

func main() {
    blocks := 10
    blockw := 40
    img := image.NewRGBA(image.Rect(0,0,blocks*blockw,200))

    c1, _ := colorful.Hex("#fdffcc")
    c2, _ := colorful.Hex("#242a42")

    // Use these colors to get invalid RGB in the gradient.
    //c1, _ := colorful.Hex("#EEEF61")
    //c2, _ := colorful.Hex("#1E3140")

    for i := 0 ; i < blocks ; i++ {
        draw.Draw(img, image.Rect(i*blockw,  0,(i+1)*blockw, 40), &image.Uniform{c1.BlendHsv(c2, float64(i)/float64(blocks-1))}, image.ZP, draw.Src)
        draw.Draw(img, image.Rect(i*blockw, 40,(i+1)*blockw, 80), &image.Uniform{c1.BlendLuv(c2, float64(i)/float64(blocks-1))}, image.ZP, draw.Src)
        draw.Draw(img, image.Rect(i*blockw, 80,(i+1)*blockw,120), &image.Uniform{c1.BlendRgb(c2, float64(i)/float64(blocks-1))}, image.ZP, draw.Src)
        draw.Draw(img, image.Rect(i*blockw,120,(i+1)*blockw,160), &image.Uniform{c1.BlendLab(c2, float64(i)/float64(blocks-1))}, image.ZP, draw.Src)
        draw.Draw(img, image.Rect(i*blockw,160,(i+1)*blockw,200), &image.Uniform{c1.BlendHcl(c2, float64(i)/float64(blocks-1))}, image.ZP, draw.Src)

        // This can be used to "fix" invalid colors in the gradient.
        //draw.Draw(img, image.Rect(i*blockw,160,(i+1)*blockw,200), &image.Uniform{c1.BlendHcl(c2, float64(i)/float64(blocks-1)).Clamped()}, image.ZP, draw.Src)
    }

    toimg, err := os.Create("colorblend.png")
    if err != nil {
        fmt.Printf("Error: %v", err)
        return
    }
    defer toimg.Close()

    png.Encode(toimg, img)
}
Generating color gradients

A very common reason to blend colors is creating gradients. There is an example program in doc/gradientgen.go; it doesn't use any API which hasn't been used in the previous example code, so I won't bother pasting the code in here. Just look at that gorgeous gradient it generated in HCL space:

Getting random colors

It is sometimes necessary to generate random colors. You could simply do this on your own by generating colors with random values. By restricting the random values to a range smaller than [0..1] and using a space such as CIE-H*C*l° or HSV, you can generate both random shades of a color or random colors of a lightness:

random_blue := colorful.Hcl(180.0+rand.Float64()*50.0, 0.2+rand.Float64()*0.8, 0.3+rand.Float64()*0.7)
random_dark := colorful.Hcl(rand.Float64()*360.0, rand.Float64(), rand.Float64()*0.4)
random_light := colorful.Hcl(rand.Float64()*360.0, rand.Float64(), 0.6+rand.Float64()*0.4)

Since getting random "warm" and "happy" colors is quite a common task, there are some helper functions:

colorful.WarmColor()
colorful.HappyColor()
colorful.FastWarmColor()
colorful.FastHappyColor()

The ones prefixed by Fast are faster but less coherent since they use the HSV space as opposed to the regular ones which use CIE-L*C*h° space. The following picture shows the warm colors in the top two rows and happy colors in the bottom two rows. Within these, the first is the regular one and the second is the fast one.

Warm, fast warm, happy and fast happy random colors, respectively.

Don't forget to initialize the random seed! You can see the code used for generating this picture in doc/colorgens/colorgens.go.

Getting random palettes

As soon as you need to generate more than one random color, you probably want them to be distinguishible. Playing against an opponent which has almost the same blue as I do is not fun. This is where random palettes can help.

These palettes are generated using an algorithm which ensures that all colors on the palette are as distinguishible as possible. Again, there is a Fast method which works in HSV and is less perceptually uniform and a non-Fast method which works in CIE spaces. For more theory on SoftPalette, check out I want hue. Yet again, there is a Happy and a Warm version, which do what you expect, but now there is an additional Soft version, which is more configurable: you can give a constraint on the color space in order to get colors within a certain feel.

Let's start with the simple methods first, all they take is the amount of colors to generate, which could, for example, be the player count. They return an array of colorful.Color objects:

pal1, err1 := colorful.WarmPalette(10)
pal2 := colorful.FastWarmPalette(10)
pal3, err3 := colorful.HappyPalette(10)
pal4 := colorful.FastHappyPalette(10)
pal5, err5 := colorful.SoftPalette(10)

Note that the non-fast methods may fail if you ask for way too many colors. Let's move on to the advanced one, namely SoftPaletteEx. Besides the color count, this function takes a SoftPaletteSettings object as argument. The interesting part here is its CheckColor member, which is a boolean function taking three floating points as arguments: l, a and b. This function should return true for colors which lie within the region you want and false otherwise. The other members are Iteration, which should be within [5..100] where higher means slower but more exact palette, and ManySamples which you should set to true in case your CheckColor constraint rejects a large part of the color space.

For example, to create a palette of 10 brownish colors, you'd call it like this:

func isbrowny(l, a, b float64) bool {
    h, c, L := colorful.LabToHcl(l, a, b)
    return 10.0 < h && h < 50.0 && 0.1 < c && c < 0.5 && L < 0.5
}
// Since the above function is pretty restrictive, we set ManySamples to true.
brownies := colorful.SoftPaletteEx(10, colorful.SoftPaletteSettings{isbrowny, 50, true})

The following picture shows the palettes generated by all of these methods (sourcecode in doc/palettegens/palettegens.go), in the order they were presented, i.e. from top to bottom: Warm, FastWarm, Happy, FastHappy, Soft, SoftEx(isbrowny). All of them contain some randomness, so YMMV.

All example palettes

Again, the code used for generating the above image is available as doc/palettegens/palettegens.go.

Sorting colors

TODO: Sort using dist fn.

Using linear RGB for computations

There are two methods for transforming RGB<->Linear RGB: a fast and almost precise one, and a slow and precise one.

r, g, b := colorful.Hex("#FF0000").FastLinearRgb()

TODO: describe some more.

Want to use some other reference point?
c := colorful.LabWhiteRef(0.507850, 0.040585,-0.370945, colorful.D50)
l, a, b := c.LabWhiteRef(colorful.D50)
Reading and writing colors from databases

The type HexColor makes it easy to store colors as strings in a database. It implements the https://godoc.org/database/sql#Scanner and database/sql/driver.Value interfaces which provide automatic type conversion.

Example:

var hc HexColor
_, err := db.QueryRow("SELECT '#ff0000';").Scan(&hc)
// hc == HexColor{R: 1, G: 0, B: 0}; err == nil

FAQ

Q: I get all f!@#ed up values! Your library sucks!

A: You probably provided values in the wrong range. For example, RGB values are expected to reside between 0 and 1, not between 0 and 255. Normalize your colors.

Q: Lab/Luv/HCl seem broken! Your library sucks!

They look like this:

A: You're likely trying to generate and display colors that can't be represented by RGB, and thus monitors. When you're trying to convert, say, HCL(190.0, 1.0, 1.0).RGB255(), you're asking for RGB values of (-2105.254 300.680 286.185), which clearly don't exist, and the RGB255 function just casts these numbers to uint8, creating wrap-around and what looks like a completely broken gradient. What you want to do, is either use more reasonable values of colors which actually exist in RGB, or just Clamp() the resulting color to its nearest existing one, living with the consequences: HCL(190.0, 1.0, 1.0).Clamp().RGB255(). It will look something like this:

Here's an issue going in-depth about this, as well as my answer, both with code and pretty pictures. Also note that this was somewhat covered above in the "Blending colors" section.

Q: In a tight loop, conversion to Lab/Luv/HCl/... are slooooow!

A: Yes, they are. This library aims for correctness, readability, and modularity; it wasn't written with speed in mind. A large part of the slowness comes from these conversions going through LinearRgb which uses powers. I implemented a fast approximation to LinearRgb called FastLinearRgb by using Taylor approximations. The approximation is roughly 5x faster and precise up to roughly 0.5%, the major caveat being that if the input values are outside the range 0-1, accuracy drops dramatically. You can use these in your conversions as follows:

col := // Get your color somehow
l, a, b := XyzToLab(LinearRgbToXyz(col.LinearRgb()))

If you need faster versions of Distance* and Blend* that make use of this fast approximation, feel free to implement them and open a pull-request, I'll happily accept.

The derivation of these functions can be followed in [this Jupyter notebook](doc/LinearRGB Approximations.ipynb). Here's the main figure showing the approximation quality:

approximation quality

More speed could be gained by using SIMD instructions in many places. You can also get more speed for specific conversions by approximating the full conversion function, but that is outside the scope of this library. Thanks to @ZirconiumX for starting this investigation, see issue #18 for details.

Q: Why would MakeColor ever fail!?

A: MakeColor fails when the alpha channel is zero. In that case, the conversion is undefined. See issue 21 as well as the short caveat note in the "The color.Color interface" section above.

Who?

This library has been developed by Lucas Beyer with contributions from Bastien Dejean (@baskerville), Phil Kulak (@pkulak) and Christian Muehlhaeuser (@muesli).

Release Notes

Version 1.0
  • API Breaking change in MakeColor: instead of panicing when alpha is zero, it now returns a secondary, boolean return value indicating success. See the color.Color interface section and this FAQ entry for details.
Version 0.9
  • Initial version number after having ignored versioning for a long time :)

License: MIT

Copyright (c) 2013 Lucas Beyer

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Documentation

Overview

The colorful package provides all kinds of functions for working with colors.

Index

Constants

View Source
const Delta = 1.0 / 255.0

This is the tolerance used when comparing colors using AlmostEqualRgb.

View Source
const LAB_DELTA = 1e-6

Variables

View Source
var D50 = [3]float64{0.96422, 1.00000, 0.82521}

And another one.

View Source
var D65 = [3]float64{0.95047, 1.00000, 1.08883}

This is the default reference white point.

Functions

func HclToLab

func HclToLab(h, c, l float64) (L, a, b float64)

func LabToHcl

func LabToHcl(L, a, b float64) (h, c, l float64)

func LabToXyz

func LabToXyz(l, a, b float64) (x, y, z float64)

func LabToXyzWhiteRef

func LabToXyzWhiteRef(l, a, b float64, wref [3]float64) (x, y, z float64)

func LinearRgbToXyz

func LinearRgbToXyz(r, g, b float64) (x, y, z float64)

func LuvToXyz

func LuvToXyz(l, u, v float64) (x, y, z float64)

func LuvToXyzWhiteRef

func LuvToXyzWhiteRef(l, u, v float64, wref [3]float64) (x, y, z float64)

func XyyToXyz

func XyyToXyz(x, y, Y float64) (X, Yout, Z float64)

func XyzToLab

func XyzToLab(x, y, z float64) (l, a, b float64)

func XyzToLabWhiteRef

func XyzToLabWhiteRef(x, y, z float64, wref [3]float64) (l, a, b float64)

func XyzToLinearRgb

func XyzToLinearRgb(x, y, z float64) (r, g, b float64)

XyzToLinearRgb converts from CIE XYZ-space to Linear RGB space.

func XyzToLuv

func XyzToLuv(x, y, z float64) (l, a, b float64)

func XyzToLuvWhiteRef

func XyzToLuvWhiteRef(x, y, z float64, wref [3]float64) (l, u, v float64)

func XyzToXyy

func XyzToXyy(X, Y, Z float64) (x, y, Yout float64)

Well, the name is bad, since it's xyY but Golang needs me to start with a capital letter to make the method public.

func XyzToXyyWhiteRef

func XyzToXyyWhiteRef(X, Y, Z float64, wref [3]float64) (x, y, Yout float64)

Types

type Color

type Color struct {
	R, G, B float64
}

A color is stored internally using sRGB (standard RGB) values in the range 0-1

func FastHappyColor

func FastHappyColor() Color

Creates a random bright, "pimpy" color through a restricted HSV space.

func FastHappyPalette

func FastHappyPalette(colorsCount int) (colors []Color)

Uses the HSV color space to generate colors with similar S,V but distributed evenly along their Hue. This is fast but not always pretty. If you've got time to spare, use Lab (the non-fast below).

func FastLinearRgb

func FastLinearRgb(r, g, b float64) Color

FastLinearRgb is much faster than and almost as accurate as LinearRgb. BUT it is important to NOTE that they only produce good results for valid inputs r,g,b in [0,1].

func FastWarmColor

func FastWarmColor() Color

Creates a random dark, "warm" color through a restricted HSV space.

func FastWarmPalette

func FastWarmPalette(colorsCount int) (colors []Color)

Uses the HSV color space to generate colors with similar S,V but distributed evenly along their Hue. This is fast but not always pretty. If you've got time to spare, use Lab (the non-fast below).

func HappyColor

func HappyColor() (c Color)

Creates a random bright, "pimpy" color through restricted HCL space. This is slower than FastHappyColor but will likely give you colors which have the same "brightness" if you run it many times.

func HappyPalette

func HappyPalette(colorsCount int) ([]Color, error)

func Hcl

func Hcl(h, c, l float64) Color

Generates a color by using data given in HCL space using D65 as reference white. H values are in [0..360], C and L values are in [0..1] WARNING: many combinations of `l`, `a`, and `b` values do not have corresponding

valid RGB values, check the FAQ in the README if you're unsure.

func HclWhiteRef

func HclWhiteRef(h, c, l float64, wref [3]float64) Color

Generates a color by using data given in HCL space, taking into account a given reference white. (i.e. the monitor's white) H values are in [0..360], C and L values are in [0..1]

func Hex

func Hex(scol string) (Color, error)

Hex parses a "html" hex color-string, either in the 3 "#f0c" or 6 "#ff1034" digits form.

func Hsl

func Hsl(h, s, l float64) Color

Hsl creates a new Color given a Hue in [0..360], a Saturation [0..1], and a Luminance (lightness) in [0..1]

func Hsv

func Hsv(H, S, V float64) Color

Hsv creates a new Color given a Hue in [0..360], a Saturation and a Value in [0..1]

func Lab

func Lab(l, a, b float64) Color

Generates a color by using data given in CIE L*a*b* space using D65 as reference white. WARNING: many combinations of `l`, `a`, and `b` values do not have corresponding

valid RGB values, check the FAQ in the README if you're unsure.

func LabWhiteRef

func LabWhiteRef(l, a, b float64, wref [3]float64) Color

Generates a color by using data given in CIE L*a*b* space, taking into account a given reference white. (i.e. the monitor's white)

func LinearRgb

func LinearRgb(r, g, b float64) Color

LinearRgb creates an sRGB color out of the given linear RGB color (see http://www.sjbrown.co.uk/2004/05/14/gamma-correct-rendering/).

func Luv

func Luv(l, u, v float64) Color

Generates a color by using data given in CIE L*u*v* space using D65 as reference white. L* is in [0..1] and both u* and v* are in about [-1..1] WARNING: many combinations of `l`, `a`, and `b` values do not have corresponding

valid RGB values, check the FAQ in the README if you're unsure.

func LuvWhiteRef

func LuvWhiteRef(l, u, v float64, wref [3]float64) Color

Generates a color by using data given in CIE L*u*v* space, taking into account a given reference white. (i.e. the monitor's white) L* is in [0..1] and both u* and v* are in about [-1..1]

func MakeColor

func MakeColor(col color.Color) (Color, bool)

Constructs a colorful.Color from something implementing color.Color

func SoftPalette

func SoftPalette(colorsCount int) ([]Color, error)

A wrapper which uses common parameters.

func SoftPaletteEx

func SoftPaletteEx(colorsCount int, settings SoftPaletteSettings) ([]Color, error)

Yeah, windows-stype Foo, FooEx, screw you golang... Uses K-means to cluster the color-space and return the means of the clusters as a new palette of distinctive colors. Falls back to K-medoid if the mean happens to fall outside of the color-space, which can only happen if you specify a CheckColor function.

func WarmColor

func WarmColor() (c Color)

Creates a random dark, "warm" color through restricted HCL space. This is slower than FastWarmColor but will likely give you colors which have the same "warmness" if you run it many times.

func WarmPalette

func WarmPalette(colorsCount int) ([]Color, error)

func Xyy

func Xyy(x, y, Y float64) Color

Generates a color by using data given in CIE xyY space. x, y and Y are in [0..1]

func Xyz

func Xyz(x, y, z float64) Color

func (Color) AlmostEqualRgb

func (c1 Color) AlmostEqualRgb(c2 Color) bool

Check for equality between colors within the tolerance Delta (1/255).

func (Color) BlendHcl

func (col1 Color) BlendHcl(col2 Color, t float64) Color

BlendHcl blends two colors in the CIE-L*C*h° color-space, which should result in a smoother blend. t == 0 results in c1, t == 1 results in c2

func (Color) BlendHsv

func (c1 Color) BlendHsv(c2 Color, t float64) Color

You don't really want to use this, do you? Go for BlendLab, BlendLuv or BlendHcl.

func (Color) BlendLab

func (c1 Color) BlendLab(c2 Color, t float64) Color

BlendLab blends two colors in the L*a*b* color-space, which should result in a smoother blend. t == 0 results in c1, t == 1 results in c2

func (Color) BlendLuv

func (c1 Color) BlendLuv(c2 Color, t float64) Color

BlendLuv blends two colors in the CIE-L*u*v* color-space, which should result in a smoother blend. t == 0 results in c1, t == 1 results in c2

func (Color) BlendRgb

func (c1 Color) BlendRgb(c2 Color, t float64) Color

You don't really want to use this, do you? Go for BlendLab, BlendLuv or BlendHcl.

func (Color) Clamped

func (c Color) Clamped() Color

Returns Clamps the color into valid range, clamping each value to [0..1] If the color is valid already, this is a no-op.

func (Color) DistanceCIE76

func (c1 Color) DistanceCIE76(c2 Color) float64

That's actually the same, but I don't want to break code.

func (Color) DistanceCIE94

func (cl Color) DistanceCIE94(cr Color) float64

Uses the CIE94 formula to calculate color distance. More accurate than DistanceLab, but also more work.

func (Color) DistanceCIEDE2000

func (cl Color) DistanceCIEDE2000(cr Color) float64

DistanceCIEDE2000 uses the Delta E 2000 formula to calculate color distance. It is more expensive but more accurate than both DistanceLab and DistanceCIE94.

func (Color) DistanceCIEDE2000klch

func (cl Color) DistanceCIEDE2000klch(cr Color, kl, kc, kh float64) float64

DistanceCIEDE2000klch uses the Delta E 2000 formula with custom values for the weighting factors kL, kC, and kH.

func (Color) DistanceLab

func (c1 Color) DistanceLab(c2 Color) float64

DistanceLab is a good measure of visual similarity between two colors! A result of 0 would mean identical colors, while a result of 1 or higher means the colors differ a lot.

func (Color) DistanceLuv

func (c1 Color) DistanceLuv(c2 Color) float64

DistanceLuv is a good measure of visual similarity between two colors! A result of 0 would mean identical colors, while a result of 1 or higher means the colors differ a lot.

func (Color) DistanceRgb

func (c1 Color) DistanceRgb(c2 Color) float64

DistanceRgb computes the distance between two colors in RGB space. This is not a good measure! Rather do it in Lab space.

func (Color) FastLinearRgb

func (col Color) FastLinearRgb() (r, g, b float64)

FastLinearRgb is much faster than and almost as accurate as LinearRgb. BUT it is important to NOTE that they only produce good results for valid colors r,g,b in [0,1].

func (Color) Hcl

func (col Color) Hcl() (h, c, l float64)

Converts the given color to HCL space using D65 as reference white. H values are in [0..360], C and L values are in [0..1] although C can overshoot 1.0

func (Color) HclWhiteRef

func (col Color) HclWhiteRef(wref [3]float64) (h, c, l float64)

Converts the given color to HCL space, taking into account a given reference white. (i.e. the monitor's white) H values are in [0..360], C and L values are in [0..1]

func (Color) Hex

func (col Color) Hex() string

Hex returns the hex "html" representation of the color, as in #ff0080.

func (Color) Hsl

func (col Color) Hsl() (h, s, l float64)

Hsl returns the Hue [0..360], Saturation [0..1], and Luminance (lightness) [0..1] of the color.

func (Color) Hsv

func (col Color) Hsv() (h, s, v float64)

Hsv returns the Hue [0..360], Saturation and Value [0..1] of the color.

func (Color) IsValid

func (c Color) IsValid() bool

Checks whether the color exists in RGB space, i.e. all values are in [0..1]

func (Color) Lab

func (col Color) Lab() (l, a, b float64)

Converts the given color to CIE L*a*b* space using D65 as reference white.

func (Color) LabWhiteRef

func (col Color) LabWhiteRef(wref [3]float64) (l, a, b float64)

Converts the given color to CIE L*a*b* space, taking into account a given reference white. (i.e. the monitor's white)

func (Color) LinearRgb

func (col Color) LinearRgb() (r, g, b float64)

LinearRgb converts the color into the linear RGB space (see http://www.sjbrown.co.uk/2004/05/14/gamma-correct-rendering/).

func (Color) Luv

func (col Color) Luv() (l, u, v float64)

Converts the given color to CIE L*u*v* space using D65 as reference white. L* is in [0..1] and both u* and v* are in about [-1..1]

func (Color) LuvWhiteRef

func (col Color) LuvWhiteRef(wref [3]float64) (l, u, v float64)

Converts the given color to CIE L*u*v* space, taking into account a given reference white. (i.e. the monitor's white) L* is in [0..1] and both u* and v* are in about [-1..1]

func (Color) RGB255

func (col Color) RGB255() (r, g, b uint8)

Might come in handy sometimes to reduce boilerplate code.

func (Color) RGBA

func (col Color) RGBA() (r, g, b, a uint32)

Implement the Go color.Color interface.

func (Color) Xyy

func (col Color) Xyy() (x, y, Y float64)

Converts the given color to CIE xyY space using D65 as reference white. (Note that the reference white is only used for black input.) x, y and Y are in [0..1]

func (Color) XyyWhiteRef

func (col Color) XyyWhiteRef(wref [3]float64) (x, y, Y float64)

Converts the given color to CIE xyY space, taking into account a given reference white. (i.e. the monitor's white) (Note that the reference white is only used for black input.) x, y and Y are in [0..1]

func (Color) Xyz

func (col Color) Xyz() (x, y, z float64)

type HexColor

type HexColor Color

A HexColor is a Color stored as a hex string "#rrggbb". It implements the database/sql.Scanner and database/sql/driver.Value interfaces.

func (*HexColor) Scan

func (hc *HexColor) Scan(value interface{}) error

func (*HexColor) Value

func (hc *HexColor) Value() (driver.Value, error)

type SoftPaletteSettings

type SoftPaletteSettings struct {
	// A function which can be used to restrict the allowed color-space.
	CheckColor func(l, a, b float64) bool

	// The higher, the better quality but the slower. Usually two figures.
	Iterations int

	// Use up to 160000 or 8000 samples of the L*a*b* space (and thus calls to CheckColor).
	// Set this to true only if your CheckColor shapes the Lab space weirdly.
	ManySamples bool
}

Directories

Path Synopsis
doc

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL