---
title: "Customising colour palettes in visdat"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Customising-colour-palettes-in-visdat}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
```

```{r setup}
library(visdat)
```

# How to provide your own colour palette?

This vignette shoes you how to provide your own colour palette with `visdat`.

A `visdat` plot is a `ggplot` object - so we can use the tools from ggplot to 
tinker with colours. In this case, that is the `scale_fill_manual` function.

A "standard" visdat plot might be like so:

```{r standard}
vis_dat(typical_data)
```

You can name the colours yourself like so (after first loading the `ggplot` package.

```{r custom}
library(ggplot2)
vis_dat(typical_data) +
  scale_fill_manual(
    values = c(
      "character" = "red",
      "factor" = "blue",
      "logical" = "green",
      "numeric" = "purple",
      "NA" = "gray"
  ))
```

This is a pretty, uh, "popping" set of colours? You can also use some hex colours instead.

Say, taken from `palette()`:

```{r show-pal}
palette()
```


```{r pal-hex-visdat}
vis_dat(typical_data) +
  scale_fill_manual(
    values = c(
      "character" = "#61D04F",
      "factor" = "#2297E6",
      "logical" = "#28E2E5",
      "numeric" = "#CD0BBC",
      "NA" = "#F5C710"
  ))
```


How can we get nicer ones?

Well, you can use any of `ggplot`'s `scale_fill_*` functions from inside ggplot2

For example:

```{r scale-fill-brewer}
vis_dat(typical_data) +
  scale_fill_brewer()
```

```{r scale-fill-viridis}
vis_dat(typical_data) +
  scale_fill_viridis_d()
```

Happy colour palette exploring! You might want to take a look at some of the following colour palettes from other packages:

- [scico](https://github.com/thomasp85/scico#ggplot2-support)
- [colorspace](https://cran.r-project.org/package=colorspace/vignettes/colorspace.html#Usage_with_ggplot2)
- [wesanderson](https://github.com/karthik/wesanderson#palettes)