---
title: Scales
---

```{r}
library(gglite)
p = g2(mtcars, hp ~ mpg)
```

Scales control how data values map to visual properties. Use helpers like
`scale_x()`, `scale_y()`, `scale_color()`, and `scale_size()` to configure
the scale for a given channel.

## Linear Scale (default for numeric)

```{r}
p |> scale_x(type = 'linear')
```

## Log Scale

```{r}
p |> scale_y(type = 'log')
```

## Power Scale

```{r}
p |> scale_y(type = 'pow')
```

## Square Root Scale

```{r}
p |> scale_y(type = 'sqrt')
```

## Point Scale (for categorical axes)

The `'point'` scale positions discrete values with even spacing suitable for
point marks. (The `'ordinal'` scale is meant for non-positional channels
such as color or shape.)

```{r}
df = data.frame(x = c('A', 'B', 'C', 'D'), y = c(3, 7, 2, 5))
p_cat = g2(df, y ~ x)
p_cat |>
  mark_point() |>
  scale_x(type = 'point')
```

## Band Scale

```{r}
p_cat |> scale_x(type = 'band')
```

## Time Scale

```{r}
df = data.frame(
  date = seq(as.Date('2024-01-01'), by = 'month', length.out = 6),
  value = c(10, 20, 15, 30, 25, 35)
)
g2(df, value ~ date) |>
  scale_x(type = 'time')
```

## Custom Domain and Range

```{r}
p |>
  scale_x(domain = c(10, 35)) |>
  scale_y(domain = c(0, 400))
```

## Nice Rounding

```{r}
p |>
  scale_x(nice = TRUE) |>
  scale_y(nice = TRUE)
```

## Zero Inclusion

```{r}
p |> scale_y(zero = TRUE)
```

## Color Palette

```{r}
p_iris = g2(iris, Sepal.Length ~ Sepal.Width, color = ~ Species)
p_iris |> scale_color(palette = 'category10')
```

### Custom color range

```{r}
p_iris |> scale_color(range = c('#e41a1c', '#377eb8', '#4daf4a'))
```

## Identity Scale

```{r}
df = data.frame(x = c('A', 'B', 'C'), y = c(3, 7, 2), col = c('red', 'green', 'blue'))
g2(df, y ~ x, color = ~ col) |>
  scale_color(type = 'identity')
```

## Multiple Scales

```{r}
g2(mtcars, hp ~ mpg, color = ~ wt, size = ~ qsec) |>
  scale_x(nice = TRUE) |>
  scale_y(type = 'sqrt') |>
  scale_color(palette = 'viridis') |>
  scale_size(range = c(2, 10))
```