---
title: "Patterns 'stripe', 'wave', 'crosshatch', 'weave' - Parameters and Examples"
output: 
  rmarkdown::html_vignette:
  toc: true
vignette: >
  %\VignetteIndexEntry{Patterns 'stripe', 'wave', 'crosshatch', 'weave' - Parameters and Examples}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  dev = "ragg_png",
  fig.width = 7, 
  fig.height = 4
)
```

```{r setup}
suppressPackageStartupMessages({
  library(ggplot2)
  library(ggpattern)
})
```

## Introduction to the 'stripe', 'wave', 'crosshatch', and 'weave' patterns

* The 'stripe' pattern draws a set of (horizontal) **straight** lines.  
* The 'wave' pattern draws a set of (horizontal) **wavy** lines.
* The 'crosshatch' pattern draws a set of vertical lines on top of a set of horizontal lines.
* The 'weave' pattern draws a set of vertical lines interlaced with a set of horizontal lines.
* For more info on these patterns see their `{gridpattern}` documentation:

  * [`help("grid.pattern_stripe", package = "gridpattern")`](https://trevorldavis.com/R/gridpattern/dev/reference/grid.pattern_stripe.html)
  * [`help("grid.pattern_wave", package = "gridpattern")`](https://trevorldavis.com/R/gridpattern/dev/reference/grid.pattern_wave.html)
  * [`help("grid.pattern_crosshatch", package = "gridpattern")`](https://trevorldavis.com/R/gridpattern/dev/reference/grid.pattern_crosshatch.html)
  * [`help("grid.pattern_weave", package = "gridpattern")`](https://trevorldavis.com/R/gridpattern/dev/reference/grid.pattern_weave.html)

## Pattern Parameters

| aesthetic                | description                                   | default   | possible values                   |
|--------------------------|-----------------------------------------------|-----------|-----------------------------------|
| pattern\_colour           | Stroke colour                                 | 'grey20'  | colour                            |
| pattern\_fill             | Fill colour                                   | 'grey80'  | colour                            |
| pattern\_angle            | Rotation angle                                | 30        | angle in degrees                  |
| pattern\_density          | Approx. fraction of area the pattern fills    | 0.2       | value in range [0, 1] (fraction)  |
| pattern\_spacing          | Spacing between repetitions of pattern        | 0.05      | value in  `pattern_units` grid units |
| pattern\_xoffset          | Shift pattern along x axis                    | 0         | value in  `pattern_units` grid units |
| pattern\_yoffset          | Shift pattern along y axis                    | 0         | value in  `pattern_units` grid units |
| pattern\_units            | Pattern grid unit                             | 'snpc'     | `grid::unit()` unit i.e. 'snpc', 'cm', and 'inches' |
| pattern\_alpha            | Alpha                                         | NA        | value in range [0, 1] or NA |
| pattern\_linetype         | Stroke linetype                               | 1         | linetype                          |
| pattern\_size             | Stroke linewidth                              | 1         | linewidth                         |
| pattern\_type             | Pattern type                                  | NA        | See [wave](https://trevorldavis.com/R/gridpattern/dev/reference/grid.pattern\_wave.html) and [weave](https://trevorldavis.com/R/gridpattern/dev/reference/pattern\_weave.html) pattern documentation |
| pattern\_subtype          | Pattern subtype                               | NA        | See [weave](https://trevorldavis.com/R/gridpattern/dev/reference/pattern\_weave.html) pattern documentation |
| pattern\_frequency        | Frequency                                     | 0.1       | Frequency of waves in 'wave' pattern |

There are also a number of parameters for extra control of legend sizing and
aspect ratio adjustments. See the 'Pattern Parameters - Common' for more 
information.

## Data

Standard data for all the example plots

```{r}
df <- data.frame(trt = c("a", "b", "c"), outcome = c(2.3, 1.9, 3.2))
df
```

## Vanilla ggplot2

```{r}
ggplot(df, aes(trt, outcome)) +
  geom_col(aes(fill=trt),colour='black') +
  theme_bw() +
  labs(title = "Plain ggplot2")
```

## Use the `{ggpattern}` geom

* Use `ggpattern::geom_col_pattern()` instead of `ggplot2::geom_col()`.
* Set `pattern = 'stripe'` 
* Default legends in `ggplot2` are usually too small to show off an example of
  the pattern, so it will usually be necessary to increase the key size.
  
```{r}
ggplot(df, aes(trt, outcome)) +
  geom_col_pattern(aes(fill=trt),colour='black',
                   pattern='stripe') +
  theme_bw() +
  labs(title = "ggpattern") +
  theme(legend.key.size = unit(1.5, 'cm'))
```

## Mixing patterns

* Often we'll want to mix patterns.  
  To get the legend right we'll often want to set them with a pattern scale function
  like 'pattern_scale_manual()' instead of simply setting them with the 'pattern' aesthetic.
* Here we'll use the 'stripe' pattern and 'wave' pattern.
* The 'wave' pattern supports two subtypes: 'triangle' waves (default) have a zig-zag appearance
  whereas 'sine' waves have a curvy appearance.

```{r}
ggplot(df, aes(trt, outcome)) +
  geom_col_pattern(aes(fill=trt, pattern=trt, pattern_type=trt),colour='black') + 
  theme_bw() +
  labs(title = "Use 'stripe' and 'wave' patterns") + 
  theme(legend.key.size = unit(1.5, 'cm')) +
  scale_pattern_manual(values=c('stripe', 'wave', 'wave')) +
  scale_pattern_type_manual(values=c(NA, 'triangle', 'sine'))
```

## The Density Aesthetic

The aesthetic `pattern_density` roughly corresponds to the fraction of the filled
area which should be covered by the pattern.

In the following plot the density of striping is increased to 50% of the fill area.

```{r}
ggplot(df, aes(trt, outcome)) +
  geom_col_pattern(
    aes(fill=trt, pattern=trt),
    colour          = 'black', 
    pattern_density = 0.5
  ) +
  theme_bw() +
  labs(title = "Fixed density of 0.5 (50% of the fill area)") + 
  scale_pattern_manual(values=c('stripe', 'crosshatch', 'weave')) +
  theme(legend.key.size = unit(1.5, 'cm'))
```

## Weave types

The 'weave' pattern supports a rich set of weave types and subtypes including irregular 'matt', 'twill' (including 'herringbone' and 'zigzag' variations), and 'satin' weaves .  See the [weave](https://trevorldavis.com/R/gridpattern/dev/reference/pattern_weave.html) pattern documentation for more information.

```{r}
ggplot(df, aes(trt, outcome)) +
  geom_col_pattern(
    aes(pattern_fill2=trt, pattern_type=trt),
    pattern = 'weave',
    colour          = 'black', 
    pattern_density = 1.0,
    pattern_fill = 'grey',
    pattern_key_scale_factor = 0.5,
  ) +
  theme_bw() +
  labs(title = "Some 'weave' types") + 
  scale_pattern_type_manual(values=c('plain', 'twill', 'satin')) +
  theme(legend.key.size = unit(1.5, 'cm'))
```

## The Density Aesthetic as a Mapped Aesthetic

```{r}
ggplot(df, aes(trt, outcome)) +
  geom_col_pattern(
    aes(fill = trt, pattern_density = trt),
    colour          = 'black', 
    pattern         = 'stripe'
  ) +
  theme_bw() +
  labs(title = "Aesthetic Mapping of 'trt' to Density") + 
  theme(legend.key.size = unit(1.5, 'cm'))
```


## The Density Aesthetic as a Mapped Aesthetic with Manual Scale


`scale_pattern_density_manual()` can be used to manually control how the variable 
is mapped to the density.


```{r}
ggplot(df, aes(trt, outcome)) +
  geom_col_pattern(
    aes(fill = trt, pattern_density = trt),
    colour          = 'black', 
    pattern         = 'stripe'
  ) +
  theme_bw() +
  labs(title = "Aesthetic Mapping of 'trt' to Density") + 
  theme(legend.key.size = unit(1.5, 'cm')) + 
  scale_pattern_density_manual(values = c(a = 0.1, b=0.3, c=0.5))
```


## The Spacing Aesthetic as a Mapped Aesthetic

```{r}
ggplot(df, aes(trt, outcome)) +
  geom_col_pattern(
    aes(fill = trt, pattern_spacing = trt),
    colour          = 'black', 
    pattern         = 'stripe'
  ) +
  theme_bw() +
  labs(title = "Aesthetic Mapping of 'trt' to Spacing") + 
  theme(legend.key.size = unit(1.5, 'cm'))
```


## The Fill Aesthetic as a Mapped Aesthetic

```{r}
ggplot(df, aes(trt, outcome)) +
  geom_col_pattern(
    aes(fill = trt, pattern_fill = trt),
    colour          = 'black', 
    pattern         = 'stripe'
  ) +
  theme_bw() +
  labs(title = "Aesthetic Mapping of 'trt' to Pattern Fill") + 
  scale_pattern_fill_viridis_d() + 
  theme(legend.key.size = unit(1.5, 'cm'))
```