---
title: "Joining charts"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Joining charts}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

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

data_barchart <- data.frame(
  dep = c('Services', 'Production', 'Marketing', 'Purchasing'),
  profit = c(12, 15, 2, -3),
  operational = c(9, 7, 1.5, -0.4),
  property = c(2, 4, 0.5, -0.6),
  bonus = c(1, 4, 0, -2),
  prev_year = c(10, 16, 4, -1),
  plan = c(11, 13, 2, -2.5)
)

styles <- data.frame(profit = rep('actual', 4),
                     plan = rep('plan', 4),
                     prev_year = rep('previous',4))
```

In this vignette we will show how to join charts in different ways. This is a useful functionality of the `tidycharts` package, as it enables the user to group charts into panels or even generate multiple charts by one command (facetting).

```{r setup, message = FALSE}
library(tidycharts)
```

# Joining charts

Combining plots in one panel is useful and easy by using the function `join_charts`. By setting `nrows` and `ncols` parameters user can control the layout of the grid.

As you can see in [Getting Started vignette](https://mi2datalab.github.io/tidycharts/articles/Getting_Started.html#grouped-1), we created grouped bar chart and two variance bar charts. We can use the data and plots from that example to create a joined chart.

```{r barchart-join}
join_charts(
  bar_chart_grouped(data = data_barchart,
                  cat = data_barchart$dep,
                  foreground = 'prev_year',
                  background = 'profit',
                  markers = 'plan',
                  series_labels = c('PY', 'AC', 'PL')),
  bar_chart_absolute_variance(cat = data_barchart$dep, 
                                baseline = data_barchart$plan,
                                real = data_barchart$profit,
                                y_title = 'Plan vs. actual',
                                y_style = 'plan'),
  bar_chart_relative_variance(cat = data_barchart$dep, 
                                baseline = data_barchart$plan,
                                real = data_barchart$profit,
                                y_title = 'Plan vs. actual',
                                y_style = 'plan'),
  nrows = 1,
  ncols = 3)
```

# Facetting

Facetting means dividing data into some categories and generating charts for each category. We will apply `facet_chart` function to visualize data from R built-in demo dataset `mtcars`.

```{r mtcars}
head(mtcars)
```

We should facet the data by a variable which doesn't have many unique values (preferably a categorical variable), so we won't end up with astounding number of charts. In this case, we will facet by `cyl` variable, that has only 3 unique values, so we will get 3 charts. 

We need to pass `FUN` argument to `facet_chart` function, which must be one of plotting functions in the package. `FUN` is responsible for creating base charts. Arguments to `FUN` can be passed through `...`. In the example, `scatter_plot` is used.

```{r facet}
facet_chart(data = mtcars, 
            facet_by = 'cyl',
            ncols = 2,
            FUN = scatter_plot,
            x = mtcars$qsec,
            y = mtcars$hp,
            cat = mtcars$gear,
            legend_title = '',
            x_names = c('1/4 mile time', 'is s'),
            y_names = c('Horsepower', '')) 
```