graphs.brew

% Gergely Daróczi
% Unified plots

This document demonstrates the features of unifying plots in `pander`. First, if you do not want to deal with styling each of your images, just activate it: `evalsOptions('graph.unify', TRUE)`.

*Note*: please install `lattice` and `ggplot2` on your computer prior to trying to run all examples of this document.

<%
## generating dataset
df <- mtcars[, c('hp', 'wt')]
df$factor <- sample(c('Foo', 'Bar', 'Foo bar'), size = nrow(df), replace = TRUE)
df$factor2 <- sample(c('Foo', 'Bar', 'Foo bar'), size = nrow(df), replace = TRUE)
df$time <- 1:nrow(df)

## loading packages
require(ggplot2, quietly = TRUE)
require(lattice, quietly = TRUE)

## saving options for restting at the end
eO <- evalsOptions()
pO <- panderOptions()

## setting options
evalsOptions('graph.unify', TRUE)
evalsOptions('cache', FALSE)
%>

# Options

There are a bunch of options you might want to check out, these are:

<%=as.list(names(panderOptions())[grepl('^graph', names(panderOptions()))])%>

Find more details on [`pander`'s homepage](http://rapporter.github.com/pander/#pander-options).

# Default options

Not touching the above ones, let us check out how different plots look like by calling `base` R graphing function, `lattice` or `ggplot2`!

## Histogram

### Base R plot

<%=hist(df$hp, main = "Histogram in base R")%>

### lattice

<%=histogram(df$hp, main = "Histogram with lattice")%>

### ggplot2

<%=ggplot(df) + geom_histogram(aes(x = hp), binwidth = 50) + ggtitle("Histogram with ggplot2")%>

## Barplot

### Base R plot

<%=barplot(table(df$factor), main = "Barplot in base R")%>

### lattice

<%=barchart(df$factor, main = "Barplot with lattice")%>

### ggplot2

<%=ggplot(df) + geom_bar(aes(x = factor)) + ggtitle("Barplot with ggplot2")%>

## Points

### Base R plot

<%=plot(df$hp, df$wt, main = "Points in base R")%>

### lattice

<%=xyplot(df$wt ~ df$hp, main = "Points with lattice")%>

### ggplot2

<%=ggplot(df) + geom_point(aes(x = hp, y = wt)) + ggtitle("Points with ggplot2")%>

## Grouped plot

### Base R plot

I have no idea how to do that besides manually adding `points`.

### lattice

<%=xyplot(wt ~ hp, group = factor, data = df, auto.key = TRUE, main = "Grouped plot with lattice")%>

### ggplot2

<%=ggplot(df) + geom_point(aes(x = hp, y = wt, colour = factor)) + ggtitle("Grouped bar with ggplot2")%>

## Facets

### Base R plot

I have no idea how to do that besides `par(mfrow=c(foo, bar))`.

### lattice

<%=barchart(table(df$factor,df$factor2), groups = FALSE, layout = c(1,3), main = "Faceted barplot with lattice")%>

### ggplot2

<%=ggplot(df) + geom_bar(aes(x = factor)) + ggtitle("Faceted barplot with ggplot2") + facet_grid(factor2 ~ .)%>

## Boxplot

### Base R plot

<%=boxplot(df$hp ~ df$factor, main = "Boxplot in base R")%>

### lattice

<%=bwplot(df$factor ~ df$hp, main = "Boxplot with lattice")%>

### ggplot2

<%=ggplot(df) + geom_boxplot(aes(y = hp, x = factor)) + ggtitle("Boxplot with ggplot2")%>


## Lines

### Base R plot

<%=plot(df$time, df$hp, type = 'l')%>

### lattice

<%=xyplot(df$hp ~ df$time, type = 'l')%>

### ggplot2

<%=ggplot(df) + geom_line(aes(x = time, y = hp))%>

# Custom options

Below you can find some images generated by the exact same commands but with some modified `panderOptions`:

```r
panderOptions('graph.fontfamily', "Comic Sans MS")
panderOptions('graph.fontsize', 18)
panderOptions('graph.fontcolor', 'blue')
panderOptions('graph.grid.color', 'blue')
panderOptions('graph.axis.angle', 3)
panderOptions('graph.boxes', T)
panderOptions('graph.legend.position', 'top')
panderOptions('graph.colors', rainbow(5))
panderOptions('graph.grid', FALSE)
panderOptions('graph.symbol', 22)
```

<%
panderOptions('graph.fontfamily', "Comic Sans MS")
panderOptions('graph.fontsize', 18)
panderOptions('graph.fontcolor', 'blue')
panderOptions('graph.grid.color', 'blue')
panderOptions('graph.axis.angle', 3)
panderOptions('graph.boxes', T)
panderOptions('graph.legend.position', 'top')
panderOptions('graph.colors', rainbow(5))
panderOptions('graph.grid', FALSE)
panderOptions('graph.symbol', 22)
%>

## Histogram

### Base R plot

<%=hist(df$hp, main = "Histogram in base R")%>

### lattice

<%=histogram(df$hp, main = "Histogram with lattice")%>

### ggplot2

<%=ggplot(df) + geom_histogram(aes(x = hp), binwidth = 50) + ggtitle("Histogram with ggplot2")%>

## Barplot

### Base R plot

<%=barplot(table(df$factor), main = "Barplot in base R")%>

Yeah, the "Foo bar" label is cropped. We need a custom `mar` option here:

<%=par(mar=c(6, 4.3, 2.1, 0.1));+barplot(table(df$factor), main = "Barplot in base R")%>

But wait, we lost the color! Right: unfortunately coloring base R plots is really hackish, `pander` is adding the `col` attribute to the calls. If you start to tweak `par` in a chunk, you should prepare to some unwanted side-effects. Solution:

<%=par(mar=c(6, 4.3, 2.1, 0.1));+barplot(table(df$factor), main = "Barplot in base R", col = panderOptions('graph.colors')[1])%>

### lattice

<%=barchart(df$factor, main = "Barplot with lattice")%>

### ggplot2

<%=ggplot(df) + geom_bar(aes(x = factor)) + ggtitle("Barplot with ggplot2")%>

## Points

### Base R plot

<%=plot(df$hp, df$wt, main = "Points in base R")%>

### lattice

<%=xyplot(df$wt ~ df$hp, main = "Points with lattice")%>

### ggplot2

<%=ggplot(df) + geom_point(aes(x = hp, y = wt)) + ggtitle("Points with ggplot2")%>

## Grouped plot

### Base R plot

I have no idea how to do that besides manually adding `points`.

### lattice

<%=xyplot(wt ~ hp, group = factor, data = df, auto.key = TRUE, main = "Grouped plot with lattice")%>

### ggplot2

<%=ggplot(df) + geom_point(aes(x = hp, y = wt, colour = factor)) + ggtitle("Grouped bar with ggplot2")%>

## Facets

### Base R plot

I have no idea how to do that besides `par(mfrow=c(foo, bar))`.

### lattice

<%=barchart(table(df$factor,df$factor2), groups = FALSE, layout = c(1,3), main = "Faceted barplot with lattice")%>

### ggplot2

<%=ggplot(df) + geom_bar(aes(x = factor)) + ggtitle("Faceted barplot with ggplot2") + facet_grid(factor2 ~ .)%>

## Boxplot

### Base R plot

<%=boxplot(df$hp ~ df$factor, main = "Boxplot in base R")%>

### lattice

<%=bwplot(df$factor ~ df$hp, main = "Boxplot with lattice")%>

### ggplot2

<%=ggplot(df) + geom_boxplot(aes(y = hp, x = factor)) + ggtitle("Boxplot with ggplot2")%>


## Lines

### Base R plot

<%=plot(df$time, df$hp, type = 'l')%>

### lattice

<%=xyplot(df$hp ~ df$time, type = 'l')%>

### ggplot2

<%=ggplot(df) + geom_line(aes(x = time, y = hp))%>

<%
## resetting options
for (o in names(eO))
    evalsOptions(o, eO[[o]])
for (o in names(pO))
    panderOptions(o, pO[[o]])
%>