I'm using ggplot2 to create quite a few facet_wrapped geom_line plot.

Although each plot only has a maximum of eight lines, when taken together, there are more like twenty categories to show on the legend.

In a similar vein to this: Recommend a scale colour for 13 or more categories and this: In R,how do I change the color value of just one value in ggplot2's scale_fill_brewer? I'd like to artificially up the number of colours I can show using colorbrewer's high-contrast colour sets.

An obvious way to do this would seem to be to 'recycle' the colours in the palette, with a different line symbol each time. So bright red with 'x's on the line could be a different category than bright red with 'o's etc.

Can anyone think how I might do this?

Thanks!

Edit

Here's some (sanitised) data to play with, and the R code I'm using to produce my plot.

Data: http://orca.casa.ucl.ac.uk/~rob/Stack%20Overflow%20question/stack%20overflow%20colours%20question%20data.csv

R code:

csvData <- read.csv("stack overflow colours question data.csv")
p <- ggplot(csvData, 
  aes(year, percentage_of_output, colour=category, group=category)) 
p + 
  geom_line(size=1.2)  + 
  labs(title = "Can I recycle the palette colours?", y = "% of output") +
  scale_colour_brewer(palette = "Set1") + 
  theme(plot.title = element_text(size = rel(1.5))) +
  facet_wrap("country_iso3", scales="free_y")
r ggplot2 colorbrewer
Edit
Report