How to write a function in r which plots data for each unique value?

Question

Based on Vancouver crime data, I am trying to write a function which plots the number of crimes per neighbourhood for all types of crime.

Below is what I have done so far:

all_type <- unique(crime$TYPE)

plot_all <- function(x) {

  m <- filter(crime, YEAR %in% c(2003, 2019), TYPE == x) %>%
  ggplot(aes(x= fct_rev(fct_infreq(NEIGHBOURHOOD))))+
  geom_bar(stat = "count")+
  coord_flip()+
  ggtitle(paste("Crime type:", x, sep=" "))+
  labs(x=" ", y="Total incidents, 2003-2019")

  for (i in seq_along(x)){
    result <- {print(m)}
  }
return(result)

}

plot_all(all_type)

But this only returns the plot for the first listed type of crime, instead of returning separate plots for each type of crime.

Answer 1

There are some issues with the function as posted. In the code below I will separate the problem of plotting one graph from the problem of plotting all graphs. To each of these problems will correspond a function.

First of all, the data I am using can be downloaded with the following instructions, that can be found at RPubs by RStudio:

Crime Data from Vancouver Police Department by Arash Tavassoli (2018-11-05)

url <- "ftp://webftp.vancouver.ca/opendata/csv/crime_csv_all_years.zip"
temp <- tempfile()
download.file(url, temp)
crime <- read.csv(unz(temp, "crime_csv_all_years.csv"))
unlink(temp)

Now the graphics code.

library(dplyr)
library(tidyr)
library(forcats)
library(ggplot2)

plot_one <- function(x, crime) {
  m <- filter(crime, YEAR %in% c(2003, 2019), TYPE == x) %>%
    ggplot(aes(x = fct_rev(fct_infreq(NEIGHBOURHOOD)))) +
    geom_bar(stat = "count") +
    coord_flip() +
    ggtitle(paste("Crime type:", x, sep=" ")) +
    labs(x = " ", y = "Total incidents, 2003-2019")
  m
}

plot_all <- function(crimedata){
  all_type <- unique(crimedata[["TYPE"]])
  g_list <- lapply(all_type, plot_one, crime = crimedata)
  lapply(g_list, print)
  g_list
}

g_all <- plot_all(crime)