How to use function to add geom_rect objects to ggplot from a data frame-CodePudding

I'm attempting to create a composite plot in r, the code for which is below:

    #Adding initial data
ggp <- ggplot(NULL, aes(x = date, y = covid))   
  geom_spline(data = onsdf, 
              aes(x = date, y = covid, colour = "ONS Modelled Estimates"), nknots = 90, size = 1.3)  
  geom_spline(data = gvtdf, 
              aes(x = date, y = covid, colour = "Gvt Reported Positive Tests"), nknots = 90, size = 1.3) 

#Creating function to add stringency bars
barfunction <- function(date1, date2, alpha){
  a <- annotate(geom = "rect",
                xmin = as.Date(date1), xmax = as.Date(date2), ymin = 0, ymax = Inf, alpha = alpha, fill = "red")
  return(a)
}

#Adding lockdown stringency bars
ggp <- ggp   
  barfunction("2020-05-03", "2020-06-01", 0.5)  
  barfunction("2020-06-01", "2020-06-15", 0.4)  
  barfunction("2020-06-15", "2020-09-14", 0.3)  
  barfunction("2020-09-14", "2020-11-05", 0.3)  
  barfunction("2020-11-05", "2020-12-02", 0.5)  
  barfunction("2020-12-02", "2021-01-06", 0.4)  
  barfunction("2021-01-06", "2021-03-29", 0.5)  
  barfunction("2021-03-29", "2021-04-12", 0.4)  
  barfunction("2021-04-12", "2021-05-17", 0.3)  
  barfunction("2021-05-17", "2021-07-19", 0.2)  
  barfunction("2021-07-19", "2021-12-08", 0.1)  
  barfunction("2021-12-08", "2022-02-24", 0.2)  


#Adding plot labels
ggp <- ggp   labs(title = "Estimated Total Covid-19 Cases vs Reported Positive Cases", 
       subtitle = "From ONS and HMGvt datasets", 
       x = "Date (year - month)", y = "Covid Levels")  
  scale_y_continuous(labels = scales::comma)  
  scale_x_date(limits = as.Date(c("2020-05-03", NA )))  
  scale_colour_manual(name = "Measurement Method",
                      values = c("ONS Modelled Estimates"="purple", 
                                 "Gvt Reported Positive Tests" = "blue"))

The output of this code looks like this:

Rendered graph

As you can see, I have a very repetitive function (barfunction) in this code that I would like to change. I thought the best way to do this was to convert the data barfunction() was applying to the graph into a dataframe, and then try to use a function on said data frame. Here is a head of the data frame (called strindf)

date1       date2       alpha
2020-05-03  2020-06-01  0.5
2020-06-01  2020-06-15  0.4
2020-06-15  2020-09-14  0.3
2020-09-14  2020-11-05  0.3

I initially tried to use apply() to add the strindf data to my plot, however I got an error message (Error in as.Date(date2) : argument "date2" is missing, with no default). Here is how I implemented it into the original code

ggptest <- ggplot(NULL, aes(x = date, y = covid))   
  geom_spline(data = onsdf, 
              aes(x = date, y = covid, colour = "ONS Modelled Estimates"), nknots = 90, size = 1.3)  
  geom_spline(data = gvtdf, 
              aes(x = date, y = covid, colour = "Gvt Reported Positive Tests"), nknots = 90, size = 1.3)  
  apply(strindf, MARGIN = 1 , barfunction)  
  theme_minimal()  
  scale_y_continuous(labels = scales::comma)  
  scale_x_date(limits = as.Date(c("2020-05-03", NA )))  
  scale_colour_manual(name = "Legend",

I'm quite new to r so I'm a bit stumped, does anyone have any suggestions?

Thanks in advance!

CodePudding user response：

Your idea was right. But you have chosen the wrong function from the apply family of functions. As you have a function of multiple arguments use mapply or as I do below purrr::pmap:

Using some fake random example data:


library(ggplot2)
library(ggformula)

barfunction <- function(date1, date2, alpha) {
  annotate(geom = "rect", xmin = as.Date(date1), xmax = as.Date(date2), ymin = 0, ymax = Inf, alpha = alpha, fill = "red")
}

ggplot(NULL, aes(x = date, y = covid))  
  geom_spline(data = df, aes(colour = "ONS Modelled Estimates"), nknots = 90, size = 1.3)  
  purrr::pmap(strindf, barfunction)  
  theme_minimal()  
  scale_y_continuous(labels = scales::comma)  
  scale_x_date(limits = as.Date(c("2020-05-03", NA)))  
  scale_colour_manual(
    name = "Measurement Method",
    values = c(
      "ONS Modelled Estimates" = "purple",
      "Gvt Reported Positive Tests" = "blue"
    )
  )
#> Warning: Removed 123 rows containing non-finite values (stat_spline).

DATA

set.seed(123)

df <- data.frame(
  date = seq.Date(as.Date("2020-01-01"), as.Date("2020-12-31"), by = "day"),
  covid = runif(366)
)

strindf <- structure(list(date1 = c(
  "2020-05-03", "2020-06-01", "2020-06-15",
  "2020-09-14"
), date2 = c(
  "2020-06-01", "2020-06-15", "2020-09-14",
  "2020-11-05"
), alpha = c(0.5, 0.4, 0.3, 0.3)), class = "data.frame", row.names = c(
  NA,
  -4L
))