Creating Functions From Arbitrary R Code?

Question

In R, I have an input dataset ("my_data"). Using this dataset, I made some plots and saved them like this:

library(htmltools)
library(plotly)

fig1 =  plot_ly(my_data, x =~ var1, y = ~var2, type = 'bar', name = 'Plot 1')
fig2 =  plot_ly(my_data, x =~ var1, y = ~var3, type = 'bar', name = 'Plot 2')
fig3 =  plot_ly(my_data, x =~ var1, y = ~var4, type = 'bar', name = 'Plot 3')
fig4 =  plot_ly(my_data, x =~ var1, y = ~var5, type = 'bar', name = 'Plot 4')

# final result
doc <- htmltools::tagList(
    div(fig1, style = "float:left;width:50%;"),
    div(fig2,style = "float:left;width:50%;"),
    div(fig3, style = "float:left;width:50%;"),
    div(fig4, style = "float:left;width:50%;")
)

# save the final result
htmltools::save_html(html = doc, file = "final.html")

Suppose now I have a new dataset called "my_data_2" which has the exact same format as "my_data". Is it somehow possible to transform the above code into a function so that the exact same process is executed on "my_data_2"?

For example:

some_function <- function (my_data) {


library(htmltools)
library(plotly)

fig1 =  plot_ly(my_data, x =~ var1, y = ~var2, type = 'bar', name = 'Plot 1')
fig2 =  plot_ly(my_data, x =~ var1, y = ~var3, type = 'bar', name = 'Plot 2')
fig3 =  plot_ly(my_data, x =~ var1, y = ~var4, type = 'bar', name = 'Plot 3')
fig4 =  plot_ly(my_data, x =~ var1, y = ~var5, type = 'bar', name = 'Plot 4')

# final result
doc <- htmltools::tagList(
    div(fig1, style = "float:left;width:50%;"),
    div(fig2,style = "float:left;width:50%;"),
    div(fig3, style = "float:left;width:50%;"),
    div(fig4, style = "float:left;width:50%;")
)

# save the final result
htmltools::save_html(html = doc, file = "final.html")

}

Then, when calling this function:

some_function(my_data_2)

Would end up producing the same results for this new file?

Thank you!

Kat · Accepted Answer · 2022-11-07 19:25:59Z

1

Assuming the columns have the same names, this would work.

some_function <- function(df1) {
  plots <- map(1:4, 
               function(k) {        # create the graphs
                 plot_ly(df1, x = ~var1, y = df1[, paste0("var", k + 1)],
                         type = "bar", name = paste0("Plot", k))
               })
  sty <- "float:left;width:50%;"    # create the style
  doc <- tagList(map(1:4,           # combine style & graphs
                     function(j) {
                       div(plots[j], style = sty)
                     }))
  save_html(html = doc, file = "final.html")
}

If you wanted to preview this before creating the graphs, you could do so like this.

library(plotly)
library(htmltools)
library(tidyverse) # for map in purrr 

some_function <- function(df1) {
  plots <- map(1:4, 
               function(k) {        # create the graphs
                 plot_ly(df1, x = ~var1, y = df1[, paste0("var", k + 1)],
                         type = "bar", name = paste0("Plot", k))
               })
  sty <- "float:left;width:50%;"    # create the style
  doc <- tagList(map(1:4,           # combine style & graphs
                     function(j) {
                       div(plots[j], style = sty)
                     }))
  # save_html(html = doc, file = "final.html") 
  doc
}
# arbitrary data to demonstrate with
df1 <- data.frame(var1 = rep(c("this", "that"), 5),
                  var2 = 1:10,
                  var3 = 10:1, 
                  var4 = 15:6,
                  var5 = 6:15)

html_print(some_function(df1))     # show in viewer pane

This is produced from that last call.

answered Nov 7, 2022 at 19:25

Kat

19k3 gold badges24 silver badges63 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

stats_noob Over a year ago

Thank you so much Kat! Is it possible to keep the original code I had written? Would my approach also work?

Kat Over a year ago

Yes, your original code would work.

Collectives™ on Stack Overflow

Creating Functions From Arbitrary R Code?

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related