Shift column names across by 10 columns

Question

I have the following code to import some data.

url <- "https://finance.yahoo.com/industry/Scientific_Technical_Instruments"

library(rvest)

read <- read_html(url) %>%
  html_table() 

library(plyr)

data <- ldply(read, data.frame)

However the data creates a data frame of 20 columns when there should be just 10. The column names of the data frame have not imported as they should and creates a number of NA values.

Is there a way in R to shift the column names across, then remove the NA columns created?

if you think that one of the replies helped you, could you approve one of them as the correct answer? — Otto Kässi, Oct 29 '18 at 04:33

score 1 · Answer 1 · answered Oct 27 '18 at 20:42

1

Your object read is a list with headers as the first element and data as the second. Your problem is that your column names in read[[1]] are not syntactically valid names for data frame columns.

You need to sanitise your names by using make.names. E.g.

data <- data.frame(read[[2]]) 
names(data) <- make.names(names(read[[1]])

An one-liner version for this can be found from here.

data <- setNames(data.frame(read[[2]]), make.names(names(read[[1]])))

answered Oct 27 '18 at 20:42

Otto Kässi

2,943
1
10
27

oh, I see what the problem is now! – user113156 Oct 27 '18 at 20:51
1

a mix of our approaches: `setNames(data.frame(read[[2]]), colnames(read[[1]]))` – G. Cocca Oct 27 '18 at 20:55

score 1 · Answer 2 · answered Oct 27 '18 at 20:47

1

my_data <- data.frame(read[[2]])
colnames(my_data) <- colnames(read[[1]])

answered Oct 27 '18 at 20:47

G. Cocca

2,456
1
12
13

Shift column names across by 10 columns

2 Answers2