CODEWITHSUNDEEP
rdataframegsub
LeMarque
LeMarque

Reputation: 783

failed without any error message while replacing part of names of columns with other names in dataframe

I have a dataframe which has column names as shown in the dput data below. 

structure(list(mosales = c(1L, 1L, 1L, 12L, 1L), sale123 = c(14.86, 
8.97, 6.44, 463.61, 15.94), totsales = c(1L, 1L, 1L, 
30L, 1L), totqty = c(1L, 1L, 1L, 34L, 2L), unqsales = c(1L, 
1L, 1L, 6L, 2L), x1_rank_1 = c(1L, 1L, 1L, 0L, 1L), x1_rank_4 = c(0L, 
0L, 0L, 1L, 0L), x1_rank_3 = c(0L, 0L, 0L, 0L, 0L), x1_rank_2 = c(0L, 
0L, 0L, 0L, 0L), x2_rank_2 = c(1L, 1L, 0L, 0L, 1L), x2_rank_1 = c(0L, 
0L, 1L, 0L, 0L), x2_rank_5 = c(0L, 0L, 0L, 1L, 0L), x2_rank_4 = c(0L, 
0L, 0L, 0L, 0L), x2_rank_3 = c(0L, 0L, 0L, 0L, 0L), x3_rank_1 = c(1L, 
1L, 1L, 0L, 1L), x3_rank_4 = c(0L, 0L, 0L, 1L, 0L), x3_rank_3 = c(0L, 
0L, 0L, 0L, 0L), x3_rank_2 = c(0L, 0L, 0L, 0L, 0L), x4_rank_1 = c(1L, 
1L, 1L, 0L, 0L), x4_rank_5 = c(0L, 0L, 0L, 1L, 0L), x4_rank_2 = c(0L, 
0L, 0L, 0L, 1L), x4_rank_4 = c(0L, 0L, 0L, 0L, 0L), x4_rank_3 = c(0L, 
0L, 0L, 0L, 0L), x5_rank_1 = c(1L, 1L, 1L, 0L, 0L), x5_rank_4 = c(0L, 
0L, 0L, 1L, 0L), x5_rank_2 = c(0L, 0L, 0L, 0L, 1L), x5_rank_3 = c(0L, 
0L, 0L, 0L, 0L)), row.names = c(36L, 41L, 72L, 79L, 137L), class = "data.frame")

What I am trying to do is to replace columns names, which start with x1_rank, x2_rank .. and so forth, by the names such that "x1_rank" should be replaced with mosales, "x2_rank" should be replaced with sales123, "x3_rank" should be replaced with "totsales", "x4_rank" should be replaced with "totqty" and "x5_rank" should be replaced with "unqsales".

So the final column names would look like:

mosales, sale123, totsales, totqty, unqsales, mosales_1, mosales_2,... sale123_1, sale123_2,... totsales_1, totsales_2,... totqty_1, totqty_2,... unqsales_1, unqsales_2,...

I tried using a for loop and gsub, as in below. This code ran without any errors but I didn't get what I was looking for. Not sure where is the error.

df1 <- df
z <- names(df)

for (i in 1:length(z)){
  gsub(paste0("x",i,"_rank"), z[i], names(df1))
}

df is dataframe which can be created using above dput data.

Upvotes: 0

Views: 39

Answers (2)

Ista
Ista

Reputation: 10437

String manipulation functions in the stringi package are vectorized over pattern, replacement, and (optionally) string. This is very convenient for your case:

library(stringi)

orig <- c(stri_c("x", 1:5, "_rank"))
repl <- c("mosales", "sales123", "totsales", "totqty", "unqsales")
names(df) <- stri_replace_all_fixed(names(df), orig, repl,
                                    vectorize_all = FALSE)

Upvotes: 2

RLave
RLave

Reputation: 8364

You can use this reg \\bx1_rank[a-zA-Z]*.

This will match x1_rank at the start of the string.

gsub lets you replace the found pattern with what you want. Repeat for all the cases you need.

In a loop:

repl <- c("mosales", "sales123", "totsales", "totqty", "unqsales")

for (i in 1:5) {
  p = paste0("\\b^x", i, "_rank[a-zA-Z]*")
  colnames(d) = gsub(pattern=p, repl[i], colnames(d))
}
colnames(d)

# 
# [1] "mosales"    "sale123"    "totsales"   "totqty"     "unqsales"   "mosales_1"  "mosales_4" 
# [8] "mosales_3"  "mosales_2"  "sales123_2" "sales123_1" "sales123_5" "sales123_4" "sales123_3"
# [15] "totsales_1" "totsales_4" "totsales_3" "totsales_2" "totqty_1"   "totqty_5"   "totqty_2"  
# [22] "totqty_4"   "totqty_3"   "unqsales_1" "unqsales_4" "unqsales_2" "unqsales_3"

Upvotes: 1

Related Questions