Change table to character vector without spaces

Question

Let's have a following datafrane:

df <- data.frame(c("a", "c ","e "), c("b ", "d", "f "))
colnames(df) <- c("X1", "X2")
df
  X1 X2
1  a b 
2 c   d
3 e  f

How to transform the given data.frame to one character vector without spaces?

#Expected result:
#vector
#[1]"abcdef"

Related post: https://stackoverflow.com/questions/2545228/converting-a-dataframe-to-a-vector-by-rows — zx8754, Feb 19 '19 at 15:00

score 5 · Accepted Answer · answered Feb 19 '19 at 14:55

5

Another solution from base R.

paste(trimws(as.vector(t(df))), collapse = "")
# [1] "abcdef"

answered Feb 19 '19 at 14:55

www

score 3 · Answer 2 · answered Feb 19 '19 at 14:56

3

Transpose df, unravel it into a vector, trim the spaces and paste the components together:

paste(trimws(c(t(df))), collapse = "")
## [1] "abcdef"

answered Feb 19 '19 at 14:56

G. Grothendieck

`c` is more efficient than `as.vector` in my solution. Thanks for sharing your solution – www Feb 19 '19 at 14:57
The documentation cautions against using `c` over `as.vector`. I agree, since it’s hard to read. `as.vector` makes the intentions clearer. – Konrad Rudolph Feb 19 '19 at 14:58
It claims as.vector is more intuitive. It does not warn against using it. Note that some of the recommendations in the documentation are not the best. For example, intidyverse would never have been developed in its current form had it followed recommendations in the documentation. – G. Grothendieck Feb 19 '19 at 15:24

score 3 · Answer 3 · answered Feb 19 '19 at 14:56

3

An idea using gsub to remove unwanted spaces,

gsub(' ', '', paste(unlist(t(df)), collapse =  ''))
#[1] "abcdef"

answered Feb 19 '19 at 14:56

Sotos

acylam · Answer 4 · 2019-02-19T15:11:13.793

3

As noted in the comments, t already converts df to a matrix, so the following works:

paste(trimws(t(df)), collapse = "")
# [1] "abcdef"

edited Feb 19 '19 at 15:11

answered Feb 19 '19 at 15:06

acylam

`t(df)` is a matrix, so I am not sure why `as.matrix` is requried – akrun Feb 19 '19 at 15:09
Actually, `t(df)` is already a matrix, which means `paste(trimws(t(df)), collapse = "") ` works! – www Feb 19 '19 at 15:09
1

Good point! I love how each answer brings some improvements to the table. – acylam Feb 19 '19 at 15:12
@avid_useR This is the most concise and efficient answer so far. Good work. – www Feb 19 '19 at 15:14

akrun · Answer 5 · 2019-02-19T16:27:34.817

2

One option is paste after removing the leading/lagging spaces with trimws

paste(do.call(paste0, lapply(df, trimws)), collapse="")
#[1] "abcdef"

Or using tidyverse

library(tidyverse)
df %>% 
  mutate_all(str_squish) %>% 
  pmap_chr(paste, sep="") %>%
  paste(., collapse="")
#[1] "abcdef"

edited Feb 19 '19 at 16:27

answered Feb 19 '19 at 14:52

akrun

5 Answers5