different quoting behavior, depending on whitespace

[ijlyttle]

Sometimes I will write out a small csv by hand - by habit I will put a space after a comma:

"foo", "bar"
"a", "b"

When I read_csv() this, it parses inconsistently.

However, if I take out the whitespace, it parses as I expect:

"foo","bar"
"a","b"

Reprex:

library("readr")

read_csv('"foo","bar"\n"a","b"\n')
#> # A tibble: 1 × 2
#>     foo   bar
#>   <chr> <chr>
#> 1     a     b

read_csv('"foo", "bar"\n"a", "b"\n')
#> # A tibble: 1 × 2
#>     foo `"bar"`
#>   <chr>   <chr>
#> 1     a     "b"

Does this rise to the level of bug? Also, may be similar to #238.

[jennybc]

I'm running up against this too. It's especially damaging if you're using the quotes to protect embedded commas. Also: something is off with the printing of the problems?

library(readr)
read.csv(text = 'x,y\n1,\"hi,there\"\n3,4')
#>   x        y
#> 1 1 hi,there
#> 2 3        4
read_csv('x,y\n1,\"hi,there\"\n3,4')
#> # A tibble: 2 x 2
#>       x        y
#>   <int>    <chr>
#> 1     1 hi,there
#> 2     3        4
  
read.csv(text = 'x,y\n1, \"hi,there\"\n3,4')
#>   x         y
#> 1 1  hi,there
#> 2 3         4
read_csv('x,y\n1, \"hi,there\"\n3,4')
#> Warning in rbind(names(probs), probs_f): number of columns of result is not
#> a multiple of vector length (arg 2)
#> Warning: 1 parsing failure.
#> row # A tibble: 1 x 5 col     row   col  expected    actual         file expected   <int> <chr>     <chr>     <chr>        <chr> actual 1     1  <NA> 2 columns 3 columns literal data file # A tibble: 1 x 5
#> # A tibble: 2 x 2
#>       x      y
#>   <int>  <chr>
#> 1     1 "\"hi"
#> 2     3      4

[ijlyttle]

@jimhester Thanks!

[lock]

This old issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with reprex) and link to this issue. https://reprex.tidyverse.org/