如何通过使用正则表达式从向量列中提取数字来创建新列？

Question

I need to extract numbers from a vector column using regex and create columns for each number. The length of the vector will not be the same for each row.

For other purposes such as counting the number of elements within the vector I have used

str_count(DATA$vectorCOL, '[0-9.+]+')

This is the data column

vectorCOL
63.
11., 36., 45+1., 79., 90+1.

45., 80., 87.

Expected output

vectorCOL                    col1  col2 col3  col4 col5
63.                          63    NA   NA    NA   NA
11., 36., 45+1., 79., 90+1.  11    36   45+1  79   90+1
                             NA    NA   NA    NA   NA
45., 80., 87.                45    80   87    NA   NA

Answer 1

We can use cSplit

splitstackshape::cSplit(df, "vectorCOL", sep = ",", drop = FALSE)
#                vectorCOL vectorCOL_1 vectorCOL_2 vectorCOL_3 vectorCOL_4 vectorCOL_5
#1:                     63.          63          NA        <NA>          NA        <NA>
#2: 11.,36.,45+1.,79.,90+1.          11          36       45+1.          79       90+1.
#3:                                  NA          NA        <NA>          NA        <NA>
#4:             45.,80.,87.          45          80         87.          NA        <NA>

If we don't want "." in the output, we can remove them first using gsub .

df$vectorCOL <- gsub("\\.", "",df$vectorCOL)

data

df <-  structure(list(vectorCOL = c("63.", "11., 36., 45+1., 79., 90+1.", 
"", "45., 80., 87.")), row.names = c(NA, -4L), class = "data.frame")

Answer 2

Using data.table :

df <- df[, c(vectorCOL = list(vectorCOL), tstrsplit(vectorCOL, ","))]
setnames(df, names(df), sub("V", "col", names(df)))
df
#                      vectorCOL col2 col3   col4 col5   col6
# 1:                         63.  63. <NA>   <NA> <NA>   <NA>
# 2: 11., 36., 45+1., 79., 90+1.  11.  36.  45+1.  79.  90+1.
# 3:                             <NA> <NA>   <NA> <NA>   <NA>
# 4:               45., 80., 87.  45.  80.    87. <NA>   <NA>