从R中向量中不同长度的字符串中提取字符值

Question

I have a vector that looks like the following:

**name**
a1
a2
a3
b1_z
b2_3z
b32z

I would like the output to only include the letters in each of these strings, not any numbers or symbols. Like this:

**name**
a
a
a
bz
bz
bz

I have tried using the following code:

df$name <- stri_extract_all_regex(df$name, "[a-z]+") 

I get this result:

**name**
a
a
a
c("b", "z")
c("b", "z")
c("b", "z") 

How do I combine the values that are two separate strings into a single string? In particular, how do I do this when some of the values in the vector already contain only one string? I am also open to other solutions for extracting characters from strings that get around this issue.

Answer 1

Please try gsub like below

df$name <- gsub("[^[:alpha:]]","",df$name)

where non-alphabet characters are replaced by "" .

We will get

> df
  name
1    a
2    a
3    a
4   bz
5   bz
6   bz

Data

> dput(df)
structure(list(name = c("a1", "a2", "a3", "b1_z", "b2_3z", "b32z"
)), class = "data.frame", row.names = c(NA, -6L))

Answer 2

You can do it like this using the gsub function:

vals = c('a1', 'a2', 'b1_z', 'b2_3z')
df = data.frame(vals)

df$name = gsub("[^[:alpha:]]", "", df$vals)
print(df)

Output will look like this:

  name
1    a
2    a
3   bz
4   bz

Answer 3

An option with str_remove

library(stringr)
str_remove_all(df$name, "[0-9_]+")
#[1] "a"  "a"  "a"  "bz" "bz" "bz"

data

df <- structure(list(name = c("a1", "a2", "a3", "b1_z", "b2_3z", "b32z"
)), class = "data.frame", row.names = c(NA, -6L))