how to create a simple query to filter a data frame for different columns using part of column name

Right now I use this line of code:

df <- df1 %>% dplyr:: select(grep("x", names(df1)), grep("y", names(df1)), grep("z", names(df1)))

I cannot figure out a way to do this without repeating the grep. It would be nice to feed grep a list or to create a peace of code where one does not have to copy and adjust the grep line. Any better ways to filter a data frame on column names containing part of a string?

>Solution :

You can use matches inside select, with a regular expression containing potential matches separated by |.

This means your existing code should be equivalent to

df <- df1 %>% select(matches('x|y|z'))

For example if df1 was this:

df1 <- data.frame(box = 1, yak = 2, lazy = 3, lab = 4, men = 5)

Then we can select columns whose names contain x, y or z by doing:

df1 %>%
  select(matches('x|y|z'))
#>   box yak lazy
#> 1   1   2    3

^{Created on 2023-09-19 with reprex v2.0.2}