CODEWITHSUNDEEP
rsubset
La Machine Infernale
La Machine Infernale

Reputation: 569

R delete columns based on character contained in the column name

how do I delete columns based on a condition, when the condition is a character contained in the column name? For instance, following a series of inner joins I get a lot of redundant columns:

client.x, client.y, age.y. age.x
  132        132       23    23
  112        112       12    12

I would like a one liner to remove all columns containing for instance the ".x", to end up with:

client.y, age.y. 
  132       23
  112       12

Thanks

Upvotes: 1

Views: 374

Answers (1)

akrun
akrun

Reputation: 887028

We can use grep to identify the column names that have .x and with the use of invert=TRUE, it returns the index of other columns

df1[grep("\\.x", names(df1), invert=TRUE)]
#    client.y age.y.
#1      132     23
#2      112     12

Upvotes: 1

Related Questions