Filtering row which contains a certain string using dplyr

Question

I have to filter a data frame using as criterion those row in which is contained the string RTB. I'm using dplyr.

d.del <- df %.%

  group_by(TrackingPixel) %.%

  summarise(MonthDelivery = as.integer(sum(Revenue))) %.%

  arrange(desc(MonthDelivery))

I know I can use the function filter in dplyr but I don't exactly how to tell it to check for the content of a string.

In particular, I want to check the content in the column TrackingPixel. If the string contains the label RTB I want to remove the row from the result.

Answer 1

To filter rows that contain a string with a specific label, you can use the grepl() function that is used to match a pattern, inside the filter function.

The basic syntax is given below:

grepl(pattern, x, ignore.case = FALSE, perl = FALSE,

      fixed = FALSE, useBytes = FALSE)

In your case, use the following:

library(dplyr)

filter(df, !grepl("RTB",TrackingPixel))

For example:

To filter rows containing “Mazda” in a string in mtcars data set:

data(mtcars)

mtcars$names <- rownames(mtcars)

filter(mtcars, grepl('Mazda',names))

Output:

  mpg cyl disp hp drat wt qsec vs am gear carb  names

1  21   6 160 110 3.9 2.620 16.46  0 1 4 4 Mazda RX4

2  21   6 160 110 3.9 2.875 17.02  0 1 4 4 Mazda RX4 Wag

Similarly, to not include rows containing “Mazda”, do the following:

filter(mtcars, !grepl('Mazda',names))