Filter First,Middle and Last string word like using grepl function-CodePudding

Trying to filter the data of the first,middle and last string word like matching using grepl function, but it is also picking words like HEV along with DEV(intended match)

Airport_ID<-c("3001","3002","3003","3004")
Airport_Name<-c("DEV Adelaide DTSUpdated","HEV Brisbane HEV Land Airport Land ADTS",
                "DEVAST Washington INC Airport DTSUpdated","DALLAS DEVASTAirport HEV INCUpdated")
dfu<-data.frame(Airport_ID,Airport_Name)

Filter_Data_F <- dfu %>%  
               dplyr::filter(grepl("^DEV" , Airport_Name , fixed = F) |
                             grepl(" \\DEV\\ " , Airport_Name , fixed = F) |
                             grepl("DEV$" , Airport_Name , fixed = F) )

CodePudding user response：

\\D has a special meaning in regex. It matches any character that is not a digit character. So in the second condition it is matching a non-digit character (H) followed by EV, hence you get HEV in the output.

Secondly, grepl has by default fixed = FALSE so you can ignore that argument.

Also, I am not sure if you should write separate grepl arguments with |. Only one grepl should do it.

library(dplyr)
dfu %>%  dplyr::filter(grepl('DEV', Airport_Name))

#  Airport_ID                             Airport_Name
#1       3001                  DEV Adelaide DTSUpdated
#2       3003 DEVAST Washington INC Airport DTSUpdated
#3       3004      DALLAS DEVASTAirport HEV INCUpdated

If you want to exactly match DEV so DEVAST does not match, use word boundaries (\\b).

dfu %>%  dplyr::filter(grepl('\\bDEV\\b', Airport_Name))

#  Airport_ID            Airport_Name
#1       3001 DEV Adelaide DTSUpdated