R data.table select rows based on partial string match from character vector

Question

I have a character vector and a data.tabe:

library(data.table)
pselection <- c("12345" , "2345", "12345678")
dt <- data.table("title"=c("First title", "Second Title", "Third Title", "Fourth Title"), 
                 "sha"=c("12345", "2345; 66543; 33423", "22222; 12345678;", "666662345; 444"))

Now I want to select all rows of the data.table which match the sha column partially based on the ; seperator. So basically I want this output:

          title                sha
1:  First title              12345
2: Second Title 2345; 66543; 33423
3:  Third Title   22222; 12345678;

How would I do this?

I tried this:

selected <- dt[sha %in% pselection]

but it only selects exact matches and using the %like% expression is just for matching one expression not many. Concatenating to a regular expression (like paste(pselection, collapse="|")) is out of the question because my pselection is > 10.000.Thanks for the help in advance!

Do you need to use data.table or are you just looking for a solution for selecting partial string matches? — Matt
– Matt, Commented May 8, 2020 at 15:49
Would be nice to use data.table but any efficient enough solution is appreciated! — Blind0ne
– Blind0ne, Commented May 8, 2020 at 15:54

linog · Accepted Answer · 2020-05-08 15:55:27Z

1

I have a solution in mind using lapply and tstrsplit. There's probably more elegant but it does the job

lapply(1:nrow(dt), function(i) {
  dt[i,'match' := any(trimws(tstrsplit(as.character(dt[i,'sha']),";")) %in% pselection)]
  })

dt[(match)]
          title                sha match
1:  First title              12345  TRUE
2: Second Title 2345; 66543; 33423  TRUE
3:  Third Title   22222; 12345678;  TRUE

The idea is to split every row of sha column (trim whitespace otherwise row 3 will not match) and check if any sha appears

answered May 8, 2020 at 15:55

linog

6,2963 gold badges17 silver badges32 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

s_baldur · Accepted Answer · 2020-05-08 16:08:56Z

1

Using regex:

pselection <- paste0("\\b", pselection) # \b is boundary and includes ; and whitespace
dt[grepl(paste(pselection, collapse = "|"), sha)]

          title                sha
1:  First title              12345
2: Second Title 2345; 66543; 33423
3:  Third Title   22222; 12345678;

answered May 8, 2020 at 16:08

s_baldur

34.5k4 gold badges43 silver badges80 bronze badges

Collectives™ on Stack Overflow

R data.table select rows based on partial string match from character vector

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related