Question

In SQL, you can limit a query result with a WHERE EXISTS qualifier. For example, this:

SELECT *
FROM Table_B
WHERE EXISTS (SELECT ID FROM TABLE_A)

Can you do the same thing in R?

I’ve got a large dataframe – Call it df.a.

str(df.a)
ClientID : Factor w/ 148892 levels
Month    : Date, format  "2012-07-31"
Visits   : int

I’ve subsetted df.a to a smaller df.b. Now, I want to go back to df.a but only get records if the clientID is in df.b.

I was able to use merge, like this, but I was hoping for a more elegant solution.

july.activity <- merge(df.a, df.b, by="ClientID", all.x=TRUE)
july.activity <- subset(july.activity, Visits.x != "NA")

Answer 1

Maybe this does what you need?

july.activity <- df.a[df.a$ClientID %in% df.b$ClientID & !is.na(df.a$Visits), ]

df.a <- data.frame(ClientID = factor(letters[c(1,2,2,3)]), Visits = c(1,NA,2,3))
df.b <- df.a[1:2,]
df.b
  ClientID Visits
1        a      1
2        b     NA
df.a
  ClientID Visits
1        a      1
2        b     NA
3        b      2
4        c      3
df.a[df.a$ClientID %in% df.b$ClientID & !is.na(df.a$Visits), ]
  ClientID Visits
1        a      1
3        b      2