使用sqldf可以将不同data.frame中列的行子集

Question

I'm translating some data.frame code into SQL by using sqldf. My goal here is to subset rows of a data.frame A using a column from B. Is this possible when A and B don't share any column names?

A = data.frame(a1 = c(1:4), a2 = c(101:104))
B = data.frame(b1 = c(1:2), b2 = c(55,56))

A[A$a1 %in% B$b1,]

##   a1  a2
## 1  1 101
## 2  2 102

I can subset A if I already know the values from B$b1, but that's not very scalable.

sqldf("select * from A where a1 in (1,2)")

Do I need an inner join and/or is it required to have identical column names?

Answer 1

We use paste twice. To concatenate the elements of the vector B$b1 separated by commas. And then to concatenate the final text string desired: [1] "select * from A where a1 in( 1,2 )"

sqldf(paste("select * from A where a1 in(", paste(B$b1, collapse = ","), ")"))

Output:

  a1  a2
1  1 101
2  2 102

Answer 2

Try this:

fn$sqldf(" select * from A where a1 in ( `toString(B$b1)` ) ")

or

sqldf("select A.* from A join B on A.a1 = B.b1")