Most efficient list to data.frame method?

Question

Just had a conversation with coworkers about this, and we thought it\'d be worth seeing what people out in SO land had to say. Suppose I had a list with N elements, where e

Answer 1

This appears to need a data.table suggestion given that efficiency for large datasets is required. Notably setattr sets by reference and does not copy

library(data.table)
set.seed(21)
n <- 1e6
h <- list(x=rnorm(n), y=rnorm(n), z=rnorm(n))
h <- c(h,h,h,h,h,h)
tracemem(h)

system.time({h <- as.data.table(h)
            setattr(h, 'names', make.names(names(h), unique=T))})

as.data.table, however does make a copy.

---

Edit - no copying version

Using @MatthewDowle's suggestion setattr(h,'class','data.frame') which will convert to data.frame by reference (no copies)

set.seed(21)
n <- 1e6
i <- list(x=rnorm(n), y=rnorm(n), z=rnorm(n))
i <- c(i,i,i,i,i,i)
tracemem(i)

system.time({  
  setattr(i, 'class', 'data.frame')
  setattr(i, "row.names", c(NA_integer_,n))

  setattr(i, "names", make.names(names(i), unique=TRUE))

})