Closes Rdatatable#2025

HughParsonage · Jan 16, 2018 · a4b6c35 · a4b6c35
1 parent 1154851
commit a4b6c35
Show file tree

Hide file tree

Showing 2 changed files with 11 additions and 2 deletions.
diff --git a/R/fread.R b/R/fread.R
@@ -16,7 +16,7 @@ fread <- function(input="",file,sep="auto",sep2="auto",dec=".",quote="\"",nrows=
   isTrueFalse = function(x) isTRUE(x) || identical(FALSE, x)
   isTrueFalseNA = function(x) isTRUE(x) || identical(FALSE, x) || identical(NA, x)
   stopifnot( isTrueFalse(strip.white), isTrueFalse(blank.lines.skip), isTrueFalse(fill), isTrueFalse(showProgress),
-             isTrueFalse(stringsAsFactors), isTrueFalse(verbose), isTrueFalse(check.names), isTrueFalse(logical01) )
+             isTrueFalse(stringsAsFactors) || is.double(stringsAsFactors), isTrueFalse(verbose), isTrueFalse(check.names), isTrueFalse(logical01) )
   stopifnot( is.numeric(nrows), length(nrows)==1L )
   if (is.na(nrows) || nrows<0) nrows=Inf   # accept -1 to mean Inf, as read.table does
   if (identical(header,"auto")) header=NA
@@ -127,7 +127,7 @@ fread <- function(input="",file,sep="auto",sep2="auto",dec=".",quote="\"",nrows=
   if (stringsAsFactors) {
     # Re Issue 2025
     if (is.double(stringsAsFactors)) {
-      should_be_factor <- function(v) is.character(v) && uniqueN(v) < nr / stringsAsFactors
+      should_be_factor <- function(v) is.character(v) && uniqueN(v) < nr * stringsAsFactors
       cols_to_factor <- which(vapply(ans, should_be_factor, logical(1L)))
       if (verbose) {
         cat("stringsAsFactors=", stringsAsFactors, ", interpreting as the minimum number",

diff --git a/inst/tests/tests.Rraw b/inst/tests/tests.Rraw
@@ -10181,6 +10181,15 @@ test(1743.307, names(fread(data1743.307, colClasses=list(NULL = c("C", "D"), NUL
 test(1743.31, names(fread(data1743, colClasses="NULL")), LETTERS[1:4], warning="colClasses.*quoted.*interpreted as colClasses.*NULL")
 test(1743.32, sapply(fread("A,B\na,0+1i", colClasses="complex"), class), c(A="character", B="complex"), warning="Column 1 was set by colClasses to be .complex")
 
+## stringsAsFactors = double Issue #2025
+fwrite(data.table(v1 = sample(letters, size = 26 * 10, replace = TRUE), V2 = paste0("A", c(1:259, 1)), V3 = sample.int(26, n = 260, replace = TRUE)), temp1743_41.csv <- tempfile(fileext = ".csv"))
+test(1743.411, sapply(fread(temp1743_41.csv), class)[[1]], "character")
+test(1743.412, sapply(fread(temp1743_41.csv, stringsAsFactors = 0), class)[[1]], "character")
+test(1743.413, sapply(fread(temp1743_41.csv, stringsAsFactors = 0.10), class)[[1]], "character")
+test(1743.414, sapply(fread(temp1743_41.csv, stringsAsFactors = 0.11), class)[[1]], "factor")
+test(1743.415, sapply(fread(temp1743_41.csv, stringsAsFactors = 0.12), class)[[1]], "factor")
+test(1743.416, sapply(fread(temp1743_41.csv, stringsAsFactors = 1.00), class)[[2]], "factor")
+invisible(file.remove(temp1743_41.csv))
 
 # rolling join stopped working for double with fractions, #1904
 DT = data.table(A=c(1999.917,2000.417,2000.917,2001.417,2001.917))