fwrite: fix integer-backed POSIXct writing epoch instead of ISO #3535

[rmwood0908]

Closes #3535

Problem
Previously, the fwrite() function would write POSIXct datetime columns
as epoch timestamps instead of the default ISO format when the column was
created with seq():

library(data.table)
Sys.setenv(TZ = 'UTC')
tmp <- tempfile()

# epoch instead of ISO
dt <- data.table(x = seq(as.POSIXct('1970-01-01'), by = '1 sec', length = 6))
fwrite(dt, tmp)
cat(readLines(tmp), sep = '\n')
#> x
#> 0
#> 1
#> 2
#> 3
#> 4
#> 5

# works correctly
dt[, x := x + 1]
fwrite(dt, tmp)
cat(readLines(tmp), sep = '\n')
#> x
#> 1970-01-01T00:00:01Z
#> 1970-01-01T00:00:02Z
#> 1970-01-01T00:00:03Z
#> 1970-01-01T00:00:04Z
#> 1970-01-01T00:00:05Z
#> 1970-01-01T00:00:06Z

This happens because seq() creates a POSIXct column backed by integers instead of doubles:

dt <- data.table(x = seq(as.POSIXct("1970-01-01"), by = "1 sec", length = 6))
typeof(dt$x)
# [1] "integer"

Fix

Two changes were required:

1. src/fwriteR.c — Added a POSIXct check so integer-backed datetime columns are routed to the correct writer instead of being treated as plain integers.

2. src/fwrite.c and src/fwrite.h — Added a new writePOSIXctInt function that reads integer values, converts them to double, then formats them as datetime strings.

Test Results

After applying the full fix, the output is now as desired

x
1970-01-01T00:00:00Z
1970-01-01T00:00:01Z
1970-01-01T00:00:02Z
1970-01-01T00:00:03Z
1970-01-01T00:00:04Z
1970-01-01T00:00:05Z

Test 2368.1 has been added to inst/tests/tests.Rraw and a NEWS entry has been added as per contributing guidelines

[aitap]

This is a good start, thank you.

[aitap]

No need to adjust TZ, just give the tz = 'UTC' argument to as.POSIXct().

[aitap]

If you don't give an output argument to fwrite(), it prints to the standard output. You can then test the value of capture.output(fwrite(...)) without going through a temporary file (which you won't have to remember to unlink()).

data.table:::test also has an output argument, but those are skipped when running with a non-English locale, and this test must succeed no matter whether or not message translation is enabled.

[aitap]

You can give &x straight to writePOSIXct() without going through a temporary variable. C allows this implicit conversion because the destination type is void *.

[aitap]

What do you think about renaming this function to writePOSIXctDouble?