Hi,
Consider the row.names of
a <- data.frame(x = runif(150000))
> row.names(a)[99998:100002]
[1] "99998" "99999" "1e+05" "100001" "100002"
not wrong, but there's room for improvement. If we replace the
expression
if (length(row.names) == 0)
row.names <- seq(length = nr)
which creates a non-integer object row.names by
if (length(row.names) == 0)
row.names <- if(nr>0) 1:nr else integer(0)
(or something to that effect) to create an integer row.names
that when coerced to character will produce
b <- data.frame(x = runif(150000))
> row.names(b)[99998:100002]
[1] "99998" "99999" "100000" "100001" "100002"
These labels have a more uniform format.
Could a change like this be made to data.frame()?
Regards,
--
David
> version
_
platform i686-pc-linux-gnu
arch i686
os linux-gnu
system i686, linux-gnu
status
major 1
minor 8.1
year 2003
month 11
day 21
language R
______________________________________________
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-devel
