On 6/27/07, ivo welch <[EMAIL PROTECTED]> wrote: > Dear R experts---I would like to write a replacement for the read.csv > function that is less general, but also more efficient. > > could someone please provide me with a skeleton function that shows me > how to read the arguments and return a data frame for a call to a C > function that handles > > returned.data.frame = my.read.csv(file, header = TRUE, sep = ",", > quote="\"", dec=".", > fill = TRUE, comment.char="", ...) > > this may be very difficult, of course, in which case writing such a > function would not be worth it. I guess I would be happy just to > learn how to return a basic data frame that holds data vectors that > are either strings or numbers---nothing more complex. > > help appreciated.
Should we assume that you have already read the relevant sections of the manual "Writing R Extensions"? The .Call interface is the easiest way to return an object like a data frame. I might pass the result back as a list and use something like do.call(data.frame, .Call("my_csv_reader", file, ...)) rather than duplicating all the error checking that is done in the data.frame function. ______________________________________________ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.