On Sun, May 6, 2012 at 3:42 PM, knavero <knav...@gmail.com> wrote: > Right, but it seems to me that the error being the NA's in the index field > are caused by the longer vector lengths of columns 4 and 5. I would think > that the EOF in the scanf() (assuming C is used for the source code) would > be called where the NA's begin in columns 1 and 2 since columns 3:5 are > nulled out. Does this sound like a possible case? > > So, if the read in data only contained columns 1 and 2, it wouldn't even > look at columns 3:5 and thus, rows 14 and so on wouldn't even be looked at > and that would be EOF already - resulting in no error. >
Don't know what "longer vector lengths" refers to but every line in your pastebin data has 5 fields -- they don't vary. > range(count.fields(URL, sep = ",")) [1] 5 5 Furthermore, the error message seems pretty clear. Its saying that the index has a bad entry and is even telling which row or rows it occurrs at. Here is another smaller example where the missing entry in row 3 triggers the same sort of message: > read.zoo(text = "1,2\n2,3\n,4\n6,7", sep = ",") Error in read.zoo(text = "1,2\n2,3\n,4\n6,7", sep = ",") : index has bad entry at data row 3 -- Statistics & Software Consulting GKX Group, GKX Associates Inc. tel: 1-877-GKX-GROUP email: ggrothendieck at gmail.com ______________________________________________ R-help@r-project.org 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.
