read.csv / read.table R 3.1.1类型转换错误

Question

Having updated to R version 3.1.1, read.table / read.csv incorporated in R code which ran under previous versions of R no longer works. My original code was simply:

read.csv("C:/................",skip = 2,as.is= TRUE)

I am getting the following error:

Error in .External2(C_typeconvert, x, na.strings, as.is, dec) :

Incorrect number of arguments (4), expecting 5 for 'typeconvert'

I have tried to incorporate all the options which are required by type.convert ie

type.convert(x, na.strings = "NA", as.is = FALSE, dec = ".")

but to no avail. In terms of data types I have a mix of character and numeric fields but, possilbly, the most striking elements of the data are:

1) Numeric fields have two missing indicators 'A' and 'NA'. Which is why I originally used the 'as.is' option and later used type.convert. For example

Mean SD
NA 0.30
NA 0.20
NA 0.30
NA 0.20
2.2 NR
1.9 NA
2.2 NA
1.9 NA
NR NR

2) The second is that some of my character fields incorporate full stops is 'A.B'.

Has something changed in this version of R?

Answer 1

I had the same problem after upgrading to version 3.1.1 from 3.1.0. The solution was simply to install R again.

I think the problem was that I was reckless when I migrated my installed packaged from the 3.1.0 library folder to the 3.1.1 library folder. I wrote over the standard packages such as base and utils (where the read.table function is) in the new version with the old version. When I reinstalled R it kept all the other packages I had installed.