[R] different outcomes using read.table vs read.csv

[Barry Rowlingson]

2009/3/13 jatwood <jatwood at montana.edu>:
> Good Afternoon
> I have noticed results similar to the following several times as I have used
> R over the past several years.
> My .csv file has a header row and 3073 rows of data.
>
>> rskreg<-read.table('D:/data/riskregions.csv',header=T,sep=",")
>> dim(rskreg)
> [1] 2722   13
>> rskreg<-read.csv('D:/data/riskregions.csv',header=T)
>> dim(rskreg)
> [1] 3073   13
>>
>
> Does someone know what could be causing the read.table and read.csv
> functions to give different results on some occasions?  The riskregions.csv
> file was generated with and saved from MS.Excel.

 read.table has 'comment.char="#"', so if a line starts with # it gets
ignored. read.csv doesn't have this set, so it might explain why
read.csv gets more than read.table...

 Do you have lines starting with #? Try read.table with
comment.char="" and see if you get the right number. See the help for
read.table for more info.

 I'd not seen this before, hope it hasn't bitten me...

Barry