Hi,

with read.fwf, it works. 


But I still don't understand why it doesn't work with read.table since the sep 
by default is "", which is the case and in one trial, I used  
read.table("myfile",colClasses = "character", stringsAsFactors=FALSE, and stil 
didn't work but it should have.

Regards,



On Thursday, June 26, 2014 9:59 AM, Ron Crump <[email protected]> wrote:
 


Hi Carol,
> It might be a primitive question but I have a file of text and there is no 
> separator between character on each line and the strings on each line have 
> the same length. The format is like the following
>
> absfjdslf
> jfdldskjff
> jfsldfjslk
>
> When I read the file with read.table("myfile",colClasses = "character"), 
> instead of putting the strings in a table of number of rows x length of 
> string, read.table saves the file in a table of number of rows x 1 and each 
> element seems to be a factor. Why does read.table not account for  colClasses 
> = "character"?
read.table relies on a separator to differentiate between columns, so it 
is not appropriate for your file, read.fwf would do the job.

Setting colClasses (in my understanding) tells read.table how to treat 
input as it comes in - so it disables some testing of data types and 
makes reading quicker, it does not disable the setting of character data 
to be factors, which is the default. You need to use the 
stringsAsFactors=FALSE option for that.

So, for your example (and I have added a letter to the first row to make 
it the same length as the others):

cf <- "absfjdslfx

jfdldskjff
jfsldfjslk"

cdf <- 
read.fwf(textConnection(cf),widths=rep(1,10),colClasses="character",stringsAsFactors=FALSE)

See ?read.fwf for more information. A width is required for each column 
(in this case 1 repeated 10 times).

Hope this helps.

Ron.
        [[alternative HTML version deleted]]

______________________________________________
[email protected] 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.

Reply via email to