R help - Mar 2006 - Using read.table() with Excel data

Help

 

I apologize if this message posts twice, but I do not think my last
attempt went through.

 

I am trying to use the read.table() function to read in a massive
alphanumeric data set.  I thought I might scan in each column
individually and do a cbind() to reconstitute the array in R, so that
maybe I could handle the numbers and the long lines of texts
separately(some columns in the Excel dataset are wrapped lines of texts,
some dates, some numbers).  However, I encountered a problem in that I
cannot seem to even read the column of texts properly.  Each line of
texts is fairly long and there are about 7000 lines.  When I try to read
the entire 7000 lines, either R will say over 50 warnings() and then
quite reading in the document or it will appear as though it
successfully read the data, but then when I look at the dimension of the
data, it is only 4000, and it will not print it to the screen.  When I
take a small chunk of the text, say 40 lines, it prints to the screen,
but much of the data is missing; usually only one true line of data
shows up and the rest are NA's.  I was just wondering if I am going
about reading this data set improperly?  Am I using the wrong function?
Thanks a million your help is much appreciated.

 

Best,

David

 



--------------------

This email message is for the sole use of the intended recip...{{dropped}}

On 3/1/06 11:31 AM, "Nealy, David" <dnealy at rand.org> wrote:
> Help
> 
>  
> 
> I apologize if this message posts twice, but I do not think my last
> attempt went through.
> 
>  
> 
> I am trying to use the read.table() function to read in a massive
> alphanumeric data set.  I thought I might scan in each column
> individually and do a cbind() to reconstitute the array in R, so that
> maybe I could handle the numbers and the long lines of texts
> separately(some columns in the Excel dataset are wrapped lines of texts,
> some dates, some numbers).  However, I encountered a problem in that I
> cannot seem to even read the column of texts properly.  Each line of
> texts is fairly long and there are about 7000 lines.  When I try to read
> the entire 7000 lines, either R will say over 50 warnings() and then
> quite reading in the document or it will appear as though it
> successfully read the data, but then when I look at the dimension of the
> data, it is only 4000, and it will not print it to the screen.  When I
> take a small chunk of the text, say 40 lines, it prints to the screen,
> but much of the data is missing; usually only one true line of data
> shows up and the rest are NA's.  I was just wondering if I am going
> about reading this data set improperly?  Am I using the wrong function?
> Thanks a million your help is much appreciated.
Are you trying to read a .xls file, or is it tab-delimited text?

Assuming that it is text, make sure that there are no quote characters in
the file.  I sometimes find that setting quote="" in a call to
read.table
does what I need.  Finally, I'm not sure how much difference reading each
column separately will make, so you might try reading as one call to
read.table to see if that works.

Sean