Thanks for the suggestions. I have had several suggestions that may provide
a way forward:
1) Put dummy #N/A codes in Excel in the top RH corner of the array, or
insert a complete dummy row to remove later.
2) cleanup.import routine in Hmisc library
3) Use VBA in Excel to pass the data. Example in XFERTODF.XLS in the
\SAMPLES\OLEAUTO\VBA folder.
4) Use some format other than Excel (though similar problems for blank
values occur for text files too)
Options 1) works, and when I have tried 2) and 3) I will report back.
The number of columns to import appears to be controlled by the first 100
(approx) rows of data, not the column headers.
Data typing appears to be controlled by ALL data in a column, so any
character data anywhere in a numeric column will force the datatype to be
either factor or character deptending on the "Import Text as Factors"
option.
Importing mixed alpha and numeric data as character data CAN be done by
specifying the "Import Text as Factors" options as "NEVER", though if the
data is all numeric it comes in as numeric.
Cheers!
Michael Camilleri
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
|