Looking for some clarification on Datasets

[Former Member]

Alright, I need to read in a csv file that contains several lines. The delimiter is ^. This file holds one line that has the header information for a sales order. The subsequent lines are individual line items (1-many). How can I read one line at a time and end when I reach the end of the file? I would be reading the line into a string and split it from there.

Regards,

Davis

[Former Member]

Alright, I need to read in a csv file that contains several lines. The delimiter is ^. This file holds one line that has the header information for a sales order. The subsequent lines are individual line items (1-many). How can I read one line at a time and end when I reach the end of the file? I would be reading the line into a string and split it from there.

Regards,

Davis

[naimesh_patel]

Hello,

LOOP AT ITAB.   " this contains your file.
at first.
*  split and  move  records to header
continue.
endat.

split and move records to item

endloop.

Regards,

Naimesh Patel

[Former Member]

I should add that this is a background job, or will be a background job. That means that I will be using datasets and can't use anything related to the GUI.

Davis

[Former Member]

Hi,

In that case, you can read the lines of file in Text mode. Get the line into a string and then use SPLIT command to split the values at '^' into the table fields.

ashish

[Former Member]

Do I just do something like this?

do.
   read dataset filepath into filestring.
   if sy-subrc <>0.
      exit.
   endif.

*split string*

enddo.

Davis

[Former Member]

Yes, but you have to use the OPEN DATASET command before trying to read it.

If you search the forum, you'll see many examples of exactly how to do this.

Rob

[Former Member]

Hi,

Check this sample program which will give you an idea -

http://www.sapfans.com/sapfans/alex.htm

ashish

[Former Member]

Hi,

Use function module GUI_UPLOAD, pass 'X' to parameter HAS_FIELD_SEPARATOR.

Check the documentation for more details.

ashish

[Former Member]

Hi,

You can use these FM RKD_WORD_WRAP or SWA_STRING_SPLIT.

Regards,

Ferry Lianto