Re: Translating CSV to Data Structure (or iterating through a data structure)

Hi Mike,

Long ago (in a distant galaxy, far far away) I wrote an example for this mailing list called gettok(). Link to archive:
http://archive.midrange.com/rpg400-l/200408/msg00372.html

With that subprocedure, you could do something like this:

x = 0;
myds.field1 = gettok(line: '|': x)
myds.field2 = gettok(line: '|': x)
myds.field3 = gettok(line: '|': x)

and so forth...

Presumably the data in 'line' would be loaded from a stream file, perhaps using fopen()/fgets()/fclose() to process the file.

Unfortunately, there really isn't any way to do this so that the whole DS is populated in one shot. To do that, you'd really need a compiler or precompiler, since the subfield definitions aren't generally known at run-time.

Technically, this is a pipe-delimited file (not CSV, which stands for "Comma Separated Values")

On 5/10/2011 8:43 AM, Mike Wills wrote:

I have a document I need to parse where each row is a different
format. The document is pipe "|" delimited. Is is possible to write a
more generic subprocedure to take each row and a data structure and
fill in the data structure? What I am thinking is iterating through
each field in the data structure and filling it with the corresponding
field from the CSV line that I parse. Is this even possible?