Differences in export function results

[Kenneth Brody]

Richard Kreiss wrote:
> Ran into an interesting problem the other day at a client's site.  There is
> an export program which creates a csv file which had been running correctly,
> that is until Tuesday.
> 
> Some of the data was being pushed right on position.

What, exactly, is in the CSV file at the point where the data is "shifted"?

> The export command used was export word.

What is the complete export line?

> Now there is a gotcha in this export.  There is an alphanumeric field which
> has numbers 4 - 8 meaning a 4 foot section of shelf was changed to 8 foot.

By that, do you mean the field contains the text "4 - 8"?

> The only way to get this out so that when it is read into Excel is to
> "\""{40.  This exports the value correctly. Otherwise it treated this like a
> date.

It sounds more like this allows Excel to import it "correctly", since it is 
Excel that treats it as a date otherwise.

Why are you prefacing one field with an open quote, and not supplying a 
close quote?  Perhaps that's the problem right there?

> I tried to correct the right shift by using export ASCII to no avail.  This
> made the situation worse.

What was your actual export ASCII line?

> I finally realized what was causing the shift.  If there was no value in
> field 40, double quotes were exports and pushed values over to the right.

What, exactly, ends up in the export file?

> When the export word  was set to:
> 
> If: 40 = ""
> Then:com(23)="";GOTO exp24
> If: 40 ne ""
> Then:com(23)="\""{40
> 
> This worked fine.
> 
> There are definite differences between the 2 functions.
> 
> export ascii shipdet =c:\shipdet r=\n f=, -x
> 
> export word shipdet=c:\shipdet
> 
> May not give the same results.

Why would you expect them to give the same results, when they're not equivalent?

The line:
     export word shipdet=c:\shipdet
is equivalent to:
     export ascii shipdet=c:\shipdet r=\n f=, o=" c="

-- 
Kenneth Brody