doedit command

Kenneth Brody kenbrody at spamcop.net
Wed May 2 08:39:26 PDT 2012


On 5/2/2012 11:22 AM, Dan McCabe wrote:
> Good morning.
>
> I tend to define my dummies all in one place at the bottom of the
> processing table.  I've never had an issue with that before.

And it's still not an issue.

> ta is defined
> with a length of 80.  t has length of 10 and a field type of .2.

As Joe and I pointed out, that's the problem.  Since you don't explicitly 
pass a length to doedit(), it uses the length of the input value -- in this 
case, 10 -- and the result can't fit into 10 characters.

> ta is
> currently defined as "*", but I also tried "rfdol" and "cheque" with the
> same results.  t has a value in my test of 54874.55.   I thought that if I
> defined the dummy with a fixed length, the doedit command will insist on
> using the length of the input, am I wrong Ken?

If you don't explicitly specify a length, it _does_ use the length of the 
input.  What did you define the input field as?

> In the debugger, it appears
> to be more than the 10 characters from the input field, even though it is
> blank.

Here, you imply that the input is still 10 characters.  Are you sure you 
changed the length of the input, and not the destination?

> If I just use the processing to print the report, t displays properly, as
> do all of the other variables I am using, but ta is blank.

See above.

> If I lookup a free record in my temp file and put the 54874.55 in a .2
> field, it works as expected.  If I put the same number in a field in
> the temp file with an edit of rfdol and a length of 80, it fails and the
> field is blank.

Can you show the processing?

>  But if I use IUA (sorry Ken, fat fingers), and type in
> 54874.55 it completes the edit with room to spare on the end.

As I would expect.  Assuming the edit is correct. :-)

> rfdol is defined as           [! " " !](zero|lots mycent |<zero>mycent)
> mycent is defined as<  and>  [!.!] num num</100 dollars>
>
> Lots is the standard edit from Global edits.
>
> I apologize for leaving so much information out.
[...]

-- 
Kenneth Brody


More information about the Filepro-list mailing list