Re: Question about variable length fields

[Scott Klement <midrange-l@xxxxxxxxxxxxxxxx>]

Hi John,

It seems to me that these commands are doing exactly what you asked them 
to do.  You told it to copy the data from one file to another, and to 
map the values into the fields with the same names.  (i.e. *MAP)

It did that.

If the old file had a field with 256 blanks in it, the new file also has 
a field with 256 blanks in it.   If the old file had a field with 50 non 
blank characters, followed by 206 blank characters, the new file has the 
same thing.

You seem to think that this isn't a logical result -- but to my mind, 
it's perfectly logical.  Blanks are characters, too.  And should be 
treated the same as any other characters.

If your goal is to trim blanks from the field, why not just run a simple 
SQL statement?

       UPDATE MYFILE SET MYFIELD=TRIM(MYFIELD)

Seems to me that this would be a one-time shot if your programs are 
writing the new data correctly...



John Allen wrote:
> I need to extend the length of a field in a physical file
>
> It currently is 256 byte field
>
> I want to make it 1028 variable length field
>
> So I changed the DDS to specify Length of 1028 with VARLEN(30) 
>
> Then I did CHGPF and when I look in the physical file I see that the length
> of the field for all of the existing records is set to
> the original fixed length of 256 even though most of the fields are blank or
> only have a few bytes of data in it
>
> New records added are fine
>
> I have tried both CHGPF and CRTPF with CPYF using *MAP *DROP and both set
> the field length of existing records to 256 (old fixed length value)
>
>  
>
> My question is:
>
> Is there any easy way to change the physical file and get the field length
> set to the length of the actual data in the field as opposed to the old
> fixed length?
>
> I suppose I could write a small program to read the old file and write to
> the new file.
>
> But CHGPF or CRTPF with CPYF are so much easier
>
>  
>
> John
>
>