Re: read/chain record format or file name?

[Scott Klement <rpg400-l@xxxxxxxxxxxxxxxx>]

Hi James,

I find using the filename to be easier to read in most cases because it 
directly links the filename to the F-spec, and to the %EOF() and 
%FOUND() bifs.   So I think it makes it easier for the next guy if you 
use the filename wherever possible.

However, there are situations where you must use the record format name.

a) When reading a file with multiple record formats.   This is 
ubiquitous when working with display files -- we typically use a 
different record format for each screen, which means you have to use the 
record format name.   This can also crop up if you use multi-format 
logical files.

b) When reading an externally-defined file directly into a qualified DS 
in V5R2.  (Older releases didn't support DSes for externally-defined 
files.  Newer releases relaxed the record format name restriction.)

The old adage in the RPG community is "Read a file, write a format".  I 
suggest you stick to that except in situations where you can't.


James.Hawkins@xxxxxxxxxx wrote:
> When performing a Read or Chain operation, is best practices to
> reference the record format name or the file name?  Why?
>
> I know what I do.  I have seen commercial software do both.  I have
> seen individual programmers do both.
>
> The reason I pose the question is I am teaching a class in RPG
> (/free, if you care) at a community college this semester and prefer
> to teach best practices from the start.
>
> Your thoughts?
>
> TIA
>
>
> Thank you, - Jim