Re: Backround processor - file access methods question

For COBOL I usually do error handling in DECLARATIVES.  Are there any
performance advantages or disadvantages to using a called routine?



Basil Zangare
Applications Specialist
Nikon Inc
1300 Walt Whitman Road
Melville, New York 11747
phone: (631) 547-4389
fax: (631) 547-4026
bzangare@xxxxxxxxx


-----Original Message-----
From: cobol400-l-bounces@xxxxxxxxxxxx
[mailto:cobol400-l-bounces@xxxxxxxxxxxx] On Behalf Of
qsrvbas@xxxxxxxxxxxx
Sent: Wednesday, February 15, 2006 6:45 PM
To: cobol400-l@xxxxxxxxxxxx
Subject: Re: [COBOL400-L] Backround processor - file access methods
question

cobol400-l-request@xxxxxxxxxxxx wrote:

>   1. Re:  Backround processor - file access methods question
>      (Adrienne McConnon)
>
>I was able to try the suggestion and it does look as if it is
>functioning, however, the document mentions that 'ILE COBOL cannot
>directly inport... Errno variable...'.  And it is suggested that a C
>module should be used to obtain the error number/description.  We do
not
>have the c compliler - what do hard core COBOL shops do for error
>handling?  Is there a way to get the error number/description in COBOL
>or CL - without having to purchase another compiler?  Where do the
error
>numbers and messages reside?

Adrienne:

Others have supplied perfectly good examples, so I'll skip that. What
might not have been clear is that the IBM example is wrong in the most
practical sense even if it has purely technical correctness for most of
it. That is, as the examples show here, COBOL has a fairly easy time of
handling errno just as RPG does.

While IBM examples are very useful, never assume that they're
actually... ummm... 'useful'.

Tom Liotta