HP3000-L Archives

July 1997, Week 3

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Larry Boyd <[log in to unmask]>
Reply To:
Larry Boyd <[log in to unmask]>
Date:
Tue, 15 Jul 1997 08:08:44 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (76 lines)
James,

What this means is that somewhere in your source you are calling FOO and
BAR where the parameters either do match (passed by value vs. passed by
address) or the number of required parameters does not match (I can't
remember exactly what the error means, but it is something like this).

Once I had to build a job file to recompile a complete application of some
200+ routines.  Several of these routines had not been recompiled in
several years.  When I received this error, I would run QEDIT through MPEX
searching for the string "FOO" or "BAR".  I then reviewed the parameters
for the routines versus all of the calls to the routines.  I was then able
to find the calls with the problem.

One thing you might have to worry about is that FOO or BAR do what the
calling routine thinks it is suppose to do.  What I found, because of the
age of several of routines were *very* old and the "new" routines didn't
process as the old ones did.  Therefore, I had to create FOO2 and BAR2 -
 one to process in the "old" manner and one to process in the "new" manner.

You may also find that after you are able to add all the routines to the RL
at run time it still fails with errors (this is where I can't remember
which of the two possible errors causes the ADDRL error).  One set of
errors can occur when you add the routine to the RL or USL, one set of
errors can occur when you PREP, and one set can occur when you RUN  (all
these depend your options during the PREP and RUN).  However, no matter
where/when the error occurred, I found that the QEDIT/MPEX solution worked.

LB
______________________________________________________________________
         ___
        /  /            Larry Boyd: Horizontal Growth Solution Manager
HEWLETT/hp/PACKARD      Commercial Systems Division (CSY)
      /__/              [log in to unmask]        (408)447-4127

All opinions are my own, and not necessarily those of Hewlett-Packard.
______________________________________________________________________


-----Original Message-----
From:   James Funk [SMTP:[log in to unmask]]
Sent:   Monday, July 14, 1997 9:36 PM
To:     [log in to unmask]
Subject:        Actual Incompatible with Formal; RL changes

HP 3000; MPE V      (classic)
-------------------------

I was making changes to a (test copy of an) RL:
..
purgerl foo
purgerl bar
..
addrl foo
addrl bar

I got error messages such as these:

*** ERROR *** FOO, FOO
ERROR #49 ACTUAL FUNCTION INCOMPATIBLE WITH FORMAL FUNCTION

*** ERROR *** BAR, BAR
ERROR #49 ACTUAL PARAMETER INCOMPATIBLE WITH FORMAL PARAMETER
2

I've seen this type of message before (and was expecting some here), but in
a context in which the named functions/subroutines on the line with ***
ERROR *** were different:  the calling routine and the called routine.

Doesn't the above imply that FOO is incompatible with itself?  (And the
same for BAR?)

I want to find those routines which are calling FOO and BAR.

== end ==

ATOM RSS1 RSS2