Yes, END-INSPECT is a good candidate for enhancement. Since I don't have
ready access to the ANSII COBOL standards manual, I have to assume that it's
not supported by hp, because it ain't supported by ANSII.
As for the complexity of use...I think it's a double edged sword...it does
take some reading/understanding to use INSPECT correctly because of all of
the options the command has, but those command options also make the INSPECT
command a very robust command.
Ray Shahan
> -----Original Message-----
> From: Tracy Pierce [SMTP:[log in to unmask]]
> Sent: Wednesday, June 27, 2001 12:36 PM
> To: [log in to unmask]
> Subject: FW: Bizarre Cobol Error
>
> I privately replied to Ray...
>
> > I was about to say I thought INSPECT was no longer 'in thar',
> > but that's clearly wrong. Are you sure there's no
> > END-INSPECT (it sure isn't doc'd in the manual)? I think
> > THAT would be a valid enhancement request, don't you?
>
> ...before realizing that his reply to me was on-list. Ten minutes later,
> oh
> yeah - EXAMINE was replaced by INSPECT. So WHY no END-INSPECT? Could it
> have been overlooked? Could it be that the code behind INSPECT is so
> horrible nobody got around to rewriting it? Or that its implementation
> was
> already so buggy that rewriting it would have broken a good portion of
> existing usage? Does this seem like a valid SIGCOBOL issue?
>
> For my money, when an INSPECT candidate situation occurs, I've always
> found
> a way around having to integrate (or maybe just read?) the NINE pages of
> prose documentation covering its operation into my programs by simply
> writing my own loops, hopefully more straightforward.
>
>
> Tracy (always hated INSPECT) Pierce
>
> PS: my minimal-periods strategy still would have caught the missing-space
> problem!
>
> > > -----Original Message-----
> > > From: Shahan, Ray [mailto:[log in to unmask]]
> > > Sent: Wednesday, June 27, 2001 9:28 AM
> > > To: Tracy Pierce; Shahan, Ray; [log in to unmask]
> > > Subject: RE: Bizarre Cobol Error
> > >
> > >
> > > I can think of a spot where no periods might sting...the INSPECT
> > > statement...I've always wondered why they didn't do an END-INSPECT?
> > >
> > > > -----Original Message-----
> > > > From: Tracy Pierce [SMTP:[log in to unmask]]
> > > > Sent: Wednesday, June 27, 2001 11:26 AM
> > > > To: 'Shahan, Ray'; [log in to unmask]
> > > > Subject: RE: Bizarre Cobol Error
> > > >
> > > > everybody's right on the money with this one, including Ray, who's
> > > > assuming
> > > > that the prior paragraph was PERFORMed.
> > > >
> > > > I generally avoid problems like this by avoiding the use of
> > > the other
> > > > "paragraph" indicator, which is the period (.). That
> > > normally ends a
> > > > sentence, but is really only necessary at the end of a
> > > paragraph. Not
> > > > using
> > > > periods to end sentences will also force you to use
> > > cobol-85 constructs
> > > > such
> > > > as END-IF, a very good thing.
> > > >
> > > > Tracy Pierce
> > > >
> > > > > -----Original Message-----
> > > > > From: Shahan, Ray [mailto:[log in to unmask]]
> > > > > Sent: Wednesday, June 27, 2001 9:16 AM
> > > > > To: [log in to unmask]
> > > > > Subject: Re: Bizarre Cobol Error
> > > > >
> > > > >
> > > > > Actually, it will GOBACK to calling paragraph at
> > > > > perform1020-put-record.
> > > > >
> > > > > Ray Shahan
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: Curt Brimacomb [SMTP:[log in to unmask]]
> > > > > > Sent: Wednesday, June 27, 2001 10:06 AM
> > > > > > To: [log in to unmask]
> > > > > > Subject: Re: Bizarre Cobol Error
> > > > > >
> > > > > > Michael,
> > > > > >
> > > > > > never thought about that one. Things that make you go
> > > > > hhhmmmmmm......
> > > > > >
> > > > > > Thanks for the enlightenment!
> > > > > >
> > > > > > curt
> > > > > > -----Original Message-----
> > > > > >
> > > > > >
> > > > > > You intended to PERFORM a paragraph name 1020-put-record,
> > > > > instead you
> > > > > > have
> > > > > > a NEW Paragraph named 'perform1020-put-record'.
> > > > > >
> > > > > > This will compile clean, but will not perform
> > > > > 1020-put-record as you
> > > > > > intended it to, instead it will fall through to perform
> > > > > > 1220-lock-next-set.
> > > > > > Probably not the results you were looking for.
> > > > > >
> > > > > >
> > > > > > --
> > > > > > Michael -
> > > > > >
> > > > > > >>> Curt Brimacomb <[log in to unmask]> 06/27/01
> > > 09:40AM >>>
> > > > > > I just had a very bizarre Cobol error.
> > > > > >
> > > > > > I was redoing paragraph names in the source code. I
> > > > > accidentally lost a
> > > > > > space after the word "perform" on one line and ended up
> > > > > with something
> > > > > > like
> > > > > > this:
> > > > > >
> > > > > > perform1020-put-record.
> > > > > > perform 1220-lock-next-set.
> > > > > > move "update" to status.
> > > > > >
> > > > > > 1000-next-para.
> > > > > >
> > > > > > the program complied with no errors, but would not run
> > > > > right. I added
> > > > > > the
> > > > > > one space back into the "bad" perform line. It still
> > > > > compiled with no
> > > > > > errors, but now does run correctly.
> > > > > >
> > > > > > Why did it compile with the missing space
> > > > > >
> > > > > > * To join/leave the list, search archives, change list
> > > settings, *
> > > > > > * etc., please visit
> > http://raven.utc.edu/archives/hp3000-l.html *
> > > >
> > > > * To join/leave the list, search archives, change list settings, *
> > > > * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
> > > >
> >
>
> * To join/leave the list, search archives, change list settings, *
> * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
|
