I agree that using a RELEASE without any arguments is not desirable
method of releasing locks, however if I look at the current
documentation for RELEASE, I cant see, either implicitly or
explicitly, any idication that it will release my LOCK statement. I
would prefer either the documentation be updated to highlight this
functionality, or the RELEASE mechanism be addressed to leave my
LOCK's  to my UNLOCK's...

Russell.

On May 23, 10:05 pm, "comp.lang.smalltalk" <[email protected]>
wrote:
> This behavior has been around since the jBASE 3 days.
>
> While it is not explicitly mentioned in the RELEASE or LOCK
> documentation, I did cover it in the 'Keys to Record Locking'
> knowledgebase article I wrote years ago.
>
> I would tend to agree that it shouldn't do that but then it is sloppy
> programming to issue a RELEASE statement with no arguments.
>
> Dan
>
> On May 22, 7:54 pm, Russell Bowes <[email protected]>
> wrote:
>
>
>
> > We've recently upgraded a client to 5.2.15 and we noticed that issuing
> > a RELEASE (with no arguments) also released a LOCK which the
> > application had also taken.
>
> > This bug may have been in previous releases, but we just hadn't
> > noticed it.
>
> > I'm planning on installing an older release, just out of interest, to
> > see if this behaviour existed in previous releases.
>
> > Has anyone else experienced this issue ?

-- 
IMPORTANT: T24/Globus posts are no longer accepted on this forum.

To post, send email to [email protected]
To unsubscribe, send email to [email protected]
For more options, visit this group at http://groups.google.com/group/jBASE?hl=en

Reply via email to