RELEASE UNLOCKS in 5.2.15

[Russell Bowes]

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 ?

[comp.lang.smalltalk]

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 <russ...@citadelcomputer.com.au>
wrote:

[Russell Bowes]

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" <danielklei...@gmail.com>
wrote:

[Daniel Klein]

The documentation for the RELEASE and LOCK statements have been updated on the KB.

Dan

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

To post, send email to jB...@googlegroups.com
To unsubscribe, send email to jBASE-un...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/jBASE?hl=en

[Bill Benedict III]

It isn't mentioned there, but there is a link to this article on The Keys to
Record Locking:

http://www.jbase.com/r5/knowledgebase/howto/general/common/RecordLocking.htm

...under the NOTES section of the RELEASE command, which states:

"...This is an important point. Avoid using the RELEASE statement without
arguments. Not only does this release all locks in all files and all
execution locks taken by the current program (which may be undesirable due
to the fact that one or more locks must remain set), it also releases all
execution locks set in the current program with the LOCK statement. Plus it
imposes additional overhead such that the entire lock table has to be
searched for all locks belonging to the current process."

-Bill

[Russell Bowes]

Dan,

Thanks. That at least will allow me to say to someone that thats the
expected behaviour.

Russell.

> >http://groups.google.com/group/jBASE?hl=en- Hide quoted text -
>
> - Show quoted text -