Valentina archives

[Christopher Bishop]

I have a request.  I think that we should have access to download all
older RELEASE versions of Valentina, even if there are bugs.

When we release a product to the market, we can no longer easily update
the xtra since the server is installed in a system folder, and you must
use the right client xtra with the right server xtra.  So we need access
to the version that we made our software with.


In my case, I'm fine with the functionality of 2.4.3, so I'm going to
stick with that for my project.  However, I know that as soon as 2.5 is
finished, I won't be able to download 2.4.3 any more, so if I ever have
a problem where I have to redownload 2.4.3 to test and fix my software,
I am screwed.


Please consider this,

Chris


-----Original Message-----
From: valentina-beta-bounces at lists.macserve.net
[mailto:valentina-beta-bounces at lists.macserve.net] On Behalf Of Ruslan
Zasukhin
Sent: Tuesday, November 07, 2006 1:53 PM
To: valentina-beta at lists.macserve.net
Subject: Re: V4MD 2.5 b11 crashes Director

On 11/7/06 11:46 PM, "Pierre Rossel" <prossel2 at myrealbox.com> wrote:

>>> OS: Win XP SP2
>>> More info: 
>>> No time to track the bug to simple example, switching back
>> to 2.4.3 which is
>>> latest stable release.
>>  
>>> Just wanted to show the problem in case it is already known.
>> 
>> This problem sounds like -- a not destroyed cursor.
> 
> OK, let's investigate. I never store cursors in persistent variables.
Only
> local variables that are destroyed when the function exits.

> I sometimes use cursor variables in loops, but I assume previous
cursor is
> destroyed when new one is assigned to the variable.

Yes, although this is NOT a best technique!  BECAUSE

If you will think deep, you will see that SECOND cursor is created
BEFORE
first is destroyed. So you have some time BOTH cursors which eat more
resources then needed. Also may happens that record locks will even
prevent
creation of second cursor.

So best technique in this case if EXPLICITLY set first cursor to VOID
before
create a new one. 

> There is no explicit
> destructor for cursors, right ?

There is:

    curs = VOID

> Just setting the variable to void or
> assigning it a new value (new cursor) will destroy the previous one.
Maybe
> it is related to the garbage collector that actually destroys object
later ?

Right, see above.
 
> The point is: very same procedure produces no crash with 2.4.3, and
crashes
> with 2.5 b11. 

This is NOT good and must be fixed on our side...
Valentina claims to have protection from such mistakes of developer.

But may be problem of your crash related to other issue..
We need do some good testing of our examples and test projects to see if
they still work fine.

> There must be something but, since I want to stay with 2.4.3
> for my next release, there is no point for me tracking down this bug
on 2.5.
> (I mean: not yet).

Do you have V4MD_Log file on hands ?

It ca


-- 

Best regards,

Ruslan Zasukhin
VP Engineering and New Technology
Paradigma Software, Inc

Valentina - Joining Worlds of Information
http://www.paradigmasoft.com

[I feel the need: the need for speed]


_______________________________________________
Valentina-beta mailing list
Valentina-beta at lists.macserve.net
http://lists.macserve.net/mailman/listinfo/valentina-beta