[Seaside] critical GOODS client bug fixes

[Sebastián Sastre]

Avi,

	I've noticed that in my dolphin port but I've omitted to
consider it cames from squeak 8-P

	Anyway, thanks for the update. 

	I want to ask if some GOODS squeak client user is experimenting
unespected instVar objects in the persistent objects.

	I have stored in a BTree a lot of aContact, and from time to
time, I have aContact who has in it's #observations instVar somethis
different of aString, aBTreeLeafNode for instance, and the seaside
application goes it DNU. 
	
	Any clue?

Sebastián Sastre
 
ssastre at seaswork.com.ar 
Seaswork 
Special Software Solutions
www.seaswork.com.ar



-----Mensaje original-----
De: seaside-bounces at lists.squeakfoundation.org
[mailto:seaside-bounces at lists.squeakfoundation.org] En nombre de Avi
Bryant
Enviado el: Domingo, 24 de Octubre de 2004 18:51
Para: The Squeak Enterprise Aubergines Server - general discussion.
CC: The general-purpose Squeak developers list
Asunto: [Seaside] critical GOODS client bug fixes


The problem below has been fixed, and a new version (78) is up on SM.  
I strongly recommend people upgrade to this - as well as fixing the 
decoding of negative small ints (would you believe this has never 
worked properly until now, and nobody noticed? Scary...), it catches a 
couple of edge cases where, after the right combination of conflicts 
and rollbacks, some changes were getting missed during commit.

Avi

On Oct 24, 2004, at 1:24 PM, Yar Hwee Boon wrote:

> On Sun, 24 Oct 2004 13:24:03 +0200, Avi Bryant <avi at beta4.com> wrote:
>
>>
>> On Oct 24, 2004, at 6:27 AM, Yar Hwee Boon wrote:
>>
>>> I noticed a problem when storing and retrieving SmallInteger(s) in
>>> GOODS. Run the following:
>>>
>>> =====
>>> db := KKDatabase onHost: 'localhost' port: 6100.
>>> db root at: #xx put: -3.
>>> db flushAll.
>>> (db root at: #xx) inspect
>>> ======
>>
>> You need a #commit before the #flushAll or you won't have anything at
>> #xx at all.  But yes, having put that in I can reproduce your bug - 
>> clearly something going wrong with the signed integer representation.

>>  I'll look into it, thanks.
>
> I thought of this, but when I try to debug and step through
> #encodeSignedInteger: and #decodeSignedInteger: by putting #halt in 
> them, it seems not to write to the database, or write differently as 
> when I remove the #halt-s. Do you know why?.

_______________________________________________
Seaside mailing list
Seaside at lists.squeakfoundation.org
http://lists.squeakfoundation.org/listinfo/seaside

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.778 / Virus Database: 525 - Release Date: 15/10/2004
 

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.778 / Virus Database: 525 - Release Date: 15/10/2004