December 2006 - jboss-dev-forums - Jboss List Archives

[Design of JBossCache] - Re: Serialization Compatibility Tests

by manik.surtani＠jboss.com

Yes, but to prevent cluttering /etc, would this be a single binary file containing a serialized Map? Or would you need multiple objects? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3995052#3995052 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3995052

19 years, 3 months

1
0
0 / 0

[Design of JBossCache] - Re: Serialization Compatibility Tests

by pavel.tsekov＠redhat.com

Ok, then. So I'd create a database for each existing release and put it under etc/ in the cvs repository, right ? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3995050#3995050 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3995050

19 years, 3 months

1
0
0 / 0

[Design of JBossCache] - Re: Serialization Compatibility Tests

by manik.surtani＠jboss.com

I would say yes, we should create a db for each of them. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3995047#3995047 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3995047

19 years, 3 months

1
0
0 / 0

[Design of JBossCache] - Re: Buddy replication cleanup in a transaction

by manik.surtani＠jboss.com

Hmm, how about a UT to prove this? Sounds like a bug. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3995044#3995044 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3995044

19 years, 3 months

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Valve on HAAspect

by timfox

That is simply not true. Acquiring a reader or writer lock only synchronizes for the duration of the acquire call. If you don't believe me look at the source. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3995043#3995043 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3995043

19 years, 3 months

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Valve on HAAspect

by clebert.suconic＠jboss.com

anonymous wrote : How does the use of Valve minimise it? Valve still needs to be synchronised. It synchronizes access to the attribute isRunning but it releases the attribute very fast, while lock will require synchronizationo for the entire call. With Locks: handleInvocation: .... lock. .... performAspect .... release lock with Valves: handleInvocation .... lock .... access atributes .... verify if the valve is free .... release locks ..... perform Aspect View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3995037#3995037 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3995037

19 years, 3 months

1
0
0 / 0

[Design of JBossCache] - Re: Last chance for any changes to the 2.0.0 API

by galder.zamarreno＠jboss.com

Guys, the change I'm currently working on to add VAM to JDBCCacheLoader would need an interface change. I haven't yet finished testing it to make sure that it's certainly more performant. The change is this: org.jboss.cache.marshall.ObjectStreamFactory public ObjectInputStream createObjectInputStream(InputStream in) throws IOException; JavaObjectStreamFactory and JBossObjectStreamFactory would implement it. VAM would add: public Object objectFromInputStream(InputStream is) throws Exception Finally, JDBCCacheLoader would call: Object marshalledNode = getMarshaller().objectFromInputStream(is); passing, InputStream is = rs.getBinaryStream(1); // ResultSet Would you guys be happy for me to include this before finishing with http://jira.jboss.com/jira/browse/JBCACHE-879? Without this methods, I would have to loop through the InputStream to create a byte[] containing this data which would a waste of time. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3995031#3995031 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3995031

19 years, 3 months

1
0
0 / 0

[Design of JBoss Web Services] - JBWS-1260 - Unwrapping empty sequences

by darran.lofthouse＠jboss.com

Hopefully the last scenario for this task :-) According to the JAX-RPC 1.1 specificiation, the third rule for an operation to be unwrappable: - anonymous wrote : such an element (a wrapper) must be of a complex type defined using the | xsd:sequence compositor and containing only elements declarations. I have one scenario with the following types defined in the WSDL: - <element name='lookup'> | <complexType> | <sequence> | </sequence> | </complexType> | </element> | | <element name='lookupResponse'> | <complexType> | <sequence> | </sequence> | </complexType> | </element> And on a second scenario I have the following types defined: - <complexType name='Person'> | <sequence> | </sequence> | </complexType> | | <complexType name='TelephoneNumber'> | <sequence> | </sequence> | </complexType> wscompile is generating the method in the SEI as: - public org.jboss.test.ws.jbws1260.LookupResponse lookup() throws | java.rmi.RemoteException; and public org.jboss.test.ws.jbws1260.TelephoneNumber lookup() throws | java.rmi.RemoteException; This apears to be handling the unwrapping of the return type differently to the unwrapping of the parameters (as discussed in a previous post). My question is - If the JAX-RPC specification only mentions complex types that contain a sequence of elements does this mean this scenario does not qualify for unwrapping as there are no elements? If this type can be unwrapped should the return type of the method be void? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3995028#3995028 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3995028

19 years, 3 months

1
0
0 / 0

[Design of JBoss Web Services] - Re: JBWS-1260 - What when we cannot unwrap the message parts

by darran.lofthouse＠jboss.com

Thinking about this further some refactoring is going to be required so that the generation of the return type and the generation of the parameters happens together as either of these can affect the decision as to if unwrapping should occur. (I can currently fall back to not unwrapping the return type or parameters independently but not both together). The current fix for JBWS-1260 contains a lot of scenarios that users are currently experiencing problems with that should be working. I will get JBWS-1260 finished to just include the scenarios that we should be unwrapping and throw an exception for the scenarios that are not supported. I will raise a feature request to cover falling back to not unwrapping if it is not possible to unwrap the message parts. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3995019#3995019 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3995019

19 years, 3 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: Running all tests at once

by adrian＠jboss.org

"alesj" wrote : | What about a single file info with which tests failed? That just involves using ant's junitreport and style (xslt) tasks see the bottom of jboss-head/testsuite/build.xml for an example. The cruisecontrol build already has this anyway (somewhere) http://cruisecontrol.jboss.com/cc/artifacts/microcontainer-head-testsuite... View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3995018#3995018 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3995018

19 years, 3 months

1
0
0 / 0

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

jboss-dev-forums December 2006