struts-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rene Gielen <>
Subject Re: Collections problem
Date Wed, 04 Jul 2007 20:42:09 GMT
Yes, I confirm this, too.

Basically, xwork does the same as before - automatically use 
GenericsObjectTypeDeterminer if found classpath. But from xwork 2.x on, 
this does not make sense any more, since it does not longer depend on 
additional tiger jar being deployed to activate Java5 support, and 
GenericsObjectTypeDeterminer will always be found in Classpath.

I've filed and fixed an issue for this in xwork:
The output should now be less confusing.

+1 fo setting GenericsObjectTypeDeterminer as the default for Struts2, 
instead of DefaultObjectTypeDeterminer as it is now...


Aymeric Levaux schrieb:
> Hi,
> I faced the same problem and found the cause.
> Prior to Xwork 2.0.3 the objectTypeDeterminer property in XWorkConverter 
> was initialised using the instance created by the static block in 
> ObjectTypeDeterminerFactory. This static block determines which 
> implementation of the ObjectTypeDeterminer must be instantiated 
> according to the Java version. This was quite useful: no need of extra 
> configuration to get some benefits when using Java5+ versions.
> In 2.0.3 the instance of ObjectTypeDeterminer in use is the one created 
> in the Dispatcher (method init_PreloadConfiguration) and the 
> implementation used depends of the value of struts.objectTypeDeterminer 
> in, which defaults to struts that points to the non 
> generics implementation.
> So applications developed with WebWork or prior Struts 2 versions that 
> are taking benefits of the GenericsObjectTypeDeterminer are broken after 
> migration to Struts 2.0.8 until the struts.objectTypeDeterminer in 
> is set to 'tiger'.
> As Struts 2 is using Java 5, I guess that the default value of 
> struts.objectTypeDeterminer should be 'tiger'.
> And maybe the static block in ObjectTypeDeterminerFactory should be 
> removed; it is a bit confusing when debugging.
> Aymeric Levaux <>
> Musachy Barroso wrote:
>> Has anybody taken a look at this?

>> regards
>> musachy
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

Rene Gielen  |
Aachen       | PGP-ID: BECB785A
Germany      | gielen at

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message