Well, I looked at that and took the approach that the URL  
serialization actually assumes URLs to be immutable and entirely  
constructible from the individual data elements. That's also why  
there's a custom section in the deserialization method. Apparently,  
URLStreamHandler are supposed to be bound to protocols and should be  
able to be found through URLStreamHandlerFactory instances. That's  
also what the set(...) methods provide.

I don't think it's needed to run the readObject() method when  
hydrating URL instances.

On 10 Oct 2007, at 20:06, Tim Eck wrote:

> Question --> URLs are mutable (but only by subclasses of  
> URLStreamHandler
> -- see set(..) methods on URL). It is an esoteric usage for sure,  
> but I
> think it is a case where someone could change the local state of an  
> URL
> but the distributed state would be unaffected.
>
> I played a trick with java.io.File where I run the readObject() method
> when hydrating instances (see FileApplicator). It is completely  
> sensitive
> to what ObjectInputStream methods File.readObject()actually calls, but
> that doesn't worry me much. Maybe with something similar URLs could be
> regular physical objects (with honored transients) and still work?
>
>
>> -----Original Message-----
>> From: [EMAIL PROTECTED] [mailto:tc-dev-
>> [EMAIL PROTECTED] On Behalf Of Geert Bevin
>> Sent: Wednesday, October 10, 2007 8:18 AM
>> To: Orion Letizi
>> Cc: [email protected]
>> Subject: Re: [tc-dev] [Tc-commits] gbevin [5866] DEV-258 : Support
>> forjava.net.URL deserialization
>>
>> Indeed, it should now just be supported as a literal value.
>>
>> On 10 Oct 2007, at 17:12, Orion Letizi wrote:
>>
>>> Does this mean that we don't need the beanshell I described anymore?
>>>
>>> On Oct 9, 2007, at 9:42 AM, Geert Bevin wrote:
>>>
>>>> Suppose these should be adapted:
>>>>
>>>> http://www.terracotta.org/confluence/display/~orion/2006/11/17/
>>>> Beanshell+for+java.net.URL
>>>> https://wiki.terracotta.org/confluence/display/wiki/Wicket
>>>>
>>>> I think there's also a list of supported (or unsupported classes)
>>>> somewhere, that should be updated too.
>>>>
>>>> ... and maybe the integration module Wicket should be updated also?
>>>>
>>>> On 09 Oct 2007, at 18:37, Alex Miller wrote:
>>>>
>>>>> Cool stuff.  Are there any docs that need to be updated on the
>>>>> web site?
>>>>>
>>>>> ----- Original Message -----
>>>>> From: [EMAIL PROTECTED]
>>>>> To: [EMAIL PROTECTED]
>>>>> Sent: Tuesday, October 9, 2007 11:32:38 AM (GMT-0600) America/
>>>>> Chicago
>>>>> Subject: [Tc-commits] gbevin [5866] DEV-258 : Support for
>>>>> java.net.URL deserialization
>>>>>
>>>>> _______________________________________________
>>>>> tc-commits mailing list
>>>>> [EMAIL PROTECTED]
>>>>> http://lists.terracotta.org/mailman/listinfo/tc-commits
>>>>>
>>>>
>>>> --
>>>> Geert Bevin
>>>> Terracotta - http://www.terracotta.org
>>>> Uwyn "Use what you need" - http://uwyn.com
>>>> RIFE Java application framework - http://rifers.org
>>>> Music and words - http://gbevin.com
>>>>
>>>
>>
>> --
>> Geert Bevin
>> Terracotta - http://www.terracotta.org
>> Uwyn "Use what you need" - http://uwyn.com
>> RIFE Java application framework - http://rifers.org
>> Music and words - http://gbevin.com
>>
>> _______________________________________________
>> tc-dev mailing list
>> [email protected]
>> http://lists.terracotta.org/mailman/listinfo/tc-dev
>
> _______________________________________________
> tc-dev mailing list
> [email protected]
> http://lists.terracotta.org/mailman/listinfo/tc-dev

--
Geert Bevin
Terracotta - http://www.terracotta.org
Uwyn "Use what you need" - http://uwyn.com
RIFE Java application framework - http://rifers.org
Music and words - http://gbevin.com

_______________________________________________
tc-dev mailing list
[email protected]
http://lists.terracotta.org/mailman/listinfo/tc-dev

Reply via email to