Quantcast

AMF3 references

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

AMF3 references

Evert Pot-3
Hi Guys,

I'm running into some issues with SabreAMF regarding object and string
references..

Way back I made a change to maintain the class definition/object
references across multiple bodies; because of a bug I encountered..
Now I'm looking at a bug report, which has the reverse problem.. but for
string references..

So, the logical conclusion is that object and class references are
maintained across AMF3 bodies, but strings aren't...

This, however, seems so weird though.. It made me wonder how you guys
deal with this...

Are your object and class references maintained across bodies?

Evert

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: AMF3 references

Nick Joyce
Hi Evert,

Afaik, AMF3 references are cleared for all types (strings, classes &  
objects) at the end of each 'context'. This includes the remoting  
header (a gotcha for me) and each body, which is a really odd design  
choice, because the whole idea of references is to cut down the bytes  
sent down the wire ..

Hope that helps sort your problem!

Cheers,

Nick



On 24 Mar 2008, at 22:08, Evert | Rooftop wrote:

> Hi Guys,
>
> I'm running into some issues with SabreAMF regarding object and string
> references..
>
> Way back I made a change to maintain the class definition/object
> references across multiple bodies; because of a bug I encountered..
> Now I'm looking at a bug report, which has the reverse problem.. but  
> for
> string references..
>
> So, the logical conclusion is that object and class references are
> maintained across AMF3 bodies, but strings aren't...
>
> This, however, seems so weird though.. It made me wonder how you guys
> deal with this...
>
> Are your object and class references maintained across bodies?
>
> Evert
> _______________________________________________
> PyAMF dev mailing list - [hidden email]
> http://lists.pyamf.org/mailman/listinfo/dev


Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: AMF3 references

Evert Pot-3
Thanks a lot! I'll reiterate my test cases, just wanted to know I'm not
crazy

Nick Joyce wrote:

> Hi Evert,
>
> Afaik, AMF3 references are cleared for all types (strings, classes &  
> objects) at the end of each 'context'. This includes the remoting  
> header (a gotcha for me) and each body, which is a really odd design  
> choice, because the whole idea of references is to cut down the bytes  
> sent down the wire ..
>
> Hope that helps sort your problem!
>
> Cheers,
>
> Nick
>
>
>
> On 24 Mar 2008, at 22:08, Evert | Rooftop wrote:
>
>  
>> Hi Guys,
>>
>> I'm running into some issues with SabreAMF regarding object and string
>> references..
>>
>> Way back I made a change to maintain the class definition/object
>> references across multiple bodies; because of a bug I encountered..
>> Now I'm looking at a bug report, which has the reverse problem.. but  
>> for
>> string references..
>>
>> So, the logical conclusion is that object and class references are
>> maintained across AMF3 bodies, but strings aren't...
>>
>> This, however, seems so weird though.. It made me wonder how you guys
>> deal with this...
>>
>> Are your object and class references maintained across bodies?
>>
>> Evert
>> _______________________________________________
>> PyAMF dev mailing list - [hidden email]
>> http://lists.pyamf.org/mailman/listinfo/dev
>>    
>
> _______________________________________________
> PyAMF dev mailing list - [hidden email]
> http://lists.pyamf.org/mailman/listinfo/dev
>  


Loading...