[Phoenix] Re: [Phoenix-commits] rev 17670 - public/ulysses/src/trunk/ulysses/prototypes/prototype1/src/chrome/content

[Michael Wechner]

Andreas Wuest wrote:

> Hi
>
> On 8.9.2006 0:03 Uhr, Michael Wechner wrote:
>
>> andi at wyona.com wrote:
>>
>>> Author: andi
>>> Date: 2006-09-07 23:30:25 +0200 (Thu, 07 Sep 2006)
>>> New Revision: 17670
>>>
>>> Added:
>>>   
>>> public/ulysses/src/trunk/ulysses/prototypes/prototype1/src/chrome/content/neutronauth.js 
>>>
>>>   
>>> public/ulysses/src/trunk/ulysses/prototypes/prototype1/src/chrome/content/neutronauthparser10.js 
>>>
>>> Modified:
>>>   
>>> public/ulysses/src/trunk/ulysses/prototypes/prototype1/src/chrome/content/authentication.js 
>>>
>>>   
>>> public/ulysses/src/trunk/ulysses/prototypes/prototype1/src/chrome/content/editor.xul 
>>>
>>>   
>>> public/ulysses/src/trunk/ulysses/prototypes/prototype1/src/chrome/content/networkservice.js 
>>>
>>> Log:
>>> Authentication rewrite part 1. (Note: may break some authentication 
>>> related stuff.)
>>>  
>>>
>>
>> if we use versioning, then it shouldn't break, resp. the server should
>> reply that it doesn't support this "Neutron version".
>>
>> I think we should really take this opportunity to implement that
>> different versions of Neutron can exist, because there will be changes
>> again.
>
>
> I haven't changed anything with regard to the syntax of Neutron-Auth.


ok, just read the word break and was associating it with protocol 
changes ;-) Should have looked more closely

>
> It's just that I have rewritten the authentication handling, but not 
> yet fixed up the callsites, i.e. some function calls are currently 
> dispatched with too few or even wrong arguments. That's why it breaks.


ok

>
> Anyway, if you'd like to reingrate Neutron-Auth into Neutron, and make 
> it independent of the underlying protocol, we should discuss this now. 
> We should find out just how independent we want to be! If you want to 
> completely decouple it from HTTP, then we had to introduce a further 
> abstraction layer.


what else than HTTP would make sense? I think Thomas had some ideas re 
other protocols.

Thanks

Michi


-- 
Michael Wechner
Wyona      -   Open Source Content Management   -    Apache Lenya
http://www.wyona.com                      http://lenya.apache.org
michael.wechner at wyona.com                        michi at apache.org
+41 44 272 91 61