Vadim Gritsenko wrote, On 29/07/2003 2.13:
Stefano Mazzocchi wrote:


On Monday, Jul 28, 2003, at 14:23 Europe/Rome, Sylvain Wallez wrote:

...
In other words, streamed requests aren't so much different from regular requests : it's just that incoming data is more complex and that decoding is not handled transparently by the servlet engine. Once decoded, the processing model can be the same as usual.

Hmmm, hmmmm, hmmmm, you are triggering second order thinking.... hmmmm... I need a whiteboard... I'll be back soon.

Are you thinking in terms of "MultipartParser, SOAPParser, ..."?

Do you remember Vadim about our discussion on adding a getStream to the Request, and how that brought also to think about this?


Basically we usually have (1):

 +------------+               .----------.
 |            |               |          |
 |  Request   |.............. | Context  |
 |            |               |          |
 +------------+               `----------'
 |            |
 |            |
 |            |          )----------------------(
 |  Payload   |          |                      |
 |            |......... |   Sitemap            |
 |            |          |                      |
 |            |          |                      |
 |            |          )----------------------(
 |            |
 +------------+


When sending xml in the POST data, as with XForms (IIUC), we have (2):



+------------+ .----------. | | | | | Request |.............. | Context | | | | | | | `----------' | )--------( | | | | | | |Payload |......., | | | | | | | | | | | )--------( | | )----------------------( +------------+ | | | | Payload |......... | Sitemap | | | | | +------------+ | | )----------------------(


And with webservices, we may have (3):



+------------+ .----------. | | | | | Request |.............. | Context | | | | | | +------------+ | `----------' | )--------( | | | | | | | | |Request |......., | | | | | | | | | )--------( | )----------------------( | | | | | Payload |......... | Sitemap | | | | | +------------+ | | )----------------------(



With our discussion we came to the conclusion that case (2) was now not possible with the current mechanism, and it came also out that the more general case was:


+------------+ .----------. | | | | | Request |.............. | Context | | | | | | | `----------' | )--------( | | |Payload1| | | |Payload2|......., | | ... | | | | |Payloadn| | | | )--------( | | )----------------------( +------------+ | | | | Payload |......... | Sitemap | | | | | +------------+ | | )----------------------(


So in reality, there should be a kind of mechanism *before* Generation that is able to present the sitemap with payloads as urls and Requests in the content as Request data.


This is basically what I'd like to see. :-)

[Made with Jave: http://www.jave.de/]

--
Nicola Ken Barozzi                   [EMAIL PROTECTED]
            - verba volant, scripta manent -
   (discussions get forgotten, just code remains)
---------------------------------------------------------------------




Reply via email to