Johan Compagner a écrit :
> first what is so special about that code?
The second implementation uses cometd which is a dojo fundation project
fully integrated in jetty 6 and using Bayeux protocol (based on json to
make server/clients communications).
> What is the protocol you are talking about
>
first what is so special about that code?
What is the protocol you are talking about
I haven't really looked at it yet at this time but i will.
How i see it if possible is just an open channel
and over it flows the wicket ajax stuff (so the xml that wicket ajax can
parse)
this way if something i
Johan Compagner a écrit :
> yes i want to have it simple and build in in wicket-ajax.js if possible
> (at least support for it it could be and extra small js: wicket-cometd.js)
well, I understand what you mean, but comted (www.cometd.com) is a dojo
fundation subproject. So if we want to write our
yes i want to have it simple and build in in wicket-ajax.js if possible
(at least support for it it could be and extra small js: wicket-cometd.js)
johan
On 3/20/07, Igor Vaynberg <[EMAIL PROTECTED]> wrote:
this is a great start. my idea vision of this implementation is something
that doesnt d
> The problem comes from cometd bus which is design to transport json and
> not xml. I was thinking of something which is able to serialize a
> component rendered as String and put all in a json object in order to
> avoid this second useless callback. but it need some additionnal work ;)
>
> --
> V
this is a great start. my idea vision of this implementation is something
that doesnt depend on dojo and something that has a very similar
requesttarget to ajaxrequesttarget that can transfer value pairs _or_
markup.
but this is definetely a great start.
-igor
On 3/20/07, Vincent Demay <[EMAIL
TECTED] [mailto:wicket-user-
> [EMAIL PROTECTED] On Behalf Of Vincent Demay
> Sent: Tuesday, March 20, 2007 11:54 AM
> To: wicket-user@lists.sourceforge.net
> Subject: Re: [Wicket-user] Serveur pushing implementation
>
> James McLaughlin a écrit :
> >> WDYT?, Any comments, feeli
On 3/20/07, Vincent Demay <[EMAIL PROTECTED]> wrote:
> Eelco Hillenius a écrit :
> > Can't the Tomcat support be merged into the same project?
> >
> I need to check for that and check tomcat behavior is the same as jetty
> but I think it should
Cool. My note was more directed to Johan actually :)
James McLaughlin a écrit :
>> WDYT?, Any comments, feelings?
>>
>
> This looks like really exciting work. I can't wait to try it out. I
> took a look at the code and you've done a really nice job. One
> question, is it possible to push rendered components instead of
> pinging those components
Eelco Hillenius a écrit :
> Can't the Tomcat support be merged into the same project?
>
I need to check for that and check tomcat behavior is the same as jetty
but I think it should
--
Vincent
> Eelco
>
>
> On 3/20/07, Johan Compagner <[EMAIL PROTECTED]> wrote:
>
>> tomcat uses this:
>>
>>
Can't the Tomcat support be merged into the same project?
Eelco
On 3/20/07, Johan Compagner <[EMAIL PROTECTED]> wrote:
> tomcat uses this:
>
> http://tomcat.apache.org/tomcat-6.0-doc/aio.html
>
> so for tomcat to work i need to create a special filter (or servlet) that
> maps a push connection
>
>
> WDYT?, Any comments, feelings?
This looks like really exciting work. I can't wait to try it out. I
took a look at the code and you've done a really nice job. One
question, is it possible to push rendered components instead of
pinging those components on the client side to call back?
-
Johan Compagner a écrit :
> tomcat uses this:
>
> http://tomcat.apache.org/tomcat-6.0-doc/aio.html
>
> so for tomcat to work i need to create a special filter (or servlet)
> that maps a push connection
>
> How does jetty work? When does jetty know that this is a nio/cometd
> connection?
It is exa
tomcat uses this:
http://tomcat.apache.org/tomcat-6.0-doc/aio.html
so for tomcat to work i need to create a special filter (or servlet) that
maps a push connection
How does jetty work? When does jetty know that this is a nio/cometd
connection?
johan
On 3/20/07, Vincent Demay <[EMAIL PROTECTE
ncent Demay
>> Gesendet: Dienstag, 20. März 2007 10:13
>> An: wicket-user@lists.sourceforge.net
>> Betreff: [Wicket-user] Serveur pushing implementation
>>
>> Hi all,
>>
>> I made a new project in wicket-stuff(sourceforge) called
>> wicket-contrib-push.
Johan Compagner a écrit :
> is the second one really push? or push by polling?
The first one is pushing simulated by polling ;) (and timer)
The second one is real pushing with cometd (using jetty6 continuation :
no locked thread on serveur side and keep open http connection).
>
> I am currently ch
TECTED]
> [mailto:[EMAIL PROTECTED] Im Auftrag
> von Vincent Demay
> Gesendet: Dienstag, 20. März 2007 10:13
> An: wicket-user@lists.sourceforge.net
> Betreff: [Wicket-user] Serveur pushing implementation
>
> Hi all,
>
> I made a new project in wicket-stuff(sourcefor
is the second one really push? or push by polling?
I am currently checking out the NIO support of tomcat 6.
I hope in the coming weeks to have an impl that is really push (so there is
really a nio socket connection from server to client all the time)
johan
On 3/20/07, Vincent Demay <[EMAIL PRO
Hi all,
I made a new project in wicket-stuff(sourceforge) called
wicket-contrib-push.
This project allows pushing of events on server side in a simulated bus
between server and all client. This pushing has 2 implementations :
* The first based on Xavier Hanin work which is based on timerBehavi
19 matches
Mail list logo
