> 1. Its now 1.4 compatible (actually wasn't before I took it)
great
> 2. Its no longer a subclass of bayeuxservice
Alright I'm quite confused. I did that yesterday and checked that change in
- also yesterday - and tried to synchronise just 2 secs ago - and there
weren't any changes. Where did you check your stuff in?
> I checked the repo before comitting, and it had the correct code... check
> how you generated the jar...
I generate the jar with mvn install on the checked out code. Again, I'm
having the resources in there, but they never get referenced in any java.
I'm missing some kind of headercontribution or renderjavascript in there ...
And now I find things even more weird. Can you confirm that your
AbstractDefaultDojoBehavior doesn't call
response.renderJavascriptReference(getDojoResourceReference());
on line 108 (and thus tries to add resources that don't exist, causing
wicket to report errors to console)?
if you can't confirm that, we're working on different code bases, however
that might work :-D
regards,
Michael
On 8/18/08, Michael Sparer <[EMAIL PROTECTED]> wrote:
>
> OK I'll take a look into it :-)
>
> Weird that you don't get any errors, because the line I pasted certainly
> throws errors as the resources are just not there - seems to be a
> remainder
> of the copied class. Are you sure you checked in all your changed sources?
>
>
> Rodolfo Hansen-2 wrote:
>>
>> On Mon, Aug 18, 2008 at 11:18 AM, Michael Sparer
>> <[EMAIL PROTECTED]>wrote:
>>
>>>
>>> so you think it wouldn't be worth the effort to do the upgrade to 1.0 in
>>> the
>>> dojo project? how much effort are we talking about? :-)
>>
>>
>> Look into it! the DOJO people have a list of compatibility changes, and
>> movements, and howto fix.
>> But, its very, very incompatible. That was the first idea I had and then
>> abandoned it
>> At the end I decided it wasn't correct in any case since push
>> functionality
>> for wicket should be independent.
>>
>>
>>>
>>>
>>> but you'll still end-up with two different dojo versions if you use both
>>> wicketstuff-push and wicketstuff-dojo in one project. that's kind of
>>> misleading and overhead, isn't it?
>>
>>
>> No, I'm planning on cleaning things up, and include only the cometd part,
>> and have it play fair with any other javascript libs...
>>
>> Hopefully you could even have other comet clients on a purely JS side...
>> we
>> need to write some test cases, as this project doesn't have any...
>>
>>>
>>>
>>> the error stuff: don't you get wicket-errors such as "shared resource
>>> org.wicketstuff.push.cometd.dojo.AbstractDefaultDojoBehavior/dojo/lfx/*.js
>>> not found" ? actually you should as AbstractDefaultDojoBehavior calls
>>
>>
>> Nope, no issues, they are included correctly in the jar I have...
>>
>>>
>>>
>>> public static final ResourceReference DOJO =
>>> new CompressedResourceReference(AbstractDefaultDojoBehavior.class,
>>> "dojo/dojo.js");
>>>
>>> on line 71 ... and there isn't a PackageResource like that - sure you
>>> checked in all your sources?
>>> shouldn't the resources in src/main/resources be added somewhere as
>>> resource
>>> in the code. either i didn't find them or i missed a point completely
>>> :-)
>>>
>>> and yes, sure, upgrading to the latest versions of cometd and dojo is
>>> definitely a good idea, no doubt about that. but i believe one should
>>> first
>>> talk about the basic stuff as we do in this thread :-)
>>>
>>> regards,
>>> Michael
>>>
>>>
>>> Rodolfo Hansen-2 wrote:
>>> >
>>> > On Mon, Aug 18, 2008 at 10:16 AM, Michael Sparer
>>> > <[EMAIL PROTECTED]>wrote:
>>> >
>>> >>
>>> >> I had a more thorough look into it now and there are some issues I'd
>>> like
>>> >> to
>>> >> address:
>>> >> - wouldn't it be better to lift wicketstuff-dojo to dojo 1.0 and keep
>>> it
>>> >> as
>>> >> dependency to wicketstuff-push? otherwise there are two seperate dojo
>>> >> versions and if you use ws-dojo and ws-push you end up in a whole
>>> mess
>>> of
>>> >> javascript libraries
>>> >
>>> >
>>> > I looked into that. The dojo API changed drastically between 0.4 and
>>> 1.0,
>>> > its a huge effort.
>>> > Albeit i think eventually it should be done, but I think
>>> wicketstuff-push
>>> > should end up independent from wicketstuff-dojo
>>> >
>>> >>
>>> >> - where in your code do you reference the new dojo libraries? i only
>>> >> found
>>> >> resourcereferences in the copied and shortened class
>>> >> AbstractDefaultDojoBehavior that point to /dev/null i.e. to
>>> references
>>> >> that
>>> >> aren't existing, leading to wicket throwing errors and to not being
>>> able
>>> >> to
>>> >> use dojo at all ...
>>> >
>>> >
>>> > ? What do you mean exactly?
>>> > The dojo files are in
>>> src/main/resources/org/wicketstuff/push/cometd/dojo
>>> > and it is working correctly in the projects I am using...
>>> >
>>> >
>>> >
>>> >>
>>> >> - you sometimes used generics for e.g. hashmaps - i think (but i'm
>>> not
>>> >> sure)
>>> >> that all wicket and wicketstuff versions lower than wicket 1.4.
>>> should
>>> be
>>> >> java 1.4 compatible and generics came with java 5 ... well i don't
>>> care,
>>> >> i'm
>>> >> using java 6, but i think there might be a few that do care :-)
>>> >
>>> >
>>> > Oops, will fix.
>>> >
>>> >>
>>> >> - speaking of users, although there's now a branch for the 1.3.0
>>> version
>>> >> existing, i think that might confuse or upset wicketstuff-push users
>>> as
>>> >> you're only chance to get wicketstuff-push is to check it out from
>>> svn.
>>> >> so
>>> >> keeping up-to-date with their version (i.e. 1.3.0) isn't easy as they
>>> >> would
>>> >> have to switch to the branched version ... which they don't know
>>> existing
>>> >> ... so shouldn't that whole upgrade thing rather be a separate
>>> project?
>>> >
>>> >
>>> > Please give 1.3.4 a chance, help me fix these small issues, and at the
>>> > end,
>>> > I think we will be better off, no?
>>> >
>>> >>
>>> >>
>>> >> and the servlet and service problem i mentioned in an earlier post
>>> (in
>>> >> the
>>> >> other push thread, i think) ...
>>> >>
>>> >> regards,
>>> >> Michael
>>> >>
>>> >> Rodolfo Hansen-2 wrote:
>>> >> >
>>> >> > No problem, thats why I bumped the version number.
>>> >> >
>>> >> > I still need to debug the RemoveListener, haven't been able to get
>>> in
>>> >> > touch
>>> >> > with the guy from jetty, and I'll fix the examples as the only real
>>> >> > incompatibility is that there is no longer a dependency on
>>> >> > wicketstuff-dojo.
>>> >> >
>>> >> >
>>> >> > On Fri, Aug 15, 2008 at 1:31 PM, Michael Sparer
>>> >> > <[EMAIL PROTECTED]>wrote:
>>> >> >
>>> >> >>
>>> >> >> I appreciate the upgrade and am certainly going to do the upgrade
>>> for
>>> >> my
>>> >> >> app
>>> >> >> and report errors etc. I hope you don't mind that I took the
>>> liberty
>>> >> of
>>> >> >> proposing a branch to have a ready-to-run version with the prior
>>> >> release
>>> >> >> of
>>> >> >> cometd and dojo (see
>>> >> >>
>>> >> >>
>>> >>
>>> http://www.nabble.com/Permission-for-branching-wicketstuff-push-to19002588.html
>>> >> >> )
>>> >> >>
>>> >> >> regards,
>>> >> >> Michael
>>> >> >>
>>> >> >>
>>> >> >> Rodolfo Hansen-2 wrote:
>>> >> >> >
>>> >> >> > Hey guys,
>>> >> >> >
>>> >> >> > I was talking in irc about updating wicketstuff-push to support
>>> the
>>> >> >> more
>>> >> >> > recent versions of the bayeux protocol and cometd servlets.
>>> >> >> >
>>> >> >> > So I endeded up with access to the wicketstuff svn and have just
>>> >> >> uploaded
>>> >> >> > a
>>> >> >> > new preliminary version of wicketstuff-push.
>>> >> >> >
>>> >> >> > there are still some issues and alot of ironing out, but the
>>> project
>>> >> >> now
>>> >> >> > works with the latest stable versions of jetty, wicket, cometd,
>>> >> there
>>> >> >> is
>>> >> >> a
>>> >> >> > new feature available where you can listen for a clients
>>> disconnect
>>> >> >> from
>>> >> >> a
>>> >> >> > web page (actually cometd session)..
>>> >> >> >
>>> >> >> > the new version is 1.3.4-SNAPSHOT if anyone is interested in
>>> cheking
>>> >> it
>>> >> >> > out,
>>> >> >> > or helping me continue the cleanup...
>>> >> >> >
>>> >> >> >
>>> >> >> > Thanks,
>>> >> >> >
>>> >> >> > Rodolfo Hansen.
>>> >> >> >
>>> >> >> >
>>> >> >>
>>> >> >>
>>> >> >> -----
>>> >> >> Michael Sparer
>>> >> >> http://talk-on-tech.blogspot.com
>>> >> >> --
>>> >> >> View this message in context:
>>> >> >>
>>> http://www.nabble.com/wicketstuff-push-update-tp18765345p19002629.html
>>> >> >> Sent from the Wicket - Dev mailing list archive at Nabble.com.
>>> >> >>
>>> >> >>
>>> >> >
>>> >> >
>>> >>
>>> >>
>>> >> -----
>>> >> Michael Sparer
>>> >> http://talk-on-tech.blogspot.com
>>> >> --
>>> >> View this message in context:
>>> >>
>>> http://www.nabble.com/wicketstuff-push-update-tp18765345p19032686.html
>>> >> Sent from the Wicket - Dev mailing list archive at Nabble.com.
>>> >>
>>> >>
>>> >
>>> >
>>>
>>>
>>> -----
>>> Michael Sparer
>>> http://talk-on-tech.blogspot.com
>>> --
>>> View this message in context:
>>> http://www.nabble.com/wicketstuff-push-update-tp18765345p19033883.html
>>> Sent from the Wicket - Dev mailing list archive at Nabble.com.
>>>
>>>
>>
>>
>
>
> -----
> Michael Sparer
> http://talk-on-tech.blogspot.com
> --
> View this message in context:
> http://www.nabble.com/wicketstuff-push-update-tp18765345p19035419.html
> Sent from the Wicket - Dev mailing list archive at Nabble.com.
>
>
-----
Michael Sparer
http://talk-on-tech.blogspot.com
--
View this message in context:
http://www.nabble.com/wicketstuff-push-update-tp18765345p19045474.html
Sent from the Wicket - Dev mailing list archive at Nabble.com.
