Hi,
I came accross the same problem some months ago (in a more generic way
because it wasn't targeting Google only).
I made the choice to use PAC4J to benefit from its principle and algorithm :
"all you have to do" is to handle page requests and decide wheither or not
it should be pushed to PAC4J
Hi,
Thank you for your code, I gave it a try and looked at it closer, and I'm
ont sure it feets a real need, well at least mine : it delegates the answer
to another thread, so it's as if we add a worker thread to the container,
kind of machine load. It is not what I expected : my very first idea
-loading pages due to background
sql select (for list item for instance) ?
2017-02-16 10:11 GMT+01:00 Martin Grigorov-4 [via Apache Wicket] <
ml-node+s1842946n4677129...@n4.nabble.com>:
> On Thu, Feb 16, 2017 at 9:50 AM, guillaume.mary <[hidden email]
> <http:///user/SendEmail.jtp?t
Ok. Thanks for the explanation.
--
View this message in context:
http://apache-wicket.1842946.n4.nabble.com/Wicket-8-and-servlet-3-1-async-nio-tp4677098p4677128.html
Sent from the Users forum mailing list archive at Nabble.com.
Martin Grigorov-4 wrote
> On Wed, Feb 15, 2017 at 9:41 PM, guillaume.mary
> guiom.mary@
>
> wrote:
>
>> Martin Grigorov-4 wrote
>> > Hi,
>> >
>> >>
>> >> Did I miss something ? if not is there anything planned about async IO
>
Martin Grigorov-4 wrote
> Hi,
>
>>
>> Did I miss something ? if not is there anything planned about async IO in
>> Wicket 8 ?
>>
>
> There is no planned support for asynchronous pages for Wicket 8!
Is there some technical (known) reason ? Or just because of roadmap decision
?
I mean, if i have
Hi there !
I had a look at the future Wicket 8 and I saw that the servlet 3.1 was the
target. So I looked a little bit at the code and I didn't see anything about
async IO usage. I was looking for some code that looks like the recommended
one by Oracle
Hi !
I'm upgrading our application from Wicket 7.4.0 to 7.5.0 to benefit from the
security issue fix.
While doing this I encounter a problem with pages that extend other one and
define a in their markup, for instance:
The parent page have a defined markup with and .
If
Thanks i'll give it a try.
Le 22 août 2016 18:45, "Sven Meier [via Apache Wicket]" <
ml-node+s1842946n4675362...@n4.nabble.com> a écrit :
> You can call #configure() on the subcomponents from your #onConfigure().
>
> Have fun
> Sven
>
> --
> If you reply to this
Well, in my use case, subcomponents use a Behavior#onConfigure() to change
their own visibility, and this method is called after
Component#onConfigure(), hence the visibility state is not uptodate.
Le 22 août 2016 13:54, "Sven Meier [via Apache Wicket]" <
Hi,
I'm a Wicket user since several years and I regularly lack a feature in
Wicket : component state change.
I know that Event/Sink exists but it doesn't feets my need : when I use a
component that wraps others from a third party, I need it to be hidden or
disabled when *all* subcomponents are.
at the moment.
Martin Grigorov-4 wrote
Hi,
On Wed, Dec 17, 2014 at 12:26 PM, guillaume.mary
guillaume.mary@
wrote:
Hi all,
With WICKET-5711 OnChangeAjaxBehavior now listen both change and
inputchange on TextField. Hence when you type in the TextField, the
behavior is called twice because both
Hi all,
With WICKET-5711 OnChangeAjaxBehavior now listen both change and
inputchange on TextField. Hence when you type in the TextField, the
behavior is called twice because both events are fired by the browser. It's
the same problem as WICKET-5603 but applied to TextField for all browsers
Hi !
I want to implement a simple use case where a notification is sent to a
client when its reports is finished, with Wicket Atmosphere.
I’m new to Atmosphere concepts, but not to Wicket. So I had a look at
Wicket-Atmosphere Quick Start.
A broadcast is very simple to create as it is done there,
Yes, thanks for the entry point.
So I came to the following solution: register a ResourceRegistrationListener
on the EventBus which fills a Map (concurrent) of SessionId to UUID. The
SessionId is took from Session.get(), and the UUID is the one from
resourceRegistered(..).
Then my reporter
issue 5190 created because of same question asked here
http://apache-wicket.1842946.n4.nabble.com/Manage-exceptions-on-a-nested-multipart-upload-form-td4658815.html
https://issues.apache.org/jira/browse/WICKET-5190
--
View this message in context:
sorry for duplicate, i will update the other post as I created a quickstart
and created issue 5190:
https://issues.apache.org/jira/browse/WICKET-5190
--
View this message in context:
Thanks for the tip Sven, I will give it a try.
For now I have another issue about the AbstractTransformerBehavior: as soon
as I add it to my Page, I get encoding problems of characters into my page,
this is weird. Here is my very simple transformer:
new AbstractTransformerBehavior() {
public
After searching a bit, I found that AbstractTransformerBehavior replaces the
RequestCycle.response temporarly to put its own, between beforeRender en
afterRender methods, whereas other classes manipulates the response during
that phase, so when AbstractTransformerBehavior puts back the original
Thanks Martin, it's easier to find the actual page with
PageRequestHandlerTracker.
Nevertheless, the redirecting case is still not filled because as I
debugged, PageRequestHandlerTracker returns null for both first and last
handler when my filter is called (am i wrong ??). That's almost the same
Because I didn't know AbstractTransformerBehavior ! :)
Many thanks !
I put it on my Page instance so it fullfills the last case of post +
redirect (but the filter is still necessary for components refreshed by
Ajax, otherwise I should add AbstractTransformerBehavior to all components
of my Page)
Martin Grigorov-4 wrote:
Because the resources are supposed to be next to WEB-INF folder.
Wicket should handle the requests to dynamic resources (pages,
callback listeners, etc.), while the static resources better should be
handled by the web container.
Why do you use 'resources' ? I
Created the 2531. Thanks.
--
View this message in context:
http://www.nabble.com/DropDownChoice-and-null-value-internationalization-key-tp25924643p25954595.html
Sent from the Wicket - User mailing list archive at Nabble.com.
well, that's for the first reason.
Doing this would allow us to distinguish wicket Ids from keys. So it would
be really better for our translators since we could concentrate our keys in
a couple of files, and prefixing our keys (or even rename them totally)
could add sense to them. Currently
Hi all,
I'm wondering why the PropertyModel sends an IndexOutOfBoundException when I
use an expression such as myList.0.name and that the 0 element doesn't
exist, whereas it doesn't sends a NullPointerException when I user
person.name with a null person.
Is it a wanted behavior or not (or a
Well, if the person is null I don't get a NPE (to check, I took a wicket
page, added an attribute A to it, displayed A.B with a Label +
PropertyModel, but never setted the attribute A = I never got the NPE)
What I meant in my first post, is since I don't get a NPE in that case, I
was thinking
Well, i'm just doing A a; But I put a trace on the getA() and a is really
null.
Can you make a try ?
Mathias P.W Nilsson wrote:
I don't know if I got it right. Is the A null? Are you just doing A a; or
A a = null; ?
If you do this I don't understand why you don't get an exception. The
I agree with you, my problem is on the fact that Wicket seems to me
inconsistent because of the 2 behaviors: if I have the IOOBE I should have
the NPE.
Wouter Huijnink wrote:
Yes, you are wrong. It's bad practice to refer to a location in a list
if you're not sure the entry exists.
28 matches
Mail list logo
