yeah, that would be helpful. maybe someone can do it :) between answering questsions here and working on wicket i have no spare time for the wiki. i am really counting on people whom i help to take the time and contribute back to the wiki.
here is an old thread that might be of help, maybe someone
the only components that can supply fragments are the ones that have markup themselves - ie a page, a panel, or a border because the fragment has to live in someone's markup and those are the only ones that have it.
so is your component a panel, a border, or a page?-IgorOn 5/7/06, Aaron Hiniker
The Pattern seems to be wrong/not working.
After quite some testing I found the following working:
PackageResource.bind(application, TinyMCEPanel.class, Pattern.compile
(.*\\.js), true);
PackageResource.bind(application, TinyMCEPanel.class, Pattern.compile
(.*\\.html), true);
Yes, the fragment provider is a Panel. The markup looks like this:
wicket:panel
/wicket:panel
fragment wicket:id=edit ... /fragment
fragment wicket:id=view ... /fragment
Since the component is providing fragments only, the main markup content is empty and I really don't wish/need to
it should work w/out the namespacethe problem is that the fragment tag is outside of the panel, it should be insidesowicket:panelwicket:fragment./wicket:fragment
/wicket:paneltry that and see if it does the trick.-IgorOn 5/7/06, Aaron Hiniker [EMAIL PROTECTED]
wrote:
Yes, the
On 5/8/06, Aaron Hiniker [EMAIL PROTECTED] wrote:
I am trying to encapsulate Edit/View/Select markup into one Component. I
want to set the mode, then add the Fragment to the parent container.
My first attempts went something like this:
parent.add( component.getFragment( markupId,
wicket:panel
wicket:fragment wicket:id=edit edit /wicket:fragment
wicket:fragment wicket:id=view view /wicket:fragment
/wicket:panel
throws:
Caused by: java.text.ParseException: Unkown tag name with Wicket namespace: 'fragment'. Might be you haven't installed the appropriate resolver?
On 5/8/06, Aaron Hiniker [EMAIL PROTECTED] wrote:
wicket:panel
wicket:fragment wicket:id=edit edit/wicket:fragment
wicket:fragment wicket:id=view view /wicket:fragment
/wicket:panel
throws:
Caused by: java.text.ParseException: Unkown tag name with Wicket namespace:
'fragment'.
why dont we activate it by default?-IgorOn 5/8/06, Juergen Donnerstag [EMAIL PROTECTED]
wrote:On 5/8/06, Aaron Hiniker [EMAIL PROTECTED]
wrote: wicket:panel wicket:fragment wicket:id=editedit/wicket:fragment wicket:fragment wicket:id=viewview/wicket:fragment
/wicket:panel throws: Caused by:
I want to create a MarkupContainer component which sets the model of contained
components. but MarkupContainer.add is final, so i can't override it. why? what
else can i do?
Thanx,
Ittay
--
===
Ittay Dror
Chief architect, openQRM TL,
RD, Qlusters Inc.
[EMAIL
why are all these wicket:xxx tags? why not span wicket:id=wicket:xxx or
similar?
thanks,
ittay
--
===
Ittay Dror
Chief architect, openQRM TL,
RD, Qlusters Inc.
[EMAIL PROTECTED]
+972-3-6081994 Fax: +972-3-6081841
http://www.openQRM.org
- Keeps your
because it hasn't been extensively tested. As far as I'm aware you are
almost the only user of it. If you think it is stable enough and
doesn't interfere with any other features and/or functionalities than
it can be enabled by default.
I made some test to see if Aaron request for a markup file
On 5/8/06, Juergen Donnerstag [EMAIL PROTECTED] wrote:
because it hasn't been extensively tested. As far as I'm aware you arealmost the only user of it. If you think it is stable enough anddoesn't interfere with any other features and/or functionalities thanit can be enabled by default.
i have
You want to do what? add() is used to add Wicket Components to the
component hierarchy. It is NOT used to set or add a model. Child
components can be access by get(childId).
Juergen
On 5/8/06, Ittay Dror [EMAIL PROTECTED] wrote:
I want to create a MarkupContainer component which sets the model
Which tags are you referring to, and why are they a problem?We have tried to minimize our specific tags, and they are nicely escaped so your browser should not complain about it.Some exist to make a clear distinction to which are components that are added to your panel, and the panel itself, or to
Martijn Dashorst wrote:
Which tags are you referring to, and why are they a problem?
We have tried to minimize our specific tags, and they are nicely escaped
so your browser should not complain about it.
Some exist to make a clear distinction to which are components that are
added to your
because it is much cleaner.
juergen
On 5/8/06, Ittay Dror [EMAIL PROTECTED] wrote:
why are all these wicket:xxx tags? why not span wicket:id=wicket:xxx or
similar?
thanks,
ittay
--
===
Ittay Dror
Chief architect, openQRM TL,
RD, Qlusters Inc.
[EMAIL
On Mon, 2006-05-08 at 00:52 -0700, Igor Vaynberg wrote:
On 5/8/06, Juergen Donnerstag [EMAIL PROTECTED] wrote:
because it hasn't been extensively tested. As far as I'm aware you are
almost the only user of it. If you think it is stable
Ittay,It seems like you need to create a Panel instead So a much cleaner design, and more reusable component would be:public class MyDatePicker extends Panel { public MyDatePicker(String id, IModel model) {
super(id); add(new DatePicker(datepicker, model)); add(new TextField(textfield, model));
Don't know wy you have to specify them once for a extentions.About that recursive thing:We do this: if (recurse){ addResources(scope, pattern, resources, new StringBuffer(relativePath
.toString()).append(file.getName()).append('/'), file, recurse);}And i guess that is wrong. We
i want to set the model of the child. i thought of doing that by:
child.setModel(myModel);
super.add(child);
what is wrong with that? how else can i do it?
here's the long story of what i'm trying to acheive:
i have a web page with two text fields, for date and time (for the
that is not my problem.
my problem is that i have two components (text fields) - one for date, one for time, whose combined value is my model value.
i can, of course, create a custom model, and onSubmit get its values, combine
them, and put in the form's model. but this is awkward, and not
it was fixed two days back, now everything should be fine. i updated to tinymce 2.0.6.1 version as well and applied Frank's patch. please do and update and let me know if it works for you.
iulianOn 5/8/06, Alexander Lohse [EMAIL PROTECTED] wrote:
The Pattern seems to be wrong/not working.After
Yeah, but a good framework should not get in the way of what you are trying
to do, right? I spent 2-3 hours trying to figure out a way to do this with
fragments :)
Others will probably hit the same wall someday. Panels did work for my
solution, but in this situation loading fragments as a
Hi all,
I want to develope
menu tree via javascript calls in wicket
How it is create
plz give me some information regarding this
Thanks in Advance
Gangadhar
---
Using Tomcat but need to do more? Need to support web services, security?
Get
a method like add is deliberately protected by final to ensure it will
work like it is supposed to, no matter what component hierarchy you're
in. So, before opening up such a method, we would like to be
absolutely sure there are no alternatives for it.
In your case there are a couple of things
i have used it quiet a bit and it seems to work fine. so i would be +1 for
making it enabled by default.
I can confirm that :) +1 for enabling by default.
Eelco
---
Using Tomcat but need to do more? Need to support web services, security?
For those people that have reacted, I want to let you know that I gave
the complete list to Manning. They didn't want to include everyone on
the list though, as that would take way to much time to handle. I
didn't have much direct influence on who is picked and who is not, so
for those that
Thanks for your reactions concerning starting up an
internationalization project guys. If I ever have a spare moment in
the next few weeks, I'll try to start it up. Might take a while
though, so I hope you're not in a hurry :)
Eelco
---
Using
On 5/8/06, Eelco Hillenius [EMAIL PROTECTED] wrote:
For those people that have reacted, I want to let you know that I gavethe complete list to Manning. They didn't want to include everyone on*sniff* ah well - roll on the publication :)
Cool, thanks for sharing that. If you're interested, it would be very
much appreciated to write a few lines about this at our WIKI:
http://wicket-wiki.org.uk/wiki/index.php/Stories
And anyone else reading this (including the guys from Servoy and
Topicus), it'd be nice to share a few lines there
Working on it at this actual moment :) Writing, writing, writing.
Eelco
On 5/8/06, Mark Derricutt [EMAIL PROTECTED] wrote:
On 5/8/06, Eelco Hillenius [EMAIL PROTECTED] wrote:
For those people that have reacted, I want to let you know that I gave
the complete list to Manning. They didn't
It's one of these components we'd still like to have in our
distribution. Didn't happen so far yet though. There's a contribution
here
http://sourceforge.net/tracker/index.php?func=detailaid=1416392group_id=119783atid=684977
(with a patch) you might find useful. There were a couple of things I
Could you please provide a patch? You already have it on your computer
right? I'll apply it right away.
Thanks,
Eelco
On 5/8/06, karthik Guru [EMAIL PROTECTED] wrote:
I'm not sure if there is any interest in
wicket.contrib.markup.html.velocity.VelocityPanel. But just
for the sake of
I've added http://www.wicket-wiki.org.uk/wiki/index.php/Using_repeaters,
although I'm not sure if there's not an optimisation I'm missing
involving a compound property model, as I've got a populateItem() like
this...
protected void populateItem(ListItem item)
{
Eelco Hillenius wrote:
Tags with wicket:id attributes are coupled to Wicket components, while
wicket:xxx are more like processing instructions. I/ we feel like
this is a big distinction and should be communicated as such.
what about my other message about a CompositeComponent. how is it so
I'm looking for the equivalent of Ognl.setValue(...) and guessing I
should use wicket.util.lang.PropertyResolver.setvalue(...);
The wicket.util.object package (referred to in the wiki migrate
instructions) does not exist.
Would it be possible to add a method performActionNamed(String
Yes it is the wicket.util.lang.PropertyResolver classWhat would that perform do?How does it look like?PropertyResolver has limited support for executing methods.something like:Object value = PropertyResolver.getValue
(getAddressArray(), person);is supported.johanOn 5/8/06, Anders Peterson [EMAIL
A different web application framework that I've worked with has ognl
available everywhere. It's extremely convenient and powerful!
http://developer.apple.com/documentation/WebObjects/Reference/API5.2.4/com/webobjects/foundation/NSKeyValueCoding.html
Seems to me it wouldn't be difficult to
Well...
public void performActionNamed(String anExpression) {
PropertyResolver.getValue(anExpression, this);
}
I'm sure it needs to return anything, but maybe it could be passed some
sort of context object (if available).
/Anders
Johan Compagner wrote:
Yes it is the
Anders Peterson wrote:
I'm sure it needs to return anything, but maybe it could be passed some
sort of context object (if available).
Should be: I'm NOT sure it needs to return anything, but maybe it could
be passed some sort of context object (if available).
You also have this now.Just call PropertyResolver.getValue(contextObject, propertyString)where should that performAction method be located?On 5/8/06,
Anders Peterson [EMAIL PROTECTED] wrote:
A different web application framework that I've worked with has ognlavailable everywhere. It's extremely
Eelco Hillenius wrote:
Tags with wicket:id attributes are coupled to Wicket components, while
wicket:xxx are more like processing instructions. I/ we feel like
this is a big distinction and should be communicated as such.
what about my other message about a CompositeComponent. how is it so
You can still use OGNL, just not by default for the PropertyModels
etc. We removed OGNL as it had seriously bad performance - it showed
up as the last hotspot we had when profiling Wicket -, and it doesn't
seem to be maintained actively anymore (in fact, I had to wait for a
bug fix for almost a
* We might consider adding event methods, like onAdd(Component) andonReplace(Component) that are called right after a component is added/
replaced. A special not on this is that the add method will dissapearanyway the next major release as we have the constructor change then.Replace will still
I did a little experiment. The code down below is all you need to
implement your requirement of finding the fragments markup snippet on
a markup stream directly associated with the class.
/**
* A special Fragment (WebMarkupContainer) which searches for the
fragment
*
Hey everybody,
Thanks for the help on the last problem. Im
implementing it as we speak and Im finally starting to dig the core
Wicket concepts now!! (and loving it)
The next problem that I need to fix to proudly show off my
wicket prototype is editing a Set of Strings in a form.
I
Please find both FreemarkerPanel and VelocityPanel attachedOn 5/8/06, Eelco Hillenius [EMAIL PROTECTED]
wrote:Could you please provide a patch? You already have it on your computer
right? I'll apply it right away.Thanks,EelcoOn 5/8/06, karthik Guru [EMAIL PROTECTED] wrote: I'm not sure if there
Provided I understand the issue you want to:
Dynamically create a TextField for each entry in the SetString. The
current value of the iterator shall be the default? The TextFied don't
have Labels? No validators are required for numbers, phone number,
email address, etc.?
= use ListView or any
What I'm asking for is not to be able to use ognl (or similar) on top of
wicket, but to have something equivalent inside it.
An interface PropertyCoding (NSKeyValueCoding) that is then implemented
by *all* wicket classes.
/Anders
Eelco Hillenius wrote:
You can still use OGNL, just not by
I assume you mean PropertyResolver.getValue(String expression, Object
object);
That's not the kind of context I meant. I was thinking there could be an
additional context parameter - maybe RequestCycle. Don't know if that is
possible or necessary. It was just an idea.
I would specifically
then we get partial scripting inside java/page.Why would we have that. What would you call with performAction?On 5/8/06, Anders Peterson
[EMAIL PROTECTED] wrote:I assume you mean
PropertyResolver.getValue(String _expression_, Objectobject);That's not the kind of context I meant. I was thinking
Johan Compagner wrote:
then we get partial scripting inside java/page.
Didn't know that was a bad thing...
Why would we have that. What would you call with performAction?
We can already do this:
PropertyResolver.getValue(expression, this.getPage());
I'd like to just be able to type:
if you quit checking your email every five seconds maybe you will get it done sooner! :)-IgorOn 5/8/06, Eelco Hillenius
[EMAIL PROTECTED] wrote:Working on it at this actual moment :) Writing, writing, writing.
EelcoOn 5/8/06, Mark Derricutt [EMAIL PROTECTED] wrote: On 5/8/06, Eelco Hillenius
I made some test to see if Aaron request for a markup file with
fragments only can be supported. There doesn't seem to be a simple
solution. A page referencing fragments defined in a Panel which is
also added to the Page defnitely doesn't work. We'd need a more
yes but why would you have that?Why not if you have the page call:page.getExpression()Then it is compile time save it refactores nicelyIn my eyes the the PropertyModel is just handy and there is no otherway to do it then by using strings.
It is a pitty that we can give method or field pointers
why not this?WebMarkupContainer composite=new WebMarkupContainer(id, new CompoundPropertyModel(new ValueMap()));composite.add(new TextField(date));composite.add(new TextField(time));
this does what you want right?composite's model is a map whose date key contains the value of the date textfield,
That's not the kind of context I meant. I was thinking there could be anadditional context parameter - maybe RequestCycle. Don't know if that is
possible or necessary. It was just an idea.RequestCycle is a threadlocal and is available from anywhere via RequestCycle.get() given that code executes
Igor Vaynberg wrote:
why not this?
WebMarkupContainer composite=new WebMarkupContainer(id, new
CompoundPropertyModel(new ValueMap()));
composite.add(new TextField(date));
composite.add(new TextField(time));
this does what you want right?
composite's model is a map whose date key contains
in that case i think you should do what martijn said because that is proper encapsulation. how else can you reuse that combination elsewhere?IModel model=new Model(new Date());WebMarkupContainer composite=new WebMarkupContainer(composite, model));
composite.add(new TextFIeld(date,
Why can't I mount my pages so they all can be bookmarkable from / ?Just a note: the WicketServlet is under url-pattern = /site, so my pages would be bookmarkable like /site/Index.Regards,
-- Bruno Borges[EMAIL PROTECTED]Sun Certified Java Programmer for 1.4Sun Certified Web Component Developer for
you cannot mount onto / because that is reserved for the homepage, so in a way homepage is always mounted on /if you want to mount something onto /site/index then you need to mount on /index not /mount(/index,
Index.class);btw, your url mapping should be /site/* not just /site-IgorOn 5/8/06,
Yeah, it's /site/*, I just cutted off the unnecessary info. Sorry... :DSo, how about something like:mountByClassName(String underPath, PackageName pkg);this would be the same as loop through all classes located under the
pkg and create a mounted path with the same name as the class.I know this
if you are talking about mounting a package on / then you should be able to do that with something like mount(/, PackageName.for(Index.class));then if you have an Index.class page in that package it should be accessible from /site/Index
at least i remember i wrote this functionality in because it
And is exactly that. it throws an exception saying that the path can't be /. I saw the source and there's an If statement checking for this. :)On 5/8/06,
Igor Vaynberg [EMAIL PROTECTED] wrote:
if you are talking about mounting a package on / then you should be able to do that with something like
can you remove that if statement and see if it works?-IgorOn 5/8/06, Bruno Borges [EMAIL PROTECTED]
wrote:And is exactly that. it throws an exception saying that the path can't be /. I saw the source and there's an If statement checking for this.
:)On 5/8/06,
Igor Vaynberg [EMAIL PROTECTED]
Igor Vaynberg wrote:
and this is also what martijn showed you. notice that in his example
both components are sharing the same model instance and thus share the
same Date object.
the problem here is that the datepicker is not aware of the time portion
of the date so it probably always
How are you adding those parameters the the image URL?In html I have an image tag like this:img wicket:id=image/and in code:Resource resource = Image image = new Image(image, resource);
I also tried:Image image = new Image(image, new Model(/images/sunshine.png));But than it tries to find the
yep, exactly what i'm trying to do.
the datepicker and timepicker should not use the form's model (which won't
neccessaraly accept them, in my case, i use CompoundPropertyModel, and it
doesn't have the corresponding methods). so, i need to set in them a Model
instance. i can do that in
try this:Image image = new Image(image, resource) { protected void onComponentTag(final ComponentTag tag) { super.onComponentTag(tag); String src=""
src=""> tag.set(src, src);};
if it works you can encapsulate it into a NonCachingImage or something-IgorOn 5/8/06, Potje rode kool
[EMAIL
On 5/8/06, Ittay Dror [EMAIL PROTECTED] wrote:
Igor Vaynberg wrote: but you can make it generic pretty easily: class AggreatingAtomicFormComponent extends FormComponent { updateModel() { visitChildren(new
FormComponent.IVisitor() { ... updatemodel(); });the sub components will try to use their
and how do i validate the resulting value? in updateModel, if i find the
combined value is not right, there's not much i can do.
also, there's a convert() method in Component, what if i want it to be employed
also?
ittay
Igor Vaynberg wrote:
On 5/8/06, *Ittay Dror* [EMAIL PROTECTED]
and how do i validate the resulting value? in updateModel, if i find the combined value is not right, there's not much i can do.
you mean the sum of correct pieces is not always correct itself?!? you can create a validator that performs the same aggregation and checks that i suppose.
also, there's
Ok, thanks all for the help on this problem for me.My Image subclass that I created based on the code of Igor looks like this:class NonCachableImage extends Image{ public NonCachableImage(String id, IModel model)
{ super(id, model); } protected void onComponentTag(final ComponentTag tag) {
Fine with me. If something thinks of a nice API.
Eelco
On 5/8/06, Martijn Dashorst [EMAIL PROTECTED] wrote:
This seems common enough, shouldn't we add this to core (of course, only
when it works :-))?
Martijn
On 5/8/06, Igor Vaynberg [EMAIL PROTECTED] wrote:
try this:
Image image =
This could go to Wicket Core. But before this happens, one note:The 'rand' parameter should be renamed to something like .wicket_rand. 'rand' is too simple... :)Regardsm
On 5/8/06, Eelco Hillenius [EMAIL PROTECTED] wrote:
Fine with me. If something thinks of a nice API.EelcoOn 5/8/06, Martijn
that is strangeare you saying that those pagemaps get deployed when a request is happening?Because all pagemaps gets the session set when the request comes in. (see RequestCycle.prepare() and Session.init())
On 5/9/06, Lusebrink, Scott E CTR DISA GIG-CS [EMAIL PROTECTED] wrote:
doing hot
I get the same exception. Here is the stack trace:
java.lang.NullPointerException
at
wicket.PageMap.visitEntries(Lwicket.PageMap$IVisitor;)V(PageMap.java:574)
at wicket.PageMap.clear()V(PageMap.java:193)
at wicket.PageMap.remove()V(PageMap.java:308)
at
I am having problem generating dynamic images in Wicket. Wicket seem to generate the same URL even when the Image's resource has changed. So the browser decides to use the cached image instead of the uptodate one. How do we tell wicket to generate a different img tag's src value when the image
see here:http://www.nabble.com/changing-images-with-Ajax-t1570490.htmli have also just checked in NonCachingImage, so if you are using trunk you can use it.
-IgorOn 5/8/06, Samyem Tuladhar [EMAIL PROTECTED] wrote:
I am having problem generating dynamic images in Wicket. Wicket seem to generate the
just checked it into corethe param looks like this wicket:antiCache=random-IgorOn 5/8/06, Bruno Borges
[EMAIL PROTECTED] wrote:This could go to Wicket Core. But before this happens, one note:
The 'rand' parameter should be renamed to something like .wicket_rand. 'rand' is too simple... :)Regardsm
the SignI.html file in wicket-examples-1.2-rc3 contains a small typo:
td alight = rightPassword:/td
which of course shoud be
td align = rightPassword:/td
--
View this message in context:
http://www.nabble.com/SignIn.html-t1581459.html#a4292267
Sent from the Wicket - User forum at
thank you, fixed.-IgorOn 5/8/06, middledot [EMAIL PROTECTED] wrote:
the SignI.html file in wicket-examples-1.2-rc3 contains a small typo:td alight = rightPassword:/tdwhich of course shoud betd align = rightPassword:/td
--View this message in context:
This is the only way I know (better: I learned from Wiki/Docs/Examples) on how to use MaskConverter:TextField zipCode = new TextField(zipCode) {
public IConverter getConverter() {MaskFormatter zipFormat = null;try {zipFormat = new MaskFormatter(#-###); // Brazillian format
} catch(Exception e)
On Sun, 07 May 2006 10:38:53 +0430, Igor Vaynberg
[EMAIL PROTECTED] wrote:
i return to first post :
2- if a user view/edit 4 book in its session , for him/her created 4
BookDetails and 4 EditBook page object.(why this needed)
i can not understand why this needed ?
in mailing list ,
Igor Vaynberg wrote:
and how do i validate the resulting value? in updateModel, if i find
the combined value is not right, there's not much i can do.
you mean the sum of correct pieces is not always correct itself?!? you
can create a validator that performs the same aggregation
Hey Guys,I think adding something to support localized attributes would be a big plus. I know that when Wicket added wicket:message it really saved a lot of extra Java code on my part. I think the same thing would happen if we came up with something to do it in attributes. AttributeModifiers
why not a format like thisinput type=submit value=wicket:i18n:buttons.save/or input type=submit value=wi18n:buttons.save/-Igor
On 5/8/06, Andrew Berman [EMAIL PROTECTED] wrote:
Hey Guys,I think adding something to support localized attributes would be a big plus. I know that when Wicket added
here is a simple examplethe user is on the BookDetails page. the book details has a tabbed view that lets user tab through general/about author/reviews/comments tabs.the user logs in and views details about book 2 - so you get the singleton page, set book id to 2, and show it to the user
the user
Ittay Dror wrote:
Igor Vaynberg wrote:
and how do i validate the resulting value? in updateModel, if i find
the combined value is not right, there's not much i can do.
you mean the sum of correct pieces is not always correct itself?!? you
can create a validator that performs the
I have changed the TypeValidator message, in several ways:foo/pkg/ MyWebApplication.properties MyWebApplication.java Application.propertiesfoo/pkg/pages FooPage.properties FooPage.html
FooPage.javaNone of these have worked. :/-- Bruno Borges[EMAIL PROTECTED]Sun Certified Java Programmer for
on the off chance that not only igor is patient with this thread
i noticed that there is an inconsistency (in my view) in how input is handled:
there's Component.getInput()
then FormComponent.inputAsStringArray()
which is used to set
convertedInput - by convert()
and
rawInput - by
it is now :)just added a public non final getInputAsArray()figured if we have a public non final getInput() the array version should be also.-IgorOn 5/8/06,
Ittay Dror [EMAIL PROTECTED] wrote:
Ittay Dror wrote: Ittay Dror wrote: Igor Vaynberg wrote: and how do i validate the resulting value? in
Igor Vaynberg wrote:
it is now :)
just added a public non final getInputAsArray()
then you can implement getInput() similar to getInputAsArray()[0] (checking
for nulls)
figured if we have a public non final getInput() the array version
should be also.
-Igor
On 5/8/06, *Ittay
94 matches
Mail list logo
