a full copy of the comment
thread,
including the initial issue submission, for this request,
not just the latest update.
Category: core
Group: 1.1
Status: Open
Priority: 5
Submitted By: Matej Knopp (knopp)
Assigned to: Nobody/Anonymous (nobody)
Summary: Introduce a factory for MostRecentlyUsedMap
wrote:
Matej Knopp wrote:
I can think of one thing, that's kind of missing. Although I don't
really need it, maybe someone alse does.
It's redirect to bookmarkable page. I think it's not possible to
redirect to a bookmarkable page (or at least not trivial), or am I wrong?
Maybe this could
)
But i agree having a choice to redirect the homepage to a bookmarkable
url would also be nice.
johan
Matej Knopp wrote:
Johan Compagner wrote:
what we could do if you do this:
setResponsePage(MyPage.class);
So setting a class instead of an instance.
That is always a redirect (why
Gili wrote:
First, I like the new setPageAlias() functionality :)
I still don't understand however why one needs to store the Page
component ID on the session (at least for bookmarkable pages the first
time they are hit). Also, Eelco previously mentioned that jsessionid is
only
Gili wrote:
http://www.google.com/webmasters/guidelines.html
Note that requiring cookies might prevent the crawler from indexing
your pages.
This is further discussed here:
http://www.searchengineguide.com/whalen/2004/0210_jw1.html and
Once again. The URLS of wicket application that need to be crawled are
bookmarkable pages (urls like app?bookmarkablePage=foo.bar.pageClass).
Bookmarkable pages have only one parameter. So they are already crawler
friendly.
All other (app?component=, app?path=, etc) should not be crawled at
...?
Gili
Matej Knopp wrote:
Gili wrote:
http://www.google.com/webmasters/guidelines.html
Note that requiring cookies might prevent the crawler from
indexing your pages.
This is further discussed here:
http://www.searchengineguide.com/whalen/2004/0210_jw1.html and
http
on the Wicket website and track how well it gets indexed? Which
pages get indexed and which don't and why...?
Gili
Matej Knopp wrote:
Gili wrote:
http://www.google.com/webmasters/guidelines.html
Note that requiring cookies might prevent the crawler from
indexing your pages
clicking on a bookmarkable page you wouldn't be able to continue working
on application.
I don't think jsessionid as a problem here. Google indexes it well.
Maybe it's not even generated for google. Maybe google crawler pretends
to support cookies and just throws them away.
-Matej
johan
Matej
bookmarkable urls that we don't
call encodeUrl()
Can we do that? Then there will be no session info on those urls.
johan
Matej Knopp wrote:
But they do index them. They just ignore the session id and everyhing
in url that follows it.
But if you really don't want want session id's in url
Matej Knopp wrote:
No, I'm afraid that wouldn't help. URLS without session id would create
new URL. I don't think it's desired behavior.
I meant ...would create new session (not new URL).
I don't even think we have any problem to solve here, as google indexes
wicket examples pretty well
and what's not.
-Matej
juergen
On 7/18/05, Matej Knopp [EMAIL PROTECTED] wrote:
Johan Compagner wrote:
who is inserting the jsessionid?
I think it happens when we call encodeUrl() maybe we shouldn't do that
for bookmarkable urls?
Don't think so. When cookies are disabled clicking on URL
Yes, I never claimed anything else. (At least not intentionally :)
-Matej
Juergen Donnerstag wrote:
yes, but it is the container's responsibilty. It is not Wicket adding it.
Juergen
On 7/18/05, Matej Knopp [EMAIL PROTECTED] wrote:
wicket.protocol.httpWebResponse#encodeURL
calls
magically support cookies, what
about the hundreds of different other crawlers that might not? Besides,
we don't even know that Google does cookies :) I'd rather not make this
assumption if I can avoid it.
Matej Knopp wrote:
No, I'm afraid that wouldn't help. URLS without session id would
create
I've attached simple filter that conditionaly disables/enables url
encoding. Maybe ther's even simpler solution I don't know about.
Only thing you have to do is to detect if user agent is a crawler. There
are lists of crawlers useragents strings on internet.
-Matej
Matej Knopp wrote:
I
Bookmarkable pages have only one argument - page class.
-Matej
Gili wrote:
Sorry, not stable URLs (we had that before). I meant we reduced the
number of query string arguments so bookmarkable pages have a greater
chance of getting indexed properly.
Gili
Gili wrote:
Stable URLs
One way is using constructors. Pages are objects like every other.
so this way you create a link that will create a new page, and redirect
to it:
class Page1 extends Page {
public Page1() {
add(new Link(link1) {
public void onClick() {
Page2 page = new Page2(Something,
Johan Compagner wrote:
This can be done by multiply ways
First construct the Page for the link and use the
public PageLink(final String id, final Page page)
constructor
Then when that link is clicked that page is returned with all the state
you inserted into it.
If you want to do lazy init
Reread the thread. The result is roughly that wicket:extend will
disapear from the output.
About doctype, I think it's developer's responsibility to ensure that
the parent and child doctypes are the same.
I think special treatment of this would only bloat wicket, while not
introducing any
About setResponsePage(Class)
I think, there should also be members of wicket.{age that would accept
class as parameters (like there is SetResponsePage(wicket.Page)
e.g.
setResponsePage(Class pageClass) {
getRequestCycle().setResponsePage(pageClass);
}
as well as one that would take
all those little shortcut methods (getRequestCycle().)
But if more people like to have that.
johan
Matej Knopp wrote:
About setResponsePage(Class)
I think, there should also be members of wicket.{age that would accept
class as parameters (like there is SetResponsePage(wicket.Page)
e.g
Better in all possible ways? Does that include using hidden forms and
not to be able using web application without javascript? Thus not be
able to use the application in more than one window? Well, doesn't seem
to me as an advantage at all.
-Matej
Gili wrote:
The only downside I see in
Massive and ugly. JSF examples in stateless mode contain form with 7kb
big hidden field (view state). Wouldn't want to use that application
with GPRS :)
-Matej
Phil Kulak wrote:
You'd also have to have everything live inside a form with some
JavaScript trickery. Right now it's nice that a
Well, I don't think it is possible to accomplish, what he wants. In
constructor, he wants to get part of parkup. (Component body)
But the markup is dealt with on rendering. In constructor, AFAIK the
markup is not known yet.
-Matej
Juergen Donnerstag wrote:
provided I understood your problem,
Hi.
I know that there's not a lot of interest in this :)
but I just wonder if anyone actually objects, if this functionality
would be introducted to wicket.
Basicaly, it would allow to specify own MostRecentlyUsedPageMap instance.
class Application
{
...
public Map
I'm not a wicket developer, but my opinion is that vast majority of
wicket users will never need it, and if some do, it's easy to acomplish
it with custom request cycle or filter. Thus, I don't see any reason for
this to be in the core.
Maybe a wiki page would be appropriate.
-Matej
Gili
is a showstopper, I
can survive it. I'll just have to use a patched version of wicket ;)
-Matej
Eelco Hillenius wrote:
Could you supply a patch please?
Eelco
Matej Knopp wrote:
Hi.
I know that there's not a lot of interest in this :)
but I just wonder if anyone actually objects
Definitely yes. I think it its important to preserve the original ID,
because it might be reffered to from CSS.
-Matej
Eelco Hillenius wrote:
Ah, ok... now I get it.
Now that I think of it... DatePicker should use the id that is attached
to the tag if there is one, and only generate one
Hmm.. When I wanted to accomplish similiar thing I used Form.validate()
and Form.markComponentsInvalid instead of default form processing. The
result was that the values were stored in FormComponents' invalidInput
properties. Nevertheless, this really isn't the cleanest approach and
seems
Hi.
I'm using datepicker and I'm constantly getting this warning
17:46:32.146 WARN!! [SocketListener0-3]
wicket.util.resource.UrlResourceStream.init(UrlResourceStream.java:92)
22 cannot convert url:
Not that my opinion should matter much, but have you considered, that if
you release 1.2 without the refactoring, you'll have to maintain at
least three different versions?
1.1, 1.2 and new 2.0.
While if you do the changes in 1.2, you'll only have to maintain 1.1 and
1.2. Maintaining previous
Would it be possible to make IResourceStreamLocator.locate take class
instead of class loader?
And is it all right that ResourceSettings.setResourceStreamLocator takes
class (ResourceStreamLocator) instead of an interface?
-Matej
---
Hi,
I want to ask, what's the reason of having all class members sorted by
alphabetical order?
I mean, look at RequestCycle. It has constants at the beginning, but
they are not sorted by value [request cycle phases] but alphabetically.
It's quite confusing. Is there a good reason for this?
Hi, is there any chance you could post it as attachment? No luck getting
wicket-stuff from anon cvs...
-Matej
Igor Vaynberg wrote:
hey guys,
just implemented partial page rendering, allowing for rendering of any
number of components on the page.
see
is in wicket not wicket-core.
-Igor
On 1/30/06, Matej Knopp [EMAIL PROTECTED] wrote:
Hi, is there any chance you could post it as attachment? No luck
getting
wicket-stuff from anon cvs...
-Matej
Igor Vaynberg wrote:
hey guys,
just implemented partial page rendering, allowing
I think that maybe stateless page should be always accessed by
bookmarkable url. The only difference between stateless page and regular
(bookmarkable) page would be that stateless page will not be put in the
pagemap.
But I don't know about setResponsePage. Maybe if
Hi,
I noticed that you can't do a partial render of a component in list
view. As well as get it's markup attributes. The problem seems to be
Component#initializeMarkupStream. IMHO the function should convert
component path from fooListX:3:fooListY:4:fooContainer to
fooListX:fooListY:fooContainer.
On 2/11/06, Juergen Donnerstag [EMAIL PROTECTED] wrote:
On 2/11/06, Matej Knopp [EMAIL PROTECTED] wrote:
Hi,
I noticed that you can't do a partial render of a component in list
view. As well as get it's markup attributes. The problem seems to be
Component#initializeMarkupStream. IMHO
Can't markup stream cache it?
Markup.findComponentIndex(path, id) {
Integer pos = cache.get(path+:+id) ;
if (pos != null)
return pos;
else {
...
}
}
On 2/11/06, Johan Compagner [EMAIL PROTECTED] wrote:
fully agree IMO This is all very much experimental only. It is the
first
Hi, I did try to implement caching of componentTag positions and as a
proof of concept it worked quite well.
Markup.java:
/**
* Initialize the index where head can be found.
*/
private void initialize()
{
if (markup != null)
association in the constructor.
Juergen
On 2/12/06, Matej Knopp [EMAIL PROTECTED] wrote:
Hi, I did try to implement caching of componentTag positions and as a
proof of concept it worked quite well.
Markup.java:
/**
* Initialize the index where head can
Juergen Donnerstag wrote:
On 2/12/06, Matej Knopp [EMAIL PROTECTED] wrote:
I see... It seemed quite easy, implementing the caching. Except that
this can't work for borders and it's very hacky with repeating
listviews. I only bother about these issues because of ajax. So the
current state
the right time to extend the Component API even further. I
predict/assume/hope the problem will be much easier/more elegant be
solved with Wicket 2 and the parent being available in the constructor
and markup association in the constructor.
Juergen
On 2/12/06, Matej Knopp [EMAIL PROTECTED] wrote:
Hi
Hi,
Is there any good reason why this shouldn't (couldn't) be implemented?
I really don't want to use versioning, but without versioning the
ListView is quite dangerous.
-Matej
---
This SF.Net email is sponsored by xPML, a groundbreaking
Johan Compagner wrote:
Hi,
Now we have pretty much solved the new tab/window issue by auto
generation pagemaps.
There is another problem with the access stack a pagemap does have.
when i have this access stack:
Page1, Page2, Page3, Page4
and i go to Page2 and press refresh
Then wicket
try to attach the tabpanel to div, instead of span. Then it works. (at
least for me). You also need to change css then, of course.
And add
var wicketGlobalAjaxErrorHandler = undefined;
to wicket-ajax.js. Without it IE is complaining that
wicketGlobalAjaxError handler is not defined.
Btw.
Matej Knopp wrote:
try to attach the tabpanel to div, instead of span. Then it works. (at
least for me). You also need to change css then, of course.
And add
var wicketGlobalAjaxErrorHandler = undefined;
to wicket-ajax.js. Without it IE is complaining that
wicketGlobalAjaxError handler
Hi,
It seems to work :)
I've just made some minor tweaks (code below). One more thing.
StringResponse still doesn't encode URLs correctly. String response is
the one responsible for rendering head. So in current state, opening
link in new tab without cookies leads to session expiration :(
One more thing. I believe the getCookie, deleteCookie, etc. should be
properly namespaced, e.g. wicketGetCookie, wicketDeleteCookie, etc. so
they wouldn't interfere with possibly existing functions.
-Matej
Matej Knopp wrote:
Hi,
It seems to work :)
I've just made some minor tweaks (code
What subclasses? There are only 3 subclasses of TextField in current
wicket core, HiddenField, PasswordTextField, and RequiredTextField.
HiddenField and PasswordTextField _do_ check for input type.
-Matej
SourceForge.net wrote:
Bugs item #1470901, was opened at 2006-04-15 16:58
Message
()
instead of using the current response. So that the url encoding is
always done on the servlet response.
wil namespace the javascript functions.
johan
On 4/16/06, *Matej Knopp* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] wrote:
Hi,
It seems to work :)
I've just made some minor
I see, it really doesn't.
Should there even be the check? I mean input without type='text' is
valid, right? Adding this check could break a lot of existing code.
-Matej
Johan Compagner wrote:
RequiredTextField doesn't check currently.
johan
On 4/16/06, *Matej Knopp* [EMAIL PROTECTED
:
it already checked everytime
Except when you use the RequiredTextField then the check was not done
suddenly.
i committed a change that the basic textfields and its basic subclasses
always checks for the type
And that can be overridden by overriding a method.
johan
On 4/16/06, *Matej Knopp
Hi,
there's a bug in calendar from wicket-contrib. All UTF-8 strings are
encoded twice. I'm attaching fixed version, can anyone please commit it?
-Matej
calendar-sk-utf8.js
Description: JavaScript source
Hi,
could anyone take a look on AbstractPageableView#getCurrentPage (line
252)? Shouldn't the line read page = Math.max(_getPageCount()_ - 1, 0)
instad?
-Matej
---
Using Tomcat but need to do more? Need to support web services, security?
Johan Compagner wrote:
Hi,
first currently we have IConverter and ITypeConverter , i want to
combine those 2 to be one interface
And then a converter must be easy to make but also to understand.
So i can do this:
IConverter.convertToObject (String, Class)
IConverter.convertToString(Object)
I am quite happy with you current proposition. It seem to be flexible
enough to allow compound converters, while also being simple enough to
make your own converter. And there is only one interface, which is IMHO
a lot better and a lot less confusing than current IConverter and
ITypeConverter
Eelco Hillenius wrote:
I'm still a bit worried about the naming etc.
In my opinion, the thing we want to do is to do input-output
conversion for models. Models could do this themselves, but pulling
the conversion out makes things way more flexible. So... as this is
what we are doing, I would
I kind of like this one. Except that it doesn't allow
CompoundConverters. But then again, with concept like this there is no
need for compound converters.
-Matej
Igor Vaynberg wrote:
yes i know it was never a stack, i propose to make it a stack
i was talking about the usecase of replacing
compound that is still possible because that is
ISupplyConveter
johan
On 5/18/06, *Matej Knopp* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] wrote:
I kind of like this one. Except that it doesn't allow
CompoundConverters. But then again, with concept like this there is no
need
Hi,
I need to specify locale for a component (not whole session). I thought
that overriding component.getLocale will do it, but it's final.
Is there a specific reason for this? What's the reason of having
component.getLocale(), if only thing it does is return
Session.getLocale() and there
You are right. It feels like misusing generics. I think generics were
not meant to be used at runtime (type erasure is not there for fun :) ).
-Matej
Eelco Hillenius wrote:
Hi,
I've got a start for a generified IVisitor implementation. Basically,
it would be used like this:
// Visit
Component.IVisitorForm(Form.class)
thats the only way to do it right now i think
-Igor
On 5/28/06, Matej Knopp [EMAIL PROTECTED] wrote:
You are right. It feels like misusing generics. I think generics were
not meant to be used at runtime (type erasure is not there for fun
:) ).
-Matej
Eelco
, Matej Knopp [EMAIL PROTECTED] mailto:[EMAIL PROTECTED]
wrote:
I never said it's not doable. It just doesn't feel right to me.
I like IVisitor the way it works now.
-Matej
Eelco Hillenius wrote:
Well, like my patch shows
think it
would be better solution.
johan
On 5/29/06, *Matej Knopp* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED]
wrote:
It's not too late to do it this way, is it?
-Matej
Jonathan Locke wrote:
true, but until we added
, i think it might be
smarter to make ITraversal a class. that way we can extend
it however we want in the future. there might be more to traversing
than just a mode.
On 5/29/06, *Matej Knopp* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] wrote:
Just one thing, considering the proposed IVisitor
Strange. SetResponseType always worked for me (jetty, tomcat, sun
application server)
-Matej
Johan Compagner wrote:
Some have problems because a container doesn't set the right header
responses when response.setCharsetEncoding() is not called
But that is a 2.4 method not a 2.3.
Can we
This is for wicket to properly support multiple tabs/windows. If you
open a link in new tab / window, page should be stored in different
pagemap. Does the javascript cause you any problems?
-Matej
[EMAIL PROTECTED] wrote:
Hello,
I'm currently working on project where are integrated
Sorry, I meant getNestedModel of course (as there's no such thing as
getWrappedModel).
Matej Knopp wrote:
Hi,
I'm going through the big model change in 2.0 and I noticed that now we
have IWrapModel, but IModel still contains getWrappedModel().
Wouldn't it be cleaner
I'm looking bit deeper and it seems that the semantics of IWrapModel is
slightly different than I though, so just silently ignore the question :)
Matej Knopp wrote:
Sorry, I meant getNestedModel of course (as there's no such thing as
getWrappedModel).
Matej Knopp wrote:
Hi,
I'm going
Hi,
we have a problem with session locking. Sometimes, the underlying
database just hangs and the response is not processed properly. The
problem is that HttpSession remains locked.
Wouldn't it be possible to replace the current locking mechanism
( synchronized(lock) { doEverything() } )
, or it really takes that long to get the
data (complicated statistics, reports, etc). In the meanwhile user would
like to work with the rest of the application but he can't.
-Matej
-Igor
On 6/21/06, *Matej Knopp* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED]
wrote:
Hi,
we have
there is a perfect solution here.
-Igor
On 6/21/06, *Matej Knopp* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED]
wrote:
Igor Vaynberg wrote:
not unless all page targets return the same mutex instance :)
The mutex would be stored in session of course
a getlock() which you can override and return something other
then session, just need a way to encode a url so it is resolved to your
own target
-Igor
On 6/21/06, *Matej Knopp* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED]
wrote:
Well, it would be your responsibility not to touch
, but this is only for
corner cases.
-Matej
Matej Knopp wrote:
Sorry, i'm lost. How is overriding getlock() supposed to help me? I need
to lock the session, but not for the whole request. Just for a part
before the db call and the part after.
-Matej
Igor Vaynberg wrote:
in that case we sort
that is not always doable, just a suggestion.
-Igor
On 6/21/06, *Matej Knopp* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED]
wrote:
If there was something like
RequestCycle.lockSession()
RequestCycle.unlockSession()
where in my onClock handler I could call
the problem at
this point, but if we have a possible deadlock in our request
handling, that's obviously a bug. If we have not, it's not a bug, and
any 'fixes' we do should be geared towards clean improvements, not
workarounds.
Eelco
On 6/21/06, Matej Knopp [EMAIL PROTECTED] wrote:
I'm also
of moving getNestedModel into a separate interface
because not all models have nested models, and im not even sure we need it
period so removing it completely is also an option imho
thoughts?
-igor
On 6/21/06, Matej Knopp [EMAIL PROTECTED] wrote:
I'm looking bit deeper and it seems
without IWrapModel though. It's used in several wrapOnxx methods,
and it's used to keep the parameterized type. I didn't look at it in a
lot of detail though, so Igor/ Johan will know that better.
Eelco
On 6/22/06, Matej Knopp [EMAIL PROTECTED] wrote:
I also think we should get rid
Btw., ICompoundModel is no longer used, is it?
Matej Knopp wrote:
public interface IInheritableModelT extends IModelT
{
C IModelC wrapOnInhertance(ComponentC component);
}
But maybe it makes sense to have a Marker interface for inherited models.
-Matej
Eelco Hillenius wrote
I think we shouldn't want to get rid of the class argument. Trying to
find out the type in runtime seems like abusing reflection to me.
Generics were not meant to be used as runtime.
I suggest the TextField constructor changes to
TextField(MarkupContainer parent, String id, ClassT type)
or
Okay, forget IModel? extends T - that doesn't really make sense.
Matej Knopp wrote:
I think we shouldn't want to get rid of the class argument. Trying to
find out the type in runtime seems like abusing reflection to me.
Generics were not meant to be used as runtime.
I suggest
Yes, window.name is specific for each frame. It can be different of the
window.name of the frameset.
-Matej
Johan Compagner wrote:
does a frame really has another window.name http://window.name even if
they are in the same window?
On 6/30/06, *Matej Knopp* [EMAIL PROTECTED] mailto:[EMAIL
://localhost:8080/we/linkomatic once again. I suppose it would
redirect to http://localhost:8080/we/linkomatic?wicket:pageMap=wicket-0
. The reason is that pagemap is no longer empty and window.name is null.
Cookies approach solve this, because onUnload deletes the pagemap cookie.
-Matej
Matej Knopp
Hi,
I've checkout the entire branch of WICKET_1_2 (wicket,
wicket-auth-roles, wicket-auth-roles-examples, ... etc) and it builds
fine. Probably because there is pom.xml in the branch folder.
I've also checkout just the wicket directory as separate project, tried
to build it and get the same
Sspitting out the javascript for SecondLevel model does make little sense.
So while I agree that changing application settings silently is not the
best practice, neither is doing nothing. Most of the user will not be
aware what multi window support does and that they should turn it off
with
Hi
There's a known problem with head contribution of components, that are
not rendered when the whole page is (are hidden or not even part of
hierarchy), but are additionally rendered using an AJAX call.
The problem is, that while the body of the component is placed in the
page, the head
test that.
johan
On 7/5/06, * Matej Knopp* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED]
wrote:
Hi
There's a known problem with head contribution of components, that are
not rendered when the whole page is (are hidden or not even part of
hierarchy
Matej Knopp wrote:
Adding both javascript and stylesheets seems to work, at least for
non-inline javascript and css. Still have to test inline stuff.
I tested inline javascript, works well. Inline styles work too, but in
IE you have to use document.createStyleSheet for it to work.
I tested
Hi, as Frank has already stated, it's not the tests that should be
fixed. It is the AjaxRequestTarget code. I'm going to look at it right now.
But maybe I should make some tests with header contribution, that would
make more sense.
I don't like the idea of storing more things in page. I just
Hi,
there's a small issue with ajax on bookmarkable pages.
I you do anything AJAX on bookmarkable page and then refresh it in
browser, new instance is created and you basically lose your state.
I think there is a solution of this. The URL hash (part starting with #)
can be modified with
No, we couldn't. Just ignore this mail (not entire thread).
Matej Knopp wrote:
Btw., taking this even further - e.g. pagemap, page id and version being
part of url hash even for regular pages, we could have history (and back
button) support even for ajax requests.
The only drawback I can
into AjaxLinks everything renders just fine. Let
me know if you need more info.
Thanks,
jim
On 7/14/06, * Frank Bille Jensen* [EMAIL PROTECTED]
mailto:[EMAIL PROTECTED] wrote:
Nice. I will test it to if this fixes the problem.
Frank
On Thu, 2006-07-13 at 20:49 +0200, Matej Knopp
for the first time
And then you close the browser and come back. And the session is still
there then the default pagemap will never be or
used anymore because it will always redirect to another pagemap.
I currently don't see away around that.
johan
On 6/30/06, *Matej Knopp* [EMAIL PROTECTED
...
johan
On 7/21/06, *Matej Knopp* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED]
wrote:
Hi,
does anyone mind these two to be added to markup container?
I know it first one could be little dangerous, because it's dealing with
object that is not fully constructed yet, but I
not a big fan of these kind of methods because of the pittfalls we
can have there. It is very bad programming to call methods on not
fully constructed objects...
johan
On 7/21/06, Matej Knopp [EMAIL PROTECTED] wrote:
Hi,
does anyone mind these two to be added to markup container?
I know
Johan Compagner wrote:
Does a minus (-) sign work?
for example if you would target that id in js:
mypanel-mycomponent
will that be parsed? A minus sign looks to me like it could have the
same problems as the divider (:)
can't we escape _ with 2 __ ?
so first all _ become __ and
I guess if you need this behavior now the easiest way is to do
target.addJavascript(window.location=+);
would be nice to have setRequestTarget though.
-Matej
SourceForge.net wrote:
Feature Requests item #1528244, was opened at 2006-07-25 10:40
Message generated for change (Tracker Item
Ah, I see igor has already responsed this on wicket-user. Should have
read that first :)
-Matej
Matej Knopp wrote:
I guess if you need this behavior now the easiest way is to do
target.addJavascript(window.location=+);
would be nice to have setRequestTarget though.
-Matej
We were hunting a bug with Johan, which turned out to be fact that
AjaxRequestTarget.EncodingResponse is derived from Response instead of
WebResponse.
WebRequestCycle.getReponse casts the response to WebResponse, which
means that all responses used in WebRequestCycle have to be
WebResponses.
1 - 100 of 152 matches
Mail list logo
