Re: [Wicket-user] Why does setResponse sometimes does notResponse?
i havent look at the code but the fix could be something like that you don't push the page after construction on the request stack because if you do that then you will overwrite the page that was pushed in the constructor of that page.. So are there really landing 2 pages on the request stack? johan On 2/19/07, Igor Vaynberg [EMAIL PROTECTED] wrote: at least if in devel mode you should log a warning if you detect this. or do you not detect its in a constructor...you just push a new target? -igor On 2/19/07, Eelco Hillenius [EMAIL PROTECTED] wrote: On 2/19/07, Igor Vaynberg [EMAIL PROTECTED] wrote: sure, but does it abort the page or do you end up with an extra page in the pagemap? No, it doesn't abort the request. If you want to do that you need to use that exception. So while it typically doesn't make sense to call setResponsePage in a constructor, it is not *wrong* nor should Wicket - like we did - ignore it altogether. Eelco - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
the fix is already in. There are no problems, everything works. This thread can be ended. Eelco On 2/20/07, Johan Compagner [EMAIL PROTECTED] wrote: i havent look at the code but the fix could be something like that you don't push the page after construction on the request stack because if you do that then you will overwrite the page that was pushed in the constructor of that page.. So are there really landing 2 pages on the request stack? johan - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
and you made sure everything is being detached properly? :) -igor On 2/20/07, Eelco Hillenius [EMAIL PROTECTED] wrote: the fix is already in. There are no problems, everything works. This thread can be ended. Eelco On 2/20/07, Johan Compagner [EMAIL PROTECTED] wrote: i havent look at the code but the fix could be something like that you don't push the page after construction on the request stack because if you do that then you will overwrite the page that was pushed in the constructor of that page.. So are there really landing 2 pages on the request stack? johan - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
On 2/20/07, Igor Vaynberg [EMAIL PROTECTED] wrote: and you made sure everything is being detached properly? :) Yes. That was the sole idea about keeping a stack with request targets to start with. Eelco - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
umm...that was a joke :) -igor On 2/20/07, Eelco Hillenius [EMAIL PROTECTED] wrote: On 2/20/07, Igor Vaynberg [EMAIL PROTECTED] wrote: and you made sure everything is being detached properly? :) Yes. That was the sole idea about keeping a stack with request targets to start with. Eelco - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
On 2/20/07, Igor Vaynberg [EMAIL PROTECTED] wrote: umm...that was a joke :) Bastard! :) - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
available, and I cannot think of a good end-user reason why this couldn't be called. The getMarkupId method can in fact be called in if it is possible to make every method work at constructor, of course this is not necessary. It's a good idea to shield people from errors that aren't immediately clear until runtime. Instead of that, wouldnt it be easier to make setResponse work forcibly in construction time similar as RestartResponseException does when explicity called? cancelling current page creation ( its component-tree) and force flow to the new page before responding... - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
i tried to do that, but the sentiment at that time was that throwing an exception internally was not explicit enough. maybe times have changed. -igor On 2/19/07, manu [EMAIL PROTECTED] wrote: available, and I cannot think of a good end-user reason why this couldn't be called. The getMarkupId method can in fact be called in if it is possible to make every method work at constructor, of course this is not necessary. It's a good idea to shield people from errors that aren't immediately clear until runtime. Instead of that, wouldnt it be easier to make setResponse work forcibly in construction time similar as RestartResponseException does when explicity called? cancelling current page creation ( its component-tree) and force flow to the new page before responding... - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
sure, but does it abort the page or do you end up with an extra page in the pagemap? -igor On 2/19/07, Eelco Hillenius [EMAIL PROTECTED] wrote: Guys, guys, did you read my reply? I fixed this and you can now use setResponsePage in your constructor or anywhere you want!. I just made this issue http://issues.apache.org/jira/browse/WICKET-298 so that the change will be in the change log. Eelco On 2/19/07, manu [EMAIL PROTECTED] wrote: available, and I cannot think of a good end-user reason why this couldn't be called. The getMarkupId method can in fact be called in if it is possible to make every method work at constructor, of course this is not necessary. It's a good idea to shield people from errors that aren't immediately clear until runtime. Instead of that, wouldnt it be easier to make setResponse work forcibly in construction time similar as RestartResponseException does when explicity called? cancelling current page creation ( its component-tree) and force flow to the new page before responding... - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
On 2/19/07, Igor Vaynberg [EMAIL PROTECTED] wrote: sure, but does it abort the page or do you end up with an extra page in the pagemap? No, it doesn't abort the request. If you want to do that you need to use that exception. So while it typically doesn't make sense to call setResponsePage in a constructor, it is not *wrong* nor should Wicket - like we did - ignore it altogether. Eelco - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
at least if in devel mode you should log a warning if you detect this. or do you not detect its in a constructor...you just push a new target? -igor On 2/19/07, Eelco Hillenius [EMAIL PROTECTED] wrote: On 2/19/07, Igor Vaynberg [EMAIL PROTECTED] wrote: sure, but does it abort the page or do you end up with an extra page in the pagemap? No, it doesn't abort the request. If you want to do that you need to use that exception. So while it typically doesn't make sense to call setResponsePage in a constructor, it is not *wrong* nor should Wicket - like we did - ignore it altogether. Eelco - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
On 2/19/07, Igor Vaynberg [EMAIL PROTECTED] wrote: at least if in devel mode you should log a warning if you detect this. or do you not detect its in a constructor...you just push a new target? Why? If users want to do that, let them. Though it's probably not the smartest thing to do, there's nothing wrong with it either. If you want to log, you should log when a request target is pushed on top of an existing one. If you're worried about efficiency or something, that's a thing you probably never really need, though throwing exceptions around isn't the most elegant thing either. I don't see the problem anyway. I don't really see a good use case for it, but like I said, there is absolutely no reason why calling setResponsePage in the constructor is 'wrong' from an end user's perspective. Go ahead making improvements if you feel like it. I'm ok with how it is now. Eelco - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
On 2/15/07, Carfield Yim [EMAIL PROTECTED] wrote: On 2/16/07, Igor Vaynberg [EMAIL PROTECTED] wrote: you shouldnt be calling setresponsepage inside constructors I think I fixed this for 1.3 and 2.0. Could you please test this and let us know whether that worked for you? It *should* work, but I'm about to take off and don't have time to test this particular instance. Other than setresponsepage(), there are also methods like getMarkupId() , which are not work properly in constructor. May be we can add flag for those object ( or all component ?) to indicate if the object is inited or not, then throws exception and recommend user to use alternative? e.g.: That's an idea, though I hopefully just fixed the first issue so that now you can call setResponsePage at any time the RequestCycle is available, and I cannot think of a good end-user reason why this couldn't be called. The getMarkupId method can in fact be called in the constructor in Wicket 2.0; enabling that was one of the key reasons for doing the constructor change. For Wicket 1.3 I guess we could add a flag, though it would mean a tiny little extra bit of state, and I'm not sure it's going to help that many cases. Maybe clear documentation could be sufficient? Eelco - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
available, and I cannot think of a good end-user reason why this couldn't be called. The getMarkupId method can in fact be called in if it is possible to make every method work at constructor, of course this is not necessary. It just my personal suggestion if this is not possible - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
Do we really need a bit? (haven't we already have a 'bit')? If getmarkupid is called but it isn't set and we want to look it up then we can already see that the markup is not attached right? And if that is not the case throw an exception On 2/17/07, Eelco Hillenius [EMAIL PROTECTED] wrote: On 2/15/07, Carfield Yim [EMAIL PROTECTED] wrote: On 2/16/07, Igor Vaynberg [EMAIL PROTECTED] wrote: you shouldnt be calling setresponsepage inside constructors I think I fixed this for 1.3 and 2.0. Could you please test this and let us know whether that worked for you? It *should* work, but I'm about to take off and don't have time to test this particular instance. Other than setresponsepage(), there are also methods like getMarkupId() , which are not work properly in constructor. May be we can add flag for those object ( or all component ?) to indicate if the object is inited or not, then throws exception and recommend user to use alternative? e.g.: That's an idea, though I hopefully just fixed the first issue so that now you can call setResponsePage at any time the RequestCycle is available, and I cannot think of a good end-user reason why this couldn't be called. The getMarkupId method can in fact be called in the constructor in Wicket 2.0; enabling that was one of the key reasons for doing the constructor change. For Wicket 1.3 I guess we could add a flag, though it would mean a tiny little extra bit of state, and I'm not sure it's going to help that many cases. Maybe clear documentation could be sufficient? Eelco - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
On 2/17/07, Carfield Yim [EMAIL PROTECTED] wrote: available, and I cannot think of a good end-user reason why this couldn't be called. The getMarkupId method can in fact be called in if it is possible to make every method work at constructor, of course this is not necessary. It just my personal suggestion if this is not possible It's a good idea to shield people from errors that aren't immediately clear until runtime. So I agree with that. If we can provide better reporting on such errors without too much costs, I'm all for that. Eelco - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
you shouldnt be calling setresponsepage inside constructors try RestartResponseException -igor On 2/15/07, Manu [EMAIL PROTECTED] wrote: Hallo, I have this web-flow sequences (two branches to the same leaf): [web-flow 1] WebPage1 WebPage2 [+Applet] WebPage3 [web-flow 2] WebPage1B [+Applet] WebPage2 [+Applet] WebPage3 The Applet is the same in all cases: is an Applet that digitally signs (windows keystore for certs) and automatically launches a javascript event onSignOK(), which does document.formNavigateToPageX.submit(). When trying web-flow 1 things work perfect: first navigate to Page1, then click-on-link and navigate to Page2, which internally checks digital signature, and in case OK it automatically navigates to Page3, by means of setResponse(Page3). That works fine. But the other web-flow 2 does not: navigate to Page1B, internally checks digital signature, and in case OK it navigates to Page2 by means of setResponse(Page2). That works fine... but when then arrives Page2, it internally checks another digital signature, and in case OK it should navigate to Page3.. but not. Internally it also does a setResponse(Page3)... but after the constructor endes, nothing happens... it executes and renders, and the flow dies at Page2. Any idea what can be wrong here? Dankeschen! Manu - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ Wicket-user mailing list Wicket-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wicket-user
Re: [Wicket-user] Why does setResponse sometimes does notResponse?
On 2/16/07, Igor Vaynberg [EMAIL PROTECTED] wrote:
you shouldnt be calling setresponsepage inside constructors
try RestartResponseException
-igor
Other than setresponsepage(), there are also methods like
getMarkupId() , which are not work properly in constructor. May be we
can add flag for those object ( or all component ?) to indicate if the
object is inited or not, then throws exception and recommend user to
use alternative? e.g.:
boolean inited = false;
[constructor] () {
.
inited = true;
}
public String getMarkupId() {
if(!inited) {
throws new illegalstateexception(Cannot use getMarkupId() at
constructor, please add AttributeModifier and specify you own Markup
ID);
} else
return markupId;
}
May be more difficult to do similar thing at
RequestCycle.setresponsepage() , what do you think about this?
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV
___
Wicket-user mailing list
Wicket-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wicket-user
