You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@wicket.apache.org by "Jan Kriesten (JIRA)" <ji...@apache.org> on 2007/08/13 11:05:43 UTC
[jira] Created: (WICKET-842) is broken
again...
<html wicket:id="html"> is broken again...
------------------------------------------
Key: WICKET-842
URL: https://issues.apache.org/jira/browse/WICKET-842
Project: Wicket
Issue Type: Bug
Components: wicket
Affects Versions: 1.3.0-beta3
Environment: JDK 1.6 / OpenSuSE 10.2
Reporter: Jan Kriesten
hi,
some changes in latest trunk broke adding a wicket:id to <html> again! :-(
actually, the wicket:id in <html> works. but extending such a basepage and then
adding a component to it doesn't work any more! the component hierarchy doesn't
seem to be resolved correctly.
test case is attached.
regards, --- jan.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (WICKET-842) is broken
again...
Posted by "Jan Kriesten (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/WICKET-842?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12519593 ]
Jan Kriesten commented on WICKET-842:
-------------------------------------
Having <html> behaving like any other tag is IMHO fine.
But I think it a problem that one has to know what component hierarchy has been set up in base Page classes (and which has been completely handled by these). If I change the base Page class to <html wicket:id> or remove that again, the whole application based on that class shouldn't break.
Shouldn't <html> be a transparent resolver by default?
Also, can <html wicket:id="xy"> be overwritten by child pages? What happens, if the child page has an <html wicket:id="yz">?
Regards, --- Jan.
> <html wicket:id="html"> is broken again...
> ------------------------------------------
>
> Key: WICKET-842
> URL: https://issues.apache.org/jira/browse/WICKET-842
> Project: Wicket
> Issue Type: Bug
> Components: wicket
> Affects Versions: 1.3.0-beta3
> Environment: JDK 1.6 / OpenSuSE 10.2
> Reporter: Jan Kriesten
> Attachments: test.tar.gz
>
>
> hi,
> some changes in latest trunk broke adding a wicket:id to <html> again! :-(
> actually, the wicket:id in <html> works. but extending such a basepage and then
> adding a component to it doesn't work any more! the component hierarchy doesn't
> seem to be resolved correctly.
> test case is attached.
> regards, --- jan.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Resolved: (WICKET-842) is broken
again...
Posted by "Juergen Donnerstag (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/WICKET-842?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Juergen Donnerstag resolved WICKET-842.
---------------------------------------
Resolution: Invalid
Fix Version/s: 1.3.0-beta4
Assignee: Juergen Donnerstag
works as expected
Juergen
> <html wicket:id="html"> is broken again...
> ------------------------------------------
>
> Key: WICKET-842
> URL: https://issues.apache.org/jira/browse/WICKET-842
> Project: Wicket
> Issue Type: Bug
> Components: wicket
> Affects Versions: 1.3.0-beta3
> Environment: JDK 1.6 / OpenSuSE 10.2
> Reporter: Jan Kriesten
> Assignee: Juergen Donnerstag
> Fix For: 1.3.0-beta4
>
> Attachments: test.tar.gz
>
>
> hi,
> some changes in latest trunk broke adding a wicket:id to <html> again! :-(
> actually, the wicket:id in <html> works. but extending such a basepage and then
> adding a component to it doesn't work any more! the component hierarchy doesn't
> seem to be resolved correctly.
> test case is attached.
> regards, --- jan.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (WICKET-842) is broken
again...
Posted by "Igor Vaynberg (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/WICKET-842?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12519604 ]
Igor Vaynberg commented on WICKET-842:
--------------------------------------
> Shouldn't <html> be a transparent resolver by default?
i dont know. the idea was to make it behave just like a regular tag - and regular tags are not transparent.
> Also, can <html wicket:id="xy"> be overwritten by child pages? What happens, if the child page has an <html wicket:id="yz">?
you should not have more then one html tag defined in your template hierarchy. the base page should define it, and every derivative page should inherit it. otherwise this problem applies to any tag and would be treated the same.
> <html wicket:id="html"> is broken again...
> ------------------------------------------
>
> Key: WICKET-842
> URL: https://issues.apache.org/jira/browse/WICKET-842
> Project: Wicket
> Issue Type: Bug
> Components: wicket
> Affects Versions: 1.3.0-beta3
> Environment: JDK 1.6 / OpenSuSE 10.2
> Reporter: Jan Kriesten
> Attachments: test.tar.gz
>
>
> hi,
> some changes in latest trunk broke adding a wicket:id to <html> again! :-(
> actually, the wicket:id in <html> works. but extending such a basepage and then
> adding a component to it doesn't work any more! the component hierarchy doesn't
> seem to be resolved correctly.
> test case is attached.
> regards, --- jan.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (WICKET-842) is broken
again...
Posted by "Igor Vaynberg (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/WICKET-842?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12519815 ]
Igor Vaynberg commented on WICKET-842:
--------------------------------------
i completely disagree, i have plenty layouts where inheritance includes things that are inside body as well...
<html>
<body>
<div wicket:id="container">
<wicket:child/>
</div>
</body>
</html>
so in that case you still have to make container div transparent. its just how wicket works. i dont think we should make anything transparent by default, it just confuses people.
> <html wicket:id="html"> is broken again...
> ------------------------------------------
>
> Key: WICKET-842
> URL: https://issues.apache.org/jira/browse/WICKET-842
> Project: Wicket
> Issue Type: Bug
> Components: wicket
> Affects Versions: 1.3.0-beta3
> Environment: JDK 1.6 / OpenSuSE 10.2
> Reporter: Jan Kriesten
> Attachments: test.tar.gz
>
>
> hi,
> some changes in latest trunk broke adding a wicket:id to <html> again! :-(
> actually, the wicket:id in <html> works. but extending such a basepage and then
> adding a component to it doesn't work any more! the component hierarchy doesn't
> seem to be resolved correctly.
> test case is attached.
> regards, --- jan.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (WICKET-842) is broken
again...
Posted by "Jan Kriesten (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/WICKET-842?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12519867 ]
Jan Kriesten commented on WICKET-842:
-------------------------------------
Igor,
your example is very different from this case. You explicitely build a hierarchy with the div - the html would live without it, too.
But you always have to have a <html> and a <body>, so these are entry points for headers and markup and so are created by wicket if not defined! And as these are defined as a container by default (else you couldn't add a header or elements to the page at all), there is actually not really a change in the hierarchy if you add an id to <html>.
But at last, it's your decision, how wicket's to behave. :-) Case closed.
> <html wicket:id="html"> is broken again...
> ------------------------------------------
>
> Key: WICKET-842
> URL: https://issues.apache.org/jira/browse/WICKET-842
> Project: Wicket
> Issue Type: Bug
> Components: wicket
> Affects Versions: 1.3.0-beta3
> Environment: JDK 1.6 / OpenSuSE 10.2
> Reporter: Jan Kriesten
> Attachments: test.tar.gz
>
>
> hi,
> some changes in latest trunk broke adding a wicket:id to <html> again! :-(
> actually, the wicket:id in <html> works. but extending such a basepage and then
> adding a component to it doesn't work any more! the component hierarchy doesn't
> seem to be resolved correctly.
> test case is attached.
> regards, --- jan.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (WICKET-842) is broken
again...
Posted by "Juergen Donnerstag (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/WICKET-842?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12519455 ]
Juergen Donnerstag commented on WICKET-842:
-------------------------------------------
Not broken. Actually it now works like it should be.
<html wicket:id="html">
<body>
<a wicket:id="label">xxx</a>
</body>
</html>
public class MyPage
{
public MyPage()
{
MarkupContainer html = new WebMarkupContainer("html");
add(html);
html.add(new Label("label", "xxxx");
....
Note that <html> is no longer treated different compared to any other tag. "label" is a child of "html" and that should be reflected in the java component hierarchy as well.
Please close the issue if you code works once modified. Thanks.
> <html wicket:id="html"> is broken again...
> ------------------------------------------
>
> Key: WICKET-842
> URL: https://issues.apache.org/jira/browse/WICKET-842
> Project: Wicket
> Issue Type: Bug
> Components: wicket
> Affects Versions: 1.3.0-beta3
> Environment: JDK 1.6 / OpenSuSE 10.2
> Reporter: Jan Kriesten
> Attachments: test.tar.gz
>
>
> hi,
> some changes in latest trunk broke adding a wicket:id to <html> again! :-(
> actually, the wicket:id in <html> works. but extending such a basepage and then
> adding a component to it doesn't work any more! the component hierarchy doesn't
> seem to be resolved correctly.
> test case is attached.
> regards, --- jan.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (WICKET-842) is broken
again...
Posted by "Igor Vaynberg (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/WICKET-842?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12519483 ]
Igor Vaynberg commented on WICKET-842:
--------------------------------------
although what you can do is
public class MyPage
{
public MyPage()
{
MarkupContainer html = new WebMarkupContainer("html") {
public boolean isTransparentResolver() { return true; }
};
add(html);
add(new Label("label", "xxxx");
> <html wicket:id="html"> is broken again...
> ------------------------------------------
>
> Key: WICKET-842
> URL: https://issues.apache.org/jira/browse/WICKET-842
> Project: Wicket
> Issue Type: Bug
> Components: wicket
> Affects Versions: 1.3.0-beta3
> Environment: JDK 1.6 / OpenSuSE 10.2
> Reporter: Jan Kriesten
> Attachments: test.tar.gz
>
>
> hi,
> some changes in latest trunk broke adding a wicket:id to <html> again! :-(
> actually, the wicket:id in <html> works. but extending such a basepage and then
> adding a component to it doesn't work any more! the component hierarchy doesn't
> seem to be resolved correctly.
> test case is attached.
> regards, --- jan.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (WICKET-842) is broken
again...
Posted by "Jan Kriesten (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/WICKET-842?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12519616 ]
Jan Kriesten commented on WICKET-842:
-------------------------------------
IMHO the problem is, that you don't always know what the parent defines. So when you extend a page class, you should always be able to add components without knowing what structure has been defined yet.
<html> (and maybe <head> and <body>, too) are special tags in that matter. They are taken as given when not specifically defined with a wicket:id. You automatically add to them - and they're generated for you if not defined. No other tags are generated by wicket, so in that respect they're outstanding anyway.
> <html wicket:id="html"> is broken again...
> ------------------------------------------
>
> Key: WICKET-842
> URL: https://issues.apache.org/jira/browse/WICKET-842
> Project: Wicket
> Issue Type: Bug
> Components: wicket
> Affects Versions: 1.3.0-beta3
> Environment: JDK 1.6 / OpenSuSE 10.2
> Reporter: Jan Kriesten
> Attachments: test.tar.gz
>
>
> hi,
> some changes in latest trunk broke adding a wicket:id to <html> again! :-(
> actually, the wicket:id in <html> works. but extending such a basepage and then
> adding a component to it doesn't work any more! the component hierarchy doesn't
> seem to be resolved correctly.
> test case is attached.
> regards, --- jan.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (WICKET-842) is broken
again...
Posted by "Igor Vaynberg (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/WICKET-842?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12520023 ]
Igor Vaynberg commented on WICKET-842:
--------------------------------------
> But you always have to have a <html> and a <body>, so these are entry points for headers and markup and so are created by wicket if not
> defined! And as these are defined as a container by default (else you couldn't add a header or elements to the page at all), there is actually
> not really a change in the hierarchy if you add an id to <html>.
what juergen tried to explain is that we no longer create those special containers. so that means you are in control of the entire markup now. if you choose to add something to it you have to do it propertly - and that means sometimes override istransparentresolver to return true.
> <html wicket:id="html"> is broken again...
> ------------------------------------------
>
> Key: WICKET-842
> URL: https://issues.apache.org/jira/browse/WICKET-842
> Project: Wicket
> Issue Type: Bug
> Components: wicket
> Affects Versions: 1.3.0-beta3
> Environment: JDK 1.6 / OpenSuSE 10.2
> Reporter: Jan Kriesten
> Attachments: test.tar.gz
>
>
> hi,
> some changes in latest trunk broke adding a wicket:id to <html> again! :-(
> actually, the wicket:id in <html> works. but extending such a basepage and then
> adding a component to it doesn't work any more! the component hierarchy doesn't
> seem to be resolved correctly.
> test case is attached.
> regards, --- jan.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (WICKET-842) is broken
again...
Posted by "Jan Kriesten (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/WICKET-842?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jan Kriesten updated WICKET-842:
--------------------------------
Attachment: test.tar.gz
TestCase
> <html wicket:id="html"> is broken again...
> ------------------------------------------
>
> Key: WICKET-842
> URL: https://issues.apache.org/jira/browse/WICKET-842
> Project: Wicket
> Issue Type: Bug
> Components: wicket
> Affects Versions: 1.3.0-beta3
> Environment: JDK 1.6 / OpenSuSE 10.2
> Reporter: Jan Kriesten
> Attachments: test.tar.gz
>
>
> hi,
> some changes in latest trunk broke adding a wicket:id to <html> again! :-(
> actually, the wicket:id in <html> works. but extending such a basepage and then
> adding a component to it doesn't work any more! the component hierarchy doesn't
> seem to be resolved correctly.
> test case is attached.
> regards, --- jan.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.