You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@wicket.apache.org by Jeremy Thomerson <je...@wickettraining.com> on 2010/04/20 22:49:57 UTC
Rules on adding classes to a release?
I forget what our exact rules are for this. I know that we won't break API
in a point release (i.e. 1.4.8). But, what about the following scenarios?
I can add these to a point release, right?
1. Add a class. In this example, I could add a class that is a custom
subclass of WebRequestCycle that someone can subclass to handle runtime
exceptions in Ajax differently than they do on regular requests. (i.e. they
would rather not redirect on an Ajax error, but instead would like to send
some JS back in a normal Wicket AJAX response)
2. Add an interface and some associated functionality. In this example,
I might have an interface to add, and some additional functionality that
only operated on instances of that interface. No functionality would change
for existing interfaces.
Those would both be okay, right?
--
Jeremy Thomerson
http://www.wickettraining.com
Re: Rules on adding classes to a release?
Posted by Ilja Pavkovic <il...@binaere-bauten.de>.
Hi,
(nearly) the same applies to the change in WICKET-2659: Only if you implement
a constructor in a class with signature
public MyErrorPage(final Throwable throwable, final Page page) { ... } this
constructor is chosen instead of MyErrorPage() { ... }.
Best Regards,
Ilja Pavkovic
Am Donnerstag, 22. April 2010, um 19:23:43 schrieb Igor Vaynberg:
> the comment from juergen implies the patch either changes api or the
> way the existing api work. such changes do not make it into 1.4. the
> changes jeremy was referring to would only activate if the cycle
> implemented an interface so it would not change existing
> functionality.
>
> -igor
>
> On Thu, Apr 22, 2010 at 9:18 AM, Ilja Pavkovic
>
> <il...@binaere-bauten.de> wrote:
> > Hi,
> >
> > applying this rule: why is this change moved to 1.5 ? :
> >
> > https://issues.apache.org/jira/browse/WICKET-2659
> >
> > Best Regards,
> > Ilja Pavkovic
> >
> > Am Mittwoch, 21. April 2010, um 00:06:40 schrieb Igor Vaynberg:
> >> yeah
> >>
> >> -igor
> >>
> >> On Tue, Apr 20, 2010 at 1:49 PM, Jeremy Thomerson
> >>
> >> <je...@wickettraining.com> wrote:
> >> > I forget what our exact rules are for this. I know that we won't
> >> > break API in a point release (i.e. 1.4.8). But, what about the
> >> > following scenarios? I can add these to a point release, right?
> >> >
> >> > 1. Add a class. In this example, I could add a class that is a
> >> > custom subclass of WebRequestCycle that someone can subclass to
> >> > handle runtime exceptions in Ajax differently than they do on regular
> >> > requests. (i.e. they would rather not redirect on an Ajax error, but
> >> > instead would like to send some JS back in a normal Wicket AJAX
> >> > response)
> >> > 2. Add an interface and some associated functionality. In this
> >> > example, I might have an interface to add, and some additional
> >> > functionality that only operated on instances of that interface. No
> >> > functionality would change for existing interfaces.
> >> >
> >> > Those would both be okay, right?
> >> >
> >> > --
> >> > Jeremy Thomerson
> >> > http://www.wickettraining.com
> >
> > --
> > binaere bauten gmbh · tempelhofer ufer 1a · 10961 berlin
> >
> > +49 · 171 · 9342 465
> >
> > Handelsregister: HRB 115854 - Amtsgericht Charlottenburg
> > Geschäftsführer: Dipl.-Inform. Ilja Pavkovic, Dipl.-Inform. Jost Becker
--
binaere bauten gmbh · tempelhofer ufer 1a · 10961 berlin
+49 · 171 · 9342 465
Handelsregister: HRB 115854 - Amtsgericht Charlottenburg
Geschäftsführer: Dipl.-Inform. Ilja Pavkovic, Dipl.-Inform. Jost Becker
Re: Rules on adding classes to a release?
Posted by Igor Vaynberg <ig...@gmail.com>.
the comment from juergen implies the patch either changes api or the
way the existing api work. such changes do not make it into 1.4. the
changes jeremy was referring to would only activate if the cycle
implemented an interface so it would not change existing
functionality.
-igor
On Thu, Apr 22, 2010 at 9:18 AM, Ilja Pavkovic
<il...@binaere-bauten.de> wrote:
> Hi,
>
> applying this rule: why is this change moved to 1.5 ? :
>
> https://issues.apache.org/jira/browse/WICKET-2659
>
> Best Regards,
> Ilja Pavkovic
>
> Am Mittwoch, 21. April 2010, um 00:06:40 schrieb Igor Vaynberg:
>> yeah
>>
>> -igor
>>
>> On Tue, Apr 20, 2010 at 1:49 PM, Jeremy Thomerson
>>
>> <je...@wickettraining.com> wrote:
>> > I forget what our exact rules are for this. I know that we won't break
>> > API in a point release (i.e. 1.4.8). But, what about the following
>> > scenarios? I can add these to a point release, right?
>> >
>> > 1. Add a class. In this example, I could add a class that is a custom
>> > subclass of WebRequestCycle that someone can subclass to handle runtime
>> > exceptions in Ajax differently than they do on regular requests. (i.e.
>> > they would rather not redirect on an Ajax error, but instead would like
>> > to send some JS back in a normal Wicket AJAX response)
>> > 2. Add an interface and some associated functionality. In this
>> > example, I might have an interface to add, and some additional
>> > functionality that only operated on instances of that interface. No
>> > functionality would change for existing interfaces.
>> >
>> > Those would both be okay, right?
>> >
>> > --
>> > Jeremy Thomerson
>> > http://www.wickettraining.com
>
>
> --
> binaere bauten gmbh · tempelhofer ufer 1a · 10961 berlin
>
> +49 · 171 · 9342 465
>
> Handelsregister: HRB 115854 - Amtsgericht Charlottenburg
> Geschäftsführer: Dipl.-Inform. Ilja Pavkovic, Dipl.-Inform. Jost Becker
>
Re: Rules on adding classes to a release?
Posted by Ilja Pavkovic <il...@binaere-bauten.de>.
Hi,
applying this rule: why is this change moved to 1.5 ? :
https://issues.apache.org/jira/browse/WICKET-2659
Best Regards,
Ilja Pavkovic
Am Mittwoch, 21. April 2010, um 00:06:40 schrieb Igor Vaynberg:
> yeah
>
> -igor
>
> On Tue, Apr 20, 2010 at 1:49 PM, Jeremy Thomerson
>
> <je...@wickettraining.com> wrote:
> > I forget what our exact rules are for this. I know that we won't break
> > API in a point release (i.e. 1.4.8). But, what about the following
> > scenarios? I can add these to a point release, right?
> >
> > 1. Add a class. In this example, I could add a class that is a custom
> > subclass of WebRequestCycle that someone can subclass to handle runtime
> > exceptions in Ajax differently than they do on regular requests. (i.e.
> > they would rather not redirect on an Ajax error, but instead would like
> > to send some JS back in a normal Wicket AJAX response)
> > 2. Add an interface and some associated functionality. In this
> > example, I might have an interface to add, and some additional
> > functionality that only operated on instances of that interface. No
> > functionality would change for existing interfaces.
> >
> > Those would both be okay, right?
> >
> > --
> > Jeremy Thomerson
> > http://www.wickettraining.com
--
binaere bauten gmbh · tempelhofer ufer 1a · 10961 berlin
+49 · 171 · 9342 465
Handelsregister: HRB 115854 - Amtsgericht Charlottenburg
Geschäftsführer: Dipl.-Inform. Ilja Pavkovic, Dipl.-Inform. Jost Becker
Re: Rules on adding classes to a release?
Posted by Igor Vaynberg <ig...@gmail.com>.
yeah
-igor
On Tue, Apr 20, 2010 at 1:49 PM, Jeremy Thomerson
<je...@wickettraining.com> wrote:
> I forget what our exact rules are for this. I know that we won't break API
> in a point release (i.e. 1.4.8). But, what about the following scenarios?
> I can add these to a point release, right?
>
> 1. Add a class. In this example, I could add a class that is a custom
> subclass of WebRequestCycle that someone can subclass to handle runtime
> exceptions in Ajax differently than they do on regular requests. (i.e. they
> would rather not redirect on an Ajax error, but instead would like to send
> some JS back in a normal Wicket AJAX response)
> 2. Add an interface and some associated functionality. In this example,
> I might have an interface to add, and some additional functionality that
> only operated on instances of that interface. No functionality would change
> for existing interfaces.
>
> Those would both be okay, right?
>
> --
> Jeremy Thomerson
> http://www.wickettraining.com
>