RE: [dsdp-tm-dev] RSEDefaultNewConnectionWizardMainPage seems to be more

["Stieber, Uwe" <Uwe.Stieber@xxxxxxxxxxxxx>]

Hi Jacques,

Well, the RSEDefaultNewConnectionWizardMainPage is (weak) bound to
RSEDefaultNewConnectionWizard, as the RSEDefaultNewConnectionWizard is
implementing the RSE wizard specific/typical behavior. 

RSEAbstractNewConnectionWizard is provided as common base class to allow
for 3rd party wizards following a complete different idea than the
default RSE wizards and more specifically, do not share UI with the RSE
default wizards.

Nonetheless, RSEDefaultNewConnectionWizardMainPage should be usable even
with a non RSEDefaultNewConnectionWizard implementation. For that
purpose, the RSEDefaultNewConnectionWizardMainPage forwards to the super
implementation.

> (RSEDefaultNewConnectionWizardMainPage[:154].canFlipToNextPage
> ()), the mainPage checks whether mainPage is a
RSEDefaultNewConnectionWizard or not. 

To be precise, the main page determines the associated wizard and checks
if the wizard is of type RSEDefaultNewConnectionWizard or not.

> If not (my case), is calls super.canFlipToNextPage() which in 
> turn seems to ask to the implementation to validate the form, 
> etc, etc...

Well, the super.canFlipToNextPage(), calls for "isPageComplete()", which
is overwriten in RSEDefaultNewConnectionWizardPage to forward the call
to the form, and "getNextPage()" which is using the same check as the
"canFlipToNextPage()". The form itself, which the two invoked methods
"isPageComplete()" and "isConnectionUnique()" does not go back to the
page to trigger a "canFlipToNextPage()" again. At least I cannot see how
from the code. You may educate me by forwarding a short extract from the
stack overflow error stack traces (run Eclipse with -Xint if you don't
have a stack trace visible for the error). Probably it is the
"getNextPage()" implementation of the wizard which is getting back to
the page?

Thanks, Best regards,

Uwe Stieber


 

> -----Original Message-----
> From: dsdp-tm-dev-bounces@xxxxxxxxxxx 
> [mailto:dsdp-tm-dev-bounces@xxxxxxxxxxx] On Behalf Of Jacques Arnoux
> Sent: Tuesday, March 30, 2010 6:37 AM
> To: dsdp-tm-dev@xxxxxxxxxxx
> Subject: [dsdp-tm-dev] RSEDefaultNewConnectionWizardMainPage 
> seems to be more specific than intended
> 
> Hi all,
> I'm currently trying to implement a NewConnectionWizard that 
> extends RSEAbstractNewConnectionWizard and uses 
> RSEDefaultNewConnectionWizardMainPage as a main page.
> While typing a hostname into the host name field, I 
> encountered a stack overflow I figured out as follow:
> while trying to validate the form/text typed and refresh the 
> buttons 
> (RSEDefaultNewConnectionWizardMainPage[:154].canFlipToNextPage
> ()), the mainPage checks whether mainPage is a 
> RSEDefaultNewConnectionWizard or not. 
> If not (my case), is calls super.canFlipToNextPage() which in 
> turn seems to ask to the implementation to validate the form, 
> etc, etc...
> Isn't RSEDefaultNewConnectionWizardMainPage usable without 
> RSEDefaultNewConnectionWizard?
> 
> My workaround is to extend RSEDefaultNewConnectionWizard 
> instead of RSEAbstractNewConnectionWizard but it's not so 
> great cause I cannot even use the super attributes since 
> there is no getters and setters and my code differs in only 
> performFinish(). I have to override everything.
> Is this intended and am I missing something here?
> Thanks,
> Jac
> 
>