call/answer problem in 2.6b1.136?

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

call/answer problem in 2.6b1.136?

Boris Popov, DeepCove Labs (SNN)
Hmm, looks like the change to add render notification had broken my
confirmation dialogs, reverting to previous implementation (attached)
fixes the problem. More specifically, 'call'ing the dialog in the new
version and waiting for a Boolean 'answer' never even got to rendering
phase and returned nil straight into 'expected boolean' error :)

I haven't had a chance to dig deeper and I need to push this release
out, but if someone could explain the motivation for this change, I may
figure it out sooner.

Cheers!

-Boris

--
+1.604.689.0322
DeepCove Labs Ltd.
4th floor 595 Howe Street
Vancouver, Canada V6C 2T5
http://tinyurl.com/r7uw4

[hidden email]

CONFIDENTIALITY NOTICE

This email is intended only for the persons named in the message
header. Unless otherwise indicated, it contains information that is
private and confidential. If you have received it in error, please
notify the sender and delete the entire message including any
attachments.

Thank you.


_______________________________________________
Seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside

WAComponent-methods.st (1K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

RE: call/answer problem in 2.6b1.136?

Bany, Michel
 

> Hmm, looks like the change to add render notification had
> broken my confirmation dialogs, reverting to previous
> implementation (attached) fixes the problem. More
> specifically, 'call'ing the dialog in the new version and
> waiting for a Boolean 'answer' never even got to rendering
> phase and returned nil straight into 'expected boolean' error :)

Hi Boris,
Sorry to hear that. Hope this is not causing too much trouble.
I could not reproduce. Could you please supply a testcase?

> I haven't had a chance to dig deeper and I need to push this
> release out, but if someone could explain the motivation for
> this change, I may figure it out sooner.

I do not know the motivation the comments for this change (lr.222) were:

- fixed bug in the interplay of #show:/#call:

Michel.
_______________________________________________
Seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
Reply | Threaded
Open this post in threaded view
|

RE: call/answer problem in 2.6b1.136?

Boris Popov, DeepCove Labs (SNN)
I'm afraid I won't get a chance to come up with a standalone case for at
least another day. Big release looming, you know how it goes :) I'll
keep the old methods as overrides for now as they seem to work fine
until I can look further into it.

Cheers!

-Boris

--
+1.604.689.0322
DeepCove Labs Ltd.
4th floor 595 Howe Street
Vancouver, Canada V6C 2T5
http://tinyurl.com/r7uw4

[hidden email]

CONFIDENTIALITY NOTICE

This email is intended only for the persons named in the message
header. Unless otherwise indicated, it contains information that is
private and confidential. If you have received it in error, please
notify the sender and delete the entire message including any
attachments.

Thank you.

> -----Original Message-----
> From: [hidden email] [mailto:seaside-
> [hidden email]] On Behalf Of Bany, Michel
> Sent: Tuesday, April 17, 2007 2:09 AM
> To: Seaside - general discussion
> Subject: RE: [Seaside] call/answer problem in 2.6b1.136?
>
>
>
> > Hmm, looks like the change to add render notification had
> > broken my confirmation dialogs, reverting to previous
> > implementation (attached) fixes the problem. More
> > specifically, 'call'ing the dialog in the new version and
> > waiting for a Boolean 'answer' never even got to rendering
> > phase and returned nil straight into 'expected boolean' error :)
>
> Hi Boris,
> Sorry to hear that. Hope this is not causing too much trouble.
> I could not reproduce. Could you please supply a testcase?
>
> > I haven't had a chance to dig deeper and I need to push this
> > release out, but if someone could explain the motivation for
> > this change, I may figure it out sooner.
>
> I do not know the motivation the comments for this change (lr.222)
were:
>
> - fixed bug in the interplay of #show:/#call:
>
> Michel.
> _______________________________________________
> Seaside mailing list
> [hidden email]
> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
_______________________________________________
Seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
Reply | Threaded
Open this post in threaded view
|

Re: call/answer problem in 2.6b1.136?

Lukas Renggli
> I'm afraid I won't get a chance to come up with a standalone case for at
> least another day. Big release looming, you know how it goes :) I'll
> keep the old methods as overrides for now as they seem to work fine
> until I can look further into it.

Are you using #show:/#show:onAnswer:/#show:onAnswer:delegation: directly?

In my opinion #show: should not block and therefor not raise the
WARenderNotification. When using #show: the web-flow continues in
aBlock, but the execution-flow is not interrupted. This is
continuation passing style.

#call: is different. It blocks the execution and resumes it after
answering. Here the web-flow and the execution-flow are unified.
That's why we need to interrupt the execution using the
WARenderNotification.

Does this make sense?

Cheers,
Lukas

--
Lukas Renggli
http://www.lukas-renggli.ch
_______________________________________________
Seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
Reply | Threaded
Open this post in threaded view
|

RE: call/answer problem in 2.6b1.136?

Boris Popov, DeepCove Labs (SNN)
Lukas,

I use my own,

WAComponent>>dialog: aComponent

^AnswerContinuation
  currentDo:
    [:cc |
      self
        show: aComponent
        onAnswer: cc
        delegation: ((DeepCoveLabs.YUIPanel new)
                         delegate: aComponent;
                         yourself)]

which was based on #lightbox: originally when I started using YUI,
should I be looking at doing something different to support the changes
you made?

Cheers!

-Boris

--
+1.604.689.0322
DeepCove Labs Ltd.
4th floor 595 Howe Street
Vancouver, Canada V6C 2T5
http://tinyurl.com/r7uw4

[hidden email]

CONFIDENTIALITY NOTICE

This email is intended only for the persons named in the message
header. Unless otherwise indicated, it contains information that is
private and confidential. If you have received it in error, please
notify the sender and delete the entire message including any
attachments.

Thank you.

> -----Original Message-----
> From: [hidden email] [mailto:seaside-
> [hidden email]] On Behalf Of Lukas Renggli
> Sent: Thursday, April 19, 2007 1:01 AM
> To: Seaside - general discussion
> Subject: Re: [Seaside] call/answer problem in 2.6b1.136?
>
> > I'm afraid I won't get a chance to come up with a standalone case
for at
> > least another day. Big release looming, you know how it goes :) I'll
> > keep the old methods as overrides for now as they seem to work fine
> > until I can look further into it.
>
> Are you using #show:/#show:onAnswer:/#show:onAnswer:delegation:
directly?

>
> In my opinion #show: should not block and therefor not raise the
> WARenderNotification. When using #show: the web-flow continues in
> aBlock, but the execution-flow is not interrupted. This is
> continuation passing style.
>
> #call: is different. It blocks the execution and resumes it after
> answering. Here the web-flow and the execution-flow are unified.
> That's why we need to interrupt the execution using the
> WARenderNotification.
>
> Does this make sense?
>
> Cheers,
> Lukas
>
> --
> Lukas Renggli
> http://www.lukas-renggli.ch
> _______________________________________________
> Seaside mailing list
> [hidden email]
> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
_______________________________________________
Seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
Reply | Threaded
Open this post in threaded view
|

Re: call/answer problem in 2.6b1.136?

Lukas Renggli
> WAComponent>>dialog: aComponent
>
> ^AnswerContinuation
>   currentDo:
>     [:cc |
>       self
>         show: aComponent
>         onAnswer: cc
>         delegation: ((DeepCoveLabs.YUIPanel new)
>                          delegate: aComponent;
>                          yourself)]

Then I think it is the best to fix your #dialog: message and I do the
same for the  #lightbox: (aka adding WARenderNotification raiseSignal
at the end of code within the continuation-block). I don't think this
is a bug in Seaside.

Lukas

--
Lukas Renggli
http://www.lukas-renggli.ch
_______________________________________________
Seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
Reply | Threaded
Open this post in threaded view
|

RE: call/answer problem in 2.6b1.136?

Boris Popov, DeepCove Labs (SNN)
Fair enough, I'll check in a bit,

Thanks!

-Boris

--
+1.604.689.0322
DeepCove Labs Ltd.
4th floor 595 Howe Street
Vancouver, Canada V6C 2T5
http://tinyurl.com/r7uw4

[hidden email]

CONFIDENTIALITY NOTICE

This email is intended only for the persons named in the message
header. Unless otherwise indicated, it contains information that is
private and confidential. If you have received it in error, please
notify the sender and delete the entire message including any
attachments.

Thank you.

> -----Original Message-----
> From: [hidden email] [mailto:seaside-
> [hidden email]] On Behalf Of Lukas Renggli
> Sent: Thursday, April 19, 2007 10:06 AM
> To: Seaside - general discussion
> Subject: Re: [Seaside] call/answer problem in 2.6b1.136?
>
> > WAComponent>>dialog: aComponent
> >
> > ^AnswerContinuation
> >   currentDo:
> >     [:cc |
> >       self
> >         show: aComponent
> >         onAnswer: cc
> >         delegation: ((DeepCoveLabs.YUIPanel new)
> >                          delegate: aComponent;
> >                          yourself)]
>
> Then I think it is the best to fix your #dialog: message and I do the
> same for the  #lightbox: (aka adding WARenderNotification raiseSignal
> at the end of code within the continuation-block). I don't think this
> is a bug in Seaside.
>
> Lukas
>
> --
> Lukas Renggli
> http://www.lukas-renggli.ch
> _______________________________________________
> Seaside mailing list
> [hidden email]
> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
_______________________________________________
Seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
Reply | Threaded
Open this post in threaded view
|

RE: call/answer problem in 2.6b1.136?

Boris Popov, DeepCove Labs (SNN)
Lukas,

Your suggestion did indeed do the trick, thanks.

Cheers!

-Boris

--
+1.604.689.0322
DeepCove Labs Ltd.
4th floor 595 Howe Street
Vancouver, Canada V6C 2T5
http://tinyurl.com/r7uw4

[hidden email]

CONFIDENTIALITY NOTICE

This email is intended only for the persons named in the message
header. Unless otherwise indicated, it contains information that is
private and confidential. If you have received it in error, please
notify the sender and delete the entire message including any
attachments.

Thank you.

> -----Original Message-----
> From: [hidden email] [mailto:seaside-
> [hidden email]] On Behalf Of Boris Popov
> Sent: Thursday, April 19, 2007 10:07 AM
> To: Seaside - general discussion
> Subject: RE: [Seaside] call/answer problem in 2.6b1.136?
>
> Fair enough, I'll check in a bit,
>
> Thanks!
>
> -Boris
>
> --
> +1.604.689.0322
> DeepCove Labs Ltd.
> 4th floor 595 Howe Street
> Vancouver, Canada V6C 2T5
> http://tinyurl.com/r7uw4
>
> [hidden email]
>
> CONFIDENTIALITY NOTICE
>
> This email is intended only for the persons named in the message
> header. Unless otherwise indicated, it contains information that is
> private and confidential. If you have received it in error, please
> notify the sender and delete the entire message including any
> attachments.
>
> Thank you.
>
> > -----Original Message-----
> > From: [hidden email] [mailto:seaside-
> > [hidden email]] On Behalf Of Lukas Renggli
> > Sent: Thursday, April 19, 2007 10:06 AM
> > To: Seaside - general discussion
> > Subject: Re: [Seaside] call/answer problem in 2.6b1.136?
> >
> > > WAComponent>>dialog: aComponent
> > >
> > > ^AnswerContinuation
> > >   currentDo:
> > >     [:cc |
> > >       self
> > >         show: aComponent
> > >         onAnswer: cc
> > >         delegation: ((DeepCoveLabs.YUIPanel new)
> > >                          delegate: aComponent;
> > >                          yourself)]
> >
> > Then I think it is the best to fix your #dialog: message and I do
the
> > same for the  #lightbox: (aka adding WARenderNotification
raiseSignal
> > at the end of code within the continuation-block). I don't think
this

> > is a bug in Seaside.
> >
> > Lukas
> >
> > --
> > Lukas Renggli
> > http://www.lukas-renggli.ch
> > _______________________________________________
> > Seaside mailing list
> > [hidden email]
> > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
> _______________________________________________
> Seaside mailing list
> [hidden email]
> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
_______________________________________________
Seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside