W3C home > Mailing lists > Public > www-forms@w3.org > July 2004

RE: use of relevance

From: David Landwehr <DLandwehr@novell.com>
Date: Fri, 23 Jul 2004 08:24:24 +0200
Message-Id: <s100bd3e.094@emea1-mh.id2.novell.com>
To: <mmystore@hotmail.com>, <mark.birbeck@x-port.net>
Cc: <JBoyer@PureEdge.com>, <www-forms@w3.org>
Hi Meghana,

I have yet another suggestion which removes the need for model item
properties and a special UI instance. A control which is not bound to a
node becomes irrelevant and could then be styled using CSS as invisible.
So the idea is to put a predicate on the single node binding for the
control in such a way that when the condition is true the control is
bound otherwise it is not bound and hence invisible.

Here is a small example where the input control is only visible when
its value is greater than 10:
<xforms:instance xmlns="">
  <data><a>11</a></data>
</xforms:instance>

<xforms:input ref="a[ . &gt; 10 ]">
  <xforms:label>A is "relevant" when its value is greater than
10</xforms:label>
</xforms:input>

Initially the input control is visible and when you change the value to
less than or equal to 10 it becomes irrelevant.

Best regards,
David


----------------------------------------------------------------
David Landwehr                              dlandwehr@novell.com 
Novell, Inc.                              Phone: +45 33 98 15 04
Amaliegade 16C                              Fax: +45 70 20 15 17
DK-1256 Copenhagen K                       http://www.novell.com 
----------------------------------------------------------------

>>> "Mark Birbeck" <mark.birbeck@x-port.net> 23-07-2004 00:38:43 >>>
Meghana,
 
The single-node binding on xf:switch does not affect the xf:case
children -
it is more like using a @ref on xf:trigger or xf:group.
 
The only way to change the currently selected xf:case is via the
toggle
action. (The spec says that the DOM can also be used, but since there
is no
definition of any interfaces with which to do this, it's not actually
possible.)
 
So, to return to your original question, I would suggest you look at
whether
you should actually be using relevant for the situation you describe.
The
whole point of relevant is to remove nodes from the serialisation that
takes
place before submission. An additional feature is that any form
controls
that are bound to those nodes can be caused not to render, but since
there
may not actually be any controls bound to the nodes, that's to a
certain
extent secondary to the serialisation/submission aspect.
 
Having said all of that, I agree with your use case, which means we
need to
flip things around, and firstly leave the instance data for submission
alone, and then ask what other ways there are to do UI switching.
 
One powerful technique is to create a completely new instance that is
used
solely to control the rendering of your form. This is a technique
called
'model-based switching' (as opposed to the UI switching of
xf:switch/xf:case), and an example of how this might be done is
available
here:
 
  <http://www.formsPlayer.com/demo/wizard/model-based-switching.htm>
 
Note that this example will also work in other XForms processors.
 
Regards,
 
Mark
 

Mark Birbeck
CEO
x-port.net Ltd.

e: Mark.Birbeck@x-port.net 
t: +44 (0) 20 7689 9232
w: http://www.formsPlayer.com/ 

Download our XForms processor from
http://www.formsPlayer.com/ 

-----Original Message-----
From: www-forms-request@w3.org [mailto:www-forms-request@w3.org] On
Behalf
Of Meghana
Sent: 22 July 2004 17:39
To: 'John Boyer'; www-forms@w3.org 
Subject: RE: use of relevance



Hi John,

 

Thanks for the reply. I checked out the switch module from Xforms
specification and the example they have given uses toggle action to
switch
among the cases. The switch tag supports single value binding but how
it can
be used to select a particular case? Can you give me some example?

 

Once again thanks for clarifying my doubts.

Meghana.


  _____  


From: John Boyer [mailto:JBoyer@PureEdge.com] 
Sent: Thursday, July 22, 2004 6:39 AM
To: Meghana; www-forms@w3.org 
Subject: RE: use of relevance

 

Hi Meghana,

 

One option is to use a switch to control what is visible.

 

Another is to create a submission instance distinct from the
processing
instance.  The processing instance would have relevant MIPs, but the
submission instance wouldn't.  Either using calculate binds or using
setvalue actions in an xforms-submit handler, copy the data to the
submission instance before submission takes place.  This will get
easier in
XForms 1.1 with a copy subtree action (which is as yet unnamed).

 

Cheers,

John Boyer

-----Original Message-----
From: Meghana [mailto:mmystore@hotmail.com] 
Sent: Wednesday, July 21, 2004 5:06 PM
To: John Boyer; www-forms@w3.org 
Subject: RE: use of relevance

I wonder how xforms handles a case when a control becomes irrelevant
based
on value of the instance data but the submission still needs to submit
that
value. Or in that case we are not supposed to use relevance and use
something else? What is that other option?

 

Meghana.

 

 


  _____  


From: John Boyer [mailto:JBoyer@PureEdge.com] 
Sent: Tuesday, July 20, 2004 11:55 AM
To: Meghana; www-forms@w3.org 
Subject: RE: use of relevance

 

Hi Meghana,

 

The first one.  Relevance applies to instance nodes.  The relevant MIP
is
then interpreted by both bound form controls and submission.

 

Cheers,

John Boyer

-----Original Message-----
From: Meghana [mailto:mmystore@hotmail.com] 
Sent: Tuesday, July 20, 2004 10:39 AM
To: www-forms@w3.org 
Subject: use of relevance

Hello,

 

When a <Bind> element has relevance attribute, is it supposed to make
the
bound element in the XForms model relevant or the bound control
relevant? In
the former case, the control will see that the model is not relevant
right
now and hide itself; also the submission will not include that value.
In the
later case, the control is hidden but model still holds that value as
valid
data. Which approach is correct?

 

Thanks,

Meghana.

 

 

 



Received on Friday, 23 July 2004 02:37:03 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 10 March 2012 06:21:58 GMT