On 01/19/2013 09:39 PM, Jonathan Wilkes wrote:
> 
> 
> 
> 
> ----- Original Message -----
>> From: Hans-Christoph Steiner <[email protected]>
>> To: Jonathan Wilkes <[email protected]>
>> Cc: "[email protected]" <[email protected]>
>> Sent: Saturday, January 19, 2013 2:22 PM
>> Subject: Re: [PD] enhance pd-extended with pd-l2ork featues ?
>>
>> On 01/19/2013 02:14 PM, Jonathan Wilkes wrote:
>>>
>>>
>>>  ----- Original Message -----
>>>>  From: Hans-Christoph Steiner <[email protected]>
>>>>  To: [email protected]
>>>>  Cc: 
>>>>  Sent: Saturday, January 19, 2013 1:33 PM
>>>>  Subject: Re: [PD] enhance pd-extended with pd-l2ork featues ?
>>>>
>>>>
>>>>  I haven't heard any objections to including them. I agree, they are 
>> nice
>>>>  features.  It is just a matter of someone doing the work of porting 
>> those
>>>>  changes to Pd-extended.
>>>
>>>  So you would be ok with adding a field to the widgetbehavior struct
>>>  (or whatever its called)?  Doesn't that break binary compatability?
>>>
>>>  I think many of his gui speedups are based off that.
>>
>> Breaking binary compatibility is not something to do lightly.  It creates lot
>> of other headaches.  Plus I don't think its necessary to speed up the GUI, I
>> think there are better approaches.
> 
> How do you change text_displace to send a single
> message to the gui to just "move everything that is currently selected" rather
> than sending a separate message for every selected t_text?

I think you take it a step further and keep pd out of the moving process all
together.  pd-gui should handle moves entirely, and then just report it to pd
once its complete.

.hc

_______________________________________________
[email protected] mailing list
UNSUBSCRIBE and account-management -> 
http://lists.puredata.info/listinfo/pd-list

Reply via email to