Luigi Castelli

let's say I am coding an object similar to groove~ but

The object can be instantiated with 1, 2 or 4 output

channels. However the number of output channels is a

parameter that can be chosen only at creation time.

The user cannot change it once the object has been

instantiated, because it directly affects the number

I want the user to be able to choose the 'number of

channels' parameter using attributes at creation time

(i.e. @numchan 4). After that, I DON'T want the user

to be able to change the parameter with a simple

message (numchan 2) like he would normally do, because

Is there a flag or some technique that I can use to

------------------------------------------------------------

THIS E-MAIL MESSAGE IS FOR THE SOLE USE OF THE INTENDED RECIPIENT AND MAY CONTAIN CONFIDENTIAL AND/OR PRIVILEGED INFORMATION. ANY UNAUTHORIZED REVIEW, USE, DISCLOSURE OR DISTRIBUTION IS PROHIBITED. IF YOU ARE NOT THE INTENDED RECIPIENT, CONTACT THE SENDER BY E-MAIL AT SUPERBIGIO@YAHOO.COM AND DESTROY ALL COPIES OF THE ORIGINAL MESSAGE. WITHOUT PREJUDICE UCC1-207.

Tired of spam? Yahoo! Mail has the best spam protection around 

let's say I am coding an object similar to groove~ but
that uses attributes.
The object can be instantiated with 1, 2 or 4 output
channels. However the number of output channels is a
parameter that can be chosen only at creation time.
The user cannot change it once the object has been
instantiated, because it directly affects the number
of outlets of the object.

I want the user to be able to choose the 'number of
channels' parameter using attributes at creation time
(i.e. @numchan 4). After that, I DON'T want the user
to be able to change the parameter with a simple
message (numchan 2) like he would normally do, because
the object would simply crash.

Is there a flag or some technique that I can use to
accomplish that?

------------------------------------------------------------
THIS E-MAIL MESSAGE IS FOR THE SOLE USE OF THE INTENDED RECIPIENT AND MAY CONTAIN CONFIDENTIAL AND/OR PRIVILEGED INFORMATION.  ANY UNAUTHORIZED REVIEW, USE, DISCLOSURE OR DISTRIBUTION IS PROHIBITED.  IF YOU ARE NOT THE INTENDED RECIPIENT, CONTACT THE SENDER BY E-MAIL AT SUPERBIGIO@YAHOO.COM AND DESTROY ALL COPIES OF THE ORIGINAL MESSAGE. WITHOUT PREJUDICE UCC1-207.
------------------------------------------------------------

Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com


question-about-attributes

Give your attribute a custom setter method. In your new method, after 

processing attributes, just set some flag to true. In the custom 

setter, if the flag is true, don't do anything.

Am 21.12.2006 um 17:19 schrieb Luigi Castelli:

> Is there a flag or some technique that I can use to

Give your attribute a custom setter method. In your new method, after  
processing attributes, just set some flag to true. In the custom  
setter, if the flag is true, don't do anything.

> Is there a flag or some technique that I can use to
> accomplish that?


Thanks for the tip, Jeremy... works like a charm.

The *class_new() method defined in ext_obex.h takes an

argument of type long for size. The old setup method

used to take an argument of type short for class size,

which was implying a limit of 32k. Does that mean that

with the new obex extension the object struct size is

> Give your attribute a custom setter method. In your

> processing attributes, just set some flag to true.

> setter, if the flag is true, don't do anything.

> Am 21.12.2006 um 17:19 schrieb Luigi Castelli:

> > Is there a flag or some technique that I can use

The *class_new() method defined in ext_obex.h takes an
argument of type long for size. The old setup method
used to take an argument of type short for class size,
which was implying a limit of 32k. Does that mean that
with the new obex extension the object struct size is
no longer limited to 32k ?

> Give your attribute a custom setter method. In your
> new method, after  
> processing attributes, just set some flag to true.
> In the custom  
> setter, if the flag is true, don't do anything.
>
> jb
>
> Am 21.12.2006 um 17:19 schrieb Luigi Castelli:
>
> > Is there a flag or some technique that I can use
> to
> > accomplish that?
>
>

> Thanks for the tip, Jeremy... works like a charm.

> The *class_new() method defined in ext_obex.h takes an

> argument of type long for size. The old setup method

> used to take an argument of type short for class size,

Right, it seems this changed from short to long, also some A_GIMME type 

functions now take long for argument count with the new API. However, 

attr_args_process() still wants shorts, so I was also wondering whether 

there still is a short limit somehwere or not.

Luigi Castelli wrote:
> Thanks for the tip, Jeremy... works like a charm.
>
> One last question for you...
>
> The *class_new() method defined in ext_obex.h takes an
> argument of type long for size. The old setup method
> used to take an argument of type short for class size,
> which was implying a limit of 32k.

Right, it seems this changed from short to long, also some A_GIMME type 
functions now take long for argument count with the new API. However, 
attr_args_process() still wants shorts, so I was also wondering whether 
there still is a short limit somehwere or not.

I believe that, yes, new-style objects declared with class_new() and 

allocated with object_alloc() can be larger as 32K. attr_args_process 

() takes an atom array, and the number of atoms in the array is 

represented by a short. You can't put more than 32K atoms in a box, 

Am 24.12.2006 um 12:45 schrieb Olaf Matthes:

>> Thanks for the tip, Jeremy... works like a charm.

>> The *class_new() method defined in ext_obex.h takes an

>> argument of type long for size. The old setup method

>> used to take an argument of type short for class size,

> Right, it seems this changed from short to long, also some A_GIMME 

> type functions now take long for argument count with the new API. 

> However, attr_args_process() still wants shorts, so I was also 

> wondering whether there still is a short limit somehwere or not.

I believe that, yes, new-style objects declared with class_new() and  
allocated with object_alloc() can be larger as 32K. attr_args_process 
() takes an atom array, and the number of atoms in the array is  
represented by a short. You can't put more than 32K atoms in a box,  
anyway.

> Luigi Castelli wrote:
>> Thanks for the tip, Jeremy... works like a charm.
>> One last question for you...
>> The *class_new() method defined in ext_obex.h takes an
>> argument of type long for size. The old setup method
>> used to take an argument of type short for class size,
>> which was implying a limit of 32k.
>
> Right, it seems this changed from short to long, also some A_GIMME  
> type functions now take long for argument count with the new API.  
> However, attr_args_process() still wants shorts, so I was also  
> wondering whether there still is a short limit somehwere or not.
>
> Olaf


question about attributes