Hi,
after creating a structure (object) with append, is it possible to delete
it? Or just better hide it with 0 on the drawing objects?
(the patch is only to display data, not to be used by the user)
João
I don't think it's possible to delete a ds object unless selecting it with the mouse (or maybe with the mouse msg).
Notice that if you turn off drawing instructions by using an inlet to [filledcurve], for example, all other visible data structures will flash off and on again. I think that's why someone suggested a while back to just change the color to 999 instead. But if you have a lot of ds instances that you want to hide, use the -v flag, because there is no flashing and it's faster than changing the color.
-Jonathan
--- On Fri, 3/20/09, João Pais jmmmpais@googlemail.com wrote:
From: João Pais jmmmpais@googlemail.com Subject: [PD] data-structures short questions To: "PD-List" pd-list@iem.at Date: Friday, March 20, 2009, 9:02 PM Hi,
after creating a structure (object) with append, is it possible to delete it? Or just better hide it with 0 on the drawing objects?
(the patch is only to display data, not to be used by the user)
João
--Friedenstr. 58 10249 Berlin (Deutschland) Tel +49 30 42020091 | Mob +49 162 6843570 jmmmpais@googlemail.com | skype: jmmmpjmmmp
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
I don't think it's possible to delete a ds object unless selecting it
with the mouse (or maybe with the mouse msg).
I thought so as well. so a better choice might be to hide it.
Notice that if you turn off drawing instructions by using an inlet to
[filledcurve], for example, all other visible data structures will flash
off and on again. I think that's why someone suggested a while back to
just change the color to 999 instead. But if you have a lot of ds
instances that you want to hide, use the -v flag, because there is no
flashing and it's faster than changing the color.
you mean flash when the 1/0 command comes in? I was just trying that, but
didn't notice anything - it was a very simple struct.
what I'm thinking uses a maximum of 64 elements, so it's not that heavy.
Here's a patch to show what I mean. Even for a two-pixel drawing in a hidden subpatch, flashing happens when the visibility is toggled. So if you use any other patches with datastructures (or the same one twice as an abstraction) it will be noticeable, regardless of the drawing complexity.
Btw: I notice in test.pd that if leave the first abstraction open and then close test.pd, I get two errors, like:
error: .x9dd9e0: no such object error: .x9cc8a0: no such object
(0.42-4, winxp sp3)
-Jonathan
--- On Fri, 3/20/09, João Pais jmmmpais@googlemail.com wrote:
From: João Pais jmmmpais@googlemail.com Subject: Re: [PD] data-structures short questions To: jancsika@yahoo.com, "PD-List" pd-list@iem.at Date: Friday, March 20, 2009, 10:52 PM
I don't think it's possible to delete a ds
object unless selecting it with the mouse (or maybe with the mouse msg).
I thought so as well. so a better choice might be to hide it.
Notice that if you turn off drawing instructions by
using an inlet to [filledcurve], for example, all other visible data structures will flash off and on again. I think that's why someone suggested a while back to just change the color to 999 instead. But if you have a lot of ds instances that you want to hide, use the -v flag, because there is no flashing and it's faster than changing the color.
you mean flash when the 1/0 command comes in? I was just trying that, but didn't notice anything - it was a very simple struct. what I'm thinking uses a maximum of 64 elements, so it's not that heavy.
ah, now I know. it never happened to me, because I never used enough to
make them visible. that's a nice example patch.
btw, in order to control the display of the individual elements I used -v
in the definitions of the base struct (that gets taken by the array). that
allows for individual control, and it should be the only example missing
from your file.
Here's a patch to show what I mean. Even for a two-pixel drawing in a
hidden subpatch, flashing happens when the visibility is toggled. So if
you use any other patches with datastructures (or the same one twice as
an abstraction) it will be noticeable, regardless of the drawing
complexity.Btw: I notice in test.pd that if leave the first abstraction open and
then close test.pd, I get two errors, like:error: .x9dd9e0: no such object error: .x9cc8a0: no such object
(0.42-4, winxp sp3)
exactly same specs here. I also notice another thing: my draw area is in a
gop, and the patch is to be used as a gop. so, when using a gop ind the
2nd degree, the setting for "hide object name and arguments" of that area
doesn't get saved. (no problem with the level above)
sorry, forgot to ask: and there's also no way of deleting an array without
deleting the whole window with "clear", is there?
ah, now I know. it never happened to me, because I never used enough to
make them visible. that's a nice example patch.btw, in order to control the display of the individual elements I used
-v in the definitions of the base struct (that gets taken by the array).
that allows for individual control, and it should be the only example
missing from your file.Here's a patch to show what I mean. Even for a two-pixel drawing in a
hidden subpatch, flashing happens when the visibility is toggled. So
if you use any other patches with datastructures (or the same one twice
as an abstraction) it will be noticeable, regardless of the drawing
complexity.Btw: I notice in test.pd that if leave the first abstraction open and
then close test.pd, I get two errors, like:error: .x9dd9e0: no such object error: .x9cc8a0: no such object
(0.42-4, winxp sp3)
exactly same specs here. I also notice another thing: my draw area is in
a gop, and the patch is to be used as a gop. so, when using a gop ind
the 2nd degree, the setting for "hide object name and arguments" of that
area doesn't get saved. (no problem with the level above)
--- On Sat, 3/21/09, João Pais jmmmpais@googlemail.com wrote:
From: João Pais jmmmpais@googlemail.com Subject: Re: [PD] data-structures short questions To: "Jonathan Wilkes" jancsika@yahoo.com, "PD-List" pd-list@iem.at Date: Saturday, March 21, 2009, 2:37 AM sorry, forgot to ask: and there's also no way of deleting an array without deleting the whole window with "clear", is there?
I believe that is the case.
I tested -v for the elements, and didn't notice any difference. -vs for [plot] has a bug, however: it doesn't properly erase the objects until you hide and reopen the window, as in the attached patch.
Additionally, arrays with graphical elements exhibit the same flashing problem I've been talking about. If you click-drag an element you'll see that all the elements after #49 start flashing. Add another drawing instruction to the template, and it's even worse: everything after the #4 flashes.
-Jonathan
ah, now I know. it never happened to me, because I
never used enough to make them visible. that's a nice example patch.
btw, in order to control the display of the individual
elements I used -v in the definitions of the base struct (that gets taken by the array). that allows for individual control, and it should be the only example missing from your file.
Here's a patch to show what I mean. Even for
a two-pixel drawing in a hidden subpatch, flashing happens when the visibility is toggled. So if you use any other patches with datastructures (or the same one twice as an abstraction) it will be noticeable, regardless of the drawing complexity.
Btw: I notice in test.pd that if leave the first
abstraction open and then close test.pd, I get two errors, like:
error: .x9dd9e0: no such object error: .x9cc8a0: no such object
(0.42-4, winxp sp3)
exactly same specs here. I also notice another thing:
my draw area is in a gop, and the patch is to be used as a gop. so, when using a gop ind the 2nd degree, the setting for "hide object name and arguments" of that area doesn't get saved. (no problem with the level above)
--Friedenstr. 58 10249 Berlin (Deutschland) Tel +49 30 42020091 | Mob +49 162 6843570 jmmmpais@googlemail.com | skype: jmmmpjmmmp