It appears I've hit the bug yet once again. Namely, Pd-gui's numbers/sliders freeze after a while, yet ironically dynamic changes to iemgui objects' attributes (e.g. background color) continue to work. Pd audio thread also continues to work uninterrupted.
Has this been fixed in newer Pd versions? If so, could someone please share the relevant bug-fix so that we can backport it to 0.41.4.
Best wishes,
Ico
It seems to be fixed in 0.42.5. I looked at backporting, its
complicated and not worth it. Its a much better use of time to put
that effort into making 0.42.5 work better. Try using 0.42.5 nightly
build of Pd-extended and see if the bug is still there.
.hc
On Oct 29, 2009, at 9:02 AM, Ivica Ico Bukvic wrote:
It appears I've hit the bug yet once again. Namely, Pd-gui's numbers/sliders freeze after a while, yet ironically dynamic changes
to iemgui objects' attributes (e.g. background color) continue to work.
Pd audio thread also continues to work uninterrupted.Has this been fixed in newer Pd versions? If so, could someone please share the relevant bug-fix so that we can backport it to 0.41.4.
Best wishes,
Ico
If nature has made any one thing less susceptible than all others of
exclusive property, it is the action of the thinking power called an
idea, which an individual may exclusively possess as long as he keeps
it to himself; but the moment it is divulged, it forces itself into
the possession of everyone, and the receiver cannot dispossess himself
of it. - Thomas Jefferson
On Thu, 2009-10-29 at 16:06 -0400, Hans-Christoph Steiner wrote:
It seems to be fixed in 0.42.5. I looked at backporting, its
complicated and not worth it. Its a much better use of time to put
that effort into making 0.42.5 work better. Try using 0.42.5 nightly
build of Pd-extended and see if the bug is still there..hc
Cool! Thanks! I have updated to 0.42.5 and so far so good!
BTW, I've looked at the 0.43 code branch and wanted to borrow a few things to improve upon 0.42.5 (kind of a mix of my hacked pd.tk with additional improvements) until dust settles around 0.43 development. One thing I noticed that was kind of nifty in the new tcl sources is a check to see whether a particular canvas has been loaded. Namely, variable ::loaded($mycanvas) which effectively (if I am not mistaken) generates a member in the global ::loaded array and which is exactly what I needed.
I am however wondering if this is a potential memory leak (however small) that never gets deallocated? If so, should I be worried if I implement a similar feature by using 4 different vars per window? FWIW, I tried closing an opening literally dozens of windows and my memory footprint did not budge at all (according to system monitor).
Any advice on this one?
P.S. the feature I am working on deals with ability to customize background, menu- and scrollbar-toggle that is patcher-specific and is retained even after the window is closed (obviously, until the main patch is closed). Abstractions rely upon the sys_gui external. Attached is the pd.tk and (the one I posted before with additional improvements) and relevant abstractions if you want to give it a try (tested on Linux only with tcl/tk >=8.5).
P.P.S. Not sure if this is already known, but here it goes anyhow: contrary to what the window_name help file suggests, when using pd-extended with compiz, most functions that were so far OSX and/or Windows-specific (as far as window_name help file is concerned) now work on Linux as well, including alpha transparency of windows. Yay for eyecandy! :-)
Best wishes,
Ico
thing I noticed that was kind of nifty in the new tcl sources is a check to see whether a particular canvas has been loaded. Namely, variable ::loaded($mycanvas) which effectively (if I am not mistaken) generates a member in the global ::loaded array and which is exactly what I needed.
Actually, this variable is a bad example as it does get deallocated but ::yscrollsomething($mycanvas) and another variable for x axis to the best of my knowledge does not get deallocated. So, I guess the question remains hopefully relevant...
Please advise.
Best wishes,
Ico
On Oct 30, 2009, at 2:02 AM, Ivica Ico Bukvic wrote:
thing I noticed that was kind of nifty in the new tcl sources is a
check to see whether a particular canvas has been loaded. Namely, variable ::loaded($mycanvas) which effectively (if I am not mistaken) generates a member in the global ::loaded array and which is exactly what I needed.Actually, this variable is a bad example as it does get deallocated
but ::yscrollsomething($mycanvas) and another variable for x axis to the
best of my knowledge does not get deallocated. So, I guess the question
remains hopefully relevant...Please advise.
Best wishes,
Ico
Like many dynamic lnaguages, Tcl has garbage collection, so its not
really a big deal. Ideally variables would be 'unset', but I don't
think you'll get a memory leak if you don't.
.hc
"It is convenient to imagine a power beyond us because that means we
don't have to examine our own lives.", from "The Idols of
Environmentalism", by Curtis White
On Wed, 4 Nov 2009, Hans-Christoph Steiner wrote:
Like many dynamic lnaguages, Tcl has garbage collection, so its not really a big deal. Ideally variables would be 'unset', but I don't think you'll get a memory leak if you don't. http://wiki.tcl.tk/3096
An endlessly growing array is just like a memory leak except that some people refuse to call it as such.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard, Montréal, Québec. téléphone: +1.514.383.3801
On Nov 5, 2009, at 12:33 AM, Mathieu Bouchard wrote:
On Wed, 4 Nov 2009, Hans-Christoph Steiner wrote:
Like many dynamic lnaguages, Tcl has garbage collection, so its not
really a big deal. Ideally variables would be 'unset', but I don't
think you'll get a memory leak if you don't. http://wiki.tcl.tk/3096An endlessly growing array is just like a memory leak except that
some people refuse to call it as such.
I don't know anything about the Tcl garbage collectiion, are you
saying it would not work on Tcl array elements?
.hc
Computer science is no more related to the computer than astronomy is
related to the telescope. -Edsger Dykstra
On Thu, 5 Nov 2009, Hans-Christoph Steiner wrote:
On Nov 5, 2009, at 12:33 AM, Mathieu Bouchard wrote:
An endlessly growing array is just like a memory leak except that some people refuse to call it as such.
I don't know anything about the Tcl garbage collectiion, are you saying it would not work on Tcl array elements?
Tcl's memory management can't guess what you mean when you don't unset array elements that you don't need. It doesn't know that you won't ever use them again and it cannot know. This is the same in any language, thus, it doesn't depend on the language, and it's not the language's fault, nor the interpreter's fault, etc.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard, Montréal, Québec. téléphone: +1.514.383.3801
-
Hans-Christoph Steiner -
Ivica Ico Bukvic -
Mathieu Bouchard