Dear list,
I am trying to develop some work simultaneously in pd (at home) and max (in school). Besides trying the most to use objects common to both, I will try to design copies of the max objects that don't exist on pd, and maybe also vice-versa.
I wanted to ask the following:
- is there any "compatibility library" between pd and max? If so, maybe I
wouldn't have to try to design the abstractions myself, since they would be already made.
- there are some irregularities in pd's pat format (at least if you try to
open it with max 4). Those can be fixed with a simple text editor and the "replace text" feature. I'm trying to see which should be substitued (like osc~ for cycle~), but would like to know if somebody has also made something similar.
- since I'm not a programmer, I would like to know if the (eventually)
resulting "file fixes" would be of interest for anybody that would like to make a kind of script, so that all the changes could be automatically made (and again ask if hasn't anybody made it so far). Since it is only something like "search"->"find"->"replace", I imagine it is not complicated.
Farewell,
João Miguel Pais
Hi,
I am trying to develop some work simultaneously in pd (at home) and max (in school). Besides trying the most to use objects common to both, I will try to design copies of the max objects that don't exist on pd, and maybe also vice-versa.
since i'm also working on both platforms i have to face the same problem. Yet, i somehow gave up to strive for complete compatibility in general, just because of the small details in differences between the respective objects that make more complicated patches impossible to port. You will sure be successful in transferring small and simple pieces but not that easily with real performance stuff, especially when using Max's colls or gui objects. The xsample objects were born due to this hindsight. Now, i try to decide beforehand what platform to use which is fairly easy because it's the choice between desktop and laptop.
all the best, Thomas
hi,
quite recently, a few comments have been made on the very same subject on pd-list. No, there is no "compatibility library", nor there is any hope of having such in the near future. The available externs which emulate several max/msp objects are scattered here and there, but most of them are not fully compatible.
Firstly, pd-lib.sourceforge.net has been created as the official site for a compatibility project, but it appears that pd-lib project is now abandoned. There are around 20 externals in pd-lib, both maxish, and mspish, but mostly unfinished.
Secondly, I would like to gather compatibility externals myself (if only I had time to do so, and some assurance it is not going to cause major licensing uproar...) Last week I have started a project named `cyclone' at http://suita.chopin.edu.pl/~czaja/miXed/externs/cyclone.html
I intend to put into it any external: 1) tested for full compatibility, 2) generally behaving well, and also such, that 3) may be released under PureData license. Currently, only six msp objects (allpass~, comb~, count~, delay~, linedrive~, and rand~) have made into cyclone.
Thirdly you may find a few other ``max/msp similarity'' externals in ggee, zexy, iemlib, ext13, maxlib, etc...
Krzysztof
João Miguel Pais wrote: ...
- is there any "compatibility library" between pd and max? If so, maybe I
wouldn't have to try to design the abstractions myself, since they would be already made.
I was searching cyclone, but the winds might have calmed down, because I couldn't find any files, just the intro page. I am not a programmer, but it would be very confortable for me to be able to work with max and pd simultaneously, so I'll try to use what's already available and make abstractions with the rest. If I give up in a week or continue trying, only time will tell.
Is there anybody who compiled the codes in the pd-lib.sourcefourge site? (into windows)
I also wanted to create a kind of command-line program (I think it might be called "script") that would make the necessary changes in the text files (osc~ -> cycle~ ; correct parameters for atoms ; detection of max signal outlets and inlets ; etc.). Basically just a "replace" command would be necessary. Does anybody know if something like that already exists and can be easily controlled by an "amateur"?
Other short question: I remember seing an object from a library that would direct bangs to two exits alternatively (1st bang -> left; 2nd bang -> right ; etc.), but I searched everything and couldn't find it. Anybody remembers which object was?
João Miguel Pais
----- Original Message ----- From: "Krzysztof Czaja" czaja@chopin.edu.pl To: "João Miguel Pais" jmmmpl@netcabo.pt Cc: "PD" pd-list@iem.kug.ac.at Sent: Monday, April 22, 2002 1:59 PM Subject: Re: [PD] pd-max compatibility
hi,
quite recently, a few comments have been made on the very same subject on pd-list. No, there is no "compatibility library", nor there is any hope of having such in the near future. The available externs which emulate several max/msp objects are scattered here and there, but most of them are not fully compatible.
Firstly, pd-lib.sourceforge.net has been created as the official site for a compatibility project, but it appears that pd-lib project is now abandoned. There are around 20 externals in pd-lib, both maxish, and mspish, but mostly unfinished.
Secondly, I would like to gather compatibility externals myself (if only I had time to do so, and some assurance it is not going to cause major licensing uproar...) Last week I have started a project named `cyclone' at http://suita.chopin.edu.pl/~czaja/miXed/externs/cyclone.html
I intend to put into it any external: 1) tested for full compatibility, 2) generally behaving well, and also such, that 3) may be released under PureData license. Currently, only six msp objects (allpass~, comb~, count~, delay~, linedrive~, and rand~) have made into cyclone.
Thirdly you may find a few other ``max/msp similarity'' externals in ggee, zexy, iemlib, ext13, maxlib, etc...
Krzysztof
João Miguel Pais wrote: ...
- is there any "compatibility library" between pd and max? If so, maybe
I
wouldn't have to try to design the abstractions myself, since they
would be
already made.
From: "João Miguel Pais" jmmmpl@netcabo.pt
Other short question: I remember seing an object from a library that would direct bangs to two exits alternatively (1st bang -> left; 2nd bang ->
right
; etc.), but I searched everything and couldn't find it. Anybody remembers which object was?
If you have Gem installed then you can use [alternate]. :) Quite useful. Otherwise you can build an abstraction using two [spigot] or [gate] command. (They do the exact same thing.) Something like and [alternate~] for audio could be useful, but you can do it as an abstraction with [gate]s and some [*~]s multiplying by 0 and 1.
-Sam
João Miguel Pais hat gesagt: // João Miguel Pais wrote:
I also wanted to create a kind of command-line program (I think it might be called "script") that would make the necessary changes in the text files (osc~ -> cycle~ ; correct parameters for atoms ; detection of max signal outlets and inlets ; etc.). Basically just a "replace" command would be necessary. Does anybody know if something like that already exists and can be easily controlled by an "amateur"?
Under Unix a standard for this is the tool "sed", the "stream editor", which now also has found it's was into Mac OS-X. A modern version of sed is perl. Well, perl can do a lot more, but it can also simply replace text in a lot of textfiles.
I think, Perl is also installed on OS-X by default, but has to be installed on W32 systems by the user. OTOH sed might come with WindowsNT/2000. With Perl you would write a textfile called for example "pd2max" and with this content:
s/osc~/cycle/gx; s/old text/new text/;
and then call
$ perl -pi pd2max *.pd
which would make the programmed changes.
Ciao,
Frank Barknecht _ _______footils__
Am Dienstag, den 23. April 2002 um 09:43:54 Uhr (+0200) schrieb Frank Barknecht:
Under Unix a standard for this is the tool "sed", the "stream editor", which now also has found it's was into Mac OS-X. A modern version of sed is perl. Well, perl can do a lot more, but it can also simply replace text in a lot of textfiles.
Another tool, which might be useful is awk (or gawk), which is standard on most Unix platforms and very easy to learn (compared to perl). It scans input line by line and does transformations based on searchstrings in that line. Considering the line oriented way, Pd patches are stored that could be another option, being somewhere between sed and perl.
Yours, Orm
hi João Miguel,
sorry for that -- after spotting a typo (linedrive~' instead oflinedrive') I have run the makefile to create a new html, causing
build counter increment (I had made some changes to the source).
But I have uploaded only html, not the tarballs -- mea culpa.
This is now corrected.
I would rather maintain only linux builds for a while. If you could not compile the objects yourself, just tell me what you need, so that I know I am not wasting time rebooting into windows.
In general, I think that, anyway, you will probably very soon move all your stuff to max/msp for windows, which now has the brand new ``projected release date'' of October 2002... oops ;-)
Krzysztof
btw, I have added lookup~.
João Miguel Pais wrote:
I was searching cyclone, but the winds might have calmed down, because I couldn't find any files, just the intro page.
<I would rather maintain only linux builds for a while. If you <could not compile the objects yourself, just tell me what you need, <so that I know I am not wasting time rebooting into windows.
I don't have any special requirement, just wanted to browse the material. If useful, I'll keep it. But is there an easy way of learning how to do this? I don't like to be bothering somebody whenever I remember to try something, I would prefer to do it myself, if possible.
<In general, I think that, anyway, you will probably very soon move <all your stuff to max/msp for windows, which now has the brand new <``projected release date'' of October 2002... oops ;-)
But will it be free or accessible as pd? (and will it work, or just something "in the middle" like jmax?)
João
On Mon, 22 Apr 2002, Krzysztof Czaja wrote:
Firstly, pd-lib.sourceforge.net has been created as the official site for a compatibility project, but it appears that pd-lib project is now abandoned. There are around 20 externals in pd-lib, both maxish, and mspish, but mostly unfinished.
Secondly, I would like to gather compatibility externals myself (if only I had time to do so, and some assurance it is not going to cause major licensing uproar...) Last week I have started a project named `cyclone' at
why did you start a new project when there is already one existing ?
Guenter
hi Guenter,
does pd-lib project still exist? I have made already so much noise about cyclone and pd-lib, that I might assume pd-lib administrator is either happy with my efforts, or does not care.
Moreover, pd-lib had been designed with clearly different aims than cyclone, and we have discussed that already on pd-list.
I have been simply gathering some code I need, and while doing that, I thought it would be nice to share that code. I do not have time, nor skills to replace Karl, sorry...
Krzysztof
btw, I have not used any pd-lib code yet (nor yours:-)
? wrote: ...
why did you start a new project when there is already one existing ?
On Tue, 2002-04-23 at 07:56, Krzysztof Czaja wrote:
hi Guenter,
does pd-lib project still exist? I have made already so much noise about cyclone and pd-lib, that I might assume pd-lib administrator is either happy with my efforts, or does not care.
Pd-lib is dead unless someone wants to take it over. It seems like I have said this many times on this list, so I will give it one more try. If no one wants to become responsible for the project I am going to get it removed from sourceforge. I still like the idea of a compatibility library, I just no longer have the time for the project.
Thanks,
Karl
Moreover, pd-lib had been designed with clearly different aims than cyclone, and we have discussed that already on pd-list.
I have been simply gathering some code I need, and while doing that, I thought it would be nice to share that code. I do not have time, nor skills to replace Karl, sorry...
Krzysztof
btw, I have not used any pd-lib code yet (nor yours:-)
? wrote: ...
why did you start a new project when there is already one existing ?
Computer Music Department
Peabody Institute of the Johns Hopkins University karlmac@peabody.jhu.edu
mambo.peabody.jhu.edu/~karlmac
On Tue, 23 Apr 2002, Krzysztof Czaja wrote:
Moreover, pd-lib had been designed with clearly different aims than cyclone, and we have discussed that already on pd-list.
I have been simply gathering some code I need, and while doing that, I thought it would be nice to share that code. I do not have time, nor skills to replace Karl, sorry...
Krzysztof
btw, I have not used any pd-lib code yet (nor yours:-)
:) ... its not really about this this time, its more about behaviour of people in the pd community.
I think that it is great that people share their code, and of course it is their own decision how they do it, ......
... but doing it in a more structured way would be a big win.
This way, I have just downloaded 65 (!) packages of pd externals, and I feel like I don't have the motivation nor the time to make this ready for demudi packaging, not speaking of debian packaging ... (which might seem they same, but packaging for debian is definitively harder)
Guenter
hi Guenter,
the same lack of time is exactly why `we' are not ``doing it in a more structured way''...
Also: most of the 65 packages are in a more or less alpha stage. I do not think it is a good idea to make people think they can depend on them (I am _not_ speaking about ggee, zexy, iemlib and some others).
Krzysztof
? wrote: ...
... but doing it in a more structured way would be a big win.
This way, I have just downloaded 65 (!) packages of pd externals, and I feel like I don't have the motivation nor the time to make this ready for demudi packaging, not speaking of debian packaging ...
On Wed, 24 Apr 2002, Krzysztof Czaja wrote:
hi Guenter,
the same lack of time is exactly why `we' are not ``doing it in a more structured way''...
... yes, it might be the lack of time that we are not even talking about doing it in a more structured way. This is a mistake in my opinion, doing things in a strucured way sometimes even saves you time :) But it only works if people agree on doing it that way.
Its like, hey do you want to have 2 hours more a week for free, just say yes and you have it. (This is very optimistic, I a agree, structuring such an effort as writing pd externals can backfire).
I would have built and hosted a CVS repository, but with no positive answer from any external developer it doesn't seem to be worth the effort.
So lets live on in the middleages of free software development :)
Guenter
Also: most of the 65 packages are in a more or less alpha stage. I do not think it is a good idea to make people think they can depend on them (I am _not_ speaking about ggee, zexy, iemlib and some others).
Krzysztof
? wrote: ...
... but doing it in a more structured way would be a big win.
This way, I have just downloaded 65 (!) packages of pd externals, and I feel like I don't have the motivation nor the time to make this ready for demudi packaging, not speaking of debian packaging ...
I would have built and hosted a CVS repository, but with no positive answer from any external developer it doesn't seem to be worth the effort.
Hey, here _is_ a positive answer. I definitely back a common code base (and CVS would also be fine for the main PD source...), and i already threw in sourceforge.net some months ago. It's just that i'm not too sure whether CVS is better suited than the existing possibilities.
all the best, Thomas
On Thu, 25 Apr 2002, Thomas Grill wrote:
It's just that i'm not too sure whether CVS is better suited than the existing possibilities.
Well, the advantage is that it would make autocompilation and collaboration a lot easier.
Making snapshot releases is a lot easier too.
..besides the advantages of a version control system of having version control.
Guenter
On Thu, 25 Apr 2002, [X-UNKNOWN] günter geiger wrote:
..besides the advantages of a version control system of having version control.
Don't mean to throw yet another weird bean on the pot, but has anyone worked with AEGIS?
Supports multiple lines of development, change sets, distributed and multiple repositories...etc. Could it be an alternative? a group of people I work with are about to switch to it from CVS.
DC
david casal --0+
d.casal@uea.ac.uk --9+
www.ariada.uea.ac.uk/~dcasal --)+
On Thu, 25 Apr 2002, david casal wrote:
On Thu, 25 Apr 2002, [X-UNKNOWN] g�nter geiger wrote:
..besides the advantages of a version control system of having version control.
Don't mean to throw yet another weird bean on the pot, but has anyone worked with AEGIS?
Supports multiple lines of development, change sets, distributed and multiple repositories...etc. Could it be an alternative? a group of people I work with are about to switch to it from CVS.
Looks pretty interesting. From reading the description, what I understand is that it forces people to do the source managment in a structured way. This is one thing that programmers using CVS have to learn first.
As a big drawback I consider the fact that they say their Windows support is weak, and as pd externals are windows too..
Another point contra aegis, we have to run our own server and can't use sourceforge for aegis.
Guenter
On Fri, 26 Apr 2002, [X-UNKNOWN] günter geiger wrote:
Looks pretty interesting. From reading the description, what I understand is that it forces people to do the source managment in a structured way. This is one thing that programmers using CVS have to learn first.
That's the case AFAIK.
As a big drawback I consider the fact that they say their Windows support is weak, and as pd externals are windows too..
Indeed. Exclusion is not really the way forward. Though actually (someone correct me here if I'm wrong), development of PD's main code is never done on a windoze platform, just tested on one. It would still affect externals development, though, you're right...
Another point contra aegis, we have to run our own server and can't use sourceforge for aegis.
I would strongly volunteer to run Aegis on UEA labs, but I see your point.
Oh well.
DC
david casal --0+
d.casal@uea.ac.uk --9+
www.ariada.uea.ac.uk/~dcasal --)+
i dunno if this is relevfant or not, but yes, I keep "my copy" of the sources on a linux machine and rely on scripts to zap them over to Windows when I need to compile&test there -- ditto for irix; for linux I actually just download the source tarball and compile like anyone else would.
One thing I'm confused about is how to deal with my one directory that contains sources, compiled objects, and all my related scripts for moving things around, testing then, etc. So far I haven't taken the time to organize them in a way that would support moving the sources to a repository.
Another source of anxiety for me is that I don't know (and would have to learn) what a good policy would be for maintaining reliability when multiple people are checking code in and out. I believe the best way is to have an exhaustive test suite, but building that would take a huge chunk of time!
cheers Miller
On Fri, Apr 26, 2002 at 04:52:52PM +0100, david casal wrote:
On Fri, 26 Apr 2002, [X-UNKNOWN] günter geiger wrote:
Looks pretty interesting. From reading the description, what I understand is that it forces people to do the source managment in a structured way. This is one thing that programmers using CVS have to learn first.
That's the case AFAIK.
As a big drawback I consider the fact that they say their Windows support is weak, and as pd externals are windows too..
Indeed. Exclusion is not really the way forward. Though actually (someone correct me here if I'm wrong), development of PD's main code is never done on a windoze platform, just tested on one. It would still affect externals development, though, you're right...
Another point contra aegis, we have to run our own server and can't use sourceforge for aegis.
I would strongly volunteer to run Aegis on UEA labs, but I see your point.
Oh well.
DC
david casal --0+
d.casal@uea.ac.uk --9+
www.ariada.uea.ac.uk/~dcasal --)+
ok here is what i found maybe you can help me diagnose this wierdness finally
i built as user instead of root and did not use my .pdrc 17 works fine with older patches su root with .pdrc pd crashes with
unknown color name "(null)" pd_gui: pd process exited
uugh
Pat Pagano, Director South East Just Intonation Society http://www.screwmusicforever.com/SHREESWIFT/
Ok
it seems that the iemlibs i have been using are incompatible past
pd-0.35-test16
as soon as i saw the color "(null)" error i figured it was them
i think they were intended for 34
sorry for my constant whining cheers
Pat Pagano, Director South East Just Intonation Society http://www.screwmusicforever.com/SHREESWIFT/ ----- Original Message ----- From: shreeswifty ppagano@bellsouth.net To: msp@ucsd.edu Cc: Pd-List pd-list@iem.kug.ac.at Sent: Friday, April 26, 2002 12:55 PM Subject: [PD] 17 crashes _unknown color name "(null)"
ok here is what i found maybe you can help me diagnose this wierdness finally
i built as user instead of root and did not use my .pdrc 17 works fine with older patches su root with .pdrc pd crashes with
unknown color name "(null)" pd_gui: pd process exited
uugh
Pat Pagano, Director South East Just Intonation Society http://www.screwmusicforever.com/SHREESWIFT/
-
david casal -
Frank Barknecht -
g�nter geiger -
João Miguel Pais -
Karl MacMillan -
Krzysztof Czaja -
Miller Puckette -
Orm Finnendahl -
Sam Watson -
shreeswifty -
Thomas Grill