swingosc-devel Mailing List for SwingOSC (Page 2)
Brought to you by:
sciss
Menu
▾
▴
swingosc-devel — List for Developers of SwingOSC
You can subscribe to this list here.
| 2007 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2008 |
Jan
|
Feb
|
Mar
(3) |
Apr
(3) |
May
|
Jun
|
Jul
(3) |
Aug
(1) |
Sep
|
Oct
|
Nov
|
Dec
(1) |
| 2009 |
Jan
(19) |
Feb
|
Mar
(9) |
Apr
(11) |
May
(6) |
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2010 |
Jan
|
Feb
(11) |
Mar
(1) |
Apr
(1) |
May
(3) |
Jun
(19) |
Jul
(6) |
Aug
(5) |
Sep
|
Oct
|
Nov
|
Dec
|
| 2011 |
Jan
|
Feb
(3) |
Mar
(1) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2012 |
Jan
|
Feb
(5) |
Mar
(1) |
Apr
|
May
(2) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: nescivi <ne...@gm...> - 2010-06-23 15:09:23
|
On Wednesday 23 June 2010 08:29:48 Jonatan Liljedahl wrote:
> James Harkins wrote:
> > On 6/23/10 9:35 AM, da...@da... wrote:
> >> 2. python needs to be told to OSC.OSCClient().connect to the
> >> (localhost-ip, NetAddr.langPort) but each time SC restarts it seems to
> >> have a different langport. How can python find out the langport
> >> dynamically so the routine can work correctly without needing to being
> >> recoded? Should I just write the number to a file, or is there an
> >> elegant way to do it?
> >
> > Oh, PS... this is odd. sclang should be on 57120 by default. If it's
> > choosing a different port each time, maybe one sclang process hasn't
> > released the port before the next one starts?
>
> This happens to me sometimes on linux, very frustrating.
>
> You can see what process is holding the port with this terminal command
> (not sure it exists on MacOSX):
>
> $ lsof -i :57120
ah, useful!
First run of sclang with SwingOSC:
pantometheus% lsof -i :57120
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
sclang 3247 nescivi 3u IPv4 42848 0t0 UDP *:57120
sh 3280 nescivi 3u IPv4 42848 0t0 UDP *:57120
java 3282 nescivi 3u IPv4 42848 0t0 UDP *:57120
After recompile, SwingOSC keeps running, sclang picks a new port:
pantometheus% lsof -i :57120
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
sh 3280 nescivi 3u IPv4 42848 0t0 UDP *:57120
java 3282 nescivi 3u IPv4 42848 0t0 UDP *:57120
pantometheus% lsof -i :57121
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
sclang 3304 nescivi 3u IPv4 44127 0t0 UDP *:57121
I'm not sure what the "sh" process is and why it has taken the UDP port...
Killing it makes no difference though (keeping SwingOSC running again, and
recompiling):
pantometheus% lsof -i :57121
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
sclang 3365 nescivi 3u IPv4 45495 0t0 UDP *:57121
pantometheus% lsof -i :57120
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
java 3282 nescivi 3u IPv4 42848 0t0 UDP *:57120
The curious thing is that SwingOSC is supposed to use TCP on port 57131
(that's the number I set it to, default is 57111).
So what could cause java to open up 57120?
hmm... this seems to happen here (SwingOSC)?
bootServerApp {
var cmd, localAddr;
// note : the -h option is used again because it significantly speeds
// up the connection.
localAddr = NetAddr.localAddr;
cmd = java + (options.javaOptions ? "") + "-jar \"" ++ program ++ "\" " ++
options.asOptionsString( addr.port ) +
("-h " ++ localAddr.ip ++ ":" ++ localAddr.port);
("booting " ++ cmd).inform;
unixCmd( cmd );
}
sincerely,
Marije
|
|
From: Sciss <co...@sc...> - 2010-06-23 14:30:35
|
hmmm, updatelistener per se is very robust and first 100% correct, i have it all over the place in my live instrument. so if there is a bug it must be somewhere outside of it, maybe a missed registration or a race condition, or....? at the moment i don't have time to look into it (also since i never witnessed this problem), but will keep it in my mind. if you have a startup sequence that reproduces the problem (without relying on extra quarks), i could see if i find the issue. best, -sciss- Am 23.06.2010 um 14:29 schrieb James Harkins: > On 6/23/10 11:43 AM, James Harkins wrote: >> Anyway... back to the main topic. I just saw from my debugging output that doWhenBooted is getting called twice during my startup sequence. Sometimes for an unknown reason, one of them does not receive the update listener notification and the cancel routine never gets stopped. The conditional would at least prevent the erroneous output, as a bandaid. But I suppose the proper fix is to figure out why UpdateListener is not always working correctly. > > To wit: > > booting java -Dapple.laf.useScreenMenuBar=true -Xdock:icon="/Users/dewdrop/Applications/SwingOSC-svn/trunk/application.icns" -Xdock:name=SwingOSC -jar "/Users/dewdrop/Applications/SwingOSC-svn/trunk/build/SwingOSC.jar" -t 57111 -L -i -h 127.0.0.1:57120 > SwingOSC v0.64. receiving TCP at address 127.0.0.1:57111 > SwingOSC : server connected. > swing updatelistener: is running: true > stopped cancel thread > alive - making server gui > cancel thread awake > ERROR: > SwingOSC server failed to start > > Whereas, a normal boot sequence in my setup looks like this: > > SwingOSC : server connected. > swing updatelistener: is running: true > stopped cancel thread > swing updatelistener: is running: true > stopped cancel thread > alive - making server gui > > Again, I'm fine with the conditional, but it's just a mask over the fact that UpdateListener isn't catching the notification all the time. > > hjh > > > -- > James Harkins /// dewdrop world > jam...@de... > http://www.dewdrop-world.net > > "Come said the Muse, > Sing me a song no poet has yet chanted, > Sing me the universal." -- Whitman > > blog: http://www.dewdrop-world.net/words > audio clips: http://www.dewdrop-world.net/audio > more audio: http://soundcloud.com/dewdrop_world/tracks > > ------------------------------------------------------------------------------ > ThinkGeek and WIRED's GeekDad team up for the Ultimate > GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the > lucky parental unit. See the prize list and enter to win: > http://p.sf.net/sfu/thinkgeek-promo_______________________________________________ > Swingosc-devel mailing list > Swi...@li... > https://lists.sourceforge.net/lists/listinfo/swingosc-devel |
|
From: James H. <jam...@gm...> - 2010-06-23 13:29:22
|
On 6/23/10 11:43 AM, James Harkins wrote: > Anyway... back to the main topic. I just saw from my debugging output > that doWhenBooted is getting called twice during my startup sequence. > Sometimes for an unknown reason, one of them does not receive the > update listener notification and the cancel routine never gets > stopped. The conditional would at least prevent the erroneous output, > as a bandaid. But I suppose the proper fix is to figure out why > UpdateListener is not always working correctly. To wit: booting java -Dapple.laf.useScreenMenuBar=true -Xdock:icon="/Users/dewdrop/Applications/SwingOSC-svn/trunk/application.icns" -Xdock:name=SwingOSC -jar "/Users/dewdrop/Applications/SwingOSC-svn/trunk/build/SwingOSC.jar" -t 57111 -L -i -h 127.0.0.1:57120 SwingOSC v0.64. receiving TCP at address 127.0.0.1:57111 SwingOSC : server connected. swing updatelistener: is running: true stopped cancel thread alive - making server gui *cancel thread awake* ERROR: SwingOSC server failed to start Whereas, a normal boot sequence in my setup looks like this: SwingOSC : server connected. swing updatelistener: is running: true stopped cancel thread swing updatelistener: is running: true stopped cancel thread alive - making server gui Again, I'm fine with the conditional, but it's just a mask over the fact that UpdateListener isn't catching the notification all the time. hjh -- James Harkins /// dewdrop world jam...@de... http://www.dewdrop-world.net "Come said the Muse, Sing me a song no poet has yet chanted, Sing me the universal." -- Whitman blog: http://www.dewdrop-world.net/words audio clips: http://www.dewdrop-world.net/audio more audio: http://soundcloud.com/dewdrop_world/tracks |
|
From: James H. <jam...@gm...> - 2010-06-23 03:43:49
|
On 6/21/10 9:37 PM, Sciss wrote: > i had used Task a lot a long time ago, and ran into problems, and people suggested on the list that i should switch to Routine as Task had several flaws. this is a while ago, and i don't find the particular posts any more... maybe things got fixed, but at least that was my last information. > I guess that's a discussion for another time and a different mailing list, but (if we're talking about the same thing) the design "flaws" in Task are really to handle the limitation that items cannot be removed from a scheduler queue. The best fix for that IMO would be to add this feature to the sc clocks, and then greatly simplify PauseStream and all its subclasses. Until the clocks allow this, it's imperative that a single PauseStream object must never be allowed to be scheduled more than once at the same time. What I don't know is, were the problems you encountered because they were getting scheduled more than once, or were the problems because of the logic to prevent that? Anyway... back to the main topic. I just saw from my debugging output that doWhenBooted is getting called twice during my startup sequence. Sometimes for an unknown reason, one of them does not receive the update listener notification and the cancel routine never gets stopped. The conditional would at least prevent the erroneous output, as a bandaid. But I suppose the proper fix is to figure out why UpdateListener is not always working correctly. hjh -- James Harkins /// dewdrop world jam...@de... http://www.dewdrop-world.net "Come said the Muse, Sing me a song no poet has yet chanted, Sing me the universal." -- Whitman blog: http://www.dewdrop-world.net/words audio clips: http://www.dewdrop-world.net/audio more audio: http://soundcloud.com/dewdrop_world/tracks |
|
From: Sciss <co...@sc...> - 2010-06-21 23:01:37
|
hi dan, looks good to me, i checked it in. (you have svn write access in fact, so feel free to adjust those things anytime). best, -sciss- Am 21.06.2010 um 23:48 schrieb Dan S: > Hi - > > Attached is a patch to add debian packaging information to the main > SwingOSC source tree. It's strongly based on Karsten Gebbert's > packaging for puredyne etc. > > In case this is unfamiliar: it makes a "debian" folder in the root of > the source tree with some files inside. They can go happily into the > distribution and should hopefully make it easy for debian and ubuntu > folks to package/install. > > (Not sure if I have commit access to the swingosc svn - anyway, this > patch is for review, comments welcome) > > Dan > -- > http://www.mcld.co.uk > <swingosc_debian_upstream.patch.txt>------------------------------------------------------------------------------ > ThinkGeek and WIRED's GeekDad team up for the Ultimate > GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the > lucky parental unit. See the prize list and enter to win: > http://p.sf.net/sfu/thinkgeek-promo_______________________________________________ > Swingosc-devel mailing list > Swi...@li... > https://lists.sourceforge.net/lists/listinfo/swingosc-devel |
|
From: Dan S <dan...@gm...> - 2010-06-21 22:48:42
|
Hi - Attached is a patch to add debian packaging information to the main SwingOSC source tree. It's strongly based on Karsten Gebbert's packaging for puredyne etc. In case this is unfamiliar: it makes a "debian" folder in the root of the source tree with some files inside. They can go happily into the distribution and should hopefully make it easy for debian and ubuntu folks to package/install. (Not sure if I have commit access to the swingosc svn - anyway, this patch is for review, comments welcome) Dan -- http://www.mcld.co.uk |
|
From: Dan S <dan...@gm...> - 2010-06-21 22:26:10
|
2010/6/21 Sciss <co...@sc...>:
> OpenJDK should be the preferred choice, i guess... i have heard there are no problems, and it is very very close to sun, so all the GUI stuff should be perfectly compatible.
Good good.
> once more, thanks for your work, dan!
And thanks for yours!
> i haven't followed the SC 3.4 stuff recently. is it out now? should we define a version of SwingOSC that is officially the accompaniment?
3.4 is not out yet. 3.4rc is out, and it seems to be going well, so
3.4 should be out soon
Dan
> Am 19.06.2010 um 17:26 schrieb Dan S:
>
>> Hi -
>>
>> I'm looking at packaging SwingOSC for recent Ubuntu (lucid). In
>> previous versions we used Sun's Java ("sun-java6-bin" package), but
>> Sun Java is officially not going to be made available through Ubuntu
>> repositories from now onwards.
>>
>> Of course it's still possible to install Sun Java on ubuntu (e.g. by
>> downloading it from the Sun website), but for someone trying to create
>> a debian package which will auto-install, this creates quite a
>> headache.
>>
>> So my question is: are any of the open Java implementations currently
>> "good enough" to build+run SwingOSC? If so, then the packaging can be
>> very straightforward. I can't remember what the problems were in the
>> past, missing features? Unexpected behaviour?
>>
>> Thanks
>> Dan
>> --
>> http://www.mcld.co.uk
>>
>> ------------------------------------------------------------------------------
>> ThinkGeek and WIRED's GeekDad team up for the Ultimate
>> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
>> lucky parental unit. See the prize list and enter to win:
>> http://p.sf.net/sfu/thinkgeek-promo
>> _______________________________________________
>> Swingosc-devel mailing list
>> Swi...@li...
>> https://lists.sourceforge.net/lists/listinfo/swingosc-devel
>
>
--
http://www.mcld.co.uk
|
|
From: Sciss <co...@sc...> - 2010-06-21 13:48:30
|
i had used Task a lot a long time ago, and ran into problems, and people suggested on the list that i should switch to Routine as Task had several flaws. this is a while ago, and i don't find the particular posts any more... maybe things got fixed, but at least that was my last information. best, -sciss- Am 21.06.2010 um 14:24 schrieb James Harkins: > On 6/21/10 8:35 PM, Sciss wrote: >> hi james, >> >> i had little time to look into this, unfortunately. if Routine:stop is not guarateed to stop the routine, i would say this is a definite case for the SC3 bugtracker. >> >> > > Go right ahead. > >> i don't like the Task class, i think it is fundamentally wrongly designed. i suggest as a workaround for now, we just don't call cancel.stop, and instead use another local boolean variable "cancelled" that the updatelistener sets to true, and when the cancel routine is executed, we need an if-clause reading the cancelled state. >> >> ok? >> >> > > Sure, if you feel better about that. I don't care how it's fixed, just that it gets fixed. > > (I find the comment about Task somewhat amusing. One could also say, cars are fundamentally wrongly designed, but sometimes you still have to take a taxi.) > > hjh > > > -- > James Harkins /// dewdrop world > jam...@de... > http://www.dewdrop-world.net > > "Come said the Muse, > Sing me a song no poet has yet chanted, > Sing me the universal." -- Whitman > > blog: http://www.dewdrop-world.net/words > audio clips: http://www.dewdrop-world.net/audio > more audio: http://soundcloud.com/dewdrop_world/tracks > > ------------------------------------------------------------------------------ > ThinkGeek and WIRED's GeekDad team up for the Ultimate > GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the > lucky parental unit. See the prize list and enter to win: > http://p.sf.net/sfu/thinkgeek-promo_______________________________________________ > Swingosc-devel mailing list > Swi...@li... > https://lists.sourceforge.net/lists/listinfo/swingosc-devel |
|
From: James H. <jam...@gm...> - 2010-06-21 13:25:01
|
On 6/21/10 8:35 PM, Sciss wrote: > hi james, > > i had little time to look into this, unfortunately. if Routine:stop is not guarateed to stop the routine, i would say this is a definite case for the SC3 bugtracker. > Go right ahead. > i don't like the Task class, i think it is fundamentally wrongly designed. i suggest as a workaround for now, we just don't call cancel.stop, and instead use another local boolean variable "cancelled" that the updatelistener sets to true, and when the cancel routine is executed, we need an if-clause reading the cancelled state. > > ok? > Sure, if you feel better about that. I don't care how it's fixed, just /that/ it gets fixed. (I find the comment about Task somewhat amusing. One could also say, cars are fundamentally wrongly designed, but sometimes you still have to take a taxi.) hjh -- James Harkins /// dewdrop world jam...@de... http://www.dewdrop-world.net "Come said the Muse, Sing me a song no poet has yet chanted, Sing me the universal." -- Whitman blog: http://www.dewdrop-world.net/words audio clips: http://www.dewdrop-world.net/audio more audio: http://soundcloud.com/dewdrop_world/tracks |
|
From: Sciss <co...@sc...> - 2010-06-21 12:36:39
|
OpenJDK should be the preferred choice, i guess... i have heard there are no problems, and it is very very close to sun, so all the GUI stuff should be perfectly compatible.
once more, thanks for your work, dan!
i haven't followed the SC 3.4 stuff recently. is it out now? should we define a version of SwingOSC that is officially the accompaniment?
best, -sciss-
Am 19.06.2010 um 17:26 schrieb Dan S:
> Hi -
>
> I'm looking at packaging SwingOSC for recent Ubuntu (lucid). In
> previous versions we used Sun's Java ("sun-java6-bin" package), but
> Sun Java is officially not going to be made available through Ubuntu
> repositories from now onwards.
>
> Of course it's still possible to install Sun Java on ubuntu (e.g. by
> downloading it from the Sun website), but for someone trying to create
> a debian package which will auto-install, this creates quite a
> headache.
>
> So my question is: are any of the open Java implementations currently
> "good enough" to build+run SwingOSC? If so, then the packaging can be
> very straightforward. I can't remember what the problems were in the
> past, missing features? Unexpected behaviour?
>
> Thanks
> Dan
> --
> http://www.mcld.co.uk
>
> ------------------------------------------------------------------------------
> ThinkGeek and WIRED's GeekDad team up for the Ultimate
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
> lucky parental unit. See the prize list and enter to win:
> http://p.sf.net/sfu/thinkgeek-promo
> _______________________________________________
> Swingosc-devel mailing list
> Swi...@li...
> https://lists.sourceforge.net/lists/listinfo/swingosc-devel
|
|
From: Sciss <co...@sc...> - 2010-06-21 12:35:57
|
hi james,
i had little time to look into this, unfortunately. if Routine:stop is not guarateed to stop the routine, i would say this is a definite case for the SC3 bugtracker.
i don't like the Task class, i think it is fundamentally wrongly designed. i suggest as a workaround for now, we just don't call cancel.stop, and instead use another local boolean variable "cancelled" that the updatelistener sets to true, and when the cancel routine is executed, we need an if-clause reading the cancelled state.
ok?
best, -sciss-
Am 20.06.2010 um 07:04 schrieb James Harkins:
> I think I see what the problem is. I put in some debugging statements
> this morning, and found that the line "cancel.stop" is not guaranteed to
> prevent the cancel action from running.
>
> SwingOSC : server connected.
> swing updatelistener: is running: true
> stopped cancel thread
>
> // later:
> cancel thread awake
>
> ^^ Whoa!! That last line shouldn't happen.
>
> My suggestion, if you don't like the conditional inside 'cancel', is to
> use a Task instead of a routine. If it's a Task, "cancel.stop" will set
> the task's stream variable to nil. The task remains in the scheduler
> queue, but when it wakes up, it looks to 'stream' to know what to do and
> finds nothing.
>
> That is, replace the current 'cancel' assignment with this:
>
> cancel = Task({
> timeout.wait;
> upd.remove;
> "SwingOSC server failed to start".error;
> serverBooting = false;
> }).play(clock);
>
> hjh
>
>
> // debug statements, for reference
>
>
> doWhenBooted { arg onComplete, timeout = 20.0;
> var cancel, upd;
>
> if( serverRunning.not, {
> upd = UpdateListener.newFor( this, {
> if( serverRunning.debug("swing updatelistener: is
> running"), {
> cancel.stop;
> "stopped cancel thread".debug;
> upd.remove;
> fork(onComplete);
> });
> }, \serverRunning);
> cancel = {
> timeout.wait;
> "cancel thread awake".debug;
> upd.remove;
> if(serverRunning.not) {
> "SwingOSC server failed to start".error;
> };
> serverBooting = false;
> }.fork( clock );
> }, { fork(onComplete) });
> }
>
>
> --
>
> James Harkins /// dewdrop world
> jam...@de...
> http://www.dewdrop-world.net
>
> "Come said the Muse,
> Sing me a song no poet has yet chanted,
> Sing me the universal." -- Whitman
>
> blog: http://www.dewdrop-world.net/words
> audio clips: http://www.dewdrop-world.net/audio
> more audio: http://soundcloud.com/dewdrop_world/tracks
>
>
> ------------------------------------------------------------------------------
> ThinkGeek and WIRED's GeekDad team up for the Ultimate
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
> lucky parental unit. See the prize list and enter to win:
> http://p.sf.net/sfu/thinkgeek-promo
> _______________________________________________
> Swingosc-devel mailing list
> Swi...@li...
> https://lists.sourceforge.net/lists/listinfo/swingosc-devel
|
|
From: James H. <jam...@gm...> - 2010-06-20 06:04:55
|
I think I see what the problem is. I put in some debugging statements
this morning, and found that the line "cancel.stop" is not guaranteed to
prevent the cancel action from running.
SwingOSC : server connected.
swing updatelistener: is running: true
stopped cancel thread
// later:
cancel thread awake
^^ Whoa!! That last line shouldn't happen.
My suggestion, if you don't like the conditional inside 'cancel', is to
use a Task instead of a routine. If it's a Task, "cancel.stop" will set
the task's stream variable to nil. The task remains in the scheduler
queue, but when it wakes up, it looks to 'stream' to know what to do and
finds nothing.
That is, replace the current 'cancel' assignment with this:
cancel = Task({
timeout.wait;
upd.remove;
"SwingOSC server failed to start".error;
serverBooting = false;
}).play(clock);
hjh
// debug statements, for reference
doWhenBooted { arg onComplete, timeout = 20.0;
var cancel, upd;
if( serverRunning.not, {
upd = UpdateListener.newFor( this, {
if( serverRunning.debug("swing updatelistener: is
running"), {
cancel.stop;
"stopped cancel thread".debug;
upd.remove;
fork(onComplete);
});
}, \serverRunning);
cancel = {
timeout.wait;
"cancel thread awake".debug;
upd.remove;
if(serverRunning.not) {
"SwingOSC server failed to start".error;
};
serverBooting = false;
}.fork( clock );
}, { fork(onComplete) });
}
--
James Harkins /// dewdrop world
jam...@de...
http://www.dewdrop-world.net
"Come said the Muse,
Sing me a song no poet has yet chanted,
Sing me the universal." -- Whitman
blog: http://www.dewdrop-world.net/words
audio clips: http://www.dewdrop-world.net/audio
more audio: http://soundcloud.com/dewdrop_world/tracks
|
|
From: Dan S <dan...@gm...> - 2010-06-19 16:26:56
|
Hi -
I'm looking at packaging SwingOSC for recent Ubuntu (lucid). In
previous versions we used Sun's Java ("sun-java6-bin" package), but
Sun Java is officially not going to be made available through Ubuntu
repositories from now onwards.
Of course it's still possible to install Sun Java on ubuntu (e.g. by
downloading it from the Sun website), but for someone trying to create
a debian package which will auto-install, this creates quite a
headache.
So my question is: are any of the open Java implementations currently
"good enough" to build+run SwingOSC? If so, then the packaging can be
very straightforward. I can't remember what the problems were in the
past, missing features? Unexpected behaviour?
Thanks
Dan
--
http://www.mcld.co.uk
|
|
From: Sciss <co...@sc...> - 2010-06-15 11:42:56
|
hi james,
the added "if" doesn't make sense. on the contrary, looking at this code, something got quite messed up with the updatelistener. the if block is superfluous, and the else block doesn't do anything. the symbol (\serverRunning) was supposed to be the third argument to UpdateListener. also, although protected by supercollider's cooperative multitasking, the cancel symbol should be defined before upd. i would try with this code:
doWhenBooted { arg onComplete, timeout = 20.0;
var cancel, upd;
if( serverRunning.not, {
cancel = {
timeout.wait;
upd.remove;
"SwingOSC server failed to start".error;
serverBooting = false;
}.fork( clock );
upd = UpdateListener.newFor( this, {
cancel.stop;
upd.remove;
onComplete.fork( clock );
}, \serverRunning );
}, { onComplete.fork( clock )});
}
if this doesn't help, i would suspect that cancel.stop is not doing what it should do. but sounds unlikely. to verify, one could add another local variable "cancelled", initialized to "false", and the updatelistener block would set it to "true", instead of stopping cancel. then in cancel check for "cancelled == false"...?
let me know if any of this helps.
best, -sciss-
Am 15.06.2010 um 06:34 schrieb James Harkins:
> Saw this again just now.
>
> Maybe all it needs in doWhenBooted is another if?
>
> doWhenBooted { arg onComplete, timeout = 20.0;
> var cancel, upd;
>
> if( serverRunning.not, {
> upd = UpdateListener.newFor( this, {
> if( serverRunning, {
> cancel.stop;
> upd.remove;
> fork(onComplete);
> }, \serverRunning );
> });
> cancel = {
> timeout.wait;
> upd.remove;
> if(serverRunning.not) {
> "SwingOSC server failed to start".error;
> };
> serverBooting = false;
> }.fork( clock );
> }, { fork(onComplete) });
> }
>
> hjh
>
> On 6/15/10 10:12 AM, James Harkins wrote:
>> Rather often, SwingOSC boots up fine (and is fully functional), then it posts "SwingOSC server failed to start."
>>
>> --
>> SwingOSC : server connected.
>> SwingOSC server failed to start
>> --
>>
>> Rather confusing to see! :)
>> hjh
>>
>>
>
>
> --
> James Harkins /// dewdrop world
> jam...@de...
> http://www.dewdrop-world.net
>
> "Come said the Muse,
> Sing me a song no poet has yet chanted,
> Sing me the universal." -- Whitman
>
> blog: http://www.dewdrop-world.net/words
> audio clips: http://www.dewdrop-world.net/audio
> more audio: http://soundcloud.com/dewdrop_world/tracks
>
> ------------------------------------------------------------------------------
> ThinkGeek and WIRED's GeekDad team up for the Ultimate
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
> lucky parental unit. See the prize list and enter to win:
> http://p.sf.net/sfu/thinkgeek-promo_______________________________________________
> Swingosc-devel mailing list
> Swi...@li...
> https://lists.sourceforge.net/lists/listinfo/swingosc-devel
|
|
From: Sciss <co...@sc...> - 2010-06-15 11:30:45
|
p.s. where do those fork(onComplete) come from? i don't have them here in my code, and i am add svn's head ... ?
Am 15.06.2010 um 06:34 schrieb James Harkins:
> Saw this again just now.
>
> Maybe all it needs in doWhenBooted is another if?
>
> doWhenBooted { arg onComplete, timeout = 20.0;
> var cancel, upd;
>
> if( serverRunning.not, {
> upd = UpdateListener.newFor( this, {
> if( serverRunning, {
> cancel.stop;
> upd.remove;
> fork(onComplete);
> }, \serverRunning );
> });
> cancel = {
> timeout.wait;
> upd.remove;
> if(serverRunning.not) {
> "SwingOSC server failed to start".error;
> };
> serverBooting = false;
> }.fork( clock );
> }, { fork(onComplete) });
> }
>
> hjh
>
> On 6/15/10 10:12 AM, James Harkins wrote:
>> Rather often, SwingOSC boots up fine (and is fully functional), then it posts "SwingOSC server failed to start."
>>
>> --
>> SwingOSC : server connected.
>> SwingOSC server failed to start
>> --
>>
>> Rather confusing to see! :)
>> hjh
>>
>>
>
>
> --
> James Harkins /// dewdrop world
> jam...@de...
> http://www.dewdrop-world.net
>
> "Come said the Muse,
> Sing me a song no poet has yet chanted,
> Sing me the universal." -- Whitman
>
> blog: http://www.dewdrop-world.net/words
> audio clips: http://www.dewdrop-world.net/audio
> more audio: http://soundcloud.com/dewdrop_world/tracks
>
> ------------------------------------------------------------------------------
> ThinkGeek and WIRED's GeekDad team up for the Ultimate
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
> lucky parental unit. See the prize list and enter to win:
> http://p.sf.net/sfu/thinkgeek-promo_______________________________________________
> Swingosc-devel mailing list
> Swi...@li...
> https://lists.sourceforge.net/lists/listinfo/swingosc-devel
|
|
From: James H. <jam...@gm...> - 2010-06-15 05:34:46
|
Saw this again just now.
Maybe all it needs in doWhenBooted is another if?
doWhenBooted { arg onComplete, timeout = 20.0;
var cancel, upd;
if( serverRunning.not, {
upd = UpdateListener.newFor( this, {
if( serverRunning, {
cancel.stop;
upd.remove;
fork(onComplete);
}, \serverRunning );
});
cancel = {
timeout.wait;
upd.remove;
*if(serverRunning.not) {*
"SwingOSC server failed to start".error;
*};*
serverBooting = false;
}.fork( clock );
}, { fork(onComplete) });
}
hjh
On 6/15/10 10:12 AM, James Harkins wrote:
> Rather often, SwingOSC boots up fine (and is fully functional), then
> it posts "SwingOSC server failed to start."
>
> --
> SwingOSC : server connected.
> SwingOSC server failed to start
> --
>
> Rather confusing to see! :)
> hjh
>
>
--
James Harkins /// dewdrop world
jam...@de...
http://www.dewdrop-world.net
"Come said the Muse,
Sing me a song no poet has yet chanted,
Sing me the universal." -- Whitman
blog: http://www.dewdrop-world.net/words
audio clips: http://www.dewdrop-world.net/audio
more audio: http://soundcloud.com/dewdrop_world/tracks
|
|
From: James H. <jam...@gm...> - 2010-06-15 02:12:36
|
I can't see what the bug is, but there's something wrong here.
doWhenBooted { arg onComplete, timeout = 20.0;
var cancel, upd;
if( serverRunning.not, {
upd = UpdateListener.newFor( this, {
if( serverRunning, {
cancel.stop;
upd.remove;
fork(onComplete);
}, \serverRunning );
});
cancel = {
timeout.wait;
upd.remove;
"SwingOSC server failed to start".error;
serverBooting = false;
}.fork( clock );
}, { fork(onComplete) });
}
Rather often, SwingOSC boots up fine (and is fully functional), then it
posts "SwingOSC server failed to start."
--
SwingOSC : server connected.
SwingOSC server failed to start
--
Rather confusing to see! :)
hjh
--
James Harkins /// dewdrop world
jam...@de...
http://www.dewdrop-world.net
"Come said the Muse,
Sing me a song no poet has yet chanted,
Sing me the universal." -- Whitman
blog: http://www.dewdrop-world.net/words
audio clips: http://www.dewdrop-world.net/audio
more audio: http://soundcloud.com/dewdrop_world/tracks
|
|
From: James H. <jam...@gm...> - 2010-05-31 12:33:16
|
I get this (sometimes) when starting a drag from a dragsink.
java.io.IOException: application/x-java-file-list
at
sun.awt.datatransfer.DataTransferer.translateTransferable(DataTransferer.java:1038)
at
apple.awt.CDataTransferer.translateTransferable(CDataTransferer.java:119)
at sun.awt.datatransfer.DataTransferer$6.run(DataTransferer.java:2216)
at java.awt.event.InvocationEvent.dispatch(InvocationEvent.java:209)
at java.awt.EventQueue.dispatchEvent(EventQueue.java:461)
at
java.awt.EventDispatchThread.pumpOneEventForHierarchy(EventDispatchThread.java:269)
at
java.awt.EventDispatchThread.pumpEventsForHierarchy(EventDispatchThread.java:190)
at
java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:184)
at
java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:176)
at java.awt.EventDispatchThread.run(EventDispatchThread.java:110)
I dragged from that dragsink this afternoon, no problem. Tonight, fail.
The dragsink, btw, is in ddwChucklib - run "ChuckableBrowser.newWindow"
then try to drag anything from "drag selection."
The exception doesn't seem to affect functionality - I can complete the
drag and dropping the object does what it should.
Thanks.
hjh
--
James Harkins /// dewdrop world
jam...@de...
http://www.dewdrop-world.net
"Come said the Muse,
Sing me a song no poet has yet chanted,
Sing me the universal." -- Whitman
|
|
From: Dan S <dan...@gm...> - 2010-05-31 09:58:44
|
Yes, sounds good to me Dan 2010/5/31 James Harkins <jam...@gm...>: > Some time ago, Dan S. changed Server's waitForBoot so that the > onComplete function is always evaluated in the context of a routine, > even if the server is already booted -- allowing s.sync to be used > seamlessly inside the func. > > Should we have the same for SwingOSC:waitForBoot? > > hjh > > > -- > > James Harkins /// dewdrop world > jam...@de... > http://www.dewdrop-world.net > > "Come said the Muse, > Sing me a song no poet has yet chanted, > Sing me the universal." -- Whitman > > > ------------------------------------------------------------------------------ > > _______________________________________________ > Swingosc-devel mailing list > Swi...@li... > https://lists.sourceforge.net/lists/listinfo/swingosc-devel > -- http://www.mcld.co.uk |
|
From: James H. <jam...@gm...> - 2010-05-31 05:49:21
|
Some time ago, Dan S. changed Server's waitForBoot so that the onComplete function is always evaluated in the context of a routine, even if the server is already booted -- allowing s.sync to be used seamlessly inside the func. Should we have the same for SwingOSC:waitForBoot? hjh -- James Harkins /// dewdrop world jam...@de... http://www.dewdrop-world.net "Come said the Muse, Sing me a song no poet has yet chanted, Sing me the universal." -- Whitman |
|
From: nescivi <ne...@gm...> - 2010-04-25 05:46:35
|
Hiho, I'm getting an error with ProxyMixer, latest revision; on SwingOSC... I'm guessing the action gets called before the complete GUI is built. The SliderBase calls step_ and performs the action... not sure whether that is the right thing to do... Isn't step a setting rather than a control? sincerely, Marije s.boot; p = ProxySpace.push(s); ProxyMixer.new; ERROR: Message 'visible_' not understood. RECEIVER: nil ARGS: false CALL STACK: DoesNotUnderstandError:reportError B5CEE940 arg this = <instance of DoesNotUnderstandError> Nil:handleError B5CEE6A0 arg this = nil arg error = <instance of DoesNotUnderstandError> Thread:handleError B5CEE580 arg this = <instance of Thread> arg error = <instance of DoesNotUnderstandError> Object:throw B5CEE220 arg this = <instance of DoesNotUnderstandError> Object:doesNotUnderstand B5CED500 arg this = nil arg selector = 'visible_' arg args = [*1] ArrayedCollection:do B5CDF260 arg this = [*2] arg function = <instance of Function> var i = 0 ProxyMixer:checkUpdate 0876DF50 arg this = <instance of ProxyMixer> var newState = <instance of Event> var arNames = nil var prevArNames = nil var fullSize = nil var krNames = nil var prevKrNames = nil var editName = nil var prevEditName = nil EZScroller:doAction B5CDF200 arg this = <instance of EZScroller> EZScroller:valueAction_ B5CDF140 arg this = <instance of EZScroller> arg val = 0 JSCView:doAction B5CDEFE0 arg this = <instance of JSCSlider> JSCView:setPropertyWithAction B5CDEF80 arg this = <instance of JSCSlider> arg symbol = 'step' arg obj = 1 JSCSliderBase:step_ B5CDEF20 arg this = <instance of JSCSlider> arg stepSize = 1 EZScroller:adjust 08769430 arg this = <instance of EZScroller> var slBounds = <instance of Rect> var maxLength = 290 var numTooMany = 1 var fractionToShow = 1 EZScroller:init 087685D0 arg this = <instance of EZScroller> arg w = <instance of JSCCompositeView> arg bounds = <instance of Rect> arg initVal = 0 ProxyMixer:makeArZone B532F260 arg this = <instance of ProxyMixer> var arLayout = <instance of FlowLayout> var dim = <instance of Point> ProxyMixer:makeViews 08738270 arg this = <instance of ProxyMixer> ... |
|
From: nescivi <ne...@gm...> - 2010-03-06 16:39:29
|
On Saturday 06 March 2010 10:09:19 Dan Stowell wrote:
> a.plotMode = \levels; a.refresh;
>
> ERROR: Message 'smoothing_' not understood.
> RECEIVER:
> class SCPen (08207D00) {
>
> setSmoothing/smoothing_, is that a known difference between swing and
> cocoa, or an accident?
setSmooting gives a deprecation error in swing.
I don't know why that is.
sincerely,
Marije
>
> Dan
>
> 2010/3/6 nescivi <ne...@gm...>:
> > Issues are fixed in this diff..
> >
> > please test on OSX.
> >
> > sincerely,
> > Marije
> >
> > On Saturday 06 March 2010 01:16:26 nescivi wrote:
> >> On Friday 05 March 2010 20:36:07 nescivi wrote:
> >> > On Friday 05 March 2010 19:20:23 nescivi wrote:
> >> > > Hiho,
> >> > >
> >> > > this is caused by the initClass of Plot.
> >> > >
> >> > > GUI calls can only be made in StartUp, not in initClass directly, as
> >> > > on systems other than OSX/scide, the GUI redirect is not set yet at
> >> > > initialisation.
> >> >
> >> > I just commited a fix (although there should also be a check in case
> >> > GUI.skin is nil, in case the GUI.scheme is nil, for people without
> >> > any graphical client).
> >> >
> >> > I am however getting this error when trying to execute the first
> >> > example in the Plotter helpfile.
> >> >
> >> > Has anyone else been testing the Plotter addition on SwingOSC?
> >>
> >> The error is probably related to this Note in the Drawing Support
> >> section in the String helpfile:
> >>
> >> Note: for cross-platform GUIs, use Pen.stringAtPoint, Pen.stringInRect,
> >> Pen.stringCenteredIn, Pen.stringLeftJustIn, Pen.stringRightJustIn
> >> instead.
> >>
> >> sincerely,
> >> Marije
> >>
> >> > Sincerely,
> >> > Marije
> >> >
> >> > ERROR: Method 'setSmoothing' of class Meta_JPen is deprecated and will
> >> > be removed. Use 'Meta_JPen:smoothing_' instead.
> >> > ERROR: A primitive was not bound. 0 547
> >> > Instance of Method { (03431390, gc=01, fmt=00, flg=11, set=04)
> >> > instance variables [15]
> >> > raw1 : Float 0.000000 00000400 04030004
> >> > raw2 : Float 0.000000 00000000 00000000
> >> > code : instance of Int8Array (0x3431570, size=4, set=0)
> >> > selectors : nil
> >> > constants : nil
> >> > prototypeFrame : instance of Array (0x34314e0, size=4, set=2)
> >> > context : nil
> >> > argNames : instance of SymbolArray (0x3431230, size=4, set=1)
> >> > varNames : nil
> >> > sourceCode : nil
> >> > ownerClass : class String (0x7ffce0fa3370)
> >> > name : Symbol 'drawAtPoint'
> >> > primitiveName : Symbol '_String_DrawAtPoint'
> >> > filenameSymbol : Symbol
> >> > '/usr/local/share/SuperCollider/SCClassLibrary/Common/GUI/PlusGUI/Coll
> >> >ect io ns/StringPlusGUI.sc' charPos : Integer 683
> >> > }
> >> > ERROR: Primitive 'none' failed.
> >> > Failed.
> >> > RECEIVER:
> >> > nil
> >> > CALL STACK:
> >> > MethodError:reportError 036A8FE8
> >> > arg this = <instance of PrimitiveFailedError>
> >> > Nil:handleError 036A5078
> >> > arg this = nil
> >> > arg error = <instance of PrimitiveFailedError>
> >> > Thread:handleError 03CEE958
> >> > arg this = <instance of Thread>
> >> > arg error = <instance of PrimitiveFailedError>
> >> > Object:throw 036A20C8
> >> > arg this = <instance of PrimitiveFailedError>
> >> > Object:primitiveFailed 03CEF628
> >> > arg this = nil
> >> > ArrayedCollection:do 0111DDD8
> >> > arg this = [*3]
> >> > arg function = <instance of Function>
> >> > var i = 0
> >> > Plot:drawOnGridX 038A8FA8
> >> > arg this = <instance of Plot>
> >> > arg func = <instance of Function>
> >> > var width = 338.25
> >> > var left = 25.875
> >> > var n = 7
> >> > var gridValues = [*3]
> >> > var xspec = <instance of ControlSpec>
> >> > Plot:drawNumbersX 014EAC98
> >> > arg this = <instance of Plot>
> >> > var top = 24.875
> >> > var base = 275.125
> >> > Plot:draw 036F37C8
> >> > arg this = <instance of Plot>
> >> > ArrayedCollection:do 036F1F98
> >> > arg this = [*1]
> >> > arg function = <instance of Function>
> >> > var i = 0
> >> > Meta_JPen:use 036F3268
> >> > arg this = class JPen
> >> > arg function = <instance of Function>
> >> > var res = nil
> >> > Plotter:draw 036F3468
> >> > arg this = <instance of Plotter>
> >> > Meta_JPen:protRefresh 037B9798
> >> > arg this = class JPen
> >> > arg func = <instance of Function>
> >> > arg view = <instance of JSCWindow>
> >> > arg server = <instance of SwingOSC>
> >> > arg penID = 1004
> >> > arg cmpID = 1000
> >> > var bndl = nil
> >> > var off = nil
> >> > var stop = nil
> >> > var len = nil
> >> > var numCmd = nil
> >> > var nextLen = nil
> >> > var maxBndlSize = nil
> >> > var floatSize = nil
> >> > JSCWindow:refresh 036F3518
> >> > arg this = <instance of JSCWindow>
> >> > Plotter:refresh 036F28A8
> >> > arg this = <instance of Plotter>
> >> > ArrayedCollection:plot2 038232B8
> >> > arg this = [*5]
> >> > arg name = nil
> >> > arg bounds = nil
> >> > arg discrete = false
> >> > arg numChannels = nil
> >> > arg minval = nil
> >> > arg maxval = nil
> >> > var array = [*5]
> >> > var plotter = <instance of Plotter>
> >> > ...
> >> >
> >> >
> >> > _______________________________________________
> >> > sc-dev mailing list
> >> >
> >> > info (subscription, etc.):
> >> > http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml archive:
> >> > http://www.listarc.bham.ac.uk/marchives/sc-dev/
> >> > search: http://www.listarc.bham.ac.uk/lists/sc-dev/search/
> >>
> >> _______________________________________________
> >> sc-dev mailing list
> >>
> >> info (subscription, etc.):
> >> http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml archive:
> >> http://www.listarc.bham.ac.uk/marchives/sc-dev/
> >> search: http://www.listarc.bham.ac.uk/lists/sc-dev/search/
>
|
|
From: Sciss <co...@sc...> - 2010-02-16 01:13:46
|
i added something for you, try it out:
(
var w;
w = JSCWindow( "JSCTextField" );
b = JSCTextField( w, Rect( 10, 10, 150, 16 ));
b.string = "hi there";
b.action = { arg field; field.value.postln };
c = JSCNumberBox( w, Rect( 10, 40, 150, 16 )).value_( 0.0 );
w.front;
)
b.setLineBorder;
b.setLineBorder( Color.blue );
b.setLineBorder( Color.red, 2 );
b.borderless = true;
b.borderless = false;
c.setLineBorder( Color.yellow, 3 );
c.setNormalBorder
ciao, -sciss-
Am 15.02.2010 um 23:38 schrieb James Harkins:
> On Mon, Feb 15, 2010 at 12:21 PM, Sciss <co...@sc...> wrote:
> hmmm.
>
> one could make the border optional:
>
> That does help. I have temporarily changed prInitView for JSCTextField and JSCNumberBox in my environment like so (here from JSCTextField):
>
> ^this.prSCViewNew([
> [ '/local', this.id, '[', '/new', "de.sciss.swingosc.TextField", ']',
> "ac" ++ this.id,
> '[', '/new', "de.sciss.swingosc.ActionResponder", this.id, \text, ']',
> "tx" ++ this.id,
> '[', '/new', "de.sciss.swingosc.DocumentResponder", this.id, ']' ]
> ], [
> [ '/set', this.id, \border ] ++ nil.asSwingArg
> ]);
>
> It's not a perfect solution - the class browser doesn't draw a border around the class name textfield anymore, but I can change the background on that view and it will be OK. It's definitely good enough for my purposes.
>
> Thanks for the help!
> hjh
>
>
> --
> James Harkins /// dewdrop world
> jam...@de...
> http://www.dewdrop-world.net
>
> "Come said the Muse,
> Sing me a song no poet has yet chanted,
> Sing me the universal." -- Whitman
> ------------------------------------------------------------------------------
> SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
> Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
> http://p.sf.net/sfu/solaris-dev2dev_______________________________________________
> Swingosc-devel mailing list
> Swi...@li...
> https://lists.sourceforge.net/lists/listinfo/swingosc-devel
|
|
From: James H. <jam...@gm...> - 2010-02-15 23:38:29
|
On Mon, Feb 15, 2010 at 12:21 PM, Sciss <co...@sc...> wrote:
> hmmm.
>
> one could make the border optional:
>
That does help. I have temporarily changed prInitView for JSCTextField and
JSCNumberBox in my environment like so (here from JSCTextField):
^this.prSCViewNew([
[ '/local', this.id, '[', '/new', "de.sciss.swingosc.TextField",
']',
"ac" ++ this.id,
'[', '/new', "de.sciss.swingosc.ActionResponder", this.id,
\text, ']',
"tx" ++ this.id,
'[', '/new', "de.sciss.swingosc.DocumentResponder", this.id,
']' ]
], [
[ '/set', this.id, \border ] ++ nil.asSwingArg
]);
It's not a perfect solution - the class browser doesn't draw a border around
the class name textfield anymore, but I can change the background on that
view and it will be OK. It's definitely good enough for my purposes.
Thanks for the help!
hjh
--
James Harkins /// dewdrop world
jam...@de...
http://www.dewdrop-world.net
"Come said the Muse,
Sing me a song no poet has yet chanted,
Sing me the universal." -- Whitman
|
|
From: Sciss <co...@sc...> - 2010-02-15 17:21:43
|
hmmm.
one could make the border optional:
g.options.javaOptions="-Dswing.defaultlaf=com.sun.java.swing.plaf.nimbus.NimbusLookAndFeel";
g.reboot;
(
var w;
w = JSCWindow( "JSCTextField" );
b = JSCTextField( w, Rect( 10, 10, 150, 16 ));
b.string = "hi there";
b.action = { arg field; field.value.postln };
c = JSCTextField( w, Rect( 10, 50, 150, 16 ));
c.string = "hi there";
c.action = { arg field; field.value.postln };
// g.sendMsg( '/method', c.id, \putClientProperty, "JComponent.sizeVariant", "mini" );
g.listSendMsg([ '/set', c.id, \border ] ++ nil.asSwingArg );
w.front;
)
Am 15.02.2010 um 17:08 schrieb James Harkins:
> On Mon, Feb 15, 2010 at 12:05 PM, James Harkins <jam...@gm...> wrote:
>
> Small static text views, not so good - even the Volume control in the server window is unreadable. I can live with it for now. Or maybe I just need more fonts.
>
> Oops, no, static text is fine. TextFields/NumberBoxes have a large inner frame in nimbus, which obscures the content. So I may need to keep looking for another LAF...
>
> hjh
>
>
> --
> James Harkins /// dewdrop world
> jam...@de...
> http://www.dewdrop-world.net
>
> "Come said the Muse,
> Sing me a song no poet has yet chanted,
> Sing me the universal." -- Whitman
> ------------------------------------------------------------------------------
> SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
> Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
> http://p.sf.net/sfu/solaris-dev2dev_______________________________________________
> Swingosc-devel mailing list
> Swi...@li...
> https://lists.sourceforge.net/lists/listinfo/swingosc-devel
|
