[#see] doesn't work everytime when you put it in a subpatch (or abstraction) with GOP and, then, click on it. I expected the same behaviour when you put it in a subpatch or in the main patch. GF 9.10, pd 0.42.5, Ubuntu 10.04 ++
Jack
I forgot : it happens when you open your main patch. You have to open the subpatch (or abstraction) where you have put [#see] inside and, then, close it to have a normal behavior (with mouse). ++
Jack
Le lundi 06 septembre 2010 à 12:05 +0200, Jack a écrit :
[#see] doesn't work everytime when you put it in a subpatch (or abstraction) with GOP and, then, click on it. I expected the same behaviour when you put it in a subpatch or in the main patch. GF 9.10, pd 0.42.5, Ubuntu 10.04 ++
Jack
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
On Mon, 6 Sep 2010, Jack wrote:
Le lundi 06 septembre 2010 à 12:05 +0200, Jack a écrit :
[#see] doesn't work everytime when you put it in a subpatch (or abstraction) with GOP and, then, click on it. I expected the same behaviour when you put it in a subpatch or in the main patch. GF 9.10, pd 0.42.5, Ubuntu 10.04
I forgot : it happens when you open your main patch. You have to open the subpatch (or abstraction) where you have put [#see] inside and, then, close it to have a normal behavior (with mouse).
But did you try with GF 9.11 ? There are new fixes about both [#see] and [display] (they share some code together), and it needs some testing before the release).
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
Le lundi 06 septembre 2010 à 15:03 -0400, Mathieu Bouchard a écrit :
On Mon, 6 Sep 2010, Jack wrote:
Le lundi 06 septembre 2010 à 12:05 +0200, Jack a écrit :
[#see] doesn't work everytime when you put it in a subpatch (or abstraction) with GOP and, then, click on it. I expected the same behaviour when you put it in a subpatch or in the main patch. GF 9.10, pd 0.42.5, Ubuntu 10.04
I forgot : it happens when you open your main patch. You have to open the subpatch (or abstraction) where you have put [#see] inside and, then, close it to have a normal behavior (with mouse).
But did you try with GF 9.11 ? There are new fixes about both [#see] and [display] (they share some code together), and it needs some testing before the release).
I have updated GF with svn. then ./configure
This is the GridFlow 9.10 configurator [gcc3] GNU C++ Compiler 3 (or 4): --------------------------------------------> found [stl] C++ Standard Template Library: -----------------------------------------> found [gcc64] GNU C++ in 64-bit mode: ----------------------------------------------> missing (runtime error) [pentium] Pentium-compatible CPU: --------------------------------------------> found [mmx] MMX-compatible CPU (using NASM): ---------------------------------------> missing (return false) [x11] X11 Display Protocol: --------------------------------------------------> found [x11_shm] X11 acceleration by shared memory (XSHM plugin): -------------------> found [sdl] Simple Directmedia Layer (experimental support): -----------------------> missing (where is SDL/SDL.h ?) [objcpp] GNU/Apple ObjectiveC++ Compiler: ------------------------------------> missing (where is objc/Object.h ?) [quartz] Apple Quartz/Cocoa Display: -----------------------------------------> disabled (would need objcpp) [aalib] Ascii Art Library: ---------------------------------------------------> missing (where is aalib.h ?) [jpeg] JPEG Library: ---------------------------------------------------------> found [png] PNG Library <libpng12/png.h>: ------------------------------------------> found [videodev] Video4linux Driver Interface Version 1 (kernel): ------------------> found [libv4l1] Video4linux I Driver Interface Library: ----------------------------> found [libv4l2] Video4linux II Driver Interface Library: ---------------------------> found [dc1394] DC1394 for Linux <libdc1394/dc1394_control.h>: ----------------------> missing (where is libdc1394/dc1394_control.h ?) [dc1394] DC1394 for Linux <dc1394/control.h>: --------------------------------> missing (where is dc1394/control.h ?) [mpeg3] HeroineWarrior LibMPEG3 <libmpeg3/libmpeg3.h>: -----------------------> missing (where is libmpeg3/libmpeg3.h ?) [mpeg3] HeroineWarrior LibMPEG3 <libmpeg3.h>: --------------------------------> found [quicktimeapple] Apple's QuickTime: ------------------------------------------> missing (gcc compilation error) [quicktimehw] Plaum's LibQuickTime (try #1) <lqt/quicktime.h>: ---------------> missing (gcc compilation error) [quicktimehw] Plaum's LibQuickTime (try #1) <quicktime/quicktime.h>: ---------> missing (where is quicktime/quicktime.h ?) [quicktimehw] Plaum's LibQuickTime (try #2) <lqt/quicktime.h>: ---------------> missing (gcc compilation error) [quicktimehw] Plaum's LibQuickTime (try #2) <quicktime/quicktime.h>: ---------> missing (where is quicktime/quicktime.h ?) [quicktimehw] Plaum's LibQuickTime (try #3) <lqt/quicktime.h>: ---------------> found [puredata] PureData (or DesireData): -----------------------------------------> found [desiredata] DesireData: -----------------------------------------------------> missing (gcc: error: ‘gobj_subscribe’ was not declared in this scope) [opencv] Intel OpenCV: -------------------------------------------------------> missing (gcc compilation error) [fftw] FFTW (Fastest Fourier Transform in the West): -------------------------> missing (gcc compilation error) [opengl] OpenGL (for GEM support, etc): --------------------------------------> found [gem] GEM support (versions 91, 92, 93, 92/93 hybrid): -----------------------> found
generating ./config.make generating config.h See ./config.log if you want the details of the configuration tests. If you are satisfied with that configuration, you may go on, and do "make". "make install" is not needed anymore, just move your gridflow directory to "lib/pd/extra". If you get stuck, you could contact the author about it, but first make sure you read "doc/install.html".
then make i get : In file included from src/gridflow.cxx.fcs:22: src/gridflow.hxx.fcs:30:26: error: bundled/m_pd.h: Aucun fichier ou dossier de ce type In file included from src/gridflow.cxx.fcs:22:
The ./configure find Pd, but GF can't find m_pd.h ? I have missed something ? Thanx. ++
Jack
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
On Mon, 6 Sep 2010, Jack wrote:
This is the GridFlow 9.10 configurator
The current version really says 9.11. You have the wrong version of ./configure
In file included from src/gridflow.cxx.fcs:22: src/gridflow.hxx.fcs:30:26: error: bundled/m_pd.h: Aucun fichier ou dossier de ce type
You're supposed to have bundled/m_pd.h now, because starting with GF 9.11, GF doesn't care anymore which Pd you have or haven't on your system, it builds anyway, and builds the same regardless of vanilla-vs-extended issues.
I have missed something ?
I bet that you only updated the «src» folder.
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
Yes, thanx, i was in src folder. ++
Jack
Le lundi 06 septembre 2010 à 17:58 -0400, Mathieu Bouchard a écrit :
On Mon, 6 Sep 2010, Jack wrote:
This is the GridFlow 9.10 configurator
The current version really says 9.11. You have the wrong version of ./configure
In file included from src/gridflow.cxx.fcs:22: src/gridflow.hxx.fcs:30:26: error: bundled/m_pd.h: Aucun fichier ou dossier de ce type
You're supposed to have bundled/m_pd.h now, because starting with GF 9.11, GF doesn't care anymore which Pd you have or haven't on your system, it builds anyway, and builds the same regardless of vanilla-vs-extended issues.
I have missed something ?
I bet that you only updated the «src» folder.
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
Hmm, very strange, when i load a picture with [#in] to [#see], all the links around [#see] disappear ! It was not the case with GF 9.10. Do you know why ? Thanx. ++
Jack
Le mardi 07 septembre 2010 à 00:16 +0200, Jack a écrit :
Yes, thanx, i was in src folder. ++
Jack
Le lundi 06 septembre 2010 à 17:58 -0400, Mathieu Bouchard a écrit :
On Mon, 6 Sep 2010, Jack wrote:
This is the GridFlow 9.10 configurator
The current version really says 9.11. You have the wrong version of ./configure
In file included from src/gridflow.cxx.fcs:22: src/gridflow.hxx.fcs:30:26: error: bundled/m_pd.h: Aucun fichier ou dossier de ce type
You're supposed to have bundled/m_pd.h now, because starting with GF 9.11, GF doesn't care anymore which Pd you have or haven't on your system, it builds anyway, and builds the same regardless of vanilla-vs-extended issues.
I have missed something ?
I bet that you only updated the «src» folder.
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Mathieu, for information, I have 10 [#see] in an abstraction (with GOP) and i load 10 different pictures with a [until]. At the beginning, i have the links but after the loading they disappear. Pd 0.42.5, GF 9.11, Ubuntu 10.04. Thanx. ++
Jack
PS : if i keep the instance of the abstraction open, it is working fine. Therefore, if it is closed when i load, there is no more links.
Le mardi 07 septembre 2010 à 00:40 +0200, Jack a écrit :
Hmm, very strange, when i load a picture with [#in] to [#see], all the links around [#see] disappear ! It was not the case with GF 9.10. Do you know why ? Thanx. ++
Jack
Le mardi 07 septembre 2010 à 00:16 +0200, Jack a écrit :
Yes, thanx, i was in src folder. ++
Jack
Le lundi 06 septembre 2010 à 17:58 -0400, Mathieu Bouchard a écrit :
On Mon, 6 Sep 2010, Jack wrote:
This is the GridFlow 9.10 configurator
The current version really says 9.11. You have the wrong version of ./configure
In file included from src/gridflow.cxx.fcs:22: src/gridflow.hxx.fcs:30:26: error: bundled/m_pd.h: Aucun fichier ou dossier de ce type
You're supposed to have bundled/m_pd.h now, because starting with GF 9.11, GF doesn't care anymore which Pd you have or haven't on your system, it builds anyway, and builds the same regardless of vanilla-vs-extended issues.
I have missed something ?
I bet that you only updated the «src» folder.
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Le mardi 07 septembre 2010 à 00:56 +0200, Jack a écrit :
Mathieu, for information, I have 10 [#see] in an abstraction (with GOP) and i load 10 different pictures with a [until]. At the beginning, i have the links but after the loading they disappear. Pd 0.42.5, GF 9.11, Ubuntu 10.04. Thanx. ++
Jack
PS : if i keep the instance of the abstraction open, it is working fine.
Sorry, i dicover an other strange behavior : When my instance of abstraction is open, it is working fine BUT when i close it and reopen it, there is no more links around [#see] ! ++
Jack
Therefore, if it is closed when i load, there is no more links.
Le mardi 07 septembre 2010 à 00:40 +0200, Jack a écrit :
Hmm, very strange, when i load a picture with [#in] to [#see], all the links around [#see] disappear ! It was not the case with GF 9.10. Do you know why ? Thanx. ++
Jack
Le mardi 07 septembre 2010 à 00:16 +0200, Jack a écrit :
Yes, thanx, i was in src folder. ++
Jack
Le lundi 06 septembre 2010 à 17:58 -0400, Mathieu Bouchard a écrit :
On Mon, 6 Sep 2010, Jack wrote:
This is the GridFlow 9.10 configurator
The current version really says 9.11. You have the wrong version of ./configure
In file included from src/gridflow.cxx.fcs:22: src/gridflow.hxx.fcs:30:26: error: bundled/m_pd.h: Aucun fichier ou dossier de ce type
You're supposed to have bundled/m_pd.h now, because starting with GF 9.11, GF doesn't care anymore which Pd you have or haven't on your system, it builds anyway, and builds the same regardless of vanilla-vs-extended issues.
I have missed something ?
I bet that you only updated the «src» folder.
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
On Tue, 7 Sep 2010, Jack wrote:
Sorry, i dicover an other strange behavior : When my instance of abstraction is open, it is working fine BUT when i close it and reopen it, there is no more links around [#see] !
I fixed the bug for [#see] and [display] and [gf/tk_button] (a line common to all three) but I think that this bug was an attempt to fix a (milder) bug so please report any problem you see. It seems to be fine at the moment.
This is a very new bug, so this won't be written in ChangeLog (it was introduced after last release).
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
Ok, it seems to work now (link problem), thanx. However, i still can't click on it when [#see] is in abstraction (problem with the mouse) : i have to open the instance of abstraction and close it to see things working. Any clue ? Thanx again. ++
Jack
Le mardi 07 septembre 2010 à 16:54 -0400, Mathieu Bouchard a écrit :
On Tue, 7 Sep 2010, Jack wrote:
Sorry, i dicover an other strange behavior : When my instance of abstraction is open, it is working fine BUT when i close it and reopen it, there is no more links around [#see] !
I fixed the bug for [#see] and [display] and [gf/tk_button] (a line common to all three) but I think that this bug was an attempt to fix a (milder) bug so please report any problem you see. It seems to be fine at the moment.
This is a very new bug, so this won't be written in ChangeLog (it was introduced after last release).
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
On Tue, 7 Sep 2010, Jack wrote:
Ok, it seems to work now (link problem), thanx. However, i still can't click on it when [#see] is in abstraction (problem with the mouse) : i have to open the instance of abstraction and close it to see things working. Any clue ? Thanx again. ++
I just refixed the connection deletion problem (becauser the previous fix caused occasional crashes) and now it seems to be alright (according to both gdb and valgrind).
I can't reproduce your problem. I have [#see] connected to [display], both in a GOP, and they correctly show the mouseover position (and clicks and drags) when the GOP starts closed. I tried it with both subpatches and abstractions.
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
Le mardi 07 septembre 2010 à 20:05 -0400, Mathieu Bouchard a écrit :
On Tue, 7 Sep 2010, Jack wrote:
Ok, it seems to work now (link problem), thanx. However, i still can't click on it when [#see] is in abstraction (problem with the mouse) : i have to open the instance of abstraction and close it to see things working. Any clue ? Thanx again. ++
I just refixed the connection deletion problem (becauser the previous fix caused occasional crashes) and now it seems to be alright (according to both gdb and valgrind).
I can't reproduce your problem. I have [#see] connected to [display], both in a GOP, and they correctly show the mouseover position (and clicks and drags) when the GOP starts closed. I tried it with both subpatches and abstractions.
Here the problem (patch is attached) : open it and click on the message [bang(. Then, you can't click on [#see]. ++
Jack
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
On Wed, 8 Sep 2010, Jack wrote:
Le mardi 07 septembre 2010 à 20:05 -0400, Mathieu Bouchard a écrit :
(becauser the previous fix
(my "r" key is a bit broken)
Here the problem (patch is attached) : open it and click on the message [bang(. Then, you can't click on [#see].
the problem is not what you say it is. the problem is that those boxes aren't supposed to be appearing at all, because they have an initial visible area three pixels higher than the size of the GOP subpatch. but they blink while loading the patch, and then the "bang" causes them to appear even though they are supposed to stay off.
I'm comparing with the behaviour of IEMGUI... maybe it's not a very good idea. It's hard to know what's a bug and what's not in this case.
So, in short, you have to make the subpatch GOP a bit bigger, and I'm going to take care of the bug later because in your case it shouldn't really be a problem (?)
Note that you can make slightly smaller [#see] boxes by instantiating them this way :
[#see, margins 1 1 1 1] [#see, margins 0 0 0 0]
or anything else that you can do with the 'margins' messages.
The visibility of the box is supposed to depend on its size just after the time of [loadbang], so, you can set the size to whatever you want before that time (either during [loadbang], [initbang], or GF's objectbox commas).
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
Le mardi 07 septembre 2010 à 22:20 -0400, Mathieu Bouchard a écrit :
On Wed, 8 Sep 2010, Jack wrote:
Le mardi 07 septembre 2010 à 20:05 -0400, Mathieu Bouchard a écrit :
(becauser the previous fix
(my "r" key is a bit broken)
Here the problem (patch is attached) : open it and click on the message [bang(. Then, you can't click on [#see].
the problem is not what you say it is. the problem is that those boxes aren't supposed to be appearing at all, because they have an initial visible area three pixels higher than the size of the GOP subpatch. but they blink while loading the patch, and then the "bang" causes them to appear even though they are supposed to stay off.
I'm comparing with the behaviour of IEMGUI... maybe it's not a very good idea. It's hard to know what's a bug and what's not in this case.
So, in short, you have to make the subpatch GOP a bit bigger, and I'm going to take care of the bug later because in your case it shouldn't really be a problem (?)
Note that you can make slightly smaller [#see] boxes by instantiating them this way :
[#see, margins 1 1 1 1] [#see, margins 0 0 0 0]
or anything else that you can do with the 'margins' messages.
OK, good to know.
The visibility of the box is supposed to depend on its size just after the time of [loadbang], so, you can set the size to whatever you want before that time (either during [loadbang], [initbang], or GF's objectbox commas).
It's OK now. Thanx Mathieu for these informations. ++
Jack
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
On Tue, 7 Sep 2010, Jack wrote:
Hmm, very strange, when i load a picture with [#in] to [#see], all the links around [#see] disappear ! It was not the case with GF 9.10. Do you know why ?
I saw it once, but couldn't figure out how to reproduce it, and I was busy doing something else.
Je rencontre Patrick aujourd'hui. On va sûrement essayer de corriger ça, entre autres choses. (je pense que ça va être surtout une journée de déboguage)
| Mathieu Bouchard ------------------------------ Villeray, Montréal, QC
-
Jack -
Mathieu Bouchard