Re: [E-devel] export of LD_LIBRARY_PATH in e_start_main.c
On Thursday 28 December 2006 21:40, Carsten Haitzler wrote: hmm - actually this will work - often, but not all the time. if prefix/bin is a symlink to for example prefix/.exec/i386/ (which i actually have seen for multi-arch shared nfs filesystems) this breaks down ok, but the way it's done now wont work either, so i dont see what you're getting at here also is this ldso feature something that exists outside of linux's ldso? as it's not documented... it is documented ... it isnt glibc-specific or linux-specific at all ... sun's ldso supports it as well since it's part of the elf library searching spec i think a more portable solution might be for e to revert its ld_lib_path changes e_start did (as its only really needed on app runtime link) or to remove it on exec of anything. they're equally portable but if you're going to pollute the env, then yes you should make sure the crap is cleansed before spawning sub processes ... of course, this wont help any modules that other people write and do exec's at the very least, there should be an option for people to remove the option for package maintainers to use since the target audience here is simply developers -mike pgpQIsuOcRcbY.pgp Description: PGP signature - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Evas GL engine patch
On Fri, 29 Dec 2006 10:55:28 +0900, Carsten Haitzler (The Rasterman) [EMAIL PROTECTED] wrote : On Sun, 10 Dec 2006 12:37:20 +0100 Simon TRENY [EMAIL PROTECTED] babbled: Hi, The GL engine of Evas has an annoying bug when you want to create several windows using this engine: when you create or resize a window, all the other windows already created are not redrawn correcly, only the last created/resized window has no problem. I've attached a shot that shows what happens when you create a new window while there is an existing window that already uses the GL engine. Please have a look at it if you want to understand the problem. So, in this example, the Etk Test Application is created first, and when the Entry button is clicked, the Etk Entry Test window is created. You can see that the Etk Test Application is no longer rendered correctly. Actually, only the part of the window that is inside the red rectangle (added with the Gimp) is correctly refreshed, the other part of the window does not refresh anymore. You can also notice that the red rectangle has exactly the same size as the Etk Entry Test window. In fact, the problem is that when you create/resize a window, _evas_gl_common_viewport_set() is called, and this function changes the GL viewport and the view matrices. The thing is, these changes affect all the windows, not only the created/resized window. A way to fix that would be to call _evas_gl_common_viewport_set() each time a window is rendered, before the rendering process begins (at the start of the function evas_render_updates_internal() for example). Only thing, I can't see a function that is called at the start of the rendering process in the engine API (something like pre_render()). I've written a small patch that calls _evas_gl_common_viewport_set() in the eng_output_redraws_rect_add() method of the gl_x11 engine (because this method is called at the start of the rendering process) just as a proof that it could fix the bug, but this is definitely not the place to put this code. The patch is attached. Please tell me what you think. that patch will do (not perfect - i think it's just masking the problem, but it shouldn't cause problems, and if ti fixes it for now - good). patch into my local tree - when i commit my latest changes it will go in. :) The patch that I submitted was just some sort of proof that it could work. As I said, it calls evas_gl_common_context_resize() in the output_redraws_rect_add() method of the gl_x11 engine, which is definitely not a good idea since this function is called several times at the start of the rendering process. A better way to avoid this would be to add a pre_render() method to the engine API, and to call this method once at the start of the rendering process. We could then call evas_gl_common_context_resize() in this method. This way, the viewport will be reset only once during the rendering process. Do you want me to write a patch that does that?? It'll be a lot cleaner imho. Simon TRENY MoOm - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Evas GL engine patch
On Fri, 29 Dec 2006 11:25:28 +0900, Carsten Haitzler (The Rasterman) [EMAIL PROTECTED] wrote : On Sat, 16 Dec 2006 10:57:48 GMT [EMAIL PROTECTED] [EMAIL PROTECTED] babbled: One thing though.. If one needs to have the 'viewport' and 'view matrices' set for each rendering pass, then I would say that this should maybe be done per obj, ie. in the internal engine render functions of each object... ?? Or does this only affect the way the gl render buffer gets 'put' on the screen? Yes, changing the GL viewport only affects the place where the whole scene is rendered in the window. You can for example create two viewports if you want to split the screen in 2 parts, like in multiplayer games on game consoles. I see... Thanks Simon. Well, even if you think your patch doesn't put those calls in the 'best' place they could be, if it solves the current gl drawing problems then I'd vote for applying it. Maybe you and rene (aka. centipede) can get together and see if you guys can do something more for the gl engine... :) i've actually been doing a bit lately. i know a few of the things it needs and as i just found recently - in order to sanely support fragment shaders for yuv-rgb i seem to need a gf6600 - my 5500fs just wont do fragment shaders (at a sane speed - software beats the 5500fx by like 200fps in software vs 0.4fps for gl in my yuv-rgb video surface test). so given this - i'm actually tempted to upgrade the iminimum requirements for gl engines. that is - REQUIRES non-power-of-2 textures. (not nv_rectangle - but npo2). requires anisotropic filtering, requires fragment shaders (glsl) etc. i can remove some nasty code to handle only power-of-2 textures for example and for nv_rectangle textures (as the text coords are different for these) Making the GL engine requires a GLSL card is really too restrictive imho, especially since (at least for now) GLSL is only used for yuv-rgb conversion, which is only used by Emotion. And this conversion can still be easily done in software if there is no GLSL support (we already have the code to do it). I have personally two GL cards and none of them support GLSL. I know that having to handle all the possible hardware configs is a lot of pain, but requiring GLSL is definitively too restrictive imho (actually even requiring anisotropic filtering is restrictive, it seems to be supported only since GeForce3 cards). jose. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Evas GL engine patch
On Fri, 29 Dec 2006 10:05:17 +0100 Simon TRENY [EMAIL PROTECTED] babbled: On Fri, 29 Dec 2006 10:55:28 +0900, Carsten Haitzler (The Rasterman) [EMAIL PROTECTED] wrote : On Sun, 10 Dec 2006 12:37:20 +0100 Simon TRENY [EMAIL PROTECTED] babbled: Hi, The GL engine of Evas has an annoying bug when you want to create several windows using this engine: when you create or resize a window, all the other windows already created are not redrawn correcly, only the last created/resized window has no problem. I've attached a shot that shows what happens when you create a new window while there is an existing window that already uses the GL engine. Please have a look at it if you want to understand the problem. So, in this example, the Etk Test Application is created first, and when the Entry button is clicked, the Etk Entry Test window is created. You can see that the Etk Test Application is no longer rendered correctly. Actually, only the part of the window that is inside the red rectangle (added with the Gimp) is correctly refreshed, the other part of the window does not refresh anymore. You can also notice that the red rectangle has exactly the same size as the Etk Entry Test window. In fact, the problem is that when you create/resize a window, _evas_gl_common_viewport_set() is called, and this function changes the GL viewport and the view matrices. The thing is, these changes affect all the windows, not only the created/resized window. A way to fix that would be to call _evas_gl_common_viewport_set() each time a window is rendered, before the rendering process begins (at the start of the function evas_render_updates_internal() for example). Only thing, I can't see a function that is called at the start of the rendering process in the engine API (something like pre_render()). I've written a small patch that calls _evas_gl_common_viewport_set() in the eng_output_redraws_rect_add() method of the gl_x11 engine (because this method is called at the start of the rendering process) just as a proof that it could fix the bug, but this is definitely not the place to put this code. The patch is attached. Please tell me what you think. that patch will do (not perfect - i think it's just masking the problem, but it shouldn't cause problems, and if ti fixes it for now - good). patch into my local tree - when i commit my latest changes it will go in. :) The patch that I submitted was just some sort of proof that it could work. As I said, it calls evas_gl_common_context_resize() in the output_redraws_rect_add() method of the gl_x11 engine, which is definitely not a good idea since this function is called several times at the start of the rendering process. A better way to avoid this would be to add a pre_render() method to the engine API, and to call this method once at the start of the rendering process. We could then call evas_gl_common_context_resize() in this method. This way, the viewport will be reset only once during the rendering process. well it's actually called only a very few times per render and so is going to be very insignificant in the scheme of things (thus why i thought it was ok) :) Do you want me to write a patch that does that?? It'll be a lot cleaner imho. Simon TRENY MoOm - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel -- - Codito, ergo sum - I code, therefore I am -- The Rasterman (Carsten Haitzler)[EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Evas GL engine patch
On Fri, 29 Dec 2006 10:21:24 +0100 Simon TRENY [EMAIL PROTECTED] babbled: On Fri, 29 Dec 2006 11:25:28 +0900, Carsten Haitzler (The Rasterman) [EMAIL PROTECTED] wrote : On Sat, 16 Dec 2006 10:57:48 GMT [EMAIL PROTECTED] [EMAIL PROTECTED] babbled: One thing though.. If one needs to have the 'viewport' and 'view matrices' set for each rendering pass, then I would say that this should maybe be done per obj, ie. in the internal engine render functions of each object... ?? Or does this only affect the way the gl render buffer gets 'put' on the screen? Yes, changing the GL viewport only affects the place where the whole scene is rendered in the window. You can for example create two viewports if you want to split the screen in 2 parts, like in multiplayer games on game consoles. I see... Thanks Simon. Well, even if you think your patch doesn't put those calls in the 'best' place they could be, if it solves the current gl drawing problems then I'd vote for applying it. Maybe you and rene (aka. centipede) can get together and see if you guys can do something more for the gl engine... :) i've actually been doing a bit lately. i know a few of the things it needs and as i just found recently - in order to sanely support fragment shaders for yuv-rgb i seem to need a gf6600 - my 5500fs just wont do fragment shaders (at a sane speed - software beats the 5500fx by like 200fps in software vs 0.4fps for gl in my yuv-rgb video surface test). so given this - i'm actually tempted to upgrade the iminimum requirements for gl engines. that is - REQUIRES non-power-of-2 textures. (not nv_rectangle - but npo2). requires anisotropic filtering, requires fragment shaders (glsl) etc. i can remove some nasty code to handle only power-of-2 textures for example and for nv_rectangle textures (as the text coords are different for these) Making the GL engine requires a GLSL card is really too restrictive imho, especially since (at least for now) GLSL is only used for yuv-rgb conversion, which is only used by Emotion. And this conversion can still be easily done in software if there is no GLSL support (we already have the code to do it). I have personally two GL cards and none of them support GLSL. I know that having to handle all the possible hardware configs is a lot of pain, but requiring GLSL is definitively too restrictive imho (actually even requiring anisotropic filtering is restrictive, it seems to be supported only since GeForce3 cards). the other side of the coin is - look at the evas benchmarks for those old cards compared to software. evas software is now pulling benchmarks that i was getting on geforce2 cards years ago with gl. (modern cpu's that is). right now the gl engine for yuv colorspace support ONLY handles it via shaders. there is no fallback support. jose. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel -- - Codito, ergo sum - I code, therefore I am -- The Rasterman (Carsten Haitzler)[EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] export of LD_LIBRARY_PATH in e_start_main.c
On Fri, 29 Dec 2006 04:03:26 -0500 Mike Frysinger [EMAIL PROTECTED] babbled: On Thursday 28 December 2006 21:40, Carsten Haitzler wrote: hmm - actually this will work - often, but not all the time. if prefix/bin is a symlink to for example prefix/.exec/i386/ (which i actually have seen for multi-arch shared nfs filesystems) this breaks down ok, but the way it's done now wont work either, so i dont see what you're getting at here yes it will - it handles this too. check the code. :) or it at least tries. also is this ldso feature something that exists outside of linux's ldso? as it's not documented... it is documented ... it isnt glibc-specific or linux-specific at all ... sun's ldso supports it as well since it's part of the elf library searching spec i think a more portable solution might be for e to revert its ld_lib_path changes e_start did (as its only really needed on app runtime link) or to remove it on exec of anything. they're equally portable but if you're going to pollute the env, then yes you should make sure the crap is cleansed before spawning sub processes ... of course, this wont help any modules that other people write and do exec's sure - though there are exec wrappers in e's code that modules should use :) at the very least, there should be an option for people to remove the option for package maintainers to use since the target audience here is simply developers i can see an option for package maintainers to build with when they know the package will never move anywhere. -mike -- - Codito, ergo sum - I code, therefore I am -- The Rasterman (Carsten Haitzler)[EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Custom screensaver to override E's desklock (work in progress)....
Carsten Haitzler (The Rasterman) wrote: I don't see any problem with this feature :) setting a command is ok- though what i think would be better is to have .desktop files with the screensaver commands and then just select from ones (maybe of Type=Screensaver) so you can pre-package configs for xscreensaver and other screensaver 3rd party apps out there. The current/new implememtation of the screensaver and desklock settings is not exactly what I was hoping for and IMHO not consistent. Although I can set a custom screensaver/desklock command, this command will be used for both screen-saving and desklocking as far as I can see. So when I set Enable X screensaver leaving Lock when the screensaver starts unchecked, but use xlock as my custom screensaver/desklock command (since I have problems with PAM), this command is used every time the screensaver activates (and of course locking my screen which I don't want). In order to have locking and screen-saving decoupled, I think there should be two custom command lines, one for the screensaver and the other one for the desklock command. The Lock when the screensaver starts option should be deactivated (greyed out) as soon as there is a custom screensaver command defined (which should take care of whether or not to lock the screen when starting). The custom desklock command should only be used when an explicit screen lock (e.g. launching Lock Screen from the system menu) is requested. Hope that makes sense. Cheers, Nick. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
[E-devel] Mixer control key bindings
Hi, the key bindings for Decrease/Increase Volume and Mute mixer don't seem to work for me (SuSE 10.1). Could somebody give me a short explanation of how they are implemented? I guess they are supposed to work and it's a problem on my side? Thanks a lot, Nick. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Custom screensaver to override E's desklock (work in progress)....
raster said: I don't see any problem with this feature :) setting a command is ok- though what i think would be better is to have .desktop files with the screensaver commands and then just select from ones (maybe of Type=Screensaver) so you can pre-package configs for xscreensaver and other screensaver 3rd party apps out there. This is a good idea, but having to pre-guess which possible screensaver programs one may want to use isn't ideal. So, we'd at least need a add screensaver option that builds the .desktop file. Nikolas Arend said: The current/new implememtation of the screensaver and desklock settings is not exactly what I was hoping for and IMHO not consistent. Although I can set a custom screensaver/desklock command, this command will be used for both screen-saving and desklocking as far as I can see. So when I set Enable X screensaver leaving Lock when the screensaver starts unchecked, but use xlock as my custom screensaver/desklock command (since I have problems with PAM), this command is used every time the screensaver activates (and of course locking my screen which I don't want). Do you have the PAM headers installed? Is pam support getting compiled in when you build e (check the value of HAVE_PAM in e17/apps/e/config.h)? (We probably need to fix desklock to not lock if pam isn't compiled in.) In order to have locking and screen-saving decoupled, I think there should be two custom command lines, one for the screensaver and the other one for the desklock command. The Lock when the screensaver starts option should be deactivated (greyed out) as soon as there is a custom screensaver command defined (which should take care of whether or not to lock the screen when starting). The custom desklock command should only be used when an explicit screen lock (e.g. launching Lock Screen from the system menu) is requested. This makes sense. Possibly coupled with the above suggestion from raster. So, a dropdown/ilist for each, with a button to add a new one? rephorm - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Custom screensaver to override E's desklock (work in progress)....
Brian Mattern wrote: Do you have the PAM headers installed? Is pam support getting compiled in when you build e (check the value of HAVE_PAM in e17/apps/e/config.h)? (We probably need to fix desklock to not lock if pam isn't compiled in.) Yes, headers are installed and are obviously found ok by configure. We've been through this in another thread and it seems that it's a SuSE specific PAM config problem. Sigh... In order to have locking and screen-saving decoupled, I think there should be two custom command lines, one for the screensaver and the other one for the desklock command. The Lock when the screensaver starts option should be deactivated (greyed out) as soon as there is a custom screensaver command defined (which should take care of whether or not to lock the screen when starting). The custom desklock command should only be used when an explicit screen lock (e.g. launching Lock Screen from the system menu) is requested. This makes sense. Possibly coupled with the above suggestion from raster. So, a dropdown/ilist for each, with a button to add a new one? That would be convenient for those who don't like to fiddle with xscreensaver/xlock/whatever command line options and flexible enough for everybody else. Nick. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Custom screensaver to override E's desklock (work in progress)....
Nikolas Arend wrote: Brian Mattern wrote: Do you have the PAM headers installed? Is pam support getting compiled in when you build e (check the value of HAVE_PAM in e17/apps/e/config.h)? (We probably need to fix desklock to not lock if pam isn't compiled in.) Yes, headers are installed and are obviously found ok by configure. We've been through this in another thread and it seems that it's a SuSE specific PAM config problem. Sigh... In order to have locking and screen-saving decoupled, I think there should be two custom command lines, one for the screensaver and the other one for the desklock command. The Lock when the screensaver starts option should be deactivated (greyed out) as soon as there is a custom screensaver command defined (which should take care of whether or not to lock the screen when starting). The custom desklock command should only be used when an explicit screen lock (e.g. launching Lock Screen from the system menu) is requested. This makes sense. Possibly coupled with the above suggestion from raster. So, a dropdown/ilist for each, with a button to add a new one? That would be convenient for those who don't like to fiddle with xscreensaver/xlock/whatever command line options and flexible enough for everybody else. Forgot... maybe s.th. like a Test screensaver button owould be good to see if the chosen custom command does what it's supposed to do? Nick - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Mixer control key bindings
Hi Nick, The implementation of the key binding in mixer module is similar to the implementation in language module. iirc, it was taken exactly from that module. The idea is, that module has to define the following things: 1) the action, which will be executed on key press, For example mixer_mute 2) the function, which is associated with the action. 3) The 'human readable' name of the action. For example Mute The Mixer. This name will appear in keybinding dialog. 4) The keybinding that uses the mixer_mute action. Actually this is optional. And has to be done if you want to have default binding. 5) Finally, you have to do reverse actions on module unload. 1) and 2) are done by those several lines of code: E_Actoin *a; a = e_action_add(mixer_mute); if (a) a-func.go = _some_binding_callback_funcition 3) is done basically by calling e_action_predef_name_set function (see e_action.h) 4) You have to define a E_Config_Binding_Key *eb variable and fill it in with the correct values. After doing that, you have to: (i) insert eb into e_config-key_bindings, and (ii) call to e_bindings_key_add to make the binding active. Note, that you have to enclose (i) and (ii) between e_manager_keys_ungrab() and e_managers_keys_grab() after. For further information look into language madule e_mod_keybindigs.c file. Feel free to contact me on IRC if you need more detailed explanation. sndev On Fri, Dec 29, 2006 at 04:12:15PM +0100, Nikolas Arend wrote: Hi, the key bindings for Decrease/Increase Volume and Mute mixer don't seem to work for me (SuSE 10.1). Could somebody give me a short explanation of how they are implemented? I guess they are supposed to work and it's a problem on my side? Thanks a lot, Nick. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel -- Aleksej Struk Master Degree Student Free University of Bozen-Bolzano Faculty of Computer Science phone: +39-0471-061749 cell phone: +39-3204627049 +370-61278908 [EMAIL PROTECTED] [EMAIL PROTECTED] - http://astruk.googlepages.com/home - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Window out of Screen?
Am Fri, 29 Dec 2006 10:50:47 +0900 schrieb Carsten Haitzler (The Rasterman): This happens because I use synaptic on my notebook with a bigger resolution. If I open it with ssh on my pc I couldn't resize the windows height. Ok, this could happen, but I expect that the maximize window button should maximize the window _not_ under the shelf. I found out that maximize behaves wrong if the window is slightly below a shelf. You could easily reproduce it: well 1. that's because the app asks to be that size as it remembers its own size and asks for it again next time it runs. Ok, but E17 is the window manager. A window couldn't force its size as I understand it. So why shoudln't E17 limit the max size to screen size even if the app requests a bigger size. Same with window position. If a app requests a position outside a window E17 should simply move it so that it's complete inside the viewable screen. I think such a policy would solve most of the problems. You should see it with the users eyes. Which E17 user hadn't yet seen this too big opened windows and missplaced dialog windows? Do you think this behaviour is good? I don't think so. it just happens to be how it works as unlike maximize routines in other wm's/desktops it is extremely generic and handles all sorts of obstacles of all shapes and sizes everywhere. it just happens to be how it's written and you see a bi-product of it. I had never problems like this with e.g. Metacity. (Ok, I've other problems with Metacity, but that's another story...) It's too easy to say: stupid application! We couldn't rewrite all apllications or use alternatives :-) regards Andreas - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Mixer control key bindings
On Fri, 29 Dec 2006 19:52:31 + Aleksej Struk [EMAIL PROTECTED] wrote: Hi Nick, The implementation of the key binding in mixer module is similar to the implementation in language module. iirc, it was taken exactly from that module. The idea is, that module has to define the following things: 1) the action, which will be executed on key press, For example mixer_mute 2) the function, which is associated with the action. 3) The 'human readable' name of the action. For example Mute The Mixer. This name will appear in keybinding dialog. 4) The keybinding that uses the mixer_mute action. Actually this is optional. And has to be done if you want to have default binding. 5) Finally, you have to do reverse actions on module unload. 1) and 2) are done by those several lines of code: E_Actoin *a; a = e_action_add(mixer_mute); if (a) a-func.go = _some_binding_callback_funcition 3) is done basically by calling e_action_predef_name_set function (see e_action.h) 4) You have to define a E_Config_Binding_Key *eb variable and fill it in with the correct values. After doing that, you have to: (i) insert eb into e_config-key_bindings, and (ii) call to e_bindings_key_add to make the binding active. Note, that you have to enclose (i) and (ii) between e_manager_keys_ungrab() and e_managers_keys_grab() after. For further information look into language madule e_mod_keybindigs.c file. Feel free to contact me on IRC if you need more detailed explanation. sndev On Fri, Dec 29, 2006 at 04:12:15PM +0100, Nikolas Arend wrote: Hi, the key bindings for Decrease/Increase Volume and Mute mixer don't seem to work for me (SuSE 10.1). Could somebody give me a short explanation of how they are implemented? I guess they are supposed to work and it's a problem on my side? Thanks a lot, Nick. Hi, as sndev said, the keybinding code in mixer was taken from the language module. The bindings in the mixer modules need at least one gadcon instance of the mixer in order to work. The mixer initialize is configuration only if an instance is created. Once the configuration is filled with the right mixer then the keybinding can works. So you need to put the mixer module in a shelf to have keybindings working. I could have move a part of the configuration code in order to make it works right after the module is loaded. But at that time I wasn't really seeing the point. The other issue can be because of the default value of the keybindings. I used some default keycodes for the media keys. Since I don't really know if there's some kind of specs making keyboards. I've asked a few people with different keyboards to give me there keysymbols and keycodes for this media keys. None of them had a keysymbols and the keycodes was the same. And so I set the keybindings by default on keycodes 160 174 and 176. It might not be always that values but you can change them on E keybindings config. lok - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Mixer control key bindings
lok wrote: On Fri, 29 Dec 2006 19:52:31 + Aleksej Struk [EMAIL PROTECTED] wrote: Hi Nick, The implementation of the key binding in mixer module is similar to the implementation in language module. iirc, it was taken exactly from that module. The idea is, that module has to define the following things: 1) the action, which will be executed on key press, For example mixer_mute 2) the function, which is associated with the action. 3) The 'human readable' name of the action. For example Mute The Mixer. This name will appear in keybinding dialog. 4) The keybinding that uses the mixer_mute action. Actually this is optional. And has to be done if you want to have default binding. 5) Finally, you have to do reverse actions on module unload. 1) and 2) are done by those several lines of code: E_Actoin *a; a = e_action_add(mixer_mute); if (a) a-func.go = _some_binding_callback_funcition 3) is done basically by calling e_action_predef_name_set function (see e_action.h) 4) You have to define a E_Config_Binding_Key *eb variable and fill it in with the correct values. After doing that, you have to: (i) insert eb into e_config-key_bindings, and (ii) call to e_bindings_key_add to make the binding active. Note, that you have to enclose (i) and (ii) between e_manager_keys_ungrab() and e_managers_keys_grab() after. For further information look into language madule e_mod_keybindigs.c file. Feel free to contact me on IRC if you need more detailed explanation. sndev On Fri, Dec 29, 2006 at 04:12:15PM +0100, Nikolas Arend wrote: Hi, the key bindings for Decrease/Increase Volume and Mute mixer don't seem to work for me (SuSE 10.1). Could somebody give me a short explanation of how they are implemented? I guess they are supposed to work and it's a problem on my side? Thanks a lot, Nick. Hi, thanks to all for your elaborate explanations! Hi, as sndev said, the keybinding code in mixer was taken from the language module. The bindings in the mixer modules need at least one gadcon instance of the mixer in order to work. The mixer initialize is configuration only if an instance is created. Once the configuration is filled with the right mixer then the keybinding can works. So you need to put the mixer module in a shelf to have keybindings working. Ah, that of course explains a lot. I don't have the mixer module running, in fact I defined the key bindings for volume increase/decrease and muting in order to _NOT_ have the need of running the mixer module (or any other mixer gui) ;-) Now that I added it to a shelf, it all works. But I'd rather like it if one wouldn't have to actually put the module into a shelf. I could have move a part of the configuration code in order to make it works right after the module is loaded. But at that time I wasn't really seeing the point. As I said above, I'd consider this a good thing. A hint somewhere that the module must be enabled/loaded to have the key bindings working maybe couldn't hurt either, IMHO. Thanks again and best regards, Nick. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Custom screensaver to override E's desklock (work in progress)....
On 12/29/2006 08:42, Nikolas Arend wrote: Carsten Haitzler (The Rasterman) wrote: I don't see any problem with this feature :) setting a command is ok- though what i think would be better is to have .desktop files with the screensaver commands and then just select from ones (maybe of Type=Screensaver) so you can pre-package configs for xscreensaver and other screensaver 3rd party apps out there. The current/new implememtation of the screensaver and desklock settings is not exactly what I was hoping for and IMHO not consistent. Although I can set a custom screensaver/desklock command, this command will be used for both screen-saving and desklocking as far as I can see. Yes. That is the present behavior. Though if you read the entire thread above... I think the desklock settings and the screensaver settings should be seperated, thus allowing unique custom settings as well. :) I'll get to it... though a little slow as my wife just gave birth to a bouncing baby boy yesterday afternoon. :D So when I set Enable X screensaver leaving Lock when the screensaver starts unchecked, but use xlock as my custom screensaver/desklock command (since I have problems with PAM), this command is used every time the screensaver activates (and of course locking my screen which I don't want). In order to have locking and screen-saving decoupled, I think there should be two custom command lines, one for the screensaver and the other one for the desklock command. The Lock when the screensaver starts option should be deactivated (greyed out) as soon as there is a custom screensaver command defined (which should take care of whether or not to lock the screen when starting). The custom desklock command should only be used when an explicit screen lock (e.g. launching Lock Screen from the system menu) is requested. Hope that makes sense. Make perfect sense. And I agree completely. :) Cheers, Nick. - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel -- Regards, Ravenlock - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Window out of Screen?
On Fri, 29 Dec 2006 19:16:39 +0100 Andreas Volz [EMAIL PROTECTED] babbled: Am Fri, 29 Dec 2006 10:50:47 +0900 schrieb Carsten Haitzler (The Rasterman): This happens because I use synaptic on my notebook with a bigger resolution. If I open it with ssh on my pc I couldn't resize the windows height. Ok, this could happen, but I expect that the maximize window button should maximize the window _not_ under the shelf. I found out that maximize behaves wrong if the window is slightly below a shelf. You could easily reproduce it: well 1. that's because the app asks to be that size as it remembers its own size and asks for it again next time it runs. Ok, but E17 is the window manager. A window couldn't force its size as I understand it. So why shoudln't E17 limit the max size to screen size even if the app requests a bigger size. because there are perfectly legitimate reasons to ask for windows bigger than the screen depending on usage patterns of the app. E can't make such quality choices for you. There are options for windows to have e override the application requests for size, location etc. (remember and lock options) for these kind of things, but e isn't going to go do this by default as it will likely cause more problems than it would solve. Same with window position. If a app requests a position outside a window E17 should simply move it so that it's complete inside the viewable screen. again - there are legitimate reasons. example:app creates window off screen then slides it onto the screen much like e does with effects when changing virtual desktops. I think such a policy would solve most of the problems. You should see it with the users eyes. Which E17 user hadn't yet seen this too big opened windows and missplaced dialog windows? as i said above - there are ways to fix this - but they are not a default. in fact no wm i have ever used does this by default. if the app is stupid - then the app will behave in a stupid manner. most application authors are very much unaware of the stupidity they code into their apps. apps that try and remember their own geometry, desktops, state etc. invariably get it wrong unless the authors are very good and can think ahead. e17 has features to say no - stupid app. don't do that! and fix it itself. Do you think this behaviour is good? I don't think so. it just happens to be how it works as unlike maximize routines in other wm's/desktops it is extremely generic and handles all sorts of obstacles of all shapes and sizes everywhere. it just happens to be how it's written and you see a bi-product of it. I had never problems like this with e.g. Metacity. (Ok, I've other problems with Metacity, but that's another story...) It's too easy to say: stupid application! We couldn't rewrite all apllications or use alternatives :-) because the INTENTIONS of the application can't always be divined in code in a wm - this, eventually leads to other nasty bugs when e gets it wrong. 'es maximize code happens to work that way because it is very generic to handle cases that simpler code doesn't. the bi-product is what you see. right now i don't have time to look into it - i am massively backlogged on even my email and i have a tonne of things to do. you will need to live with it as-is, or send patches (and such patches would then limit functionality - i won't go into that) regards Andreas -- - Codito, ergo sum - I code, therefore I am -- The Rasterman (Carsten Haitzler)[EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] export of LD_LIBRARY_PATH in e_start_main.c
On Fri, 29 Dec 2006 17:30:50 -0500 Mike Frysinger [EMAIL PROTECTED] babbled: On Friday 29 December 2006 06:33, Carsten Haitzler wrote: yes it will - it handles this too. check the code. :) or it at least tries. last i looked the code simply exported LD_LIBRARY_PATH=$bindir/../ ... so if it's changed since, then sure, i'd agree that the current code handles it well the code in e_prefix.c is devoted to trying to figure this stuff out so that e can be moved about and figure out its prefix simply from base execution without requiring the system to be set up (this helps people install it as a user in ~/ when they don't have root access etc.). i'd like to keep this feature and not play with compile flags (as this percolates down to compile flags for everything e may need). i'm loathe to play with compile flags... when i can play with code instead :) either way - the code removes its magic ld lib path stuff except when executing its own utils (which will want it). :) sure - though there are exec wrappers in e's code that modules should use :) if that's true then modules calling exec functions are broken so i wont belabor that point -mike -- - Codito, ergo sum - I code, therefore I am -- The Rasterman (Carsten Haitzler)[EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
