Re: [E-devel] Nightly builds - Was: Is this a community project?
On Sat, 21 Jul 2007 13:33:51 +1000 David Seikel [EMAIL PROTECTED] babbled: On Sat, 21 Jul 2007 12:24:25 +0900 Carsten Haitzler (The Rasterman) [EMAIL PROTECTED] wrote: On Sat, 21 Jul 2007 12:43:39 +1000 David Seikel [EMAIL PROTECTED] babbled: On Wed, 18 Jul 2007 12:30:53 +0900 Carsten Haitzler (The Rasterman) [EMAIL PROTECTED] wrote: can you... put the script in cvs somewhere? tough to talk about something we can't see. After much experimentation, we ended up with easy_e17.sh and a post install script. easy_e17.sh is available from morlenxus' server as I've mentioned before. The post install script relies on the current svn version, but that will be the next release soon enough. I'll put the post install script into cvs somewhere. The real question is - where? Create a new project in e17/apps/nightly_build? e17/test perhaps? The major problem at the moment is that sf.net is really fussy about emails you send to it. Seems like a typical workstation with a dynamic IP and default install will not be able to use sendmail mail.txt from a script to send to sf.net. Looks like you need a real mail server to do it for you. So some config of local mail systems may be needed if we want volunteers to use this to cover various distros etc. Once we are happy with the output generated and where it goes we can worry about OS / distro / CPU coverage details. tus why you should not worry and just run this on e2.enlightenment.org - it already can send email as part of bugzilla requirements. :) Yes, that is planned. Do I have a shell account on that and enough yes :) access to set up a cron job? Being able to send email, and being able to convince sf.net that we are allowed to send email to the list are two separate things. :/ true - but its easier for e.org as its an official server - your adsl link will be on an RBL somewhere as a dialup/user account ip thus be blocked. Does e2.enlightenment.org have enough grunt to run a half dozen qemu sessions for nightly testing on a few distro/CPU combinations? If not this is why volunteers may be needed later. Like I said, nothing to worry about now. it's a dual cpu dual core xeon @ 3ghz. 2gb ram. it's got enough. we also have e1.enlightenment.org too - same specs. we can split builds across them. e2 runs all our www stuff - e1 is anon cvs right now. also just run the build with really low priority - niced down so www or later anoncvs don't get too impacted. builds can just roll along in the background - the time they take is not important. -- - Codito, ergo sum - I code, therefore I am -- The Rasterman (Carsten Haitzler)[EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Deprecating Ecore_Desktop
On Sat, 21 Jul 2007 01:09:20 -0400 dan sinclair [EMAIL PROTECTED] babbled: On 21-Jul-07, at 12:29 AM, Carsten Haitzler (The Rasterman) wrote: On Tue, 17 Jul 2007 10:07:11 -0400 dan sinclair [EMAIL PROTECTED] babbled: All, We've had Efreet in CVS for a while now and it seems to be working well. Taking a look at the other apps in CVS the only one still using Ecore_Desktop that I noticed was Engage. Can the Engage maintainer please update the app to use Efreet so we can move Ecore_Desktop over into proto? also ecore_dbus should get deprecated soon. This was deprecated a while ago by rephorm I believe. It's been moved into proto/ oh dear me - it has too. i missed that! yay! -- - Codito, ergo sum - I code, therefore I am -- The Rasterman (Carsten Haitzler)[EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] E CVS: libs/evas raster
On Sat, 21 Jul 2007 00:32:27 -0300 Gustavo Sverzut Barbieri [EMAIL PROTECTED] babbled: On 7/21/07, Enlightenment CVS [EMAIL PROTECTED] wrote: Enlightenment CVS committal Author : raster Project : e17 Module : libs/evas Dir : e17/libs/evas/src/lib/canvas Modified Files: evas_object_textblock.c Log Message: no need for extra nul byte check - while terminated IF its null. so known condition. formatting too. === RCS file: /cvs/e/e17/libs/evas/src/lib/canvas/evas_object_textblock.c,v retrieving revision 1.141 retrieving revision 1.142 diff -u -3 -r1.141 -r1.142 --- evas_object_textblock.c 20 Jul 2007 22:30:12 - 1.141 +++ evas_object_textblock.c 21 Jul 2007 03:06:08 - 1.142 [...] @@ -2463,7 +2462,7 @@ evas_textblock_cursor_text_append(cur, map_itr); return; } - + if (map_itr map_itr) _advance_after_end_of_string(map_itr); } is this trailing whitespace really required? :-) I have emacs to highlight these on my files. not needed - but it seems jed's indenting replaced it. still about minor optimizations, do you think it worth to remove and ; since they're always present? which ones? -- Gustavo Sverzut Barbieri -- Jabber: [EMAIL PROTECTED] MSN: [EMAIL PROTECTED] ICQ#: 17249123 Skype: gsbarbieri Mobile: +55 (81) 9927 0010 - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ 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) - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] E CVS: libs/evas raster
On 7/21/07, The Rasterman Carsten Haitzler [EMAIL PROTECTED] wrote: On Sat, 21 Jul 2007 00:32:27 -0300 Gustavo Sverzut Barbieri still about minor optimizations, do you think it worth to remove and ; since they're always present? which ones? from escape_strings: lt;\0\x3c\0 - lt\0\x3c\0 this will require some changes, but will save (176 * 2) = 352 bytes, plus less data to look/process. -- Gustavo Sverzut Barbieri -- Jabber: [EMAIL PROTECTED] MSN: [EMAIL PROTECTED] ICQ#: 17249123 Skype: gsbarbieri Mobile: +55 (81) 9927 0010 - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Nightly builds - Was: Is this a community project?
On Sat, 21 Jul 2007 15:32:32 +0900 Carsten Haitzler (The Rasterman) [EMAIL PROTECTED] wrote: On Sat, 21 Jul 2007 13:33:51 +1000 David Seikel [EMAIL PROTECTED] babbled: access to set up a cron job? Being able to send email, and being able to convince sf.net that we are allowed to send email to the list are two separate things. :/ true - but its easier for e.org as its an official server - your adsl link will be on an RBL somewhere as a dialup/user account ip thus be blocked. I don't have an ADSL link, and I have a static IP. It's not RBL blocking that is the problem from here. It will be a problem for Random E. Developer though. I'll run the script from e2 right now and see how it goes. signature.asc Description: PGP signature - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] E CVS: libs/evas raster
On Sat, 21 Jul 2007 03:54:18 -0300 Gustavo Sverzut Barbieri [EMAIL PROTECTED] babbled: aaah. the standard escape delimiters - i guess it makes sense to remove them. i can't think of needing to have a different escape handler at the moment. On 7/21/07, The Rasterman Carsten Haitzler [EMAIL PROTECTED] wrote: On Sat, 21 Jul 2007 00:32:27 -0300 Gustavo Sverzut Barbieri still about minor optimizations, do you think it worth to remove and ; since they're always present? which ones? from escape_strings: lt;\0\x3c\0 - lt\0\x3c\0 this will require some changes, but will save (176 * 2) = 352 bytes, plus less data to look/process. -- Gustavo Sverzut Barbieri -- Jabber: [EMAIL PROTECTED] MSN: [EMAIL PROTECTED] ICQ#: 17249123 Skype: gsbarbieri Mobile: +55 (81) 9927 0010 -- - Codito, ergo sum - I code, therefore I am -- The Rasterman (Carsten Haitzler)[EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Nightly builds - Was: Is this a community project?
On Sat, 21 Jul 2007 17:07:57 +1000 David Seikel [EMAIL PROTECTED] wrote: I'll run the script from e2 right now and see how it goes. Who do I talk to about getting stuff installed on e2? Gonna need a few things, starting with autoconf, automake, ... you get the idea. mc would be nice as well. B-) signature.asc Description: PGP signature - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] patch for a better wallpaper dialog layout
Hannes Janetzek wrote:
hi,
if this patch is ok iwould go on and fix some of the other dialog
layouts.
a screenshot is here:
http://www.informatik.uni-bremen.de/~jeff/conf2.jpg
Regards,
Hannes jeffdameth Janetzek
Index: e_int_config_wallpaper.c
===
RCS file:
/var/cvs/e/e17/apps/e/src/modules/conf_wallpaper/e_int_config_wallpaper.c,v
retrieving revision 1.1
diff -u -r1.1 e_int_config_wallpaper.c
--- e_int_config_wallpaper.c 4 Jul 2007 15:09:24 - 1.1
+++ e_int_config_wallpaper.c 14 Jul 2007 18:07:58 -
@@ -57,7 +57,7 @@
e_int_config_wallpaper_desk(E_Container *con, const char *params)
{
int con_num, zone_num, desk_x, desk_y;
-
+
if (!params) return NULL;
con_num = zone_num = desk_x = desk_y = -1;
if (sscanf(params, %i %i %i %i, con_num, zone_num, desk_x, desk_y)
!= 4)
@@ -75,7 +75,7 @@
if (e_config_dialog_find(E, _config_wallpaper_dialog)) return NULL;
v = E_NEW(E_Config_Dialog_View, 1);
cw = E_NEW(E_Config_Wallpaper, 1);
-
+
v-create_cfdata = _create_data;
v-free_cfdata = _free_data;
v-basic.apply_cfdata = _basic_apply_data;
@@ -109,7 +109,7 @@
E_Config_Dialog_Data *cfdata;
char path[4096];
const char *homedir;
-
+
cfdata = dia-cfdata;
homedir = e_user_homedir_get();
cfdata-fmdir = 1;
@@ -132,7 +132,7 @@
e_int_config_wallpaper_import_done(E_Config_Dialog *dia)
{
E_Config_Dialog_Data *cfdata;
-
+
cfdata = dia-cfdata;
cfdata-win_import = NULL;
}
@@ -141,7 +141,7 @@
e_int_config_wallpaper_gradient_done(E_Config_Dialog *dia)
{
E_Config_Dialog_Data *cfdata;
-
+
cfdata = dia-cfdata;
cfdata-dia_gradient = NULL;
}
@@ -150,7 +150,7 @@
_cb_button_up(void *data1, void *data2)
{
E_Config_Dialog_Data *cfdata;
-
+
cfdata = data1;
if (cfdata-o_fm)
e_fm2_parent_go(cfdata-o_fm);
@@ -162,7 +162,7 @@
_cb_files_changed(void *data, Evas_Object *obj, void *event_info)
{
E_Config_Dialog_Data *cfdata;
-
+
cfdata = data;
if (!cfdata-o_fm) return;
if (!e_fm2_has_parent_get(cfdata-o_fm))
@@ -187,7 +187,7 @@
E_Fm2_Icon_Info *ici;
const char *realpath;
char buf[4096];
-
+
cfdata = data;
if (!cfdata-o_fm) return;
selected = e_fm2_selected_list_get(cfdata-o_fm);
@@ -225,7 +225,7 @@
E_Config_Dialog_Data *cfdata;
const char *p, *homedir;
char buf[4096];
-
+
cfdata = data;
if (!cfdata-bg) return;
if (!cfdata-o_fm) return;
@@ -247,18 +247,18 @@
else
p = cfdata-bg;
}
-
+
e_fm2_select_set(cfdata-o_fm, p, 1);
e_fm2_file_show(cfdata-o_fm, p);
}
static void
-_cb_files_files_deleted(void *data, Evas_Object *obj, void *event_info)
+_cb_files_files_deleted(void *data, Evas_Object *obj, void *event_info)
{
E_Config_Dialog_Data *cfdata;
Evas_List *sel, *all, *n;
E_Fm2_Icon_Info *ici, *ic;
-
+
cfdata = data;
if (!cfdata-bg) return;
if (!cfdata-o_fm) return;
@@ -269,23 +269,23 @@
if (!sel) return;
ici = sel-data;
-
+
all = evas_list_find_list(all, ici);
n = evas_list_next(all);
- if (!n)
+ if (!n)
{
n = evas_list_prev(all);
if (!n) return;
}
-
+
ic = n-data;
if (!ic) return;
-
+
e_fm2_select_set(cfdata-o_fm, ic-file, 1);
e_fm2_file_show(cfdata-o_fm, ic-file);
-
+
evas_list_free(n);
-
+
evas_object_smart_callback_call(cfdata-o_fm, selection_change, cfdata);
}
@@ -294,7 +294,7 @@
{
E_Config_Dialog_Data *cfdata;
const char *f;
-
+
cfdata = data;
if (cfdata-use_theme_bg)
{
@@ -320,7 +320,7 @@
E_Config_Dialog_Data *cfdata;
char path[4096];
const char *homedir;
-
+
cfdata = data;
if (cfdata-fmdir == 1)
{
@@ -338,11 +338,11 @@
_cb_import(void *data1, void *data2)
{
E_Config_Dialog_Data *cfdata;
-
+
cfdata = data1;
if (cfdata-win_import)
e_win_raise(cfdata-win_import);
- else
+ else
cfdata-win_import = e_int_config_wallpaper_import(cfdata-cfd);
}
@@ -350,11 +350,11 @@
_cb_gradient(void *data1, void *data2)
{
E_Config_Dialog_Data *cfdata;
-
+
cfdata = data1;
if (cfdata-dia_gradient)
e_win_raise(cfdata-dia_gradient-win);
- else
+ else
cfdata-dia_gradient = e_int_config_wallpaper_gradient(cfdata-cfd);
}
@@ -364,7 +364,7 @@
char path[4096];
E_Config_Wallpaper *cw;
const E_Config_Desktop_Background *cfbg;
-
+
cw = cfdata-cfd-data;
if (cw-specific_config)
@@ -400,10 +400,10 @@
cfdata-bg = strdup(cfbg-file);
}
}
-
- if ((!cfdata-bg)
Re: [E-devel] patch for a better wallpaper dialog layout
Alberto Castro wrote: Hannes Janetzek wrote: hi, if this patch is ok iwould go on and fix some of the other dialog layouts. a screenshot is here: http://www.informatik.uni-bremen.de/~jeff/conf2.jpg Regards, Hannes jeffdameth Janetzek (accidently i replied to Raster instead of the list.) I'm going to be really anal about this. Do we really really need to have a go up a directory button ? Yes, we need the go up button because a user may have sub-dirs in their Personal folder (ie: ~/.e/e/backgrounds/Space Images). isn't that what the file dialog is for ? Which file dialog are you referring to ? That's enough questions, but don't you think both the personal, system, including the theme wallpaper should be part of the list itself (at all times) (you could even add a header, to separate them or make them more visible to the user). We actually had it like this a long time ago, not sure why the change tho. I don't see why we need 3 different sources when they could all be selected from the list itself. I'm not seeing 3 different sources...I see 2...Personal and System. Now if you allow me to be more critical, the arrow on the go up a directory button never once changes states. Its always the same image, regardless of the button state. Is it supposed to ? devilhorns - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Nightly builds - Was: Is this a community project?
On Sat, 21 Jul 2007 17:22:39 +1000 David Seikel [EMAIL PROTECTED] babbled: On Sat, 21 Jul 2007 17:07:57 +1000 David Seikel [EMAIL PROTECTED] wrote: I'll run the script from e2 right now and see how it goes. Who do I talk to about getting stuff installed on e2? Gonna need a few things, starting with autoconf, automake, ... you get the idea. mc would be nice as well. B-) log out and in - use sudo. :) -- - Codito, ergo sum - I code, therefore I am -- The Rasterman (Carsten Haitzler)[EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] patch for a better wallpaper dialog layout
On Sat, 21 Jul 2007 04:40:55 -0400 Christopher Michael [EMAIL PROTECTED] babbled: Alberto Castro wrote: Hannes Janetzek wrote: hi, if this patch is ok iwould go on and fix some of the other dialog layouts. a screenshot is here: http://www.informatik.uni-bremen.de/~jeff/conf2.jpg Regards, Hannes jeffdameth Janetzek (accidently i replied to Raster instead of the list.) I'm going to be really anal about this. Do we really really need to have a go up a directory button ? Yes, we need the go up button because a user may have sub-dirs in their Personal folder (ie: ~/.e/e/backgrounds/Space Images). isn't that what the file dialog is for ? Which file dialog are you referring to ? this is the wallpaper and theme selector dialogs (for example) - not file selector. chris is right - you can create subdirs and use them to store files/wallpapers etc. That's enough questions, but don't you think both the personal, system, including the theme wallpaper should be part of the list itself (at all times) (you could even add a header, to separate them or make them more visible to the user). We actually had it like this a long time ago, not sure why the change tho. it changed because it became a filemanager backed list - and it will only display 1 directory at a time. personal and system are in completely separate places on the fs - thus a switch to swap between them. it was a quick way top keep the functionality with the new ui widgets. I don't see why we need 3 different sources when they could all be selected from the list itself. I'm not seeing 3 different sources...I see 2...Personal and System. Now if you allow me to be more critical, the arrow on the go up a directory button never once changes states. Its always the same image, regardless of the button state. Is it supposed to ? devilhorns - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ 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) - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Nightly builds - Was: Is this a community project?
On Sat, 21 Jul 2007 18:04:35 +0900 Carsten Haitzler (The Rasterman) [EMAIL PROTECTED] wrote: On Sat, 21 Jul 2007 17:22:39 +1000 David Seikel [EMAIL PROTECTED] babbled: On Sat, 21 Jul 2007 17:07:57 +1000 David Seikel [EMAIL PROTECTED] wrote: I'll run the script from e2 right now and see how it goes. Who do I talk to about getting stuff installed on e2? Gonna need a few things, starting with autoconf, automake, ... you get the idea. mc would be nice as well. B-) log out and in - use sudo. :) It's basically working on e2 now. Compiles as an ordinary user, no sudo required. Running at nice 19, so low impact on the rest of the box. Installing to my home directory. Most of the dependencies are installed, still some obscure ones left though. Most of the packages are building fine. Most of the failures are missing dependencies, or packages that could do with some pkg-config lovin'. No email out, not even to myself. Have to figure out where in the web site to put the results. Once it is all sorted, then I'll hook it up to cron. Getting tired, I'll sort the rest out tomorrow. signature.asc Description: PGP signature - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Nightly build log for E17
Hi list, I'm sorry if this is actually a repost of a message I wrote a couple of days ago. However, it may be that my MDA (which wasn't properly configured back then) failed to send that message. If it is a repost, I deeply apologize. I have a couple of comments about the format of this build log. Obviously, this is a bikeshed thing and I'll avoid posting about it again. Obviously, these comments are just my impressions and you needn't share them. Excerpts from David Seikel's message of Sat Jul 21 06:13:57 +0300 2007: Build log for Enlightenment DR 0.17 on 2007-07-21 13:11:41 +1000 Ubuntu 6.10 \n \l Linux cluster 2.6.17-11-generic #2 SMP Fri May 18 23:39:08 UTC 2007 i686 GNU/Linux autoconf (GNU Autoconf) 2.60 automake (GNU automake) 1.9.6 ltmain.sh (GNU libtool) 1.5.22 Debian 1.5.22-4 (1.1220.2.365 2005/12/18 22:14:06) gettext (GNU gettext-runtime) 0.15 1. The list of projects that fail should probably be the first thing to see after the name of the OS. The detailed list of autotools versions and such should probably only be detailed in the log on the web page, or near the end of this message. Failed Packages (and command that failed): eflpp make -j 2 elapse make -j 2 epdf ./autogen.sh --prefix=/opt/e17 --cache-file=/tmp/easy_e17/easy_e17.cache e_phys ./bootstrap etk_extra ./autogen.sh --prefix=/opt/e17 --cache-file=/tmp/easy_e17/easy_e17.cache etk-perl make install Evas_Perl make -j 2 2. The list of failed projects should (at least in the beginning of the message) be a single line (possibly wrapped) of text. example: Failed Packages: eflpp elapse epdf e_phys etk_extra etk-perl Evas_Perl 3. Each project that failed to get built should have a link to its log something like: Details about failed packages: eflpp: failed command: make -j 2 build log: http://onefang.humbug.org.au/E-build-logs/eflpp.log ... Packages with no supported build system: eflame, enthrall, esmart_rsvg, euphoria, exorcist, nexus, ruby-efl, Skipped Packages: camE, embrace, enotes, enscribe, epbb, eplay, erss, etk_server, etox, evoak, gfx_routines, lvs-gui, med, notgame, webcam, Successful Packages: alarm, bling, cpu, deskshow, eclair, ecore, edb, e_dbus, edje_editor, edje, edje_viewer, edvi, eet, efreet, elation, elicit, elitaire, e, embryo, emotion, emphasis, empower, emu, engage, engrave, engycad, enhance, enity, enterminus, entice, entrance_edit_gui, entrance, entropy, envision, epeg, ephoto, epsilon, equate, esmart, estickies, etk, e_utils, evas, evfs, evolve, ewl, examine, exhibit, exml, expedite, express, extrackt, feh, flame, forecasts, iconbar, imlib2_loaders, imlib2, Imlib2_Perl, imlib2_tools, language, mail, mem, mixer, moon, net, news, pesh, photo, rage, rain, screenshot, scrot, slideshow, snow, taskbar, tclock, uptime, weather, winselector, wlan, 4. The list of projects that were successfully built shouldn't be written, I think. The whole point of this system is to tell people what projects haven't been built. A lot of lines of text are taken up by the projects that haven been successfully built. Build logs are available at http://onefang.humbug.org.au/E-build-logs/ 5. The build logs should be dated, but I'm sure that will happen (something like: http://onefang.humbug.org.au/E-build-logs/2007/07/21) The nightly build system tries to build everything in the Enlightenment CVS or give a reason why. Currently eterm, e16, e17/docs, e17/test, and web are not built. Feel free to enlighten me about any decisions. The web directory does not contain anything that needs to be built. The e17/test directory is just a dumping ground for old testing stuff that is not meant to be built automatically. The e17/docs directory contains documentation. Some of it can be built but I have not bothered yet. The eterm and e16 directories are not currently being built because the build script used as the basis for the nightly build system didn't support those. This may change. Maybe I'm not understanding this and this part isn't really a part of the log. But if it is, I think this is maybe a bit too verbose. How about something like: The nightly build system tries to build everything in the Enlightenment CVS or give a reason why it isn't being built. For more information about this system, please go to this link: ... The skipped result means that the nightly build system was configured to skip that particular project for some reason. Those reasons should be listed here - etox- Obsolete. evoak - Obsolete. enscribe- Unmaintained, broken, and too old to fix. med - Unmaintained, broken, and too old to fix. camE- Unmaintained, broken, and too old to fix. embrace - Unmaintained, broken, and too old to fix. epbb- Unmaintained, broken, and too old to fix. eplay - Unmaintained, broken, and too old to fix. erss- Unmaintained, broken, and too old to fix. lvs-gui
Re: [E-devel] [PATCH] adding get workarea to Ecore_X
Carsten Haitzler wrote: On Fri, 20 Jul 2007 16:55:09 +0200 Dr. Michael 'Mickey' Lauer [EMAIL PROTECTED] babbled: I found this patch to be necessary when working with Ecore X windows that register with the window manager as the desktop window -- otherwise I can't get the correct size of the desktop minus the space that's occupied by panels etc. (The only glitch is that the semantics is not symmetrical to set_workareas where you do additional processing depending on your amount of virtual workspaces) could you make it symmetrical? basically you could have multiple areas... it's bad not to be symmetrical I agree, however, the semantics of the NETWM_WORKAREA atom is clearly defined in http://standards.freedesktop.org/wm-spec/wm-spec-1.4.html#id2510516 So this is really just the workarea for _one_ desktop. It might be wise to add ecore_desk_get_workareas (plural) in addition, but I wouldn't know what to compute there. Regards, :M: -- Michael 'Mickey' Lauer | IT-Freelancer | http://www.vanille-media.de - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] [PATCH] adding get workarea to Ecore_X
Dr. Michael 'Mickey' Lauer wrote: Carsten Haitzler wrote: On Fri, 20 Jul 2007 16:55:09 +0200 Dr. Michael 'Mickey' Lauer [EMAIL PROTECTED] babbled: I found this patch to be necessary when working with Ecore X windows that register with the window manager as the desktop window -- otherwise I can't get the correct size of the desktop minus the space that's occupied by panels etc. (The only glitch is that the semantics is not symmetrical to set_workareas where you do additional processing depending on your amount of virtual workspaces) could you make it symmetrical? basically you could have multiple areas... it's bad not to be symmetrical I agree, however, the semantics of the NETWM_WORKAREA atom is clearly defined in http://standards.freedesktop.org/wm-spec/wm-spec-1.4.html#id2510516 So this is really just the workarea for _one_ desktop. It might be wise to add ecore_desk_get_workareas (plural) in addition, but I wouldn't know what to compute there. Not correct, read the spec again. _NET_WORKAREA (x, y, width, height CARDINAL[][4]/32) holds the work area for *each* desktop, i.e. an array of [x, y, width, height]'s. /Kim - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] [PATCH] adding get workarea to Ecore_X
Kim Woelders wrote: Dr. Michael 'Mickey' Lauer wrote: Carsten Haitzler wrote: On Fri, 20 Jul 2007 16:55:09 +0200 Dr. Michael 'Mickey' Lauer [EMAIL PROTECTED] babbled: I found this patch to be necessary when working with Ecore X windows that register with the window manager as the desktop window -- otherwise I can't get the correct size of the desktop minus the space that's occupied by panels etc. (The only glitch is that the semantics is not symmetrical to set_workareas where you do additional processing depending on your amount of virtual workspaces) could you make it symmetrical? basically you could have multiple areas... it's bad not to be symmetrical I agree, however, the semantics of the NETWM_WORKAREA atom is clearly defined in http://standards.freedesktop.org/wm-spec/wm-spec-1.4.html#id2510516 So this is really just the workarea for _one_ desktop. It might be wise to add ecore_desk_get_workareas (plural) in addition, but I wouldn't know what to compute there. Not correct, read the spec again. _NET_WORKAREA (x, y, width, height CARDINAL[][4]/32) holds the work area for *each* desktop, i.e. an array of [x, y, width, height]'s. Ah, that's how it's meant -- thanks. I see, we should expose the complete array to the caller then. Regards, :M: -- Michael 'Mickey' Lauer | IT-Freelancer | http://www.vanille-media.de - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
[E-devel] Worldmap with Devs - GPS Coords in info.txt
Hi guys, today there was an idea of a hackfest in the #etk channel on freenode.net. For managing this the idea came up to have an overview where the devs are located on the earth. So I searched a bit and found xplanet that seems perfect for this job. It will get a list of devs with abolute world GPS coords and names and create a nice looking world map with the devs on it. Here is an example: Small version: http://thunder-development.de/test/etk_developers-small.png High-Res version: http://thunder-development.de/test/etk_developers.png (attention 11mb) Morlenxus had the idea to have the devs coords inside the info.txt in CVS. So a script on the server could fetch all coords out of this files and run xplanet to generate the image. At least for the high-res version it could maybe be a hard task for a webserver, here it takes 30 sec of 100% load on a Celeron M 1.6 to generate it. So maybe it would be better to just have the coords in the info.txt file, but the collection of them and generation of the image will be done from time to time manually. Would like to hear your opinions :) Regards Christian 'Thunder2002' Hunke - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Worldmap with Devs - GPS Coords in info.txt
On Sa, 21.07.07 21:48 Christian Hunke [EMAIL PROTECTED] wrote: Hi guys, today there was an idea of a hackfest in the #etk channel on freenode.net. For managing this the idea came up to have an overview where the devs are located on the earth. So I searched a bit and found xplanet that seems perfect for this job. It will get a list of devs with abolute world GPS coords and names and create a nice looking world map with the devs on it. Here is an example: Small version: http://thunder-development.de/test/etk_developers-small.png High-Res version: http://thunder-development.de/test/etk_developers.png (attention 11mb) Morlenxus had the idea to have the devs coords inside the info.txt in CVS. So a script on the server could fetch all coords out of this files and run xplanet to generate the image. At least for the high-res version it could maybe be a hard task for a webserver, here it takes 30 sec of 100% load on a Celeron M 1.6 to generate it. So maybe it would be better to just have the coords in the info.txt file, but the collection of them and generation of the image will be done from time to time manually. Would like to hear your opinions :) Regards Christian 'Thunder2002' Hunke Cool stuff! Having these coordinates one could also create a shapefile and import it to google earth, as for example an e-devs layer.. like: click on raster, and GE starts spinning globe a zooming to raster's location, hehe signature.asc Description: PGP signature - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Worldmap with Devs - GPS Coords in info.txt
Am Sat, 21 Jul 2007 21:48:50 +0200 schrieb Christian Hunke: Hi guys, today there was an idea of a hackfest in the #etk channel on freenode.net. For managing this the idea came up to have an overview where the devs are located on the earth. So I searched a bit and found xplanet that seems perfect for this job. It will get a list of devs with abolute world GPS coords and names and create a nice looking world map with the devs on it. Here is an example: Small version: http://thunder-development.de/test/etk_developers-small.png High-Res version: http://thunder-development.de/test/etk_developers.png (attention 11mb) Morlenxus had the idea to have the devs coords inside the info.txt in CVS. So a script on the server could fetch all coords out of this files and run xplanet to generate the image. At least for the high-res version it could maybe be a hard task for a webserver, here it takes 30 sec of 100% load on a Celeron M 1.6 to generate it. So maybe it would be better to just have the coords in the info.txt file, but the collection of them and generation of the image will be done from time to time manually. Would like to hear your opinions :) Cool! Please tell me how you format the GPS coords in your info.txt. Perhaps we could display it later on the OSM[1] map. http://openstreetmap.org/ regards Andreas - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] logout
Carsten Haitzler (The Rasterman) írta: On Mon, 16 Jul 2007 03:45:34 +1000 David Seikel [EMAIL PROTECTED] babbled: On Sun, 15 Jul 2007 17:26:59 +0200 Peter Parkanyi [EMAIL PROTECTED] wrote: I have a rather annoying problem with e17 on my Gentoo box. Wait. This isn't an e17 bug. This is obviously a skype bug, but skype is closed, so I can't really do anything with it. When I hit Logout, Shut Down, Reboot, and so on in the E menu, every application quits, except skype, because it only closes its main window. My suggestion is that it would be good if I could set somewhere to autokill specified stuff when logout, or specify a timeout for logout, and if the applications doesn't quit in that time, e17 autokills (not -9) them, or simply quit (if e17 starts applications as children, then all of the started apps are forced to quit, right?). This would be very handy. There is already a timeout, then E17 asks the user if it's OK to kill everything. This gives the user a chance to manually close down things properly or otherwise respond appropriately. The Exit menu option does an unconditional exit. Maybe a configuration option to unconditionally kill things after the timeout would do the trick? generally i think this is a bad idea - a lot of apps will put up are you sure you want to exit - you haven't saved your files dialogs. you want the user to go handle all of these first - then logout/shutdown etc. this is a skype specific issue due to it wanting a system tray unconditionally. This is skype specific, yes, but the interesting thing about it that if i click menu-system-logout on the same virtual desktop that skype's on, E kills it, no problem. But if i do it on another virtual desk, skype doesn't quit. Is this still a skype or rather an E related bug? rhapsodhy - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
[E-devel] E_DBus - IPC replacement
Hello Developers, Recently there has been a move towards modularizing the config dialogs within E17. With the development in this area raster and I were thinking it would be a nice time to modularize the IPC code. That is, the enlightenment binary should only contain a minimal amount of IPC functionality (Module, Restart, Profiles) while the rest should be extended via modules. The problem with the current IPC framework is that the handlers are not dynamically appendable. This could be fixed. However, I believe that dbus should be the way to advertise out IPC as it seems like the standard way to do things now (look at, pidgin, gnome, rhythmbox etc. if you will). The final goal is to remove the e_ipc_handlers* files completely. e_ipc is still needed for e_fm and thumbnailing. Attached is a file that implements the core dbus subsystem in E17. The next step is to expose register / unregister API to allow modules to register interfaces with the enlightenment window manager core RemoteObject. Currently any feedback and opinions will be helpful. Current API: !DOCTYPE node PUBLIC -//freedesktop//DTD D-BUS Object Introspection 1.0//EN http://www.freedesktop.org/standards/dbus/1.0/introspect.dtd; node name=/org/enlightenment/wm/RemoteObject interface name=org.freedesktop.DBus.Introspectable method name=Introspect arg type=s direction=out/ /method /interface interface name=org.enlightenment.wm.Core method name=Restart /method method name=Shutdown /method /interface interface name=org.enlightenment.wm.Module method name=Load arg type=s direction=in/ /method method name=Unload arg type=s direction=in/ /method method name=Enable arg type=s direction=in/ /method method name=Disable arg type=s direction=in/ /method method name=List arg type=a(si) direction=out/ /method /interface /node -Stafford - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Nightly builds - Was: Is this a community project?
On Sat, 21 Jul 2007 22:24:50 +1000 David Seikel [EMAIL PROTECTED] babbled: On Sat, 21 Jul 2007 18:04:35 +0900 Carsten Haitzler (The Rasterman) [EMAIL PROTECTED] wrote: On Sat, 21 Jul 2007 17:22:39 +1000 David Seikel [EMAIL PROTECTED] babbled: On Sat, 21 Jul 2007 17:07:57 +1000 David Seikel [EMAIL PROTECTED] wrote: I'll run the script from e2 right now and see how it goes. Who do I talk to about getting stuff installed on e2? Gonna need a few things, starting with autoconf, automake, ... you get the idea. mc would be nice as well. B-) log out and in - use sudo. :) It's basically working on e2 now. Compiles as an ordinary user, no sudo required. Running at nice 19, so low impact on the rest of the box. Installing to my home directory. Most of the dependencies are installed, still some obscure ones left though. Most of the packages are building fine. Most of the failures are missing dependencies, or packages that could do with some pkg-config lovin'. No email out, not even to myself. Have to figure out where in the web site to put the results. Once it is all sorted, then I'll hook it up to cron. /var/www has all the websites - you might want to put it in download for now. Getting tired, I'll sort the rest out tomorrow. p.s. - might also want to try ionice as well as nice :) -- - Codito, ergo sum - I code, therefore I am -- The Rasterman (Carsten Haitzler)[EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] logout
On Sat, 21 Jul 2007 23:45:20 +0200 Peter Parkanyi [EMAIL PROTECTED] babbled: Carsten Haitzler (The Rasterman) írta: On Mon, 16 Jul 2007 03:45:34 +1000 David Seikel [EMAIL PROTECTED] babbled: On Sun, 15 Jul 2007 17:26:59 +0200 Peter Parkanyi [EMAIL PROTECTED] wrote: I have a rather annoying problem with e17 on my Gentoo box. Wait. This isn't an e17 bug. This is obviously a skype bug, but skype is closed, so I can't really do anything with it. When I hit Logout, Shut Down, Reboot, and so on in the E menu, every application quits, except skype, because it only closes its main window. My suggestion is that it would be good if I could set somewhere to autokill specified stuff when logout, or specify a timeout for logout, and if the applications doesn't quit in that time, e17 autokills (not -9) them, or simply quit (if e17 starts applications as children, then all of the started apps are forced to quit, right?). This would be very handy. There is already a timeout, then E17 asks the user if it's OK to kill everything. This gives the user a chance to manually close down things properly or otherwise respond appropriately. The Exit menu option does an unconditional exit. Maybe a configuration option to unconditionally kill things after the timeout would do the trick? generally i think this is a bad idea - a lot of apps will put up are you sure you want to exit - you haven't saved your files dialogs. you want the user to go handle all of these first - then logout/shutdown etc. this is a skype specific issue due to it wanting a system tray unconditionally. This is skype specific, yes, but the interesting thing about it that if i click menu-system-logout on the same virtual desktop that skype's on, E kills it, no problem. But if i do it on another virtual desk, skype doesn't quit. Is this still a skype or rather an E related bug? unsure there. at this point i don't know what in e would cause it as it tries to delete ALL windows it tracks - irrespective of desktop. it sets up a timer then when that times out, if any windows are left, e complains (the timeout is 20 seconds). it monitors windows being closed/deleted - and once the window list is empty, it actually exits or begins system shutdown/reboot etc. -- - Codito, ergo sum - I code, therefore I am -- The Rasterman (Carsten Haitzler)[EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] [PATCH] adding get workarea to Ecore_X
On Sat, 21 Jul 2007 18:51:34 +0200 Dr. Michael 'Mickey' Lauer [EMAIL PROTECTED] babbled: Kim Woelders wrote: Dr. Michael 'Mickey' Lauer wrote: Carsten Haitzler wrote: On Fri, 20 Jul 2007 16:55:09 +0200 Dr. Michael 'Mickey' Lauer [EMAIL PROTECTED] babbled: I found this patch to be necessary when working with Ecore X windows that register with the window manager as the desktop window -- otherwise I can't get the correct size of the desktop minus the space that's occupied by panels etc. (The only glitch is that the semantics is not symmetrical to set_workareas where you do additional processing depending on your amount of virtual workspaces) could you make it symmetrical? basically you could have multiple areas... it's bad not to be symmetrical I agree, however, the semantics of the NETWM_WORKAREA atom is clearly defined in http://standards.freedesktop.org/wm-spec/wm-spec-1.4.html#id2510516 So this is really just the workarea for _one_ desktop. It might be wise to add ecore_desk_get_workareas (plural) in addition, but I wouldn't know what to compute there. Not correct, read the spec again. _NET_WORKAREA (x, y, width, height CARDINAL[][4]/32) holds the work area for *each* desktop, i.e. an array of [x, y, width, height]'s. Ah, that's how it's meant -- thanks. I see, we should expose the complete array to the caller then. correct - thus the set api call being able to set an array - get should do the same. Regards, :M: -- Michael 'Mickey' Lauer | IT-Freelancer | http://www.vanille-media.de - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ 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) - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Worldmap with Devs - GPS Coords in info.txt
On Sat, 21 Jul 2007 21:48:50 +0200 Christian Hunke [EMAIL PROTECTED] babbled: Hi guys, today there was an idea of a hackfest in the #etk channel on freenode.net. For managing this the idea came up to have an overview where the devs are located on the earth. So I searched a bit and found xplanet that seems perfect for this job. It will get a list of devs with abolute world GPS coords and names and create a nice looking world map with the devs on it. Here is an example: Small version: http://thunder-development.de/test/etk_developers-small.png High-Res version: http://thunder-development.de/test/etk_developers.png (attention 11mb) Morlenxus had the idea to have the devs coords inside the info.txt in CVS. So a script on the server could fetch all coords out of this files and run xplanet to generate the image. At least for the high-res version it could maybe be a hard task for a webserver, here it takes 30 sec of 100% load on a Celeron M 1.6 to generate it. So maybe it would be better to just have the coords in the info.txt file, but the collection of them and generation of the image will be done from time to time manually. Would like to hear your opinions :) i'd be happy to put my co-ords in info.txt. we can just run locations through google maps to get co-ords for now and put them in info.txt's Regards Christian 'Thunder2002' Hunke - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ 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) - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] Worldmap with Devs - GPS Coords in info.txt
Andreas Volz wrote: Am Sat, 21 Jul 2007 21:48:50 +0200 schrieb Christian Hunke: Hi guys, today there was an idea of a hackfest in the #etk channel on freenode.net. BB :) For managing this the idea came up to have an overview where the devs are located on the earth. So I searched a bit and found xplanet that seems perfect for this job. Even an early e17 version of this :) Triple W00T Ubber-Moo Werd Score !! http://www.squaregoldfish.co.uk/software/e17xplanet/ It will get a list of devs with abolute world GPS coords and names and create a nice looking world map with the devs on it. Have done something like this in the past w/ Google map Api. Don't have the page anymore tho, but pretty simple code... Here is an example: Small version: http://thunder-development.de/test/etk_developers-small.png High-Res version: http://thunder-development.de/test/etk_developers.png (attention 11mb) Morlenxus had the idea to have the devs coords inside the info.txt in CVS. Simple, effect, clean, efficient...what's not to like :) So a script on the server could fetch all coords out of this files and run xplanet to generate the image. Hmmm, seems a nightly build script maybe coming soon to. At least for the high-res version it could maybe be a hard task for a webserver, here it takes 30 sec of 100% load on a Celeron M 1.6 to generate it. Proally not much disk space if done html via google map... So maybe it would be better to just have the coords in the info.txt file, but the collection of them and generation of the image will be done from time to time manually. Sounds like a great plan !! This should be jumped on... Would like to hear your opinions :) Cool! Please tell me how you format the GPS coords in your info.txt. Perhaps we could display it later on the OSM[1] map. IIRC, the map api works best in lat-long. http://openstreetmap.org/ regards Andreas dh - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] E_DBus - IPC replacement
Stafford Horne wrote: Hello Developers, Recently there has been a move towards modularizing the config dialogs within E17. With the development in this area raster and I were thinking it would be a nice time to modularize the IPC code. That is, the enlightenment binary should only contain a minimal amount of IPC functionality (Module, Restart, Profiles) Perhaps some sort of Session Management here ? Incorporate w/ remembers startup... while the rest should be extended via modules. The problem with the current IPC framework is that the handlers are not dynamically appendable. This could be fixed. However, I believe that dbus should be the way to advertise out IPC as it seems like the standard way to do things now (look at, pidgin, gnome, rhythmbox etc. if you will). But are we gtk? Can we not come up w/ something better? Yes a dbus layer may/will be needed to integrate w/ other apps...But really, should we rely on dbus for socket comms.?, or do something ourselves. Not totally against the idea, but dbus is..:( The final goal is to remove the e_ipc_handlers* files completely. e_ipc is still needed for e_fm and thumbnailing. Attached is a file that implements the core dbus subsystem in E17. The next step is to expose register / unregister API to allow modules to register interfaces with the enlightenment window manager core RemoteObject. Currently any feedback and opinions will be helpful. Current API: !DOCTYPE node PUBLIC -//freedesktop//DTD D-BUS Object Introspection 1.0//EN http://www.freedesktop.org/standards/dbus/1.0/introspect.dtd; node name=/org/enlightenment/wm/RemoteObject interface name=org.freedesktop.DBus.Introspectable method name=Introspect arg type=s direction=out/ /method /interface interface name=org.enlightenment.wm.Core method name=Restart /method method name=Shutdown /method /interface interface name=org.enlightenment.wm.Module method name=Load arg type=s direction=in/ /method method name=Unload arg type=s direction=in/ /method method name=Enable arg type=s direction=in/ /method method name=Disable arg type=s direction=in/ /method method name=List arg type=a(si) direction=out/ /method /interface /node -Stafford If the work is done, then we have to at least check it out.:) Tying ourselves to dbus thoeh. Granted, it would probably get us out the door faster, but deps++(On a side note, can we get something like applications.menu which E can 'make install' itself? This way we're not bound to any package...) dh - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] E_DBus - IPC replacement
On Sun, 22 Jul 2007 00:04:44 -0400 Christopher Michael [EMAIL PROTECTED] babbled: Stafford Horne wrote: Hello Developers, Recently there has been a move towards modularizing the config dialogs within E17. With the development in this area raster and I were thinking it would be a nice time to modularize the IPC code. That is, the enlightenment binary should only contain a minimal amount of IPC functionality (Module, Restart, Profiles) Perhaps some sort of Session Management here ? Incorporate w/ remembers startup... that's up to modules being written to advertise dbus control over anything anyone wants. the idea is to remove it from core e and put it into modules and transition to dbus for the ipc. we need to keep as little as possible in core e. i would say session.remember stuff is a module thing. while the rest should be extended via modules. The problem with the current IPC framework is that the handlers are not dynamically appendable. This could be fixed. However, I believe that dbus should be the way to advertise out IPC as it seems like the standard way to do things now (look at, pidgin, gnome, rhythmbox etc. if you will). But are we gtk? Can we not come up w/ something better? Yes a dbus layer may/will be needed to integrate w/ other apps...But really, should we rely on dbus for socket comms.?, or do something ourselves. Not totally against the idea, but dbus is..:( the thing is - we don't need much better. we do need to rationalise the existing ipc - dbus is as good as anything for this purpose (these days). imho we will end up with minimal dbus ipc anyway - because no one will bother writing all the ipc modules ad frankly- i don't see there being enough need for it for people to justify the effort. The final goal is to remove the e_ipc_handlers* files completely. e_ipc is still needed for e_fm and thumbnailing. yup. e_ipc is still all for internal stuff - like fm, thumbnailer etc. might move to epsilon for thumbnailing eventually too - but it does its own ipc anyway. Attached is a file that implements the core dbus subsystem in E17. The next step is to expose register / unregister API to allow modules to register interfaces with the enlightenment window manager core RemoteObject. Currently any feedback and opinions will be helpful. Current API: !DOCTYPE node PUBLIC -//freedesktop//DTD D-BUS Object Introspection 1.0//EN http://www.freedesktop.org/standards/dbus/1.0/introspect.dtd; node name=/org/enlightenment/wm/RemoteObject interface name=org.freedesktop.DBus.Introspectable method name=Introspect arg type=s direction=out/ /method /interface interface name=org.enlightenment.wm.Core method name=Restart /method method name=Shutdown /method /interface interface name=org.enlightenment.wm.Module method name=Load arg type=s direction=in/ /method method name=Unload arg type=s direction=in/ /method method name=Enable arg type=s direction=in/ /method method name=Disable arg type=s direction=in/ /method method name=List arg type=a(si) direction=out/ /method /interface /node -Stafford If the work is done, then we have to at least check it out.:) Tying ourselves to dbus thoeh. Granted, it would probably get us out the door faster, but deps++(On a side note, can we get something like applications.menu which E can 'make install' itself? This way we're not bound to any package...) and what will you put in applications.menu? as we don't provide the apps - how can we provide an applications.menu with any meaningful content? dh - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ 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) - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net
Re: [E-devel] E_DBus - IPC replacement
Carsten Haitzler (The Rasterman) wrote: On Sun, 22 Jul 2007 00:04:44 -0400 Christopher Michael [EMAIL PROTECTED] babbled: Stafford Horne wrote: Hello Developers, Recently there has been a move towards modularizing the config dialogs within E17. With the development in this area raster and I were thinking it would be a nice time to modularize the IPC code. That is, the enlightenment binary should only contain a minimal amount of IPC functionality (Module, Restart, Profiles) Perhaps some sort of Session Management here ? Incorporate w/ remembers startup... that's up to modules being written to advertise dbus control over anything anyone wants. the idea is to remove it from core e and put it into modules and transition to dbus for the ipc. we need to keep as little as possible in core e. i would say session.remember stuff is a module thing. Yea, certainly modular for session stuff... while the rest should be extended via modules. The problem with the current IPC framework is that the handlers are not dynamically appendable. This could be fixed. However, I believe that dbus should be the way to advertise out IPC as it seems like the standard way to do things now (look at, pidgin, gnome, rhythmbox etc. if you will). But are we gtk? Can we not come up w/ something better? Yes a dbus layer may/will be needed to integrate w/ other apps...But really, should we rely on dbus for socket comms.?, or do something ourselves. Not totally against the idea, but dbus is..:( the thing is - we don't need much better. we do need to rationalise the existing ipc - dbus is as good as anything for this purpose (these days). imho we will end up with minimal dbus ipc anyway - because no one will bother writing all the ipc modules ad frankly- i don't see there being enough need for it for people to justify the effort. Ok, no worries. Was just thinking about deps... The final goal is to remove the e_ipc_handlers* files completely. e_ipc is still needed for e_fm and thumbnailing. yup. e_ipc is still all for internal stuff - like fm, thumbnailer etc. might move to epsilon for thumbnailing eventually too - but it does its own ipc anyway. Attached is a file that implements the core dbus subsystem in E17. The next step is to expose register / unregister API to allow modules to register interfaces with the enlightenment window manager core RemoteObject. Currently any feedback and opinions will be helpful. Current API: !DOCTYPE node PUBLIC -//freedesktop//DTD D-BUS Object Introspection 1.0//EN http://www.freedesktop.org/standards/dbus/1.0/introspect.dtd; node name=/org/enlightenment/wm/RemoteObject interface name=org.freedesktop.DBus.Introspectable method name=Introspect arg type=s direction=out/ /method /interface interface name=org.enlightenment.wm.Core method name=Restart /method method name=Shutdown /method /interface interface name=org.enlightenment.wm.Module method name=Load arg type=s direction=in/ /method method name=Unload arg type=s direction=in/ /method method name=Enable arg type=s direction=in/ /method method name=Disable arg type=s direction=in/ /method method name=List arg type=a(si) direction=out/ /method /interface /node -Stafford If the work is done, then we have to at least check it out.:) Tying ourselves to dbus thoeh. Granted, it would probably get us out the door faster, but deps++(On a side note, can we get something like applications.menu which E can 'make install' itself? This way we're not bound to any package...) and what will you put in applications.menu? as we don't provide the apps - how can we provide an applications.menu with any meaningful content? Well, we used to provide a .order file that had some standard things...xterm, xmms, etc, etc...my thoughts were along those lines. dh - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
