Re: [E-devel] terminal application based on the efl.
using freetype directly probably. Again, this is easier to do IMO. Doing that inside Evas will be bit harder I guess, maybe not because engine stuff remains unchanged you know better than me, does it worth to have such restricted-use object built-in in Evas? much easier to do inside evas. trust me. much easier :) and more efficient. Yes, for current evas - without a doubt more efficient. The 'problem' with this is that it's really 'covering up' a more general issue that evas needs to address - make it easier/possible to do efficient 'custom' rendering or define custom 'self-rendered' objs. Right now there's really only one way, software rendering to an argb image. We can extend this to a similar buffered solution which is more engine specific via things like native surfaces (which is useful in its own right), but still an 'indirect' solution. well smart object with a grid of NxM text objects... but thats just nasty overhead... :) Yes, a more 'direct' rendering solution is likely preferable here... We know evas needs to be able to support more direct custom rendering/objs, the question is simply how to go about doing that.. because there are several ways in which one could realize it (I'll refrain from going into those here). yeah - absolutely. need to be able to create 1st-class evas objects by providing a paint method basically. but as such this is exactly what is done inside of evas anyway - just there is no external access to that api, that's all. you'll end up implementing the same thing. the only question will be where does that implementation live :) ... And what kind of 'api', if any, should be provided for custom rendering - eg. an api of some sort for immediate-mode to an 'abstract' update surface? or per-engine update surfaces you draw to as fits that engine? or somehow use the canvas objs to do immediate-mode rendering (change state and set a 'draw')? or others... But if we're talking about whether evas should extend its current api to include something like a 'multi-line text obj' or something of that sort... well, that depends. Does one want to keep adding more and more specialized stuff right now, though it would be useful/efficient for this or that thing one wants now.. or take the time to fix the 'real' underlying issues the lib must invariably face long-term? of course not - but there is a limit. i think evas has very much not suffered from much of this and has remained very clean and simple - a textgrid object does make a lot of sense. a buffer u render to and keep is a little silly as u keep argb pixels around for stuff u can generate on the fly pretty fast anyway from glyphs. :) Doing it by drawing to an image buffer would be, generally, less efficient.. and one could agree that a 'textgrid' obj is fairly basic. But these kinds of things have a way of snow-balling.. I mean, first text objs, then it was argued textblock objs, now it's yet another text related obj.. and there arguments for all sorts of 'extras' one might want along with that, and ... What would be 'nice', and is doable now, would be the basics of custom objs. ie. separately declared objs (ie. their apis) which the lib could load.. even if their implementation would be much like what is done now and would still depend on knowing evas internals. It'd be a step towards a more complete solution and gives the option to later move such objs' implementation (if desired) to a true 'custom rendering' one. It would allow for such things as a 'cairo' obj (which could draw to native update surfaces in some cases), for your text-grid, for funky text, for... and doesn't require having such objs as part of the 'evas' api or as a dependency on such, nor deciding on some abstract api or exported engine funcs. p.s. still have transforms patch on my list. just been dragged off into fixing some bugs and dealing with paid stuff for elementary. end of the month (start of next) definitely will get a bit of time i think! (just hesitant as this does break a few engines and that needs fixing) Transforms? I thought we were into stroked/filled, textured stuff already... :) As far as engine breaks.. well man, we knew that was inevitable - one way or another. Problem is, there are so *many* engines people have been adding, with a very primitive interface - we both knew this. It's going to take everyone pitching in - or just forget about extending much of the gfx capabilities in any real efficient way. :( Click to find information on your credit score and your credit report. http://thirdpartyoffers.juno.com/TGL2141/fc/Ioyw6i3m2PcNwkpvcNdklpPRqCveOTlDZ5RGERkCyZzAVpVGtkOJMA/ - This SF.Net email is sponsored by the Moblin Your Move
Re: [E-devel] [Evil, svn trunk] autogen.sh failed on WinXP
On Fri, 17 Oct 2008, Lionel ORRY wrote: The dirtiest part of the patch is declaring eina_mempool_register and eina_mempool_unregister with EAPI, because it helps passing the link stage. I don't know much about the differences between .dll and .so (there must me many), but gcc was not happy with these functions prototypes. I don't know exactly how to correct that, it's just a workaround. EAPI needs to be added in front of these mempool functions. I've also fixed the problem with ffs(). I have added an implementation for everything except cegcc (which already implements it). After the addition of EAPI, i think that eina will compile correctly with the latest Evil. Vincent - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] terminal application based on the efl.
I wrote: What would be 'nice', and is doable now, would be the basics of custom objs. ie. separately declared objs (ie. their apis) which the lib could load.. even if their implementation would be much like what is done now and would still depend on knowing evas internals. It'd be a step towards a more complete solution and gives the option to later move such objs' implementation (if desired) to a true 'custom rendering' one. It would allow for such things as a 'cairo' obj (which could draw to native update surfaces in some cases), for your text-grid, for funky text, for... and doesn't require having such objs as part of the 'evas' api or as a dependency on such, nor deciding on some abstract api or exported engine funcs. Just to give you a quick idea of what I mean by these evas object modules. First of all, we add ONE new engine function to get certain funcs for a named obj type - this returns a void* which holds all the 'engine funcs' that a given obj type will use. The lib will look somewhere for such named obj modules and get this void* from the obj's 'engine-level' implementation (may need one per engine backend). Then, say you want your text-grid obj with whatever its api. That's declared in its own header file you need to include if you want to use this type of object. In the canvas-level implementation, say you have an Evas_Object_Text_Grid struct which will contain its cur/prev states, etc.. but now also add a pointer to say an Evas_Object_Text_Grid_Funcs, which are the 'engine funcs' for this obj type. This will get assigned when you create an instance of the object by getting it from the above mentioned new evas engine func. Then in the actual implementation of the obj's api funcs you basically do what is done now (you have acess to evas internals - these are to be seen as direct evas obect modules for now), and when you need to use an 'engine func' specific to the object, well that's what you have in that Evas_Object_Text_Grid_Funcs pointer. So, we're just moving the new obj's api into its own header and have the lib 'load' the object's engine funcs (if any) which the canvas-level implementation of the object's api will use as it needs. Take a break - you deserve it. Click here to find a great vacation. http://thirdpartyoffers.juno.com/TGL2141/fc/Ioyw6i3nJgyAfHyvgCGRTUkMRokrAvNTzztt6rfJAlaE3CjmPQi2uw/ - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] ecore: ecore_con_server_add() fails to unlink stale socket
On Sat, Oct 18, 2008 at 2:02 PM, Shulga Nikita [EMAIL PROTECTED] wrote:
Hi all,
There is logic glitch in bind() error handling logic in
ecore_con_server_add() implemented in ecore/src/lib/ecore_con/
ecore_con.c ( revision 36139 ), that results in stale unix socket not
being unlinked.
Please take a look on following conditional:
if ((type == ECORE_CON_LOCAL_USER) || (type ==
ECORE_CON_LOCAL_SYSTEM)
(connect(svr-fd, (struct sockaddr *)socket_unix,
socket_unix_len) 0)
(unlink(buf) = 0))
goto start;
If type argument is equal to ECORE_CON_LOCAL_USER (i.e. first
condition is true ) than rest conditions will not be evaluated, i.e.
connect() and unlink() will not be called for socket in question. To
ensure that connect() and unlink() will be called if type either
equal to ECORE_CON_LOCAL_USER or ECORE_CON_LOCAL_SYSTEM one should put
parentheses around type checks. That is:
===
--- src/lib/ecore_con/ecore_con.c (revision 36772)
+++ src/lib/ecore_con/ecore_con.c (working copy)
@@ -278,7 +278,7 @@
}
if (bind(svr-fd, (struct sockaddr *)socket_unix, socket_unix_len)
0)
{
- if ((type == ECORE_CON_LOCAL_USER) || (type ==
ECORE_CON_LOCAL_SYSTEM)
+ if (((type == ECORE_CON_LOCAL_USER) || (type ==
ECORE_CON_LOCAL_SYSTEM))
(connect(svr-fd, (struct sockaddr *)socket_unix,
socket_unix_len) 0)
(unlink(buf) = 0))
goto start;
In SVN, thanks for catching it !
--
Cedric BAIL
-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100url=/
___
enlightenment-devel mailing list
enlightenment-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] [Fwd: Entrance problems]
On Sun, Oct 19, 2008 at 6:11 PM, Thanatermesis [EMAIL PROTECTED] wrote: Hello Gustavo, I would like to tell you that there's a bug in entrance that does a black screen in the big percentage of computers with a X for mouse, after to see the commits and go back to the previous version of src/daemon/entranced_display.c , everything looks like to work perfectly hum... could you point me which commit? And it would be better to mail enlightenment-devel for such (in CC now) For now i will use this old version for my own usage (for the Elive package), but i should like that we can discuss about this in order to solve correctly the problem since your commits looks like to be for compatibility in hard-loaded or old systems, but by other side made an uncompatibility in others Do you actually maintain entrance ? i know that it doesn't has much development ;) No, I just fixed that problem. PS: your email [EMAIL PROTECTED] is wrong, please update it on the SVN info it's on purpose, it should omit my mail from web listings, that's how it worked some time ago at least :-) -- Gustavo Sverzut Barbieri http://profusion.mobi embedded systems -- MSN: [EMAIL PROTECTED] Skype: gsbarbieri Mobile: +55 (19) 9225-2202 - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
[E-devel] python-etk lists broken in rev 36540 (FSO testing on Openmoko)
Hi, I just land here because I'm a Openmoko Freerunner owner (1 1/2 month) and a python developper. The FSO (freesmartphone.org) last upgrade and last image (18/10/2008) bring me revision 36540 (svnr36540) of Enlightenment. Now, python-etk lists no more appear, even with sample 15-list.py of python-etk-samples. C example etk_test works as expected. I don't know the origin of the problem and don't know what to do. Do I have to file a ticket on FSO tracker or on Enlightenment tracker ? Thanks. Yves Mahe PS : from where come this svn revision code ? Current revision code of Enlightenment's svn seems to be slightly greater (36770) than 36540 ( http://downloads.freesmartphone.org/fso-testing/feeds/armv4t/python-etk_0.3.0+svnr36540-r0.1_armv4t.ipk ) - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] [Fwd: Entrance problems]
The exact commit is: http://trac.enlightenment.org/e/changeset?new=36417%40trunk%2Fentranceold=36364%40trunk%2Fentrance Thanatermesis 2008/10/20 Gustavo Sverzut Barbieri [EMAIL PROTECTED] On Sun, Oct 19, 2008 at 6:11 PM, Thanatermesis [EMAIL PROTECTED] wrote: Hello Gustavo, I would like to tell you that there's a bug in entrance that does a black screen in the big percentage of computers with a X for mouse, after to see the commits and go back to the previous version of src/daemon/entranced_display.c , everything looks like to work perfectly hum... could you point me which commit? And it would be better to mail enlightenment-devel for such (in CC now) For now i will use this old version for my own usage (for the Elive package), but i should like that we can discuss about this in order to solve correctly the problem since your commits looks like to be for compatibility in hard-loaded or old systems, but by other side made an uncompatibility in others Do you actually maintain entrance ? i know that it doesn't has much development ;) No, I just fixed that problem. PS: your email [EMAIL PROTECTED] is wrong, please update it on the SVN info it's on purpose, it should omit my mail from web listings, that's how it worked some time ago at least :-) -- Gustavo Sverzut Barbieri http://profusion.mobi embedded systems -- MSN: [EMAIL PROTECTED] Skype: gsbarbieri Mobile: +55 (19) 9225-2202 - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] python-etk lists broken in rev 36540 (FSO testing on Openmoko)
I got your last mail as well, just need some time to work on it. but maybe other guys can take a look before me. On Mon, Oct 20, 2008 at 11:04 AM, yves mahe [EMAIL PROTECTED] wrote: Hi, I just land here because I'm a Openmoko Freerunner owner (1 1/2 month) and a python developper. The FSO (freesmartphone.org) last upgrade and last image (18/10/2008) bring me revision 36540 (svnr36540) of Enlightenment. Now, python-etk lists no more appear, even with sample 15-list.py of python-etk-samples. C example etk_test works as expected. I don't know the origin of the problem and don't know what to do. Do I have to file a ticket on FSO tracker or on Enlightenment tracker ? Thanks. Yves Mahe PS : from where come this svn revision code ? Current revision code of Enlightenment's svn seems to be slightly greater (36770) than 36540 ( http://downloads.freesmartphone.org/fso-testing/feeds/armv4t/python-etk_0.3.0+svnr36540-r0.1_armv4t.ipk ) - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel -- Gustavo Sverzut Barbieri http://profusion.mobi embedded systems -- MSN: [EMAIL PROTECTED] Skype: gsbarbieri Mobile: +55 (19) 9225-2202 - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
[E-devel] Environnement de bureau - Wikipédia / Desktop environment
Hi all, shall E deserve a word in these pages : http://fr.wikipedia.org/wiki/Environnement_de_bureau http://en.wikipedia.org/wiki/Desktop_environment Cheers, Michel - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
[E-devel] Found a small issue when compiling from today's SVN
Hi, I'd like to post a problem ( and solution ) that I found while trying to update my e17 installation today on Ubuntu. I use the easy_e17.sh script ( of course! ), and while embryo was trying to compile, make kept complaining about not being able to find eina_types.h. I fixed this by changing the following: in embryo_private.h, line 33: #include eina_types.h to #include eina-0/eina_types.h I then changed embryo_cc_sc.h, line 39: #include eina_types.h to #include eina-0/eina_types.h I apologize if my message is lengthy, but I thought you all would want to know. Thanks, Mike -- - Mike McNeil Sr. Network Engineer Information Systems and Technology University of California Berkeley - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] E SVN: urandom trunk/E-MODULES-EXTRA/forecasts
On Mon, Oct 20, 2008 at 7:28 PM, Enlightenment SVN
[EMAIL PROTECTED] wrote:
Log:
unswallow the icons, or there will be visible duplicates, even if the gadman
window is hidden.
+if (swallow)
+ {
+ edje_object_part_unswallow(inst-forecasts-forecasts_obj,
swallow);
+ evas_object_del(swallow);
+ }
you don't need to unswallow if you del, that's automatic.
--
Gustavo Sverzut Barbieri
http://profusion.mobi embedded systems
--
MSN: [EMAIL PROTECTED]
Skype: gsbarbieri
Mobile: +55 (19) 9225-2202
-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100url=/
___
enlightenment-devel mailing list
enlightenment-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
