On Thu, 30 Oct 2003 01:11:52 +0100 Hallvar Helleseth <[EMAIL PROTECTED]>
babbled:
> On Wed, 2003-10-22 at 00:54, Carsten Haitzler wrote:
> > On Tue, 21 Oct 2003 18:15:02 +0200 Hallvar Helleseth <[EMAIL PROTECTED]>
> > babbled:
> >
> > > Hi!
> > >
> > > Ive been fooling a bit around, adding ecore_directfb and
> > > ecore_evas_directfb (mostly just copied from fb modules)
> > >
> > > Ive been trying hard to be able to use edje with directfb.. Ive only
> > > gotten to the part where I can use the "edje" program to display it self
> > > in a DirectFB surface/window. Though I find it really odd, that the only
> > > graphics I see is the test programs (the part selection and the border
> > > window of the edje) not the edje itself.. I modified edje_main.c to
> > > display "e_logo.eet test" by default (without having to click in the
> > > selection) It works well with both x11 and fb, but not directfb...
> > >
> > > Is there anything obvios I might have overlooked that you can tell
> > > without having to look at the code I did..?
> > >
> > > I started out by basically just rename fb to dfb.. then added code to
> > > open DirectFB and create a window, then modifed evas to use directfb
> > > engine instead of fb engine.
> > >
> > > A note on evas too. I noticed the text objects dont work.. I tried
> > > modifing code to use DirectFBs text natively instead of evases common
> > > font software code. I got it to work to some extend though not
> > > satisfactory (skipped difficult parts like querying positions of
> > > characters and such) Is there a reason why the evas software code was
> > > chosen over DirectFBs? I guess text drawing cant really be accelerated,
> > > but still..
> >
> > it's easier to get an accurate port. ie text is 1:1 the same for every
> > character, glyph, pixel, spacing, kerning etc. also you get the character
> > querying for free. :)
> Does your text in DirectFB-evas show up correct? Doesnt here, tried
> differnt stuff with the code but just wont work....
last time i looked (like 6 months ago) it did! :)
> > > My interest is to be able to run edjes in DirectFB surfaces, including
> > > all the input stuff, which means a bit of work on a directfb ecore
> > > module too as I understand.. Im ready to give it a go, but Ive been
> > > using a bit too much time trying to understand why the "edje" program
> > > only shows its own graphics and not the edje...
> > >
> > > sorry for the long email! and the poor english. But Ill give back fully
> > > functional edje on DirectFB :) atleast Ill give it my best try!
> >
> > ok. check timers in ecore work. check ilders, and idle enterers work. last
> > time i played with directfb the real problem was dfb didn't have select()
> > support. this means i couldn't sit in a select and wait on fd's and use it
> > as a timer system. i'd need to have dfb's event call wait until a timeout
> > and have a separate thread do select() and then post a fake "wake up" event
> > to the dfb event queue to wake up the dfb "next event" call. this was a bit
> > of a PITA.
> >
> > ecore_evas uses idle handlers to render with evas. edje uses ecore_evas's
> > pre_render and post_render calls to freeze then thaw the edje - which
> > basically means the edje wont update change or be recalculated as long as
> > its frozen. it could be this? (which would lead back to idle enterers not
> > working and thus timers and idlers too). check on this. you also need to
> > build the eet image loader in evas - maybe you didn't build it? try these
> > things... maybe it'll get you up and running.
>
> The problem was that DirectFB-evas couldnt load eet images.. I added
> support for doing that in DirectFB-evas. Or do you think it would be a
> better idea to add an eet-imageprovider to DFB? (sort of difficult with
> the keys although I could try to persuade the DFB developers to let me
> add support for spescifing keys)
aaah yeah. dfb's loaders... yup yup. forgot.. :) for now i'd just let evas do
the loading and just dump the pixel data to a dfb surface :)
> oh and it be REALLY helpfull if evas was more verbose on errors such as
> not beeing able to load an image of a certain format.. or some other
> sort of error handling (or am I missing something as always..?) maybe
> edje should be the one telling about the error..
yeah. it does have error checking. but its pretty damn basic :) i need to
formalize the whole loader api at some point - thats when i'll fix all of that
up :)
> Ive gotten the whole edje test program to function with input and all,
> although at the moment the code is not very clean. Ill clean up quite a
> bit and send a patch :)
>
> But before that Id like to know from you if my solution to DFBs lack of
> select is approriate:
>
> - Create a unix socket (/tmp/.ecore_dfb ?), create a thread.
>
> - in thread: connect to socket, while(wait for even) { send a byte to
> socket; wait for socket to return a byte }
>
> - the other end of the socket is connected to ecore.
>
> - in the fd/event handler: get event, process event, send a byte to
> socket (the thread receives it)
>
> comments?
hmm yup. works. definitely. but its synchronous lock-step. dont think i like
that. heres' what i'd do. i would have the "event" thread wait for the next
event and when it happens read the event info and write a data struct down the
socket to the main thread and then have the main thread read the data form the
socket and use that. now you get pipelining and buffering and no need to
lock-step back and forth PER event. :) that'd be the way imho to do it :)
> by god edje is a nice peice of work dude! im luuuvin it, dadadadadah
thanks :):):)
> Hallvar Helleseth
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: SF.net Giveback Program.
> Does SourceForge.net help you be more productive? Does it
> help you create better code? SHARE THE LOVE, and help us help
> YOU! Click Here: http://sourceforge.net/donate/
> _______________________________________________
> enlightenment-devel mailing list
> [EMAIL PROTECTED]
> https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
--
--------------- Codito, ergo sum - "I code, therefore I am" --------------------
The Rasterman (Carsten Haitzler) [EMAIL PROTECTED]
熊耳 - 車君 [EMAIL PROTECTED]
Mobile Phone: +61 (0)413 451 899 Home Phone: 02 9698 8615
-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/
_______________________________________________
enlightenment-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
