On Sat, May 26, 2012 at 4:16 AM, Bluezery <[email protected]> wrote:
>
> 2012/5/26 Vincent Torri <[email protected]>:
> > On Sat, May 26, 2012 at 4:03 AM, Bluezery <[email protected]> wrote:
> >> Hum.. I also do not want to guarantee that code :)
> >> I only want to discussed about int/shutdown. These can be called &
> >> mixed because we can use many libraries (EFL or not) which use
> >> libraries (EFL or not) which use libraries (EFL or not), and so on....
> >
> > i'm wondering if you have read what i previously said. Look at
> > elm_init() :
> >
> > elm_init calls (i give only some of them) :
> > eina_init
> > ecore_init
> > ecore_file_init
> > evas_init
> > edje_init
> > ecore_con_init
> > ecore_evas_init
> >
> > ecore_evas_init calls
> > evas_init
> > ecore_init
> >
> > and ecore_evas_sdl_new calls SDL_Init
> >
> > ecore_init calls
> > eina_init
> >
> > evas_init calls
> > eina_init
> > eet_init
> >
> > eet_init call
> > eina_init
> > the initialisation function of libgcrypt
> >
> > look also edje_init, ecore_con_init, etc.. and you'll see that they
> > also call ecore_init and eina_init and other ones.
> >
> > so what you described above is exactly what we are doing. Is there a
> > problem ? No. Why ? Because we use correctly the init functions of the
> > EFL and of other libs
> >
> > Vincent
>
> Yes, I have read :)
> Anyway, We have different point of view. I do not want to point out
> bugs or problems in EFL. I just want to say about something like
> conceptual points.
> I agreed with that EFL developers use correctly and also EFL libraries
> do. But I just say about OTHERS (MANY good & bad developers in the
> world).
> If the EFL became more and more prosperous, I cannot predict how
> others use EFL. Even though I can use correctly, I cannot assert
> underlying libraries. (not of the EFL, but these libraries can use EFL
> somewhere). I want to find out places where we can improve without
> performance cost, conformable to general way.
> This is just my humble suggestion. I'm sorry to disturb you. I would
> better to end these discussion. :)
Ouch, this got a bit out of hand.
Bluezery confirm your suggestion so far is:
- in every _shutdown() function, check if counter is already 0 and
call EINA_LOG_CRITICAL("module MODNAME already shutdown") and do not
make counter negative (<0). [maybe not eina_log_critical() as maybe
eina is already down? just fputs("CRITICAL: MODNAME already
shutdown\n", stderr);]
If it is, please do so as there is no problem with that and should
make everyone happy.
changes to init and other stuff will break, and people will complain.
But this simple check should be all fine.
--
Gustavo Sverzut Barbieri
http://profusion.mobi embedded systems
--------------------------------------
MSN: [email protected]
Skype: gsbarbieri
Mobile: +55 (19) 9225-2202
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
enlightenment-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
