On Sat, 29 Dec 2007, Christopher Michael wrote:

> Hisham Mardam Bey wrote:
>>  ~/code/e17/libs/ecore/src/lib/ecore_x
>>
>> .oOo. grep prefetch *.[ch] | grep xy
>>
>> ecore_xcb_window.c:ecore_x_pointer_xy_get_prefetch(Ecore_X_Window window)
>> Ecore_X.h:EAPI void ecore_x_pointer_xy_get_prefetch(Ecore_X_Window window);
>>
>> Which effectively places it in Ecore_X.h without having stubs for it
>> in the C files. People that try to use it will get compile time
>> errors:
>>
>> /tmp/ccwsbRaX.o: In function `_eg_grab_region_start':
>> egrab.c:(.text+0x426): undefined reference to 
>> `ecore_x_pointer_xy_get_prefetch'
>> collect2: ld returned 1 exit status
>>
>> This needs to be fixed, pronto.
>>
>
> WTH !!!
>
> Who broke ecore_x ??

me

> This also effects ecore_x_default_screen_get.
> These functions are not specific to ecore_xcb, but rather are generic
> ecore_x functions, so this needs to be fixed like NOW !!!!!!!!

that kind of functions are used only for xcb (well, for 
ecore_x_default_screen, i can add it for Xlib too, even if it is really 
needed for XCB as it is not stored at all, contrary to Xlib which stores 
the default screen in the Display structure), as mentioned in the 
documentation. So, except me (I don't think that someone else is using 
ecore with xcb), nobody should use those files.

I *know* that i need to add them for the Xlib backend.

BUT

ecore_xcb is not perfect. I've sent several mails about that and i am 
waiting for some people to answer them (that is, mainly, raster, Nathan 
and maybe rephorm, but he does not read the e-dev ML anymore, iirc. 
Other people can comment too, of course).

"fixing" all those files is VERY long. I don't want to do some work once, 
and see that I need to modify all that changes again (i've already written 
ecore_xcb at least 3 times).

So, one solution would be to disable ecore_xcb at all, and use some 
#define in Ecore_X.h around xcb-specific functions (or removing them)

comments ?

Vincent

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
enlightenment-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/enlightenment-devel

Reply via email to