[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#16930: [ac02f3a7] fails to build without HAVE_WINDOW_SYSTEM
From: |
Eli Zaretskii |
Subject: |
bug#16930: [ac02f3a7] fails to build without HAVE_WINDOW_SYSTEM |
Date: |
Mon, 03 Mar 2014 22:02:34 +0200 |
> From: Ivan Shmakov <ivan@siamics.net>
> Date: Mon, 03 Mar 2014 19:29:39 +0000
>
> ../../src/font.c: In function 'Fframe_font_cache':
> ../../src/font.c:4853:3: error: implicit declaration of function
> 'FRAME_DISPLAY_INFO' [-Werror=implicit-function-declaration]
> return FRAME_DISPLAY_INFO (decode_live_frame (frame))->name_list_element;
> ^
> ../../src/font.c:4853:56: error: invalid type argument of '->' (have 'int')
> return FRAME_DISPLAY_INFO (decode_live_frame (frame))->name_list_element;
> ^
> ../../src/font.c:4854:1: error: control reaches end of non-void function
> [-Werror=return-type]
> }
> ^
> cc1: all warnings being treated as errors
> make[1]: *** [font.o] Error 1
Should be fixed now (trunk revision 116646).
> As all the other invocations of FRAME_DISPLAY_INFO () seem to be
> guarded with #ifdef HAVE_WINDOW_SYSTEM, I’d rather assume that
> the whole DEFUN ("frame-font-cache", …) should also be within
> such guards, along with the reference to Sframe_font_cache
> within syms_of_font ().
No, it's not enough to #ifdef away that function when
HAVE_WINDOW_SYSTEM is not defined, because if HAVE_WINDOW_SYSTEM _is_
defined, you can still call the function for a TTY frame.
In general, one has always think about multi-tty when handling such
issues: a build that supports window-system can have frames that
don't. It's no longer a compile-time issue.