terminal bypassing optimisations. Previously if curses thought curscr
was in sync with virtscr (curses concept of what is on the screen) then
nothing would be output. This change forces an update out to the terminal
regardless.
parameter that controls whether or not certain characters in the
string are interpreted or not (things like tab being expanded).
Make __waddbytes a wrapper for _cursesi_waddbytes that passes all
parameters and sets the flag for character interpretation for backward
compatibility.
Fix an incipient bug in _cursesi_waddbytes where garbage would have
been written to the terminal if the terminal TABSIZE was set > 8 and
character interpretation is on.
Convert all internal __waddbytes calls to use _cursesi_waddbytes, fix
the function prototypes and add a new flag that will be used later.
Fix the addchstr family functions so that they call _cursesi_waddbytes
with character interpretation off as per SUSV2.
The previous version of this file changed a terminal initialisation test on
the exit_attribute_mode capability, checking for the exit_alt_charset_mode
capability as a substring, rather than performing a search for the hard-coded
^O character.
That works better on terminals where ^O is not the correct value for
exit_alt_charset_mode. But it works worse on terminals that don't have a
definition specified for exit_alt_charset_mode.
For example:
% TERMCAP='xterm:me=\E[m:' TERM=xterm vi
segmentation fault (core dumped) TERMCAP='xterm:me=\E[m:' TERM=xterm vi
The crash can be avoided (without fixing the bug) by defining
exit_alt_charset_mode:
% TERMCAP='xterm|:me=\E[m:ae=:' TERM=xterm vi
ex/vi: Error: xterm: No such process
We now test exit_alt_charset_mode for NULL before continuing with the fatal
test, restoring the original no-crash behaviour.
XXX does_ctrl_o() is now just a naive reimplementation of strstr(), so should
probably just use strstr() instead.
Adjust various man pages and other documentation to point to capfile(5)
instead of termcap(5).
Remove getcap(3) as curses hasn't been building it for a long time.
Punt wrterm.c as tset no longer uses it.
sys/stdarg.h and expect compiler to provide proper builtins, defaulting
to the GCC interface. lint still has a special fallback.
Reduce abuse of _BSD_VA_LIST_ by defining __va_list by default and
derive va_list as required by standards.
