bool means something to C++
From: Chris Torek <torek@BSDI.COM>

typo, wrong parameter to __mvcur in endwin()

handle tgoto functions that return NULL, not "OOPS"

use 8K buffering for stdout -- some C libraries use 1K

the cursor must be at the screen bottom for SF/sf

don't leave the screen in stadnout mode
bug fix taken from FreeBSD sources, fixes Sun console bug for nvi

don't reenter standout mode on every character
(test for already in standout mode was wrong)

wstandend/wstandout return an int (that's what System V does)

simplify the code, return an int (that's what System V does)

rework scrolling to us AL/DL if scrolling most of the screen, in
preference to sf/sr

I got the sense of the scroll test backwards....

__tscroll has to take a second integer argument for CS string

add support for CS SF/sf/SR/sr combinations as scrolling commands

add support for CS SF/sf/SR/sr combinations as scrolling commands

delete __FULLLINE, it's never used

delete __FULLLINE, it's never used
__tscroll has to take a second integer argument for CS string

page breaks for the 4.4BSD manuals

fix running head

syntax fixes

add ioctl.h back in, TIOCGWINSZ on old systems.

fix return values for tty function

upgrade some includes, make curses.h local
helps out nvi, but it's cleaner regardless

add %p, so we can load against System V terminfo files
lots of random cleanup

It seems to me that addbytes() (and thus addch()) should return ERR if
you try to go past the end of the window and SCROLLOK is not set. The
old code also used to do this; the current code does not.

It seems to me that addbytes() (and thus addch()) should return ERR if
you try to go past the end of the window and SCROLLOK is not set. The
old code also used to do this; the current code does not.
This breaks at least one program, mille(6).
From: Charles Hannum <mycroft@sun-lamp.cs.berkeley.edu>

show more ...

don't need termios.h any longer