Blame - doc/html/ncurses-intro.html - android_external_libncurses

blob: 451e7ab4d32dba74f0b72854dd9f9c9fb495476b [file] [log] [blame]

Amit Daniel Kachhap	e6a01f5	2011-07-20 11:45:59 +0530	[diff] [blame]	1	<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.0//EN">
				2	<!--
				3	$Id: ncurses-intro.html,v 1.43 2007/03/03 19:31:50 tom Exp $
				4	****************************************************************************
				5	* Copyright (c) 1998-2006,2007 Free Software Foundation, Inc. *
				6	* *
				7	* Permission is hereby granted, free of charge, to any person obtaining a *
				8	* copy of this software and associated documentation files (the *
				9	* "Software"), to deal in the Software without restriction, including *
				10	* without limitation the rights to use, copy, modify, merge, publish, *
				11	* distribute, distribute with modifications, sublicense, and/or sell *
				12	* copies of the Software, and to permit persons to whom the Software is *
				13	* furnished to do so, subject to the following conditions: *
				14	* *
				15	* The above copyright notice and this permission notice shall be included *
				16	* in all copies or substantial portions of the Software. *
				17	* *
				18	* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS *
				19	* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF *
				20	* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. *
				21	* IN NO EVENT SHALL THE ABOVE COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, *
				22	* DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR *
				23	* OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR *
				24	* THE USE OR OTHER DEALINGS IN THE SOFTWARE. *
				25	* *
				26	* Except as contained in this notice, the name(s) of the above copyright *
				27	* holders shall not be used in advertising or otherwise to promote the *
				28	* sale, use or other dealings in this Software without prior written *
				29	* authorization. *
				30	****************************************************************************
				31	-->
				32	<HTML>
				33	<HEAD>
				34	<TITLE>Writing Programs with NCURSES</TITLE>
				35	<link rev="made" href="mailto:bugs-ncurses@gnu.org">
				36	<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
				37	</HEAD>
				38	<BODY>
				39
				40	<H1>Writing Programs with NCURSES</H1>
				41
				42	<BLOCKQUOTE>
				43	by Eric S. Raymond and Zeyd M. Ben-Halim<BR>
				44	updates since release 1.9.9e by Thomas Dickey
				45	</BLOCKQUOTE>
				46
				47	<H1>Contents</H1>
				48	<UL>
				49	<LI><A HREF="#introduction">Introduction</A>
				50	<UL>
				51	<LI><A HREF="#history">A Brief History of Curses</A>
				52	<LI><A HREF="#scope">Scope of This Document</A>
				53	<LI><A HREF="#terminology">Terminology</A>
				54	</UL>
				55	<LI><A HREF="#curses">The Curses Library</A>
				56	<UL>
				57	<LI><A HREF="#overview">An Overview of Curses</A>
				58	<UL>
				59	<LI><A HREF="#compiling">Compiling Programs using Curses</A>
				60	<LI><A HREF="#updating">Updating the Screen</A>
				61	<LI><A HREF="#stdscr">Standard Windows and Function Naming Conventions</A>
				62	<LI><A HREF="#variables">Variables</A>
				63	</UL>
				64	<LI><A HREF="#using">Using the Library</A>
				65	<UL>
				66	<LI><A HREF="#starting">Starting up</A>
				67	<LI><A HREF="#output">Output</A>
				68	<LI><A HREF="#input">Input</A>
				69	<LI><A HREF="#formschars">Using Forms Characters</A>
				70	<LI><A HREF="#attributes">Character Attributes and Color</A>
				71	<LI><A HREF="#mouse">Mouse Interfacing</A>
				72	<LI><A HREF="#finishing">Finishing Up</A>
				73	</UL>
				74	<LI><A HREF="#functions">Function Descriptions</A>
				75	<UL>
				76	<LI><A HREF="#init">Initialization and Wrapup</A>
				77	<LI><A HREF="#flush">Causing Output to the Terminal</A>
				78	<LI><A HREF="#lowlevel">Low-Level Capability Access</A>
				79	<LI><A HREF="#debugging">Debugging</A>
				80	</UL>
				81	<LI><A HREF="#hints">Hints, Tips, and Tricks</A>
				82	<UL>
				83	<LI><A HREF="#caution">Some Notes of Caution</A>
				84	<LI><A HREF="#leaving">Temporarily Leaving ncurses Mode</A>
				85	<LI><A HREF="#xterm">Using <CODE>ncurses</CODE> under <CODE>xterm</CODE></A>
				86	<LI><A HREF="#screens">Handling Multiple Terminal Screens</A>
				87	<LI><A HREF="#testing">Testing for Terminal Capabilities</A>
				88	<LI><A HREF="#tuning">Tuning for Speed</A>
				89	<LI><A HREF="#special">Special Features of <CODE>ncurses</CODE></A>
				90	</UL>
				91	<LI><A HREF="#compat">Compatibility with Older Versions</A>
				92	<UL>
				93	<LI><A HREF="#refbug">Refresh of Overlapping Windows</A>
				94	<LI><A HREF="#backbug">Background Erase</A>
				95	</UL>
				96	<LI><A HREF="#xsifuncs">XSI Curses Conformance</A>
				97	</UL>
				98	<LI><A HREF="#panels">The Panels Library</A>
				99	<UL>
				100	<LI><A HREF="#pcompile">Compiling With the Panels Library</A>
				101	<LI><A HREF="#poverview">Overview of Panels</A>
				102	<LI><A HREF="#pstdscr">Panels, Input, and the Standard Screen</A>
				103	<LI><A HREF="#hiding">Hiding Panels</A>
				104	<LI><A HREF="#pmisc">Miscellaneous Other Facilities</A>
				105	</UL>
				106	<LI><A HREF="#menu">The Menu Library</A>
				107	<UL>
				108	<LI><A HREF="#mcompile">Compiling with the menu Library</A>
				109	<LI><A HREF="#moverview">Overview of Menus</A>
				110	<LI><A HREF="#mselect">Selecting items</A>
				111	<LI><A HREF="#mdisplay">Menu Display</A>
				112	<LI><A HREF="#mwindows">Menu Windows</A>
				113	<LI><A HREF="#minput">Processing Menu Input</A>
				114	<LI><A HREF="#mmisc">Miscellaneous Other Features</A>
				115	</UL>
				116	<LI><A HREF="#form">The Forms Library</A>
				117	<UL>
				118	<LI><A HREF="#fcompile">Compiling with the forms Library</A>
				119	<LI><A HREF="#foverview">Overview of Forms</A>
				120	<LI><A HREF="#fcreate">Creating and Freeing Fields and Forms</A>
				121	<LI><A HREF="#fattributes">Fetching and Changing Field Attributes</A>
				122	<UL>
				123	<LI><A HREF="#fsizes">Fetching Size and Location Data</A>
				124	<LI><A HREF="#flocation">Changing the Field Location</A>
				125	<LI><A HREF="#fjust">The Justification Attribute</A>
				126	<LI><A HREF="#fdispatts">Field Display Attributes</A>
				127	<LI><A HREF="#foptions">Field Option Bits</A>
				128	<LI><A HREF="#fstatus">Field Status</A>
				129	<LI><A HREF="#fuser">Field User Pointer</A>
				130	</UL>
				131	<LI><A HREF="#fdynamic">Variable-Sized Fields</A>
				132	<LI><A HREF="#fvalidation">Field Validation</A>
				133	<UL>
				134	<LI><A HREF="#ftype_alpha">TYPE_ALPHA</A>
				135	<LI><A HREF="#ftype_alnum">TYPE_ALNUM</A>
				136	<LI><A HREF="#ftype_enum">TYPE_ENUM</A>
				137	<LI><A HREF="#ftype_integer">TYPE_INTEGER</A>
				138	<LI><A HREF="#ftype_numeric">TYPE_NUMERIC</A>
				139	<LI><A HREF="#ftype_regexp">TYPE_REGEXP</A>
				140	</UL>
				141	<LI><A HREF="#fbuffer">Direct Field Buffer Manipulation</A>
				142	<LI><A HREF="#formattrs">Attributes of Forms</A>
				143	<LI><A HREF="#fdisplay">Control of Form Display</A>
				144	<LI><A HREF="#fdriver">Input Processing in the Forms Driver</A>
				145	<UL>
				146	<LI><A HREF="#fpage">Page Navigation Requests</A>
				147	<LI><A HREF="#ffield">Inter-Field Navigation Requests</A>
				148	<LI><A HREF="#fifield">Intra-Field Navigation Requests</A>
				149	<LI><A HREF="#fscroll">Scrolling Requests</A>
				150	<LI><A HREF="#fedit">Field Editing Requests</A>
				151	<LI><A HREF="#forder">Order Requests</A>
				152	<LI><A HREF="#fappcmds">Application Commands</A>
				153	</UL>
				154	<LI><A HREF="#fhooks">Field Change Hooks</A>
				155	<LI><A HREF="#ffocus">Field Change Commands</A>
				156	<LI><A HREF="#frmoptions">Form Options</A>
				157	<LI><A HREF="#fcustom">Custom Validation Types</A>
				158	<UL>
				159	<LI><A HREF="#flinktypes">Union Types</A>
				160	<LI><A HREF="#fnewtypes">New Field Types</A>
				161	<LI><A HREF="#fcheckargs">Validation Function Arguments</A>
				162	<LI><A HREF="#fcustorder">Order Functions For Custom Types</A>
				163	<LI><A HREF="#fcustprobs">Avoiding Problems</A>
				164	</UL>
				165	</UL>
				166	</UL>
				167
				168	<HR>
				169	<H1><A NAME="introduction">Introduction</A></H1>
				170
				171	This document is an introduction to programming with <CODE>curses</CODE>. It is
				172	not an exhaustive reference for the curses Application Programming Interface
				173	(API); that role is filled by the <CODE>curses</CODE> manual pages. Rather, it
				174	is intended to help C programmers ease into using the package. <P>
				175
				176	This document is aimed at C applications programmers not yet specifically
				177	familiar with ncurses. If you are already an experienced <CODE>curses</CODE>
				178	programmer, you should nevertheless read the sections on
				179	<A HREF="#mouse">Mouse Interfacing</A>, <A HREF="#debugging">Debugging</A>,
				180	<A HREF="#compat">Compatibility with Older Versions</A>,
				181	and <A HREF="#hints">Hints, Tips, and Tricks</A>. These will bring you up
				182	to speed on the special features and quirks of the <CODE>ncurses</CODE>
				183	implementation. If you are not so experienced, keep reading. <P>
				184
				185	The <CODE>curses</CODE> package is a subroutine library for
				186	terminal-independent screen-painting and input-event handling which
				187	presents a high level screen model to the programmer, hiding differences
				188	between terminal types and doing automatic optimization of output to change
				189	one screen full of text into another. <CODE>Curses</CODE> uses terminfo, which
				190	is a database format that can describe the capabilities of thousands of
				191	different terminals. <P>
				192
				193	The <CODE>curses</CODE> API may seem something of an archaism on UNIX desktops
				194	increasingly dominated by X, Motif, and Tcl/Tk. Nevertheless, UNIX still
				195	supports tty lines and X supports <EM>xterm(1)</EM>; the <CODE>curses</CODE>
				196	API has the advantage of (a) back-portability to character-cell terminals,
				197	and (b) simplicity. For an application that does not require bit-mapped
				198	graphics and multiple fonts, an interface implementation using <CODE>curses</CODE>
				199	will typically be a great deal simpler and less expensive than one using an
				200	X toolkit.
				201
				202	<H2><A NAME="history">A Brief History of Curses</A></H2>
				203
				204	Historically, the first ancestor of <CODE>curses</CODE> was the routines written to
				205	provide screen-handling for the game <CODE>rogue</CODE>; these used the
				206	already-existing <CODE>termcap</CODE> database facility for describing terminal
				207	capabilities. These routines were abstracted into a documented library and
				208	first released with the early BSD UNIX versions. <P>
				209
				210	System III UNIX from Bell Labs featured a rewritten and much-improved
				211	<CODE>curses</CODE> library. It introduced the terminfo format. Terminfo is based
				212	on Berkeley's termcap database, but contains a number of improvements and
				213	extensions. Parameterized capabilities strings were introduced, making it
				214	possible to describe multiple video attributes, and colors and to handle far
				215	more unusual terminals than possible with termcap. In the later AT&T
				216	System V releases, <CODE>curses</CODE> evolved to use more facilities and offer
				217	more capabilities, going far beyond BSD curses in power and flexibility.
				218
				219	<H2><A NAME="scope">Scope of This Document</A></H2>
				220
				221	This document describes <CODE>ncurses</CODE>, a free implementation of
				222	the System V <CODE>curses</CODE> API with some clearly marked extensions.
				223	It includes the following System V curses features:
				224	<UL>
				225	<LI>Support for multiple screen highlights (BSD curses could only
				226	handle one `standout' highlight, usually reverse-video).
				227	<LI>Support for line- and box-drawing using forms characters.
				228	<LI>Recognition of function keys on input.
				229	<LI>Color support.
				230	<LI>Support for pads (windows of larger than screen size on which the
				231	screen or a subwindow defines a viewport).
				232	</UL>
				233
				234	Also, this package makes use of the insert and delete line and character
				235	features of terminals so equipped, and determines how to optimally use these
				236	features with no help from the programmer. It allows arbitrary combinations of
				237	video attributes to be displayed, even on terminals that leave ``magic
				238	cookies'' on the screen to mark changes in attributes. <P>
				239
				240	The <CODE>ncurses</CODE> package can also capture and use event reports from a
				241	mouse in some environments (notably, xterm under the X window system). This
				242	document includes tips for using the mouse. <P>
				243
				244	The <CODE>ncurses</CODE> package was originated by Pavel Curtis. The original
				245	maintainer of this package is
				246	<A HREF="mailto:zmbenhal@netcom.com">Zeyd Ben-Halim</A>
				247	<zmbenhal@netcom.com>.
				248	<A HREF="mailto:esr@snark.thyrsus.com">Eric S. Raymond</A>
				249	<esr@snark.thyrsus.com>
				250	wrote many of the new features in versions after 1.8.1
				251	and wrote most of this introduction.
				252	Jürgen Pfeifer
				253	wrote all of the menu and forms code as well as the
				254	<A HREF="http://www.adahome.com">Ada95</A> binding.
				255	Ongoing work is being done by
				256	<A HREF="mailto:dickey@invisible-island.net">Thomas Dickey</A> (maintainer).
				257	Contact the current maintainers at
				258	<A HREF="mailto:bug-ncurses@gnu.org">bug-ncurses@gnu.org</A>.
				259	<P>
				260
				261	This document also describes the <A HREF="#panels">panels</A> extension library,
				262	similarly modeled on the SVr4 panels facility. This library allows you to
				263	associate backing store with each of a stack or deck of overlapping windows,
				264	and provides operations for moving windows around in the stack that change
				265	their visibility in the natural way (handling window overlaps). <P>
				266
				267	Finally, this document describes in detail the <A HREF="#menu">menus</A> and <A
				268	HREF="#form">forms</A> extension libraries, also cloned from System V,
				269	which support easy construction and sequences of menus and fill-in
				270	forms.
				271
				272
				273	<H2><A NAME="terminology">Terminology</A></H2>
				274
				275	In this document, the following terminology is used with reasonable
				276	consistency:
				277
				278	<DL>
				279	<DT> window
				280	<DD>
				281	A data structure describing a sub-rectangle of the screen (possibly the
				282	entire screen). You can write to a window as though it were a miniature
				283	screen, scrolling independently of other windows on the physical screen.
				284	<DT> screens
				285	<DD>
				286	A subset of windows which are as large as the terminal screen, i.e., they start
				287	at the upper left hand corner and encompass the lower right hand corner. One
				288	of these, <CODE>stdscr</CODE>, is automatically provided for the programmer.
				289	<DT> terminal screen
				290	<DD>
				291	The package's idea of what the terminal display currently looks like, i.e.,
				292	what the user sees now. This is a special screen.
				293	</DL>
				294
				295	<H1><A NAME="curses">The Curses Library</A></H1>
				296
				297	<H2><A NAME="overview">An Overview of Curses</A></H2>
				298
				299	<H3><A NAME="compiling">Compiling Programs using Curses</A></H3>
				300
				301	In order to use the library, it is necessary to have certain types and
				302	variables defined. Therefore, the programmer must have a line:
				303
				304	<PRE>
				305	#include <curses.h>
				306	</PRE>
				307
				308	at the top of the program source. The screen package uses the Standard I/O
				309	library, so <CODE><curses.h></CODE> includes
				310	<CODE><stdio.h></CODE>. <CODE><curses.h></CODE> also includes
				311	<CODE><termios.h></CODE>, <CODE><termio.h></CODE>, or
				312	<CODE><sgtty.h></CODE> depending on your system. It is redundant (but
				313	harmless) for the programmer to do these includes, too. In linking with
				314	<CODE>curses</CODE> you need to have <CODE>-lncurses</CODE> in your LDFLAGS or on the
				315	command line. There is no need for any other libraries.
				316
				317	<H3><A NAME="updating">Updating the Screen</A></H3>
				318
				319	In order to update the screen optimally, it is necessary for the routines to
				320	know what the screen currently looks like and what the programmer wants it to
				321	look like next. For this purpose, a data type (structure) named WINDOW is
				322	defined which describes a window image to the routines, including its starting
				323	position on the screen (the (y, x) coordinates of the upper left hand corner)
				324	and its size. One of these (called <CODE>curscr</CODE>, for current screen) is a
				325	screen image of what the terminal currently looks like. Another screen (called
				326	<CODE>stdscr</CODE>, for standard screen) is provided by default to make changes
				327	on. <P>
				328
				329	A window is a purely internal representation. It is used to build and store a
				330	potential image of a portion of the terminal. It doesn't bear any necessary
				331	relation to what is really on the terminal screen; it's more like a
				332	scratchpad or write buffer. <P>
				333
				334	To make the section of physical screen corresponding to a window reflect the
				335	contents of the window structure, the routine <CODE>refresh()</CODE> (or
				336	<CODE>wrefresh()</CODE> if the window is not <CODE>stdscr</CODE>) is called. <P>
				337
				338	A given physical screen section may be within the scope of any number of
				339	overlapping windows. Also, changes can be made to windows in any order,
				340	without regard to motion efficiency. Then, at will, the programmer can
				341	effectively say ``make it look like this,'' and let the package implementation
				342	determine the most efficient way to repaint the screen.
				343
				344	<H3><A NAME="stdscr">Standard Windows and Function Naming Conventions</A></H3>
				345
				346	As hinted above, the routines can use several windows, but two are
				347	automatically given: <CODE>curscr</CODE>, which knows what the terminal looks like,
				348	and <CODE>stdscr</CODE>, which is what the programmer wants the terminal to look
				349	like next. The user should never actually access <CODE>curscr</CODE> directly.
				350	Changes should be made to through the API, and then the routine
				351	<CODE>refresh()</CODE> (or <CODE>wrefresh()</CODE>) called. <P>
				352
				353	Many functions are defined to use <CODE>stdscr</CODE> as a default screen. For
				354	example, to add a character to <CODE>stdscr</CODE>, one calls <CODE>addch()</CODE> with
				355	the desired character as argument. To write to a different window. use the
				356	routine <CODE>waddch()</CODE> (for `w'indow-specific addch()) is provided. This
				357	convention of prepending function names with a `w' when they are to be
				358	applied to specific windows is consistent. The only routines which do not
				359	follow it are those for which a window must always be specified. <P>
				360
				361	In order to move the current (y, x) coordinates from one point to another, the
				362	routines <CODE>move()</CODE> and <CODE>wmove()</CODE> are provided. However, it is
				363	often desirable to first move and then perform some I/O operation. In order to
				364	avoid clumsiness, most I/O routines can be preceded by the prefix 'mv' and
				365	the desired (y, x) coordinates prepended to the arguments to the function. For
				366	example, the calls
				367
				368	<PRE>
				369	move(y, x);
				370	addch(ch);
				371	</PRE>
				372
				373	can be replaced by
				374
				375	<PRE>
				376	mvaddch(y, x, ch);
				377	</PRE>
				378
				379	and
				380
				381	<PRE>
				382	wmove(win, y, x);
				383	waddch(win, ch);
				384	</PRE>
				385
				386	can be replaced by
				387
				388	<PRE>
				389	mvwaddch(win, y, x, ch);
				390	</PRE>
				391
				392	Note that the window description pointer (win) comes before the added (y, x)
				393	coordinates. If a function requires a window pointer, it is always the first
				394	parameter passed.
				395
				396	<H3><A NAME="variables">Variables</A></H3>
				397
				398	The <CODE>curses</CODE> library sets some variables describing the terminal
				399	capabilities.
				400
				401	<PRE>
				402	type name description
				403	------------------------------------------------------------------
				404	int LINES number of lines on the terminal
				405	int COLS number of columns on the terminal
				406	</PRE>
				407
				408	The <CODE>curses.h</CODE> also introduces some <CODE>#define</CODE> constants and types
				409	of general usefulness:
				410
				411	<DL>
				412	<DT> <CODE>bool</CODE>
				413	<DD> boolean type, actually a `char' (e.g., <CODE>bool doneit;</CODE>)
				414	<DT> <CODE>TRUE</CODE>
				415	<DD> boolean `true' flag (1).
				416	<DT> <CODE>FALSE</CODE>
				417	<DD> boolean `false' flag (0).
				418	<DT> <CODE>ERR</CODE>
				419	<DD> error flag returned by routines on a failure (-1).
				420	<DT> <CODE>OK</CODE>
				421	<DD> error flag returned by routines when things go right.
				422	</DL>
				423
				424	<H2><A NAME="using">Using the Library</A></H2>
				425
				426	Now we describe how to actually use the screen package. In it, we assume all
				427	updating, reading, etc. is applied to <CODE>stdscr</CODE>. These instructions will
				428	work on any window, providing you change the function names and parameters as
				429	mentioned above. <P>
				430
				431	Here is a sample program to motivate the discussion:
				432
				433	<PRE>
				434	#include <curses.h>
				435	#include <signal.h>
				436
				437	static void finish(int sig);
				438
				439	int
				440	main(int argc, char *argv[])
				441	{
				442	int num = 0;
				443
				444	/* initialize your non-curses data structures here */
				445
				446	(void) signal(SIGINT, finish); /* arrange interrupts to terminate */
				447
				448	(void) initscr(); /* initialize the curses library */
				449	keypad(stdscr, TRUE); /* enable keyboard mapping */
				450	(void) nonl(); /* tell curses not to do NL->CR/NL on output */
				451	(void) cbreak(); /* take input chars one at a time, no wait for \n */
				452	(void) echo(); /* echo input - in color */
				453
				454	if (has_colors())
				455	{
				456	start_color();
				457
				458	/*
				459	* Simple color assignment, often all we need. Color pair 0 cannot
				460	* be redefined. This example uses the same value for the color
				461	* pair as for the foreground color, though of course that is not
				462	* necessary:
				463	*/
				464	init_pair(1, COLOR_RED, COLOR_BLACK);
				465	init_pair(2, COLOR_GREEN, COLOR_BLACK);
				466	init_pair(3, COLOR_YELLOW, COLOR_BLACK);
				467	init_pair(4, COLOR_BLUE, COLOR_BLACK);
				468	init_pair(5, COLOR_CYAN, COLOR_BLACK);
				469	init_pair(6, COLOR_MAGENTA, COLOR_BLACK);
				470	init_pair(7, COLOR_WHITE, COLOR_BLACK);
				471	}
				472
				473	for (;;)
				474	{
				475	int c = getch(); /* refresh, accept single keystroke of input */
				476	attrset(COLOR_PAIR(num % 8));
				477	num++;
				478
				479	/* process the command keystroke */
				480	}
				481
				482	finish(0); /* we're done */
				483	}
				484
				485	static void finish(int sig)
				486	{
				487	endwin();
				488
				489	/* do your non-curses wrapup here */
				490
				491	exit(0);
				492	}
				493	</PRE>
				494
				495	<H3><A NAME="starting">Starting up</A></H3>
				496
				497	In order to use the screen package, the routines must know about terminal
				498	characteristics, and the space for <CODE>curscr</CODE> and <CODE>stdscr</CODE> must be
				499	allocated. These function <CODE>initscr()</CODE> does both these things. Since it
				500	must allocate space for the windows, it can overflow memory when attempting to
				501	do so. On the rare occasions this happens, <CODE>initscr()</CODE> will terminate
				502	the program with an error message. <CODE>initscr()</CODE> must always be called
				503	before any of the routines which affect windows are used. If it is not, the
				504	program will core dump as soon as either <CODE>curscr</CODE> or <CODE>stdscr</CODE> are
				505	referenced. However, it is usually best to wait to call it until after you are
				506	sure you will need it, like after checking for startup errors. Terminal status
				507	changing routines like <CODE>nl()</CODE> and <CODE>cbreak()</CODE> should be called
				508	after <CODE>initscr()</CODE>. <P>
				509
				510	Once the screen windows have been allocated, you can set them up for
				511	your program. If you want to, say, allow a screen to scroll, use
				512	<CODE>scrollok()</CODE>. If you want the cursor to be left in place after
				513	the last change, use <CODE>leaveok()</CODE>. If this isn't done,
				514	<CODE>refresh()</CODE> will move the cursor to the window's current (y, x)
				515	coordinates after updating it. <P>
				516
				517	You can create new windows of your own using the functions <CODE>newwin()</CODE>,
				518	<CODE>derwin()</CODE>, and <CODE>subwin()</CODE>. The routine <CODE>delwin()</CODE> will
				519	allow you to get rid of old windows. All the options described above can be
				520	applied to any window.
				521
				522	<H3><A NAME="output">Output</A></H3>
				523
				524	Now that we have set things up, we will want to actually update the terminal.
				525	The basic functions used to change what will go on a window are
				526	<CODE>addch()</CODE> and <CODE>move()</CODE>. <CODE>addch()</CODE> adds a character at the
				527	current (y, x) coordinates. <CODE>move()</CODE> changes the current (y, x)
				528	coordinates to whatever you want them to be. It returns <CODE>ERR</CODE> if you
				529	try to move off the window. As mentioned above, you can combine the two into
				530	<CODE>mvaddch()</CODE> to do both things at once. <P>
				531
				532	The other output functions, such as <CODE>addstr()</CODE> and <CODE>printw()</CODE>,
				533	all call <CODE>addch()</CODE> to add characters to the window. <P>
				534
				535	After you have put on the window what you want there, when you want the portion
				536	of the terminal covered by the window to be made to look like it, you must call
				537	<CODE>refresh()</CODE>. In order to optimize finding changes, <CODE>refresh()</CODE>
				538	assumes that any part of the window not changed since the last
				539	<CODE>refresh()</CODE> of that window has not been changed on the terminal, i.e.,
				540	that you have not refreshed a portion of the terminal with an overlapping
				541	window. If this is not the case, the routine <CODE>touchwin()</CODE> is provided
				542	to make it look like the entire window has been changed, thus making
				543	<CODE>refresh()</CODE> check the whole subsection of the terminal for changes. <P>
				544
				545	If you call <CODE>wrefresh()</CODE> with <CODE>curscr</CODE> as its argument, it will
				546	make the screen look like <CODE>curscr</CODE> thinks it looks like. This is useful
				547	for implementing a command which would redraw the screen in case it get messed
				548	up.
				549
				550	<H3><A NAME="input">Input</A></H3>
				551
				552	The complementary function to <CODE>addch()</CODE> is <CODE>getch()</CODE> which, if
				553	echo is set, will call <CODE>addch()</CODE> to echo the character. Since the
				554	screen package needs to know what is on the terminal at all times, if
				555	characters are to be echoed, the tty must be in raw or cbreak mode. Since
				556	initially the terminal has echoing enabled and is in ordinary ``cooked'' mode,
				557	one or the other has to changed before calling <CODE>getch()</CODE>; otherwise,
				558	the program's output will be unpredictable. <P>
				559
				560	When you need to accept line-oriented input in a window, the functions
				561	<CODE>wgetstr()</CODE> and friends are available. There is even a <CODE>wscanw()</CODE>
				562	function that can do <CODE>scanf()</CODE>(3)-style multi-field parsing on window
				563	input. These pseudo-line-oriented functions turn on echoing while they
				564	execute. <P>
				565
				566	The example code above uses the call <CODE>keypad(stdscr, TRUE)</CODE> to enable
				567	support for function-key mapping. With this feature, the <CODE>getch()</CODE> code
				568	watches the input stream for character sequences that correspond to arrow and
				569	function keys. These sequences are returned as pseudo-character values. The
				570	<CODE>#define</CODE> values returned are listed in the <CODE>curses.h</CODE> The
				571	mapping from sequences to <CODE>#define</CODE> values is determined by
				572	<CODE>key_</CODE> capabilities in the terminal's terminfo entry.
				573
				574	<H3><A NAME="formschars">Using Forms Characters</A></H3>
				575
				576	The <CODE>addch()</CODE> function (and some others, including <CODE>box()</CODE> and
				577	<CODE>border()</CODE>) can accept some pseudo-character arguments which are specially
				578	defined by <CODE>ncurses</CODE>. These are <CODE>#define</CODE> values set up in
				579	the <CODE>curses.h</CODE> header; see there for a complete list (look for
				580	the prefix <CODE>ACS_</CODE>). <P>
				581
				582	The most useful of the ACS defines are the forms-drawing characters. You can
				583	use these to draw boxes and simple graphs on the screen. If the terminal
				584	does not have such characters, <CODE>curses.h</CODE> will map them to a
				585	recognizable (though ugly) set of ASCII defaults.
				586
				587	<H3><A NAME="attributes">Character Attributes and Color</A></H3>
				588
				589	The <CODE>ncurses</CODE> package supports screen highlights including standout,
				590	reverse-video, underline, and blink. It also supports color, which is treated
				591	as another kind of highlight. <P>
				592
				593	Highlights are encoded, internally, as high bits of the pseudo-character type
				594	(<CODE>chtype</CODE>) that <CODE>curses.h</CODE> uses to represent the contents of a
				595	screen cell. See the <CODE>curses.h</CODE> header file for a complete list of
				596	highlight mask values (look for the prefix <CODE>A_</CODE>).<P>
				597
				598	There are two ways to make highlights. One is to logical-or the value of the
				599	highlights you want into the character argument of an <CODE>addch()</CODE> call,
				600	or any other output call that takes a <CODE>chtype</CODE> argument. <P>
				601
				602	The other is to set the current-highlight value. This is logical-or'ed with
				603	any highlight you specify the first way. You do this with the functions
				604	<CODE>attron()</CODE>, <CODE>attroff()</CODE>, and <CODE>attrset()</CODE>; see the manual
				605	pages for details.
				606
				607	Color is a special kind of highlight. The package actually thinks in terms
				608	of color pairs, combinations of foreground and background colors. The sample
				609	code above sets up eight color pairs, all of the guaranteed-available colors
				610	on black. Note that each color pair is, in effect, given the name of its
				611	foreground color. Any other range of eight non-conflicting values could
				612	have been used as the first arguments of the <CODE>init_pair()</CODE> values. <P>
				613
				614	Once you've done an <CODE>init_pair()</CODE> that creates color-pair N, you can
				615	use <CODE>COLOR_PAIR(N)</CODE> as a highlight that invokes that particular
				616	color combination. Note that <CODE>COLOR_PAIR(N)</CODE>, for constant N,
				617	is itself a compile-time constant and can be used in initializers.
				618
				619	<H3><A NAME="mouse">Mouse Interfacing</A></H3>
				620
				621	The <CODE>ncurses</CODE> library also provides a mouse interface.
				622	<!-- The 'note' tag is not portable enough -->
				623	<blockquote>
				624	<strong>NOTE:</strong> this facility is specific to <CODE>ncurses</CODE>, it is not part of either
				625	the XSI Curses standard, nor of System V Release 4, nor BSD curses.
				626	System V Release 4 curses contains code with similar interface definitions,
				627	however it is not documented. Other than by disassembling the library, we
				628	have no way to determine exactly how that mouse code works.
				629	Thus, we recommend that you wrap mouse-related code in an #ifdef using the
				630	feature macro NCURSES_MOUSE_VERSION so it will not be compiled and linked
				631	on non-ncurses systems.
				632	</blockquote>
				633
				634	Presently, mouse event reporting works in the following environments:
				635	<ul>
				636	<li>xterm and similar programs such as rxvt.
				637	<li>Linux console, when configured with <CODE>gpm</CODE>(1), Alessandro
				638	Rubini's mouse server.
				639	<li>FreeBSD sysmouse (console)
				640	<li>OS/2 EMX
				641	</ul>
				642	<P>
				643	The mouse interface is very simple. To activate it, you use the function
				644	<CODE>mousemask()</CODE>, passing it as first argument a bit-mask that specifies
				645	what kinds of events you want your program to be able to see. It will
				646	return the bit-mask of events that actually become visible, which may differ
				647	from the argument if the mouse device is not capable of reporting some of
				648	the event types you specify. <P>
				649
				650	Once the mouse is active, your application's command loop should watch
				651	for a return value of <CODE>KEY_MOUSE</CODE> from <CODE>wgetch()</CODE>. When
				652	you see this, a mouse event report has been queued. To pick it off
				653	the queue, use the function <CODE>getmouse()</CODE> (you must do this before
				654	the next <CODE>wgetch()</CODE>, otherwise another mouse event might come
				655	in and make the first one inaccessible). <P>
				656
				657	Each call to <CODE>getmouse()</CODE> fills a structure (the address of which you'll
				658	pass it) with mouse event data. The event data includes zero-origin,
				659	screen-relative character-cell coordinates of the mouse pointer. It also
				660	includes an event mask. Bits in this mask will be set, corresponding
				661	to the event type being reported. <P>
				662
				663	The mouse structure contains two additional fields which may be
				664	significant in the future as ncurses interfaces to new kinds of
				665	pointing device. In addition to x and y coordinates, there is a slot
				666	for a z coordinate; this might be useful with touch-screens that can
				667	return a pressure or duration parameter. There is also a device ID
				668	field, which could be used to distinguish between multiple pointing
				669	devices. <P>
				670
				671	The class of visible events may be changed at any time via <CODE>mousemask()</CODE>.
				672	Events that can be reported include presses, releases, single-, double- and
				673	triple-clicks (you can set the maximum button-down time for clicks). If
				674	you don't make clicks visible, they will be reported as press-release
				675	pairs. In some environments, the event mask may include bits reporting
				676	the state of shift, alt, and ctrl keys on the keyboard during the event. <P>
				677
				678	A function to check whether a mouse event fell within a given window is
				679	also supplied. You can use this to see whether a given window should
				680	consider a mouse event relevant to it. <P>
				681
				682	Because mouse event reporting will not be available in all
				683	environments, it would be unwise to build <CODE>ncurses</CODE>
				684	applications that <EM>require</EM> the use of a mouse. Rather, you should
				685	use the mouse as a shortcut for point-and-shoot commands your application
				686	would normally accept from the keyboard. Two of the test games in the
				687	<CODE>ncurses</CODE> distribution (<CODE>bs</CODE> and <CODE>knight</CODE>) contain
				688	code that illustrates how this can be done. <P>
				689
				690	See the manual page <CODE>curs_mouse(3X)</CODE> for full details of the
				691	mouse-interface functions.
				692
				693	<H3><A NAME="finishing">Finishing Up</A></H3>
				694
				695	In order to clean up after the <CODE>ncurses</CODE> routines, the routine
				696	<CODE>endwin()</CODE> is provided. It restores tty modes to what they were when
				697	<CODE>initscr()</CODE> was first called, and moves the cursor down to the
				698	lower-left corner. Thus, anytime after the call to initscr, <CODE>endwin()</CODE>
				699	should be called before exiting.
				700
				701	<H2><A NAME="functions">Function Descriptions</A></H2>
				702
				703	We describe the detailed behavior of some important curses functions here, as a
				704	supplement to the manual page descriptions.
				705
				706	<H3><A NAME="init">Initialization and Wrapup</A></H3>
				707
				708	<DL>
				709	<DT> <CODE>initscr()</CODE>
				710	<DD> The first function called should almost always be <CODE>initscr()</CODE>.
				711	This will determine the terminal type and
				712	initialize curses data structures. <CODE>initscr()</CODE> also arranges that
				713	the first call to <CODE>refresh()</CODE> will clear the screen. If an error
				714	occurs a message is written to standard error and the program
				715	exits. Otherwise it returns a pointer to stdscr. A few functions may be
				716	called before initscr (<CODE>slk_init()</CODE>, <CODE>filter()</CODE>,
				717	<CODE>ripoffline()</CODE>, <CODE>use_env()</CODE>, and, if you are using multiple
				718	terminals, <CODE>newterm()</CODE>.)
				719	<DT> <CODE>endwin()</CODE>
				720	<DD> Your program should always call <CODE>endwin()</CODE> before exiting or
				721	shelling out of the program. This function will restore tty modes,
				722	move the cursor to the lower left corner of the screen, reset the
				723	terminal into the proper non-visual mode. Calling <CODE>refresh()</CODE>
				724	or <CODE>doupdate()</CODE> after a temporary escape from the program will
				725	restore the ncurses screen from before the escape.
				726	<DT> <CODE>newterm(type, ofp, ifp)</CODE>
				727	<DD> A program which outputs to more than one terminal should use
				728	<CODE>newterm()</CODE> instead of <CODE>initscr()</CODE>. <CODE>newterm()</CODE> should
				729	be called once for each terminal. It returns a variable of type
				730	<CODE>SCREEN *</CODE> which should be saved as a reference to that
				731	terminal.
				732	(NOTE: a SCREEN variable is not a <em>screen</em> in the sense we
				733	are describing in this introduction, but a collection of
				734	parameters used to assist in optimizing the display.)
				735	The arguments are the type of the terminal (a string) and
				736	<CODE>FILE</CODE> pointers for the output and input of the terminal. If
				737	type is NULL then the environment variable <CODE>$TERM</CODE> is used.
				738	<CODE>endwin()</CODE> should called once at wrapup time for each terminal
				739	opened using this function.
				740	<DT> <CODE>set_term(new)</CODE>
				741	<DD> This function is used to switch to a different terminal previously
				742	opened by <CODE>newterm()</CODE>. The screen reference for the new terminal
				743	is passed as the parameter. The previous terminal is returned by the
				744	function. All other calls affect only the current terminal.
				745	<DT> <CODE>delscreen(sp)</CODE>
				746	<DD> The inverse of <CODE>newterm()</CODE>; deallocates the data structures
				747	associated with a given <CODE>SCREEN</CODE> reference.
				748	</DL>
				749
				750	<H3><A NAME="flush">Causing Output to the Terminal</A></H3>
				751
				752	<DL>
				753	<DT> <CODE>refresh()</CODE> and <CODE>wrefresh(win)</CODE>
				754	<DD> These functions must be called to actually get any output on
				755	the terminal, as other routines merely manipulate data
				756	structures. <CODE>wrefresh()</CODE> copies the named window to the physical
				757	terminal screen, taking into account what is already
				758	there in order to do optimizations. <CODE>refresh()</CODE> does a
				759	refresh of <CODE>stdscr</CODE>. Unless <CODE>leaveok()</CODE> has been
				760	enabled, the physical cursor of the terminal is left at the
				761	location of the window's cursor.
				762	<DT> <CODE>doupdate()</CODE> and <CODE>wnoutrefresh(win)</CODE>
				763	<DD> These two functions allow multiple updates with more efficiency
				764	than wrefresh. To use them, it is important to understand how curses
				765	works. In addition to all the window structures, curses keeps two
				766	data structures representing the terminal screen: a physical screen,
				767	describing what is actually on the screen, and a virtual screen,
				768	describing what the programmer wants to have on the screen. wrefresh
				769	works by first copying the named window to the virtual screen
				770	(<CODE>wnoutrefresh()</CODE>), and then calling the routine to update the
				771	screen (<CODE>doupdate()</CODE>). If the programmer wishes to output
				772	several windows at once, a series of calls to <CODE>wrefresh</CODE> will result
				773	in alternating calls to <CODE>wnoutrefresh()</CODE> and <CODE>doupdate()</CODE>,
				774	causing several bursts of output to the screen. By calling
				775	<CODE>wnoutrefresh()</CODE> for each window, it is then possible to call
				776	<CODE>doupdate()</CODE> once, resulting in only one burst of output, with
				777	fewer total characters transmitted (this also avoids a visually annoying
				778	flicker at each update).
				779	</DL>
				780
				781	<H3><A NAME="lowlevel">Low-Level Capability Access</A></H3>
				782
				783	<DL>
				784	<DT> <CODE>setupterm(term, filenum, errret)</CODE>
				785	<DD> This routine is called to initialize a terminal's description, without setting
				786	up the curses screen structures or changing the tty-driver mode bits.
				787	<CODE>term</CODE> is the character string representing the name of the terminal
				788	being used. <CODE>filenum</CODE> is the UNIX file descriptor of the terminal to
				789	be used for output. <CODE>errret</CODE> is a pointer to an integer, in which a
				790	success or failure indication is returned. The values returned can be 1 (all
				791	is well), 0 (no such terminal), or -1 (some problem locating the terminfo
				792	database). <P>
				793
				794	The value of <CODE>term</CODE> can be given as NULL, which will cause the value of
				795	<CODE>TERM</CODE> in the environment to be used. The <CODE>errret</CODE> pointer can
				796	also be given as NULL, meaning no error code is wanted. If <CODE>errret</CODE> is
				797	defaulted, and something goes wrong, <CODE>setupterm()</CODE> will print an
				798	appropriate error message and exit, rather than returning. Thus, a simple
				799	program can call setupterm(0, 1, 0) and not worry about initialization
				800	errors. <P>
				801
				802	After the call to <CODE>setupterm()</CODE>, the global variable <CODE>cur_term</CODE> is
				803	set to point to the current structure of terminal capabilities. By calling
				804	<CODE>setupterm()</CODE> for each terminal, and saving and restoring
				805	<CODE>cur_term</CODE>, it is possible for a program to use two or more terminals at
				806	once. <CODE>Setupterm()</CODE> also stores the names section of the terminal
				807	description in the global character array <CODE>ttytype[]</CODE>. Subsequent calls
				808	to <CODE>setupterm()</CODE> will overwrite this array, so you'll have to save it
				809	yourself if need be.
				810	</DL>
				811
				812	<H3><A NAME="debugging">Debugging</A></H3>
				813
				814	<!-- The 'note' tag is not portable enough -->
				815	<blockquote>
				816	<strong>NOTE:</strong> These functions are not part of the standard curses API!
				817	</blockquote>
				818
				819	<DL>
				820	<DT> <CODE>trace()</CODE>
				821	<DD>
				822	This function can be used to explicitly set a trace level. If the
				823	trace level is nonzero, execution of your program will generate a file
				824	called `trace' in the current working directory containing a report on
				825	the library's actions. Higher trace levels enable more detailed (and
				826	verbose) reporting -- see comments attached to <CODE>TRACE_</CODE> defines
				827	in the <CODE>curses.h</CODE> file for details. (It is also possible to set
				828	a trace level by assigning a trace level value to the environment variable
				829	<CODE>NCURSES_TRACE</CODE>).
				830	<DT> <CODE>_tracef()</CODE>
				831	<DD>
				832	This function can be used to output your own debugging information. It is only
				833	available only if you link with -lncurses_g. It can be used the same way as
				834	<CODE>printf()</CODE>, only it outputs a newline after the end of arguments.
				835	The output goes to a file called <CODE>trace</CODE> in the current directory.
				836	</DL>
				837
				838	Trace logs can be difficult to interpret due to the sheer volume of
				839	data dumped in them. There is a script called <STRONG>tracemunch</STRONG>
				840	included with the <CODE>ncurses</CODE> distribution that can alleviate
				841	this problem somewhat; it compacts long sequences of similar operations into
				842	more succinct single-line pseudo-operations. These pseudo-ops can be
				843	distinguished by the fact that they are named in capital letters.
				844
				845	<H2><A NAME="hints">Hints, Tips, and Tricks</A></H2>
				846
				847	The <CODE>ncurses</CODE> manual pages are a complete reference for this library.
				848	In the remainder of this document, we discuss various useful methods that
				849	may not be obvious from the manual page descriptions.
				850
				851	<H3><A NAME="caution">Some Notes of Caution</A></H3>
				852
				853	If you find yourself thinking you need to use <CODE>noraw()</CODE> or
				854	<CODE>nocbreak()</CODE>, think again and move carefully. It's probably
				855	better design to use <CODE>getstr()</CODE> or one of its relatives to
				856	simulate cooked mode. The <CODE>noraw()</CODE> and <CODE>nocbreak()</CODE>
				857	functions try to restore cooked mode, but they may end up clobbering
				858	some control bits set before you started your application. Also, they
				859	have always been poorly documented, and are likely to hurt your
				860	application's usability with other curses libraries. <P>
				861
				862	Bear in mind that <CODE>refresh()</CODE> is a synonym for <CODE>wrefresh(stdscr)</CODE>.
				863	Don't try to mix use of <CODE>stdscr</CODE> with use of windows declared
				864	by <CODE>newwin()</CODE>; a <CODE>refresh()</CODE> call will blow them off the
				865	screen. The right way to handle this is to use <CODE>subwin()</CODE>, or
				866	not touch <CODE>stdscr</CODE> at all and tile your screen with declared
				867	windows which you then <CODE>wnoutrefresh()</CODE> somewhere in your program
				868	event loop, with a single <CODE>doupdate()</CODE> call to trigger actual
				869	repainting. <P>
				870
				871	You are much less likely to run into problems if you design your screen
				872	layouts to use tiled rather than overlapping windows. Historically,
				873	curses support for overlapping windows has been weak, fragile, and poorly
				874	documented. The <CODE>ncurses</CODE> library is not yet an exception to this
				875	rule. <P>
				876
				877	There is a panels library included in the <CODE>ncurses</CODE>
				878	distribution that does a pretty good job of strengthening the
				879	overlapping-windows facilities. <P>
				880
				881	Try to avoid using the global variables LINES and COLS. Use
				882	<CODE>getmaxyx()</CODE> on the <CODE>stdscr</CODE> context instead. Reason:
				883	your code may be ported to run in an environment with window resizes,
				884	in which case several screens could be open with different sizes.
				885
				886	<H3><A NAME="leaving">Temporarily Leaving NCURSES Mode</A></H3>
				887
				888	Sometimes you will want to write a program that spends most of its time in
				889	screen mode, but occasionally returns to ordinary `cooked' mode. A common
				890	reason for this is to support shell-out. This behavior is simple to arrange
				891	in <CODE>ncurses</CODE>. <P>
				892
				893	To leave <CODE>ncurses</CODE> mode, call <CODE>endwin()</CODE> as you would if you
				894	were intending to terminate the program. This will take the screen back to
				895	cooked mode; you can do your shell-out. When you want to return to
				896	<CODE>ncurses</CODE> mode, simply call <CODE>refresh()</CODE> or <CODE>doupdate()</CODE>.
				897	This will repaint the screen. <P>
				898
				899	There is a boolean function, <CODE>isendwin()</CODE>, which code can use to
				900	test whether <CODE>ncurses</CODE> screen mode is active. It returns <CODE>TRUE</CODE>
				901	in the interval between an <CODE>endwin()</CODE> call and the following
				902	<CODE>refresh()</CODE>, <CODE>FALSE</CODE> otherwise. <P>
				903
				904	Here is some sample code for shellout:
				905
				906	<PRE>
				907	addstr("Shelling out...");
				908	def_prog_mode(); /* save current tty modes */
				909	endwin(); /* restore original tty modes */
				910	system("sh"); /* run shell */
				911	addstr("returned.\n"); /* prepare return message */
				912	refresh(); /* restore save modes, repaint screen */
				913	</PRE>
				914
				915	<H3><A NAME="xterm">Using NCURSES under XTERM</A></H3>
				916
				917	A resize operation in X sends <CODE>SIGWINCH</CODE> to the application running
				918	under xterm.
				919
				920	The easiest way to handle <CODE>SIGWINCH</CODE>
				921	is to do an <CODE>endwin</CODE>,
				922	followed by an <CODE>refresh</CODE> and a screen repaint you code
				923	yourself.
				924	The <CODE>refresh</CODE> will pick up the new screen size from the
				925	xterm's environment. <P>
				926
				927	That is the standard way, of course (it even works with some vendor's curses
				928	implementations).
				929	Its drawback is that it clears the screen to reinitialize the display, and does
				930	not resize subwindows which must be shrunk.
				931	<CODE>Ncurses</CODE> provides an extension which works better, the
				932	<CODE>resizeterm</CODE> function. That function ensures that all windows
				933	are limited to the new screen dimensions, and pads <CODE>stdscr</CODE>
				934	with blanks if the screen is larger. <P>
				935
				936	The <CODE>ncurses</CODE> library provides a SIGWINCH signal handler,
				937	which pushes a <CODE>KEY_RESIZE</CODE> via the wgetch() calls.
				938	When <CODE>ncurses</CODE> returns that code,
				939	it calls <code>resizeterm</CODE>
				940	to update the size of the standard screen's window, repainting that
				941	(filling with blanks or truncating as needed).
				942	It also resizes other windows,
				943	but its effect may be less satisfactory because it cannot
				944	know how you want the screen re-painted.
				945	You will usually have to write special-purpose code to handle
				946	<CODE>KEY_RESIZE</CODE> yourself.
				947
				948	<H3><A NAME="screens">Handling Multiple Terminal Screens</A></H3>
				949
				950	The <CODE>initscr()</CODE> function actually calls a function named
				951	<CODE>newterm()</CODE> to do most of its work. If you are writing a program that
				952	opens multiple terminals, use <CODE>newterm()</CODE> directly. <P>
				953
				954	For each call, you will have to specify a terminal type and a pair of file
				955	pointers; each call will return a screen reference, and <CODE>stdscr</CODE> will be
				956	set to the last one allocated. You will switch between screens with the
				957	<CODE>set_term</CODE> call. Note that you will also have to call
				958	<CODE>def_shell_mode</CODE> and <CODE>def_prog_mode</CODE> on each tty yourself.
				959
				960	<H3><A NAME="testing">Testing for Terminal Capabilities</A></H3>
				961
				962	Sometimes you may want to write programs that test for the presence of various
				963	capabilities before deciding whether to go into <CODE>ncurses</CODE> mode. An easy
				964	way to do this is to call <CODE>setupterm()</CODE>, then use the functions
				965	<CODE>tigetflag()</CODE>, <CODE>tigetnum()</CODE>, and <CODE>tigetstr()</CODE> to do your
				966	testing. <P>
				967
				968	A particularly useful case of this often comes up when you want to
				969	test whether a given terminal type should be treated as `smart'
				970	(cursor-addressable) or `stupid'. The right way to test this is to see
				971	if the return value of <CODE>tigetstr("cup")</CODE> is non-NULL. Alternatively,
				972	you can include the <CODE>term.h</CODE> file and test the value of the
				973	macro <CODE>cursor_address</CODE>.
				974
				975	<H3><A NAME="tuning">Tuning for Speed</A></H3>
				976
				977	Use the <CODE>addchstr()</CODE> family of functions for fast
				978	screen-painting of text when you know the text doesn't contain any
				979	control characters. Try to make attribute changes infrequent on your
				980	screens. Don't use the <CODE>immedok()</CODE> option!
				981
				982	<H3><A NAME="special">Special Features of NCURSES</A></H3>
				983
				984	The <CODE>wresize()</CODE> function allows you to resize a window in place.
				985	The associated <CODE>resizeterm()</CODE> function simplifies the construction
				986	of <a HREF="#xterm">SIGWINCH</a> handlers, for resizing all windows. <P>
				987
				988	The <CODE>define_key()</CODE> function allows you
				989	to define at runtime function-key control sequences which are not in the
				990	terminal description.
				991	The <CODE>keyok()</CODE> function allows you to temporarily
				992	enable or disable interpretation of any function-key control sequence. <P>
				993
				994	The <CODE>use_default_colors()</CODE> function allows you to construct
				995	applications which can use the terminal's default foreground and
				996	background colors as an additional "default" color.
				997	Several terminal emulators support this feature, which is based on ISO 6429. <P>
				998
				999	Ncurses supports up 16 colors, unlike SVr4 curses which defines only 8.
				1000	While most terminals which provide color allow only 8 colors, about
				1001	a quarter (including XFree86 xterm) support 16 colors.
				1002
				1003	<H2><A NAME="compat">Compatibility with Older Versions</A></H2>
				1004
				1005	Despite our best efforts, there are some differences between <CODE>ncurses</CODE>
				1006	and the (undocumented!) behavior of older curses implementations. These arise
				1007	from ambiguities or omissions in the documentation of the API.
				1008
				1009	<H3><A NAME="refbug">Refresh of Overlapping Windows</A></H3>
				1010
				1011	If you define two windows A and B that overlap, and then alternately scribble
				1012	on and refresh them, the changes made to the overlapping region under historic
				1013	<CODE>curses</CODE> versions were often not documented precisely. <P>
				1014
				1015	To understand why this is a problem, remember that screen updates are
				1016	calculated between two representations of the <EM>entire</EM> display. The
				1017	documentation says that when you refresh a window, it is first copied to the
				1018	virtual screen, and then changes are calculated to update the physical screen
				1019	(and applied to the terminal). But "copied to" is not very specific, and
				1020	subtle differences in how copying works can produce different behaviors in the
				1021	case where two overlapping windows are each being refreshed at unpredictable
				1022	intervals. <P>
				1023
				1024	What happens to the overlapping region depends on what <CODE>wnoutrefresh()</CODE>
				1025	does with its argument -- what portions of the argument window it copies to the
				1026	virtual screen. Some implementations do "change copy", copying down only
				1027	locations in the window that have changed (or been marked changed with
				1028	<CODE>wtouchln()</CODE> and friends). Some implementations do "entire copy",
				1029	copying <EM>all</EM> window locations to the virtual screen whether or not
				1030	they have changed. <P>
				1031
				1032	The <CODE>ncurses</CODE> library itself has not always been consistent on this
				1033	score. Due to a bug, versions 1.8.7 to 1.9.8a did entire copy. Versions
				1034	1.8.6 and older, and versions 1.9.9 and newer, do change copy. <P>
				1035
				1036	For most commercial curses implementations, it is not documented and not known
				1037	for sure (at least not to the <CODE>ncurses</CODE> maintainers) whether they do
				1038	change copy or entire copy. We know that System V release 3 curses has logic
				1039	in it that looks like an attempt to do change copy, but the surrounding logic
				1040	and data representations are sufficiently complex, and our knowledge
				1041	sufficiently indirect, that it's hard to know whether this is reliable.
				1042
				1043	It is not clear what the SVr4 documentation and XSI standard intend. The XSI
				1044	Curses standard barely mentions wnoutrefresh(); the SVr4 documents seem to be
				1045	describing entire-copy, but it is possible with some effort and straining to
				1046	read them the other way. <P>
				1047
				1048	It might therefore be unwise to rely on either behavior in programs that might
				1049	have to be linked with other curses implementations. Instead, you can do an
				1050	explicit <CODE>touchwin()</CODE> before the <CODE>wnoutrefresh()</CODE> call to
				1051	guarantee an entire-contents copy anywhere. <P>
				1052
				1053	The really clean way to handle this is to use the panels library. If,
				1054	when you want a screen update, you do <CODE>update_panels()</CODE>, it will
				1055	do all the necessary <CODE>wnoutrefresh()</CODE> calls for whatever panel
				1056	stacking order you have defined. Then you can do one <CODE>doupdate()</CODE>
				1057	and there will be a <EM>single</EM> burst of physical I/O that will do
				1058	all your updates.
				1059
				1060	<H3><A NAME="backbug">Background Erase</A></H3>
				1061
				1062	If you have been using a very old versions of <CODE>ncurses</CODE> (1.8.7 or
				1063	older) you may be surprised by the behavior of the erase functions. In older
				1064	versions, erased areas of a window were filled with a blank modified by the
				1065	window's current attribute (as set by <STRONG>wattrset()</STRONG>, <STRONG>wattron()</STRONG>,
				1066	<STRONG>wattroff()</STRONG> and friends). <P>
				1067
				1068	In newer versions, this is not so. Instead, the attribute of erased blanks
				1069	is normal unless and until it is modified by the functions <CODE>bkgdset()</CODE>
				1070	or <CODE>wbkgdset()</CODE>. <P>
				1071
				1072	This change in behavior conforms <CODE>ncurses</CODE> to System V Release 4 and
				1073	the XSI Curses standard.
				1074
				1075	<H2><A NAME="xsifuncs">XSI Curses Conformance</A></H2>
				1076
				1077	The <CODE>ncurses</CODE> library is intended to be base-level conformant with the
				1078	XSI Curses standard from X/Open. Many extended-level features (in fact, almost
				1079	all features not directly concerned with wide characters and
				1080	internationalization) are also supported. <P>
				1081
				1082	One effect of XSI conformance is the change in behavior described under
				1083	<A HREF="#backbug">"Background Erase -- Compatibility with Old Versions"</A>. <P>
				1084
				1085	Also, <CODE>ncurses</CODE> meets the XSI requirement that every macro
				1086	entry point have a corresponding function which may be linked (and
				1087	will be prototype-checked) if the macro definition is disabled with
				1088	<CODE>#undef</CODE>.
				1089
				1090	<H1><A NAME="panels">The Panels Library</A></H1>
				1091
				1092	The <CODE>ncurses</CODE> library by itself provides good support for screen
				1093	displays in which the windows are tiled (non-overlapping). In the more
				1094	general case that windows may overlap, you have to use a series of
				1095	<CODE>wnoutrefresh()</CODE> calls followed by a <CODE>doupdate()</CODE>, and be
				1096	careful about the order you do the window refreshes in. It has to be
				1097	bottom-upwards, otherwise parts of windows that should be obscured will
				1098	show through. <P>
				1099
				1100	When your interface design is such that windows may dive deeper into the
				1101	visibility stack or pop to the top at runtime, the resulting book-keeping
				1102	can be tedious and difficult to get right. Hence the panels library. <P>
				1103
				1104	The <CODE>panel</CODE> library first appeared in AT&T System V. The
				1105	version documented here is the <CODE>panel</CODE> code distributed
				1106	with <CODE>ncurses</CODE>.
				1107
				1108	<H2><A NAME="pcompile">Compiling With the Panels Library</A></H2>
				1109
				1110	Your panels-using modules must import the panels library declarations with
				1111
				1112	<PRE>
				1113	#include <panel.h>
				1114	</PRE>
				1115
				1116	and must be linked explicitly with the panels library using an
				1117	<CODE>-lpanel</CODE> argument. Note that they must also link the
				1118	<CODE>ncurses</CODE> library with <CODE>-lncurses</CODE>. Many linkers
				1119	are two-pass and will accept either order, but it is still good practice
				1120	to put <CODE>-lpanel</CODE> first and <CODE>-lncurses</CODE> second.
				1121
				1122	<H2><A NAME="poverview">Overview of Panels</A></H2>
				1123
				1124	A panel object is a window that is implicitly treated as part of a
				1125	<DFN>deck</DFN> including all other panel objects. The deck has an implicit
				1126	bottom-to-top visibility order. The panels library includes an update
				1127	function (analogous to <CODE>refresh()</CODE>) that displays all panels in the
				1128	deck in the proper order to resolve overlaps. The standard window,
				1129	<CODE>stdscr</CODE>, is considered below all panels. <P>
				1130
				1131	Details on the panels functions are available in the man pages. We'll just
				1132	hit the highlights here. <P>
				1133
				1134	You create a panel from a window by calling <CODE>new_panel()</CODE> on a
				1135	window pointer. It then becomes the top of the deck. The panel's window
				1136	is available as the value of <CODE>panel_window()</CODE> called with the
				1137	panel pointer as argument.<P>
				1138
				1139	You can delete a panel (removing it from the deck) with <CODE>del_panel</CODE>.
				1140	This will not deallocate the associated window; you have to do that yourself.
				1141
				1142	You can replace a panel's window with a different window by calling
				1143	<CODE>replace_window</CODE>. The new window may be of different size;
				1144	the panel code will re-compute all overlaps. This operation doesn't
				1145	change the panel's position in the deck. <P>
				1146
				1147	To move a panel's window, use <CODE>move_panel()</CODE>. The
				1148	<CODE>mvwin()</CODE> function on the panel's window isn't sufficient because it
				1149	doesn't update the panels library's representation of where the windows are.
				1150	This operation leaves the panel's depth, contents, and size unchanged. <P>
				1151
				1152	Two functions (<CODE>top_panel()</CODE>, <CODE>bottom_panel()</CODE>) are
				1153	provided for rearranging the deck. The first pops its argument window to the
				1154	top of the deck; the second sends it to the bottom. Either operation leaves
				1155	the panel's screen location, contents, and size unchanged. <P>
				1156
				1157	The function <CODE>update_panels()</CODE> does all the
				1158	<CODE>wnoutrefresh()</CODE> calls needed to prepare for
				1159	<CODE>doupdate()</CODE> (which you must call yourself, afterwards). <P>
				1160
				1161	Typically, you will want to call <CODE>update_panels()</CODE> and
				1162	<CODE>doupdate()</CODE> just before accepting command input, once in each cycle
				1163	of interaction with the user. If you call <CODE>update_panels()</CODE> after
				1164	each and every panel write, you'll generate a lot of unnecessary refresh
				1165	activity and screen flicker.
				1166
				1167	<H2><A NAME="pstdscr">Panels, Input, and the Standard Screen</A></H2>
				1168
				1169	You shouldn't mix <CODE>wnoutrefresh()</CODE> or <CODE>wrefresh()</CODE>
				1170	operations with panels code; this will work only if the argument window
				1171	is either in the top panel or unobscured by any other panels. <P>
				1172
				1173	The <CODE>stsdcr</CODE> window is a special case. It is considered below all
				1174	panels. Because changes to panels may obscure parts of <CODE>stdscr</CODE>,
				1175	though, you should call <CODE>update_panels()</CODE> before
				1176	<CODE>doupdate()</CODE> even when you only change <CODE>stdscr</CODE>. <P>
				1177
				1178	Note that <CODE>wgetch</CODE> automatically calls <CODE>wrefresh</CODE>.
				1179	Therefore, before requesting input from a panel window, you need to be sure
				1180	that the panel is totally unobscured. <P>
				1181
				1182	There is presently no way to display changes to one obscured panel without
				1183	repainting all panels.
				1184
				1185	<H2><A NAME="hiding">Hiding Panels</A></H2>
				1186
				1187	It's possible to remove a panel from the deck temporarily; use
				1188	<CODE>hide_panel</CODE> for this. Use <CODE>show_panel()</CODE> to render it
				1189	visible again. The predicate function <CODE>panel_hidden</CODE>
				1190	tests whether or not a panel is hidden. <P>
				1191
				1192	The <CODE>panel_update</CODE> code ignores hidden panels. You cannot do
				1193	<CODE>top_panel()</CODE> or <CODE>bottom_panel</CODE> on a hidden panel().
				1194	Other panels operations are applicable.
				1195
				1196	<H2><A NAME="pmisc">Miscellaneous Other Facilities</A></H2>
				1197
				1198	It's possible to navigate the deck using the functions
				1199	<CODE>panel_above()</CODE> and <CODE>panel_below</CODE>. Handed a panel
				1200	pointer, they return the panel above or below that panel. Handed
				1201	<CODE>NULL</CODE>, they return the bottom-most or top-most panel. <P>
				1202
				1203	Every panel has an associated user pointer, not used by the panel code, to
				1204	which you can attach application data. See the man page documentation
				1205	of <CODE>set_panel_userptr()</CODE> and <CODE>panel_userptr</CODE> for
				1206	details.
				1207
				1208	<H1><A NAME="menu">The Menu Library</A></H1>
				1209
				1210	A menu is a screen display that assists the user to choose some subset
				1211	of a given set of items. The <CODE>menu</CODE> library is a curses
				1212	extension that supports easy programming of menu hierarchies with a
				1213	uniform but flexible interface. <P>
				1214
				1215	The <CODE>menu</CODE> library first appeared in AT&T System V. The
				1216	version documented here is the <CODE>menu</CODE> code distributed
				1217	with <CODE>ncurses</CODE>.
				1218
				1219	<H2><A NAME="mcompile">Compiling With the menu Library</A></H2>
				1220
				1221	Your menu-using modules must import the menu library declarations with
				1222
				1223	<PRE>
				1224	#include <menu.h>
				1225	</PRE>
				1226
				1227	and must be linked explicitly with the menus library using an
				1228	<CODE>-lmenu</CODE> argument. Note that they must also link the
				1229	<CODE>ncurses</CODE> library with <CODE>-lncurses</CODE>. Many linkers
				1230	are two-pass and will accept either order, but it is still good practice
				1231	to put <CODE>-lmenu</CODE> first and <CODE>-lncurses</CODE> second.
				1232
				1233	<H2><A NAME="moverview">Overview of Menus</A></H2>
				1234
				1235	The menus created by this library consist of collections of
				1236	<DFN>items</DFN> including a name string part and a description string
				1237	part. To make menus, you create groups of these items and connect
				1238	them with menu frame objects. <P>
				1239
				1240	The menu can then by <DFN>posted</DFN>, that is written to an
				1241	associated window. Actually, each menu has two associated windows; a
				1242	containing window in which the programmer can scribble titles or
				1243	borders, and a subwindow in which the menu items proper are displayed.
				1244	If this subwindow is too small to display all the items, it will be a
				1245	scrollable viewport on the collection of items. <P>
				1246
				1247	A menu may also be <DFN>unposted</DFN> (that is, undisplayed), and finally
				1248	freed to make the storage associated with it and its items available for
				1249	re-use. <P>
				1250
				1251	The general flow of control of a menu program looks like this:
				1252
				1253	<OL>
				1254	<LI>Initialize <CODE>curses</CODE>.
				1255	<LI>Create the menu items, using <CODE>new_item()</CODE>.
				1256	<LI>Create the menu using <CODE>new_menu()</CODE>.
				1257	<LI>Post the menu using <CODE>post_menu()</CODE>.
				1258	<LI>Refresh the screen.
				1259	<LI>Process user requests via an input loop.
				1260	<LI>Unpost the menu using <CODE>unpost_menu()</CODE>.
				1261	<LI>Free the menu, using <CODE>free_menu()</CODE>.
				1262	<LI>Free the items using <CODE>free_item()</CODE>.
				1263	<LI>Terminate <CODE>curses</CODE>.
				1264	</OL>
				1265
				1266	<H2><A NAME="mselect">Selecting items</A></H2>
				1267
				1268	Menus may be multi-valued or (the default) single-valued (see the manual
				1269	page <CODE>menu_opts(3x)</CODE> to see how to change the default).
				1270	Both types always have a <DFN>current item</DFN>. <P>
				1271
				1272	From a single-valued menu you can read the selected value simply by looking
				1273	at the current item. From a multi-valued menu, you get the selected set
				1274	by looping through the items applying the <CODE>item_value()</CODE>
				1275	predicate function. Your menu-processing code can use the function
				1276	<CODE>set_item_value()</CODE> to flag the items in the select set. <P>
				1277
				1278	Menu items can be made unselectable using <CODE>set_item_opts()</CODE>
				1279	or <CODE>item_opts_off()</CODE> with the <CODE>O_SELECTABLE</CODE>
				1280	argument. This is the only option so far defined for menus, but it
				1281	is good practice to code as though other option bits might be on.
				1282
				1283	<H2><A NAME="mdisplay">Menu Display</A></H2>
				1284
				1285	The menu library calculates a minimum display size for your window, based
				1286	on the following variables:
				1287
				1288	<UL>
				1289	<LI>The number and maximum length of the menu items
				1290	<LI>Whether the O_ROWMAJOR option is enabled
				1291	<LI>Whether display of descriptions is enabled
				1292	<LI>Whatever menu format may have been set by the programmer
				1293	<LI>The length of the menu mark string used for highlighting selected items
				1294	</UL>
				1295
				1296	The function <CODE>set_menu_format()</CODE> allows you to set the
				1297	maximum size of the viewport or <DFN>menu page</DFN> that will be used
				1298	to display menu items. You can retrieve any format associated with a
				1299	menu with <CODE>menu_format()</CODE>. The default format is rows=16,
				1300	columns=1. <P>
				1301
				1302	The actual menu page may be smaller than the format size. This depends
				1303	on the item number and size and whether O_ROWMAJOR is on. This option
				1304	(on by default) causes menu items to be displayed in a `raster-scan'
				1305	pattern, so that if more than one item will fit horizontally the first
				1306	couple of items are side-by-side in the top row. The alternative is
				1307	column-major display, which tries to put the first several items in
				1308	the first column. <P>
				1309
				1310	As mentioned above, a menu format not large enough to allow all items to fit
				1311	on-screen will result in a menu display that is vertically scrollable. <P>
				1312	You can scroll it with requests to the menu driver, which will be described
				1313	in the section on <A HREF="#minput">menu input handling</A>. <P>
				1314
				1315	Each menu has a <DFN>mark string</DFN> used to visually tag selected items;
				1316	see the <CODE>menu_mark(3x)</CODE> manual page for details. The mark
				1317	string length also influences the menu page size. <P>
				1318
				1319	The function <CODE>scale_menu()</CODE> returns the minimum display size
				1320	that the menu code computes from all these factors.
				1321
				1322	There are other menu display attributes including a select attribute,
				1323	an attribute for selectable items, an attribute for unselectable items,
				1324	and a pad character used to separate item name text from description
				1325	text. These have reasonable defaults which the library allows you to
				1326	change (see the <CODE>menu_attribs(3x)</CODE> manual page.
				1327
				1328	<H2><A NAME="mwindows">Menu Windows</A></H2>
				1329
				1330	Each menu has, as mentioned previously, a pair of associated windows.
				1331	Both these windows are painted when the menu is posted and erased when
				1332	the menu is unposted. <P>
				1333
				1334	The outer or frame window is not otherwise touched by the menu
				1335	routines. It exists so the programmer can associate a title, a
				1336	border, or perhaps help text with the menu and have it properly
				1337	refreshed or erased at post/unpost time. The inner window or
				1338	<DFN>subwindow</DFN> is where the current menu page is displayed. <P>
				1339
				1340	By default, both windows are <CODE>stdscr</CODE>. You can set them with the
				1341	functions in <CODE>menu_win(3x)</CODE>. <P>
				1342
				1343	When you call <CODE>post_menu()</CODE>, you write the menu to its
				1344	subwindow. When you call <CODE>unpost_menu()</CODE>, you erase the
				1345	subwindow, However, neither of these actually modifies the screen. To
				1346	do that, call <CODE>wrefresh()</CODE> or some equivalent.
				1347
				1348	<H2><A NAME="minput">Processing Menu Input</A></H2>
				1349
				1350	The main loop of your menu-processing code should call
				1351	<CODE>menu_driver()</CODE> repeatedly. The first argument of this routine
				1352	is a menu pointer; the second is a menu command code. You should write an
				1353	input-fetching routine that maps input characters to menu command codes, and
				1354	pass its output to <CODE>menu_driver()</CODE>. The menu command codes are
				1355	fully documented in <CODE>menu_driver(3x)</CODE>. <P>
				1356
				1357	The simplest group of command codes is <CODE>REQ_NEXT_ITEM</CODE>,
				1358	<CODE>REQ_PREV_ITEM</CODE>, <CODE>REQ_FIRST_ITEM</CODE>,
				1359	<CODE>REQ_LAST_ITEM</CODE>, <CODE>REQ_UP_ITEM</CODE>,
				1360	<CODE>REQ_DOWN_ITEM</CODE>, <CODE>REQ_LEFT_ITEM</CODE>,
				1361	<CODE>REQ_RIGHT_ITEM</CODE>. These change the currently selected
				1362	item. These requests may cause scrolling of the menu page if it only
				1363	partially displayed. <P>
				1364
				1365	There are explicit requests for scrolling which also change the
				1366	current item (because the select location does not change, but the
				1367	item there does). These are <CODE>REQ_SCR_DLINE</CODE>,
				1368	<CODE>REQ_SCR_ULINE</CODE>, <CODE>REQ_SCR_DPAGE</CODE>, and
				1369	<CODE>REQ_SCR_UPAGE</CODE>. <P>
				1370
				1371	The <CODE>REQ_TOGGLE_ITEM</CODE> selects or deselects the current item.
				1372	It is for use in multi-valued menus; if you use it with <CODE>O_ONEVALUE</CODE>
				1373	on, you'll get an error return (<CODE>E_REQUEST_DENIED</CODE>). <P>
				1374
				1375	Each menu has an associated pattern buffer. The
				1376	<CODE>menu_driver()</CODE> logic tries to accumulate printable ASCII
				1377	characters passed in in that buffer; when it matches a prefix of an
				1378	item name, that item (or the next matching item) is selected. If
				1379	appending a character yields no new match, that character is deleted
				1380	from the pattern buffer, and <CODE>menu_driver()</CODE> returns
				1381	<CODE>E_NO_MATCH</CODE>. <P>
				1382
				1383	Some requests change the pattern buffer directly:
				1384	<CODE>REQ_CLEAR_PATTERN</CODE>, <CODE>REQ_BACK_PATTERN</CODE>,
				1385	<CODE>REQ_NEXT_MATCH</CODE>, <CODE>REQ_PREV_MATCH</CODE>. The latter
				1386	two are useful when pattern buffer input matches more than one item
				1387	in a multi-valued menu. <P>
				1388
				1389	Each successful scroll or item navigation request clears the pattern
				1390	buffer. It is also possible to set the pattern buffer explicitly
				1391	with <CODE>set_menu_pattern()</CODE>. <P>
				1392
				1393	Finally, menu driver requests above the constant <CODE>MAX_COMMAND</CODE>
				1394	are considered application-specific commands. The <CODE>menu_driver()</CODE>
				1395	code ignores them and returns <CODE>E_UNKNOWN_COMMAND</CODE>.
				1396
				1397	<H2><A NAME="mmisc">Miscellaneous Other Features</A></H2>
				1398
				1399	Various menu options can affect the processing and visual appearance
				1400	and input processing of menus. See <CODE>menu_opts(3x) for
				1401	details.</CODE> <P>
				1402
				1403	It is possible to change the current item from application code; this
				1404	is useful if you want to write your own navigation requests. It is
				1405	also possible to explicitly set the top row of the menu display. See
				1406	<CODE>mitem_current(3x)</CODE>.
				1407
				1408	If your application needs to change the menu subwindow cursor for
				1409	any reason, <CODE>pos_menu_cursor()</CODE> will restore it to the
				1410	correct location for continuing menu driver processing. <P>
				1411
				1412	It is possible to set hooks to be called at menu initialization and
				1413	wrapup time, and whenever the selected item changes. See
				1414	<CODE>menu_hook(3x)</CODE>. <P>
				1415
				1416	Each item, and each menu, has an associated user pointer on which you
				1417	can hang application data. See <CODE>mitem_userptr(3x)</CODE> and
				1418	<CODE>menu_userptr(3x)</CODE>.
				1419
				1420	<H1><A NAME="form">The Forms Library</A></H1>
				1421
				1422	The <CODE>form</CODE> library is a curses extension that supports easy
				1423	programming of on-screen forms for data entry and program control. <P>
				1424
				1425	The <CODE>form</CODE> library first appeared in AT&T System V. The
				1426	version documented here is the <CODE>form</CODE> code distributed
				1427	with <CODE>ncurses</CODE>.
				1428
				1429	<H2><A NAME="fcompile">Compiling With the form Library</A></H2>
				1430
				1431	Your form-using modules must import the form library declarations with
				1432
				1433	<PRE>
				1434	#include <form.h>
				1435	</PRE>
				1436
				1437	and must be linked explicitly with the forms library using an
				1438	<CODE>-lform</CODE> argument. Note that they must also link the
				1439	<CODE>ncurses</CODE> library with <CODE>-lncurses</CODE>. Many linkers
				1440	are two-pass and will accept either order, but it is still good practice
				1441	to put <CODE>-lform</CODE> first and <CODE>-lncurses</CODE> second.
				1442
				1443	<H2><A NAME="foverview">Overview of Forms</A></H2>
				1444
				1445	A form is a collection of fields; each field may be either a label
				1446	(explanatory text) or a data-entry location. Long forms may be
				1447	segmented into pages; each entry to a new page clears the screen. <P>
				1448	To make forms, you create groups of fields and connect them with form
				1449	frame objects; the form library makes this relatively simple. <P>
				1450
				1451	Once defined, a form can be <DFN>posted</DFN>, that is written to an
				1452	associated window. Actually, each form has two associated windows; a
				1453	containing window in which the programmer can scribble titles or
				1454	borders, and a subwindow in which the form fields proper are displayed. <P>
				1455
				1456	As the form user fills out the posted form, navigation and editing
				1457	keys support movement between fields, editing keys support modifying
				1458	field, and plain text adds to or changes data in a current field. The
				1459	form library allows you (the forms designer) to bind each navigation
				1460	and editing key to any keystroke accepted by <CODE>curses</CODE>
				1461
				1462	Fields may have validation conditions on them, so that they check input
				1463	data for type and value. The form library supplies a rich set of
				1464	pre-defined field types, and makes it relatively easy to define new ones. <P>
				1465
				1466	Once its transaction is completed (or aborted), a form may be
				1467	<DFN>unposted</DFN> (that is, undisplayed), and finally freed to make
				1468	the storage associated with it and its items available for re-use. <P>
				1469
				1470	The general flow of control of a form program looks like this:
				1471
				1472	<OL>
				1473	<LI>Initialize <CODE>curses</CODE>.
				1474	<LI>Create the form fields, using <CODE>new_field()</CODE>.
				1475	<LI>Create the form using <CODE>new_form()</CODE>.
				1476	<LI>Post the form using <CODE>post_form()</CODE>.
				1477	<LI>Refresh the screen.
				1478	<LI>Process user requests via an input loop.
				1479	<LI>Unpost the form using <CODE>unpost_form()</CODE>.
				1480	<LI>Free the form, using <CODE>free_form()</CODE>.
				1481	<LI>Free the fields using <CODE>free_field()</CODE>.
				1482	<LI>Terminate <CODE>curses</CODE>.
				1483	</OL>
				1484
				1485	Note that this looks much like a menu program; the form library handles
				1486	tasks which are in many ways similar, and its interface was obviously
				1487	designed to resemble that of the <A HREF="#menu">menu library</A>
				1488	wherever possible. <P>
				1489
				1490	In forms programs, however, the `process user requests' is somewhat more
				1491	complicated than for menus. Besides menu-like navigation operations,
				1492	the menu driver loop has to support field editing and data validation.
				1493
				1494	<H2><A NAME="fcreate">Creating and Freeing Fields and Forms</A></H2>
				1495
				1496	The basic function for creating fields is <CODE>new_field()</CODE>:
				1497
				1498	<PRE>
				1499	FIELD new_field(int height, int width, / new field size */
				1500	int top, int left, /* upper left corner */
				1501	int offscreen, /* number of offscreen rows */
				1502	int nbuf); /* number of working buffers */
				1503	</PRE>
				1504
				1505	Menu items always occupy a single row, but forms fields may have
				1506	multiple rows. So <CODE>new_field()</CODE> requires you to specify a
				1507	width and height (the first two arguments, which mist both be greater
				1508	than zero). <P>
				1509
				1510	You must also specify the location of the field's upper left corner on
				1511	the screen (the third and fourth arguments, which must be zero or
				1512	greater). Note that these coordinates are relative to the form
				1513	subwindow, which will coincide with <CODE>stdscr</CODE> by default but
				1514	need not be <CODE>stdscr</CODE> if you've done an explicit
				1515	<CODE>set_form_win()</CODE> call. <P>
				1516
				1517	The fifth argument allows you to specify a number of off-screen rows. If
				1518	this is zero, the entire field will always be displayed. If it is
				1519	nonzero, the form will be scrollable, with only one screen-full (initially
				1520	the top part) displayed at any given time. If you make a field dynamic
				1521	and grow it so it will no longer fit on the screen, the form will become
				1522	scrollable even if the <CODE>offscreen</CODE> argument was initially zero. <P>
				1523
				1524	The forms library allocates one working buffer per field; the size of
				1525	each buffer is <CODE>((height + offscreen)*width + 1</CODE>, one character
				1526	for each position in the field plus a NUL terminator. The sixth
				1527	argument is the number of additional data buffers to allocate for the
				1528	field; your application can use them for its own purposes.
				1529
				1530	<PRE>
				1531	FIELD dup_field(FIELD field, /* field to copy */
				1532	int top, int left); /* location of new copy */
				1533	</PRE>
				1534
				1535	The function <CODE>dup_field()</CODE> duplicates an existing field at a
				1536	new location. Size and buffering information are copied; some
				1537	attribute flags and status bits are not (see the
				1538	<CODE>form_field_new(3X)</CODE> for details).
				1539
				1540	<PRE>
				1541	FIELD link_field(FIELD field, /* field to copy */
				1542	int top, int left); /* location of new copy */
				1543	</PRE>
				1544
				1545	The function <CODE>link_field()</CODE> also duplicates an existing field
				1546	at a new location. The difference from <CODE>dup_field()</CODE> is that
				1547	it arranges for the new field's buffer to be shared with the old one. <P>
				1548
				1549	Besides the obvious use in making a field editable from two different
				1550	form pages, linked fields give you a way to hack in dynamic labels. If
				1551	you declare several fields linked to an original, and then make them
				1552	inactive, changes from the original will still be propagated to the
				1553	linked fields. <P>
				1554
				1555	As with duplicated fields, linked fields have attribute bits separate
				1556	from the original. <P>
				1557
				1558	As you might guess, all these field-allocations return <CODE>NULL</CODE> if
				1559	the field allocation is not possible due to an out-of-memory error or
				1560	out-of-bounds arguments. <P>
				1561
				1562	To connect fields to a form, use
				1563
				1564	<PRE>
				1565	FORM new_form(FIELD *fields);
				1566	</PRE>
				1567
				1568	This function expects to see a NULL-terminated array of field pointers.
				1569	Said fields are connected to a newly-allocated form object; its address
				1570	is returned (or else NULL if the allocation fails). <P>
				1571
				1572	Note that <CODE>new_field()</CODE> does <EM>not</EM> copy the pointer array
				1573	into private storage; if you modify the contents of the pointer array
				1574	during forms processing, all manner of bizarre things might happen. Also
				1575	note that any given field may only be connected to one form. <P>
				1576
				1577	The functions <CODE>free_field()</CODE> and <CODE>free_form</CODE> are available
				1578	to free field and form objects. It is an error to attempt to free a field
				1579	connected to a form, but not vice-versa; thus, you will generally free
				1580	your form objects first.
				1581
				1582	<H2><A NAME="fattributes">Fetching and Changing Field Attributes</A></H2>
				1583
				1584	Each form field has a number of location and size attributes
				1585	associated with it. There are other field attributes used to control
				1586	display and editing of the field. Some (for example, the <CODE>O_STATIC</CODE> bit)
				1587	involve sufficient complications to be covered in sections of their own
				1588	later on. We cover the functions used to get and set several basic
				1589	attributes here. <P>
				1590
				1591	When a field is created, the attributes not specified by the
				1592	<CODE>new_field</CODE> function are copied from an invisible system
				1593	default field. In attribute-setting and -fetching functions, the
				1594	argument NULL is taken to mean this field. Changes to it persist
				1595	as defaults until your forms application terminates.
				1596
				1597	<H3><A NAME="fsizes">Fetching Size and Location Data</A></H3>
				1598
				1599	You can retrieve field sizes and locations through:
				1600
				1601	<PRE>
				1602	int field_info(FIELD field, / field from which to fetch */
				1603	int height, int width, /* field size */
				1604	int top, int left, /* upper left corner */
				1605	int offscreen, / number of offscreen rows */
				1606	int nbuf); / number of working buffers */
				1607	</PRE>
				1608
				1609	This function is a sort of inverse of <CODE>new_field()</CODE>; instead of
				1610	setting size and location attributes of a new field, it fetches them
				1611	from an existing one.
				1612
				1613	<H3><A NAME="flocation">Changing the Field Location</A></H3>
				1614
				1615	It is possible to move a field's location on the screen:
				1616
				1617	<PRE>
				1618	int move_field(FIELD field, / field to alter */
				1619	int top, int left); /* new upper-left corner */
				1620	</PRE>
				1621
				1622	You can, of course. query the current location through <CODE>field_info()</CODE>.
				1623
				1624	<H3><A NAME="fjust">The Justification Attribute</A></H3>
				1625
				1626	One-line fields may be unjustified, justified right, justified left,
				1627	or centered. Here is how you manipulate this attribute:
				1628
				1629	<PRE>
				1630	int set_field_just(FIELD field, / field to alter */
				1631	int justmode); /* mode to set */
				1632
				1633	int field_just(FIELD field); / fetch mode of field */
				1634	</PRE>
				1635
				1636	The mode values accepted and returned by this functions are
				1637	preprocessor macros <CODE>NO_JUSTIFICATION</CODE>, <CODE>JUSTIFY_RIGHT</CODE>,
				1638	<CODE>JUSTIFY_LEFT</CODE>, or <CODE>JUSTIFY_CENTER</CODE>.
				1639
				1640	<H3><A NAME="fdispatts">Field Display Attributes</A></H3>
				1641
				1642	For each field, you can set a foreground attribute for entered
				1643	characters, a background attribute for the entire field, and a pad
				1644	character for the unfilled portion of the field. You can also
				1645	control pagination of the form. <P>
				1646
				1647	This group of four field attributes controls the visual appearance
				1648	of the field on the screen, without affecting in any way the data
				1649	in the field buffer.
				1650
				1651	<PRE>
				1652	int set_field_fore(FIELD field, / field to alter */
				1653	chtype attr); /* attribute to set */
				1654
				1655	chtype field_fore(FIELD field); / field to query */
				1656
				1657	int set_field_back(FIELD field, / field to alter */
				1658	chtype attr); /* attribute to set */
				1659
				1660	chtype field_back(FIELD field); / field to query */
				1661
				1662	int set_field_pad(FIELD field, / field to alter */
				1663	int pad); /* pad character to set */
				1664
				1665	chtype field_pad(FIELD *field);
				1666
				1667	int set_new_page(FIELD field, / field to alter */
				1668	int flag); /* TRUE to force new page */
				1669
				1670	chtype new_page(FIELD field); / field to query */
				1671	</PRE>
				1672
				1673	The attributes set and returned by the first four functions are normal
				1674	<CODE>curses(3x)</CODE> display attribute values (<CODE>A_STANDOUT</CODE>,
				1675	<CODE>A_BOLD</CODE>, <CODE>A_REVERSE</CODE> etc).
				1676
				1677	The page bit of a field controls whether it is displayed at the start of
				1678	a new form screen.
				1679
				1680	<H3><A NAME="foptions">Field Option Bits</A></H3>
				1681
				1682	There is also a large collection of field option bits you can set to control
				1683	various aspects of forms processing. You can manipulate them with these
				1684	functions:
				1685
				1686	<PRE>
				1687	int set_field_opts(FIELD field, / field to alter */
				1688	int attr); /* attribute to set */
				1689
				1690	int field_opts_on(FIELD field, / field to alter */
				1691	int attr); /* attributes to turn on */
				1692
				1693	int field_opts_off(FIELD field, / field to alter */
				1694	int attr); /* attributes to turn off */
				1695
				1696	int field_opts(FIELD field); / field to query */
				1697	</PRE>
				1698
				1699	By default, all options are on. Here are the available option bits:
				1700	<DL>
				1701	<DT> O_VISIBLE
				1702	<DD> Controls whether the field is visible on the screen. Can be used
				1703	during form processing to hide or pop up fields depending on the value
				1704	of parent fields.
				1705	<DT> O_ACTIVE
				1706	<DD> Controls whether the field is active during forms processing (i.e.
				1707	visited by form navigation keys). Can be used to make labels or derived
				1708	fields with buffer values alterable by the forms application, not the user.
				1709	<DT> O_PUBLIC
				1710	<DD> Controls whether data is displayed during field entry. If this option is
				1711	turned off on a field, the library will accept and edit data in that field,
				1712	but it will not be displayed and the visible field cursor will not move.
				1713	You can turn off the O_PUBLIC bit to define password fields.
				1714	<DT> O_EDIT
				1715	<DD> Controls whether the field's data can be modified. When this option is
				1716	off, all editing requests except <CODE>REQ_PREV_CHOICE</CODE> and
				1717	<CODE>REQ_NEXT_CHOICE</CODE> will fail. Such read-only fields may be useful for
				1718	help messages.
				1719	<DT> O_WRAP
				1720	<DD> Controls word-wrapping in multi-line fields. Normally, when any
				1721	character of a (blank-separated) word reaches the end of the current line, the
				1722	entire word is wrapped to the next line (assuming there is one). When this
				1723	option is off, the word will be split across the line break.
				1724	<DT> O_BLANK
				1725	<DD> Controls field blanking. When this option is on, entering a character at
				1726	the first field position erases the entire field (except for the just-entered
				1727	character).
				1728	<DT> O_AUTOSKIP
				1729	<DD> Controls automatic skip to next field when this one fills. Normally,
				1730	when the forms user tries to type more data into a field than will fit,
				1731	the editing location jumps to next field. When this option is off, the
				1732	user's cursor will hang at the end of the field. This option is ignored
				1733	in dynamic fields that have not reached their size limit.
				1734	<DT> O_NULLOK
				1735	<DD> Controls whether <A HREF="#fvalidation">validation</A> is applied to
				1736	blank fields. Normally, it is not; the user can leave a field blank
				1737	without invoking the usual validation check on exit. If this option is
				1738	off on a field, exit from it will invoke a validation check.
				1739	<DT> O_PASSOK
				1740	<DD> Controls whether validation occurs on every exit, or only after
				1741	the field is modified. Normally the latter is true. Setting O_PASSOK
				1742	may be useful if your field's validation function may change during
				1743	forms processing.
				1744	<DT> O_STATIC
				1745	<DD> Controls whether the field is fixed to its initial dimensions. If you
				1746	turn this off, the field becomes <A HREF="#fdynamic">dynamic</A> and will
				1747	stretch to fit entered data.
				1748	</DL>
				1749
				1750	A field's options cannot be changed while the field is currently selected.
				1751	However, options may be changed on posted fields that are not current. <P>
				1752
				1753	The option values are bit-masks and can be composed with logical-or in
				1754	the obvious way.
				1755
				1756	<H2><A NAME="fstatus">Field Status</A></H2>
				1757
				1758	Every field has a status flag, which is set to FALSE when the field is
				1759	created and TRUE when the value in field buffer 0 changes. This flag can
				1760	be queried and set directly:
				1761
				1762	<PRE>
				1763	int set_field_status(FIELD field, / field to alter */
				1764	int status); /* mode to set */
				1765
				1766	int field_status(FIELD field); / fetch mode of field */
				1767	</PRE>
				1768
				1769	Setting this flag under program control can be useful if you use the same
				1770	form repeatedly, looking for modified fields each time. <P>
				1771
				1772	Calling <CODE>field_status()</CODE> on a field not currently selected
				1773	for input will return a correct value. Calling <CODE>field_status()</CODE> on a
				1774	field that is currently selected for input may not necessarily give a
				1775	correct field status value, because entered data isn't necessarily copied to
				1776	buffer zero before the exit validation check.
				1777
				1778	To guarantee that the returned status value reflects reality, call
				1779	<CODE>field_status()</CODE> either (1) in the field's exit validation check
				1780	routine, (2) from the field's or form's initialization or termination
				1781	hooks, or (3) just after a <CODE>REQ_VALIDATION</CODE> request has been
				1782	processed by the forms driver.
				1783
				1784	<H2><A NAME="fuser">Field User Pointer</A></H2>
				1785
				1786	Each field structure contains one character pointer slot that is not used
				1787	by the forms library. It is intended to be used by applications to store
				1788	private per-field data. You can manipulate it with:
				1789
				1790	<PRE>
				1791	int set_field_userptr(FIELD field, / field to alter */
				1792	char userptr); / mode to set */
				1793
				1794	char field_userptr(FIELD field); /* fetch mode of field */
				1795	</PRE>
				1796
				1797	(Properly, this user pointer field ought to have <CODE>(void *)</CODE> type.
				1798	The <CODE>(char *)</CODE> type is retained for System V compatibility.) <P>
				1799
				1800	It is valid to set the user pointer of the default field (with a
				1801	<CODE>set_field_userptr()</CODE> call passed a NULL field pointer.)
				1802	When a new field is created, the default-field user pointer is copied
				1803	to initialize the new field's user pointer.
				1804
				1805	<H2><A NAME="fdynamic">Variable-Sized Fields</A></H2>
				1806
				1807	Normally, a field is fixed at the size specified for it at creation
				1808	time. If, however, you turn off its O_STATIC bit, it becomes
				1809	<DFN>dynamic</DFN> and will automatically resize itself to accommodate
				1810	data as it is entered. If the field has extra buffers associated with it,
				1811	they will grow right along with the main input buffer. <P>
				1812
				1813	A one-line dynamic field will have a fixed height (1) but variable
				1814	width, scrolling horizontally to display data within the field area as
				1815	originally dimensioned and located. A multi-line dynamic field will
				1816	have a fixed width, but variable height (number of rows), scrolling
				1817	vertically to display data within the field area as originally
				1818	dimensioned and located. <P>
				1819
				1820	Normally, a dynamic field is allowed to grow without limit. But it is
				1821	possible to set an upper limit on the size of a dynamic field. You do
				1822	it with this function:
				1823
				1824	<PRE>
				1825	int set_max_field(FIELD field, / field to alter (may not be NULL) */
				1826	int max_size); /* upper limit on field size */
				1827	</PRE>
				1828
				1829	If the field is one-line, <CODE>max_size</CODE> is taken to be a column size
				1830	limit; if it is multi-line, it is taken to be a line size limit. To disable
				1831	any limit, use an argument of zero. The growth limit can be changed whether
				1832	or not the O_STATIC bit is on, but has no effect until it is. <P>
				1833
				1834	The following properties of a field change when it becomes dynamic:
				1835
				1836	<UL>
				1837	<LI>If there is no growth limit, there is no final position of the field;
				1838	therefore <CODE>O_AUTOSKIP</CODE> and <CODE>O_NL_OVERLOAD</CODE> are ignored.
				1839	<LI>Field justification will be ignored (though whatever justification is
				1840	set up will be retained internally and can be queried).
				1841	<LI>The <CODE>dup_field()</CODE> and <CODE>link_field()</CODE> calls copy
				1842	dynamic-buffer sizes. If the <CODE>O_STATIC</CODE> option is set on one of a
				1843	collection of links, buffer resizing will occur only when the field is
				1844	edited through that link.
				1845	<LI>The call <CODE>field_info()</CODE> will retrieve the original static size of
				1846	the field; use <CODE>dynamic_field_info()</CODE> to get the actual dynamic size.
				1847	</UL>
				1848
				1849	<H2><A NAME="fvalidation">Field Validation</A></H2>
				1850
				1851	By default, a field will accept any data that will fit in its input buffer.
				1852	However, it is possible to attach a validation type to a field. If you do
				1853	this, any attempt to leave the field while it contains data that doesn't
				1854	match the validation type will fail. Some validation types also have a
				1855	character-validity check for each time a character is entered in the field. <P>
				1856
				1857	A field's validation check (if any) is not called when
				1858	<CODE>set_field_buffer()</CODE> modifies the input buffer, nor when that buffer
				1859	is changed through a linked field. <P>
				1860
				1861	The <CODE>form</CODE> library provides a rich set of pre-defined validation
				1862	types, and gives you the capability to define custom ones of your own. You
				1863	can examine and change field validation attributes with the following
				1864	functions:
				1865
				1866	<PRE>
				1867	int set_field_type(FIELD field, / field to alter */
				1868	FIELDTYPE ftype, / type to associate */
				1869	...); /* additional arguments*/
				1870
				1871	FIELDTYPE field_type(FIELD field); /* field to query */
				1872	</PRE>
				1873
				1874	The validation type of a field is considered an attribute of the field. As
				1875	with other field attributes, Also, doing <CODE>set_field_type()</CODE> with a
				1876	<CODE>NULL</CODE> field default will change the system default for validation of
				1877	newly-created fields. <P>
				1878
				1879	Here are the pre-defined validation types:
				1880
				1881	<H3><A NAME="ftype_alpha">TYPE_ALPHA</A></H3>
				1882
				1883	This field type accepts alphabetic data; no blanks, no digits, no special
				1884	characters (this is checked at character-entry time). It is set up with:
				1885
				1886	<PRE>
				1887	int set_field_type(FIELD field, / field to alter */
				1888	TYPE_ALPHA, /* type to associate */
				1889	int width); /* maximum width of field */
				1890	</PRE>
				1891
				1892	The <CODE>width</CODE> argument sets a minimum width of data. Typically
				1893	you'll want to set this to the field width; if it's greater than the
				1894	field width, the validation check will always fail. A minimum width
				1895	of zero makes field completion optional.
				1896
				1897	<H3><A NAME="ftype_alnum">TYPE_ALNUM</A></H3>
				1898
				1899	This field type accepts alphabetic data and digits; no blanks, no special
				1900	characters (this is checked at character-entry time). It is set up with:
				1901
				1902	<PRE>
				1903	int set_field_type(FIELD field, / field to alter */
				1904	TYPE_ALNUM, /* type to associate */
				1905	int width); /* maximum width of field */
				1906	</PRE>
				1907
				1908	The <CODE>width</CODE> argument sets a minimum width of data. As with
				1909	TYPE_ALPHA, typically you'll want to set this to the field width; if it's
				1910	greater than the field width, the validation check will always fail. A
				1911	minimum width of zero makes field completion optional.
				1912
				1913	<H3><A NAME="ftype_enum">TYPE_ENUM</A></H3>
				1914
				1915	This type allows you to restrict a field's values to be among a specified
				1916	set of string values (for example, the two-letter postal codes for U.S.
				1917	states). It is set up with:
				1918
				1919	<PRE>
				1920	int set_field_type(FIELD field, / field to alter */
				1921	TYPE_ENUM, /* type to associate */
				1922	char *valuelist; / list of possible values */
				1923	int checkcase; /* case-sensitive? */
				1924	int checkunique); /* must specify uniquely? */
				1925	</PRE>
				1926
				1927	The <CODE>valuelist</CODE> parameter must point at a NULL-terminated list of
				1928	valid strings. The <CODE>checkcase</CODE> argument, if true, makes comparison
				1929	with the string case-sensitive. <P>
				1930
				1931	When the user exits a TYPE_ENUM field, the validation procedure tries to
				1932	complete the data in the buffer to a valid entry. If a complete choice string
				1933	has been entered, it is of course valid. But it is also possible to enter a
				1934	prefix of a valid string and have it completed for you. <P>
				1935
				1936	By default, if you enter such a prefix and it matches more than one value
				1937	in the string list, the prefix will be completed to the first matching
				1938	value. But the <CODE>checkunique</CODE> argument, if true, requires prefix
				1939	matches to be unique in order to be valid. <P>
				1940
				1941	The <CODE>REQ_NEXT_CHOICE</CODE> and <CODE>REQ_PREV_CHOICE</CODE> input requests
				1942	can be particularly useful with these fields.
				1943
				1944	<H3><A NAME="ftype_integer">TYPE_INTEGER</A></H3>
				1945
				1946	This field type accepts an integer. It is set up as follows:
				1947
				1948	<PRE>
				1949	int set_field_type(FIELD field, / field to alter */
				1950	TYPE_INTEGER, /* type to associate */
				1951	int padding, /* # places to zero-pad to */
				1952	int vmin, int vmax); /* valid range */
				1953	</PRE>
				1954
				1955	Valid characters consist of an optional leading minus and digits.
				1956	The range check is performed on exit. If the range maximum is less
				1957	than or equal to the minimum, the range is ignored. <P>
				1958
				1959	If the value passes its range check, it is padded with as many leading
				1960	zero digits as necessary to meet the padding argument. <P>
				1961
				1962	A <CODE>TYPE_INTEGER</CODE> value buffer can conveniently be interpreted
				1963	with the C library function <CODE>atoi(3)</CODE>.
				1964
				1965	<H3><A NAME="ftype_numeric">TYPE_NUMERIC</A></H3>
				1966
				1967	This field type accepts a decimal number. It is set up as follows:
				1968
				1969	<PRE>
				1970	int set_field_type(FIELD field, / field to alter */
				1971	TYPE_NUMERIC, /* type to associate */
				1972	int padding, /* # places of precision */
				1973	double vmin, double vmax); /* valid range */
				1974	</PRE>
				1975
				1976	Valid characters consist of an optional leading minus and digits. possibly
				1977	including a decimal point. If your system supports locale's, the decimal point
				1978	character used must be the one defined by your locale. The range check is
				1979	performed on exit. If the range maximum is less than or equal to the minimum,
				1980	the range is ignored. <P>
				1981
				1982	If the value passes its range check, it is padded with as many trailing
				1983	zero digits as necessary to meet the padding argument. <P>
				1984
				1985	A <CODE>TYPE_NUMERIC</CODE> value buffer can conveniently be interpreted
				1986	with the C library function <CODE>atof(3)</CODE>.
				1987
				1988	<H3><A NAME="ftype_regexp">TYPE_REGEXP</A></H3>
				1989
				1990	This field type accepts data matching a regular expression. It is set up
				1991	as follows:
				1992
				1993	<PRE>
				1994	int set_field_type(FIELD field, / field to alter */
				1995	TYPE_REGEXP, /* type to associate */
				1996	char regexp); / expression to match */
				1997	</PRE>
				1998
				1999	The syntax for regular expressions is that of <CODE>regcomp(3)</CODE>.
				2000	The check for regular-expression match is performed on exit.
				2001
				2002	<H2><A NAME="fbuffer">Direct Field Buffer Manipulation</A></H2>
				2003
				2004	The chief attribute of a field is its buffer contents. When a form has
				2005	been completed, your application usually needs to know the state of each
				2006	field buffer. You can find this out with:
				2007
				2008	<PRE>
				2009	char field_buffer(FIELD field, /* field to query */
				2010	int bufindex); /* number of buffer to query */
				2011	</PRE>
				2012
				2013	Normally, the state of the zero-numbered buffer for each field is set by
				2014	the user's editing actions on that field. It's sometimes useful to be able
				2015	to set the value of the zero-numbered (or some other) buffer from your
				2016	application:
				2017
				2018	<PRE>
				2019	int set_field_buffer(FIELD field, / field to alter */
				2020	int bufindex, /* number of buffer to alter */
				2021	char value); / string value to set */
				2022	</PRE>
				2023
				2024	If the field is not large enough and cannot be resized to a sufficiently
				2025	large size to contain the specified value, the value will be truncated
				2026	to fit. <P>
				2027
				2028	Calling <CODE>field_buffer()</CODE> with a null field pointer will raise an
				2029	error. Calling <CODE>field_buffer()</CODE> on a field not currently selected
				2030	for input will return a correct value. Calling <CODE>field_buffer()</CODE> on a
				2031	field that is currently selected for input may not necessarily give a
				2032	correct field buffer value, because entered data isn't necessarily copied to
				2033	buffer zero before the exit validation check.
				2034
				2035	To guarantee that the returned buffer value reflects on-screen reality,
				2036	call <CODE>field_buffer()</CODE> either (1) in the field's exit validation
				2037	check routine, (2) from the field's or form's initialization or termination
				2038	hooks, or (3) just after a <CODE>REQ_VALIDATION</CODE> request has been processed
				2039	by the forms driver.
				2040
				2041	<H2><A NAME="formattrs">Attributes of Forms</A></H2>
				2042
				2043	As with field attributes, form attributes inherit a default from a
				2044	system default form structure. These defaults can be queried or set by
				2045	of these functions using a form-pointer argument of <CODE>NULL</CODE>. <P>
				2046
				2047	The principal attribute of a form is its field list. You can query
				2048	and change this list with:
				2049
				2050	<PRE>
				2051	int set_form_fields(FORM form, / form to alter */
				2052	FIELD *fields); / fields to connect */
				2053
				2054	char form_fields(FORM form); /* fetch fields of form */
				2055
				2056	int field_count(FORM form); / count connect fields */
				2057	</PRE>
				2058
				2059	The second argument of <CODE>set_form_fields()</CODE> may be a
				2060	NULL-terminated field pointer array like the one required by
				2061	<CODE>new_form()</CODE>. In that case, the old fields of the form are
				2062	disconnected but not freed (and eligible to be connected to other
				2063	forms), then the new fields are connected. <P>
				2064
				2065	It may also be null, in which case the old fields are disconnected
				2066	(and not freed) but no new ones are connected. <P>
				2067
				2068	The <CODE>field_count()</CODE> function simply counts the number of fields
				2069	connected to a given from. It returns -1 if the form-pointer argument
				2070	is NULL.
				2071
				2072	<H2><A NAME="fdisplay">Control of Form Display</A></H2>
				2073
				2074	In the overview section, you saw that to display a form you normally
				2075	start by defining its size (and fields), posting it, and refreshing
				2076	the screen. There is an hidden step before posting, which is the
				2077	association of the form with a frame window (actually, a pair of
				2078	windows) within which it will be displayed. By default, the forms
				2079	library associates every form with the full-screen window
				2080	<CODE>stdscr</CODE>. <P>
				2081
				2082	By making this step explicit, you can associate a form with a declared
				2083	frame window on your screen display. This can be useful if you want to
				2084	adapt the form display to different screen sizes, dynamically tile
				2085	forms on the screen, or use a form as part of an interface layout
				2086	managed by <A HREF="#panels">panels</A>. <P>
				2087
				2088	The two windows associated with each form have the same functions as
				2089	their analogues in the <A HREF="#menu">menu library</A>. Both these
				2090	windows are painted when the form is posted and erased when the form
				2091	is unposted. <P>
				2092
				2093	The outer or frame window is not otherwise touched by the form
				2094	routines. It exists so the programmer can associate a title, a
				2095	border, or perhaps help text with the form and have it properly
				2096	refreshed or erased at post/unpost time. The inner window or subwindow
				2097	is where the current form page is actually displayed. <P>
				2098
				2099	In order to declare your own frame window for a form, you'll need to
				2100	know the size of the form's bounding rectangle. You can get this
				2101	information with:
				2102
				2103	<PRE>
				2104	int scale_form(FORM form, / form to query */
				2105	int rows, / form rows */
				2106	int cols); / form cols */
				2107	</PRE>
				2108
				2109	The form dimensions are passed back in the locations pointed to by
				2110	the arguments. Once you have this information, you can use it to
				2111	declare of windows, then use one of these functions:
				2112
				2113	<PRE>
				2114	int set_form_win(FORM form, / form to alter */
				2115	WINDOW win); / frame window to connect */
				2116
				2117	WINDOW form_win(FORM form); /* fetch frame window of form */
				2118
				2119	int set_form_sub(FORM form, / form to alter */
				2120	WINDOW win); / form subwindow to connect */
				2121
				2122	WINDOW form_sub(FORM form); /* fetch form subwindow of form */
				2123	</PRE>
				2124
				2125	Note that curses operations, including <CODE>refresh()</CODE>, on the form,
				2126	should be done on the frame window, not the form subwindow. <P>
				2127
				2128	It is possible to check from your application whether all of a
				2129	scrollable field is actually displayed within the menu subwindow. Use
				2130	these functions:
				2131
				2132	<PRE>
				2133	int data_ahead(FORM form); / form to be queried */
				2134
				2135	int data_behind(FORM form); / form to be queried */
				2136	</PRE>
				2137
				2138	The function <CODE>data_ahead()</CODE> returns TRUE if (a) the current
				2139	field is one-line and has undisplayed data off to the right, (b) the current
				2140	field is multi-line and there is data off-screen below it. <P>
				2141
				2142	The function <CODE>data_behind()</CODE> returns TRUE if the first (upper
				2143	left hand) character position is off-screen (not being displayed). <P>
				2144
				2145	Finally, there is a function to restore the form window's cursor to the
				2146	value expected by the forms driver:
				2147
				2148	<PRE>
				2149	int pos_form_cursor(FORM ) / form to be queried */
				2150	</PRE>
				2151
				2152	If your application changes the form window cursor, call this function before
				2153	handing control back to the forms driver in order to re-synchronize it.
				2154
				2155	<H2><A NAME="fdriver">Input Processing in the Forms Driver</A></H2>
				2156
				2157	The function <CODE>form_driver()</CODE> handles virtualized input requests
				2158	for form navigation, editing, and validation requests, just as
				2159	<CODE>menu_driver</CODE> does for menus (see the section on <A
				2160	HREF="#minput">menu input handling</A>).
				2161
				2162	<PRE>
				2163	int form_driver(FORM form, / form to pass input to */
				2164	int request); /* form request code */
				2165	</PRE>
				2166
				2167	Your input virtualization function needs to take input and then convert it
				2168	to either an alphanumeric character (which is treated as data to be
				2169	entered in the currently-selected field), or a forms processing request. <P>
				2170
				2171	The forms driver provides hooks (through input-validation and
				2172	field-termination functions) with which your application code can check
				2173	that the input taken by the driver matched what was expected.
				2174
				2175	<H3><A NAME="fpage">Page Navigation Requests</A></H3>
				2176
				2177	These requests cause page-level moves through the form,
				2178	triggering display of a new form screen.
				2179
				2180	<DL>
				2181	<DT> <CODE>REQ_NEXT_PAGE</CODE>
				2182	<DD> Move to the next form page.
				2183	<DT> <CODE>REQ_PREV_PAGE</CODE>
				2184	<DD> Move to the previous form page.
				2185	<DT> <CODE>REQ_FIRST_PAGE</CODE>
				2186	<DD> Move to the first form page.
				2187	<DT> <CODE>REQ_LAST_PAGE</CODE>
				2188	<DD> Move to the last form page.
				2189	</DL>
				2190
				2191	These requests treat the list as cyclic; that is, <CODE>REQ_NEXT_PAGE</CODE>
				2192	from the last page goes to the first, and <CODE>REQ_PREV_PAGE</CODE> from
				2193	the first page goes to the last.
				2194
				2195	<H3><A NAME="ffield">Inter-Field Navigation Requests</A></H3>
				2196
				2197	These requests handle navigation between fields on the same page.
				2198
				2199	<DL>
				2200	<DT> <CODE>REQ_NEXT_FIELD</CODE>
				2201	<DD> Move to next field.
				2202	<DT> <CODE>REQ_PREV_FIELD</CODE>
				2203	<DD> Move to previous field.
				2204	<DT> <CODE>REQ_FIRST_FIELD</CODE>
				2205	<DD> Move to the first field.
				2206	<DT> <CODE>REQ_LAST_FIELD</CODE>
				2207	<DD> Move to the last field.
				2208	<DT> <CODE>REQ_SNEXT_FIELD</CODE>
				2209	<DD> Move to sorted next field.
				2210	<DT> <CODE>REQ_SPREV_FIELD</CODE>
				2211	<DD> Move to sorted previous field.
				2212	<DT> <CODE>REQ_SFIRST_FIELD</CODE>
				2213	<DD> Move to the sorted first field.
				2214	<DT> <CODE>REQ_SLAST_FIELD</CODE>
				2215	<DD> Move to the sorted last field.
				2216	<DT> <CODE>REQ_LEFT_FIELD</CODE>
				2217	<DD> Move left to field.
				2218	<DT> <CODE>REQ_RIGHT_FIELD</CODE>
				2219	<DD> Move right to field.
				2220	<DT> <CODE>REQ_UP_FIELD</CODE>
				2221	<DD> Move up to field.
				2222	<DT> <CODE>REQ_DOWN_FIELD</CODE>
				2223	<DD> Move down to field.
				2224	</DL>
				2225
				2226	These requests treat the list of fields on a page as cyclic; that is,
				2227	<CODE>REQ_NEXT_FIELD</CODE> from the last field goes to the first, and
				2228	<CODE>REQ_PREV_FIELD</CODE> from the first field goes to the last. The
				2229	order of the fields for these (and the <CODE>REQ_FIRST_FIELD</CODE> and
				2230	<CODE>REQ_LAST_FIELD</CODE> requests) is simply the order of the field
				2231	pointers in the form array (as set up by <CODE>new_form()</CODE> or
				2232	<CODE>set_form_fields()</CODE> <P>
				2233
				2234	It is also possible to traverse the fields as if they had been sorted in
				2235	screen-position order, so the sequence goes left-to-right and top-to-bottom.
				2236	To do this, use the second group of four sorted-movement requests. <P>
				2237
				2238	Finally, it is possible to move between fields using visual directions up,
				2239	down, right, and left. To accomplish this, use the third group of four
				2240	requests. Note, however, that the position of a form for purposes of these
				2241	requests is its upper-left corner. <P>
				2242
				2243	For example, suppose you have a multi-line field B, and two
				2244	single-line fields A and C on the same line with B, with A to the left
				2245	of B and C to the right of B. A <CODE>REQ_MOVE_RIGHT</CODE> from A will
				2246	go to B only if A, B, and C <EM>all</EM> share the same first line;
				2247	otherwise it will skip over B to C.
				2248
				2249	<H3><A NAME="fifield">Intra-Field Navigation Requests</A></H3>
				2250
				2251	These requests drive movement of the edit cursor within the currently
				2252	selected field.
				2253
				2254	<DL>
				2255	<DT> <CODE>REQ_NEXT_CHAR</CODE>
				2256	<DD> Move to next character.
				2257	<DT> <CODE>REQ_PREV_CHAR</CODE>
				2258	<DD> Move to previous character.
				2259	<DT> <CODE>REQ_NEXT_LINE</CODE>
				2260	<DD> Move to next line.
				2261	<DT> <CODE>REQ_PREV_LINE</CODE>
				2262	<DD> Move to previous line.
				2263	<DT> <CODE>REQ_NEXT_WORD</CODE>
				2264	<DD> Move to next word.
				2265	<DT> <CODE>REQ_PREV_WORD</CODE>
				2266	<DD> Move to previous word.
				2267	<DT> <CODE>REQ_BEG_FIELD</CODE>
				2268	<DD> Move to beginning of field.
				2269	<DT> <CODE>REQ_END_FIELD</CODE>
				2270	<DD> Move to end of field.
				2271	<DT> <CODE>REQ_BEG_LINE</CODE>
				2272	<DD> Move to beginning of line.
				2273	<DT> <CODE>REQ_END_LINE</CODE>
				2274	<DD> Move to end of line.
				2275	<DT> <CODE>REQ_LEFT_CHAR</CODE>
				2276	<DD> Move left in field.
				2277	<DT> <CODE>REQ_RIGHT_CHAR</CODE>
				2278	<DD> Move right in field.
				2279	<DT> <CODE>REQ_UP_CHAR</CODE>
				2280	<DD> Move up in field.
				2281	<DT> <CODE>REQ_DOWN_CHAR</CODE>
				2282	<DD> Move down in field.
				2283	</DL>
				2284
				2285	Each <EM>word</EM> is separated from the previous and next characters
				2286	by whitespace. The commands to move to beginning and end of line or field
				2287	look for the first or last non-pad character in their ranges.
				2288
				2289	<H3><A NAME="fscroll">Scrolling Requests</A></H3>
				2290
				2291	Fields that are dynamic and have grown and fields explicitly created
				2292	with offscreen rows are scrollable. One-line fields scroll horizontally;
				2293	multi-line fields scroll vertically. Most scrolling is triggered by
				2294	editing and intra-field movement (the library scrolls the field to keep the
				2295	cursor visible). It is possible to explicitly request scrolling with the
				2296	following requests:
				2297
				2298	<DL>
				2299	<DT> <CODE>REQ_SCR_FLINE</CODE>
				2300	<DD> Scroll vertically forward a line.
				2301	<DT> <CODE>REQ_SCR_BLINE</CODE>
				2302	<DD> Scroll vertically backward a line.
				2303	<DT> <CODE>REQ_SCR_FPAGE</CODE>
				2304	<DD> Scroll vertically forward a page.
				2305	<DT> <CODE>REQ_SCR_BPAGE</CODE>
				2306	<DD> Scroll vertically backward a page.
				2307	<DT> <CODE>REQ_SCR_FHPAGE</CODE>
				2308	<DD> Scroll vertically forward half a page.
				2309	<DT> <CODE>REQ_SCR_BHPAGE</CODE>
				2310	<DD> Scroll vertically backward half a page.
				2311	<DT> <CODE>REQ_SCR_FCHAR</CODE>
				2312	<DD> Scroll horizontally forward a character.
				2313	<DT> <CODE>REQ_SCR_BCHAR</CODE>
				2314	<DD> Scroll horizontally backward a character.
				2315	<DT> <CODE>REQ_SCR_HFLINE</CODE>
				2316	<DD> Scroll horizontally one field width forward.
				2317	<DT> <CODE>REQ_SCR_HBLINE</CODE>
				2318	<DD> Scroll horizontally one field width backward.
				2319	<DT> <CODE>REQ_SCR_HFHALF</CODE>
				2320	<DD> Scroll horizontally one half field width forward.
				2321	<DT> <CODE>REQ_SCR_HBHALF</CODE>
				2322	<DD> Scroll horizontally one half field width backward.
				2323	</DL>
				2324
				2325	For scrolling purposes, a <EM>page</EM> of a field is the height
				2326	of its visible part.
				2327
				2328	<H3><A NAME="fedit">Editing Requests</A></H3>
				2329
				2330	When you pass the forms driver an ASCII character, it is treated as a
				2331	request to add the character to the field's data buffer. Whether this
				2332	is an insertion or a replacement depends on the field's edit mode
				2333	(insertion is the default. <P>
				2334
				2335	The following requests support editing the field and changing the edit
				2336	mode:
				2337
				2338	<DL>
				2339	<DT> <CODE>REQ_INS_MODE</CODE>
				2340	<DD> Set insertion mode.
				2341	<DT> <CODE>REQ_OVL_MODE</CODE>
				2342	<DD> Set overlay mode.
				2343	<DT> <CODE>REQ_NEW_LINE</CODE>
				2344	<DD> New line request (see below for explanation).
				2345	<DT> <CODE>REQ_INS_CHAR</CODE>
				2346	<DD> Insert space at character location.
				2347	<DT> <CODE>REQ_INS_LINE</CODE>
				2348	<DD> Insert blank line at character location.
				2349	<DT> <CODE>REQ_DEL_CHAR</CODE>
				2350	<DD> Delete character at cursor.
				2351	<DT> <CODE>REQ_DEL_PREV</CODE>
				2352	<DD> Delete previous word at cursor.
				2353	<DT> <CODE>REQ_DEL_LINE</CODE>
				2354	<DD> Delete line at cursor.
				2355	<DT> <CODE>REQ_DEL_WORD</CODE>
				2356	<DD> Delete word at cursor.
				2357	<DT> <CODE>REQ_CLR_EOL</CODE>
				2358	<DD> Clear to end of line.
				2359	<DT> <CODE>REQ_CLR_EOF</CODE>
				2360	<DD> Clear to end of field.
				2361	<DT> <CODE>REQ_CLEAR_FIELD</CODE>
				2362	<DD> Clear entire field.
				2363	</DL>
				2364
				2365	The behavior of the <CODE>REQ_NEW_LINE</CODE> and <CODE>REQ_DEL_PREV</CODE> requests
				2366	is complicated and partly controlled by a pair of forms options.
				2367	The special cases are triggered when the cursor is at the beginning of
				2368	a field, or on the last line of the field. <P>
				2369
				2370	First, we consider <CODE>REQ_NEW_LINE</CODE>: <P>
				2371
				2372	The normal behavior of <CODE>REQ_NEW_LINE</CODE> in insert mode is to break the
				2373	current line at the position of the edit cursor, inserting the portion of
				2374	the current line after the cursor as a new line following the current
				2375	and moving the cursor to the beginning of that new line (you may think
				2376	of this as inserting a newline in the field buffer). <P>
				2377
				2378	The normal behavior of <CODE>REQ_NEW_LINE</CODE> in overlay mode is to clear the
				2379	current line from the position of the edit cursor to end of line.
				2380	The cursor is then moved to the beginning of the next line. <P>
				2381
				2382	However, <CODE>REQ_NEW_LINE</CODE> at the beginning of a field, or on the
				2383	last line of a field, instead does a <CODE>REQ_NEXT_FIELD</CODE>.
				2384	<CODE>O_NL_OVERLOAD</CODE> option is off, this special action is
				2385	disabled. <P>
				2386
				2387	Now, let us consider <CODE>REQ_DEL_PREV</CODE>: <P>
				2388
				2389	The normal behavior of <CODE>REQ_DEL_PREV</CODE> is to delete the previous
				2390	character. If insert mode is on, and the cursor is at the start of a
				2391	line, and the text on that line will fit on the previous one, it
				2392	instead appends the contents of the current line to the previous one
				2393	and deletes the current line (you may think of this as deleting a
				2394	newline from the field buffer). <P>
				2395
				2396	However, <CODE>REQ_DEL_PREV</CODE> at the beginning of a field is instead
				2397	treated as a <CODE>REQ_PREV_FIELD</CODE>. <P> If the
				2398	<CODE>O_BS_OVERLOAD</CODE> option is off, this special action is
				2399	disabled and the forms driver just returns <CODE>E_REQUEST_DENIED</CODE>. <P>
				2400
				2401	See <A HREF="#frmoptions">Form Options</A> for discussion of how to set
				2402	and clear the overload options.
				2403
				2404	<H3><A NAME="forder">Order Requests</A></H3>
				2405
				2406	If the type of your field is ordered, and has associated functions
				2407	for getting the next and previous values of the type from a given value,
				2408	there are requests that can fetch that value into the field buffer:
				2409
				2410	<DL>
				2411	<DT> <CODE>REQ_NEXT_CHOICE</CODE>
				2412	<DD> Place the successor value of the current value in the buffer.
				2413	<DT> <CODE>REQ_PREV_CHOICE</CODE>
				2414	<DD> Place the predecessor value of the current value in the buffer.
				2415	</DL>
				2416
				2417	Of the built-in field types, only <CODE>TYPE_ENUM</CODE> has built-in successor
				2418	and predecessor functions. When you define a field type of your own
				2419	(see <A HREF="#fcustom">Custom Validation Types</A>), you can associate
				2420	our own ordering functions.
				2421
				2422	<H3><A NAME="fappcmds">Application Commands</A></H3>
				2423
				2424	Form requests are represented as integers above the <CODE>curses</CODE> value
				2425	greater than <CODE>KEY_MAX</CODE> and less than or equal to the constant
				2426	<CODE>MAX_COMMAND</CODE>. If your input-virtualization routine returns a
				2427	value above <CODE>MAX_COMMAND</CODE>, the forms driver will ignore it.
				2428
				2429	<H2><A NAME="fhooks">Field Change Hooks</A></H2>
				2430
				2431	It is possible to set function hooks to be executed whenever the
				2432	current field or form changes. Here are the functions that support this:
				2433
				2434	<PRE>
				2435	typedef void (HOOK)(); / pointer to function returning void */
				2436
				2437	int set_form_init(FORM form, / form to alter */
				2438	HOOK hook); /* initialization hook */
				2439
				2440	HOOK form_init(FORM form); / form to query */
				2441
				2442	int set_form_term(FORM form, / form to alter */
				2443	HOOK hook); /* termination hook */
				2444
				2445	HOOK form_term(FORM form); / form to query */
				2446
				2447	int set_field_init(FORM form, / form to alter */
				2448	HOOK hook); /* initialization hook */
				2449
				2450	HOOK field_init(FORM form); / form to query */
				2451
				2452	int set_field_term(FORM form, / form to alter */
				2453	HOOK hook); /* termination hook */
				2454
				2455	HOOK field_term(FORM form); / form to query */
				2456	</PRE>
				2457
				2458	These functions allow you to either set or query four different hooks.
				2459	In each of the set functions, the second argument should be the
				2460	address of a hook function. These functions differ only in the timing
				2461	of the hook call.
				2462
				2463	<DL>
				2464	<DT> form_init
				2465	<DD> This hook is called when the form is posted; also, just after
				2466	each page change operation.
				2467	<DT> field_init
				2468	<DD> This hook is called when the form is posted; also, just after
				2469	each field change
				2470	<DT> field_term
				2471	<DD> This hook is called just after field validation; that is, just before
				2472	the field is altered. It is also called when the form is unposted.
				2473	<DT> form_term
				2474	<DD> This hook is called when the form is unposted; also, just before
				2475	each page change operation.
				2476	</DL>
				2477
				2478	Calls to these hooks may be triggered
				2479	<OL>
				2480	<LI>When user editing requests are processed by the forms driver
				2481	<LI>When the current page is changed by <CODE>set_current_field()</CODE> call
				2482	<LI>When the current field is changed by a <CODE>set_form_page()</CODE> call
				2483	</OL>
				2484
				2485	See <A NAME="ffocus">Field Change Commands</A> for discussion of the latter
				2486	two cases. <P>
				2487
				2488	You can set a default hook for all fields by passing one of the set functions
				2489	a NULL first argument. <P>
				2490
				2491	You can disable any of these hooks by (re)setting them to NULL, the default
				2492	value.
				2493
				2494	<H2><A HREF="#ffocus">Field Change Commands</A></H2>
				2495
				2496	Normally, navigation through the form will be driven by the user's
				2497	input requests. But sometimes it is useful to be able to move the
				2498	focus for editing and viewing under control of your application, or
				2499	ask which field it currently is in. The following functions help you
				2500	accomplish this:
				2501
				2502	<PRE>
				2503	int set_current_field(FORM form, / form to alter */
				2504	FIELD field); / field to shift to */
				2505
				2506	FIELD current_field(FORM form); /* form to query */
				2507
				2508	int field_index(FORM form, / form to query */
				2509	FIELD field); / field to get index of */
				2510	</PRE>
				2511
				2512	The function <CODE>field_index()</CODE> returns the index of the given field
				2513	in the given form's field array (the array passed to <CODE>new_form()</CODE> or
				2514	<CODE>set_form_fields()</CODE>). <P>
				2515
				2516	The initial current field of a form is the first active field on the
				2517	first page. The function <CODE>set_form_fields()</CODE> resets this.<P>
				2518
				2519	It is also possible to move around by pages.
				2520
				2521	<PRE>
				2522	int set_form_page(FORM form, / form to alter */
				2523	int page); /* page to go to (0-origin) */
				2524
				2525	int form_page(FORM form); / return form's current page */
				2526	</PRE>
				2527
				2528	The initial page of a newly-created form is 0. The function
				2529	<CODE>set_form_fields()</CODE> resets this.
				2530
				2531	<H2><A NAME="frmoptions">Form Options</A></H2>
				2532
				2533	Like fields, forms may have control option bits. They can be changed
				2534	or queried with these functions:
				2535
				2536	<PRE>
				2537	int set_form_opts(FORM form, / form to alter */
				2538	int attr); /* attribute to set */
				2539
				2540	int form_opts_on(FORM form, / form to alter */
				2541	int attr); /* attributes to turn on */
				2542
				2543	int form_opts_off(FORM form, / form to alter */
				2544	int attr); /* attributes to turn off */
				2545
				2546	int form_opts(FORM form); / form to query */
				2547	</PRE>
				2548
				2549	By default, all options are on. Here are the available option bits:
				2550
				2551	<DL>
				2552	<DT> O_NL_OVERLOAD
				2553	<DD> Enable overloading of <CODE>REQ_NEW_LINE</CODE> as described in <A
				2554	href="#fedit">Editing Requests</A>. The value of this option is
				2555	ignored on dynamic fields that have not reached their size limit;
				2556	these have no last line, so the circumstances for triggering a
				2557	<CODE>REQ_NEXT_FIELD</CODE> never arise.
				2558	<DT> O_BS_OVERLOAD
				2559	<DD> Enable overloading of <CODE>REQ_DEL_PREV</CODE> as described in
				2560	<A href="#fedit">Editing Requests</A>.
				2561	</DL>
				2562
				2563	The option values are bit-masks and can be composed with logical-or in
				2564	the obvious way.
				2565
				2566	<H2><A NAME="fcustom">Custom Validation Types</A></H2>
				2567
				2568	The <CODE>form</CODE> library gives you the capability to define custom
				2569	validation types of your own. Further, the optional additional arguments
				2570	of <CODE>set_field_type</CODE> effectively allow you to parameterize validation
				2571	types. Most of the complications in the validation-type interface have to
				2572	do with the handling of the additional arguments within custom validation
				2573	functions.
				2574
				2575	<H3><A NAME="flinktypes">Union Types</A></H3>
				2576
				2577	The simplest way to create a custom data type is to compose it from two
				2578	preexisting ones:
				2579
				2580	<PRE>
				2581	FIELD link_fieldtype(FIELDTYPE type1,
				2582	FIELDTYPE *type2);
				2583	</PRE>
				2584
				2585	This function creates a field type that will accept any of the values
				2586	legal for either of its argument field types (which may be either
				2587	predefined or programmer-defined).
				2588
				2589	If a <CODE>set_field_type()</CODE> call later requires arguments, the new
				2590	composite type expects all arguments for the first type, than all arguments
				2591	for the second. Order functions (see <A HREF="#forder">Order Requests</A>)
				2592	associated with the component types will work on the composite; what it does
				2593	is check the validation function for the first type, then for the second, to
				2594	figure what type the buffer contents should be treated as.
				2595
				2596	<H3><A NAME="fnewtypes">New Field Types</A></H3>
				2597
				2598	To create a field type from scratch, you need to specify one or both of the
				2599	following things:
				2600
				2601	<UL>
				2602	<LI>A character-validation function, to check each character as it is entered.
				2603	<LI>A field-validation function to be applied on exit from the field.
				2604	</UL>
				2605
				2606	Here's how you do that:
				2607	<PRE>
				2608	typedef int (HOOK)(); / pointer to function returning int */
				2609
				2610	FIELDTYPE new_fieldtype(HOOK f_validate, / field validator */
				2611	HOOK c_validate) /* character validator */
				2612
				2613
				2614	int free_fieldtype(FIELDTYPE ftype); / type to free */
				2615	</PRE>
				2616
				2617	At least one of the arguments of <CODE>new_fieldtype()</CODE> must be
				2618	non-NULL. The forms driver will automatically call the new type's
				2619	validation functions at appropriate points in processing a field of
				2620	the new type. <P>
				2621
				2622	The function <CODE>free_fieldtype()</CODE> deallocates the argument
				2623	fieldtype, freeing all storage associated with it. <P>
				2624
				2625	Normally, a field validator is called when the user attempts to
				2626	leave the field. Its first argument is a field pointer, from which it
				2627	can get to field buffer 0 and test it. If the function returns TRUE,
				2628	the operation succeeds; if it returns FALSE, the edit cursor stays in
				2629	the field. <P>
				2630
				2631	A character validator gets the character passed in as a first argument.
				2632	It too should return TRUE if the character is valid, FALSE otherwise.
				2633
				2634	<H3><A NAME="fcheckargs">Validation Function Arguments</A></H3>
				2635
				2636	Your field- and character- validation functions will be passed a
				2637	second argument as well. This second argument is the address of a
				2638	structure (which we'll call a <EM>pile</EM>) built from any of the
				2639	field-type-specific arguments passed to <CODE>set_field_type()</CODE>. If
				2640	no such arguments are defined for the field type, this pile pointer
				2641	argument will be NULL. <P>
				2642
				2643	In order to arrange for such arguments to be passed to your validation
				2644	functions, you must associate a small set of storage-management functions
				2645	with the type. The forms driver will use these to synthesize a pile
				2646	from the trailing arguments of each <CODE>set_field_type()</CODE> argument, and
				2647	a pointer to the pile will be passed to the validation functions. <P>
				2648
				2649	Here is how you make the association:
				2650
				2651	<PRE>
				2652	typedef char (PTRHOOK)(); /* pointer to function returning (char ) /
				2653	typedef void (VOIDHOOK)(); / pointer to function returning void */
				2654
				2655	int set_fieldtype_arg(FIELDTYPE type, / type to alter */
				2656	PTRHOOK make_str, /* make structure from args */
				2657	PTRHOOK copy_str, /* make copy of structure */
				2658	VOIDHOOK free_str); /* free structure storage */
				2659	</PRE>
				2660
				2661	Here is how the storage-management hooks are used:
				2662
				2663	<DL>
				2664	<DT> <CODE>make_str</CODE>
				2665	<DD> This function is called by <CODE>set_field_type()</CODE>. It gets one
				2666	argument, a <CODE>va_list</CODE> of the type-specific arguments passed to
				2667	<CODE>set_field_type()</CODE>. It is expected to return a pile pointer to a data
				2668	structure that encapsulates those arguments.
				2669	<DT> <CODE>copy_str</CODE>
				2670	<DD> This function is called by form library functions that allocate new
				2671	field instances. It is expected to take a pile pointer, copy the pile
				2672	to allocated storage, and return the address of the pile copy.
				2673	<DT> <CODE>free_str</CODE>
				2674	<DD> This function is called by field- and type-deallocation routines in the
				2675	library. It takes a pile pointer argument, and is expected to free the
				2676	storage of that pile.
				2677	</DL>
				2678
				2679	The <CODE>make_str</CODE> and <CODE>copy_str</CODE> functions may return NULL to
				2680	signal allocation failure. The library routines will that call them will
				2681	return error indication when this happens. Thus, your validation functions
				2682	should never see a NULL file pointer and need not check specially for it.
				2683
				2684	<H3><A NAME="fcustorder">Order Functions For Custom Types</A></H3>
				2685
				2686	Some custom field types are simply ordered in the same well-defined way
				2687	that <CODE>TYPE_ENUM</CODE> is. For such types, it is possible to define
				2688	successor and predecessor functions to support the <CODE>REQ_NEXT_CHOICE</CODE>
				2689	and <CODE>REQ_PREV_CHOICE</CODE> requests. Here's how:
				2690
				2691	<PRE>
				2692	typedef int (INTHOOK)(); / pointer to function returning int */
				2693
				2694	int set_fieldtype_arg(FIELDTYPE type, / type to alter */
				2695	INTHOOK succ, /* get successor value */
				2696	INTHOOK pred); /* get predecessor value */
				2697	</PRE>
				2698
				2699	The successor and predecessor arguments will each be passed two arguments;
				2700	a field pointer, and a pile pointer (as for the validation functions). They
				2701	are expected to use the function <CODE>field_buffer()</CODE> to read the
				2702	current value, and <CODE>set_field_buffer()</CODE> on buffer 0 to set the next
				2703	or previous value. Either hook may return TRUE to indicate success (a
				2704	legal next or previous value was set) or FALSE to indicate failure.
				2705
				2706	<H3><A NAME="fcustprobs">Avoiding Problems</A></H3>
				2707
				2708	The interface for defining custom types is complicated and tricky.
				2709	Rather than attempting to create a custom type entirely from scratch,
				2710	you should start by studying the library source code for whichever of
				2711	the pre-defined types seems to be closest to what you want. <P>
				2712
				2713	Use that code as a model, and evolve it towards what you really want.
				2714	You will avoid many problems and annoyances that way. The code
				2715	in the <CODE>ncurses</CODE> library has been specifically exempted from
				2716	the package copyright to support this. <P>
				2717
				2718	If your custom type defines order functions, have do something intuitive
				2719	with a blank field. A useful convention is to make the successor of a
				2720	blank field the types minimum value, and its predecessor the maximum.
				2721	</BODY>
				2722	</HTML>