2 Writing Programs with NCURSES
4 by Eric S. Raymond and Zeyd M. Ben-Halim
5 updates since release 1.9.9e by Thomas Dickey
10 + A Brief History of Curses
11 + Scope of This Document
14 + An Overview of Curses
15 o Compiling Programs using Curses
17 o Standard Windows and Function Naming Conventions
23 o Using Forms Characters
24 o Character Attributes and Color
27 + Function Descriptions
28 o Initialization and Wrapup
29 o Causing Output to the Terminal
30 o Low-Level Capability Access
32 + Hints, Tips, and Tricks
33 o Some Notes of Caution
34 o Temporarily Leaving ncurses Mode
35 o Using ncurses under xterm
36 o Handling Multiple Terminal Screens
37 o Testing for Terminal Capabilities
39 o Special Features of ncurses
40 + Compatibility with Older Versions
41 o Refresh of Overlapping Windows
43 + XSI Curses Conformance
45 + Compiling With the Panels Library
47 + Panels, Input, and the Standard Screen
49 + Miscellaneous Other Facilities
51 + Compiling with the menu Library
56 + Processing Menu Input
57 + Miscellaneous Other Features
59 + Compiling with the forms Library
61 + Creating and Freeing Fields and Forms
62 + Fetching and Changing Field Attributes
63 o Fetching Size and Location Data
64 o Changing the Field Location
65 o The Justification Attribute
66 o Field Display Attributes
70 + Variable-Sized Fields
78 + Direct Field Buffer Manipulation
80 + Control of Form Display
81 + Input Processing in the Forms Driver
82 o Page Navigation Requests
83 o Inter-Field Navigation Requests
84 o Intra-Field Navigation Requests
86 o Field Editing Requests
88 o Application Commands
90 + Field Change Commands
92 + Custom Validation Types
95 o Validation Function Arguments
96 o Order Functions For Custom Types
98 _________________________________________________________________
102 This document is an introduction to programming with curses. It is not
103 an exhaustive reference for the curses Application Programming
104 Interface (API); that role is filled by the curses manual pages.
105 Rather, it is intended to help C programmers ease into using the
108 This document is aimed at C applications programmers not yet
109 specifically familiar with ncurses. If you are already an experienced
110 curses programmer, you should nevertheless read the sections on Mouse
111 Interfacing, Debugging, Compatibility with Older Versions, and Hints,
112 Tips, and Tricks. These will bring you up to speed on the special
113 features and quirks of the ncurses implementation. If you are not so
114 experienced, keep reading.
116 The curses package is a subroutine library for terminal-independent
117 screen-painting and input-event handling which presents a high level
118 screen model to the programmer, hiding differences between terminal
119 types and doing automatic optimization of output to change one screen
120 full of text into another. Curses uses terminfo, which is a database
121 format that can describe the capabilities of thousands of different
124 The curses API may seem something of an archaism on UNIX desktops
125 increasingly dominated by X, Motif, and Tcl/Tk. Nevertheless, UNIX
126 still supports tty lines and X supports xterm(1); the curses API has
127 the advantage of (a) back-portability to character-cell terminals, and
128 (b) simplicity. For an application that does not require bit-mapped
129 graphics and multiple fonts, an interface implementation using curses
130 will typically be a great deal simpler and less expensive than one
133 A Brief History of Curses
135 Historically, the first ancestor of curses was the routines written to
136 provide screen-handling for the game rogue; these used the
137 already-existing termcap database facility for describing terminal
138 capabilities. These routines were abstracted into a documented library
139 and first released with the early BSD UNIX versions.
141 System III UNIX from Bell Labs featured a rewritten and much-improved
142 curses library. It introduced the terminfo format. Terminfo is based
143 on Berkeley's termcap database, but contains a number of improvements
144 and extensions. Parameterized capabilities strings were introduced,
145 making it possible to describe multiple video attributes, and colors
146 and to handle far more unusual terminals than possible with termcap.
147 In the later AT&T System V releases, curses evolved to use more
148 facilities and offer more capabilities, going far beyond BSD curses in
149 power and flexibility.
151 Scope of This Document
153 This document describes ncurses, a free implementation of the System V
154 curses API with some clearly marked extensions. It includes the
155 following System V curses features:
157 * Support for multiple screen highlights (BSD curses could only
158 handle one `standout' highlight, usually reverse-video).
159 * Support for line- and box-drawing using forms characters.
160 * Recognition of function keys on input.
162 * Support for pads (windows of larger than screen size on which the
163 screen or a subwindow defines a viewport).
165 Also, this package makes use of the insert and delete line and
166 character features of terminals so equipped, and determines how to
167 optimally use these features with no help from the programmer. It
168 allows arbitrary combinations of video attributes to be displayed,
169 even on terminals that leave ``magic cookies'' on the screen to mark
170 changes in attributes.
172 The ncurses package can also capture and use event reports from a
173 mouse in some environments (notably, xterm under the X window system).
174 This document includes tips for using the mouse.
176 The ncurses package was originated by Pavel Curtis. The original
177 maintainer of this package is Zeyd Ben-Halim <zmbenhal@netcom.com>.
178 Eric S. Raymond <esr@snark.thyrsus.com> wrote many of the new features
179 in versions after 1.8.1 and wrote most of this introduction. Jürgen
180 Pfeifer wrote all of the menu and forms code as well as the Ada95
181 binding. Ongoing work is being done by Thomas Dickey and Jürgen
182 Pfeifer. Florian La Roche acts as the maintainer for the Free Software
183 Foundation, which holds the copyright on ncurses. Contact the current
184 maintainers at bug-ncurses@gnu.org.
186 This document also describes the panels extension library, similarly
187 modeled on the SVr4 panels facility. This library allows you to
188 associate backing store with each of a stack or deck of overlapping
189 windows, and provides operations for moving windows around in the
190 stack that change their visibility in the natural way (handling window
193 Finally, this document describes in detail the menus and forms
194 extension libraries, also cloned from System V, which support easy
195 construction and sequences of menus and fill-in forms.
199 In this document, the following terminology is used with reasonable
203 A data structure describing a sub-rectangle of the screen
204 (possibly the entire screen). You can write to a window as
205 though it were a miniature screen, scrolling independently of
206 other windows on the physical screen.
209 A subset of windows which are as large as the terminal screen,
210 i.e., they start at the upper left hand corner and encompass
211 the lower right hand corner. One of these, stdscr, is
212 automatically provided for the programmer.
215 The package's idea of what the terminal display currently looks
216 like, i.e., what the user sees now. This is a special screen.
220 An Overview of Curses
222 Compiling Programs using Curses
224 In order to use the library, it is necessary to have certain types and
225 variables defined. Therefore, the programmer must have a line:
228 at the top of the program source. The screen package uses the Standard
229 I/O library, so <curses.h> includes <stdio.h>. <curses.h> also
230 includes <termios.h>, <termio.h>, or <sgtty.h> depending on your
231 system. It is redundant (but harmless) for the programmer to do these
232 includes, too. In linking with curses you need to have -lncurses in
233 your LDFLAGS or on the command line. There is no need for any other
238 In order to update the screen optimally, it is necessary for the
239 routines to know what the screen currently looks like and what the
240 programmer wants it to look like next. For this purpose, a data type
241 (structure) named WINDOW is defined which describes a window image to
242 the routines, including its starting position on the screen (the (y,
243 x) coordinates of the upper left hand corner) and its size. One of
244 these (called curscr, for current screen) is a screen image of what
245 the terminal currently looks like. Another screen (called stdscr, for
246 standard screen) is provided by default to make changes on.
248 A window is a purely internal representation. It is used to build and
249 store a potential image of a portion of the terminal. It doesn't bear
250 any necessary relation to what is really on the terminal screen; it's
251 more like a scratchpad or write buffer.
253 To make the section of physical screen corresponding to a window
254 reflect the contents of the window structure, the routine refresh()
255 (or wrefresh() if the window is not stdscr) is called.
257 A given physical screen section may be within the scope of any number
258 of overlapping windows. Also, changes can be made to windows in any
259 order, without regard to motion efficiency. Then, at will, the
260 programmer can effectively say ``make it look like this,'' and let the
261 package implementation determine the most efficient way to repaint the
264 Standard Windows and Function Naming Conventions
266 As hinted above, the routines can use several windows, but two are
267 automatically given: curscr, which knows what the terminal looks like,
268 and stdscr, which is what the programmer wants the terminal to look
269 like next. The user should never actually access curscr directly.
270 Changes should be made to through the API, and then the routine
271 refresh() (or wrefresh()) called.
273 Many functions are defined to use stdscr as a default screen. For
274 example, to add a character to stdscr, one calls addch() with the
275 desired character as argument. To write to a different window. use the
276 routine waddch() (for `w'indow-specific addch()) is provided. This
277 convention of prepending function names with a `w' when they are to be
278 applied to specific windows is consistent. The only routines which do
279 not follow it are those for which a window must always be specified.
281 In order to move the current (y, x) coordinates from one point to
282 another, the routines move() and wmove() are provided. However, it is
283 often desirable to first move and then perform some I/O operation. In
284 order to avoid clumsiness, most I/O routines can be preceded by the
285 prefix 'mv' and the desired (y, x) coordinates prepended to the
286 arguments to the function. For example, the calls
298 mvwaddch(win, y, x, ch);
300 Note that the window description pointer (win) comes before the added
301 (y, x) coordinates. If a function requires a window pointer, it is
302 always the first parameter passed.
306 The curses library sets some variables describing the terminal
308 type name description
309 ------------------------------------------------------------------
310 int LINES number of lines on the terminal
311 int COLS number of columns on the terminal
313 The curses.h also introduces some #define constants and types of
317 boolean type, actually a `char' (e.g., bool doneit;)
320 boolean `true' flag (1).
323 boolean `false' flag (0).
326 error flag returned by routines on a failure (-1).
329 error flag returned by routines when things go right.
333 Now we describe how to actually use the screen package. In it, we
334 assume all updating, reading, etc. is applied to stdscr. These
335 instructions will work on any window, providing you change the
336 function names and parameters as mentioned above.
338 Here is a sample program to motivate the discussion:
343 static void finish(int sig);
345 main(int argc, char *argv[])
347 /* initialize your non-curses data structures here */
349 (void) signal(SIGINT, finish); /* arrange interrupts to terminate */
351 (void) initscr(); /* initialize the curses library */
352 keypad(stdscr, TRUE); /* enable keyboard mapping */
353 (void) nonl(); /* tell curses not to do NL->CR/NL on output */
354 (void) cbreak(); /* take input chars one at a time, no wait for \n */
355 (void) noecho(); /* don't echo input */
362 * Simple color assignment, often all we need.
364 init_pair(COLOR_BLACK, COLOR_BLACK, COLOR_BLACK);
365 init_pair(COLOR_GREEN, COLOR_GREEN, COLOR_BLACK);
366 init_pair(COLOR_RED, COLOR_RED, COLOR_BLACK);
367 init_pair(COLOR_CYAN, COLOR_CYAN, COLOR_BLACK);
368 init_pair(COLOR_WHITE, COLOR_WHITE, COLOR_BLACK);
369 init_pair(COLOR_MAGENTA, COLOR_MAGENTA, COLOR_BLACK);
370 init_pair(COLOR_BLUE, COLOR_BLUE, COLOR_BLACK);
371 init_pair(COLOR_YELLOW, COLOR_YELLOW, COLOR_BLACK);
376 int c = getch(); /* refresh, accept single keystroke of input */
378 /* process the command keystroke */
381 finish(0); /* we're done */
384 static void finish(int sig)
388 /* do your non-curses wrapup here */
395 In order to use the screen package, the routines must know about
396 terminal characteristics, and the space for curscr and stdscr must be
397 allocated. These function initscr() does both these things. Since it
398 must allocate space for the windows, it can overflow memory when
399 attempting to do so. On the rare occasions this happens, initscr()
400 will terminate the program with an error message. initscr() must
401 always be called before any of the routines which affect windows are
402 used. If it is not, the program will core dump as soon as either
403 curscr or stdscr are referenced. However, it is usually best to wait
404 to call it until after you are sure you will need it, like after
405 checking for startup errors. Terminal status changing routines like
406 nl() and cbreak() should be called after initscr().
408 Once the screen windows have been allocated, you can set them up for
409 your program. If you want to, say, allow a screen to scroll, use
410 scrollok(). If you want the cursor to be left in place after the last
411 change, use leaveok(). If this isn't done, refresh() will move the
412 cursor to the window's current (y, x) coordinates after updating it.
414 You can create new windows of your own using the functions newwin(),
415 derwin(), and subwin(). The routine delwin() will allow you to get rid
416 of old windows. All the options described above can be applied to any
421 Now that we have set things up, we will want to actually update the
422 terminal. The basic functions used to change what will go on a window
423 are addch() and move(). addch() adds a character at the current (y, x)
424 coordinates. move() changes the current (y, x) coordinates to whatever
425 you want them to be. It returns ERR if you try to move off the window.
426 As mentioned above, you can combine the two into mvaddch() to do both
429 The other output functions, such as addstr() and printw(), all call
430 addch() to add characters to the window.
432 After you have put on the window what you want there, when you want
433 the portion of the terminal covered by the window to be made to look
434 like it, you must call refresh(). In order to optimize finding
435 changes, refresh() assumes that any part of the window not changed
436 since the last refresh() of that window has not been changed on the
437 terminal, i.e., that you have not refreshed a portion of the terminal
438 with an overlapping window. If this is not the case, the routine
439 touchwin() is provided to make it look like the entire window has been
440 changed, thus making refresh() check the whole subsection of the
441 terminal for changes.
443 If you call wrefresh() with curscr as its argument, it will make the
444 screen look like curscr thinks it looks like. This is useful for
445 implementing a command which would redraw the screen in case it get
450 The complementary function to addch() is getch() which, if echo is
451 set, will call addch() to echo the character. Since the screen package
452 needs to know what is on the terminal at all times, if characters are
453 to be echoed, the tty must be in raw or cbreak mode. Since initially
454 the terminal has echoing enabled and is in ordinary ``cooked'' mode,
455 one or the other has to changed before calling getch(); otherwise, the
456 program's output will be unpredictable.
458 When you need to accept line-oriented input in a window, the functions
459 wgetstr() and friends are available. There is even a wscanw() function
460 that can do scanf()(3)-style multi-field parsing on window input.
461 These pseudo-line-oriented functions turn on echoing while they
464 The example code above uses the call keypad(stdscr, TRUE) to enable
465 support for function-key mapping. With this feature, the getch() code
466 watches the input stream for character sequences that correspond to
467 arrow and function keys. These sequences are returned as
468 pseudo-character values. The #define values returned are listed in the
469 curses.h The mapping from sequences to #define values is determined by
470 key_ capabilities in the terminal's terminfo entry.
472 Using Forms Characters
474 The addch() function (and some others, including box() and border())
475 can accept some pseudo-character arguments which are specially defined
476 by ncurses. These are #define values set up in the curses.h header;
477 see there for a complete list (look for the prefix ACS_).
479 The most useful of the ACS defines are the forms-drawing characters.
480 You can use these to draw boxes and simple graphs on the screen. If
481 the terminal does not have such characters, curses.h will map them to
482 a recognizable (though ugly) set of ASCII defaults.
484 Character Attributes and Color
486 The ncurses package supports screen highlights including standout,
487 reverse-video, underline, and blink. It also supports color, which is
488 treated as another kind of highlight.
490 Highlights are encoded, internally, as high bits of the
491 pseudo-character type (chtype) that curses.h uses to represent the
492 contents of a screen cell. See the curses.h header file for a complete
493 list of highlight mask values (look for the prefix A_).
495 There are two ways to make highlights. One is to logical-or the value
496 of the highlights you want into the character argument of an addch()
497 call, or any other output call that takes a chtype argument.
499 The other is to set the current-highlight value. This is logical-or'ed
500 with any highlight you specify the first way. You do this with the
501 functions attron(), attroff(), and attrset(); see the manual pages for
502 details. Color is a special kind of highlight. The package actually
503 thinks in terms of color pairs, combinations of foreground and
504 background colors. The sample code above sets up eight color pairs,
505 all of the guaranteed-available colors on black. Note that each color
506 pair is, in effect, given the name of its foreground color. Any other
507 range of eight non-conflicting values could have been used as the
508 first arguments of the init_pair() values.
510 Once you've done an init_pair() that creates color-pair N, you can use
511 COLOR_PAIR(N) as a highlight that invokes that particular color
512 combination. Note that COLOR_PAIR(N), for constant N, is itself a
513 compile-time constant and can be used in initializers.
517 The ncurses library also provides a mouse interface.
519 NOTE: this facility is specific to ncurses, it is not part of
520 either the XSI Curses standard, nor of System V Release 4, nor BSD
521 curses. System V Release 4 curses contains code with similar
522 interface definitions, however it is not documented. Other than by
523 disassembling the library, we have no way to determine exactly how
524 that mouse code works. Thus, we recommend that you wrap
525 mouse-related code in an #ifdef using the feature macro
526 NCURSES_MOUSE_VERSION so it will not be compiled and linked on
529 Presently, mouse event reporting works in the following environments:
530 * xterm and similar programs such as rxvt.
531 * Linux console, when configured with gpm(1), Alessandro Rubini's
535 The mouse interface is very simple. To activate it, you use the
536 function mousemask(), passing it as first argument a bit-mask that
537 specifies what kinds of events you want your program to be able to
538 see. It will return the bit-mask of events that actually become
539 visible, which may differ from the argument if the mouse device is not
540 capable of reporting some of the event types you specify.
542 Once the mouse is active, your application's command loop should watch
543 for a return value of KEY_MOUSE from wgetch(). When you see this, a
544 mouse event report has been queued. To pick it off the queue, use the
545 function getmouse() (you must do this before the next wgetch(),
546 otherwise another mouse event might come in and make the first one
549 Each call to getmouse() fills a structure (the address of which you'll
550 pass it) with mouse event data. The event data includes zero-origin,
551 screen-relative character-cell coordinates of the mouse pointer. It
552 also includes an event mask. Bits in this mask will be set,
553 corresponding to the event type being reported.
555 The mouse structure contains two additional fields which may be
556 significant in the future as ncurses interfaces to new kinds of
557 pointing device. In addition to x and y coordinates, there is a slot
558 for a z coordinate; this might be useful with touch-screens that can
559 return a pressure or duration parameter. There is also a device ID
560 field, which could be used to distinguish between multiple pointing
563 The class of visible events may be changed at any time via
564 mousemask(). Events that can be reported include presses, releases,
565 single-, double- and triple-clicks (you can set the maximum
566 button-down time for clicks). If you don't make clicks visible, they
567 will be reported as press-release pairs. In some environments, the
568 event mask may include bits reporting the state of shift, alt, and
569 ctrl keys on the keyboard during the event.
571 A function to check whether a mouse event fell within a given window
572 is also supplied. You can use this to see whether a given window
573 should consider a mouse event relevant to it.
575 Because mouse event reporting will not be available in all
576 environments, it would be unwise to build ncurses applications that
577 require the use of a mouse. Rather, you should use the mouse as a
578 shortcut for point-and-shoot commands your application would normally
579 accept from the keyboard. Two of the test games in the ncurses
580 distribution (bs and knight) contain code that illustrates how this
583 See the manual page curs_mouse(3X) for full details of the
584 mouse-interface functions.
588 In order to clean up after the ncurses routines, the routine endwin()
589 is provided. It restores tty modes to what they were when initscr()
590 was first called, and moves the cursor down to the lower-left corner.
591 Thus, anytime after the call to initscr, endwin() should be called
594 Function Descriptions
596 We describe the detailed behavior of some important curses functions
597 here, as a supplement to the manual page descriptions.
599 Initialization and Wrapup
602 The first function called should almost always be initscr().
603 This will determine the terminal type and initialize curses
604 data structures. initscr() also arranges that the first call to
605 refresh() will clear the screen. If an error occurs a message
606 is written to standard error and the program exits. Otherwise
607 it returns a pointer to stdscr. A few functions may be called
608 before initscr (slk_init(), filter(), ripofflines(), use_env(),
609 and, if you are using multiple terminals, newterm().)
612 Your program should always call endwin() before exiting or
613 shelling out of the program. This function will restore tty
614 modes, move the cursor to the lower left corner of the screen,
615 reset the terminal into the proper non-visual mode. Calling
616 refresh() or doupdate() after a temporary escape from the
617 program will restore the ncurses screen from before the escape.
619 newterm(type, ofp, ifp)
620 A program which outputs to more than one terminal should use
621 newterm() instead of initscr(). newterm() should be called once
622 for each terminal. It returns a variable of type SCREEN * which
623 should be saved as a reference to that terminal. The arguments
624 are the type of the terminal (a string) and FILE pointers for
625 the output and input of the terminal. If type is NULL then the
626 environment variable $TERM is used. endwin() should called once
627 at wrapup time for each terminal opened using this function.
630 This function is used to switch to a different terminal
631 previously opened by newterm(). The screen reference for the
632 new terminal is passed as the parameter. The previous terminal
633 is returned by the function. All other calls affect only the
637 The inverse of newterm(); deallocates the data structures
638 associated with a given SCREEN reference.
640 Causing Output to the Terminal
642 refresh() and wrefresh(win)
643 These functions must be called to actually get any output on
644 the terminal, as other routines merely manipulate data
645 structures. wrefresh() copies the named window to the physical
646 terminal screen, taking into account what is already there in
647 order to do optimizations. refresh() does a refresh of
648 stdscr(). Unless leaveok() has been enabled, the physical
649 cursor of the terminal is left at the location of the window's
652 doupdate() and wnoutrefresh(win)
653 These two functions allow multiple updates with more efficiency
654 than wrefresh. To use them, it is important to understand how
655 curses works. In addition to all the window structures, curses
656 keeps two data structures representing the terminal screen: a
657 physical screen, describing what is actually on the screen, and
658 a virtual screen, describing what the programmer wants to have
659 on the screen. wrefresh works by first copying the named window
660 to the virtual screen (wnoutrefresh()), and then calling the
661 routine to update the screen (doupdate()). If the programmer
662 wishes to output several windows at once, a series of calls to
663 wrefresh will result in alternating calls to wnoutrefresh() and
664 doupdate(), causing several bursts of output to the screen. By
665 calling wnoutrefresh() for each window, it is then possible to
666 call doupdate() once, resulting in only one burst of output,
667 with fewer total characters transmitted (this also avoids a
668 visually annoying flicker at each update).
670 Low-Level Capability Access
672 setupterm(term, filenum, errret)
673 This routine is called to initialize a terminal's description,
674 without setting up the curses screen structures or changing the
675 tty-driver mode bits. term is the character string representing
676 the name of the terminal being used. filenum is the UNIX file
677 descriptor of the terminal to be used for output. errret is a
678 pointer to an integer, in which a success or failure indication
679 is returned. The values returned can be 1 (all is well), 0 (no
680 such terminal), or -1 (some problem locating the terminfo
683 The value of term can be given as NULL, which will cause the
684 value of TERM in the environment to be used. The errret pointer
685 can also be given as NULL, meaning no error code is wanted. If
686 errret is defaulted, and something goes wrong, setupterm() will
687 print an appropriate error message and exit, rather than
688 returning. Thus, a simple program can call setupterm(0, 1, 0)
689 and not worry about initialization errors.
691 After the call to setupterm(), the global variable cur_term is
692 set to point to the current structure of terminal capabilities.
693 By calling setupterm() for each terminal, and saving and
694 restoring cur_term, it is possible for a program to use two or
695 more terminals at once. Setupterm() also stores the names
696 section of the terminal description in the global character
697 array ttytype[]. Subsequent calls to setupterm() will overwrite
698 this array, so you'll have to save it yourself if need be.
702 NOTE: These functions are not part of the standard curses API!
705 This function can be used to explicitly set a trace level. If
706 the trace level is nonzero, execution of your program will
707 generate a file called `trace' in the current working directory
708 containing a report on the library's actions. Higher trace
709 levels enable more detailed (and verbose) reporting -- see
710 comments attached to TRACE_ defines in the curses.h file for
711 details. (It is also possible to set a trace level by assigning
712 a trace level value to the environment variable NCURSES_TRACE).
715 This function can be used to output your own debugging
716 information. It is only available only if you link with
717 -lncurses_g. It can be used the same way as printf(), only it
718 outputs a newline after the end of arguments. The output goes
719 to a file called trace in the current directory.
721 Trace logs can be difficult to interpret due to the sheer volume of
722 data dumped in them. There is a script called tracemunch included with
723 the ncurses distribution that can alleviate this problem somewhat; it
724 compacts long sequences of similar operations into more succinct
725 single-line pseudo-operations. These pseudo-ops can be distinguished
726 by the fact that they are named in capital letters.
728 Hints, Tips, and Tricks
730 The ncurses manual pages are a complete reference for this library. In
731 the remainder of this document, we discuss various useful methods that
732 may not be obvious from the manual page descriptions.
734 Some Notes of Caution
736 If you find yourself thinking you need to use noraw() or nocbreak(),
737 think again and move carefully. It's probably better design to use
738 getstr() or one of its relatives to simulate cooked mode. The noraw()
739 and nocbreak() functions try to restore cooked mode, but they may end
740 up clobbering some control bits set before you started your
741 application. Also, they have always been poorly documented, and are
742 likely to hurt your application's usability with other curses
745 Bear in mind that refresh() is a synonym for wrefresh(stdscr). Don't
746 try to mix use of stdscr with use of windows declared by newwin(); a
747 refresh() call will blow them off the screen. The right way to handle
748 this is to use subwin(), or not touch stdscr at all and tile your
749 screen with declared windows which you then wnoutrefresh() somewhere
750 in your program event loop, with a single doupdate() call to trigger
753 You are much less likely to run into problems if you design your
754 screen layouts to use tiled rather than overlapping windows.
755 Historically, curses support for overlapping windows has been weak,
756 fragile, and poorly documented. The ncurses library is not yet an
757 exception to this rule.
759 There is a panels library included in the ncurses distribution that
760 does a pretty good job of strengthening the overlapping-windows
763 Try to avoid using the global variables LINES and COLS. Use getmaxyx()
764 on the stdscr context instead. Reason: your code may be ported to run
765 in an environment with window resizes, in which case several screens
766 could be open with different sizes.
768 Temporarily Leaving NCURSES Mode
770 Sometimes you will want to write a program that spends most of its
771 time in screen mode, but occasionally returns to ordinary `cooked'
772 mode. A common reason for this is to support shell-out. This behavior
773 is simple to arrange in ncurses.
775 To leave ncurses mode, call endwin() as you would if you were
776 intending to terminate the program. This will take the screen back to
777 cooked mode; you can do your shell-out. When you want to return to
778 ncurses mode, simply call refresh() or doupdate(). This will repaint
781 There is a boolean function, isendwin(), which code can use to test
782 whether ncurses screen mode is active. It returns TRUE in the interval
783 between an endwin() call and the following refresh(), FALSE otherwise.
785 Here is some sample code for shellout:
786 addstr("Shelling out...");
787 def_prog_mode(); /* save current tty modes */
788 endwin(); /* restore original tty modes */
789 system("sh"); /* run shell */
790 addstr("returned.\n"); /* prepare return message */
791 refresh(); /* restore save modes, repaint screen */
793 Using NCURSES under XTERM
795 A resize operation in X sends SIGWINCH to the application running
796 under xterm. The ncurses library provides an experimental signal
797 handler, but in general does not catch this signal, because it cannot
798 know how you want the screen re-painted. You will usually have to
799 write the SIGWINCH handler yourself. Ncurses can give you some help.
801 The easiest way to code your SIGWINCH handler is to have it do an
802 endwin, followed by an refresh and a screen repaint you code yourself.
803 The refresh will pick up the new screen size from the xterm's
806 That is the standard way, of course (it even works with some vendor's
807 curses implementations). Its drawback is that it clears the screen to
808 reinitialize the display, and does not resize subwindows which must be
809 shrunk. Ncurses provides an extension which works better, the
810 resizeterm function. That function ensures that all windows are
811 limited to the new screen dimensions, and pads stdscr with blanks if
812 the screen is larger.
814 Finally, ncurses can be configured to provide its own SIGWINCH
815 handler, based on resizeterm.
817 Handling Multiple Terminal Screens
819 The initscr() function actually calls a function named newterm() to do
820 most of its work. If you are writing a program that opens multiple
821 terminals, use newterm() directly.
823 For each call, you will have to specify a terminal type and a pair of
824 file pointers; each call will return a screen reference, and stdscr
825 will be set to the last one allocated. You will switch between screens
826 with the set_term call. Note that you will also have to call
827 def_shell_mode and def_prog_mode on each tty yourself.
829 Testing for Terminal Capabilities
831 Sometimes you may want to write programs that test for the presence of
832 various capabilities before deciding whether to go into ncurses mode.
833 An easy way to do this is to call setupterm(), then use the functions
834 tigetflag(), tigetnum(), and tigetstr() to do your testing.
836 A particularly useful case of this often comes up when you want to
837 test whether a given terminal type should be treated as `smart'
838 (cursor-addressable) or `stupid'. The right way to test this is to see
839 if the return value of tigetstr("cup") is non-NULL. Alternatively, you
840 can include the term.h file and test the value of the macro
845 Use the addchstr() family of functions for fast screen-painting of
846 text when you know the text doesn't contain any control characters.
847 Try to make attribute changes infrequent on your screens. Don't use
848 the immedok() option!
850 Special Features of NCURSES
852 The wresize() function allows you to resize a window in place. The
853 associated resizeterm() function simplifies the construction of
854 SIGWINCH handlers, for resizing all windows.
856 The define_key() function allows you to define at runtime function-key
857 control sequences which are not in the terminal description. The
858 keyok() function allows you to temporarily enable or disable
859 interpretation of any function-key control sequence.
861 The use_default_colors() function allows you to construct applications
862 which can use the terminal's default foreground and background colors
863 as an additional "default" color. Several terminal emulators support
864 this feature, which is based on ISO 6429.
866 Ncurses supports up 16 colors, unlike SVr4 curses which defines only
867 8. While most terminals which provide color allow only 8 colors, about
868 a quarter (including XFree86 xterm) support 16 colors.
870 Compatibility with Older Versions
872 Despite our best efforts, there are some differences between ncurses
873 and the (undocumented!) behavior of older curses implementations.
874 These arise from ambiguities or omissions in the documentation of the
877 Refresh of Overlapping Windows
879 If you define two windows A and B that overlap, and then alternately
880 scribble on and refresh them, the changes made to the overlapping
881 region under historic curses versions were often not documented
884 To understand why this is a problem, remember that screen updates are
885 calculated between two representations of the entire display. The
886 documentation says that when you refresh a window, it is first copied
887 to to the virtual screen, and then changes are calculated to update
888 the physical screen (and applied to the terminal). But "copied to" is
889 not very specific, and subtle differences in how copying works can
890 produce different behaviors in the case where two overlapping windows
891 are each being refreshed at unpredictable intervals.
893 What happens to the overlapping region depends on what wnoutrefresh()
894 does with its argument -- what portions of the argument window it
895 copies to the virtual screen. Some implementations do "change copy",
896 copying down only locations in the window that have changed (or been
897 marked changed with wtouchln() and friends). Some implementations do
898 "entire copy", copying all window locations to the virtual screen
899 whether or not they have changed.
901 The ncurses library itself has not always been consistent on this
902 score. Due to a bug, versions 1.8.7 to 1.9.8a did entire copy.
903 Versions 1.8.6 and older, and versions 1.9.9 and newer, do change
906 For most commercial curses implementations, it is not documented and
907 not known for sure (at least not to the ncurses maintainers) whether
908 they do change copy or entire copy. We know that System V release 3
909 curses has logic in it that looks like an attempt to do change copy,
910 but the surrounding logic and data representations are sufficiently
911 complex, and our knowledge sufficiently indirect, that it's hard to
912 know whether this is reliable. It is not clear what the SVr4
913 documentation and XSI standard intend. The XSI Curses standard barely
914 mentions wnoutrefresh(); the SVr4 documents seem to be describing
915 entire-copy, but it is possible with some effort and straining to read
918 It might therefore be unwise to rely on either behavior in programs
919 that might have to be linked with other curses implementations.
920 Instead, you can do an explicit touchwin() before the wnoutrefresh()
921 call to guarantee an entire-contents copy anywhere.
923 The really clean way to handle this is to use the panels library. If,
924 when you want a screen update, you do update_panels(), it will do all
925 the necessary wnoutrfresh() calls for whatever panel stacking order
926 you have defined. Then you can do one doupdate() and there will be a
927 single burst of physical I/O that will do all your updates.
931 If you have been using a very old versions of ncurses (1.8.7 or older)
932 you may be surprised by the behavior of the erase functions. In older
933 versions, erased areas of a window were filled with a blank modified
934 by the window's current attribute (as set by wattrset(), wattron(),
935 wattroff() and friends).
937 In newer versions, this is not so. Instead, the attribute of erased
938 blanks is normal unless and until it is modified by the functions
939 bkgdset() or wbkgdset().
941 This change in behavior conforms ncurses to System V Release 4 and the
944 XSI Curses Conformance
946 The ncurses library is intended to be base-level conformant with the
947 XSI Curses standard from X/Open. Many extended-level features (in
948 fact, almost all features not directly concerned with wide characters
949 and internationalization) are also supported.
951 One effect of XSI conformance is the change in behavior described
952 under "Background Erase -- Compatibility with Old Versions".
954 Also, ncurses meets the XSI requirement that every macro entry point
955 have a corresponding function which may be linked (and will be
956 prototype-checked) if the macro definition is disabled with #undef.
960 The ncurses library by itself provides good support for screen
961 displays in which the windows are tiled (non-overlapping). In the more
962 general case that windows may overlap, you have to use a series of
963 wnoutrefresh() calls followed by a doupdate(), and be careful about
964 the order you do the window refreshes in. It has to be bottom-upwards,
965 otherwise parts of windows that should be obscured will show through.
967 When your interface design is such that windows may dive deeper into
968 the visibility stack or pop to the top at runtime, the resulting
969 book-keeping can be tedious and difficult to get right. Hence the
972 The panel library first appeared in AT&T System V. The version
973 documented here is the panel code distributed with ncurses.
975 Compiling With the Panels Library
977 Your panels-using modules must import the panels library declarations
981 and must be linked explicitly with the panels library using an -lpanel
982 argument. Note that they must also link the ncurses library with
983 -lncurses. Many linkers are two-pass and will accept either order, but
984 it is still good practice to put -lpanel first and -lncurses second.
988 A panel object is a window that is implicitly treated as part of a
989 deck including all other panel objects. The deck has an implicit
990 bottom-to-top visibility order. The panels library includes an update
991 function (analogous to refresh()) that displays all panels in the deck
992 in the proper order to resolve overlaps. The standard window, stdscr,
993 is considered below all panels.
995 Details on the panels functions are available in the man pages. We'll
996 just hit the highlights here.
998 You create a panel from a window by calling new_panel() on a window
999 pointer. It then becomes the top of the deck. The panel's window is
1000 available as the value of panel_window() called with the panel pointer
1003 You can delete a panel (removing it from the deck) with del_panel.
1004 This will not deallocate the associated window; you have to do that
1005 yourself. You can replace a panel's window with a different window by
1006 calling replace_window. The new window may be of different size; the
1007 panel code will re-compute all overlaps. This operation doesn't change
1008 the panel's position in the deck.
1010 To move a panel's window, use move_panel(). The mvwin() function on
1011 the panel's window isn't sufficient because it doesn't update the
1012 panels library's representation of where the windows are. This
1013 operation leaves the panel's depth, contents, and size unchanged.
1015 Two functions (top_panel(), bottom_panel()) are provided for
1016 rearranging the deck. The first pops its argument window to the top of
1017 the deck; the second sends it to the bottom. Either operation leaves
1018 the panel's screen location, contents, and size unchanged.
1020 The function update_panels() does all the wnoutrefresh() calls needed
1021 to prepare for doupdate() (which you must call yourself, afterwards).
1023 Typically, you will want to call update_panels() and doupdate() just
1024 before accepting command input, once in each cycle of interaction with
1025 the user. If you call update_panels() after each and every panel
1026 write, you'll generate a lot of unnecessary refresh activity and
1029 Panels, Input, and the Standard Screen
1031 You shouldn't mix wnoutrefresh() or wrefresh() operations with panels
1032 code; this will work only if the argument window is either in the top
1033 panel or unobscured by any other panels.
1035 The stsdcr window is a special case. It is considered below all
1036 panels. Because changes to panels may obscure parts of stdscr, though,
1037 you should call update_panels() before doupdate() even when you only
1040 Note that wgetch automatically calls wrefresh. Therefore, before
1041 requesting input from a panel window, you need to be sure that the
1042 panel is totally unobscured.
1044 There is presently no way to display changes to one obscured panel
1045 without repainting all panels.
1049 It's possible to remove a panel from the deck temporarily; use
1050 hide_panel for this. Use show_panel() to render it visible again. The
1051 predicate function panel_hidden tests whether or not a panel is
1054 The panel_update code ignores hidden panels. You cannot do top_panel()
1055 or bottom_panel on a hidden panel(). Other panels operations are
1058 Miscellaneous Other Facilities
1060 It's possible to navigate the deck using the functions panel_above()
1061 and panel_below. Handed a panel pointer, they return the panel above
1062 or below that panel. Handed NULL, they return the bottom-most or
1065 Every panel has an associated user pointer, not used by the panel
1066 code, to which you can attach application data. See the man page
1067 documentation of set_panel_userptr() and panel_userptr for details.
1071 A menu is a screen display that assists the user to choose some subset
1072 of a given set of items. The menu library is a curses extension that
1073 supports easy programming of menu hierarchies with a uniform but
1076 The menu library first appeared in AT&T System V. The version
1077 documented here is the menu code distributed with ncurses.
1079 Compiling With the menu Library
1081 Your menu-using modules must import the menu library declarations with
1084 and must be linked explicitly with the menus library using an -lmenu
1085 argument. Note that they must also link the ncurses library with
1086 -lncurses. Many linkers are two-pass and will accept either order, but
1087 it is still good practice to put -lmenu first and -lncurses second.
1091 The menus created by this library consist of collections of items
1092 including a name string part and a description string part. To make
1093 menus, you create groups of these items and connect them with menu
1096 The menu can then by posted, that is written to an associated window.
1097 Actually, each menu has two associated windows; a containing window in
1098 which the programmer can scribble titles or borders, and a subwindow
1099 in which the menu items proper are displayed. If this subwindow is too
1100 small to display all the items, it will be a scrollable viewport on
1101 the collection of items.
1103 A menu may also be unposted (that is, undisplayed), and finally freed
1104 to make the storage associated with it and its items available for
1107 The general flow of control of a menu program looks like this:
1108 1. Initialize curses.
1109 2. Create the menu items, using new_item().
1110 3. Create the menu using new_menu().
1111 4. Post the menu using menu_post().
1112 5. Refresh the screen.
1113 6. Process user requests via an input loop.
1114 7. Unpost the menu using menu_unpost().
1115 8. Free the menu, using free_menu().
1116 9. Free the items using free_item().
1117 10. Terminate curses.
1121 Menus may be multi-valued or (the default) single-valued (see the
1122 manual page menu_opts(3x) to see how to change the default). Both
1123 types always have a current item.
1125 From a single-valued menu you can read the selected value simply by
1126 looking at the current item. From a multi-valued menu, you get the
1127 selected set by looping through the items applying the item_value()
1128 predicate function. Your menu-processing code can use the function
1129 set_item_value() to flag the items in the select set.
1131 Menu items can be made unselectable using set_item_opts() or
1132 item_opts_off() with the O_SELECTABLE argument. This is the only
1133 option so far defined for menus, but it is good practice to code as
1134 though other option bits might be on.
1138 The menu library calculates a minimum display size for your window,
1139 based on the following variables:
1141 * The number and maximum length of the menu items
1142 * Whether the O_ROWMAJOR option is enabled
1143 * Whether display of descriptions is enabled
1144 * Whatever menu format may have been set by the programmer
1145 * The length of the menu mark string used for highlighting selected
1148 The function set_menu_format() allows you to set the maximum size of
1149 the viewport or menu page that will be used to display menu items. You
1150 can retrieve any format associated with a menu with menu_format(). The
1151 default format is rows=16, columns=1.
1153 The actual menu page may be smaller than the format size. This depends
1154 on the item number and size and whether O_ROWMAJOR is on. This option
1155 (on by default) causes menu items to be displayed in a `raster-scan'
1156 pattern, so that if more than one item will fit horizontally the first
1157 couple of items are side-by-side in the top row. The alternative is
1158 column-major display, which tries to put the first several items in
1161 As mentioned above, a menu format not large enough to allow all items
1162 to fit on-screen will result in a menu display that is vertically
1165 You can scroll it with requests to the menu driver, which will be
1166 described in the section on menu input handling.
1168 Each menu has a mark string used to visually tag selected items; see
1169 the menu_mark(3x) manual page for details. The mark string length also
1170 influences the menu page size.
1172 The function scale_menu() returns the minimum display size that the
1173 menu code computes from all these factors. There are other menu
1174 display attributes including a select attribute, an attribute for
1175 selectable items, an attribute for unselectable items, and a pad
1176 character used to separate item name text from description text. These
1177 have reasonable defaults which the library allows you to change (see
1178 the menu_attribs(3x) manual page.
1182 Each menu has, as mentioned previously, a pair of associated windows.
1183 Both these windows are painted when the menu is posted and erased when
1184 the menu is unposted.
1186 The outer or frame window is not otherwise touched by the menu
1187 routines. It exists so the programmer can associate a title, a border,
1188 or perhaps help text with the menu and have it properly refreshed or
1189 erased at post/unpost time. The inner window or subwindow is where the
1190 current menu page is displayed.
1192 By default, both windows are stdscr. You can set them with the
1193 functions in menu_win(3x).
1195 When you call menu_post(), you write the menu to its subwindow. When
1196 you call menu_unpost(), you erase the subwindow, However, neither of
1197 these actually modifies the screen. To do that, call wrefresh() or
1200 Processing Menu Input
1202 The main loop of your menu-processing code should call menu_driver()
1203 repeatedly. The first argument of this routine is a menu pointer; the
1204 second is a menu command code. You should write an input-fetching
1205 routine that maps input characters to menu command codes, and pass its
1206 output to menu_driver(). The menu command codes are fully documented
1209 The simplest group of command codes is REQ_NEXT_ITEM, REQ_PREV_ITEM,
1210 REQ_FIRST_ITEM, REQ_LAST_ITEM, REQ_UP_ITEM, REQ_DOWN_ITEM,
1211 REQ_LEFT_ITEM, REQ_RIGHT_ITEM. These change the currently selected
1212 item. These requests may cause scrolling of the menu page if it only
1213 partially displayed.
1215 There are explicit requests for scrolling which also change the
1216 current item (because the select location does not change, but the
1217 item there does). These are REQ_SCR_DLINE, REQ_SCR_ULINE,
1218 REQ_SCR_DPAGE, and REQ_SCR_UPAGE.
1220 The REQ_TOGGLE_ITEM selects or deselects the current item. It is for
1221 use in multi-valued menus; if you use it with O_ONEVALUE on, you'll
1222 get an error return (E_REQUEST_DENIED).
1224 Each menu has an associated pattern buffer. The menu_driver() logic
1225 tries to accumulate printable ASCII characters passed in in that
1226 buffer; when it matches a prefix of an item name, that item (or the
1227 next matching item) is selected. If appending a character yields no
1228 new match, that character is deleted from the pattern buffer, and
1229 menu_driver() returns E_NO_MATCH.
1231 Some requests change the pattern buffer directly: REQ_CLEAR_PATTERN,
1232 REQ_BACK_PATTERN, REQ_NEXT_MATCH, REQ_PREV_MATCH. The latter two are
1233 useful when pattern buffer input matches more than one item in a
1236 Each successful scroll or item navigation request clears the pattern
1237 buffer. It is also possible to set the pattern buffer explicitly with
1240 Finally, menu driver requests above the constant MAX_COMMAND are
1241 considered application-specific commands. The menu_driver() code
1242 ignores them and returns E_UNKNOWN_COMMAND.
1244 Miscellaneous Other Features
1246 Various menu options can affect the processing and visual appearance
1247 and input processing of menus. See menu_opts(3x) for details.
1249 It is possible to change the current item from application code; this
1250 is useful if you want to write your own navigation requests. It is
1251 also possible to explicitly set the top row of the menu display. See
1252 mitem_current(3x). If your application needs to change the menu
1253 subwindow cursor for any reason, pos_menu_cursor() will restore it to
1254 the correct location for continuing menu driver processing.
1256 It is possible to set hooks to be called at menu initialization and
1257 wrapup time, and whenever the selected item changes. See
1260 Each item, and each menu, has an associated user pointer on which you
1261 can hang application data. See mitem_userptr(3x) and menu_userptr(3x).
1265 The form library is a curses extension that supports easy programming
1266 of on-screen forms for data entry and program control.
1268 The form library first appeared in AT&T System V. The version
1269 documented here is the form code distributed with ncurses.
1271 Compiling With the form Library
1273 Your form-using modules must import the form library declarations with
1276 and must be linked explicitly with the forms library using an -lform
1277 argument. Note that they must also link the ncurses library with
1278 -lncurses. Many linkers are two-pass and will accept either order, but
1279 it is still good practice to put -lform first and -lncurses second.
1283 A form is a collection of fields; each field may be either a label
1284 (explanatory text) or a data-entry location. Long forms may be
1285 segmented into pages; each entry to a new page clears the screen.
1287 To make forms, you create groups of fields and connect them with form
1288 frame objects; the form library makes this relatively simple.
1290 Once defined, a form can be posted, that is written to an associated
1291 window. Actually, each form has two associated windows; a containing
1292 window in which the programmer can scribble titles or borders, and a
1293 subwindow in which the form fields proper are displayed.
1295 As the form user fills out the posted form, navigation and editing
1296 keys support movement between fields, editing keys support modifying
1297 field, and plain text adds to or changes data in a current field. The
1298 form library allows you (the forms designer) to bind each navigation
1299 and editing key to any keystroke accepted by curses Fields may have
1300 validation conditions on them, so that they check input data for type
1301 and value. The form library supplies a rich set of pre-defined field
1302 types, and makes it relatively easy to define new ones.
1304 Once its transaction is completed (or aborted), a form may be unposted
1305 (that is, undisplayed), and finally freed to make the storage
1306 associated with it and its items available for re-use.
1308 The general flow of control of a form program looks like this:
1309 1. Initialize curses.
1310 2. Create the form fields, using new_field().
1311 3. Create the form using new_form().
1312 4. Post the form using form_post().
1313 5. Refresh the screen.
1314 6. Process user requests via an input loop.
1315 7. Unpost the form using form_unpost().
1316 8. Free the form, using free_form().
1317 9. Free the fields using free_field().
1318 10. Terminate curses.
1320 Note that this looks much like a menu program; the form library
1321 handles tasks which are in many ways similar, and its interface was
1322 obviously designed to resemble that of the menu library wherever
1325 In forms programs, however, the `process user requests' is somewhat
1326 more complicated than for menus. Besides menu-like navigation
1327 operations, the menu driver loop has to support field editing and data
1330 Creating and Freeing Fields and Forms
1332 The basic function for creating fields is new_field():
1334 FIELD *new_field(int height, int width, /* new field size */
1335 int top, int left, /* upper left corner */
1336 int offscreen, /* number of offscreen rows */
1337 int nbuf); /* number of working buffers */
1339 Menu items always occupy a single row, but forms fields may have
1340 multiple rows. So new_field() requires you to specify a width and
1341 height (the first two arguments, which mist both be greater than
1344 You must also specify the location of the field's upper left corner on
1345 the screen (the third and fourth arguments, which must be zero or
1346 greater). Note that these coordinates are relative to the form
1347 subwindow, which will coincide with stdscr by default but need not be
1348 stdscr if you've done an explicit set_form_window() call.
1350 The fifth argument allows you to specify a number of off-screen rows.
1351 If this is zero, the entire field will always be displayed. If it is
1352 nonzero, the form will be scrollable, with only one screen-full
1353 (initially the top part) displayed at any given time. If you make a
1354 field dynamic and grow it so it will no longer fit on the screen, the
1355 form will become scrollable even if the offscreen argument was
1358 The forms library allocates one working buffer per field; the size of
1359 each buffer is ((height + offscreen)*width + 1, one character for each
1360 position in the field plus a NUL terminator. The sixth argument is the
1361 number of additional data buffers to allocate for the field; your
1362 application can use them for its own purposes.
1364 FIELD *dup_field(FIELD *field, /* field to copy */
1365 int top, int left); /* location of new copy */
1367 The function dup_field() duplicates an existing field at a new
1368 location. Size and buffering information are copied; some attribute
1369 flags and status bits are not (see the form_field_new(3X) for
1372 FIELD *link_field(FIELD *field, /* field to copy */
1373 int top, int left); /* location of new copy */
1375 The function link_field() also duplicates an existing field at a new
1376 location. The difference from dup_field() is that it arranges for the
1377 new field's buffer to be shared with the old one.
1379 Besides the obvious use in making a field editable from two different
1380 form pages, linked fields give you a way to hack in dynamic labels. If
1381 you declare several fields linked to an original, and then make them
1382 inactive, changes from the original will still be propagated to the
1385 As with duplicated fields, linked fields have attribute bits separate
1388 As you might guess, all these field-allocations return NULL if the
1389 field allocation is not possible due to an out-of-memory error or
1390 out-of-bounds arguments.
1392 To connect fields to a form, use
1394 FORM *new_form(FIELD **fields);
1396 This function expects to see a NULL-terminated array of field
1397 pointers. Said fields are connected to a newly-allocated form object;
1398 its address is returned (or else NULL if the allocation fails).
1400 Note that new_field() does not copy the pointer array into private
1401 storage; if you modify the contents of the pointer array during forms
1402 processing, all manner of bizarre things might happen. Also note that
1403 any given field may only be connected to one form.
1405 The functions free_field() and free_form are available to free field
1406 and form objects. It is an error to attempt to free a field connected
1407 to a form, but not vice-versa; thus, you will generally free your form
1410 Fetching and Changing Field Attributes
1412 Each form field has a number of location and size attributes
1413 associated with it. There are other field attributes used to control
1414 display and editing of the field. Some (for example, the O_STATIC bit)
1415 involve sufficient complications to be covered in sections of their
1416 own later on. We cover the functions used to get and set several basic
1419 When a field is created, the attributes not specified by the new_field
1420 function are copied from an invisible system default field. In
1421 attribute-setting and -fetching functions, the argument NULL is taken
1422 to mean this field. Changes to it persist as defaults until your forms
1423 application terminates.
1425 Fetching Size and Location Data
1427 You can retrieve field sizes and locations through:
1429 int field_info(FIELD *field, /* field from which to fetch */
1430 int *height, *int width, /* field size */
1431 int *top, int *left, /* upper left corner */
1432 int *offscreen, /* number of offscreen rows */
1433 int *nbuf); /* number of working buffers */
1435 This function is a sort of inverse of new_field(); instead of setting
1436 size and location attributes of a new field, it fetches them from an
1439 Changing the Field Location
1441 It is possible to move a field's location on the screen:
1443 int move_field(FIELD *field, /* field to alter */
1444 int top, int left); /* new upper-left corner */
1446 You can, of course. query the current location through field_info().
1448 The Justification Attribute
1450 One-line fields may be unjustified, justified right, justified left,
1451 or centered. Here is how you manipulate this attribute:
1453 int set_field_just(FIELD *field, /* field to alter */
1454 int justmode); /* mode to set */
1456 int field_just(FIELD *field); /* fetch mode of field */
1458 The mode values accepted and returned by this functions are
1459 preprocessor macros NO_JUSTIFICATION, JUSTIFY_RIGHT, JUSTIFY_LEFT, or
1462 Field Display Attributes
1464 For each field, you can set a foreground attribute for entered
1465 characters, a background attribute for the entire field, and a pad
1466 character for the unfilled portion of the field. You can also control
1467 pagination of the form.
1469 This group of four field attributes controls the visual appearance of
1470 the field on the screen, without affecting in any way the data in the
1473 int set_field_fore(FIELD *field, /* field to alter */
1474 chtype attr); /* attribute to set */
1476 chtype field_fore(FIELD *field); /* field to query */
1478 int set_field_back(FIELD *field, /* field to alter */
1479 chtype attr); /* attribute to set */
1481 chtype field_back(FIELD *field); /* field to query */
1483 int set_field_pad(FIELD *field, /* field to alter */
1484 int pad); /* pad character to set */
1486 chtype field_pad(FIELD *field);
1488 int set_new_page(FIELD *field, /* field to alter */
1489 int flag); /* TRUE to force new page */
1491 chtype new_page(FIELD *field); /* field to query */
1493 The attributes set and returned by the first four functions are normal
1494 curses(3x) display attribute values (A_STANDOUT, A_BOLD, A_REVERSE
1495 etc). The page bit of a field controls whether it is displayed at the
1496 start of a new form screen.
1500 There is also a large collection of field option bits you can set to
1501 control various aspects of forms processing. You can manipulate them
1502 with these functions:
1503 int set_field_opts(FIELD *field, /* field to alter */
1504 int attr); /* attribute to set */
1506 int field_opts_on(FIELD *field, /* field to alter */
1507 int attr); /* attributes to turn on */
1509 int field_opts_off(FIELD *field, /* field to alter */
1510 int attr); /* attributes to turn off */
1512 int field_opts(FIELD *field); /* field to query */
1514 By default, all options are on. Here are the available option bits:
1517 Controls whether the field is visible on the screen. Can be
1518 used during form processing to hide or pop up fields depending
1519 on the value of parent fields.
1522 Controls whether the field is active during forms processing
1523 (i.e. visited by form navigation keys). Can be used to make
1524 labels or derived fields with buffer values alterable by the
1525 forms application, not the user.
1528 Controls whether data is displayed during field entry. If this
1529 option is turned off on a field, the library will accept and
1530 edit data in that field, but it will not be displayed and the
1531 visible field cursor will not move. You can turn off the
1532 O_PUBLIC bit to define password fields.
1535 Controls whether the field's data can be modified. When this
1536 option is off, all editing requests except REQ_PREV_CHOICE and
1537 REQ_NEXT_CHOICE will fail. Such read-only fields may be useful
1541 Controls word-wrapping in multi-line fields. Normally, when any
1542 character of a (blank-separated) word reaches the end of the
1543 current line, the entire word is wrapped to the next line
1544 (assuming there is one). When this option is off, the word will
1545 be split across the line break.
1548 Controls field blanking. When this option is on, entering a
1549 character at the first field position erases the entire field
1550 (except for the just-entered character).
1553 Controls automatic skip to next field when this one fills.
1554 Normally, when the forms user tries to type more data into a
1555 field than will fit, the editing location jumps to next field.
1556 When this option is off, the user's cursor will hang at the end
1557 of the field. This option is ignored in dynamic fields that
1558 have not reached their size limit.
1561 Controls whether validation is applied to blank fields.
1562 Normally, it is not; the user can leave a field blank without
1563 invoking the usual validation check on exit. If this option is
1564 off on a field, exit from it will invoke a validation check.
1567 Controls whether validation occurs on every exit, or only after
1568 the field is modified. Normally the latter is true. Setting
1569 O_PASSOK may be useful if your field's validation function may
1570 change during forms processing.
1573 Controls whether the field is fixed to its initial dimensions.
1574 If you turn this off, the field becomes dynamic and will
1575 stretch to fit entered data.
1577 A field's options cannot be changed while the field is currently
1578 selected. However, options may be changed on posted fields that are
1581 The option values are bit-masks and can be composed with logical-or in
1586 Every field has a status flag, which is set to FALSE when the field is
1587 created and TRUE when the value in field buffer 0 changes. This flag
1588 can be queried and set directly:
1590 int set_field_status(FIELD *field, /* field to alter */
1591 int status); /* mode to set */
1593 int field_status(FIELD *field); /* fetch mode of field */
1595 Setting this flag under program control can be useful if you use the
1596 same form repeatedly, looking for modified fields each time.
1598 Calling field_status() on a field not currently selected for input
1599 will return a correct value. Calling field_status() on a field that is
1600 currently selected for input may not necessarily give a correct field
1601 status value, because entered data isn't necessarily copied to buffer
1602 zero before the exit validation check. To guarantee that the returned
1603 status value reflects reality, call field_status() either (1) in the
1604 field's exit validation check routine, (2) from the field's or form's
1605 initialization or termination hooks, or (3) just after a
1606 REQ_VALIDATION request has been processed by the forms driver.
1610 Each field structure contains one character pointer slot that is not
1611 used by the forms library. It is intended to be used by applications
1612 to store private per-field data. You can manipulate it with:
1613 int set_field_userptr(FIELD *field, /* field to alter */
1614 char *userptr); /* mode to set */
1616 char *field_userptr(FIELD *field); /* fetch mode of field */
1618 (Properly, this user pointer field ought to have (void *) type. The
1619 (char *) type is retained for System V compatibility.)
1621 It is valid to set the user pointer of the default field (with a
1622 set_field_userptr() call passed a NULL field pointer.) When a new
1623 field is created, the default-field user pointer is copied to
1624 initialize the new field's user pointer.
1626 Variable-Sized Fields
1628 Normally, a field is fixed at the size specified for it at creation
1629 time. If, however, you turn off its O_STATIC bit, it becomes dynamic
1630 and will automatically resize itself to accommodate data as it is
1631 entered. If the field has extra buffers associated with it, they will
1632 grow right along with the main input buffer.
1634 A one-line dynamic field will have a fixed height (1) but variable
1635 width, scrolling horizontally to display data within the field area as
1636 originally dimensioned and located. A multi-line dynamic field will
1637 have a fixed width, but variable height (number of rows), scrolling
1638 vertically to display data within the field area as originally
1639 dimensioned and located.
1641 Normally, a dynamic field is allowed to grow without limit. But it is
1642 possible to set an upper limit on the size of a dynamic field. You do
1643 it with this function:
1645 int set_max_field(FIELD *field, /* field to alter (may not be NULL) */
1646 int max_size); /* upper limit on field size */
1648 If the field is one-line, max_size is taken to be a column size limit;
1649 if it is multi-line, it is taken to be a line size limit. To disable
1650 any limit, use an argument of zero. The growth limit can be changed
1651 whether or not the O_STATIC bit is on, but has no effect until it is.
1653 The following properties of a field change when it becomes dynamic:
1654 * If there is no growth limit, there is no final position of the
1655 field; therefore O_AUTOSKIP and O_NL_OVERLOAD are ignored.
1656 * Field justification will be ignored (though whatever justification
1657 is set up will be retained internally and can be queried).
1658 * The dup_field() and link_field() calls copy dynamic-buffer sizes.
1659 If the O_STATIC option is set on one of a collection of links,
1660 buffer resizing will occur only when the field is edited through
1662 * The call field_info() will retrieve the original static size of
1663 the field; use dynamic_field_info() to get the actual dynamic
1668 By default, a field will accept any data that will fit in its input
1669 buffer. However, it is possible to attach a validation type to a
1670 field. If you do this, any attempt to leave the field while it
1671 contains data that doesn't match the validation type will fail. Some
1672 validation types also have a character-validity check for each time a
1673 character is entered in the field.
1675 A field's validation check (if any) is not called when
1676 set_field_buffer() modifies the input buffer, nor when that buffer is
1677 changed through a linked field.
1679 The form library provides a rich set of pre-defined validation types,
1680 and gives you the capability to define custom ones of your own. You
1681 can examine and change field validation attributes with the following
1684 int set_field_type(FIELD *field, /* field to alter */
1685 FIELDTYPE *ftype, /* type to associate */
1686 ...); /* additional arguments*/
1688 FIELDTYPE *field_type(FIELD *field); /* field to query */
1690 The validation type of a field is considered an attribute of the
1691 field. As with other field attributes, Also, doing set_field_type()
1692 with a NULL field default will change the system default for
1693 validation of newly-created fields.
1695 Here are the pre-defined validation types:
1699 This field type accepts alphabetic data; no blanks, no digits, no
1700 special characters (this is checked at character-entry time). It is
1703 int set_field_type(FIELD *field, /* field to alter */
1704 TYPE_ALPHA, /* type to associate */
1705 int width); /* maximum width of field */
1707 The width argument sets a minimum width of data. Typically you'll want
1708 to set this to the field width; if it's greater than the field width,
1709 the validation check will always fail. A minimum width of zero makes
1710 field completion optional.
1714 This field type accepts alphabetic data and digits; no blanks, no
1715 special characters (this is checked at character-entry time). It is
1718 int set_field_type(FIELD *field, /* field to alter */
1719 TYPE_ALNUM, /* type to associate */
1720 int width); /* maximum width of field */
1722 The width argument sets a minimum width of data. As with TYPE_ALPHA,
1723 typically you'll want to set this to the field width; if it's greater
1724 than the field width, the validation check will always fail. A minimum
1725 width of zero makes field completion optional.
1729 This type allows you to restrict a field's values to be among a
1730 specified set of string values (for example, the two-letter postal
1731 codes for U.S. states). It is set up with:
1733 int set_field_type(FIELD *field, /* field to alter */
1734 TYPE_ENUM, /* type to associate */
1735 char **valuelist; /* list of possible values */
1736 int checkcase; /* case-sensitive? */
1737 int checkunique); /* must specify uniquely? */
1739 The valuelist parameter must point at a NULL-terminated list of valid
1740 strings. The checkcase argument, if true, makes comparison with the
1741 string case-sensitive.
1743 When the user exits a TYPE_ENUM field, the validation procedure tries
1744 to complete the data in the buffer to a valid entry. If a complete
1745 choice string has been entered, it is of course valid. But it is also
1746 possible to enter a prefix of a valid string and have it completed for
1749 By default, if you enter such a prefix and it matches more than one
1750 value in the string list, the prefix will be completed to the first
1751 matching value. But the checkunique argument, if true, requires prefix
1752 matches to be unique in order to be valid.
1754 The REQ_NEXT_CHOICE and REQ_PREV_CHOICE input requests can be
1755 particularly useful with these fields.
1759 This field type accepts an integer. It is set up as follows:
1761 int set_field_type(FIELD *field, /* field to alter */
1762 TYPE_INTEGER, /* type to associate */
1763 int padding, /* # places to zero-pad to */
1764 int vmin, int vmax); /* valid range */
1766 Valid characters consist of an optional leading minus and digits. The
1767 range check is performed on exit. If the range maximum is less than or
1768 equal to the minimum, the range is ignored.
1770 If the value passes its range check, it is padded with as many leading
1771 zero digits as necessary to meet the padding argument.
1773 A TYPE_INTEGER value buffer can conveniently be interpreted with the C
1774 library function atoi(3).
1778 This field type accepts a decimal number. It is set up as follows:
1780 int set_field_type(FIELD *field, /* field to alter */
1781 TYPE_NUMERIC, /* type to associate */
1782 int padding, /* # places of precision */
1783 double vmin, double vmax); /* valid range */
1785 Valid characters consist of an optional leading minus and digits.
1786 possibly including a decimal point. If your system supports locale's,
1787 the decimal point character used must be the one defined by your
1788 locale. The range check is performed on exit. If the range maximum is
1789 less than or equal to the minimum, the range is ignored.
1791 If the value passes its range check, it is padded with as many
1792 trailing zero digits as necessary to meet the padding argument.
1794 A TYPE_NUMERIC value buffer can conveniently be interpreted with the C
1795 library function atof(3).
1799 This field type accepts data matching a regular expression. It is set
1802 int set_field_type(FIELD *field, /* field to alter */
1803 TYPE_REGEXP, /* type to associate */
1804 char *regexp); /* expression to match */
1806 The syntax for regular expressions is that of regcomp(3). The check
1807 for regular-expression match is performed on exit.
1809 Direct Field Buffer Manipulation
1811 The chief attribute of a field is its buffer contents. When a form has
1812 been completed, your application usually needs to know the state of
1813 each field buffer. You can find this out with:
1815 char *field_buffer(FIELD *field, /* field to query */
1816 int bufindex); /* number of buffer to query */
1818 Normally, the state of the zero-numbered buffer for each field is set
1819 by the user's editing actions on that field. It's sometimes useful to
1820 be able to set the value of the zero-numbered (or some other) buffer
1821 from your application:
1822 int set_field_buffer(FIELD *field, /* field to alter */
1823 int bufindex, /* number of buffer to alter */
1824 char *value); /* string value to set */
1826 If the field is not large enough and cannot be resized to a
1827 sufficiently large size to contain the specified value, the value will
1828 be truncated to fit.
1830 Calling field_buffer() with a null field pointer will raise an error.
1831 Calling field_buffer() on a field not currently selected for input
1832 will return a correct value. Calling field_buffer() on a field that is
1833 currently selected for input may not necessarily give a correct field
1834 buffer value, because entered data isn't necessarily copied to buffer
1835 zero before the exit validation check. To guarantee that the returned
1836 buffer value reflects on-screen reality, call field_buffer() either
1837 (1) in the field's exit validation check routine, (2) from the field's
1838 or form's initialization or termination hooks, or (3) just after a
1839 REQ_VALIDATION request has been processed by the forms driver.
1843 As with field attributes, form attributes inherit a default from a
1844 system default form structure. These defaults can be queried or set by
1845 of these functions using a form-pointer argument of NULL.
1847 The principal attribute of a form is its field list. You can query and
1848 change this list with:
1850 int set_form_fields(FORM *form, /* form to alter */
1851 FIELD **fields); /* fields to connect */
1853 char *form_fields(FORM *form); /* fetch fields of form */
1855 int field_count(FORM *form); /* count connect fields */
1857 The second argument of set_form_fields() may be a NULL-terminated
1858 field pointer array like the one required by new_form(). In that case,
1859 the old fields of the form are disconnected but not freed (and
1860 eligible to be connected to other forms), then the new fields are
1863 It may also be null, in which case the old fields are disconnected
1864 (and not freed) but no new ones are connected.
1866 The field_count() function simply counts the number of fields
1867 connected to a given from. It returns -1 if the form-pointer argument
1870 Control of Form Display
1872 In the overview section, you saw that to display a form you normally
1873 start by defining its size (and fields), posting it, and refreshing
1874 the screen. There is an hidden step before posting, which is the
1875 association of the form with a frame window (actually, a pair of
1876 windows) within which it will be displayed. By default, the forms
1877 library associates every form with the full-screen window stdscr.
1879 By making this step explicit, you can associate a form with a declared
1880 frame window on your screen display. This can be useful if you want to
1881 adapt the form display to different screen sizes, dynamically tile
1882 forms on the screen, or use a form as part of an interface layout
1885 The two windows associated with each form have the same functions as
1886 their analogues in the menu library. Both these windows are painted
1887 when the form is posted and erased when the form is unposted.
1889 The outer or frame window is not otherwise touched by the form
1890 routines. It exists so the programmer can associate a title, a border,
1891 or perhaps help text with the form and have it properly refreshed or
1892 erased at post/unpost time. The inner window or subwindow is where the
1893 current form page is actually displayed.
1895 In order to declare your own frame window for a form, you'll need to
1896 know the size of the form's bounding rectangle. You can get this
1899 int scale_form(FORM *form, /* form to query */
1900 int *rows, /* form rows */
1901 int *cols); /* form cols */
1903 The form dimensions are passed back in the locations pointed to by the
1904 arguments. Once you have this information, you can use it to declare
1905 of windows, then use one of these functions:
1906 int set_form_win(FORM *form, /* form to alter */
1907 WINDOW *win); /* frame window to connect */
1909 WINDOW *form_win(FORM *form); /* fetch frame window of form */
1911 int set_form_sub(FORM *form, /* form to alter */
1912 WINDOW *win); /* form subwindow to connect */
1914 WINDOW *form_sub(FORM *form); /* fetch form subwindow of form */
1916 Note that curses operations, including refresh(), on the form, should
1917 be done on the frame window, not the form subwindow.
1919 It is possible to check from your application whether all of a
1920 scrollable field is actually displayed within the menu subwindow. Use
1923 int data_ahead(FORM *form); /* form to be queried */
1925 int data_behind(FORM *form); /* form to be queried */
1927 The function data_ahead() returns TRUE if (a) the current field is
1928 one-line and has undisplayed data off to the right, (b) the current
1929 field is multi-line and there is data off-screen below it.
1931 The function data_behind() returns TRUE if the first (upper left hand)
1932 character position is off-screen (not being displayed).
1934 Finally, there is a function to restore the form window's cursor to
1935 the value expected by the forms driver:
1937 int pos_form_cursor(FORM *) /* form to be queried */
1939 If your application changes the form window cursor, call this function
1940 before handing control back to the forms driver in order to
1943 Input Processing in the Forms Driver
1945 The function form_driver() handles virtualized input requests for form
1946 navigation, editing, and validation requests, just as menu_driver does
1947 for menus (see the section on menu input handling).
1949 int form_driver(FORM *form, /* form to pass input to */
1950 int request); /* form request code */
1952 Your input virtualization function needs to take input and then
1953 convert it to either an alphanumeric character (which is treated as
1954 data to be entered in the currently-selected field), or a forms
1957 The forms driver provides hooks (through input-validation and
1958 field-termination functions) with which your application code can
1959 check that the input taken by the driver matched what was expected.
1961 Page Navigation Requests
1963 These requests cause page-level moves through the form, triggering
1964 display of a new form screen.
1967 Move to the next form page.
1970 Move to the previous form page.
1973 Move to the first form page.
1976 Move to the last form page.
1978 These requests treat the list as cyclic; that is, REQ_NEXT_PAGE from
1979 the last page goes to the first, and REQ_PREV_PAGE from the first page
1982 Inter-Field Navigation Requests
1984 These requests handle navigation between fields on the same page.
1990 Move to previous field.
1993 Move to the first field.
1996 Move to the last field.
1999 Move to sorted next field.
2002 Move to sorted previous field.
2005 Move to the sorted first field.
2008 Move to the sorted last field.
2014 Move right to field.
2022 These requests treat the list of fields on a page as cyclic; that is,
2023 REQ_NEXT_FIELD from the last field goes to the first, and
2024 REQ_PREV_FIELD from the first field goes to the last. The order of the
2025 fields for these (and the REQ_FIRST_FIELD and REQ_LAST_FIELD requests)
2026 is simply the order of the field pointers in the form array (as set up
2027 by new_form() or set_form_fields()
2029 It is also possible to traverse the fields as if they had been sorted
2030 in screen-position order, so the sequence goes left-to-right and
2031 top-to-bottom. To do this, use the second group of four
2032 sorted-movement requests.
2034 Finally, it is possible to move between fields using visual directions
2035 up, down, right, and left. To accomplish this, use the third group of
2036 four requests. Note, however, that the position of a form for purposes
2037 of these requests is its upper-left corner.
2039 For example, suppose you have a multi-line field B, and two
2040 single-line fields A and C on the same line with B, with A to the left
2041 of B and C to the right of B. A REQ_MOVE_RIGHT from A will go to B
2042 only if A, B, and C all share the same first line; otherwise it will
2045 Intra-Field Navigation Requests
2047 These requests drive movement of the edit cursor within the currently
2051 Move to next character.
2054 Move to previous character.
2060 Move to previous line.
2066 Move to previous word.
2069 Move to beginning of field.
2072 Move to end of field.
2075 Move to beginning of line.
2078 Move to end of line.
2084 Move right in field.
2092 Each word is separated from the previous and next characters by
2093 whitespace. The commands to move to beginning and end of line or field
2094 look for the first or last non-pad character in their ranges.
2098 Fields that are dynamic and have grown and fields explicitly created
2099 with offscreen rows are scrollable. One-line fields scroll
2100 horizontally; multi-line fields scroll vertically. Most scrolling is
2101 triggered by editing and intra-field movement (the library scrolls the
2102 field to keep the cursor visible). It is possible to explicitly
2103 request scrolling with the following requests:
2106 Scroll vertically forward a line.
2109 Scroll vertically backward a line.
2112 Scroll vertically forward a page.
2115 Scroll vertically backward a page.
2118 Scroll vertically forward half a page.
2121 Scroll vertically backward half a page.
2124 Scroll horizontally forward a character.
2127 Scroll horizontally backward a character.
2130 Scroll horizontally one field width forward.
2133 Scroll horizontally one field width backward.
2136 Scroll horizontally one half field width forward.
2139 Scroll horizontally one half field width backward.
2141 For scrolling purposes, a page of a field is the height of its visible
2146 When you pass the forms driver an ASCII character, it is treated as a
2147 request to add the character to the field's data buffer. Whether this
2148 is an insertion or a replacement depends on the field's edit mode
2149 (insertion is the default.
2151 The following requests support editing the field and changing the edit
2161 New line request (see below for explanation).
2164 Insert space at character location.
2167 Insert blank line at character location.
2170 Delete character at cursor.
2173 Delete previous word at cursor.
2176 Delete line at cursor.
2179 Delete word at cursor.
2182 Clear to end of line.
2185 Clear to end of field.
2190 The behavior of the REQ_NEW_LINE and REQ_DEL_PREV requests is
2191 complicated and partly controlled by a pair of forms options. The
2192 special cases are triggered when the cursor is at the beginning of a
2193 field, or on the last line of the field.
2195 First, we consider REQ_NEW_LINE:
2197 The normal behavior of REQ_NEW_LINE in insert mode is to break the
2198 current line at the position of the edit cursor, inserting the portion
2199 of the current line after the cursor as a new line following the
2200 current and moving the cursor to the beginning of that new line (you
2201 may think of this as inserting a newline in the field buffer).
2203 The normal behavior of REQ_NEW_LINE in overlay mode is to clear the
2204 current line from the position of the edit cursor to end of line. The
2205 cursor is then moved to the beginning of the next line.
2207 However, REQ_NEW_LINE at the beginning of a field, or on the last line
2208 of a field, instead does a REQ_NEXT_FIELD. O_NL_OVERLOAD option is
2209 off, this special action is disabled.
2211 Now, let us consider REQ_DEL_PREV:
2213 The normal behavior of REQ_DEL_PREV is to delete the previous
2214 character. If insert mode is on, and the cursor is at the start of a
2215 line, and the text on that line will fit on the previous one, it
2216 instead appends the contents of the current line to the previous one
2217 and deletes the current line (you may think of this as deleting a
2218 newline from the field buffer).
2220 However, REQ_DEL_PREV at the beginning of a field is instead treated
2221 as a REQ_PREV_FIELD.
2223 If the O_BS_OVERLOAD option is off, this special action is disabled
2224 and the forms driver just returns E_REQUEST_DENIED.
2226 See Form Options for discussion of how to set and clear the overload
2231 If the type of your field is ordered, and has associated functions for
2232 getting the next and previous values of the type from a given value,
2233 there are requests that can fetch that value into the field buffer:
2236 Place the successor value of the current value in the buffer.
2239 Place the predecessor value of the current value in the buffer.
2241 Of the built-in field types, only TYPE_ENUM has built-in successor and
2242 predecessor functions. When you define a field type of your own (see
2243 Custom Validation Types), you can associate our own ordering
2246 Application Commands
2248 Form requests are represented as integers above the curses value
2249 greater than KEY_MAX and less than or equal to the constant
2250 MAX_COMMAND. If your input-virtualization routine returns a value
2251 above MAX_COMMAND, the forms driver will ignore it.
2255 It is possible to set function hooks to be executed whenever the
2256 current field or form changes. Here are the functions that support
2259 typedef void (*HOOK)(); /* pointer to function returning void */
2261 int set_form_init(FORM *form, /* form to alter */
2262 HOOK hook); /* initialization hook */
2264 HOOK form_init(FORM *form); /* form to query */
2266 int set_form_term(FORM *form, /* form to alter */
2267 HOOK hook); /* termination hook */
2269 HOOK form_term(FORM *form); /* form to query */
2271 int set_field_init(FORM *form, /* form to alter */
2272 HOOK hook); /* initialization hook */
2274 HOOK field_init(FORM *form); /* form to query */
2276 int set_field_term(FORM *form, /* form to alter */
2277 HOOK hook); /* termination hook */
2279 HOOK field_term(FORM *form); /* form to query */
2281 These functions allow you to either set or query four different hooks.
2282 In each of the set functions, the second argument should be the
2283 address of a hook function. These functions differ only in the timing
2287 This hook is called when the form is posted; also, just after
2288 each page change operation.
2291 This hook is called when the form is posted; also, just after
2295 This hook is called just after field validation; that is, just
2296 before the field is altered. It is also called when the form is
2300 This hook is called when the form is unposted; also, just
2301 before each page change operation.
2303 Calls to these hooks may be triggered
2304 1. When user editing requests are processed by the forms driver
2305 2. When the current page is changed by set_current_field() call
2306 3. When the current field is changed by a set_form_page() call
2308 See Field Change Commands for discussion of the latter two cases.
2310 You can set a default hook for all fields by passing one of the set
2311 functions a NULL first argument.
2313 You can disable any of these hooks by (re)setting them to NULL, the
2316 Field Change Commands
2318 Normally, navigation through the form will be driven by the user's
2319 input requests. But sometimes it is useful to be able to move the
2320 focus for editing and viewing under control of your application, or
2321 ask which field it currently is in. The following functions help you
2324 int set_current_field(FORM *form, /* form to alter */
2325 FIELD *field); /* field to shift to */
2327 FIELD *current_field(FORM *form); /* form to query */
2329 int field_index(FORM *form, /* form to query */
2330 FIELD *field); /* field to get index of */
2332 The function field_index() returns the index of the given field in the
2333 given form's field array (the array passed to new_form() or
2336 The initial current field of a form is the first active field on the
2337 first page. The function set_form_fields() resets this.
2339 It is also possible to move around by pages.
2341 int set_form_page(FORM *form, /* form to alter */
2342 int page); /* page to go to (0-origin) */
2344 int form_page(FORM *form); /* return form's current page */
2346 The initial page of a newly-created form is 0. The function
2347 set_form_fields() resets this.
2351 Like fields, forms may have control option bits. They can be changed
2352 or queried with these functions:
2354 int set_form_opts(FORM *form, /* form to alter */
2355 int attr); /* attribute to set */
2357 int form_opts_on(FORM *form, /* form to alter */
2358 int attr); /* attributes to turn on */
2360 int form_opts_off(FORM *form, /* form to alter */
2361 int attr); /* attributes to turn off */
2363 int form_opts(FORM *form); /* form to query */
2365 By default, all options are on. Here are the available option bits:
2368 Enable overloading of REQ_NEW_LINE as described in Editing
2369 Requests. The value of this option is ignored on dynamic fields
2370 that have not reached their size limit; these have no last
2371 line, so the circumstances for triggering a REQ_NEXT_FIELD
2375 Enable overloading of REQ_DEL_PREV as described in Editing
2378 The option values are bit-masks and can be composed with logical-or in
2381 Custom Validation Types
2383 The form library gives you the capability to define custom validation
2384 types of your own. Further, the optional additional arguments of
2385 set_field_type effectively allow you to parameterize validation types.
2386 Most of the complications in the validation-type interface have to do
2387 with the handling of the additional arguments within custom validation
2392 The simplest way to create a custom data type is to compose it from
2393 two preexisting ones:
2395 FIELD *link_fieldtype(FIELDTYPE *type1,
2398 This function creates a field type that will accept any of the values
2399 legal for either of its argument field types (which may be either
2400 predefined or programmer-defined). If a set_field_type() call later
2401 requires arguments, the new composite type expects all arguments for
2402 the first type, than all arguments for the second. Order functions
2403 (see Order Requests) associated with the component types will work on
2404 the composite; what it does is check the validation function for the
2405 first type, then for the second, to figure what type the buffer
2406 contents should be treated as.
2410 To create a field type from scratch, you need to specify one or both
2411 of the following things:
2413 * A character-validation function, to check each character as it is
2415 * A field-validation function to be applied on exit from the field.
2417 Here's how you do that:
2419 typedef int (*HOOK)(); /* pointer to function returning int */
2421 FIELDTYPE *new_fieldtype(HOOK f_validate, /* field validator */
2422 HOOK c_validate) /* character validator */
2425 int free_fieldtype(FIELDTYPE *ftype); /* type to free */
2427 At least one of the arguments of new_fieldtype() must be non-NULL. The
2428 forms driver will automatically call the new type's validation
2429 functions at appropriate points in processing a field of the new type.
2431 The function free_fieldtype() deallocates the argument fieldtype,
2432 freeing all storage associated with it.
2434 Normally, a field validator is called when the user attempts to leave
2435 the field. Its first argument is a field pointer, from which it can
2436 get to field buffer 0 and test it. If the function returns TRUE, the
2437 operation succeeds; if it returns FALSE, the edit cursor stays in the
2440 A character validator gets the character passed in as a first
2441 argument. It too should return TRUE if the character is valid, FALSE
2444 Validation Function Arguments
2446 Your field- and character- validation functions will be passed a
2447 second argument as well. This second argument is the address of a
2448 structure (which we'll call a pile) built from any of the
2449 field-type-specific arguments passed to set_field_type(). If no such
2450 arguments are defined for the field type, this pile pointer argument
2453 In order to arrange for such arguments to be passed to your validation
2454 functions, you must associate a small set of storage-management
2455 functions with the type. The forms driver will use these to synthesize
2456 a pile from the trailing arguments of each set_field_type() argument,
2457 and a pointer to the pile will be passed to the validation functions.
2459 Here is how you make the association:
2461 typedef char *(*PTRHOOK)(); /* pointer to function returning (char *) */
2462 typedef void (*VOIDHOOK)(); /* pointer to function returning void */
2464 int set_fieldtype_arg(FIELDTYPE *type, /* type to alter */
2465 PTRHOOK make_str, /* make structure from args */
2466 PTRHOOK copy_str, /* make copy of structure */
2467 VOIDHOOK free_str); /* free structure storage */
2469 Here is how the storage-management hooks are used:
2472 This function is called by set_field_type(). It gets one
2473 argument, a va_list of the type-specific arguments passed to
2474 set_field_type(). It is expected to return a pile pointer to a
2475 data structure that encapsulates those arguments.
2478 This function is called by form library functions that allocate
2479 new field instances. It is expected to take a pile pointer,
2480 copy the pile to allocated storage, and return the address of
2484 This function is called by field- and type-deallocation
2485 routines in the library. It takes a pile pointer argument, and
2486 is expected to free the storage of that pile.
2488 The make_str and copy_str functions may return NULL to signal
2489 allocation failure. The library routines will that call them will
2490 return error indication when this happens. Thus, your validation
2491 functions should never see a NULL file pointer and need not check
2494 Order Functions For Custom Types
2496 Some custom field types are simply ordered in the same well-defined
2497 way that TYPE_ENUM is. For such types, it is possible to define
2498 successor and predecessor functions to support the REQ_NEXT_CHOICE and
2499 REQ_PREV_CHOICE requests. Here's how:
2501 typedef int (*INTHOOK)(); /* pointer to function returning int */
2503 int set_fieldtype_arg(FIELDTYPE *type, /* type to alter */
2504 INTHOOK succ, /* get successor value */
2505 INTHOOK pred); /* get predecessor value */
2507 The successor and predecessor arguments will each be passed two
2508 arguments; a field pointer, and a pile pointer (as for the validation
2509 functions). They are expected to use the function field_buffer() to
2510 read the current value, and set_field_buffer() on buffer 0 to set the
2511 next or previous value. Either hook may return TRUE to indicate
2512 success (a legal next or previous value was set) or FALSE to indicate
2517 The interface for defining custom types is complicated and tricky.
2518 Rather than attempting to create a custom type entirely from scratch,
2519 you should start by studying the library source code for whichever of
2520 the pre-defined types seems to be closest to what you want.
2522 Use that code as a model, and evolve it towards what you really want.
2523 You will avoid many problems and annoyances that way. The code in the
2524 ncurses library has been specifically exempted from the package
2525 copyright to support this.
2527 If your custom type defines order functions, have do something
2528 intuitive with a blank field. A useful convention is to make the
2529 successor of a blank field the types minimum value, and its
2530 predecessor the maximum.