1 Writing Programs with NCURSES
3 Writing Programs with NCURSES
5 by Eric S. Raymond and Zeyd M. Ben-Halim
6 updates since release 1.9.9e by Thomas Dickey
11 + A Brief History of Curses
12 + Scope of This Document
15 + An Overview of Curses
16 o Compiling Programs using Curses
18 o Standard Windows and Function Naming Conventions
24 o Using Forms Characters
25 o Character Attributes and Color
28 + Function Descriptions
29 o Initialization and Wrapup
30 o Causing Output to the Terminal
31 o Low-Level Capability Access
33 + Hints, Tips, and Tricks
34 o Some Notes of Caution
35 o Temporarily Leaving ncurses Mode
36 o Using ncurses under xterm
37 o Handling Multiple Terminal Screens
38 o Testing for Terminal Capabilities
40 o Special Features of ncurses
41 + Compatibility with Older Versions
42 o Refresh of Overlapping Windows
44 + XSI Curses Conformance
46 + Compiling With the Panels Library
48 + Panels, Input, and the Standard Screen
50 + Miscellaneous Other Facilities
52 + Compiling with the menu Library
57 + Processing Menu Input
58 + Miscellaneous Other Features
60 + Compiling with the forms Library
62 + Creating and Freeing Fields and Forms
63 + Fetching and Changing Field Attributes
64 o Fetching Size and Location Data
65 o Changing the Field Location
66 o The Justification Attribute
67 o Field Display Attributes
71 + Variable-Sized Fields
79 + Direct Field Buffer Manipulation
81 + Control of Form Display
82 + Input Processing in the Forms Driver
83 o Page Navigation Requests
84 o Inter-Field Navigation Requests
85 o Intra-Field Navigation Requests
87 o Field Editing Requests
89 o Application Commands
91 + Field Change Commands
93 + Custom Validation Types
96 o Validation Function Arguments
97 o Order Functions For Custom Types
99 _________________________________________________________________
103 This document is an introduction to programming with curses. It is not
104 an exhaustive reference for the curses Application Programming
105 Interface (API); that role is filled by the curses manual pages.
106 Rather, it is intended to help C programmers ease into using the
109 This document is aimed at C applications programmers not yet
110 specifically familiar with ncurses. If you are already an experienced
111 curses programmer, you should nevertheless read the sections on Mouse
112 Interfacing, Debugging, Compatibility with Older Versions, and Hints,
113 Tips, and Tricks. These will bring you up to speed on the special
114 features and quirks of the ncurses implementation. If you are not so
115 experienced, keep reading.
117 The curses package is a subroutine library for terminal-independent
118 screen-painting and input-event handling which presents a high level
119 screen model to the programmer, hiding differences between terminal
120 types and doing automatic optimization of output to change one screen
121 full of text into another. Curses uses terminfo, which is a database
122 format that can describe the capabilities of thousands of different
125 The curses API may seem something of an archaism on UNIX desktops
126 increasingly dominated by X, Motif, and Tcl/Tk. Nevertheless, UNIX
127 still supports tty lines and X supports xterm(1); the curses API has
128 the advantage of (a) back-portability to character-cell terminals, and
129 (b) simplicity. For an application that does not require bit-mapped
130 graphics and multiple fonts, an interface implementation using curses
131 will typically be a great deal simpler and less expensive than one
134 A Brief History of Curses
136 Historically, the first ancestor of curses was the routines written to
137 provide screen-handling for the vi editor; these used the termcap
138 database facility (both released in 3BSD) for describing terminal
139 capabilities. These routines were abstracted into a documented library
140 and first released with the early BSD UNIX versions. All of this work
141 was done by students at the University of California (Berkeley
142 campus). The curses library was first published in 4.0BSD, a year
143 after 3BSD (i.e., late 1980).
145 After graduation, one of those students went to work at AT&T Bell
146 Labs, and made an improved termcap library called terminfo (i.e.,
147 "libterm"), and adapted the curses library to use this. That was
148 subsequently released in System V Release 2 (early 1984). Thereafter,
149 other developers added to the curses and terminfo libraries. For
150 instance, a student at Cornell University wrote an improved terminfo
151 library as well as a tool (tic) to compile the terminal descriptions.
152 As a general rule, AT&T did not identify the developers in the
153 source-code or documentation; the tic and infocmp programs are the
156 System V Release 3 (System III UNIX) from Bell Labs featured a
157 rewritten and much-improved curses library, along with the tic program
160 To recap, terminfo is based on Berkeley's termcap database, but
161 contains a number of improvements and extensions. Parameterized
162 capabilities strings were introduced, making it possible to describe
163 multiple video attributes, and colors and to handle far more unusual
164 terminals than possible with termcap. In the later AT&T System V
165 releases, curses evolved to use more facilities and offer more
166 capabilities, going far beyond BSD curses in power and flexibility.
168 Scope of This Document
170 This document describes ncurses, a free implementation of the System V
171 curses API with some clearly marked extensions. It includes the
172 following System V curses features:
173 * Support for multiple screen highlights (BSD curses could only
174 handle one "standout" highlight, usually reverse-video).
175 * Support for line- and box-drawing using forms characters.
176 * Recognition of function keys on input.
178 * Support for pads (windows of larger than screen size on which the
179 screen or a subwindow defines a viewport).
181 Also, this package makes use of the insert and delete line and
182 character features of terminals so equipped, and determines how to
183 optimally use these features with no help from the programmer. It
184 allows arbitrary combinations of video attributes to be displayed,
185 even on terminals that leave "magic cookies" on the screen to mark
186 changes in attributes.
188 The ncurses package can also capture and use event reports from a
189 mouse in some environments (notably, xterm under the X window system).
190 This document includes tips for using the mouse.
192 The ncurses package was originated by Pavel Curtis. The original
193 maintainer of this package is Zeyd Ben-Halim <zmbenhal@netcom.com>.
194 Eric S. Raymond <esr@snark.thyrsus.com> wrote many of the new features
195 in versions after 1.8.1 and wrote most of this introduction. Juergen
196 Pfeifer wrote all of the menu and forms code as well as the Ada95
197 binding. Ongoing work is being done by Thomas Dickey (maintainer).
198 Contact the current maintainers at bug-ncurses@gnu.org.
200 This document also describes the panels extension library, similarly
201 modeled on the SVr4 panels facility. This library allows you to
202 associate backing store with each of a stack or deck of overlapping
203 windows, and provides operations for moving windows around in the
204 stack that change their visibility in the natural way (handling window
207 Finally, this document describes in detail the menus and forms
208 extension libraries, also cloned from System V, which support easy
209 construction and sequences of menus and fill-in forms.
213 In this document, the following terminology is used with reasonable
217 A data structure describing a sub-rectangle of the screen
218 (possibly the entire screen). You can write to a window as
219 though it were a miniature screen, scrolling independently of
220 other windows on the physical screen.
223 A subset of windows which are as large as the terminal screen,
224 i.e., they start at the upper left hand corner and encompass
225 the lower right hand corner. One of these, stdscr, is
226 automatically provided for the programmer.
229 The package's idea of what the terminal display currently looks
230 like, i.e., what the user sees now. This is a special screen.
234 An Overview of Curses
236 Compiling Programs using Curses
238 In order to use the library, it is necessary to have certain types and
239 variables defined. Therefore, the programmer must have a line:
242 at the top of the program source. The screen package uses the Standard
243 I/O library, so <curses.h> includes <stdio.h>. <curses.h> also
244 includes <termios.h>, <termio.h>, or <sgtty.h> depending on your
245 system. It is redundant (but harmless) for the programmer to do these
246 includes, too. In linking with curses you need to have -lncurses in
247 your LDFLAGS or on the command line. There is no need for any other
252 In order to update the screen optimally, it is necessary for the
253 routines to know what the screen currently looks like and what the
254 programmer wants it to look like next. For this purpose, a data type
255 (structure) named WINDOW is defined which describes a window image to
256 the routines, including its starting position on the screen (the (y,
257 x) coordinates of the upper left hand corner) and its size. One of
258 these (called curscr, for current screen) is a screen image of what
259 the terminal currently looks like. Another screen (called stdscr, for
260 standard screen) is provided by default to make changes on.
262 A window is a purely internal representation. It is used to build and
263 store a potential image of a portion of the terminal. It does not bear
264 any necessary relation to what is really on the terminal screen; it is
265 more like a scratchpad or write buffer.
267 To make the section of physical screen corresponding to a window
268 reflect the contents of the window structure, the routine refresh()
269 (or wrefresh() if the window is not stdscr) is called.
271 A given physical screen section may be within the scope of any number
272 of overlapping windows. Also, changes can be made to windows in any
273 order, without regard to motion efficiency. Then, at will, the
274 programmer can effectively say "make it look like this," and let the
275 package implementation determine the most efficient way to repaint the
278 Standard Windows and Function Naming Conventions
280 As hinted above, the routines can use several windows, but two are
281 automatically given: curscr, which knows what the terminal looks like,
282 and stdscr, which is what the programmer wants the terminal to look
283 like next. The user should never actually access curscr directly.
284 Changes should be made to through the API, and then the routine
285 refresh() (or wrefresh()) called.
287 Many functions are defined to use stdscr as a default screen. For
288 example, to add a character to stdscr, one calls addch() with the
289 desired character as argument. To write to a different window. use the
290 routine waddch() (for window-specific addch()) is provided. This
291 convention of prepending function names with a "w" when they are to be
292 applied to specific windows is consistent. The only routines which do
293 not follow it are those for which a window must always be specified.
295 In order to move the current (y, x) coordinates from one point to
296 another, the routines move() and wmove() are provided. However, it is
297 often desirable to first move and then perform some I/O operation. In
298 order to avoid clumsiness, most I/O routines can be preceded by the
299 prefix "mv" and the desired (y, x) coordinates prepended to the
300 arguments to the function. For example, the calls
312 mvwaddch(win, y, x, ch);
314 Note that the window description pointer (win) comes before the added
315 (y, x) coordinates. If a function requires a window pointer, it is
316 always the first parameter passed.
320 The curses library sets some variables describing the terminal
322 type name description
323 ------------------------------------------------------------------
324 int LINES number of lines on the terminal
325 int COLS number of columns on the terminal
327 The curses.h also introduces some #define constants and types of
331 boolean type, actually a "char" (e.g., bool doneit;)
334 boolean "true" flag (1).
337 boolean "false" flag (0).
340 error flag returned by routines on a failure (-1).
343 error flag returned by routines when things go right.
347 Now we describe how to actually use the screen package. In it, we
348 assume all updating, reading, etc. is applied to stdscr. These
349 instructions will work on any window, providing you change the
350 function names and parameters as mentioned above.
352 Here is a sample program to motivate the discussion:
357 static void finish(int sig);
360 main(int argc, char *argv[])
364 /* initialize your non-curses data structures here */
366 (void) signal(SIGINT, finish); /* arrange interrupts to terminate */
368 (void) initscr(); /* initialize the curses library */
369 keypad(stdscr, TRUE); /* enable keyboard mapping */
370 (void) nonl(); /* tell curses not to do NL->CR/NL on output */
371 (void) cbreak(); /* take input chars one at a time, no wait for \n */
372 (void) echo(); /* echo input - in color */
379 * Simple color assignment, often all we need. Color pair 0 cannot
380 * be redefined. This example uses the same value for the color
381 * pair as for the foreground color, though of course that is not
384 init_pair(1, COLOR_RED, COLOR_BLACK);
385 init_pair(2, COLOR_GREEN, COLOR_BLACK);
386 init_pair(3, COLOR_YELLOW, COLOR_BLACK);
387 init_pair(4, COLOR_BLUE, COLOR_BLACK);
388 init_pair(5, COLOR_CYAN, COLOR_BLACK);
389 init_pair(6, COLOR_MAGENTA, COLOR_BLACK);
390 init_pair(7, COLOR_WHITE, COLOR_BLACK);
395 int c = getch(); /* refresh, accept single keystroke of input */
396 attrset(COLOR_PAIR(num % 8));
399 /* process the command keystroke */
402 finish(0); /* we are done */
405 static void finish(int sig)
409 /* do your non-curses wrapup here */
416 In order to use the screen package, the routines must know about
417 terminal characteristics, and the space for curscr and stdscr must be
418 allocated. These function initscr() does both these things. Since it
419 must allocate space for the windows, it can overflow memory when
420 attempting to do so. On the rare occasions this happens, initscr()
421 will terminate the program with an error message. initscr() must
422 always be called before any of the routines which affect windows are
423 used. If it is not, the program will core dump as soon as either
424 curscr or stdscr are referenced. However, it is usually best to wait
425 to call it until after you are sure you will need it, like after
426 checking for startup errors. Terminal status changing routines like
427 nl() and cbreak() should be called after initscr().
429 Once the screen windows have been allocated, you can set them up for
430 your program. If you want to, say, allow a screen to scroll, use
431 scrollok(). If you want the cursor to be left in place after the last
432 change, use leaveok(). If this is not done, refresh() will move the
433 cursor to the window's current (y, x) coordinates after updating it.
435 You can create new windows of your own using the functions newwin(),
436 derwin(), and subwin(). The routine delwin() will allow you to get rid
437 of old windows. All the options described above can be applied to any
442 Now that we have set things up, we will want to actually update the
443 terminal. The basic functions used to change what will go on a window
444 are addch() and move(). addch() adds a character at the current (y, x)
445 coordinates. move() changes the current (y, x) coordinates to whatever
446 you want them to be. It returns ERR if you try to move off the window.
447 As mentioned above, you can combine the two into mvaddch() to do both
450 The other output functions, such as addstr() and printw(), all call
451 addch() to add characters to the window.
453 After you have put on the window what you want there, when you want
454 the portion of the terminal covered by the window to be made to look
455 like it, you must call refresh(). In order to optimize finding
456 changes, refresh() assumes that any part of the window not changed
457 since the last refresh() of that window has not been changed on the
458 terminal, i.e., that you have not refreshed a portion of the terminal
459 with an overlapping window. If this is not the case, the routine
460 touchwin() is provided to make it look like the entire window has been
461 changed, thus making refresh() check the whole subsection of the
462 terminal for changes.
464 If you call wrefresh() with curscr as its argument, it will make the
465 screen look like curscr thinks it looks like. This is useful for
466 implementing a command which would redraw the screen in case it get
471 The complementary function to addch() is getch() which, if echo is
472 set, will call addch() to echo the character. Since the screen package
473 needs to know what is on the terminal at all times, if characters are
474 to be echoed, the tty must be in raw or cbreak mode. Since initially
475 the terminal has echoing enabled and is in ordinary "cooked" mode, one
476 or the other has to changed before calling getch(); otherwise, the
477 program's output will be unpredictable.
479 When you need to accept line-oriented input in a window, the functions
480 wgetstr() and friends are available. There is even a wscanw() function
481 that can do scanf()(3)-style multi-field parsing on window input.
482 These pseudo-line-oriented functions turn on echoing while they
485 The example code above uses the call keypad(stdscr, TRUE) to enable
486 support for function-key mapping. With this feature, the getch() code
487 watches the input stream for character sequences that correspond to
488 arrow and function keys. These sequences are returned as
489 pseudo-character values. The #define values returned are listed in the
490 curses.h The mapping from sequences to #define values is determined by
491 key_ capabilities in the terminal's terminfo entry.
493 Using Forms Characters
495 The addch() function (and some others, including box() and border())
496 can accept some pseudo-character arguments which are specially defined
497 by ncurses. These are #define values set up in the curses.h header;
498 see there for a complete list (look for the prefix ACS_).
500 The most useful of the ACS defines are the forms-drawing characters.
501 You can use these to draw boxes and simple graphs on the screen. If
502 the terminal does not have such characters, curses.h will map them to
503 a recognizable (though ugly) set of ASCII defaults.
505 Character Attributes and Color
507 The ncurses package supports screen highlights including standout,
508 reverse-video, underline, and blink. It also supports color, which is
509 treated as another kind of highlight.
511 Highlights are encoded, internally, as high bits of the
512 pseudo-character type (chtype) that curses.h uses to represent the
513 contents of a screen cell. See the curses.h header file for a complete
514 list of highlight mask values (look for the prefix A_).
516 There are two ways to make highlights. One is to logical-or the value
517 of the highlights you want into the character argument of an addch()
518 call, or any other output call that takes a chtype argument.
520 The other is to set the current-highlight value. This is logical-ORed
521 with any highlight you specify the first way. You do this with the
522 functions attron(), attroff(), and attrset(); see the manual pages for
523 details. Color is a special kind of highlight. The package actually
524 thinks in terms of color pairs, combinations of foreground and
525 background colors. The sample code above sets up eight color pairs,
526 all of the guaranteed-available colors on black. Note that each color
527 pair is, in effect, given the name of its foreground color. Any other
528 range of eight non-conflicting values could have been used as the
529 first arguments of the init_pair() values.
531 Once you have done an init_pair() that creates color-pair N, you can
532 use COLOR_PAIR(N) as a highlight that invokes that particular color
533 combination. Note that COLOR_PAIR(N), for constant N, is itself a
534 compile-time constant and can be used in initializers.
538 The ncurses library also provides a mouse interface.
540 NOTE: this facility is specific to ncurses, it is not part of
541 either the XSI Curses standard, nor of System V Release 4, nor BSD
542 curses. System V Release 4 curses contains code with similar
543 interface definitions, however it is not documented. Other than by
544 disassembling the library, we have no way to determine exactly how
545 that mouse code works. Thus, we recommend that you wrap
546 mouse-related code in an #ifdef using the feature macro
547 NCURSES_MOUSE_VERSION so it will not be compiled and linked on
550 Presently, mouse event reporting works in the following environments:
551 * xterm and similar programs such as rxvt.
552 * Linux console, when configured with gpm(1), Alessandro Rubini's
554 * FreeBSD sysmouse (console)
557 The mouse interface is very simple. To activate it, you use the
558 function mousemask(), passing it as first argument a bit-mask that
559 specifies what kinds of events you want your program to be able to
560 see. It will return the bit-mask of events that actually become
561 visible, which may differ from the argument if the mouse device is not
562 capable of reporting some of the event types you specify.
564 Once the mouse is active, your application's command loop should watch
565 for a return value of KEY_MOUSE from wgetch(). When you see this, a
566 mouse event report has been queued. To pick it off the queue, use the
567 function getmouse() (you must do this before the next wgetch(),
568 otherwise another mouse event might come in and make the first one
571 Each call to getmouse() fills a structure (the address of which you
572 will pass it) with mouse event data. The event data includes
573 zero-origin, screen-relative character-cell coordinates of the mouse
574 pointer. It also includes an event mask. Bits in this mask will be
575 set, corresponding to the event type being reported.
577 The mouse structure contains two additional fields which may be
578 significant in the future as ncurses interfaces to new kinds of
579 pointing device. In addition to x and y coordinates, there is a slot
580 for a z coordinate; this might be useful with touch-screens that can
581 return a pressure or duration parameter. There is also a device ID
582 field, which could be used to distinguish between multiple pointing
585 The class of visible events may be changed at any time via
586 mousemask(). Events that can be reported include presses, releases,
587 single-, double- and triple-clicks (you can set the maximum
588 button-down time for clicks). If you do not make clicks visible, they
589 will be reported as press-release pairs. In some environments, the
590 event mask may include bits reporting the state of shift, alt, and
591 ctrl keys on the keyboard during the event.
593 A function to check whether a mouse event fell within a given window
594 is also supplied. You can use this to see whether a given window
595 should consider a mouse event relevant to it.
597 Because mouse event reporting will not be available in all
598 environments, it would be unwise to build ncurses applications that
599 require the use of a mouse. Rather, you should use the mouse as a
600 shortcut for point-and-shoot commands your application would normally
601 accept from the keyboard. Two of the test games in the ncurses
602 distribution (bs and knight) contain code that illustrates how this
605 See the manual page curs_mouse(3X) for full details of the
606 mouse-interface functions.
610 In order to clean up after the ncurses routines, the routine endwin()
611 is provided. It restores tty modes to what they were when initscr()
612 was first called, and moves the cursor down to the lower-left corner.
613 Thus, anytime after the call to initscr, endwin() should be called
616 Function Descriptions
618 We describe the detailed behavior of some important curses functions
619 here, as a supplement to the manual page descriptions.
621 Initialization and Wrapup
624 The first function called should almost always be initscr().
625 This will determine the terminal type and initialize curses
626 data structures. initscr() also arranges that the first call to
627 refresh() will clear the screen. If an error occurs a message
628 is written to standard error and the program exits. Otherwise
629 it returns a pointer to stdscr. A few functions may be called
630 before initscr (slk_init(), filter(), ripoffline(), use_env(),
631 and, if you are using multiple terminals, newterm().)
634 Your program should always call endwin() before exiting or
635 shelling out of the program. This function will restore tty
636 modes, move the cursor to the lower left corner of the screen,
637 reset the terminal into the proper non-visual mode. Calling
638 refresh() or doupdate() after a temporary escape from the
639 program will restore the ncurses screen from before the escape.
641 newterm(type, ofp, ifp)
642 A program which outputs to more than one terminal should use
643 newterm() instead of initscr(). newterm() should be called once
644 for each terminal. It returns a variable of type SCREEN * which
645 should be saved as a reference to that terminal. (NOTE: a
646 SCREEN variable is not a screen in the sense we are describing
647 in this introduction, but a collection of parameters used to
648 assist in optimizing the display.) The arguments are the type
649 of the terminal (a string) and FILE pointers for the output and
650 input of the terminal. If type is NULL then the environment
651 variable $TERM is used. endwin() should called once at wrapup
652 time for each terminal opened using this function.
655 This function is used to switch to a different terminal
656 previously opened by newterm(). The screen reference for the
657 new terminal is passed as the parameter. The previous terminal
658 is returned by the function. All other calls affect only the
662 The inverse of newterm(); deallocates the data structures
663 associated with a given SCREEN reference.
665 Causing Output to the Terminal
667 refresh() and wrefresh(win)
668 These functions must be called to actually get any output on
669 the terminal, as other routines merely manipulate data
670 structures. wrefresh() copies the named window to the physical
671 terminal screen, taking into account what is already there in
672 order to do optimizations. refresh() does a refresh of stdscr.
673 Unless leaveok() has been enabled, the physical cursor of the
674 terminal is left at the location of the window's cursor.
676 doupdate() and wnoutrefresh(win)
677 These two functions allow multiple updates with more efficiency
678 than wrefresh. To use them, it is important to understand how
679 curses works. In addition to all the window structures, curses
680 keeps two data structures representing the terminal screen: a
681 physical screen, describing what is actually on the screen, and
682 a virtual screen, describing what the programmer wants to have
683 on the screen. wrefresh works by first copying the named window
684 to the virtual screen (wnoutrefresh()), and then calling the
685 routine to update the screen (doupdate()). If the programmer
686 wishes to output several windows at once, a series of calls to
687 wrefresh will result in alternating calls to wnoutrefresh() and
688 doupdate(), causing several bursts of output to the screen. By
689 calling wnoutrefresh() for each window, it is then possible to
690 call doupdate() once, resulting in only one burst of output,
691 with fewer total characters transmitted (this also avoids a
692 visually annoying flicker at each update).
694 Low-Level Capability Access
696 setupterm(term, filenum, errret)
697 This routine is called to initialize a terminal's description,
698 without setting up the curses screen structures or changing the
699 tty-driver mode bits. term is the character string representing
700 the name of the terminal being used. filenum is the UNIX file
701 descriptor of the terminal to be used for output. errret is a
702 pointer to an integer, in which a success or failure indication
703 is returned. The values returned can be 1 (all is well), 0 (no
704 such terminal), or -1 (some problem locating the terminfo
707 The value of term can be given as NULL, which will cause the
708 value of TERM in the environment to be used. The errret pointer
709 can also be given as NULL, meaning no error code is wanted. If
710 errret is defaulted, and something goes wrong, setupterm() will
711 print an appropriate error message and exit, rather than
712 returning. Thus, a simple program can call setupterm(0, 1, 0)
713 and not worry about initialization errors.
715 After the call to setupterm(), the global variable cur_term is
716 set to point to the current structure of terminal capabilities.
717 By calling setupterm() for each terminal, and saving and
718 restoring cur_term, it is possible for a program to use two or
719 more terminals at once. Setupterm() also stores the names
720 section of the terminal description in the global character
721 array ttytype[]. Subsequent calls to setupterm() will overwrite
722 this array, so you will have to save it yourself if need be.
726 NOTE: These functions are not part of the standard curses API!
729 This function can be used to explicitly set a trace level. If
730 the trace level is nonzero, execution of your program will
731 generate a file called "trace" in the current working directory
732 containing a report on the library's actions. Higher trace
733 levels enable more detailed (and verbose) reporting -- see
734 comments attached to TRACE_ defines in the curses.h file for
735 details. (It is also possible to set a trace level by assigning
736 a trace level value to the environment variable NCURSES_TRACE).
739 This function can be used to output your own debugging
740 information. It is only available only if you link with
741 -lncurses_g. It can be used the same way as printf(), only it
742 outputs a newline after the end of arguments. The output goes
743 to a file called trace in the current directory.
745 Trace logs can be difficult to interpret due to the sheer volume of
746 data dumped in them. There is a script called tracemunch included with
747 the ncurses distribution that can alleviate this problem somewhat; it
748 compacts long sequences of similar operations into more succinct
749 single-line pseudo-operations. These pseudo-ops can be distinguished
750 by the fact that they are named in capital letters.
752 Hints, Tips, and Tricks
754 The ncurses manual pages are a complete reference for this library. In
755 the remainder of this document, we discuss various useful methods that
756 may not be obvious from the manual page descriptions.
758 Some Notes of Caution
760 If you find yourself thinking you need to use noraw() or nocbreak(),
761 think again and move carefully. It is probably better design to use
762 getstr() or one of its relatives to simulate cooked mode. The noraw()
763 and nocbreak() functions try to restore cooked mode, but they may end
764 up clobbering some control bits set before you started your
765 application. Also, they have always been poorly documented, and are
766 likely to hurt your application's usability with other curses
769 Bear in mind that refresh() is a synonym for wrefresh(stdscr). Do not
770 try to mix use of stdscr with use of windows declared by newwin(); a
771 refresh() call will blow them off the screen. The right way to handle
772 this is to use subwin(), or not touch stdscr at all and tile your
773 screen with declared windows which you then wnoutrefresh() somewhere
774 in your program event loop, with a single doupdate() call to trigger
777 You are much less likely to run into problems if you design your
778 screen layouts to use tiled rather than overlapping windows.
779 Historically, curses support for overlapping windows has been weak,
780 fragile, and poorly documented. The ncurses library is not yet an
781 exception to this rule.
783 There is a panels library included in the ncurses distribution that
784 does a pretty good job of strengthening the overlapping-windows
787 Try to avoid using the global variables LINES and COLS. Use getmaxyx()
788 on the stdscr context instead. Reason: your code may be ported to run
789 in an environment with window resizes, in which case several screens
790 could be open with different sizes.
792 Temporarily Leaving NCURSES Mode
794 Sometimes you will want to write a program that spends most of its
795 time in screen mode, but occasionally returns to ordinary "cooked"
796 mode. A common reason for this is to support shell-out. This behavior
797 is simple to arrange in ncurses.
799 To leave ncurses mode, call endwin() as you would if you were
800 intending to terminate the program. This will take the screen back to
801 cooked mode; you can do your shell-out. When you want to return to
802 ncurses mode, simply call refresh() or doupdate(). This will repaint
805 There is a boolean function, isendwin(), which code can use to test
806 whether ncurses screen mode is active. It returns TRUE in the interval
807 between an endwin() call and the following refresh(), FALSE otherwise.
809 Here is some sample code for shellout:
810 addstr("Shelling out...");
811 def_prog_mode(); /* save current tty modes */
812 endwin(); /* restore original tty modes */
813 system("sh"); /* run shell */
814 addstr("returned.\n"); /* prepare return message */
815 refresh(); /* restore save modes, repaint screen */
817 Using NCURSES under XTERM
819 A resize operation in X sends SIGWINCH to the application running
820 under xterm. The easiest way to handle SIGWINCH is to do an endwin,
821 followed by an refresh and a screen repaint you code yourself. The
822 refresh will pick up the new screen size from the xterm's environment.
824 That is the standard way, of course (it even works with some vendor's
825 curses implementations). Its drawback is that it clears the screen to
826 reinitialize the display, and does not resize subwindows which must be
827 shrunk. Ncurses provides an extension which works better, the
828 resizeterm function. That function ensures that all windows are
829 limited to the new screen dimensions, and pads stdscr with blanks if
830 the screen is larger.
832 The ncurses library provides a SIGWINCH signal handler, which pushes a
833 KEY_RESIZE via the wgetch() calls. When ncurses returns that code, it
834 calls resizeterm to update the size of the standard screen's window,
835 repainting that (filling with blanks or truncating as needed). It also
836 resizes other windows, but its effect may be less satisfactory because
837 it cannot know how you want the screen re-painted. You will usually
838 have to write special-purpose code to handle KEY_RESIZE yourself.
840 Handling Multiple Terminal Screens
842 The initscr() function actually calls a function named newterm() to do
843 most of its work. If you are writing a program that opens multiple
844 terminals, use newterm() directly.
846 For each call, you will have to specify a terminal type and a pair of
847 file pointers; each call will return a screen reference, and stdscr
848 will be set to the last one allocated. You will switch between screens
849 with the set_term call. Note that you will also have to call
850 def_shell_mode and def_prog_mode on each tty yourself.
852 Testing for Terminal Capabilities
854 Sometimes you may want to write programs that test for the presence of
855 various capabilities before deciding whether to go into ncurses mode.
856 An easy way to do this is to call setupterm(), then use the functions
857 tigetflag(), tigetnum(), and tigetstr() to do your testing.
859 A particularly useful case of this often comes up when you want to
860 test whether a given terminal type should be treated as "smart"
861 (cursor-addressable) or "stupid". The right way to test this is to see
862 if the return value of tigetstr("cup") is non-NULL. Alternatively, you
863 can include the term.h file and test the value of the macro
868 Use the addchstr() family of functions for fast screen-painting of
869 text when you know the text does not contain any control characters.
870 Try to make attribute changes infrequent on your screens. Do not use
871 the immedok() option!
873 Special Features of NCURSES
875 The wresize() function allows you to resize a window in place. The
876 associated resizeterm() function simplifies the construction of
877 SIGWINCH handlers, for resizing all windows.
879 The define_key() function allows you to define at runtime function-key
880 control sequences which are not in the terminal description. The
881 keyok() function allows you to temporarily enable or disable
882 interpretation of any function-key control sequence.
884 The use_default_colors() function allows you to construct applications
885 which can use the terminal's default foreground and background colors
886 as an additional "default" color. Several terminal emulators support
887 this feature, which is based on ISO 6429.
889 Ncurses supports up 16 colors, unlike SVr4 curses which defines only
890 8. While most terminals which provide color allow only 8 colors, about
891 a quarter (including XFree86 xterm) support 16 colors.
893 Compatibility with Older Versions
895 Despite our best efforts, there are some differences between ncurses
896 and the (undocumented!) behavior of older curses implementations.
897 These arise from ambiguities or omissions in the documentation of the
900 Refresh of Overlapping Windows
902 If you define two windows A and B that overlap, and then alternately
903 scribble on and refresh them, the changes made to the overlapping
904 region under historic curses versions were often not documented
907 To understand why this is a problem, remember that screen updates are
908 calculated between two representations of the entire display. The
909 documentation says that when you refresh a window, it is first copied
910 to the virtual screen, and then changes are calculated to update the
911 physical screen (and applied to the terminal). But "copied to" is not
912 very specific, and subtle differences in how copying works can produce
913 different behaviors in the case where two overlapping windows are each
914 being refreshed at unpredictable intervals.
916 What happens to the overlapping region depends on what wnoutrefresh()
917 does with its argument -- what portions of the argument window it
918 copies to the virtual screen. Some implementations do "change copy",
919 copying down only locations in the window that have changed (or been
920 marked changed with wtouchln() and friends). Some implementations do
921 "entire copy", copying all window locations to the virtual screen
922 whether or not they have changed.
924 The ncurses library itself has not always been consistent on this
925 score. Due to a bug, versions 1.8.7 to 1.9.8a did entire copy.
926 Versions 1.8.6 and older, and versions 1.9.9 and newer, do change
929 For most commercial curses implementations, it is not documented and
930 not known for sure (at least not to the ncurses maintainers) whether
931 they do change copy or entire copy. We know that System V release 3
932 curses has logic in it that looks like an attempt to do change copy,
933 but the surrounding logic and data representations are sufficiently
934 complex, and our knowledge sufficiently indirect, that it is hard to
935 know whether this is reliable. It is not clear what the SVr4
936 documentation and XSI standard intend. The XSI Curses standard barely
937 mentions wnoutrefresh(); the SVr4 documents seem to be describing
938 entire-copy, but it is possible with some effort and straining to read
941 It might therefore be unwise to rely on either behavior in programs
942 that might have to be linked with other curses implementations.
943 Instead, you can do an explicit touchwin() before the wnoutrefresh()
944 call to guarantee an entire-contents copy anywhere.
946 The really clean way to handle this is to use the panels library. If,
947 when you want a screen update, you do update_panels(), it will do all
948 the necessary wnoutrefresh() calls for whatever panel stacking order
949 you have defined. Then you can do one doupdate() and there will be a
950 single burst of physical I/O that will do all your updates.
954 If you have been using a very old versions of ncurses (1.8.7 or older)
955 you may be surprised by the behavior of the erase functions. In older
956 versions, erased areas of a window were filled with a blank modified
957 by the window's current attribute (as set by wattrset(), wattron(),
958 wattroff() and friends).
960 In newer versions, this is not so. Instead, the attribute of erased
961 blanks is normal unless and until it is modified by the functions
962 bkgdset() or wbkgdset().
964 This change in behavior conforms ncurses to System V Release 4 and the
967 XSI Curses Conformance
969 The ncurses library is intended to be base-level conformant with the
970 XSI Curses standard from X/Open. Many extended-level features (in
971 fact, almost all features not directly concerned with wide characters
972 and internationalization) are also supported.
974 One effect of XSI conformance is the change in behavior described
975 under "Background Erase -- Compatibility with Old Versions".
977 Also, ncurses meets the XSI requirement that every macro entry point
978 have a corresponding function which may be linked (and will be
979 prototype-checked) if the macro definition is disabled with #undef.
983 The ncurses library by itself provides good support for screen
984 displays in which the windows are tiled (non-overlapping). In the more
985 general case that windows may overlap, you have to use a series of
986 wnoutrefresh() calls followed by a doupdate(), and be careful about
987 the order you do the window refreshes in. It has to be bottom-upwards,
988 otherwise parts of windows that should be obscured will show through.
990 When your interface design is such that windows may dive deeper into
991 the visibility stack or pop to the top at runtime, the resulting
992 book-keeping can be tedious and difficult to get right. Hence the
995 The panel library first appeared in AT&T System V. The version
996 documented here is the panel code distributed with ncurses.
998 Compiling With the Panels Library
1000 Your panels-using modules must import the panels library declarations
1004 and must be linked explicitly with the panels library using an -lpanel
1005 argument. Note that they must also link the ncurses library with
1006 -lncurses. Many linkers are two-pass and will accept either order, but
1007 it is still good practice to put -lpanel first and -lncurses second.
1011 A panel object is a window that is implicitly treated as part of a
1012 deck including all other panel objects. The deck has an implicit
1013 bottom-to-top visibility order. The panels library includes an update
1014 function (analogous to refresh()) that displays all panels in the deck
1015 in the proper order to resolve overlaps. The standard window, stdscr,
1016 is considered below all panels.
1018 Details on the panels functions are available in the man pages. We
1019 will just hit the highlights here.
1021 You create a panel from a window by calling new_panel() on a window
1022 pointer. It then becomes the top of the deck. The panel's window is
1023 available as the value of panel_window() called with the panel pointer
1026 You can delete a panel (removing it from the deck) with del_panel.
1027 This will not deallocate the associated window; you have to do that
1028 yourself. You can replace a panel's window with a different window by
1029 calling replace_window. The new window may be of different size; the
1030 panel code will re-compute all overlaps. This operation does not
1031 change the panel's position in the deck.
1033 To move a panel's window, use move_panel(). The mvwin() function on
1034 the panel's window is not sufficient because it does not update the
1035 panels library's representation of where the windows are. This
1036 operation leaves the panel's depth, contents, and size unchanged.
1038 Two functions (top_panel(), bottom_panel()) are provided for
1039 rearranging the deck. The first pops its argument window to the top of
1040 the deck; the second sends it to the bottom. Either operation leaves
1041 the panel's screen location, contents, and size unchanged.
1043 The function update_panels() does all the wnoutrefresh() calls needed
1044 to prepare for doupdate() (which you must call yourself, afterwards).
1046 Typically, you will want to call update_panels() and doupdate() just
1047 before accepting command input, once in each cycle of interaction with
1048 the user. If you call update_panels() after each and every panel
1049 write, you will generate a lot of unnecessary refresh activity and
1052 Panels, Input, and the Standard Screen
1054 You should not mix wnoutrefresh() or wrefresh() operations with panels
1055 code; this will work only if the argument window is either in the top
1056 panel or unobscured by any other panels.
1058 The stsdcr window is a special case. It is considered below all
1059 panels. Because changes to panels may obscure parts of stdscr, though,
1060 you should call update_panels() before doupdate() even when you only
1063 Note that wgetch automatically calls wrefresh. Therefore, before
1064 requesting input from a panel window, you need to be sure that the
1065 panel is totally unobscured.
1067 There is presently no way to display changes to one obscured panel
1068 without repainting all panels.
1072 It is possible to remove a panel from the deck temporarily; use
1073 hide_panel for this. Use show_panel() to render it visible again. The
1074 predicate function panel_hidden tests whether or not a panel is
1077 The panel_update code ignores hidden panels. You cannot do top_panel()
1078 or bottom_panel on a hidden panel(). Other panels operations are
1081 Miscellaneous Other Facilities
1083 It is possible to navigate the deck using the functions panel_above()
1084 and panel_below. Handed a panel pointer, they return the panel above
1085 or below that panel. Handed NULL, they return the bottom-most or
1088 Every panel has an associated user pointer, not used by the panel
1089 code, to which you can attach application data. See the man page
1090 documentation of set_panel_userptr() and panel_userptr for details.
1094 A menu is a screen display that assists the user to choose some subset
1095 of a given set of items. The menu library is a curses extension that
1096 supports easy programming of menu hierarchies with a uniform but
1099 The menu library first appeared in AT&T System V. The version
1100 documented here is the menu code distributed with ncurses.
1102 Compiling With the menu Library
1104 Your menu-using modules must import the menu library declarations with
1107 and must be linked explicitly with the menus library using an -lmenu
1108 argument. Note that they must also link the ncurses library with
1109 -lncurses. Many linkers are two-pass and will accept either order, but
1110 it is still good practice to put -lmenu first and -lncurses second.
1114 The menus created by this library consist of collections of items
1115 including a name string part and a description string part. To make
1116 menus, you create groups of these items and connect them with menu
1119 The menu can then by posted, that is written to an associated window.
1120 Actually, each menu has two associated windows; a containing window in
1121 which the programmer can scribble titles or borders, and a subwindow
1122 in which the menu items proper are displayed. If this subwindow is too
1123 small to display all the items, it will be a scrollable viewport on
1124 the collection of items.
1126 A menu may also be unposted (that is, undisplayed), and finally freed
1127 to make the storage associated with it and its items available for
1130 The general flow of control of a menu program looks like this:
1131 1. Initialize curses.
1132 2. Create the menu items, using new_item().
1133 3. Create the menu using new_menu().
1134 4. Post the menu using post_menu().
1135 5. Refresh the screen.
1136 6. Process user requests via an input loop.
1137 7. Unpost the menu using unpost_menu().
1138 8. Free the menu, using free_menu().
1139 9. Free the items using free_item().
1140 10. Terminate curses.
1144 Menus may be multi-valued or (the default) single-valued (see the
1145 manual page menu_opts(3x) to see how to change the default). Both
1146 types always have a current item.
1148 From a single-valued menu you can read the selected value simply by
1149 looking at the current item. From a multi-valued menu, you get the
1150 selected set by looping through the items applying the item_value()
1151 predicate function. Your menu-processing code can use the function
1152 set_item_value() to flag the items in the select set.
1154 Menu items can be made unselectable using set_item_opts() or
1155 item_opts_off() with the O_SELECTABLE argument. This is the only
1156 option so far defined for menus, but it is good practice to code as
1157 though other option bits might be on.
1161 The menu library calculates a minimum display size for your window,
1162 based on the following variables:
1163 * The number and maximum length of the menu items
1164 * Whether the O_ROWMAJOR option is enabled
1165 * Whether display of descriptions is enabled
1166 * Whatever menu format may have been set by the programmer
1167 * The length of the menu mark string used for highlighting selected
1170 The function set_menu_format() allows you to set the maximum size of
1171 the viewport or menu page that will be used to display menu items. You
1172 can retrieve any format associated with a menu with menu_format(). The
1173 default format is rows=16, columns=1.
1175 The actual menu page may be smaller than the format size. This depends
1176 on the item number and size and whether O_ROWMAJOR is on. This option
1177 (on by default) causes menu items to be displayed in a "raster-scan"
1178 pattern, so that if more than one item will fit horizontally the first
1179 couple of items are side-by-side in the top row. The alternative is
1180 column-major display, which tries to put the first several items in
1183 As mentioned above, a menu format not large enough to allow all items
1184 to fit on-screen will result in a menu display that is vertically
1187 You can scroll it with requests to the menu driver, which will be
1188 described in the section on menu input handling.
1190 Each menu has a mark string used to visually tag selected items; see
1191 the menu_mark(3x) manual page for details. The mark string length also
1192 influences the menu page size.
1194 The function scale_menu() returns the minimum display size that the
1195 menu code computes from all these factors. There are other menu
1196 display attributes including a select attribute, an attribute for
1197 selectable items, an attribute for unselectable items, and a pad
1198 character used to separate item name text from description text. These
1199 have reasonable defaults which the library allows you to change (see
1200 the menu_attribs(3x) manual page.
1204 Each menu has, as mentioned previously, a pair of associated windows.
1205 Both these windows are painted when the menu is posted and erased when
1206 the menu is unposted.
1208 The outer or frame window is not otherwise touched by the menu
1209 routines. It exists so the programmer can associate a title, a border,
1210 or perhaps help text with the menu and have it properly refreshed or
1211 erased at post/unpost time. The inner window or subwindow is where the
1212 current menu page is displayed.
1214 By default, both windows are stdscr. You can set them with the
1215 functions in menu_win(3x).
1217 When you call post_menu(), you write the menu to its subwindow. When
1218 you call unpost_menu(), you erase the subwindow, However, neither of
1219 these actually modifies the screen. To do that, call wrefresh() or
1222 Processing Menu Input
1224 The main loop of your menu-processing code should call menu_driver()
1225 repeatedly. The first argument of this routine is a menu pointer; the
1226 second is a menu command code. You should write an input-fetching
1227 routine that maps input characters to menu command codes, and pass its
1228 output to menu_driver(). The menu command codes are fully documented
1231 The simplest group of command codes is REQ_NEXT_ITEM, REQ_PREV_ITEM,
1232 REQ_FIRST_ITEM, REQ_LAST_ITEM, REQ_UP_ITEM, REQ_DOWN_ITEM,
1233 REQ_LEFT_ITEM, REQ_RIGHT_ITEM. These change the currently selected
1234 item. These requests may cause scrolling of the menu page if it only
1235 partially displayed.
1237 There are explicit requests for scrolling which also change the
1238 current item (because the select location does not change, but the
1239 item there does). These are REQ_SCR_DLINE, REQ_SCR_ULINE,
1240 REQ_SCR_DPAGE, and REQ_SCR_UPAGE.
1242 The REQ_TOGGLE_ITEM selects or deselects the current item. It is for
1243 use in multi-valued menus; if you use it with O_ONEVALUE on, you will
1244 get an error return (E_REQUEST_DENIED).
1246 Each menu has an associated pattern buffer. The menu_driver() logic
1247 tries to accumulate printable ASCII characters passed in in that
1248 buffer; when it matches a prefix of an item name, that item (or the
1249 next matching item) is selected. If appending a character yields no
1250 new match, that character is deleted from the pattern buffer, and
1251 menu_driver() returns E_NO_MATCH.
1253 Some requests change the pattern buffer directly: REQ_CLEAR_PATTERN,
1254 REQ_BACK_PATTERN, REQ_NEXT_MATCH, REQ_PREV_MATCH. The latter two are
1255 useful when pattern buffer input matches more than one item in a
1258 Each successful scroll or item navigation request clears the pattern
1259 buffer. It is also possible to set the pattern buffer explicitly with
1262 Finally, menu driver requests above the constant MAX_COMMAND are
1263 considered application-specific commands. The menu_driver() code
1264 ignores them and returns E_UNKNOWN_COMMAND.
1266 Miscellaneous Other Features
1268 Various menu options can affect the processing and visual appearance
1269 and input processing of menus. See menu_opts(3x) for details.
1271 It is possible to change the current item from application code; this
1272 is useful if you want to write your own navigation requests. It is
1273 also possible to explicitly set the top row of the menu display. See
1274 mitem_current(3x). If your application needs to change the menu
1275 subwindow cursor for any reason, pos_menu_cursor() will restore it to
1276 the correct location for continuing menu driver processing.
1278 It is possible to set hooks to be called at menu initialization and
1279 wrapup time, and whenever the selected item changes. See
1282 Each item, and each menu, has an associated user pointer on which you
1283 can hang application data. See mitem_userptr(3x) and menu_userptr(3x).
1287 The form library is a curses extension that supports easy programming
1288 of on-screen forms for data entry and program control.
1290 The form library first appeared in AT&T System V. The version
1291 documented here is the form code distributed with ncurses.
1293 Compiling With the form Library
1295 Your form-using modules must import the form library declarations with
1298 and must be linked explicitly with the forms library using an -lform
1299 argument. Note that they must also link the ncurses library with
1300 -lncurses. Many linkers are two-pass and will accept either order, but
1301 it is still good practice to put -lform first and -lncurses second.
1305 A form is a collection of fields; each field may be either a label
1306 (explanatory text) or a data-entry location. Long forms may be
1307 segmented into pages; each entry to a new page clears the screen.
1309 To make forms, you create groups of fields and connect them with form
1310 frame objects; the form library makes this relatively simple.
1312 Once defined, a form can be posted, that is written to an associated
1313 window. Actually, each form has two associated windows; a containing
1314 window in which the programmer can scribble titles or borders, and a
1315 subwindow in which the form fields proper are displayed.
1317 As the form user fills out the posted form, navigation and editing
1318 keys support movement between fields, editing keys support modifying
1319 field, and plain text adds to or changes data in a current field. The
1320 form library allows you (the forms designer) to bind each navigation
1321 and editing key to any keystroke accepted by curses Fields may have
1322 validation conditions on them, so that they check input data for type
1323 and value. The form library supplies a rich set of pre-defined field
1324 types, and makes it relatively easy to define new ones.
1326 Once its transaction is completed (or aborted), a form may be unposted
1327 (that is, undisplayed), and finally freed to make the storage
1328 associated with it and its items available for re-use.
1330 The general flow of control of a form program looks like this:
1331 1. Initialize curses.
1332 2. Create the form fields, using new_field().
1333 3. Create the form using new_form().
1334 4. Post the form using post_form().
1335 5. Refresh the screen.
1336 6. Process user requests via an input loop.
1337 7. Unpost the form using unpost_form().
1338 8. Free the form, using free_form().
1339 9. Free the fields using free_field().
1340 10. Terminate curses.
1342 Note that this looks much like a menu program; the form library
1343 handles tasks which are in many ways similar, and its interface was
1344 obviously designed to resemble that of the menu library wherever
1347 In forms programs, however, the "process user requests" is somewhat
1348 more complicated than for menus. Besides menu-like navigation
1349 operations, the menu driver loop has to support field editing and data
1352 Creating and Freeing Fields and Forms
1354 The basic function for creating fields is new_field():
1355 FIELD *new_field(int height, int width, /* new field size */
1356 int top, int left, /* upper left corner */
1357 int offscreen, /* number of offscreen rows */
1358 int nbuf); /* number of working buffers */
1360 Menu items always occupy a single row, but forms fields may have
1361 multiple rows. So new_field() requires you to specify a width and
1362 height (the first two arguments, which mist both be greater than
1365 You must also specify the location of the field's upper left corner on
1366 the screen (the third and fourth arguments, which must be zero or
1367 greater). Note that these coordinates are relative to the form
1368 subwindow, which will coincide with stdscr by default but need not be
1369 stdscr if you have done an explicit set_form_win() call.
1371 The fifth argument allows you to specify a number of off-screen rows.
1372 If this is zero, the entire field will always be displayed. If it is
1373 nonzero, the form will be scrollable, with only one screen-full
1374 (initially the top part) displayed at any given time. If you make a
1375 field dynamic and grow it so it will no longer fit on the screen, the
1376 form will become scrollable even if the offscreen argument was
1379 The forms library allocates one working buffer per field; the size of
1380 each buffer is ((height + offscreen)*width + 1, one character for each
1381 position in the field plus a NUL terminator. The sixth argument is the
1382 number of additional data buffers to allocate for the field; your
1383 application can use them for its own purposes.
1384 FIELD *dup_field(FIELD *field, /* field to copy */
1385 int top, int left); /* location of new copy */
1387 The function dup_field() duplicates an existing field at a new
1388 location. Size and buffering information are copied; some attribute
1389 flags and status bits are not (see the form_field_new(3X) for
1391 FIELD *link_field(FIELD *field, /* field to copy */
1392 int top, int left); /* location of new copy */
1394 The function link_field() also duplicates an existing field at a new
1395 location. The difference from dup_field() is that it arranges for the
1396 new field's buffer to be shared with the old one.
1398 Besides the obvious use in making a field editable from two different
1399 form pages, linked fields give you a way to hack in dynamic labels. If
1400 you declare several fields linked to an original, and then make them
1401 inactive, changes from the original will still be propagated to the
1404 As with duplicated fields, linked fields have attribute bits separate
1407 As you might guess, all these field-allocations return NULL if the
1408 field allocation is not possible due to an out-of-memory error or
1409 out-of-bounds arguments.
1411 To connect fields to a form, use
1412 FORM *new_form(FIELD **fields);
1414 This function expects to see a NULL-terminated array of field
1415 pointers. Said fields are connected to a newly-allocated form object;
1416 its address is returned (or else NULL if the allocation fails).
1418 Note that new_field() does not copy the pointer array into private
1419 storage; if you modify the contents of the pointer array during forms
1420 processing, all manner of bizarre things might happen. Also note that
1421 any given field may only be connected to one form.
1423 The functions free_field() and free_form are available to free field
1424 and form objects. It is an error to attempt to free a field connected
1425 to a form, but not vice-versa; thus, you will generally free your form
1428 Fetching and Changing Field Attributes
1430 Each form field has a number of location and size attributes
1431 associated with it. There are other field attributes used to control
1432 display and editing of the field. Some (for example, the O_STATIC bit)
1433 involve sufficient complications to be covered in sections of their
1434 own later on. We cover the functions used to get and set several basic
1437 When a field is created, the attributes not specified by the new_field
1438 function are copied from an invisible system default field. In
1439 attribute-setting and -fetching functions, the argument NULL is taken
1440 to mean this field. Changes to it persist as defaults until your forms
1441 application terminates.
1443 Fetching Size and Location Data
1445 You can retrieve field sizes and locations through:
1446 int field_info(FIELD *field, /* field from which to fetch */
1447 int *height, *int width, /* field size */
1448 int *top, int *left, /* upper left corner */
1449 int *offscreen, /* number of offscreen rows */
1450 int *nbuf); /* number of working buffers */
1452 This function is a sort of inverse of new_field(); instead of setting
1453 size and location attributes of a new field, it fetches them from an
1456 Changing the Field Location
1458 It is possible to move a field's location on the screen:
1459 int move_field(FIELD *field, /* field to alter */
1460 int top, int left); /* new upper-left corner */
1462 You can, of course. query the current location through field_info().
1464 The Justification Attribute
1466 One-line fields may be unjustified, justified right, justified left,
1467 or centered. Here is how you manipulate this attribute:
1468 int set_field_just(FIELD *field, /* field to alter */
1469 int justmode); /* mode to set */
1471 int field_just(FIELD *field); /* fetch mode of field */
1473 The mode values accepted and returned by this functions are
1474 preprocessor macros NO_JUSTIFICATION, JUSTIFY_RIGHT, JUSTIFY_LEFT, or
1477 Field Display Attributes
1479 For each field, you can set a foreground attribute for entered
1480 characters, a background attribute for the entire field, and a pad
1481 character for the unfilled portion of the field. You can also control
1482 pagination of the form.
1484 This group of four field attributes controls the visual appearance of
1485 the field on the screen, without affecting in any way the data in the
1487 int set_field_fore(FIELD *field, /* field to alter */
1488 chtype attr); /* attribute to set */
1490 chtype field_fore(FIELD *field); /* field to query */
1492 int set_field_back(FIELD *field, /* field to alter */
1493 chtype attr); /* attribute to set */
1495 chtype field_back(FIELD *field); /* field to query */
1497 int set_field_pad(FIELD *field, /* field to alter */
1498 int pad); /* pad character to set */
1500 chtype field_pad(FIELD *field);
1502 int set_new_page(FIELD *field, /* field to alter */
1503 int flag); /* TRUE to force new page */
1505 chtype new_page(FIELD *field); /* field to query */
1507 The attributes set and returned by the first four functions are normal
1508 curses(3x) display attribute values (A_STANDOUT, A_BOLD, A_REVERSE
1509 etc). The page bit of a field controls whether it is displayed at the
1510 start of a new form screen.
1514 There is also a large collection of field option bits you can set to
1515 control various aspects of forms processing. You can manipulate them
1516 with these functions:
1517 int set_field_opts(FIELD *field, /* field to alter */
1518 int attr); /* attribute to set */
1520 int field_opts_on(FIELD *field, /* field to alter */
1521 int attr); /* attributes to turn on */
1523 int field_opts_off(FIELD *field, /* field to alter */
1524 int attr); /* attributes to turn off */
1526 int field_opts(FIELD *field); /* field to query */
1528 By default, all options are on. Here are the available option bits:
1531 Controls whether the field is visible on the screen. Can be
1532 used during form processing to hide or pop up fields depending
1533 on the value of parent fields.
1536 Controls whether the field is active during forms processing
1537 (i.e. visited by form navigation keys). Can be used to make
1538 labels or derived fields with buffer values alterable by the
1539 forms application, not the user.
1542 Controls whether data is displayed during field entry. If this
1543 option is turned off on a field, the library will accept and
1544 edit data in that field, but it will not be displayed and the
1545 visible field cursor will not move. You can turn off the
1546 O_PUBLIC bit to define password fields.
1549 Controls whether the field's data can be modified. When this
1550 option is off, all editing requests except REQ_PREV_CHOICE and
1551 REQ_NEXT_CHOICE will fail. Such read-only fields may be useful
1555 Controls word-wrapping in multi-line fields. Normally, when any
1556 character of a (blank-separated) word reaches the end of the
1557 current line, the entire word is wrapped to the next line
1558 (assuming there is one). When this option is off, the word will
1559 be split across the line break.
1562 Controls field blanking. When this option is on, entering a
1563 character at the first field position erases the entire field
1564 (except for the just-entered character).
1567 Controls automatic skip to next field when this one fills.
1568 Normally, when the forms user tries to type more data into a
1569 field than will fit, the editing location jumps to next field.
1570 When this option is off, the user's cursor will hang at the end
1571 of the field. This option is ignored in dynamic fields that
1572 have not reached their size limit.
1575 Controls whether validation is applied to blank fields.
1576 Normally, it is not; the user can leave a field blank without
1577 invoking the usual validation check on exit. If this option is
1578 off on a field, exit from it will invoke a validation check.
1581 Controls whether validation occurs on every exit, or only after
1582 the field is modified. Normally the latter is true. Setting
1583 O_PASSOK may be useful if your field's validation function may
1584 change during forms processing.
1587 Controls whether the field is fixed to its initial dimensions.
1588 If you turn this off, the field becomes dynamic and will
1589 stretch to fit entered data.
1591 A field's options cannot be changed while the field is currently
1592 selected. However, options may be changed on posted fields that are
1595 The option values are bit-masks and can be composed with logical-or in
1600 Every field has a status flag, which is set to FALSE when the field is
1601 created and TRUE when the value in field buffer 0 changes. This flag
1602 can be queried and set directly:
1603 int set_field_status(FIELD *field, /* field to alter */
1604 int status); /* mode to set */
1606 int field_status(FIELD *field); /* fetch mode of field */
1608 Setting this flag under program control can be useful if you use the
1609 same form repeatedly, looking for modified fields each time.
1611 Calling field_status() on a field not currently selected for input
1612 will return a correct value. Calling field_status() on a field that is
1613 currently selected for input may not necessarily give a correct field
1614 status value, because entered data is not necessarily copied to buffer
1615 zero before the exit validation check. To guarantee that the returned
1616 status value reflects reality, call field_status() either (1) in the
1617 field's exit validation check routine, (2) from the field's or form's
1618 initialization or termination hooks, or (3) just after a
1619 REQ_VALIDATION request has been processed by the forms driver.
1623 Each field structure contains one character pointer slot that is not
1624 used by the forms library. It is intended to be used by applications
1625 to store private per-field data. You can manipulate it with:
1626 int set_field_userptr(FIELD *field, /* field to alter */
1627 char *userptr); /* mode to set */
1629 char *field_userptr(FIELD *field); /* fetch mode of field */
1631 (Properly, this user pointer field ought to have (void *) type. The
1632 (char *) type is retained for System V compatibility.)
1634 It is valid to set the user pointer of the default field (with a
1635 set_field_userptr() call passed a NULL field pointer.) When a new
1636 field is created, the default-field user pointer is copied to
1637 initialize the new field's user pointer.
1639 Variable-Sized Fields
1641 Normally, a field is fixed at the size specified for it at creation
1642 time. If, however, you turn off its O_STATIC bit, it becomes dynamic
1643 and will automatically resize itself to accommodate data as it is
1644 entered. If the field has extra buffers associated with it, they will
1645 grow right along with the main input buffer.
1647 A one-line dynamic field will have a fixed height (1) but variable
1648 width, scrolling horizontally to display data within the field area as
1649 originally dimensioned and located. A multi-line dynamic field will
1650 have a fixed width, but variable height (number of rows), scrolling
1651 vertically to display data within the field area as originally
1652 dimensioned and located.
1654 Normally, a dynamic field is allowed to grow without limit. But it is
1655 possible to set an upper limit on the size of a dynamic field. You do
1656 it with this function:
1657 int set_max_field(FIELD *field, /* field to alter (may not be NULL) */
1658 int max_size); /* upper limit on field size */
1660 If the field is one-line, max_size is taken to be a column size limit;
1661 if it is multi-line, it is taken to be a line size limit. To disable
1662 any limit, use an argument of zero. The growth limit can be changed
1663 whether or not the O_STATIC bit is on, but has no effect until it is.
1665 The following properties of a field change when it becomes dynamic:
1666 * If there is no growth limit, there is no final position of the
1667 field; therefore O_AUTOSKIP and O_NL_OVERLOAD are ignored.
1668 * Field justification will be ignored (though whatever justification
1669 is set up will be retained internally and can be queried).
1670 * The dup_field() and link_field() calls copy dynamic-buffer sizes.
1671 If the O_STATIC option is set on one of a collection of links,
1672 buffer resizing will occur only when the field is edited through
1674 * The call field_info() will retrieve the original static size of
1675 the field; use dynamic_field_info() to get the actual dynamic
1680 By default, a field will accept any data that will fit in its input
1681 buffer. However, it is possible to attach a validation type to a
1682 field. If you do this, any attempt to leave the field while it
1683 contains data that does not match the validation type will fail. Some
1684 validation types also have a character-validity check for each time a
1685 character is entered in the field.
1687 A field's validation check (if any) is not called when
1688 set_field_buffer() modifies the input buffer, nor when that buffer is
1689 changed through a linked field.
1691 The form library provides a rich set of pre-defined validation types,
1692 and gives you the capability to define custom ones of your own. You
1693 can examine and change field validation attributes with the following
1695 int set_field_type(FIELD *field, /* field to alter */
1696 FIELDTYPE *ftype, /* type to associate */
1697 ...); /* additional arguments*/
1699 FIELDTYPE *field_type(FIELD *field); /* field to query */
1701 The validation type of a field is considered an attribute of the
1702 field. As with other field attributes, Also, doing set_field_type()
1703 with a NULL field default will change the system default for
1704 validation of newly-created fields.
1706 Here are the pre-defined validation types:
1710 This field type accepts alphabetic data; no blanks, no digits, no
1711 special characters (this is checked at character-entry time). It is
1713 int set_field_type(FIELD *field, /* field to alter */
1714 TYPE_ALPHA, /* type to associate */
1715 int width); /* maximum width of field */
1717 The width argument sets a minimum width of data. Typically you will
1718 want to set this to the field width; if it is greater than the field
1719 width, the validation check will always fail. A minimum width of zero
1720 makes field completion optional.
1724 This field type accepts alphabetic data and digits; no blanks, no
1725 special characters (this is checked at character-entry time). It is
1727 int set_field_type(FIELD *field, /* field to alter */
1728 TYPE_ALNUM, /* type to associate */
1729 int width); /* maximum width of field */
1731 The width argument sets a minimum width of data. As with TYPE_ALPHA,
1732 typically you will want to set this to the field width; if it is
1733 greater than the field width, the validation check will always fail. A
1734 minimum width of zero makes field completion optional.
1738 This type allows you to restrict a field's values to be among a
1739 specified set of string values (for example, the two-letter postal
1740 codes for U.S. states). It is set up with:
1741 int set_field_type(FIELD *field, /* field to alter */
1742 TYPE_ENUM, /* type to associate */
1743 char **valuelist; /* list of possible values */
1744 int checkcase; /* case-sensitive? */
1745 int checkunique); /* must specify uniquely? */
1747 The valuelist parameter must point at a NULL-terminated list of valid
1748 strings. The checkcase argument, if true, makes comparison with the
1749 string case-sensitive.
1751 When the user exits a TYPE_ENUM field, the validation procedure tries
1752 to complete the data in the buffer to a valid entry. If a complete
1753 choice string has been entered, it is of course valid. But it is also
1754 possible to enter a prefix of a valid string and have it completed for
1757 By default, if you enter such a prefix and it matches more than one
1758 value in the string list, the prefix will be completed to the first
1759 matching value. But the checkunique argument, if true, requires prefix
1760 matches to be unique in order to be valid.
1762 The REQ_NEXT_CHOICE and REQ_PREV_CHOICE input requests can be
1763 particularly useful with these fields.
1767 This field type accepts an integer. It is set up as follows:
1768 int set_field_type(FIELD *field, /* field to alter */
1769 TYPE_INTEGER, /* type to associate */
1770 int padding, /* # places to zero-pad to */
1771 int vmin, int vmax); /* valid range */
1773 Valid characters consist of an optional leading minus and digits. The
1774 range check is performed on exit. If the range maximum is less than or
1775 equal to the minimum, the range is ignored.
1777 If the value passes its range check, it is padded with as many leading
1778 zero digits as necessary to meet the padding argument.
1780 A TYPE_INTEGER value buffer can conveniently be interpreted with the C
1781 library function atoi(3).
1785 This field type accepts a decimal number. It is set up as follows:
1786 int set_field_type(FIELD *field, /* field to alter */
1787 TYPE_NUMERIC, /* type to associate */
1788 int padding, /* # places of precision */
1789 double vmin, double vmax); /* valid range */
1791 Valid characters consist of an optional leading minus and digits.
1792 possibly including a decimal point. If your system supports locale's,
1793 the decimal point character used must be the one defined by your
1794 locale. The range check is performed on exit. If the range maximum is
1795 less than or equal to the minimum, the range is ignored.
1797 If the value passes its range check, it is padded with as many
1798 trailing zero digits as necessary to meet the padding argument.
1800 A TYPE_NUMERIC value buffer can conveniently be interpreted with the C
1801 library function atof(3).
1805 This field type accepts data matching a regular expression. It is set
1807 int set_field_type(FIELD *field, /* field to alter */
1808 TYPE_REGEXP, /* type to associate */
1809 char *regexp); /* expression to match */
1811 The syntax for regular expressions is that of regcomp(3). The check
1812 for regular-expression match is performed on exit.
1814 Direct Field Buffer Manipulation
1816 The chief attribute of a field is its buffer contents. When a form has
1817 been completed, your application usually needs to know the state of
1818 each field buffer. You can find this out with:
1819 char *field_buffer(FIELD *field, /* field to query */
1820 int bufindex); /* number of buffer to query */
1822 Normally, the state of the zero-numbered buffer for each field is set
1823 by the user's editing actions on that field. It is sometimes useful to
1824 be able to set the value of the zero-numbered (or some other) buffer
1825 from your application:
1826 int set_field_buffer(FIELD *field, /* field to alter */
1827 int bufindex, /* number of buffer to alter */
1828 char *value); /* string value to set */
1830 If the field is not large enough and cannot be resized to a
1831 sufficiently large size to contain the specified value, the value will
1832 be truncated to fit.
1834 Calling field_buffer() with a null field pointer will raise an error.
1835 Calling field_buffer() on a field not currently selected for input
1836 will return a correct value. Calling field_buffer() on a field that is
1837 currently selected for input may not necessarily give a correct field
1838 buffer value, because entered data is not necessarily copied to buffer
1839 zero before the exit validation check. To guarantee that the returned
1840 buffer value reflects on-screen reality, call field_buffer() either
1841 (1) in the field's exit validation check routine, (2) from the field's
1842 or form's initialization or termination hooks, or (3) just after a
1843 REQ_VALIDATION request has been processed by the forms driver.
1847 As with field attributes, form attributes inherit a default from a
1848 system default form structure. These defaults can be queried or set by
1849 of these functions using a form-pointer argument of NULL.
1851 The principal attribute of a form is its field list. You can query and
1852 change this list with:
1853 int set_form_fields(FORM *form, /* form to alter */
1854 FIELD **fields); /* fields to connect */
1856 char *form_fields(FORM *form); /* fetch fields of form */
1858 int field_count(FORM *form); /* count connect fields */
1860 The second argument of set_form_fields() may be a NULL-terminated
1861 field pointer array like the one required by new_form(). In that case,
1862 the old fields of the form are disconnected but not freed (and
1863 eligible to be connected to other forms), then the new fields are
1866 It may also be null, in which case the old fields are disconnected
1867 (and not freed) but no new ones are connected.
1869 The field_count() function simply counts the number of fields
1870 connected to a given from. It returns -1 if the form-pointer argument
1873 Control of Form Display
1875 In the overview section, you saw that to display a form you normally
1876 start by defining its size (and fields), posting it, and refreshing
1877 the screen. There is an hidden step before posting, which is the
1878 association of the form with a frame window (actually, a pair of
1879 windows) within which it will be displayed. By default, the forms
1880 library associates every form with the full-screen window stdscr.
1882 By making this step explicit, you can associate a form with a declared
1883 frame window on your screen display. This can be useful if you want to
1884 adapt the form display to different screen sizes, dynamically tile
1885 forms on the screen, or use a form as part of an interface layout
1888 The two windows associated with each form have the same functions as
1889 their analogues in the menu library. Both these windows are painted
1890 when the form is posted and erased when the form is unposted.
1892 The outer or frame window is not otherwise touched by the form
1893 routines. It exists so the programmer can associate a title, a border,
1894 or perhaps help text with the form and have it properly refreshed or
1895 erased at post/unpost time. The inner window or subwindow is where the
1896 current form page is actually displayed.
1898 In order to declare your own frame window for a form, you will need to
1899 know the size of the form's bounding rectangle. You can get this
1901 int scale_form(FORM *form, /* form to query */
1902 int *rows, /* form rows */
1903 int *cols); /* form cols */
1905 The form dimensions are passed back in the locations pointed to by the
1906 arguments. Once you have this information, you can use it to declare
1907 of windows, then use one of these functions:
1908 int set_form_win(FORM *form, /* form to alter */
1909 WINDOW *win); /* frame window to connect */
1911 WINDOW *form_win(FORM *form); /* fetch frame window of form */
1913 int set_form_sub(FORM *form, /* form to alter */
1914 WINDOW *win); /* form subwindow to connect */
1916 WINDOW *form_sub(FORM *form); /* fetch form subwindow of form */
1918 Note that curses operations, including refresh(), on the form, should
1919 be done on the frame window, not the form subwindow.
1921 It is possible to check from your application whether all of a
1922 scrollable field is actually displayed within the menu subwindow. Use
1924 int data_ahead(FORM *form); /* form to be queried */
1926 int data_behind(FORM *form); /* form to be queried */
1928 The function data_ahead() returns TRUE if (a) the current field is
1929 one-line and has undisplayed data off to the right, (b) the current
1930 field is multi-line and there is data off-screen below it.
1932 The function data_behind() returns TRUE if the first (upper left hand)
1933 character position is off-screen (not being displayed).
1935 Finally, there is a function to restore the form window's cursor to
1936 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).
1948 int form_driver(FORM *form, /* form to pass input to */
1949 int request); /* form request code */
1951 Your input virtualization function needs to take input and then
1952 convert it to either an alphanumeric character (which is treated as
1953 data to be entered in the currently-selected field), or a forms
1956 The forms driver provides hooks (through input-validation and
1957 field-termination functions) with which your application code can
1958 check that the input taken by the driver matched what was expected.
1960 Page Navigation Requests
1962 These requests cause page-level moves through the form, triggering
1963 display of a new form screen.
1966 Move to the next form page.
1969 Move to the previous form page.
1972 Move to the first form page.
1975 Move to the last form page.
1977 These requests treat the list as cyclic; that is, REQ_NEXT_PAGE from
1978 the last page goes to the first, and REQ_PREV_PAGE from the first page
1981 Inter-Field Navigation Requests
1983 These requests handle navigation between fields on the same page.
1989 Move to previous field.
1992 Move to the first field.
1995 Move to the last field.
1998 Move to sorted next field.
2001 Move to sorted previous field.
2004 Move to the sorted first field.
2007 Move to the sorted last field.
2013 Move right to field.
2021 These requests treat the list of fields on a page as cyclic; that is,
2022 REQ_NEXT_FIELD from the last field goes to the first, and
2023 REQ_PREV_FIELD from the first field goes to the last. The order of the
2024 fields for these (and the REQ_FIRST_FIELD and REQ_LAST_FIELD requests)
2025 is simply the order of the field pointers in the form array (as set up
2026 by new_form() or set_form_fields()
2028 It is also possible to traverse the fields as if they had been sorted
2029 in screen-position order, so the sequence goes left-to-right and
2030 top-to-bottom. To do this, use the second group of four
2031 sorted-movement requests.
2033 Finally, it is possible to move between fields using visual directions
2034 up, down, right, and left. To accomplish this, use the third group of
2035 four requests. Note, however, that the position of a form for purposes
2036 of these requests is its upper-left corner.
2038 For example, suppose you have a multi-line field B, and two
2039 single-line fields A and C on the same line with B, with A to the left
2040 of B and C to the right of B. A REQ_MOVE_RIGHT from A will go to B
2041 only if A, B, and C all share the same first line; otherwise it will
2044 Intra-Field Navigation Requests
2046 These requests drive movement of the edit cursor within the currently
2050 Move to next character.
2053 Move to previous character.
2059 Move to previous line.
2065 Move to previous word.
2068 Move to beginning of field.
2071 Move to end of field.
2074 Move to beginning of line.
2077 Move to end of line.
2083 Move right in field.
2091 Each word is separated from the previous and next characters by
2092 whitespace. The commands to move to beginning and end of line or field
2093 look for the first or last non-pad character in their ranges.
2097 Fields that are dynamic and have grown and fields explicitly created
2098 with offscreen rows are scrollable. One-line fields scroll
2099 horizontally; multi-line fields scroll vertically. Most scrolling is
2100 triggered by editing and intra-field movement (the library scrolls the
2101 field to keep the cursor visible). It is possible to explicitly
2102 request scrolling with the following requests:
2105 Scroll vertically forward a line.
2108 Scroll vertically backward a line.
2111 Scroll vertically forward a page.
2114 Scroll vertically backward a page.
2117 Scroll vertically forward half a page.
2120 Scroll vertically backward half a page.
2123 Scroll horizontally forward a character.
2126 Scroll horizontally backward a character.
2129 Scroll horizontally one field width forward.
2132 Scroll horizontally one field width backward.
2135 Scroll horizontally one half field width forward.
2138 Scroll horizontally one half field width backward.
2140 For scrolling purposes, a page of a field is the height of its visible
2145 When you pass the forms driver an ASCII character, it is treated as a
2146 request to add the character to the field's data buffer. Whether this
2147 is an insertion or a replacement depends on the field's edit mode
2148 (insertion is the default.
2150 The following requests support editing the field and changing the edit
2160 New line request (see below for explanation).
2163 Insert space at character location.
2166 Insert blank line at character location.
2169 Delete character at cursor.
2172 Delete previous word at cursor.
2175 Delete line at cursor.
2178 Delete word at cursor.
2181 Clear to end of line.
2184 Clear to end of field.
2189 The behavior of the REQ_NEW_LINE and REQ_DEL_PREV requests is
2190 complicated and partly controlled by a pair of forms options. The
2191 special cases are triggered when the cursor is at the beginning of a
2192 field, or on the last line of the field.
2194 First, we consider REQ_NEW_LINE:
2196 The normal behavior of REQ_NEW_LINE in insert mode is to break the
2197 current line at the position of the edit cursor, inserting the portion
2198 of the current line after the cursor as a new line following the
2199 current and moving the cursor to the beginning of that new line (you
2200 may think of this as inserting a newline in the field buffer).
2202 The normal behavior of REQ_NEW_LINE in overlay mode is to clear the
2203 current line from the position of the edit cursor to end of line. The
2204 cursor is then moved to the beginning of the next line.
2206 However, REQ_NEW_LINE at the beginning of a field, or on the last line
2207 of a field, instead does a REQ_NEXT_FIELD. O_NL_OVERLOAD option is
2208 off, this special action is disabled.
2210 Now, let us consider REQ_DEL_PREV:
2212 The normal behavior of REQ_DEL_PREV is to delete the previous
2213 character. If insert mode is on, and the cursor is at the start of a
2214 line, and the text on that line will fit on the previous one, it
2215 instead appends the contents of the current line to the previous one
2216 and deletes the current line (you may think of this as deleting a
2217 newline from the field buffer).
2219 However, REQ_DEL_PREV at the beginning of a field is instead treated
2220 as a REQ_PREV_FIELD.
2222 If the O_BS_OVERLOAD option is off, this special action is disabled
2223 and the forms driver just returns E_REQUEST_DENIED.
2225 See Form Options for discussion of how to set and clear the overload
2230 If the type of your field is ordered, and has associated functions for
2231 getting the next and previous values of the type from a given value,
2232 there are requests that can fetch that value into the field buffer:
2235 Place the successor value of the current value in the buffer.
2238 Place the predecessor value of the current value in the buffer.
2240 Of the built-in field types, only TYPE_ENUM has built-in successor and
2241 predecessor functions. When you define a field type of your own (see
2242 Custom Validation Types), you can associate our own ordering
2245 Application Commands
2247 Form requests are represented as integers above the curses value
2248 greater than KEY_MAX and less than or equal to the constant
2249 MAX_COMMAND. If your input-virtualization routine returns a value
2250 above MAX_COMMAND, the forms driver will ignore it.
2254 It is possible to set function hooks to be executed whenever the
2255 current field or form changes. Here are the functions that support
2257 typedef void (*HOOK)(); /* pointer to function returning void */
2259 int set_form_init(FORM *form, /* form to alter */
2260 HOOK hook); /* initialization hook */
2262 HOOK form_init(FORM *form); /* form to query */
2264 int set_form_term(FORM *form, /* form to alter */
2265 HOOK hook); /* termination hook */
2267 HOOK form_term(FORM *form); /* form to query */
2269 int set_field_init(FORM *form, /* form to alter */
2270 HOOK hook); /* initialization hook */
2272 HOOK field_init(FORM *form); /* form to query */
2274 int set_field_term(FORM *form, /* form to alter */
2275 HOOK hook); /* termination hook */
2277 HOOK field_term(FORM *form); /* form to query */
2279 These functions allow you to either set or query four different hooks.
2280 In each of the set functions, the second argument should be the
2281 address of a hook function. These functions differ only in the timing
2285 This hook is called when the form is posted; also, just after
2286 each page change operation.
2289 This hook is called when the form is posted; also, just after
2293 This hook is called just after field validation; that is, just
2294 before the field is altered. It is also called when the form is
2298 This hook is called when the form is unposted; also, just
2299 before each page change operation.
2301 Calls to these hooks may be triggered
2302 1. When user editing requests are processed by the forms driver
2303 2. When the current page is changed by set_current_field() call
2304 3. When the current field is changed by a set_form_page() call
2306 See Field Change Commands for discussion of the latter two cases.
2308 You can set a default hook for all fields by passing one of the set
2309 functions a NULL first argument.
2311 You can disable any of these hooks by (re)setting them to NULL, the
2314 Field Change Commands
2316 Normally, navigation through the form will be driven by the user's
2317 input requests. But sometimes it is useful to be able to move the
2318 focus for editing and viewing under control of your application, or
2319 ask which field it currently is in. The following functions help you
2321 int set_current_field(FORM *form, /* form to alter */
2322 FIELD *field); /* field to shift to */
2324 FIELD *current_field(FORM *form); /* form to query */
2326 int field_index(FORM *form, /* form to query */
2327 FIELD *field); /* field to get index of */
2329 The function field_index() returns the index of the given field in the
2330 given form's field array (the array passed to new_form() or
2333 The initial current field of a form is the first active field on the
2334 first page. The function set_form_fields() resets this.
2336 It is also possible to move around by pages.
2337 int set_form_page(FORM *form, /* form to alter */
2338 int page); /* page to go to (0-origin) */
2340 int form_page(FORM *form); /* return form's current page */
2342 The initial page of a newly-created form is 0. The function
2343 set_form_fields() resets this.
2347 Like fields, forms may have control option bits. They can be changed
2348 or queried with these functions:
2349 int set_form_opts(FORM *form, /* form to alter */
2350 int attr); /* attribute to set */
2352 int form_opts_on(FORM *form, /* form to alter */
2353 int attr); /* attributes to turn on */
2355 int form_opts_off(FORM *form, /* form to alter */
2356 int attr); /* attributes to turn off */
2358 int form_opts(FORM *form); /* form to query */
2360 By default, all options are on. Here are the available option bits:
2363 Enable overloading of REQ_NEW_LINE as described in Editing
2364 Requests. The value of this option is ignored on dynamic fields
2365 that have not reached their size limit; these have no last
2366 line, so the circumstances for triggering a REQ_NEXT_FIELD
2370 Enable overloading of REQ_DEL_PREV as described in Editing
2373 The option values are bit-masks and can be composed with logical-or in
2376 Custom Validation Types
2378 The form library gives you the capability to define custom validation
2379 types of your own. Further, the optional additional arguments of
2380 set_field_type effectively allow you to parameterize validation types.
2381 Most of the complications in the validation-type interface have to do
2382 with the handling of the additional arguments within custom validation
2387 The simplest way to create a custom data type is to compose it from
2388 two preexisting ones:
2389 FIELD *link_fieldtype(FIELDTYPE *type1,
2392 This function creates a field type that will accept any of the values
2393 legal for either of its argument field types (which may be either
2394 predefined or programmer-defined). If a set_field_type() call later
2395 requires arguments, the new composite type expects all arguments for
2396 the first type, than all arguments for the second. Order functions
2397 (see Order Requests) associated with the component types will work on
2398 the composite; what it does is check the validation function for the
2399 first type, then for the second, to figure what type the buffer
2400 contents should be treated as.
2404 To create a field type from scratch, you need to specify one or both
2405 of the following things:
2406 * A character-validation function, to check each character as it is
2408 * A field-validation function to be applied on exit from the field.
2410 Here is how you do that:
2411 typedef int (*HOOK)(); /* pointer to function returning int */
2413 FIELDTYPE *new_fieldtype(HOOK f_validate, /* field validator */
2414 HOOK c_validate) /* character validator */
2416 int free_fieldtype(FIELDTYPE *ftype); /* type to free */
2418 At least one of the arguments of new_fieldtype() must be non-NULL. The
2419 forms driver will automatically call the new type's validation
2420 functions at appropriate points in processing a field of the new type.
2422 The function free_fieldtype() deallocates the argument fieldtype,
2423 freeing all storage associated with it.
2425 Normally, a field validator is called when the user attempts to leave
2426 the field. Its first argument is a field pointer, from which it can
2427 get to field buffer 0 and test it. If the function returns TRUE, the
2428 operation succeeds; if it returns FALSE, the edit cursor stays in the
2431 A character validator gets the character passed in as a first
2432 argument. It too should return TRUE if the character is valid, FALSE
2435 Validation Function Arguments
2437 Your field- and character- validation functions will be passed a
2438 second argument as well. This second argument is the address of a
2439 structure (which we will call a pile) built from any of the
2440 field-type-specific arguments passed to set_field_type(). If no such
2441 arguments are defined for the field type, this pile pointer argument
2444 In order to arrange for such arguments to be passed to your validation
2445 functions, you must associate a small set of storage-management
2446 functions with the type. The forms driver will use these to synthesize
2447 a pile from the trailing arguments of each set_field_type() argument,
2448 and a pointer to the pile will be passed to the validation functions.
2450 Here is how you make the association:
2451 typedef char *(*PTRHOOK)(); /* pointer to function returning (char *) */
2452 typedef void (*VOIDHOOK)(); /* pointer to function returning void */
2454 int set_fieldtype_arg(FIELDTYPE *type, /* type to alter */
2455 PTRHOOK make_str, /* make structure from args */
2456 PTRHOOK copy_str, /* make copy of structure */
2457 VOIDHOOK free_str); /* free structure storage */
2459 Here is how the storage-management hooks are used:
2462 This function is called by set_field_type(). It gets one
2463 argument, a va_list of the type-specific arguments passed to
2464 set_field_type(). It is expected to return a pile pointer to a
2465 data structure that encapsulates those arguments.
2468 This function is called by form library functions that allocate
2469 new field instances. It is expected to take a pile pointer,
2470 copy the pile to allocated storage, and return the address of
2474 This function is called by field- and type-deallocation
2475 routines in the library. It takes a pile pointer argument, and
2476 is expected to free the storage of that pile.
2478 The make_str and copy_str functions may return NULL to signal
2479 allocation failure. The library routines will that call them will
2480 return error indication when this happens. Thus, your validation
2481 functions should never see a NULL file pointer and need not check
2484 Order Functions For Custom Types
2486 Some custom field types are simply ordered in the same well-defined
2487 way that TYPE_ENUM is. For such types, it is possible to define
2488 successor and predecessor functions to support the REQ_NEXT_CHOICE and
2489 REQ_PREV_CHOICE requests. Here is how:
2490 typedef int (*INTHOOK)(); /* pointer to function returning int */
2492 int set_fieldtype_arg(FIELDTYPE *type, /* type to alter */
2493 INTHOOK succ, /* get successor value */
2494 INTHOOK pred); /* get predecessor value */
2496 The successor and predecessor arguments will each be passed two
2497 arguments; a field pointer, and a pile pointer (as for the validation
2498 functions). They are expected to use the function field_buffer() to
2499 read the current value, and set_field_buffer() on buffer 0 to set the
2500 next or previous value. Either hook may return TRUE to indicate
2501 success (a legal next or previous value was set) or FALSE to indicate
2506 The interface for defining custom types is complicated and tricky.
2507 Rather than attempting to create a custom type entirely from scratch,
2508 you should start by studying the library source code for whichever of
2509 the pre-defined types seems to be closest to what you want.
2511 Use that code as a model, and evolve it towards what you really want.
2512 You will avoid many problems and annoyances that way. The code in the
2513 ncurses library has been specifically exempted from the package
2514 copyright to support this.
2516 If your custom type defines order functions, have do something
2517 intuitive with a blank field. A useful convention is to make the
2518 successor of a blank field the types minimum value, and its
2519 predecessor the maximum.