1 Writing Programs with NCURSES
3 by Eric S. Raymond and Zeyd M. Ben-Halim
4 updates since release 1.9.9e by Thomas Dickey
9 + A Brief History of Curses
10 + Scope of This Document
13 + An Overview of Curses
14 o Compiling Programs using Curses
16 o Standard Windows and Function Naming Conventions
22 o Using Forms Characters
23 o Character Attributes and Color
26 + Function Descriptions
27 o Initialization and Wrapup
28 o Causing Output to the Terminal
29 o Low-Level Capability Access
31 + Hints, Tips, and Tricks
32 o Some Notes of Caution
33 o Temporarily Leaving ncurses Mode
34 o Using ncurses under xterm
35 o Handling Multiple Terminal Screens
36 o Testing for Terminal Capabilities
38 o Special Features of ncurses
39 + Compatibility with Older Versions
40 o Refresh of Overlapping Windows
42 + XSI Curses Conformance
44 + Compiling With the Panels Library
46 + Panels, Input, and the Standard Screen
48 + Miscellaneous Other Facilities
50 + Compiling with the menu Library
55 + Processing Menu Input
56 + Miscellaneous Other Features
58 + Compiling with the forms Library
60 + Creating and Freeing Fields and Forms
61 + Fetching and Changing Field Attributes
62 o Fetching Size and Location Data
63 o Changing the Field Location
64 o The Justification Attribute
65 o Field Display Attributes
69 + Variable-Sized Fields
77 + Direct Field Buffer Manipulation
79 + Control of Form Display
80 + Input Processing in the Forms Driver
81 o Page Navigation Requests
82 o Inter-Field Navigation Requests
83 o Intra-Field Navigation Requests
85 o Field Editing Requests
87 o Application Commands
89 + Field Change Commands
91 + Custom Validation Types
94 o Validation Function Arguments
95 o Order Functions For Custom Types
97 _________________________________________________________________
101 This document is an introduction to programming with curses. It is not
102 an exhaustive reference for the curses Application Programming
103 Interface (API); that role is filled by the curses manual pages.
104 Rather, it is intended to help C programmers ease into using the
107 This document is aimed at C applications programmers not yet
108 specifically familiar with ncurses. If you are already an experienced
109 curses programmer, you should nevertheless read the sections on Mouse
110 Interfacing, Debugging, Compatibility with Older Versions, and Hints,
111 Tips, and Tricks. These will bring you up to speed on the special
112 features and quirks of the ncurses implementation. If you are not so
113 experienced, keep reading.
115 The curses package is a subroutine library for terminal-independent
116 screen-painting and input-event handling which presents a high level
117 screen model to the programmer, hiding differences between terminal
118 types and doing automatic optimization of output to change one screen
119 full of text into another. Curses uses terminfo, which is a database
120 format that can describe the capabilities of thousands of different
123 The curses API may seem something of an archaism on UNIX desktops
124 increasingly dominated by X, Motif, and Tcl/Tk. Nevertheless, UNIX
125 still supports tty lines and X supports xterm(1); the curses API has
126 the advantage of (a) back-portability to character-cell terminals, and
127 (b) simplicity. For an application that does not require bit-mapped
128 graphics and multiple fonts, an interface implementation using curses
129 will typically be a great deal simpler and less expensive than one
132 A Brief History of Curses
134 Historically, the first ancestor of curses was the routines written to
135 provide screen-handling for the vi editor; these used the termcap
136 database facility (both released in 3BSD) for describing terminal
137 capabilities. These routines were abstracted into a documented library
138 and first released with the early BSD UNIX versions. All of this work
139 was done by students at the University of California (Berkeley
140 campus). The curses library was first published in 4.0BSD, a year
141 after 3BSD (i.e., late 1980).
143 After graduation, one of those students went to work at AT&T Bell
144 Labs, and made an improved termcap library called terminfo (i.e.,
145 "libterm"), and adapted the curses library to use this. That was
146 subsequently released in System V Release 2 (early 1984). Thereafter,
147 other developers added to the curses and terminfo libraries. For
148 instance, a student at Cornell University wrote an improved terminfo
149 library as well as a tool (tic) to compile the terminal descriptions.
150 As a general rule, AT&T did not identify the developers in the
151 source-code or documentation; the tic and infocmp programs are the
154 System V Release 3 (System III UNIX) from Bell Labs featured a
155 rewritten and much-improved curses library, along with the tic program
158 To recap, terminfo is based on Berkeley's termcap database, but
159 contains a number of improvements and extensions. Parameterized
160 capabilities strings were introduced, making it possible to describe
161 multiple video attributes, and colors and to handle far more unusual
162 terminals than possible with termcap. In the later AT&T System V
163 releases, curses evolved to use more facilities and offer more
164 capabilities, going far beyond BSD curses in power and flexibility.
166 Scope of This Document
168 This document describes ncurses, a free implementation of the System V
169 curses API with some clearly marked extensions. It includes the
170 following System V curses features:
171 * Support for multiple screen highlights (BSD curses could only
172 handle one "standout" highlight, usually reverse-video).
173 * Support for line- and box-drawing using forms characters.
174 * Recognition of function keys on input.
176 * Support for pads (windows of larger than screen size on which the
177 screen or a subwindow defines a viewport).
179 Also, this package makes use of the insert and delete line and
180 character features of terminals so equipped, and determines how to
181 optimally use these features with no help from the programmer. It
182 allows arbitrary combinations of video attributes to be displayed,
183 even on terminals that leave "magic cookies" on the screen to mark
184 changes in attributes.
186 The ncurses package can also capture and use event reports from a
187 mouse in some environments (notably, xterm under the X window system).
188 This document includes tips for using the mouse.
190 The ncurses package was originated by Pavel Curtis. The original
191 maintainer of this package is Zeyd Ben-Halim <zmbenhal@netcom.com>.
192 Eric S. Raymond <esr@snark.thyrsus.com> wrote many of the new features
193 in versions after 1.8.1 and wrote most of this introduction. Juergen
194 Pfeifer wrote all of the menu and forms code as well as the Ada95
195 binding. Ongoing work is being done by Thomas Dickey (maintainer).
196 Contact the current maintainers at bug-ncurses@gnu.org.
198 This document also describes the panels extension library, similarly
199 modeled on the SVr4 panels facility. This library allows you to
200 associate backing store with each of a stack or deck of overlapping
201 windows, and provides operations for moving windows around in the
202 stack that change their visibility in the natural way (handling window
205 Finally, this document describes in detail the menus and forms
206 extension libraries, also cloned from System V, which support easy
207 construction and sequences of menus and fill-in forms.
211 In this document, the following terminology is used with reasonable
215 A data structure describing a sub-rectangle of the screen
216 (possibly the entire screen). You can write to a window as
217 though it were a miniature screen, scrolling independently of
218 other windows on the physical screen.
221 A subset of windows which are as large as the terminal screen,
222 i.e., they start at the upper left hand corner and encompass
223 the lower right hand corner. One of these, stdscr, is
224 automatically provided for the programmer.
227 The package's idea of what the terminal display currently looks
228 like, i.e., what the user sees now. This is a special screen.
232 An Overview of Curses
234 Compiling Programs using Curses
236 In order to use the library, it is necessary to have certain types and
237 variables defined. Therefore, the programmer must have a line:
240 at the top of the program source. The screen package uses the Standard
241 I/O library, so <curses.h> includes <stdio.h>. <curses.h> also
242 includes <termios.h>, <termio.h>, or <sgtty.h> depending on your
243 system. It is redundant (but harmless) for the programmer to do these
244 includes, too. In linking with curses you need to have -lncurses in
245 your LDFLAGS or on the command line. There is no need for any other
250 In order to update the screen optimally, it is necessary for the
251 routines to know what the screen currently looks like and what the
252 programmer wants it to look like next. For this purpose, a data type
253 (structure) named WINDOW is defined which describes a window image to
254 the routines, including its starting position on the screen (the (y,
255 x) coordinates of the upper left hand corner) and its size. One of
256 these (called curscr, for current screen) is a screen image of what
257 the terminal currently looks like. Another screen (called stdscr, for
258 standard screen) is provided by default to make changes on.
260 A window is a purely internal representation. It is used to build and
261 store a potential image of a portion of the terminal. It does not bear
262 any necessary relation to what is really on the terminal screen; it is
263 more like a scratchpad or write buffer.
265 To make the section of physical screen corresponding to a window
266 reflect the contents of the window structure, the routine refresh()
267 (or wrefresh() if the window is not stdscr) is called.
269 A given physical screen section may be within the scope of any number
270 of overlapping windows. Also, changes can be made to windows in any
271 order, without regard to motion efficiency. Then, at will, the
272 programmer can effectively say "make it look like this," and let the
273 package implementation determine the most efficient way to repaint the
276 Standard Windows and Function Naming Conventions
278 As hinted above, the routines can use several windows, but two are
279 automatically given: curscr, which knows what the terminal looks like,
280 and stdscr, which is what the programmer wants the terminal to look
281 like next. The user should never actually access curscr directly.
282 Changes should be made to through the API, and then the routine
283 refresh() (or wrefresh()) called.
285 Many functions are defined to use stdscr as a default screen. For
286 example, to add a character to stdscr, one calls addch() with the
287 desired character as argument. To write to a different window. use the
288 routine waddch() (for window-specific addch()) is provided. This
289 convention of prepending function names with a "w" when they are to be
290 applied to specific windows is consistent. The only routines which do
291 not follow it are those for which a window must always be specified.
293 In order to move the current (y, x) coordinates from one point to
294 another, the routines move() and wmove() are provided. However, it is
295 often desirable to first move and then perform some I/O operation. In
296 order to avoid clumsiness, most I/O routines can be preceded by the
297 prefix "mv" and the desired (y, x) coordinates prepended to the
298 arguments to the function. For example, the calls
310 mvwaddch(win, y, x, ch);
312 Note that the window description pointer (win) comes before the added
313 (y, x) coordinates. If a function requires a window pointer, it is
314 always the first parameter passed.
318 The curses library sets some variables describing the terminal
320 type name description
321 ------------------------------------------------------------------
322 int LINES number of lines on the terminal
323 int COLS number of columns on the terminal
325 The curses.h also introduces some #define constants and types of
329 boolean type, actually a "char" (e.g., bool doneit;)
332 boolean "true" flag (1).
335 boolean "false" flag (0).
338 error flag returned by routines on a failure (-1).
341 error flag returned by routines when things go right.
345 Now we describe how to actually use the screen package. In it, we
346 assume all updating, reading, etc. is applied to stdscr. These
347 instructions will work on any window, providing you change the
348 function names and parameters as mentioned above.
350 Here is a sample program to motivate the discussion:
355 static void finish(int sig);
358 main(int argc, char *argv[])
362 /* initialize your non-curses data structures here */
364 (void) signal(SIGINT, finish); /* arrange interrupts to terminate */
366 (void) initscr(); /* initialize the curses library */
367 keypad(stdscr, TRUE); /* enable keyboard mapping */
368 (void) nonl(); /* tell curses not to do NL->CR/NL on output */
369 (void) cbreak(); /* take input chars one at a time, no wait for \n */
370 (void) echo(); /* echo input - in color */
377 * Simple color assignment, often all we need. Color pair 0 cannot
378 * be redefined. This example uses the same value for the color
379 * pair as for the foreground color, though of course that is not
382 init_pair(1, COLOR_RED, COLOR_BLACK);
383 init_pair(2, COLOR_GREEN, COLOR_BLACK);
384 init_pair(3, COLOR_YELLOW, COLOR_BLACK);
385 init_pair(4, COLOR_BLUE, COLOR_BLACK);
386 init_pair(5, COLOR_CYAN, COLOR_BLACK);
387 init_pair(6, COLOR_MAGENTA, COLOR_BLACK);
388 init_pair(7, COLOR_WHITE, COLOR_BLACK);
393 int c = getch(); /* refresh, accept single keystroke of input */
394 attrset(COLOR_PAIR(num % 8));
397 /* process the command keystroke */
400 finish(0); /* we are done */
403 static void finish(int sig)
407 /* do your non-curses wrapup here */
414 In order to use the screen package, the routines must know about
415 terminal characteristics, and the space for curscr and stdscr must be
416 allocated. These function initscr() does both these things. Since it
417 must allocate space for the windows, it can overflow memory when
418 attempting to do so. On the rare occasions this happens, initscr()
419 will terminate the program with an error message. initscr() must
420 always be called before any of the routines which affect windows are
421 used. If it is not, the program will core dump as soon as either
422 curscr or stdscr are referenced. However, it is usually best to wait
423 to call it until after you are sure you will need it, like after
424 checking for startup errors. Terminal status changing routines like
425 nl() and cbreak() should be called after initscr().
427 Once the screen windows have been allocated, you can set them up for
428 your program. If you want to, say, allow a screen to scroll, use
429 scrollok(). If you want the cursor to be left in place after the last
430 change, use leaveok(). If this is not done, refresh() will move the
431 cursor to the window's current (y, x) coordinates after updating it.
433 You can create new windows of your own using the functions newwin(),
434 derwin(), and subwin(). The routine delwin() will allow you to get rid
435 of old windows. All the options described above can be applied to any
440 Now that we have set things up, we will want to actually update the
441 terminal. The basic functions used to change what will go on a window
442 are addch() and move(). addch() adds a character at the current (y, x)
443 coordinates. move() changes the current (y, x) coordinates to whatever
444 you want them to be. It returns ERR if you try to move off the window.
445 As mentioned above, you can combine the two into mvaddch() to do both
448 The other output functions, such as addstr() and printw(), all call
449 addch() to add characters to the window.
451 After you have put on the window what you want there, when you want
452 the portion of the terminal covered by the window to be made to look
453 like it, you must call refresh(). In order to optimize finding
454 changes, refresh() assumes that any part of the window not changed
455 since the last refresh() of that window has not been changed on the
456 terminal, i.e., that you have not refreshed a portion of the terminal
457 with an overlapping window. If this is not the case, the routine
458 touchwin() is provided to make it look like the entire window has been
459 changed, thus making refresh() check the whole subsection of the
460 terminal for changes.
462 If you call wrefresh() with curscr as its argument, it will make the
463 screen look like curscr thinks it looks like. This is useful for
464 implementing a command which would redraw the screen in case it get
469 The complementary function to addch() is getch() which, if echo is
470 set, will call addch() to echo the character. Since the screen package
471 needs to know what is on the terminal at all times, if characters are
472 to be echoed, the tty must be in raw or cbreak mode. Since initially
473 the terminal has echoing enabled and is in ordinary "cooked" mode, one
474 or the other has to changed before calling getch(); otherwise, the
475 program's output will be unpredictable.
477 When you need to accept line-oriented input in a window, the functions
478 wgetstr() and friends are available. There is even a wscanw() function
479 that can do scanf()(3)-style multi-field parsing on window input.
480 These pseudo-line-oriented functions turn on echoing while they
483 The example code above uses the call keypad(stdscr, TRUE) to enable
484 support for function-key mapping. With this feature, the getch() code
485 watches the input stream for character sequences that correspond to
486 arrow and function keys. These sequences are returned as
487 pseudo-character values. The #define values returned are listed in the
488 curses.h The mapping from sequences to #define values is determined by
489 key_ capabilities in the terminal's terminfo entry.
491 Using Forms Characters
493 The addch() function (and some others, including box() and border())
494 can accept some pseudo-character arguments which are specially defined
495 by ncurses. These are #define values set up in the curses.h header;
496 see there for a complete list (look for the prefix ACS_).
498 The most useful of the ACS defines are the forms-drawing characters.
499 You can use these to draw boxes and simple graphs on the screen. If
500 the terminal does not have such characters, curses.h will map them to
501 a recognizable (though ugly) set of ASCII defaults.
503 Character Attributes and Color
505 The ncurses package supports screen highlights including standout,
506 reverse-video, underline, and blink. It also supports color, which is
507 treated as another kind of highlight.
509 Highlights are encoded, internally, as high bits of the
510 pseudo-character type (chtype) that curses.h uses to represent the
511 contents of a screen cell. See the curses.h header file for a complete
512 list of highlight mask values (look for the prefix A_).
514 There are two ways to make highlights. One is to logical-or the value
515 of the highlights you want into the character argument of an addch()
516 call, or any other output call that takes a chtype argument.
518 The other is to set the current-highlight value. This is logical-ORed
519 with any highlight you specify the first way. You do this with the
520 functions attron(), attroff(), and attrset(); see the manual pages for
521 details. Color is a special kind of highlight. The package actually
522 thinks in terms of color pairs, combinations of foreground and
523 background colors. The sample code above sets up eight color pairs,
524 all of the guaranteed-available colors on black. Note that each color
525 pair is, in effect, given the name of its foreground color. Any other
526 range of eight non-conflicting values could have been used as the
527 first arguments of the init_pair() values.
529 Once you have done an init_pair() that creates color-pair N, you can
530 use COLOR_PAIR(N) as a highlight that invokes that particular color
531 combination. Note that COLOR_PAIR(N), for constant N, is itself a
532 compile-time constant and can be used in initializers.
536 The ncurses library also provides a mouse interface.
538 NOTE: this facility is specific to ncurses, it is not part of
539 either the XSI Curses standard, nor of System V Release 4, nor BSD
540 curses. System V Release 4 curses contains code with similar
541 interface definitions, however it is not documented. Other than by
542 disassembling the library, we have no way to determine exactly how
543 that mouse code works. Thus, we recommend that you wrap
544 mouse-related code in an #ifdef using the feature macro
545 NCURSES_MOUSE_VERSION so it will not be compiled and linked on
548 Presently, mouse event reporting works in the following environments:
549 * xterm and similar programs such as rxvt.
550 * Linux console, when configured with gpm(1), Alessandro Rubini's
552 * FreeBSD sysmouse (console)
555 The mouse interface is very simple. To activate it, you use the
556 function mousemask(), passing it as first argument a bit-mask that
557 specifies what kinds of events you want your program to be able to
558 see. It will return the bit-mask of events that actually become
559 visible, which may differ from the argument if the mouse device is not
560 capable of reporting some of the event types you specify.
562 Once the mouse is active, your application's command loop should watch
563 for a return value of KEY_MOUSE from wgetch(). When you see this, a
564 mouse event report has been queued. To pick it off the queue, use the
565 function getmouse() (you must do this before the next wgetch(),
566 otherwise another mouse event might come in and make the first one
569 Each call to getmouse() fills a structure (the address of which you
570 will pass it) with mouse event data. The event data includes
571 zero-origin, screen-relative character-cell coordinates of the mouse
572 pointer. It also includes an event mask. Bits in this mask will be
573 set, corresponding to the event type being reported.
575 The mouse structure contains two additional fields which may be
576 significant in the future as ncurses interfaces to new kinds of
577 pointing device. In addition to x and y coordinates, there is a slot
578 for a z coordinate; this might be useful with touch-screens that can
579 return a pressure or duration parameter. There is also a device ID
580 field, which could be used to distinguish between multiple pointing
583 The class of visible events may be changed at any time via
584 mousemask(). Events that can be reported include presses, releases,
585 single-, double- and triple-clicks (you can set the maximum
586 button-down time for clicks). If you do not make clicks visible, they
587 will be reported as press-release pairs. In some environments, the
588 event mask may include bits reporting the state of shift, alt, and
589 ctrl keys on the keyboard during the event.
591 A function to check whether a mouse event fell within a given window
592 is also supplied. You can use this to see whether a given window
593 should consider a mouse event relevant to it.
595 Because mouse event reporting will not be available in all
596 environments, it would be unwise to build ncurses applications that
597 require the use of a mouse. Rather, you should use the mouse as a
598 shortcut for point-and-shoot commands your application would normally
599 accept from the keyboard. Two of the test games in the ncurses
600 distribution (bs and knight) contain code that illustrates how this
603 See the manual page curs_mouse(3X) for full details of the
604 mouse-interface functions.
608 In order to clean up after the ncurses routines, the routine endwin()
609 is provided. It restores tty modes to what they were when initscr()
610 was first called, and moves the cursor down to the lower-left corner.
611 Thus, anytime after the call to initscr, endwin() should be called
614 Function Descriptions
616 We describe the detailed behavior of some important curses functions
617 here, as a supplement to the manual page descriptions.
619 Initialization and Wrapup
622 The first function called should almost always be initscr().
623 This will determine the terminal type and initialize curses
624 data structures. initscr() also arranges that the first call to
625 refresh() will clear the screen. If an error occurs a message
626 is written to standard error and the program exits. Otherwise
627 it returns a pointer to stdscr. A few functions may be called
628 before initscr (slk_init(), filter(), ripoffline(), use_env(),
629 and, if you are using multiple terminals, newterm().)
632 Your program should always call endwin() before exiting or
633 shelling out of the program. This function will restore tty
634 modes, move the cursor to the lower left corner of the screen,
635 reset the terminal into the proper non-visual mode. Calling
636 refresh() or doupdate() after a temporary escape from the
637 program will restore the ncurses screen from before the escape.
639 newterm(type, ofp, ifp)
640 A program which outputs to more than one terminal should use
641 newterm() instead of initscr(). newterm() should be called once
642 for each terminal. It returns a variable of type SCREEN * which
643 should be saved as a reference to that terminal. (NOTE: a
644 SCREEN variable is not a screen in the sense we are describing
645 in this introduction, but a collection of parameters used to
646 assist in optimizing the display.) The arguments are the type
647 of the terminal (a string) and FILE pointers for the output and
648 input of the terminal. If type is NULL then the environment
649 variable $TERM is used. endwin() should called once at wrapup
650 time for each terminal opened using this function.
653 This function is used to switch to a different terminal
654 previously opened by newterm(). The screen reference for the
655 new terminal is passed as the parameter. The previous terminal
656 is returned by the function. All other calls affect only the
660 The inverse of newterm(); deallocates the data structures
661 associated with a given SCREEN reference.
663 Causing Output to the Terminal
665 refresh() and wrefresh(win)
666 These functions must be called to actually get any output on
667 the terminal, as other routines merely manipulate data
668 structures. wrefresh() copies the named window to the physical
669 terminal screen, taking into account what is already there in
670 order to do optimizations. refresh() does a refresh of stdscr.
671 Unless leaveok() has been enabled, the physical cursor of the
672 terminal is left at the location of the window's cursor.
674 doupdate() and wnoutrefresh(win)
675 These two functions allow multiple updates with more efficiency
676 than wrefresh. To use them, it is important to understand how
677 curses works. In addition to all the window structures, curses
678 keeps two data structures representing the terminal screen: a
679 physical screen, describing what is actually on the screen, and
680 a virtual screen, describing what the programmer wants to have
681 on the screen. wrefresh works by first copying the named window
682 to the virtual screen (wnoutrefresh()), and then calling the
683 routine to update the screen (doupdate()). If the programmer
684 wishes to output several windows at once, a series of calls to
685 wrefresh will result in alternating calls to wnoutrefresh() and
686 doupdate(), causing several bursts of output to the screen. By
687 calling wnoutrefresh() for each window, it is then possible to
688 call doupdate() once, resulting in only one burst of output,
689 with fewer total characters transmitted (this also avoids a
690 visually annoying flicker at each update).
692 Low-Level Capability Access
694 setupterm(term, filenum, errret)
695 This routine is called to initialize a terminal's description,
696 without setting up the curses screen structures or changing the
697 tty-driver mode bits. term is the character string representing
698 the name of the terminal being used. filenum is the UNIX file
699 descriptor of the terminal to be used for output. errret is a
700 pointer to an integer, in which a success or failure indication
701 is returned. The values returned can be 1 (all is well), 0 (no
702 such terminal), or -1 (some problem locating the terminfo
705 The value of term can be given as NULL, which will cause the
706 value of TERM in the environment to be used. The errret pointer
707 can also be given as NULL, meaning no error code is wanted. If
708 errret is defaulted, and something goes wrong, setupterm() will
709 print an appropriate error message and exit, rather than
710 returning. Thus, a simple program can call setupterm(0, 1, 0)
711 and not worry about initialization errors.
713 After the call to setupterm(), the global variable cur_term is
714 set to point to the current structure of terminal capabilities.
715 By calling setupterm() for each terminal, and saving and
716 restoring cur_term, it is possible for a program to use two or
717 more terminals at once. Setupterm() also stores the names
718 section of the terminal description in the global character
719 array ttytype[]. Subsequent calls to setupterm() will overwrite
720 this array, so you will have to save it yourself if need be.
724 NOTE: These functions are not part of the standard curses API!
727 This function can be used to explicitly set a trace level. If
728 the trace level is nonzero, execution of your program will
729 generate a file called "trace" in the current working directory
730 containing a report on the library's actions. Higher trace
731 levels enable more detailed (and verbose) reporting -- see
732 comments attached to TRACE_ defines in the curses.h file for
733 details. (It is also possible to set a trace level by assigning
734 a trace level value to the environment variable NCURSES_TRACE).
737 This function can be used to output your own debugging
738 information. It is only available only if you link with
739 -lncurses_g. It can be used the same way as printf(), only it
740 outputs a newline after the end of arguments. The output goes
741 to a file called trace in the current directory.
743 Trace logs can be difficult to interpret due to the sheer volume of
744 data dumped in them. There is a script called tracemunch included with
745 the ncurses distribution that can alleviate this problem somewhat; it
746 compacts long sequences of similar operations into more succinct
747 single-line pseudo-operations. These pseudo-ops can be distinguished
748 by the fact that they are named in capital letters.
750 Hints, Tips, and Tricks
752 The ncurses manual pages are a complete reference for this library. In
753 the remainder of this document, we discuss various useful methods that
754 may not be obvious from the manual page descriptions.
756 Some Notes of Caution
758 If you find yourself thinking you need to use noraw() or nocbreak(),
759 think again and move carefully. It is probably better design to use
760 getstr() or one of its relatives to simulate cooked mode. The noraw()
761 and nocbreak() functions try to restore cooked mode, but they may end
762 up clobbering some control bits set before you started your
763 application. Also, they have always been poorly documented, and are
764 likely to hurt your application's usability with other curses
767 Bear in mind that refresh() is a synonym for wrefresh(stdscr). Do not
768 try to mix use of stdscr with use of windows declared by newwin(); a
769 refresh() call will blow them off the screen. The right way to handle
770 this is to use subwin(), or not touch stdscr at all and tile your
771 screen with declared windows which you then wnoutrefresh() somewhere
772 in your program event loop, with a single doupdate() call to trigger
775 You are much less likely to run into problems if you design your
776 screen layouts to use tiled rather than overlapping windows.
777 Historically, curses support for overlapping windows has been weak,
778 fragile, and poorly documented. The ncurses library is not yet an
779 exception to this rule.
781 There is a panels library included in the ncurses distribution that
782 does a pretty good job of strengthening the overlapping-windows
785 Try to avoid using the global variables LINES and COLS. Use getmaxyx()
786 on the stdscr context instead. Reason: your code may be ported to run
787 in an environment with window resizes, in which case several screens
788 could be open with different sizes.
790 Temporarily Leaving NCURSES Mode
792 Sometimes you will want to write a program that spends most of its
793 time in screen mode, but occasionally returns to ordinary "cooked"
794 mode. A common reason for this is to support shell-out. This behavior
795 is simple to arrange in ncurses.
797 To leave ncurses mode, call endwin() as you would if you were
798 intending to terminate the program. This will take the screen back to
799 cooked mode; you can do your shell-out. When you want to return to
800 ncurses mode, simply call refresh() or doupdate(). This will repaint
803 There is a boolean function, isendwin(), which code can use to test
804 whether ncurses screen mode is active. It returns TRUE in the interval
805 between an endwin() call and the following refresh(), FALSE otherwise.
807 Here is some sample code for shellout:
808 addstr("Shelling out...");
809 def_prog_mode(); /* save current tty modes */
810 endwin(); /* restore original tty modes */
811 system("sh"); /* run shell */
812 addstr("returned.\n"); /* prepare return message */
813 refresh(); /* restore save modes, repaint screen */
815 Using NCURSES under XTERM
817 A resize operation in X sends SIGWINCH to the application running
818 under xterm. The easiest way to handle SIGWINCH is to do an endwin,
819 followed by an refresh and a screen repaint you code yourself. The
820 refresh will pick up the new screen size from the xterm's environment.
822 That is the standard way, of course (it even works with some vendor's
823 curses implementations). Its drawback is that it clears the screen to
824 reinitialize the display, and does not resize subwindows which must be
825 shrunk. Ncurses provides an extension which works better, the
826 resizeterm function. That function ensures that all windows are
827 limited to the new screen dimensions, and pads stdscr with blanks if
828 the screen is larger.
830 The ncurses library provides a SIGWINCH signal handler, which pushes a
831 KEY_RESIZE via the wgetch() calls. When ncurses returns that code, it
832 calls resizeterm to update the size of the standard screen's window,
833 repainting that (filling with blanks or truncating as needed). It also
834 resizes other windows, but its effect may be less satisfactory because
835 it cannot know how you want the screen re-painted. You will usually
836 have to write special-purpose code to handle KEY_RESIZE yourself.
838 Handling Multiple Terminal Screens
840 The initscr() function actually calls a function named newterm() to do
841 most of its work. If you are writing a program that opens multiple
842 terminals, use newterm() directly.
844 For each call, you will have to specify a terminal type and a pair of
845 file pointers; each call will return a screen reference, and stdscr
846 will be set to the last one allocated. You will switch between screens
847 with the set_term call. Note that you will also have to call
848 def_shell_mode and def_prog_mode on each tty yourself.
850 Testing for Terminal Capabilities
852 Sometimes you may want to write programs that test for the presence of
853 various capabilities before deciding whether to go into ncurses mode.
854 An easy way to do this is to call setupterm(), then use the functions
855 tigetflag(), tigetnum(), and tigetstr() to do your testing.
857 A particularly useful case of this often comes up when you want to
858 test whether a given terminal type should be treated as "smart"
859 (cursor-addressable) or "stupid". The right way to test this is to see
860 if the return value of tigetstr("cup") is non-NULL. Alternatively, you
861 can include the term.h file and test the value of the macro
866 Use the addchstr() family of functions for fast screen-painting of
867 text when you know the text does not contain any control characters.
868 Try to make attribute changes infrequent on your screens. Do not use
869 the immedok() option!
871 Special Features of NCURSES
873 The wresize() function allows you to resize a window in place. The
874 associated resizeterm() function simplifies the construction of
875 SIGWINCH handlers, for resizing all windows.
877 The define_key() function allows you to define at runtime function-key
878 control sequences which are not in the terminal description. The
879 keyok() function allows you to temporarily enable or disable
880 interpretation of any function-key control sequence.
882 The use_default_colors() function allows you to construct applications
883 which can use the terminal's default foreground and background colors
884 as an additional "default" color. Several terminal emulators support
885 this feature, which is based on ISO 6429.
887 Ncurses supports up 16 colors, unlike SVr4 curses which defines only
888 8. While most terminals which provide color allow only 8 colors, about
889 a quarter (including XFree86 xterm) support 16 colors.
891 Compatibility with Older Versions
893 Despite our best efforts, there are some differences between ncurses
894 and the (undocumented!) behavior of older curses implementations.
895 These arise from ambiguities or omissions in the documentation of the
898 Refresh of Overlapping Windows
900 If you define two windows A and B that overlap, and then alternately
901 scribble on and refresh them, the changes made to the overlapping
902 region under historic curses versions were often not documented
905 To understand why this is a problem, remember that screen updates are
906 calculated between two representations of the entire display. The
907 documentation says that when you refresh a window, it is first copied
908 to the virtual screen, and then changes are calculated to update the
909 physical screen (and applied to the terminal). But "copied to" is not
910 very specific, and subtle differences in how copying works can produce
911 different behaviors in the case where two overlapping windows are each
912 being refreshed at unpredictable intervals.
914 What happens to the overlapping region depends on what wnoutrefresh()
915 does with its argument -- what portions of the argument window it
916 copies to the virtual screen. Some implementations do "change copy",
917 copying down only locations in the window that have changed (or been
918 marked changed with wtouchln() and friends). Some implementations do
919 "entire copy", copying all window locations to the virtual screen
920 whether or not they have changed.
922 The ncurses library itself has not always been consistent on this
923 score. Due to a bug, versions 1.8.7 to 1.9.8a did entire copy.
924 Versions 1.8.6 and older, and versions 1.9.9 and newer, do change
927 For most commercial curses implementations, it is not documented and
928 not known for sure (at least not to the ncurses maintainers) whether
929 they do change copy or entire copy. We know that System V release 3
930 curses has logic in it that looks like an attempt to do change copy,
931 but the surrounding logic and data representations are sufficiently
932 complex, and our knowledge sufficiently indirect, that it is hard to
933 know whether this is reliable. It is not clear what the SVr4
934 documentation and XSI standard intend. The XSI Curses standard barely
935 mentions wnoutrefresh(); the SVr4 documents seem to be describing
936 entire-copy, but it is possible with some effort and straining to read
939 It might therefore be unwise to rely on either behavior in programs
940 that might have to be linked with other curses implementations.
941 Instead, you can do an explicit touchwin() before the wnoutrefresh()
942 call to guarantee an entire-contents copy anywhere.
944 The really clean way to handle this is to use the panels library. If,
945 when you want a screen update, you do update_panels(), it will do all
946 the necessary wnoutrefresh() calls for whatever panel stacking order
947 you have defined. Then you can do one doupdate() and there will be a
948 single burst of physical I/O that will do all your updates.
952 If you have been using a very old versions of ncurses (1.8.7 or older)
953 you may be surprised by the behavior of the erase functions. In older
954 versions, erased areas of a window were filled with a blank modified
955 by the window's current attribute (as set by wattrset(), wattron(),
956 wattroff() and friends).
958 In newer versions, this is not so. Instead, the attribute of erased
959 blanks is normal unless and until it is modified by the functions
960 bkgdset() or wbkgdset().
962 This change in behavior conforms ncurses to System V Release 4 and the
965 XSI Curses Conformance
967 The ncurses library is intended to be base-level conformant with the
968 XSI Curses standard from X/Open. Many extended-level features (in
969 fact, almost all features not directly concerned with wide characters
970 and internationalization) are also supported.
972 One effect of XSI conformance is the change in behavior described
973 under "Background Erase -- Compatibility with Old Versions".
975 Also, ncurses meets the XSI requirement that every macro entry point
976 have a corresponding function which may be linked (and will be
977 prototype-checked) if the macro definition is disabled with #undef.
981 The ncurses library by itself provides good support for screen
982 displays in which the windows are tiled (non-overlapping). In the more
983 general case that windows may overlap, you have to use a series of
984 wnoutrefresh() calls followed by a doupdate(), and be careful about
985 the order you do the window refreshes in. It has to be bottom-upwards,
986 otherwise parts of windows that should be obscured will show through.
988 When your interface design is such that windows may dive deeper into
989 the visibility stack or pop to the top at runtime, the resulting
990 book-keeping can be tedious and difficult to get right. Hence the
993 The panel library first appeared in AT&T System V. The version
994 documented here is the panel code distributed with ncurses.
996 Compiling With the Panels Library
998 Your panels-using modules must import the panels library declarations
1002 and must be linked explicitly with the panels library using an -lpanel
1003 argument. Note that they must also link the ncurses library with
1004 -lncurses. Many linkers are two-pass and will accept either order, but
1005 it is still good practice to put -lpanel first and -lncurses second.
1009 A panel object is a window that is implicitly treated as part of a
1010 deck including all other panel objects. The deck has an implicit
1011 bottom-to-top visibility order. The panels library includes an update
1012 function (analogous to refresh()) that displays all panels in the deck
1013 in the proper order to resolve overlaps. The standard window, stdscr,
1014 is considered below all panels.
1016 Details on the panels functions are available in the man pages. We
1017 will just hit the highlights here.
1019 You create a panel from a window by calling new_panel() on a window
1020 pointer. It then becomes the top of the deck. The panel's window is
1021 available as the value of panel_window() called with the panel pointer
1024 You can delete a panel (removing it from the deck) with del_panel.
1025 This will not deallocate the associated window; you have to do that
1026 yourself. You can replace a panel's window with a different window by
1027 calling replace_window. The new window may be of different size; the
1028 panel code will re-compute all overlaps. This operation does not
1029 change the panel's position in the deck.
1031 To move a panel's window, use move_panel(). The mvwin() function on
1032 the panel's window is not sufficient because it does not update the
1033 panels library's representation of where the windows are. This
1034 operation leaves the panel's depth, contents, and size unchanged.
1036 Two functions (top_panel(), bottom_panel()) are provided for
1037 rearranging the deck. The first pops its argument window to the top of
1038 the deck; the second sends it to the bottom. Either operation leaves
1039 the panel's screen location, contents, and size unchanged.
1041 The function update_panels() does all the wnoutrefresh() calls needed
1042 to prepare for doupdate() (which you must call yourself, afterwards).
1044 Typically, you will want to call update_panels() and doupdate() just
1045 before accepting command input, once in each cycle of interaction with
1046 the user. If you call update_panels() after each and every panel
1047 write, you will generate a lot of unnecessary refresh activity and
1050 Panels, Input, and the Standard Screen
1052 You should not mix wnoutrefresh() or wrefresh() operations with panels
1053 code; this will work only if the argument window is either in the top
1054 panel or unobscured by any other panels.
1056 The stsdcr window is a special case. It is considered below all
1057 panels. Because changes to panels may obscure parts of stdscr, though,
1058 you should call update_panels() before doupdate() even when you only
1061 Note that wgetch automatically calls wrefresh. Therefore, before
1062 requesting input from a panel window, you need to be sure that the
1063 panel is totally unobscured.
1065 There is presently no way to display changes to one obscured panel
1066 without repainting all panels.
1070 It is possible to remove a panel from the deck temporarily; use
1071 hide_panel for this. Use show_panel() to render it visible again. The
1072 predicate function panel_hidden tests whether or not a panel is
1075 The panel_update code ignores hidden panels. You cannot do top_panel()
1076 or bottom_panel on a hidden panel(). Other panels operations are
1079 Miscellaneous Other Facilities
1081 It is possible to navigate the deck using the functions panel_above()
1082 and panel_below. Handed a panel pointer, they return the panel above
1083 or below that panel. Handed NULL, they return the bottom-most or
1086 Every panel has an associated user pointer, not used by the panel
1087 code, to which you can attach application data. See the man page
1088 documentation of set_panel_userptr() and panel_userptr for details.
1092 A menu is a screen display that assists the user to choose some subset
1093 of a given set of items. The menu library is a curses extension that
1094 supports easy programming of menu hierarchies with a uniform but
1097 The menu library first appeared in AT&T System V. The version
1098 documented here is the menu code distributed with ncurses.
1100 Compiling With the menu Library
1102 Your menu-using modules must import the menu library declarations with
1105 and must be linked explicitly with the menus library using an -lmenu
1106 argument. Note that they must also link the ncurses library with
1107 -lncurses. Many linkers are two-pass and will accept either order, but
1108 it is still good practice to put -lmenu first and -lncurses second.
1112 The menus created by this library consist of collections of items
1113 including a name string part and a description string part. To make
1114 menus, you create groups of these items and connect them with menu
1117 The menu can then by posted, that is written to an associated window.
1118 Actually, each menu has two associated windows; a containing window in
1119 which the programmer can scribble titles or borders, and a subwindow
1120 in which the menu items proper are displayed. If this subwindow is too
1121 small to display all the items, it will be a scrollable viewport on
1122 the collection of items.
1124 A menu may also be unposted (that is, undisplayed), and finally freed
1125 to make the storage associated with it and its items available for
1128 The general flow of control of a menu program looks like this:
1129 1. Initialize curses.
1130 2. Create the menu items, using new_item().
1131 3. Create the menu using new_menu().
1132 4. Post the menu using post_menu().
1133 5. Refresh the screen.
1134 6. Process user requests via an input loop.
1135 7. Unpost the menu using unpost_menu().
1136 8. Free the menu, using free_menu().
1137 9. Free the items using free_item().
1138 10. Terminate curses.
1142 Menus may be multi-valued or (the default) single-valued (see the
1143 manual page menu_opts(3x) to see how to change the default). Both
1144 types always have a current item.
1146 From a single-valued menu you can read the selected value simply by
1147 looking at the current item. From a multi-valued menu, you get the
1148 selected set by looping through the items applying the item_value()
1149 predicate function. Your menu-processing code can use the function
1150 set_item_value() to flag the items in the select set.
1152 Menu items can be made unselectable using set_item_opts() or
1153 item_opts_off() with the O_SELECTABLE argument. This is the only
1154 option so far defined for menus, but it is good practice to code as
1155 though other option bits might be on.
1159 The menu library calculates a minimum display size for your window,
1160 based on the following variables:
1161 * The number and maximum length of the menu items
1162 * Whether the O_ROWMAJOR option is enabled
1163 * Whether display of descriptions is enabled
1164 * Whatever menu format may have been set by the programmer
1165 * The length of the menu mark string used for highlighting selected
1168 The function set_menu_format() allows you to set the maximum size of
1169 the viewport or menu page that will be used to display menu items. You
1170 can retrieve any format associated with a menu with menu_format(). The
1171 default format is rows=16, columns=1.
1173 The actual menu page may be smaller than the format size. This depends
1174 on the item number and size and whether O_ROWMAJOR is on. This option
1175 (on by default) causes menu items to be displayed in a "raster-scan"
1176 pattern, so that if more than one item will fit horizontally the first
1177 couple of items are side-by-side in the top row. The alternative is
1178 column-major display, which tries to put the first several items in
1181 As mentioned above, a menu format not large enough to allow all items
1182 to fit on-screen will result in a menu display that is vertically
1185 You can scroll it with requests to the menu driver, which will be
1186 described in the section on menu input handling.
1188 Each menu has a mark string used to visually tag selected items; see
1189 the menu_mark(3x) manual page for details. The mark string length also
1190 influences the menu page size.
1192 The function scale_menu() returns the minimum display size that the
1193 menu code computes from all these factors. There are other menu
1194 display attributes including a select attribute, an attribute for
1195 selectable items, an attribute for unselectable items, and a pad
1196 character used to separate item name text from description text. These
1197 have reasonable defaults which the library allows you to change (see
1198 the menu_attribs(3x) manual page.
1202 Each menu has, as mentioned previously, a pair of associated windows.
1203 Both these windows are painted when the menu is posted and erased when
1204 the menu is unposted.
1206 The outer or frame window is not otherwise touched by the menu
1207 routines. It exists so the programmer can associate a title, a border,
1208 or perhaps help text with the menu and have it properly refreshed or
1209 erased at post/unpost time. The inner window or subwindow is where the
1210 current menu page is displayed.
1212 By default, both windows are stdscr. You can set them with the
1213 functions in menu_win(3x).
1215 When you call post_menu(), you write the menu to its subwindow. When
1216 you call unpost_menu(), you erase the subwindow, However, neither of
1217 these actually modifies the screen. To do that, call wrefresh() or
1220 Processing Menu Input
1222 The main loop of your menu-processing code should call menu_driver()
1223 repeatedly. The first argument of this routine is a menu pointer; the
1224 second is a menu command code. You should write an input-fetching
1225 routine that maps input characters to menu command codes, and pass its
1226 output to menu_driver(). The menu command codes are fully documented
1229 The simplest group of command codes is REQ_NEXT_ITEM, REQ_PREV_ITEM,
1230 REQ_FIRST_ITEM, REQ_LAST_ITEM, REQ_UP_ITEM, REQ_DOWN_ITEM,
1231 REQ_LEFT_ITEM, REQ_RIGHT_ITEM. These change the currently selected
1232 item. These requests may cause scrolling of the menu page if it only
1233 partially displayed.
1235 There are explicit requests for scrolling which also change the
1236 current item (because the select location does not change, but the
1237 item there does). These are REQ_SCR_DLINE, REQ_SCR_ULINE,
1238 REQ_SCR_DPAGE, and REQ_SCR_UPAGE.
1240 The REQ_TOGGLE_ITEM selects or deselects the current item. It is for
1241 use in multi-valued menus; if you use it with O_ONEVALUE on, you will
1242 get an error return (E_REQUEST_DENIED).
1244 Each menu has an associated pattern buffer. The menu_driver() logic
1245 tries to accumulate printable ASCII characters passed in in that
1246 buffer; when it matches a prefix of an item name, that item (or the
1247 next matching item) is selected. If appending a character yields no
1248 new match, that character is deleted from the pattern buffer, and
1249 menu_driver() returns E_NO_MATCH.
1251 Some requests change the pattern buffer directly: REQ_CLEAR_PATTERN,
1252 REQ_BACK_PATTERN, REQ_NEXT_MATCH, REQ_PREV_MATCH. The latter two are
1253 useful when pattern buffer input matches more than one item in a
1256 Each successful scroll or item navigation request clears the pattern
1257 buffer. It is also possible to set the pattern buffer explicitly with
1260 Finally, menu driver requests above the constant MAX_COMMAND are
1261 considered application-specific commands. The menu_driver() code
1262 ignores them and returns E_UNKNOWN_COMMAND.
1264 Miscellaneous Other Features
1266 Various menu options can affect the processing and visual appearance
1267 and input processing of menus. See menu_opts(3x) for details.
1269 It is possible to change the current item from application code; this
1270 is useful if you want to write your own navigation requests. It is
1271 also possible to explicitly set the top row of the menu display. See
1272 mitem_current(3x). If your application needs to change the menu
1273 subwindow cursor for any reason, pos_menu_cursor() will restore it to
1274 the correct location for continuing menu driver processing.
1276 It is possible to set hooks to be called at menu initialization and
1277 wrapup time, and whenever the selected item changes. See
1280 Each item, and each menu, has an associated user pointer on which you
1281 can hang application data. See mitem_userptr(3x) and menu_userptr(3x).
1285 The form library is a curses extension that supports easy programming
1286 of on-screen forms for data entry and program control.
1288 The form library first appeared in AT&T System V. The version
1289 documented here is the form code distributed with ncurses.
1291 Compiling With the form Library
1293 Your form-using modules must import the form library declarations with
1296 and must be linked explicitly with the forms library using an -lform
1297 argument. Note that they must also link the ncurses library with
1298 -lncurses. Many linkers are two-pass and will accept either order, but
1299 it is still good practice to put -lform first and -lncurses second.
1303 A form is a collection of fields; each field may be either a label
1304 (explanatory text) or a data-entry location. Long forms may be
1305 segmented into pages; each entry to a new page clears the screen.
1307 To make forms, you create groups of fields and connect them with form
1308 frame objects; the form library makes this relatively simple.
1310 Once defined, a form can be posted, that is written to an associated
1311 window. Actually, each form has two associated windows; a containing
1312 window in which the programmer can scribble titles or borders, and a
1313 subwindow in which the form fields proper are displayed.
1315 As the form user fills out the posted form, navigation and editing
1316 keys support movement between fields, editing keys support modifying
1317 field, and plain text adds to or changes data in a current field. The
1318 form library allows you (the forms designer) to bind each navigation
1319 and editing key to any keystroke accepted by curses Fields may have
1320 validation conditions on them, so that they check input data for type
1321 and value. The form library supplies a rich set of pre-defined field
1322 types, and makes it relatively easy to define new ones.
1324 Once its transaction is completed (or aborted), a form may be unposted
1325 (that is, undisplayed), and finally freed to make the storage
1326 associated with it and its items available for re-use.
1328 The general flow of control of a form program looks like this:
1329 1. Initialize curses.
1330 2. Create the form fields, using new_field().
1331 3. Create the form using new_form().
1332 4. Post the form using post_form().
1333 5. Refresh the screen.
1334 6. Process user requests via an input loop.
1335 7. Unpost the form using unpost_form().
1336 8. Free the form, using free_form().
1337 9. Free the fields using free_field().
1338 10. Terminate curses.
1340 Note that this looks much like a menu program; the form library
1341 handles tasks which are in many ways similar, and its interface was
1342 obviously designed to resemble that of the menu library wherever
1345 In forms programs, however, the "process user requests" is somewhat
1346 more complicated than for menus. Besides menu-like navigation
1347 operations, the menu driver loop has to support field editing and data
1350 Creating and Freeing Fields and Forms
1352 The basic function for creating fields is new_field():
1353 FIELD *new_field(int height, int width, /* new field size */
1354 int top, int left, /* upper left corner */
1355 int offscreen, /* number of offscreen rows */
1356 int nbuf); /* number of working buffers */
1358 Menu items always occupy a single row, but forms fields may have
1359 multiple rows. So new_field() requires you to specify a width and
1360 height (the first two arguments, which mist both be greater than
1363 You must also specify the location of the field's upper left corner on
1364 the screen (the third and fourth arguments, which must be zero or
1365 greater). Note that these coordinates are relative to the form
1366 subwindow, which will coincide with stdscr by default but need not be
1367 stdscr if you have done an explicit set_form_win() call.
1369 The fifth argument allows you to specify a number of off-screen rows.
1370 If this is zero, the entire field will always be displayed. If it is
1371 nonzero, the form will be scrollable, with only one screen-full
1372 (initially the top part) displayed at any given time. If you make a
1373 field dynamic and grow it so it will no longer fit on the screen, the
1374 form will become scrollable even if the offscreen argument was
1377 The forms library allocates one working buffer per field; the size of
1378 each buffer is ((height + offscreen)*width + 1, one character for each
1379 position in the field plus a NUL terminator. The sixth argument is the
1380 number of additional data buffers to allocate for the field; your
1381 application can use them for its own purposes.
1382 FIELD *dup_field(FIELD *field, /* field to copy */
1383 int top, int left); /* location of new copy */
1385 The function dup_field() duplicates an existing field at a new
1386 location. Size and buffering information are copied; some attribute
1387 flags and status bits are not (see the form_field_new(3X) for
1389 FIELD *link_field(FIELD *field, /* field to copy */
1390 int top, int left); /* location of new copy */
1392 The function link_field() also duplicates an existing field at a new
1393 location. The difference from dup_field() is that it arranges for the
1394 new field's buffer to be shared with the old one.
1396 Besides the obvious use in making a field editable from two different
1397 form pages, linked fields give you a way to hack in dynamic labels. If
1398 you declare several fields linked to an original, and then make them
1399 inactive, changes from the original will still be propagated to the
1402 As with duplicated fields, linked fields have attribute bits separate
1405 As you might guess, all these field-allocations return NULL if the
1406 field allocation is not possible due to an out-of-memory error or
1407 out-of-bounds arguments.
1409 To connect fields to a form, use
1410 FORM *new_form(FIELD **fields);
1412 This function expects to see a NULL-terminated array of field
1413 pointers. Said fields are connected to a newly-allocated form object;
1414 its address is returned (or else NULL if the allocation fails).
1416 Note that new_field() does not copy the pointer array into private
1417 storage; if you modify the contents of the pointer array during forms
1418 processing, all manner of bizarre things might happen. Also note that
1419 any given field may only be connected to one form.
1421 The functions free_field() and free_form are available to free field
1422 and form objects. It is an error to attempt to free a field connected
1423 to a form, but not vice-versa; thus, you will generally free your form
1426 Fetching and Changing Field Attributes
1428 Each form field has a number of location and size attributes
1429 associated with it. There are other field attributes used to control
1430 display and editing of the field. Some (for example, the O_STATIC bit)
1431 involve sufficient complications to be covered in sections of their
1432 own later on. We cover the functions used to get and set several basic
1435 When a field is created, the attributes not specified by the new_field
1436 function are copied from an invisible system default field. In
1437 attribute-setting and -fetching functions, the argument NULL is taken
1438 to mean this field. Changes to it persist as defaults until your forms
1439 application terminates.
1441 Fetching Size and Location Data
1443 You can retrieve field sizes and locations through:
1444 int field_info(FIELD *field, /* field from which to fetch */
1445 int *height, *int width, /* field size */
1446 int *top, int *left, /* upper left corner */
1447 int *offscreen, /* number of offscreen rows */
1448 int *nbuf); /* number of working buffers */
1450 This function is a sort of inverse of new_field(); instead of setting
1451 size and location attributes of a new field, it fetches them from an
1454 Changing the Field Location
1456 It is possible to move a field's location on the screen:
1457 int move_field(FIELD *field, /* field to alter */
1458 int top, int left); /* new upper-left corner */
1460 You can, of course. query the current location through field_info().
1462 The Justification Attribute
1464 One-line fields may be unjustified, justified right, justified left,
1465 or centered. Here is how you manipulate this attribute:
1466 int set_field_just(FIELD *field, /* field to alter */
1467 int justmode); /* mode to set */
1469 int field_just(FIELD *field); /* fetch mode of field */
1471 The mode values accepted and returned by this functions are
1472 preprocessor macros NO_JUSTIFICATION, JUSTIFY_RIGHT, JUSTIFY_LEFT, or
1475 Field Display Attributes
1477 For each field, you can set a foreground attribute for entered
1478 characters, a background attribute for the entire field, and a pad
1479 character for the unfilled portion of the field. You can also control
1480 pagination of the form.
1482 This group of four field attributes controls the visual appearance of
1483 the field on the screen, without affecting in any way the data in the
1485 int set_field_fore(FIELD *field, /* field to alter */
1486 chtype attr); /* attribute to set */
1488 chtype field_fore(FIELD *field); /* field to query */
1490 int set_field_back(FIELD *field, /* field to alter */
1491 chtype attr); /* attribute to set */
1493 chtype field_back(FIELD *field); /* field to query */
1495 int set_field_pad(FIELD *field, /* field to alter */
1496 int pad); /* pad character to set */
1498 chtype field_pad(FIELD *field);
1500 int set_new_page(FIELD *field, /* field to alter */
1501 int flag); /* TRUE to force new page */
1503 chtype new_page(FIELD *field); /* field to query */
1505 The attributes set and returned by the first four functions are normal
1506 curses(3x) display attribute values (A_STANDOUT, A_BOLD, A_REVERSE
1507 etc). The page bit of a field controls whether it is displayed at the
1508 start of a new form screen.
1512 There is also a large collection of field option bits you can set to
1513 control various aspects of forms processing. You can manipulate them
1514 with these functions:
1515 int set_field_opts(FIELD *field, /* field to alter */
1516 int attr); /* attribute to set */
1518 int field_opts_on(FIELD *field, /* field to alter */
1519 int attr); /* attributes to turn on */
1521 int field_opts_off(FIELD *field, /* field to alter */
1522 int attr); /* attributes to turn off */
1524 int field_opts(FIELD *field); /* field to query */
1526 By default, all options are on. Here are the available option bits:
1529 Controls whether the field is visible on the screen. Can be
1530 used during form processing to hide or pop up fields depending
1531 on the value of parent fields.
1534 Controls whether the field is active during forms processing
1535 (i.e. visited by form navigation keys). Can be used to make
1536 labels or derived fields with buffer values alterable by the
1537 forms application, not the user.
1540 Controls whether data is displayed during field entry. If this
1541 option is turned off on a field, the library will accept and
1542 edit data in that field, but it will not be displayed and the
1543 visible field cursor will not move. You can turn off the
1544 O_PUBLIC bit to define password fields.
1547 Controls whether the field's data can be modified. When this
1548 option is off, all editing requests except REQ_PREV_CHOICE and
1549 REQ_NEXT_CHOICE will fail. Such read-only fields may be useful
1553 Controls word-wrapping in multi-line fields. Normally, when any
1554 character of a (blank-separated) word reaches the end of the
1555 current line, the entire word is wrapped to the next line
1556 (assuming there is one). When this option is off, the word will
1557 be split across the line break.
1560 Controls field blanking. When this option is on, entering a
1561 character at the first field position erases the entire field
1562 (except for the just-entered character).
1565 Controls automatic skip to next field when this one fills.
1566 Normally, when the forms user tries to type more data into a
1567 field than will fit, the editing location jumps to next field.
1568 When this option is off, the user's cursor will hang at the end
1569 of the field. This option is ignored in dynamic fields that
1570 have not reached their size limit.
1573 Controls whether validation is applied to blank fields.
1574 Normally, it is not; the user can leave a field blank without
1575 invoking the usual validation check on exit. If this option is
1576 off on a field, exit from it will invoke a validation check.
1579 Controls whether validation occurs on every exit, or only after
1580 the field is modified. Normally the latter is true. Setting
1581 O_PASSOK may be useful if your field's validation function may
1582 change during forms processing.
1585 Controls whether the field is fixed to its initial dimensions.
1586 If you turn this off, the field becomes dynamic and will
1587 stretch to fit entered data.
1589 A field's options cannot be changed while the field is currently
1590 selected. However, options may be changed on posted fields that are
1593 The option values are bit-masks and can be composed with logical-or in
1598 Every field has a status flag, which is set to FALSE when the field is
1599 created and TRUE when the value in field buffer 0 changes. This flag
1600 can be queried and set directly:
1601 int set_field_status(FIELD *field, /* field to alter */
1602 int status); /* mode to set */
1604 int field_status(FIELD *field); /* fetch mode of field */
1606 Setting this flag under program control can be useful if you use the
1607 same form repeatedly, looking for modified fields each time.
1609 Calling field_status() on a field not currently selected for input
1610 will return a correct value. Calling field_status() on a field that is
1611 currently selected for input may not necessarily give a correct field
1612 status value, because entered data is not necessarily copied to buffer
1613 zero before the exit validation check. To guarantee that the returned
1614 status value reflects reality, call field_status() either (1) in the
1615 field's exit validation check routine, (2) from the field's or form's
1616 initialization or termination hooks, or (3) just after a
1617 REQ_VALIDATION request has been processed by the forms driver.
1621 Each field structure contains one character pointer slot that is not
1622 used by the forms library. It is intended to be used by applications
1623 to store private per-field data. You can manipulate it with:
1624 int set_field_userptr(FIELD *field, /* field to alter */
1625 char *userptr); /* mode to set */
1627 char *field_userptr(FIELD *field); /* fetch mode of field */
1629 (Properly, this user pointer field ought to have (void *) type. The
1630 (char *) type is retained for System V compatibility.)
1632 It is valid to set the user pointer of the default field (with a
1633 set_field_userptr() call passed a NULL field pointer.) When a new
1634 field is created, the default-field user pointer is copied to
1635 initialize the new field's user pointer.
1637 Variable-Sized Fields
1639 Normally, a field is fixed at the size specified for it at creation
1640 time. If, however, you turn off its O_STATIC bit, it becomes dynamic
1641 and will automatically resize itself to accommodate data as it is
1642 entered. If the field has extra buffers associated with it, they will
1643 grow right along with the main input buffer.
1645 A one-line dynamic field will have a fixed height (1) but variable
1646 width, scrolling horizontally to display data within the field area as
1647 originally dimensioned and located. A multi-line dynamic field will
1648 have a fixed width, but variable height (number of rows), scrolling
1649 vertically to display data within the field area as originally
1650 dimensioned and located.
1652 Normally, a dynamic field is allowed to grow without limit. But it is
1653 possible to set an upper limit on the size of a dynamic field. You do
1654 it with this function:
1655 int set_max_field(FIELD *field, /* field to alter (may not be NULL) */
1656 int max_size); /* upper limit on field size */
1658 If the field is one-line, max_size is taken to be a column size limit;
1659 if it is multi-line, it is taken to be a line size limit. To disable
1660 any limit, use an argument of zero. The growth limit can be changed
1661 whether or not the O_STATIC bit is on, but has no effect until it is.
1663 The following properties of a field change when it becomes dynamic:
1664 * If there is no growth limit, there is no final position of the
1665 field; therefore O_AUTOSKIP and O_NL_OVERLOAD are ignored.
1666 * Field justification will be ignored (though whatever justification
1667 is set up will be retained internally and can be queried).
1668 * The dup_field() and link_field() calls copy dynamic-buffer sizes.
1669 If the O_STATIC option is set on one of a collection of links,
1670 buffer resizing will occur only when the field is edited through
1672 * The call field_info() will retrieve the original static size of
1673 the field; use dynamic_field_info() to get the actual dynamic
1678 By default, a field will accept any data that will fit in its input
1679 buffer. However, it is possible to attach a validation type to a
1680 field. If you do this, any attempt to leave the field while it
1681 contains data that does not match the validation type will fail. Some
1682 validation types also have a character-validity check for each time a
1683 character is entered in the field.
1685 A field's validation check (if any) is not called when
1686 set_field_buffer() modifies the input buffer, nor when that buffer is
1687 changed through a linked field.
1689 The form library provides a rich set of pre-defined validation types,
1690 and gives you the capability to define custom ones of your own. You
1691 can examine and change field validation attributes with the following
1693 int set_field_type(FIELD *field, /* field to alter */
1694 FIELDTYPE *ftype, /* type to associate */
1695 ...); /* additional arguments*/
1697 FIELDTYPE *field_type(FIELD *field); /* field to query */
1699 The validation type of a field is considered an attribute of the
1700 field. As with other field attributes, Also, doing set_field_type()
1701 with a NULL field default will change the system default for
1702 validation of newly-created fields.
1704 Here are the pre-defined validation types:
1708 This field type accepts alphabetic data; no blanks, no digits, no
1709 special characters (this is checked at character-entry time). It is
1711 int set_field_type(FIELD *field, /* field to alter */
1712 TYPE_ALPHA, /* type to associate */
1713 int width); /* maximum width of field */
1715 The width argument sets a minimum width of data. Typically you will
1716 want to set this to the field width; if it is greater than the field
1717 width, the validation check will always fail. A minimum width of zero
1718 makes field completion optional.
1722 This field type accepts alphabetic data and digits; no blanks, no
1723 special characters (this is checked at character-entry time). It is
1725 int set_field_type(FIELD *field, /* field to alter */
1726 TYPE_ALNUM, /* type to associate */
1727 int width); /* maximum width of field */
1729 The width argument sets a minimum width of data. As with TYPE_ALPHA,
1730 typically you will want to set this to the field width; if it is
1731 greater than the field width, the validation check will always fail. A
1732 minimum width of zero makes field completion optional.
1736 This type allows you to restrict a field's values to be among a
1737 specified set of string values (for example, the two-letter postal
1738 codes for U.S. states). It is set up with:
1739 int set_field_type(FIELD *field, /* field to alter */
1740 TYPE_ENUM, /* type to associate */
1741 char **valuelist; /* list of possible values */
1742 int checkcase; /* case-sensitive? */
1743 int checkunique); /* must specify uniquely? */
1745 The valuelist parameter must point at a NULL-terminated list of valid
1746 strings. The checkcase argument, if true, makes comparison with the
1747 string case-sensitive.
1749 When the user exits a TYPE_ENUM field, the validation procedure tries
1750 to complete the data in the buffer to a valid entry. If a complete
1751 choice string has been entered, it is of course valid. But it is also
1752 possible to enter a prefix of a valid string and have it completed for
1755 By default, if you enter such a prefix and it matches more than one
1756 value in the string list, the prefix will be completed to the first
1757 matching value. But the checkunique argument, if true, requires prefix
1758 matches to be unique in order to be valid.
1760 The REQ_NEXT_CHOICE and REQ_PREV_CHOICE input requests can be
1761 particularly useful with these fields.
1765 This field type accepts an integer. It is set up as follows:
1766 int set_field_type(FIELD *field, /* field to alter */
1767 TYPE_INTEGER, /* type to associate */
1768 int padding, /* # places to zero-pad to */
1769 int vmin, int vmax); /* valid range */
1771 Valid characters consist of an optional leading minus and digits. The
1772 range check is performed on exit. If the range maximum is less than or
1773 equal to the minimum, the range is ignored.
1775 If the value passes its range check, it is padded with as many leading
1776 zero digits as necessary to meet the padding argument.
1778 A TYPE_INTEGER value buffer can conveniently be interpreted with the C
1779 library function atoi(3).
1783 This field type accepts a decimal number. It is set up as follows:
1784 int set_field_type(FIELD *field, /* field to alter */
1785 TYPE_NUMERIC, /* type to associate */
1786 int padding, /* # places of precision */
1787 double vmin, double vmax); /* valid range */
1789 Valid characters consist of an optional leading minus and digits.
1790 possibly including a decimal point. If your system supports locale's,
1791 the decimal point character used must be the one defined by your
1792 locale. The range check is performed on exit. If the range maximum is
1793 less than or equal to the minimum, the range is ignored.
1795 If the value passes its range check, it is padded with as many
1796 trailing zero digits as necessary to meet the padding argument.
1798 A TYPE_NUMERIC value buffer can conveniently be interpreted with the C
1799 library function atof(3).
1803 This field type accepts data matching a regular expression. It is set
1805 int set_field_type(FIELD *field, /* field to alter */
1806 TYPE_REGEXP, /* type to associate */
1807 char *regexp); /* expression to match */
1809 The syntax for regular expressions is that of regcomp(3). The check
1810 for regular-expression match is performed on exit.
1812 Direct Field Buffer Manipulation
1814 The chief attribute of a field is its buffer contents. When a form has
1815 been completed, your application usually needs to know the state of
1816 each field buffer. You can find this out with:
1817 char *field_buffer(FIELD *field, /* field to query */
1818 int bufindex); /* number of buffer to query */
1820 Normally, the state of the zero-numbered buffer for each field is set
1821 by the user's editing actions on that field. It is sometimes useful to
1822 be able to set the value of the zero-numbered (or some other) buffer
1823 from your application:
1824 int set_field_buffer(FIELD *field, /* field to alter */
1825 int bufindex, /* number of buffer to alter */
1826 char *value); /* string value to set */
1828 If the field is not large enough and cannot be resized to a
1829 sufficiently large size to contain the specified value, the value will
1830 be truncated to fit.
1832 Calling field_buffer() with a null field pointer will raise an error.
1833 Calling field_buffer() on a field not currently selected for input
1834 will return a correct value. Calling field_buffer() on a field that is
1835 currently selected for input may not necessarily give a correct field
1836 buffer value, because entered data is not necessarily copied to buffer
1837 zero before the exit validation check. To guarantee that the returned
1838 buffer value reflects on-screen reality, call field_buffer() either
1839 (1) in the field's exit validation check routine, (2) from the field's
1840 or form's initialization or termination hooks, or (3) just after a
1841 REQ_VALIDATION request has been processed by the forms driver.
1845 As with field attributes, form attributes inherit a default from a
1846 system default form structure. These defaults can be queried or set by
1847 of these functions using a form-pointer argument of NULL.
1849 The principal attribute of a form is its field list. You can query and
1850 change this list with:
1851 int set_form_fields(FORM *form, /* form to alter */
1852 FIELD **fields); /* fields to connect */
1854 char *form_fields(FORM *form); /* fetch fields of form */
1856 int field_count(FORM *form); /* count connect fields */
1858 The second argument of set_form_fields() may be a NULL-terminated
1859 field pointer array like the one required by new_form(). In that case,
1860 the old fields of the form are disconnected but not freed (and
1861 eligible to be connected to other forms), then the new fields are
1864 It may also be null, in which case the old fields are disconnected
1865 (and not freed) but no new ones are connected.
1867 The field_count() function simply counts the number of fields
1868 connected to a given from. It returns -1 if the form-pointer argument
1871 Control of Form Display
1873 In the overview section, you saw that to display a form you normally
1874 start by defining its size (and fields), posting it, and refreshing
1875 the screen. There is an hidden step before posting, which is the
1876 association of the form with a frame window (actually, a pair of
1877 windows) within which it will be displayed. By default, the forms
1878 library associates every form with the full-screen window stdscr.
1880 By making this step explicit, you can associate a form with a declared
1881 frame window on your screen display. This can be useful if you want to
1882 adapt the form display to different screen sizes, dynamically tile
1883 forms on the screen, or use a form as part of an interface layout
1886 The two windows associated with each form have the same functions as
1887 their analogues in the menu library. Both these windows are painted
1888 when the form is posted and erased when the form is unposted.
1890 The outer or frame window is not otherwise touched by the form
1891 routines. It exists so the programmer can associate a title, a border,
1892 or perhaps help text with the form and have it properly refreshed or
1893 erased at post/unpost time. The inner window or subwindow is where the
1894 current form page is actually displayed.
1896 In order to declare your own frame window for a form, you will need to
1897 know the size of the form's bounding rectangle. You can get this
1899 int scale_form(FORM *form, /* form to query */
1900 int *rows, /* form rows */
1901 int *cols); /* form cols */
1903 The form dimensions are passed back in the locations pointed to by the
1904 arguments. Once you have this information, you can use it to declare
1905 of windows, then use one of these functions:
1906 int set_form_win(FORM *form, /* form to alter */
1907 WINDOW *win); /* frame window to connect */
1909 WINDOW *form_win(FORM *form); /* fetch frame window of form */
1911 int set_form_sub(FORM *form, /* form to alter */
1912 WINDOW *win); /* form subwindow to connect */
1914 WINDOW *form_sub(FORM *form); /* fetch form subwindow of form */
1916 Note that curses operations, including refresh(), on the form, should
1917 be done on the frame window, not the form subwindow.
1919 It is possible to check from your application whether all of a
1920 scrollable field is actually displayed within the menu subwindow. Use
1922 int data_ahead(FORM *form); /* form to be queried */
1924 int data_behind(FORM *form); /* form to be queried */
1926 The function data_ahead() returns TRUE if (a) the current field is
1927 one-line and has undisplayed data off to the right, (b) the current
1928 field is multi-line and there is data off-screen below it.
1930 The function data_behind() returns TRUE if the first (upper left hand)
1931 character position is off-screen (not being displayed).
1933 Finally, there is a function to restore the form window's cursor to
1934 the value expected by the forms driver:
1935 int pos_form_cursor(FORM *) /* form to be queried */
1937 If your application changes the form window cursor, call this function
1938 before handing control back to the forms driver in order to
1941 Input Processing in the Forms Driver
1943 The function form_driver() handles virtualized input requests for form
1944 navigation, editing, and validation requests, just as menu_driver does
1945 for menus (see the section on menu input handling).
1946 int form_driver(FORM *form, /* form to pass input to */
1947 int request); /* form request code */
1949 Your input virtualization function needs to take input and then
1950 convert it to either an alphanumeric character (which is treated as
1951 data to be entered in the currently-selected field), or a forms
1954 The forms driver provides hooks (through input-validation and
1955 field-termination functions) with which your application code can
1956 check that the input taken by the driver matched what was expected.
1958 Page Navigation Requests
1960 These requests cause page-level moves through the form, triggering
1961 display of a new form screen.
1964 Move to the next form page.
1967 Move to the previous form page.
1970 Move to the first form page.
1973 Move to the last form page.
1975 These requests treat the list as cyclic; that is, REQ_NEXT_PAGE from
1976 the last page goes to the first, and REQ_PREV_PAGE from the first page
1979 Inter-Field Navigation Requests
1981 These requests handle navigation between fields on the same page.
1987 Move to previous field.
1990 Move to the first field.
1993 Move to the last field.
1996 Move to sorted next field.
1999 Move to sorted previous field.
2002 Move to the sorted first field.
2005 Move to the sorted last field.
2011 Move right to field.
2019 These requests treat the list of fields on a page as cyclic; that is,
2020 REQ_NEXT_FIELD from the last field goes to the first, and
2021 REQ_PREV_FIELD from the first field goes to the last. The order of the
2022 fields for these (and the REQ_FIRST_FIELD and REQ_LAST_FIELD requests)
2023 is simply the order of the field pointers in the form array (as set up
2024 by new_form() or set_form_fields()
2026 It is also possible to traverse the fields as if they had been sorted
2027 in screen-position order, so the sequence goes left-to-right and
2028 top-to-bottom. To do this, use the second group of four
2029 sorted-movement requests.
2031 Finally, it is possible to move between fields using visual directions
2032 up, down, right, and left. To accomplish this, use the third group of
2033 four requests. Note, however, that the position of a form for purposes
2034 of these requests is its upper-left corner.
2036 For example, suppose you have a multi-line field B, and two
2037 single-line fields A and C on the same line with B, with A to the left
2038 of B and C to the right of B. A REQ_MOVE_RIGHT from A will go to B
2039 only if A, B, and C all share the same first line; otherwise it will
2042 Intra-Field Navigation Requests
2044 These requests drive movement of the edit cursor within the currently
2048 Move to next character.
2051 Move to previous character.
2057 Move to previous line.
2063 Move to previous word.
2066 Move to beginning of field.
2069 Move to end of field.
2072 Move to beginning of line.
2075 Move to end of line.
2081 Move right in field.
2089 Each word is separated from the previous and next characters by
2090 whitespace. The commands to move to beginning and end of line or field
2091 look for the first or last non-pad character in their ranges.
2095 Fields that are dynamic and have grown and fields explicitly created
2096 with offscreen rows are scrollable. One-line fields scroll
2097 horizontally; multi-line fields scroll vertically. Most scrolling is
2098 triggered by editing and intra-field movement (the library scrolls the
2099 field to keep the cursor visible). It is possible to explicitly
2100 request scrolling with the following requests:
2103 Scroll vertically forward a line.
2106 Scroll vertically backward a line.
2109 Scroll vertically forward a page.
2112 Scroll vertically backward a page.
2115 Scroll vertically forward half a page.
2118 Scroll vertically backward half a page.
2121 Scroll horizontally forward a character.
2124 Scroll horizontally backward a character.
2127 Scroll horizontally one field width forward.
2130 Scroll horizontally one field width backward.
2133 Scroll horizontally one half field width forward.
2136 Scroll horizontally one half field width backward.
2138 For scrolling purposes, a page of a field is the height of its visible
2143 When you pass the forms driver an ASCII character, it is treated as a
2144 request to add the character to the field's data buffer. Whether this
2145 is an insertion or a replacement depends on the field's edit mode
2146 (insertion is the default.
2148 The following requests support editing the field and changing the edit
2158 New line request (see below for explanation).
2161 Insert space at character location.
2164 Insert blank line at character location.
2167 Delete character at cursor.
2170 Delete previous word at cursor.
2173 Delete line at cursor.
2176 Delete word at cursor.
2179 Clear to end of line.
2182 Clear to end of field.
2187 The behavior of the REQ_NEW_LINE and REQ_DEL_PREV requests is
2188 complicated and partly controlled by a pair of forms options. The
2189 special cases are triggered when the cursor is at the beginning of a
2190 field, or on the last line of the field.
2192 First, we consider REQ_NEW_LINE:
2194 The normal behavior of REQ_NEW_LINE in insert mode is to break the
2195 current line at the position of the edit cursor, inserting the portion
2196 of the current line after the cursor as a new line following the
2197 current and moving the cursor to the beginning of that new line (you
2198 may think of this as inserting a newline in the field buffer).
2200 The normal behavior of REQ_NEW_LINE in overlay mode is to clear the
2201 current line from the position of the edit cursor to end of line. The
2202 cursor is then moved to the beginning of the next line.
2204 However, REQ_NEW_LINE at the beginning of a field, or on the last line
2205 of a field, instead does a REQ_NEXT_FIELD. O_NL_OVERLOAD option is
2206 off, this special action is disabled.
2208 Now, let us consider REQ_DEL_PREV:
2210 The normal behavior of REQ_DEL_PREV is to delete the previous
2211 character. If insert mode is on, and the cursor is at the start of a
2212 line, and the text on that line will fit on the previous one, it
2213 instead appends the contents of the current line to the previous one
2214 and deletes the current line (you may think of this as deleting a
2215 newline from the field buffer).
2217 However, REQ_DEL_PREV at the beginning of a field is instead treated
2218 as a REQ_PREV_FIELD.
2220 If the O_BS_OVERLOAD option is off, this special action is disabled
2221 and the forms driver just returns E_REQUEST_DENIED.
2223 See Form Options for discussion of how to set and clear the overload
2228 If the type of your field is ordered, and has associated functions for
2229 getting the next and previous values of the type from a given value,
2230 there are requests that can fetch that value into the field buffer:
2233 Place the successor value of the current value in the buffer.
2236 Place the predecessor value of the current value in the buffer.
2238 Of the built-in field types, only TYPE_ENUM has built-in successor and
2239 predecessor functions. When you define a field type of your own (see
2240 Custom Validation Types), you can associate our own ordering
2243 Application Commands
2245 Form requests are represented as integers above the curses value
2246 greater than KEY_MAX and less than or equal to the constant
2247 MAX_COMMAND. If your input-virtualization routine returns a value
2248 above MAX_COMMAND, the forms driver will ignore it.
2252 It is possible to set function hooks to be executed whenever the
2253 current field or form changes. Here are the functions that support
2255 typedef void (*HOOK)(); /* pointer to function returning void */
2257 int set_form_init(FORM *form, /* form to alter */
2258 HOOK hook); /* initialization hook */
2260 HOOK form_init(FORM *form); /* form to query */
2262 int set_form_term(FORM *form, /* form to alter */
2263 HOOK hook); /* termination hook */
2265 HOOK form_term(FORM *form); /* form to query */
2267 int set_field_init(FORM *form, /* form to alter */
2268 HOOK hook); /* initialization hook */
2270 HOOK field_init(FORM *form); /* form to query */
2272 int set_field_term(FORM *form, /* form to alter */
2273 HOOK hook); /* termination hook */
2275 HOOK field_term(FORM *form); /* form to query */
2277 These functions allow you to either set or query four different hooks.
2278 In each of the set functions, the second argument should be the
2279 address of a hook function. These functions differ only in the timing
2283 This hook is called when the form is posted; also, just after
2284 each page change operation.
2287 This hook is called when the form is posted; also, just after
2291 This hook is called just after field validation; that is, just
2292 before the field is altered. It is also called when the form is
2296 This hook is called when the form is unposted; also, just
2297 before each page change operation.
2299 Calls to these hooks may be triggered
2300 1. When user editing requests are processed by the forms driver
2301 2. When the current page is changed by set_current_field() call
2302 3. When the current field is changed by a set_form_page() call
2304 See Field Change Commands for discussion of the latter two cases.
2306 You can set a default hook for all fields by passing one of the set
2307 functions a NULL first argument.
2309 You can disable any of these hooks by (re)setting them to NULL, the
2312 Field Change Commands
2314 Normally, navigation through the form will be driven by the user's
2315 input requests. But sometimes it is useful to be able to move the
2316 focus for editing and viewing under control of your application, or
2317 ask which field it currently is in. The following functions help you
2319 int set_current_field(FORM *form, /* form to alter */
2320 FIELD *field); /* field to shift to */
2322 FIELD *current_field(FORM *form); /* form to query */
2324 int field_index(FORM *form, /* form to query */
2325 FIELD *field); /* field to get index of */
2327 The function field_index() returns the index of the given field in the
2328 given form's field array (the array passed to new_form() or
2331 The initial current field of a form is the first active field on the
2332 first page. The function set_form_fields() resets this.
2334 It is also possible to move around by pages.
2335 int set_form_page(FORM *form, /* form to alter */
2336 int page); /* page to go to (0-origin) */
2338 int form_page(FORM *form); /* return form's current page */
2340 The initial page of a newly-created form is 0. The function
2341 set_form_fields() resets this.
2345 Like fields, forms may have control option bits. They can be changed
2346 or queried with these functions:
2347 int set_form_opts(FORM *form, /* form to alter */
2348 int attr); /* attribute to set */
2350 int form_opts_on(FORM *form, /* form to alter */
2351 int attr); /* attributes to turn on */
2353 int form_opts_off(FORM *form, /* form to alter */
2354 int attr); /* attributes to turn off */
2356 int form_opts(FORM *form); /* form to query */
2358 By default, all options are on. Here are the available option bits:
2361 Enable overloading of REQ_NEW_LINE as described in Editing
2362 Requests. The value of this option is ignored on dynamic fields
2363 that have not reached their size limit; these have no last
2364 line, so the circumstances for triggering a REQ_NEXT_FIELD
2368 Enable overloading of REQ_DEL_PREV as described in Editing
2371 The option values are bit-masks and can be composed with logical-or in
2374 Custom Validation Types
2376 The form library gives you the capability to define custom validation
2377 types of your own. Further, the optional additional arguments of
2378 set_field_type effectively allow you to parameterize validation types.
2379 Most of the complications in the validation-type interface have to do
2380 with the handling of the additional arguments within custom validation
2385 The simplest way to create a custom data type is to compose it from
2386 two preexisting ones:
2387 FIELD *link_fieldtype(FIELDTYPE *type1,
2390 This function creates a field type that will accept any of the values
2391 legal for either of its argument field types (which may be either
2392 predefined or programmer-defined). If a set_field_type() call later
2393 requires arguments, the new composite type expects all arguments for
2394 the first type, than all arguments for the second. Order functions
2395 (see Order Requests) associated with the component types will work on
2396 the composite; what it does is check the validation function for the
2397 first type, then for the second, to figure what type the buffer
2398 contents should be treated as.
2402 To create a field type from scratch, you need to specify one or both
2403 of the following things:
2404 * A character-validation function, to check each character as it is
2406 * A field-validation function to be applied on exit from the field.
2408 Here is how you do that:
2409 typedef int (*HOOK)(); /* pointer to function returning int */
2411 FIELDTYPE *new_fieldtype(HOOK f_validate, /* field validator */
2412 HOOK c_validate) /* character validator */
2414 int free_fieldtype(FIELDTYPE *ftype); /* type to free */
2416 At least one of the arguments of new_fieldtype() must be non-NULL. The
2417 forms driver will automatically call the new type's validation
2418 functions at appropriate points in processing a field of the new type.
2420 The function free_fieldtype() deallocates the argument fieldtype,
2421 freeing all storage associated with it.
2423 Normally, a field validator is called when the user attempts to leave
2424 the field. Its first argument is a field pointer, from which it can
2425 get to field buffer 0 and test it. If the function returns TRUE, the
2426 operation succeeds; if it returns FALSE, the edit cursor stays in the
2429 A character validator gets the character passed in as a first
2430 argument. It too should return TRUE if the character is valid, FALSE
2433 Validation Function Arguments
2435 Your field- and character- validation functions will be passed a
2436 second argument as well. This second argument is the address of a
2437 structure (which we will call a pile) built from any of the
2438 field-type-specific arguments passed to set_field_type(). If no such
2439 arguments are defined for the field type, this pile pointer argument
2442 In order to arrange for such arguments to be passed to your validation
2443 functions, you must associate a small set of storage-management
2444 functions with the type. The forms driver will use these to synthesize
2445 a pile from the trailing arguments of each set_field_type() argument,
2446 and a pointer to the pile will be passed to the validation functions.
2448 Here is how you make the association:
2449 typedef char *(*PTRHOOK)(); /* pointer to function returning (char *) */
2450 typedef void (*VOIDHOOK)(); /* pointer to function returning void */
2452 int set_fieldtype_arg(FIELDTYPE *type, /* type to alter */
2453 PTRHOOK make_str, /* make structure from args */
2454 PTRHOOK copy_str, /* make copy of structure */
2455 VOIDHOOK free_str); /* free structure storage */
2457 Here is how the storage-management hooks are used:
2460 This function is called by set_field_type(). It gets one
2461 argument, a va_list of the type-specific arguments passed to
2462 set_field_type(). It is expected to return a pile pointer to a
2463 data structure that encapsulates those arguments.
2466 This function is called by form library functions that allocate
2467 new field instances. It is expected to take a pile pointer,
2468 copy the pile to allocated storage, and return the address of
2472 This function is called by field- and type-deallocation
2473 routines in the library. It takes a pile pointer argument, and
2474 is expected to free the storage of that pile.
2476 The make_str and copy_str functions may return NULL to signal
2477 allocation failure. The library routines will that call them will
2478 return error indication when this happens. Thus, your validation
2479 functions should never see a NULL file pointer and need not check
2482 Order Functions For Custom Types
2484 Some custom field types are simply ordered in the same well-defined
2485 way that TYPE_ENUM is. For such types, it is possible to define
2486 successor and predecessor functions to support the REQ_NEXT_CHOICE and
2487 REQ_PREV_CHOICE requests. Here is how:
2488 typedef int (*INTHOOK)(); /* pointer to function returning int */
2490 int set_fieldtype_arg(FIELDTYPE *type, /* type to alter */
2491 INTHOOK succ, /* get successor value */
2492 INTHOOK pred); /* get predecessor value */
2494 The successor and predecessor arguments will each be passed two
2495 arguments; a field pointer, and a pile pointer (as for the validation
2496 functions). They are expected to use the function field_buffer() to
2497 read the current value, and set_field_buffer() on buffer 0 to set the
2498 next or previous value. Either hook may return TRUE to indicate
2499 success (a legal next or previous value was set) or FALSE to indicate
2504 The interface for defining custom types is complicated and tricky.
2505 Rather than attempting to create a custom type entirely from scratch,
2506 you should start by studying the library source code for whichever of
2507 the pre-defined types seems to be closest to what you want.
2509 Use that code as a model, and evolve it towards what you really want.
2510 You will avoid many problems and annoyances that way. The code in the
2511 ncurses library has been specifically exempted from the package
2512 copyright to support this.
2514 If your custom type defines order functions, have do something
2515 intuitive with a blank field. A useful convention is to make the
2516 successor of a blank field the types minimum value, and its
2517 predecessor the maximum.