table of contents
PC, UP, BC, ospeed, tgetent, tgetflag, tgetnum, tgetstr, tgoto, tputs - curses emulation of termcap
extern char PC;
extern char * UP;
extern char * BC;
extern short ospeed;
int tgetent(char *bp, const char
int tgetflag(const char *id);
int tgetnum(const char *id);
char *tgetstr(const char *id, char **area);
char *tgoto(const char *cap, int col, int row);
int tputs(const char *str, int affcnt, int (*putc)(int));
These routines are included as a conversion aid for programs that use the termcap library. Their parameters are the same, but the routines are emulated using the terminfo database. Thus, they can only be used to query the capabilities of entries for which a terminfo entry has been compiled.
The tgetent routine loads the entry for name. It returns:
- on success,
- if there is no such entry (or that it is a generic type, having too little information for curses applications to run), and
- if the terminfo database could not be found.
This differs from the termcap library in two ways:
- The emulation ignores the buffer pointer bp. The termcap library would store a copy of the terminal description in the area referenced by this pointer. However, ncurses stores its terminal descriptions in compiled binary form, which is not the same thing.
- There is a difference in return codes. The termcap library does not check if the terminal description is marked with the generic capability, or if the terminal description has cursor-addressing.
The tgetflag routine gets the boolean entry for id, or zero if it is not available.
The tgetnum routine gets the numeric entry for id, or -1 if it is not available.
The tgetstr routine returns the string entry for id, or zero if it is not available. Use tputs to output the returned string. The area parameter is used as follows:
- It is assumed to be the address of a pointer to a buffer managed by the calling application.
- However, ncurses checks to ensure that area is not NULL, and also that the resulting buffer pointer is not NULL. If either check fails, the area parameter is ignored.
- If the checks succeed, ncurses also copies the return value to the buffer pointed to by area, and the area value will be updated to point past the null ending this value.
- The return value itself is an address in the terminal description which is loaded into memory.
Only the first two characters of the id parameter of tgetflag, tgetnum and tgetstr are compared in lookups.
The tgoto routine expands the given capability using the parameters.
- Because the capability may have padding characters, the output of tgoto should be passed to tputs rather than some other output function such as printf(3).
- While tgoto is assumed to be used for the two-parameter cursor positioning capability, termcap applications also use it for single-parameter capabilities.
- Doing this shows a quirk in tgoto: most hardware terminals use cursor addressing with row first, but the original developers of the termcap interface chose to put the column parameter first. The tgoto function swaps the order of parameters. It does this also for calls requiring only a single parameter. In that case, the first parameter is merely a placeholder.
- Normally the ncurses library is compiled with terminfo support. In that case, tgoto uses tparm(3X) (a more capable formatter).
- However, tparm is not a termcap feature, and portable termcap applications should not rely upon its availability.
The tputs routine is described on the terminfo(3NCURSES) manual page. It can retrieve capabilities by either termcap or terminfo name.
The variables PC, UP and BC are set by tgetent to the terminfo entry's data for pad_char, cursor_up and backspace_if_not_bs, respectively. UP is not used by ncurses. PC is used in the tdelay_output function. BC is used in the tgoto emulation. The variable ospeed is set by ncurses in a system-specific coding to reflect the terminal speed.
The termcap functions provide no means for freeing memory, because legacy termcap implementations used only the buffer areas provided by the caller via tgetent and tgetstr. Those buffers are unused in terminfo.
On the other hand, terminfo allocates memory. It uses setupterm to retrieve the data used by tgetent and the functions which return capability values such as tgetstr. One could use
to free this memory, but there is an additional complication with ncurses. It uses a fixed-size pool of storage locations, one per setting of the TERM variable when tgetent is called. The screen(1) program relies upon this arrangement, to improve its performance.
An application which uses only the low-level termcap functions could free the memory using del_curterm, because the pool is freed using other functions (see memleaks(3NCURSES)).
Except where explicitly noted, routines that return an integer return ERR upon failure and OK (SVr4 only specifies "an integer value other than ERR") upon successful completion.
Routines that return pointers return NULL on error.
If you call tgetstr to fetch ca or any other parameterized string, be aware that it will be returned in terminfo notation, not the older and not-quite-compatible termcap notation. This will not cause problems if all you do with it is call tgoto or tparm, which both expand terminfo-style strings as terminfo. (The tgoto function, if configured to support termcap, will check if the string is indeed terminfo-style by looking for "%p" parameters or "$<..>" delays, and invoke a termcap-style parser if the string does not appear to be terminfo).
Because terminfo conventions for representing padding in string capabilities differ from termcap's, users can be surprised:
- tputs("50") in a terminfo system will put out a literal “50” rather than busy-waiting for 50 milliseconds.
- However, if ncurses is configured to support termcap, it may also have been configured to support the BSD-style padding.
- In that case, tputs inspects strings passed to it, looking for digits at the beginning of the string.
- tputs("50") in a termcap system may wait for 50 milliseconds rather than put out a literal “50”
Note that termcap has nothing analogous to terminfo's sgr string. One consequence of this is that termcap applications assume me (terminfo sgr0) does not reset the alternate character set. This implementation checks for, and modifies the data shown to the termcap interface to accommodate termcap's limitation in this respect.
These functions are provided for supporting legacy applications, and should not be used in new programs:
- The XSI Curses standard, Issue 4 describes these functions. However, they are marked TO BE WITHDRAWN and may be removed in future versions.
- X/Open Curses, Issue 5 (December 2007) marked the termcap interface (along with vwprintw and vwscanw) as withdrawn.
Neither the XSI Curses standard nor the SVr4 man pages documented the return values of tgetent correctly, though all three were in fact returned ever since SVr1. In particular, an omission in the XSI Curses documentation has been misinterpreted to mean that tgetent returns OK or ERR. Because the purpose of these functions is to provide compatibility with the termcap library, that is a defect in XCurses, Issue 4, Version 2 rather than in ncurses.
Compatibility with BSD Termcap¶
External variables are provided for support of certain termcap applications. However, termcap applications' use of those variables is poorly documented, e.g., not distinguishing between input and output. In particular, some applications are reported to declare and/or modify ospeed.
The comment that only the first two characters of the id parameter are used escapes many application developers. The original BSD 4.2 termcap library (and historical relics thereof) did not require a trailing null NUL on the parameter name passed to tgetstr, tgetnum and tgetflag. Some applications assume that the termcap interface does not require the trailing NUL for the parameter name. Taking into account these issues:
- As a special case, tgetflag matched against a single-character identifier provided that was at the end of the terminal description. You should not rely upon this behavior in portable programs. This implementation disallows matches against single-character capability names.
- This implementation disallows matches by the termcap interface against extended capability names which are longer than two characters.
The BSD termcap function tgetent returns the text of a termcap entry in the buffer passed as an argument. This library (like other terminfo implementations) does not store terminal descriptions as text. It sets the buffer contents to a null-terminated string.
This library includes a termcap.h header, for compatibility with other implementations. But the header is rarely used because the other implementations are not strictly compatible.
The original BSD termcap (through 4.3BSD) had no header file which gave function prototypes, because that was a feature of ANSI C. BSD termcap was written several years before C was standardized. However, there were two different termcap.h header files in the BSD sources:
- One was used internally by the jove editor in 2BSD through 4.4BSD. It defined global symbols for the termcap variables which it used.
- The other appeared in 4.4BSD Lite Release 2 (mid-1993) as part of libedit (also known as the editline library). The CSRG source history shows that this was added in mid-1992. The libedit header file was used internally, as a convenience for compiling the editline library. It declared function prototypes, but no global variables.
The header file from libedit was added to NetBSD's termcap library in mid-1994.
Meanwhile, GNU termcap was under development, starting in 1990. The first release (termcap 1.0) in 1991 included a termcap.h header. The second release (termcap 1.1) in September 1992 modified the header to use const for the function prototypes in the header where one would expect the parameters to be read-only. This was a difference versus the original BSD termcap. The prototype for tputs also differed, but in that instance, it was libedit which differed from BSD termcap.
A copy of GNU termcap 1.3 was bundled with bash in mid-1993, to support the readline(3) library.
A termcap.h file was provided in ncurses 1.8.1 (November 1993). That reflected influence by emacs(1) (rather than jove(1)) and GNU termcap:
- it provided declarations for a few global symbols used by emacs
- it provided function prototypes (using const).
- a prototype for tparam (a GNU termcap feature) was provided.
Later (in mid-1996) the tparam function was removed from ncurses. As a result, there are differences between any of the four implementations, which must be taken into account by programs which can work with all termcap library interfaces.
ncurses(3NCURSES), putc(3), terminfo_variables(3NCURSES), terminfo(5).