xref: /freebsd/contrib/ncurses/man/tic.1m (revision db33c6f3ae9d1231087710068ee4ea5398aacca7)
***************************************************************************
Copyright 2018-2023,2024 Thomas E. Dickey *
Copyright 1998-2016,2017 Free Software Foundation, Inc. *
*
Permission is hereby granted, free of charge, to any person obtaining a *
copy of this software and associated documentation files (the *
"Software"), to deal in the Software without restriction, including *
without limitation the rights to use, copy, modify, merge, publish, *
distribute, distribute with modifications, sublicense, and/or sell *
copies of the Software, and to permit persons to whom the Software is *
furnished to do so, subject to the following conditions: *
*
The above copyright notice and this permission notice shall be included *
in all copies or substantial portions of the Software. *
*
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS *
OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF *
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. *
IN NO EVENT SHALL THE ABOVE COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, *
DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR *
OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR *
THE USE OR OTHER DEALINGS IN THE SOFTWARE. *
*
Except as contained in this notice, the name(s) of the above copyright *
holders shall not be used in advertising or otherwise to promote the *
sale, use or other dealings in this Software without prior written *
authorization. *
***************************************************************************

$Id: tic.1m,v 1.110 2024/04/27 17:57:06 tom Exp $
@TIC@ 1M 2024-04-27 "ncurses @NCURSES_MAJOR@.@NCURSES_MINOR@" "User commands"
.\} .\} . .. .
NAME
\%@TIC@ - compile terminal descriptions for terminfo or termcap
SYNOPSIS
@TIC@ [-\ 0\ 1\ a\ c\ C\ D\ f\ g\ G\ I\ K\ L\ N\ q\ r\ s\ t\ T\ U\ V\ W\ x\ ] [-e terminal-type-list] [-o dir] [-Q[n]] [-R subset] [-v[n]] [-w[n]] file
DESCRIPTION
The @TIC@ command translates a terminfo file from source format into compiled format. The compiled format is necessary for use with the library routines in \%ncurses(3X).

As described in term(5), the database may be either a directory tree (one file per terminal entry) or a hashed database (one record per entry). The @TIC@ command writes only one type of entry, depending on how it was built: .bP For directory trees, the top-level directory, e.g., /usr/share/terminfo, specifies the location of the database. .bP For hashed databases, a filename is needed. If the given file is not found by that name, but can be found by adding the suffix ".db", then that is used.

The default name for the hashed database is the same as the default directory name (only adding a ".db" suffix).

In either case (directory or hashed database), @TIC@ will create the container if it does not exist. For a directory, this would be the \*(``terminfo\*('' leaf, versus a "terminfo.db" file.

The results are normally placed in the system terminfo database \*d. The compiled terminal description can be placed in a different terminfo database. There are two ways to achieve this: .bP First, you may override the system default either by using the -o option, or by setting the variable \%TERMINFO in your shell environment to a valid database location. .bP Secondly, if @TIC@ cannot write in \*d or the location specified using your \%TERMINFO variable, it looks for the directory $HOME/.terminfo (or hashed database $HOME/.terminfo.db); if that location exists, the entry is placed there.

Libraries that read terminfo entries are expected to check in succession .bP a location specified with the \%TERMINFO environment variable, .bP $HOME/.terminfo, .bP directories listed in the \%TERMINFO_DIRS environment variable, .bP a compiled-in list of directories (@TERMINFO_DIRS@), and .bP the system terminfo database (\*d).

The Fetching Compiled Descriptions section in the terminfo(5) manual goes into further detail.

Aliases
This is the same program as @INFOTOCAP@ and @CAPTOINFO@; usually those are linked to, or copied from this program: .bP When invoked as @INFOTOCAP@, @TIC@ sets the -I option. .bP When invoked as @CAPTOINFO@, @TIC@ sets the -C option.
OPTIONS

-0 restricts the output to a single line

-1 restricts the output to a single column

-a tells @TIC@ to retain commented-out capabilities rather than discarding them. Capabilities are commented by prefixing them with a period. This sets the -x option, because it treats the commented-out entries as user-defined names. If the source is termcap, accept the 2-character names required by version 6. Otherwise these are ignored.

-C Force source translation to termcap format. Note: this differs from the -C option of @INFOCMP@(1M) in that it does not merely translate capability names, but also translates terminfo strings to termcap format. Capabilities that are not translatable are left in the entry under their terminfo names but commented out with two preceding dots. The actual format used incorporates some improvements for escaped characters from terminfo format. For a stricter BSD-compatible translation, add the -K option.

If this is combined with -c, @TIC@ makes additional checks to report cases where the terminfo values do not have an exact equivalent in termcap form. For example:

.bP sgr usually will not convert, because termcap lacks the ability to work with more than two parameters, and because termcap lacks many of the arithmetic/logical operators used in terminfo. .bP capabilities with more than one delay or with delays before the end of the string will not convert completely.

-c tells @TIC@ to only check file for errors, including syntax problems and bad use-links. If you specify -C (-I) with this option, the code will print warnings about entries which, after use resolution, are more than 1023 (4096) bytes long. Due to a fixed buffer length in older termcap libraries, as well as buggy checking for the buffer length (and a documented limit in terminfo), these entries may cause core dumps with other implementations.

@TIC@ checks string capabilities to ensure that those with parameters will be valid expressions. It does this check only for the predefined string capabilities; those which are defined with the -x option are ignored.

-D tells @TIC@ to print the database locations that it knows about, and exit. The first location shown is the one to which it would write compiled terminal descriptions. If @TIC@ is not able to find a writable database location according to the rules summarized above, it will print a diagnostic and exit with an error rather than printing a list of database locations.

-e list Limit writes and translations to the comma-separated list of terminal types. If any name or alias of a terminal matches one of the names in the list, the entry will be written or translated as normal. Otherwise no output will be generated for it. The option value is interpreted as a file containing the list if it contains a '/'. (Note: depending on how @TIC@ was compiled, this option may require -I or -C.)

-f Display complex terminfo strings which contain if/then/else/endif expressions indented for readability.

-G Display constant literals in decimal form rather than their character equivalents.

-g Display constant character literals in quoted form rather than their decimal equivalents.

-I Force source translation to terminfo format.

-K Suppress some longstanding \%ncurses extensions to termcap format, e.g., "\es" for space.

-L Force source translation to terminfo format using the long C variable names listed in <term.h>

-N Disable smart defaults. Normally, when translating from termcap to terminfo, the compiler makes a number of assumptions about the defaults of string capabilities reset1_string, carriage_return, cursor_left, cursor_down, scroll_forward, tab, newline, key_backspace, key_left, and key_down, then attempts to use obsolete termcap capabilities to deduce correct values. It also normally suppresses output of obsolete termcap capabilities such as bs. This option forces a more literal translation that also preserves the obsolete capabilities.

-odir Write compiled entries to given database location. Overrides the \%TERMINFO environment variable.

-Qn Rather than show source in terminfo (text) format, print the compiled (binary) format in hexadecimal or base64 form, depending on the option's value:

3 1 hexadecimal

3 2 base64

3 3 hexadecimal and base64

-q Suppress comments and blank lines when showing translated source.

-Rsubset Restrict output to a given subset. This option is for use with archaic versions of terminfo like those on SVr1, Ultrix, or HP-UX that do not support the full set of SVR4/XSI Curses terminfo; and outright broken ports like AIX 3.x that have their own extensions incompatible with SVr4/XSI.

Available subsets are

\*(``SVr1\*('', \*(``Ultrix\*('', \*(``HP\*('', \*(``BSD\*('', and \*(``AIX\*(''

See terminfo(5) for details.

-r Force entry resolution (so there are no remaining tc capabilities) even when doing translation to termcap format. This may be needed if you are preparing a termcap file for a termcap library (such as GNU termcap through version 1.3 or BSD termcap through 4.3BSD) that does not handle multiple tc capabilities per entry.

-s Summarize the compile by showing the database location into which entries are written, and the number of entries which are compiled.

-T eliminates size-restrictions on the generated text. This is mainly useful for testing and analysis, since the compiled descriptions are limited (e.g., 1023 for termcap, 4096 for terminfo).

-t tells @TIC@ to discard commented-out capabilities. Normally when translating from terminfo to termcap, untranslatable capabilities are commented-out.

-U tells @TIC@ to not post-process the data after parsing the source file. Normally, it infers data which is commonly missing in older terminfo data, or in termcaps.

-V reports the version of \%ncurses which was used in this program, and exits.

-vn specifies that (verbose) output be written to standard error trace information showing @TIC@'s progress.

The optional parameter n is a number from 1 to 9, inclusive, indicating the desired level of detail of information.

.bP If \%ncurses is built without tracing support, the optional parameter is ignored. .bP If n is omitted, the default level is 1. .bP If n is specified and greater than 1, the level of detail is increased, and the output is written (with tracing information) to the \*(``trace\*('' file.

The debug flag levels are as follows:

4 1 Names of files created and linked

2 Information related to the \*(``use\*('' facility

3 Statistics from the hashing algorithm

4 Details of extended capabilities

5 (unused)

6 (unused)

7 Entries into the string-table

8 List of tokens encountered by scanner

9 All values computed in construction of the hash table

-W By itself, the -w option will not force long strings to be wrapped. Use the -W option to do this.

If you specify both -f and -W options, the latter is ignored when -f has already split the line.

-wn specifies the width of the output. The parameter is optional. If it is omitted, it defaults to 60.

-x Treat unknown capabilities as user-defined (see user_caps(5)). That is, if you supply a capability name which @TIC@ does not recognize, it will infer its type (Boolean, number or string) from the syntax and make an extended table entry for that. User-defined capability strings whose name begins with \*(``k\*('' are treated as function keys.

Parameters

file contains one or more terminfo terminal descriptions in source format [see terminfo(5)]. Each description in the file describes the capabilities of a particular terminal.

If file is \*(``-\*('', then the data is read from the standard input. The file parameter may also be the path of a character-device.
Processing
All but one of the capabilities recognized by @TIC@ are documented in terminfo(5). The exception is the use capability.

When a use=entry-name field is discovered in a terminal entry currently being compiled, @TIC@ reads in the binary from \*d to complete the entry. (Entries created from file will be used first. @TIC@ duplicates the capabilities in entry-name for the current entry, with the exception of those capabilities that explicitly are defined in the current entry.

When an entry, e.g., entry_name_1, contains a use=entry_name_2 field, any canceled capabilities in entry_name_2 must also appear in entry_name_1 before use= for these capabilities to be canceled in entry_name_1.

Total compiled entries cannot exceed 4096 bytes in the legacy storage format, or 32768 using the extended number format. The name field cannot exceed 512 bytes. Terminal names exceeding the maximum alias length (32 characters on systems with long filenames, 14 characters otherwise) will be truncated to the maximum alias length and a warning message will be printed.

FILES

\*d compiled terminal description database

NOTES
There is some evidence that historic @TIC@ implementations treated description fields with no whitespace in them as additional aliases or short names. This @TIC@ does not do that, but it does warn when description fields may be treated that way and check them for dangerous characters.
EXTENSIONS
Unlike the SVr4 @TIC@ command, this implementation can actually compile termcap sources. In fact, entries in terminfo and termcap syntax can be mixed in a single source file. See terminfo(5) for the list of termcap names taken to be equivalent to terminfo names.

The SVr4 manual pages are not clear on the resolution rules for use capabilities. This implementation of @TIC@ will find use targets anywhere in the source file, or anywhere in the file tree rooted at \%TERMINFO (if \%TERMINFO is defined), or in the user's $HOME/.terminfo database (if it exists), or (finally) anywhere in the system's file tree of compiled entries.

The error messages from this @TIC@ have the same format as GNU C error messages, and can be parsed by GNU Emacs's compile facility.

Aside from -c and -v, options are not portable: .bP Most of @TIC@'s options are not supported by SVr4 tic:

-0 -1 -C -G -I -N -R -T -V -a -e -f -g -o -r -s -t -x

.bP The NetBSD tic supports a few of the \%ncurses options

-a -o -x

and adds -S (a feature which does the same thing as @INFOCMP@'s -e and -E options).

The SVr4 -c mode does not report bad \*(``use=\*('' links.

System V does not compile entries to or read entries from your $HOME/.terminfo database unless \%TERMINFO is explicitly set to it.

PORTABILITY
X/Open Curses, Issue 7 (2009) provides a brief description of tic. It lists one option: -c. The omission of -v is unexpected. The change history states that the description is derived from Tru64. According to its manual pages, that system also supported the -v option.

Shortly after Issue 7 was released, Tru64 was discontinued. As of 2019, the surviving implementations of tic are SVr4 (AIX, HP-UX and Solaris), \%ncurses and NetBSD curses. The SVr4 tic programs all support the -v option. The NetBSD tic program follows X/Open's documentation, omitting the -v option.

The X/Open rationale states that some implementations of tic read terminal descriptions from the standard input if the file parameter is omitted. None of these implementations do that. Further, it comments that some may choose to read from \*(''./terminfo.src\*('' but that is obsolescent behavior from SVr2, and is not (for example) a documented feature of SVr3.

HISTORY
System V Release 2 provided a tic utility. It accepted a single option: -v (optionally followed by a number). According to Ross Ridge's comment in mytinfo, this version of tic was unable to represent cancelled capabilities.

System V Release 3 provided a different tic utility, written by Pavel Curtis, (originally named \*(``compile\*('' in pcurses). This added an option -c to check the file for errors, with the caveat that errors in \*(``use=\*('' links would not be reported. System V Release 3 documented a few warning messages which did not appear in pcurses. While the program itself was changed little as development continued with System V Release 4, the table of capabilities grew from 180 (pcurses) to 464 (Solaris).

In early development of \%ncurses (1993), Zeyd Ben-Halim used the table from mytinfo to extend the pcurses table to 469 capabilities (456 matched SVr4, 8 were only in SVr4, 13 were not in SVr4). Of those 13, 11 were ultimately discarded (perhaps to match the draft of X/Open Curses). The exceptions were \%memory_lock_above and \%memory_unlock (see \%user_caps(5)).

Eric Raymond incorporated parts of mytinfo into \%ncurses to implement the termcap-to-terminfo source conversion, and extended that to begin development of the corresponding terminfo-to-termcap source conversion, Thomas Dickey completed that development over the course of several years.

In 1999, Thomas Dickey added the -x option to support user-defined capabilities.

In 2010, Roy Marples provided a tic program and terminfo library for NetBSD. That implementation adapts several features from \%ncurses, including @TIC@'s -x option.

The -c option tells @TIC@ to check for problems in the terminfo source file. Continued development provides additional checks: .bP pcurses had 8 warnings .bP \%ncurses in 1996 had 16 warnings .bP Solaris (SVr4) curses has 28 warnings .bP NetBSD tic in 2019 has 19 warnings. .bP \%ncurses in 2019 has 96 warnings

The checking done in \%ncurses' @TIC@ helps with the conversion to termcap, as well as pointing out errors and inconsistencies. It is also used to ensure consistency with the user-defined capabilities. There are 527 distinct capabilities in \%ncurses' terminal database; 128 of those are user-defined.

AUTHORS
Eric S. Raymond <esr@snark.thyrsus.com> and

Thomas E. Dickey <dickey@invisible-island.net>

SEE ALSO
\%@CAPTOINFO@(1M), \%@INFOCMP@(1M), \%@INFOTOCAP@(1M), \%@TOE@(1M), \%curses(3X), \%term(5), \%terminfo(5), \%user_caps(5)