Codebase list mksh / lintian-fixes/main lksh.1
lintian-fixes/main

Tree @lintian-fixes/main (Download .tar.gz)

lksh.1 @lintian-fixes/mainraw · history · blame

.\" $MirOS: src/bin/mksh/lksh.1,v 1.30 2022/01/28 10:28:19 tg Exp $
.\"-
.\" Copyright © 2008, 2009, 2010, 2012, 2013, 2015, 2016, 2017,
.\"		2018
.\"	mirabilos <m@mirbsd.org>
.\"
.\" Provided that these terms and disclaimer and all copyright notices
.\" are retained or reproduced in an accompanying document, permission
.\" is granted to deal in this work without restriction, including un‐
.\" limited rights to use, publicly perform, distribute, sell, modify,
.\" merge, give away, or sublicence.
.\"
.\" This work is provided “AS IS” and WITHOUT WARRANTY of any kind, to
.\" the utmost extent permitted by applicable law, neither express nor
.\" implied; without malicious intent or gross negligence. In no event
.\" may a licensor, author or contributor be held liable for indirect,
.\" direct, other damage, loss, or other issues arising in any way out
.\" of dealing in the work, even if advised of the possibility of such
.\" damage or existence of a defect, except proven that it results out
.\" of said person’s immediate fault when using the work as intended.
.\"-
.\" Try to make GNU groff and AT&T nroff more compatible
.\" * ` generates ‘ in gnroff, so use \`
.\" * ' generates ’ in gnroff, \' generates ´, so use \*(aq
.\" * - generates ‐ in gnroff, \- generates −, so .tr it to -
.\"   thus use - for hyphens and \- for minus signs and option dashes
.\" * ~ is size-reduced and placed atop in groff, so use \*(TI
.\" * ^ is size-reduced and placed atop in groff, so use \*(ha
.\" * \(en does not work in nroff, so use \*(en for a solo en dash
.\" *   and \*(EM for a correctly spaced em dash
.\" * <>| are problematic, so redefine and use \*(Lt\*(Gt\*(Ba
.\" Also make sure to use \& *before* a punctuation char that is to not
.\" be interpreted as punctuation, and especially with two-letter words
.\" but also (after) a period that does not end a sentence (“e.g.\&”).
.\" The section after the "doc" macropackage has been loaded contains
.\" additional code to convene between the UCB mdoc macropackage (and
.\" its variant as BSD mdoc in groff) and the GNU mdoc macropackage.
.\"
.ie \n(.g \{\
.	if \*[.T]ascii .tr \-\N'45'
.	if \*[.T]latin1 .tr \-\N'45'
.	if \*[.T]utf8 .tr \-\N'45'
.	ds <= \[<=]
.	ds >= \[>=]
.	ds Rq \[rq]
.	ds Lq \[lq]
.	ds sL \(aq
.	ds sR \(aq
.	if \*[.T]utf8 .ds sL `
.	if \*[.T]ps .ds sL `
.	if \*[.T]utf8 .ds sR '
.	if \*[.T]ps .ds sR '
.	ds aq \(aq
.	ds TI \(ti
.	ds ha \(ha
.	ds en \(en
.\}
.el \{\
.	ds aq '
.	ds TI ~
.	ds ha ^
.	ds en \(em
.\}
.ie n \{\
.	ds EM \ \(em\ \&
.\}
.el \{\
.	ds EM \f(TR\^\(em\^\fP
.\}
.\"
.\" Implement .Dd with the Mdocdate RCS keyword
.\"
.rn Dd xD
.de Dd
.ie \\$1$Mdocdate: \{\
.	xD \\$2 \\$3, \\$4
.\}
.el .xD \\$1 \\$2 \\$3 \\$4 \\$5 \\$6 \\$7 \\$8
..
.\"
.\" .Dd must come before definition of .Mx, because when called
.\" with -mandoc, it might implement .Mx itself, but we want to
.\" use our own definition. And .Dd must come *first*, always.
.\"
.Dd $Mdocdate: January 28 2022 $
.\"
.\" Check which macro package we use, and do other -mdoc setup.
.\"
.ie \n(.g \{\
.	if \*[.T]utf8 .tr \[la]\*(Lt
.	if \*[.T]utf8 .tr \[ra]\*(Gt
.	ie d volume-ds-1 .ds tT gnu
.	el .ie d doc-volume-ds-1 .ds tT gnp
.	el .ds tT bsd
.\}
.el .ds tT ucb
.\"
.\" Implement .Mx (MirBSD)
.\"
.ie "\*(tT"gnu" \{\
.	eo
.	de Mx
.	nr curr-font \n[.f]
.	nr curr-size \n[.ps]
.	ds str-Mx \f[\n[curr-font]]\s[\n[curr-size]u]
.	ds str-Mx1 \*[Tn-font-size]\%MirBSD\*[str-Mx]
.	if !\n[arg-limit] \
.	if \n[.$] \{\
.	ds macro-name Mx
.	parse-args \$@
.	\}
.	if (\n[arg-limit] > \n[arg-ptr]) \{\
.	nr arg-ptr +1
.	ie (\n[type\n[arg-ptr]] == 2) \
.	as str-Mx1 \~\*[arg\n[arg-ptr]]
.	el \
.	nr arg-ptr -1
.	\}
.	ds arg\n[arg-ptr] "\*[str-Mx1]
.	nr type\n[arg-ptr] 2
.	ds space\n[arg-ptr] "\*[space]
.	nr num-args (\n[arg-limit] - \n[arg-ptr])
.	nr arg-limit \n[arg-ptr]
.	if \n[num-args] \
.	parse-space-vector
.	print-recursive
..
.	ec
.	ds sP \s0
.	ds tN \*[Tn-font-size]
.\}
.el .ie "\*(tT"gnp" \{\
.	eo
.	de Mx
.	nr doc-curr-font \n[.f]
.	nr doc-curr-size \n[.ps]
.	ds doc-str-Mx \f[\n[doc-curr-font]]\s[\n[doc-curr-size]u]
.	ds doc-str-Mx1 \*[doc-Tn-font-size]\%MirBSD\*[doc-str-Mx]
.	if !\n[doc-arg-limit] \
.	if \n[.$] \{\
.	ds doc-macro-name Mx
.	doc-parse-args \$@
.	\}
.	if (\n[doc-arg-limit] > \n[doc-arg-ptr]) \{\
.	nr doc-arg-ptr +1
.	ie (\n[doc-type\n[doc-arg-ptr]] == 2) \
.	as doc-str-Mx1 \~\*[doc-arg\n[doc-arg-ptr]]
.	el \
.	nr doc-arg-ptr -1
.	\}
.	ds doc-arg\n[doc-arg-ptr] "\*[doc-str-Mx1]
.	nr doc-type\n[doc-arg-ptr] 2
.	ds doc-space\n[doc-arg-ptr] "\*[doc-space]
.	nr doc-num-args (\n[doc-arg-limit] - \n[doc-arg-ptr])
.	nr doc-arg-limit \n[doc-arg-ptr]
.	if \n[doc-num-args] \
.	doc-parse-space-vector
.	doc-print-recursive
..
.	ec
.	ds sP \s0
.	ds tN \*[doc-Tn-font-size]
.\}
.el \{\
.	de Mx
.	nr cF \\n(.f
.	nr cZ \\n(.s
.	ds aa \&\f\\n(cF\s\\n(cZ
.	if \\n(aC==0 \{\
.		ie \\n(.$==0 \&MirBSD\\*(aa
.		el .aV \\$1 \\$2 \\$3 \\$4 \\$5 \\$6 \\$7 \\$8 \\$9
.	\}
.	if \\n(aC>\\n(aP \{\
.		nr aP \\n(aP+1
.		ie \\n(C\\n(aP==2 \{\
.			as b1 \&MirBSD\ #\&\\*(A\\n(aP\\*(aa
.			ie \\n(aC>\\n(aP \{\
.				nr aP \\n(aP+1
.				nR
.			\}
.			el .aZ
.		\}
.		el \{\
.			as b1 \&MirBSD\\*(aa
.			nR
.		\}
.	\}
..
.\}
.\"-
.Dt LKSH 1
.Os MirBSD
.Sh NAME
.Nm lksh
.Nd Legacy Korn shell built on mksh
.Sh SYNOPSIS
.Nm
.Bk -words
.Op Fl +abCefhiklmnprUuvXx
.Op Fl +o Ar opt
.Oo
.Fl c Ar string \*(Ba
.Fl s \*(Ba
.Ar file
.Op Ar args ...
.Oc
.Ek
.Sh DESCRIPTION
.Nm
is a command interpreter intended exclusively for running legacy
shell scripts.
It is built on
.Nm mksh ;
refer to its manual page for details on the scripting language.
It is recommended to port scripts to
.Nm mksh
instead of relying on legacy or objectionable POSIX-mandated behaviour,
since the MirBSD Korn Shell scripting language is much more consistent.
.Pp
Do not use
.Nm
as an interactive or login shell; use
.Nm mksh
instead.
.Pp
Note that it's strongly recommended to invoke
.Nm
with
.Fl o Ic posix
to fully enjoy better compatibility to the
.Tn POSIX
standard (which is probably why you use
.Nm
over
.Nm mksh
in the first place);
.Fl o Ic sh
(possibly additionally to the above) may be needed for some legacy scripts.
.Sh LEGACY MODE
.Nm
currently has the following differences from
.Nm mksh :
.Bl -bullet
.It
The
.Ev KSH_VERSION
string identifies
.Nm
as
.Dq Li LEGACY KSH
instead of
.Dq Li MIRBSD KSH .
Note that the rest of the version string is identical between
the two shell flavours, and the behaviour and differences can
change between versions; see the accompanying manual page
.Xr mksh 1
for the versions this document applies to.
.It
.Nm
uses
.Tn POSIX
arithmetic, which has quite a few implications:
The data type for arithmetic operations is the host
.Tn ISO
C
.Vt long
data type.
Signed integer wraparound is Undefined Behaviour; this means that...
.Bd -literal -offset indent
$ echo $((2147483647 + 1))
.Ed
.Pp
\&... is permitted to, e.g. delete all files on your system
(the figure differs for non-32-bit systems, the rule doesn't).
The sign of the result of a modulo operation with at least one
negative operand is unspecified.
Shift operations on negative numbers are unspecified.
Division of the largest negative number by \-1 is Undefined Behaviour.
The compiler is permitted to delete all data and crash the system
if Undefined Behaviour occurs (see above for an example).
.It
The rotation arithmetic operators are not available.
.It
The shift arithmetic operators take all bits of the second operand into
account; if they exceed permitted precision, the result is unspecified.
.It
Unless
.Ic set -o posix
is active,
.Nm
always uses traditional mode for constructs like:
.Bd -literal -offset indent
$ set -- $(getopt ab:c "$@")
$ echo $?
.Ed
.Pp
POSIX mandates this to show 0, but traditional mode
passes through the errorlevel from the
.Xr getopt 1
command.
.It
Functions defined with the
.Ic function
reserved word share the shell options
.Pq Ic set -o
instead of locally scoping them.
.El
.Sh SEE ALSO
.Xr mksh 1
.Pp
.Pa http://www.mirbsd.org/mksh.htm
.Pp
.Pa http://www.mirbsd.org/ksh\-chan.htm
.Sh CAVEATS
To use
.Nm
as
.Pa /bin/sh ,
compilation to enable
.Ic set -o posix
by default if called as
.Nm sh
.Pq adding Dv \-DMKSH_BINSHPOSIX to Dv CPPFLAGS
is highly recommended for better standards compliance.
.Pp
For better compatibility with legacy scripts, such as many
.Tn Debian
maintainer scripts, Upstart and SYSV init scripts, and other
unfixed scripts, also adding the
.Dv \-DMKSH_BINSHREDUCED
compile-time option to enable
.Em both
.Ic set -o posix -o sh
when the shell is run as
.Nm sh ,
as well as integrating the optional disrecommended
.Xr printf 1
builtin, might be necessary.
.Pp
.Nm
tries to make a cross between a legacy bourne/posix compatibl-ish
shell and a legacy pdksh-alike but
.Dq legacy
is not exactly specified.
.Pp
Talk to the
.Mx
development team and users using the mailing list at
.Aq Mt miros\-mksh@mirbsd.org
or in the
.Li \&#\&!/bin/mksh
IRC channel; mind the infos from
.Pa http://www.mirbsd.org/mksh\-faq.htm#contact
for either.
Consider migrating your legacy scripts to work with
.Nm mksh
instead of requiring
.Nm .