Improve markup in some cases where .Li was abused.
Triggered by a smaller diff from jan@.
OK jmc@ jan@

ksh: remove broken special handling of test -t

Drop the vestiges of the pre-POSIX support of `test -t' defaulting to fd
1. It doesn't work and it always succeed since "-t" is treated as a
string by default when no argument (fd) is specified.

diff by Lucas (lucas [at] sexy [dot] is) with minor change by me.
ok millert@

reformat some multiline markup, which had been causing an unwanted
space (specifically in the "alias", "readonly" and "typeset"
commands);

from josiah frentsos

== in [[ does pattern matching as well

OK millert

man pages: add missing commas between subordinate and main clauses

jmc@ dislikes a comma before "then" in a conditional, so leave those
untouched.

ok jmc@

shell scripts should use getopts instead of getopt

Add a prominent deprecation notice to getopt.1.
Add examples of the getopts idiom to sh.1 and ksh.1.

Requested by and ok espie@, ok jmc@

groff complains about the word "An" in an Rs/Re block, believing it a macro,
so escape it;

Add support for ^R (redraw) in insert mode too.
From gotroyb127, OK tb@

use a journal reference instead of cstr when possible

Add some references, most of these were removed when we stopped building
and installing USD/SMM/PSD docs.

jmc@ agrees with the direction, ok millert@ on an earlier diff

Clear screen before redrawing the line with ^L, also in input mode.
This is similar to the emacs mode clear-screen command. Unlike
bash, but like zsh, ^L also works in input mode, not just command
mode. OK kn@ tb@

Add support for set -o pipefail

With the pipefail option set, the exit status of a pipeline is 0 if all
commands succeed, or the return status of the rightmost command that
fails. This can help stronger error checking, but is not a silver
bullet. For example, commands will exhibit a non-zero exit status if
they're killed by a SIGPIPE when writing to a pipe. Yet pipefail was
considered useful enough to be included in the next POSIX standard.

This implementation remembers the value of the pipefail option when
a pipeline is started, as described as option 1) in

https://www.austingroupbugs.net/view.php?id=789#c4102

Requested by ajacoutot@, ok millert@

some corrections to CDPATH;
from chohag

Fix spelling

Partial revert of rev. 1.151:
Reference the First Edition (1989) of Bolsky/Korn which is about ksh88,
the shell the OpenBSD ksh(1) descends from (via pdksh).
The Second Edition (1995) of the book is about ksh93 which we don't provide.
Pointed out by Andras Farkas on bugs@.

zap trailing whitespace;

Allow string greater/less than than operators to work with test aka [.
Previously they were only recognized in [[ ... ]] expressions. This
changes sh/ksh to be consistent with test(1) as well as shells like
bash and dash. OK jca@ jmc@

Bind ^L (C-l) to clear-screen instead of redraw

Slightly more useful for some, same defaults as bash.
No objection deraadt@ phessler@, ok tb@ kn@ benno@

Tweak the syntax displays to show that the list of words
in "for name in [word ...]; do list; done" can be empty.
In sh(1), clarify what happens in that case.
In ksh(1), clarify how it can happen that the list is never executed.
OK jmc@ tb@

Add clear-screen emacs editing command, currently not bound to a
key by default. The shell will query the terminfo database to
find the escape sequence to clear the screen. OK deraadt@

point readers to the correct place instead of having them
jump around.

suggested and ok jmc@, ok jcm@

remove the alias type='whence -v' and replace it with
a buildin command, that just calls into c_whence().
This makes type look like the buildin in other shells and makes
things like system("'type' 'git'"); work.

With lots of suggestions and feedback from anton@, kn@ and jca@.
ok kn@ jca@

Clarify that the "bind" built-in command only affects Emacs editing
mode, to avoid potential confusion pointed out by andreas dot kahari
at icm dot uu dot se on misc@.
Patch minimally tweaked based on a comment from jmc@.
OK anton@ jca@ jmc@ on the previous version without the tweak.

Document change in the default shell prompts.

Discussed with and ok jmc

Fix some incorrectness related to Emacs editing mode in ksh:

- Keep the order of bindings in sync between the manual and implementation
- Fix wrongly documented bindings in the manual
- Break out commands without a default binding in the manual

ok jmc@ tb@

Implement HISTCONTROL ignoredups & ignorespace features

ignoredups: don't save the current line if it is identical to the last
history line.
ignorespace: don't save the current line if it starts with a space

ok anton@ millert@

Stop exposing the emacs-usemeta option, and warn when trying to set it.

Unused since 2012, to be removed after 6.2. Input from anton@,
ok anton@ millert@

\$ in PS1 gets you the default prompt character, which differs from the
default prompt by not including a space after the character;

diff from scott cheloha

Retire old behavior of requiring root prompt to contain # or \!

Requested by akoshibe and phessler
ok phessler@ anton@ jca@

fix broken cross references; found with mandoc -Tlint

Backout previous due to a bug discovered by zhuk@ that requires some tinkering
and is not an easy fix for now.

Add support for pattern substitution to variables in ksh using a common syntax
borrowed from ksh93.

Survived a ports build performed by naddy@ and encouraged by many.

'^T' is documented to be bounded to transpose-chars while in emacs
mode, not "stuff". While here, remove the poorly described "stuff"
and "stuff" bind functions as well..

In the unlikely event anyone has those in .kshrc, they'll need to
be removed.

ok anton@

Assigning to RANDOM calls srand_deterministic(3), not srand(3),
and simplify the next sentence; triggered by a different patch
from Anton dot Lindqvist at gmail dot com.

OK millert@ tb@

Markup improvements:
1. Mark up reserved words intended to appear at the beginning of command
lines with .Ic.
2. Mark up syntax elements intended to appear in the middle of command
lines with .Cm.

This is not only more systematic, but also makes internal search
commands like :tif<enter>, :tstop<enter>, and :t[[<enter> work in
man(1).

remove the "in the future" notes from the past;
from ray lai

Correct description of the "bind" builtin command.

jmc@ finds the key binding stuff "disgusting" (i can't argue with that)
and doesn't want to comment on the content, but agrees with the wording.

note that ulimit -m is not enforced, spotted by jmc

remove unneccessary .Pp; from jan stary

Move bug description to the BUGS section

OK natano@, jmc@

(I forgot to commit it back in mid-August when it was discussed.)

Allow simple custom completions by creating an array named
"complete_commandname_argnum", for example:

set -A complete_kill_1 -- -9 -HUP -INFO -KILL -TERM

To set completions for the first argument to kill(1). If no complete_*
arrays are present, the normal filename completion is offered.

positive comments from many; man page ok/tweaks jmc; ok tedu

Do not handle echo "`echo \"hi\"`" in POSIX mode differently than in
traditional mode. This aligns ksh's behavior with bash and FreeBSD sh.
The interpretation of the POSIX text is disputed, but it is unlikely
that a change from the traditional behavior was intended.

ok millert@

More ksh POSIX compliance fixes by Martijn Dekker:

This simple patch makes the 'command' builtin POSIX-compliant and
consistent with other current shells. It fixes two things:

a) 'command -v' does not find shell reserved words (a.k.a. keywords).
For instance, 'command -v select' outputs nothing but should output
'select'.

b) 'command -pv' always outputs the path of an external command, even
if 'command -p' would execute a builtin. For instance, 'command -p kill'
executes the 'kill' builtin, as expected, but 'command -pv kill' outputs
'/bin/kill'. The '-v' option is supposed to reflect what would actually
be executed, so 'command -pv kill' should output 'kill'. The -p option
sets the PATH to a default system value before doing the search, but
that has no bearing on the fact that builtins take precedence over
external commands.

The patch fixes both issues for 'command' without affecting the
behaviour of the ksh-specific builtin 'whence', which is handled by the
same C function.

Regression test added to obsd-regress.t.

Issues found and fixed by Martijn Dekker, ok millert@

tweak previous a little;

Make set +o conform with POSIX.

It should be possible to use set +o to save the current shell options,
then modify them at will and later restore them to the saved value.
Therefore the output must not only contain the options which are
currently set, but also the ones that are _not_ set, so a restore
disables them if they are enabled.

from Martijn Dekker (martijn (at) inlv (dot) org)
ok sthen@, tb@

Fix a typo. The command is to get the default system PATH value
is "getconf PATH" not "getconf CS_PATH".

remove CSRG BSD-licensed mknod builtin code which was previously used to
accelerate install/upgrade time MAKEDEV runs. That allows ramdisk ksh
to be pledged, without needing "dpath". We'll solve the performance issues
a different way (soon).
ok otto espie natano tb tobiasu rpe

Remove reference to $ERRNO, which was been unimplemented for >= 19
years.

From Michael Reed. ok halex@

remove special characters; diff from michael reed

Revert previous tweak, pending a more complete rewording. Discussed with
halex@, sthen@, jmc@.

This man page uses "Note that" profusely, which is generally considered
bad writing style. Also, add a paragraph break and split up a
megasentence.

Don't use pdksh for self-reference and remove needless historical
context.

Remove commented-out BUGS section that directs reports to long-extinct
upstream.

"one" -> "one or more"

add missing NAME entries; ok schwarze

zap trailing whitespace;

Try to clarify the fc section a tad; fc apprently stands for "fix command".
Also the "fc -e -" syntax is obsoleted by "fc -s". OK tedu@

'fc -e -' is deprecated in favor of 'fc -s', so update the built-in alias

ok millert@

add missing underscore; from theo buehler

Avoid .Ns right after .Pf, it's pointless.
In some cases, do additional cleanup in the immediate vicinity.

Use "alternation" consistently

ok jmc@

Bind the Delete key (ESC[3~) to delete-char-forward

ok mpi@

from zhuk: paths not beginning dot (as well as slash) use CDPATH;

remove the first comma from constructs like ", and," and ", or,": you can use
"and" and "or" to join sentence clauses, and you can use commas, but both hinders
reading;

remove useless escaping; mandoc warned about some of this

put back some information what the character classes actually mean;
while here, remove the lie that regex(3) character classes would
depend on the locale;
ok jmc@

no more ctype(3);

update the built-ins list:
- "times" is both promoted to posix and special (ooh!)
- "pwd" is promoted to posix, but not special (aah!)

while here, jiggle the text somewhat to clarify that "non-POSIX"
actually meant when posix mode is off, not mandated by posix

joint work with guenther

missing .An macros, based on a patch from Jan Stary <hans at stare dot cz>

update book references;
From: Jan Stary
tweaked a bit by myself

Obvious .Xr fixes, found while testing mandocdb(8).

Remove artificial limit on the max array index.
Adapted from a bitrig commit from Martin Natano.
OK zhuk@

Add a proper suspend builtin that saves/restores the tty and pgrp
as needed instead of an alias that just sends SIGSTOP. Login shells
may be suspended if they are not running in an orphan process group.
OK guenther@ jmc@

Document correct interaction of -e flag with && and ||. OK jmc@

Keep documentation in sync with reality and update binding examples.
This should have been part of the keybinding rewrote when support for
multi-character sequences has been added. Pointed out by mikeb@

remove some uesless Pp;

Document ${var[@]} and ${var[*]}

ok espie@

for arithmetic expressions, note that:
+A parameter that is NULL or unset evaluates to 0.

original diff from robert peichaer
this version from naddy

generalise ksh keybinding system - bindings are no longer limited in
length and need not start with ESC or ^X; patch from marco@ feedback mpi@

make -column lists pretty again;

specifically, rewrite them to permit some markup in the column headers,
and use "Ta" instead of literal tabs; mandoc does not currently match groff
100%, but a mandoc fix may be some time off, and we've gone enough releases
with poorly formatting column lists.

in some cases i have rewritten the lists as -tag, where -column made
little sense.

improve line breaking in SYNOPSIS now that semantics of the .Bk macro
has changed to conform with modern groff releases.

diff for route6d(8) written mostly by schwarze@, with lots of useful
advice from jmc@.

ok jmc@, schwarze@

substitute '~' for $HOME in the \W prompt case; matches bash's behavior.

feedback from Dan Harnett

OK deraadt@

according to ingo, it isn;t a problem to have empty [BDS]q macros, so
revert that part of yesterday's diffs;

some macro cleanup:
- remove some old groff workarounds in SYNOPSIS
- avoid "Op ... Oc" constructions
- avoid empty Sq/Dq
- change some incorrect Xr

When the first argument or arguments of a macro are opening delimiters
(parentheses and/or square brackets), both modern groff and mandoc first
output those leading delimiters as plain text, then start the macro scope
after these opening delimiters. This is similar to printing trailing
punctuation and trailing closing delimiters on a macro line outside and
after the macro scope. For example, ".Sq ( text )" is "(`text')",
not "`(text)'". Thus, we now need to quote leading opening delimiters
when we want them inside the macro scope.
These are the cases in src/bin.

"makes sense" jmc@

ksh.1: add ((, )), ]], and `name' to the list of reserved words for compounds
sh.1: add ]] and `name' to the list; additionally document the commands
`select' (requested by Ross L Richardson) and [[

ok millert sobrado

tweak previous;

unset doesn't fail except for read-only variables and functions

ok guenther@

`b' and `c' arguments to mknod are not optional; from LEVAI Daniel
c_sh.c part from sobrado;

ok millert guenther

Array indices are limited to 0 through 10239 now.

ok jmc deraadt

get rid of overlapping blocks discouraged by mdoc.samples(7);
found by and unbreaking the build with mandoc(1);
"fine" jmc@ sobrado@

fix missing bracket by slightly rewriting; from Alan R. S. Bueno

after some discussion with millert, pull in the character class info
from glob(7) to the shell pages - they have their own glob stuff,
and pointing to glob(7) is just not helpful.

ok millert

Add glob(7) and refer to it. OK jmc@ sobrado@

Add POSIX character class support ([:alpha:] and friends) to ksh globbing.
OK deraadt@ krw@ jmc@ sobrado@

some tweaks for previous that got missed;

Make built-in echo behave according to POSIX when set -o posix is in effect:
the only option is -n, and only one of those is parsed.

diff from Ingo Schwarze
ok otto@ kili@; manpage changes ok jmc@

pass "xerrok" status across the execution call stack to more closely
match what both POSIX and ksh.1 already describe in regards to set
-e/errexit's behavior in determining when to exit from nonzero return
values.

specifically, the truth values tested as operands to `&&' and `||', as
well as the resulting compound expression itself, along with the truth
value resulting from a negated command (i.e. a pipeline prefixed `!'),
should not make the shell exit when -e is in effect.

issue reported by matthieu.
testing matthieu, naddy.
ok miod (earlier version), otto.
man page ok jmc.

documentation tweaks.

ok (some time ago) jmc@

Make ulimit able to get and set multiple limits in a single invocation
like bash and zsh do. Requested by espie@, OK deraadt@

Only process ENV when in interactive mode as SU says. Started by
tedu to fix make -j, SU thing spotted by naddy@; ok millert@ @tedu
jmc@ for the man page bit.

convert to new .Dd format;

typos; from alf schlichting

comment fix; from tamas tevesz

clarifications for OPTIND:

hunk #1 from tamas tevesz;
hunk #2 from otto;

ok otto

simplify the PS1 section somewhat by moving the ! and \! stuff together;
ok otto

document some advice when \$ is part of your prompt;
from otto and myself

Implement \$ and \# expansion for PS1. Whoever thought it a clever
idea to assign a special meaning to "\$" -- two chars that are already
(very) special -- deserves a spanking.

back out last - it was not a command line prompt;
thanks matthias kilian

missing command line prompt;

simplify the description of what a restricted shell is:
sh.1 does not need to talk about rksh;
ksh.1 does not need to talk about rsh;
also references to rpdksh are not really helpful;

from tamas tevesz;
ok jaredy

make the PS1 section a little easier to read;

document root PS1 feature:

from matthias kilian (pr #4924), but slightly tweaked version of his diff;
ok otto

document the procedure for providing a startup file for interactive
(but non-login) shells;

document /etc/ksh.kshrc and sort FILES;

fix + ok otto@

document new max array size

ok otto@

brief description of mknod built-in;
add mknod to list of (regular ksh) built-ins;

ok otto@

Introducing mknod as a built-in. It might be against the unix
philosophy, but in this case it's worth it. ok deraadt@

LINES and \\ in PS1 prompt are implemented, so remove information
which says they are not;

from matthias kilian (pr #4476);
ok otto@

Comment out description for 'ulimit -v' since we don't support
RLIMIT_VMEM / RLIMIT_AS at the moment. Support for those may
go in post-3.8.

more `version' removal;
ok otto@

Document ^V change from version to quote command

add missing detail from IFS example

From PR4220/TAKAHASHI Tamotsu <ttakah@lapis.plala.or.jp>

.Xr script 7

document that $HOME is shown as `~' for PS1 \w substitutions;
diff from niall o'higgins, pr 4132 (slightly reworded);

ok otto@

sort options to echo builtin;

a note about why we have both EDITOR and VISUAL env vars;
ok otto@ jaredy@

document that `emacs' is the default editing mode, and that as well
as being set explicitly w/ `set', modes can also be set using
EDITOR and VISUAL;

ok jaredy@ otto@

oops, missed a prompt;

use command prompts and literal for commands;

big overhaul of these pages, to make them clearer to read;
various other issues fixed along the way;

lots of stuff from jaredy@ in here too..

note that symlinks in test file are followed, except -h and -L
(just like test(1) does);

from paul de weerd;

Introduce POSIX hex and octal (0x... and 0...) constants in arithmetic
expressions. Work by Matthias Kilian, based on an old diff by myself.
Note: MAKEDEV should be updated. Tested by many, thanks.

ok millert@ deraadt@

typo;

although the description for the `comment' bind was in the page,
the bind itself (^[#) was missing;

typo;

typo;

These man pages are not derived from UCB and are actually public domain.
Verified with aaron@ who did the conversion from man -> mdoc.

Our times(3) just calls getrusage(2) and gettimeofday(2), converting seconds
to ticks. Since ksh needs things in seconds it then converted them back.
Avoid the silliness and use the getrusage(2) and gettimeofday(2) directly.
With man page help from jmc@

- document IFS behavior changes
- grammar, punctuation, xref, mdoc tweaks
- sort keywords
- indent example displays
- minimize against diffing between sh.1tbl
- document set -o sh
- tweak the displays of some editing and builtin commands
- rework some operator descriptions

ok jmc

escape punctuation;

csh-style ! history completion, which can be activated by using
set -o csh-history (off by default)
this is not feature complete, and likely never will be. no ^
and ! has some oddities ... ksh's internal history stuff has got
some very odd behaviours that are rather nasty
ok various developers

minor tweaks; ok jmc

add a more complex PS1 example, which does not use the new backslash
sequences;

from naddy@, via deraadt@

correct my description of \[ and \]
found by (and ok) deraadt@

document PS1 expansions;

incorrect markup was messing the description of IFS. i've rearranged it
slightly so that no one (me, most likely) falls for it again.

also, make sh(1)'s IFS description a little more consistent.

HISTSIZE cranked to 500

fix spacing after period at EOL. from ray@cyth.net

Document ulimit -v; mickey@ OK

- mdoc and macro fixes and simplification
- punctuation and grammar fixes
- show prompts for example shell commands

help and ok jmc

Document new behavior of $RANDOM. ok deraadt@

typo;
from ray at cyth dot net (PR 3762);

missing commas;

reversed word -> reserved word;
from Karl O. Pinc

escape full stops where necessary to kill erroneous double spaces;
softly, softly, catchy monkey...

`Ns' implies `No', so `Ns No' -> `Ns'; (even simpler in adduser(8))
discussed with todd@

escape full stops properly to avoid double spaces;

- revert a typo introduced when this page was mdoc'ed
- use .Sq rather than .Dq for single letters

note that `&&' and `||' are left-associative;
from vax at carolina dot rr com (NetBSD PR 23814);

agreed (in principal) millert@ deraadt@

typo; ok jmc@

standard section order;

macros with too many args;

escapes in the wrong place;

escape punctuation;
ok deraadt@

In emacs editing mode, ksh by default interprets a set 8th bit as
meta prefix, i.e. all characters with the top bit set (>= 0x80) are
taken as commands and cannot be entered literally.

Introduce a new shell option, emacs-usemeta, that allows to toggle
this behavior. The default is the traditional behavior; to enter
8-bit characters use "set +o emacs-usemeta".

ok fgsch@, henning@

redrew -> redraw;
from Andrey Matveev.

- convert some lists to displays
- .Ql -> .Sq where necessary
- remove some .Pp's before lists
- add missing .El
- remove some wrong macros

liternal -> literal;

Remove the advertising clause in the UCB license which Berkeley
rescinded 22 July 1999. Proofed by myself and Theo.

typos;

suggestions, corrections and ok millert@

typos;

from NetBSD (Soren Jacobsen)
ok millert@

better document how getopts handles options;
plus typos;
from J.A. Neitzel;

Fixes PR 3141;
ok millert@

fix Xr refs; frisco@blackant.net

The 'newgrp' alias is explicitly disabled in our ksh, so do not mention
it in the manpage.

ok millert@

Typo; addresses pr/2038 from tim@robbins.dropbear.id.au

OpenBSD doesn't have RLIMIT_VMEM or RLIMIT_SWAP, so we don't
have ulimit -v or -w either.
ok millert@

fix a typo; tim@robbins.dropbear.id.au

Typo; heko@iki.fi

Documentation fixes per PR 1979 by Kaspar Toomik.

more .%A handling; bk@rt.fm

Various fixes; mpech@prosoft.org.lv

Insert some missing commas.

Typo (introduced by me in mdoc conversion of Rev 1.15); d@

typo; bk@rt.fm

Where we talk about $$, also say mention that PIDs are not safe to use for
generating temporary file names, and point people to mktemp(1) instead.

fix another, similar typo

typo

Sometimes it is confusing when a sentence begins with lowercase; d@

Try to synchronize the ksh(1) and sh(1) man pages somewhat.

Remove hard sentence breaks.

Better mdoc formatting.

'braceexapnd' -> 'braceexpand'

Minor formatting nits.

Formatting fix.

cetain -> certain; kwesterback@home.com

Turn <tab> completion on by default for people used to bash, tcsh, etc...

update man page to pdksh v5.2.14

mdoc'ify syntax of the more complicated commands; when I converted this man
page originally, I didn't know how to do such funky things

remove trailing spaces; missed from a previous commit

patches from pdksh 5.2.13.11

Fixed a bunch of typos, grammar errors, and unclear wording in the sh
man page, and the corresponding problems in the ksh page.

use the hyphenated form "command-line" in the case of a compound adjective;
pjanzen@

repairs

be more conservative with the .Dq macro; in particular, use .Sq macros most of
the time instead when quoting 1-2 characters (makes it much easier to read);
based on a patch received awhile ago from d@

``that that'' -> ``than that''; nash@mcs.net

Finally, new and improved mandoc ksh/sh man pages. This took hours, and hours,
and hours... Someone please read it and find any typos I may have missed.

obliveous -> oblivious; d@

Updates from pdksh-unstable-5.2.13.7. Most notable change is:
trap: exit traps now executed in subshells (without explicit exit call).
See the Changelog for a full list of changes.

Bug fixes from pdksh-unstable-5.2.13.4, including "official" versions of
some that we had already fixed locally.
o typeset -f FUNC doesn't print follows command (and expression) substitutions.
o when re-allocating memory, too much may be copied from old memory.
o set -o printed some options sans names.
o emacs mode: <esc>. in very fist command causes core dump.
o pdksh dumps core after a cd command.
o typeset -i reports on array elements that have no value
(at&t ksh reports on array base name - no index).
o ulimit -ctn unlimittttted kills shell (resource exceeded).
o ". /dev/null" says access denied.
o flag field in aliases incorrectly changed (all flags set instead of
clearing ISSET) in exec.c(flushcom).
o ${#array[*]} prints largest index instead of number of (set) elements
in an array (ksh88 does the former).
o sys_siglist[] doesn't always have NSIG non-null entries...

preceeding -> preceding

More man page fixes. Particularly fix uses of it's/its, affect/effect,
then/than and such. Other miscellaneous problems fixed across these source
files.

more man page fixes; aaron@ug.cs.dal.ca

fix spelling of separate; aaron@ug.cs.dal.ca

pdksh-5.2.13 + local changes

do not nest RS/RE

Document what sh/ksh does when the cd command matches a non empty path
in the CDPATH environment variable.

Don't document ulimit flags that don't exist

sync

Update to 5.2.12.

Move ksh.1 to ksh.1tbl, fixing all of it's problems in one massive paradigm
shift.

Or something like that.

ksh: remove broken special handling of test -t

Drop the vestiges of the pre-POSIX support of `test -t' defaulting to fd
1. It doesn't work and it always succeed since "-t" is treated as a
string by default when no argument (fd) is specified.

diff by Lucas (lucas [at] sexy [dot] is) with minor change by me.
ok millert@

reformat some multiline markup, which had been causing an unwanted
space (specifically in the "alias", "readonly" and "typeset"
commands);

from josiah frentsos

== in [[ does pattern matching as well

OK millert

man pages: add missing commas between subordinate and main clauses

jmc@ dislikes a comma before "then" in a conditional, so leave those
untouched.

ok jmc@

shell scripts should use getopts instead of getopt

Add a prominent deprecation notice to getopt.1.
Add examples of the getopts idiom to sh.1 and ksh.1.

Requested by and ok espie@, ok jmc@

groff complains about the word "An" in an Rs/Re block, believing it a macro,
so escape it;

Add support for ^R (redraw) in insert mode too.
From gotroyb127, OK tb@

use a journal reference instead of cstr when possible

Add some references, most of these were removed when we stopped building
and installing USD/SMM/PSD docs.

jmc@ agrees with the direction, ok millert@ on an earlier diff

Clear screen before redrawing the line with ^L, also in input mode.
This is similar to the emacs mode clear-screen command. Unlike
bash, but like zsh, ^L also works in input mode, not just command
mode. OK kn@ tb@

Add support for set -o pipefail

With the pipefail option set, the exit status of a pipeline is 0 if all
commands succeed, or the return status of the rightmost command that
fails. This can help stronger error checking, but is not a silver
bullet. For example, commands will exhibit a non-zero exit status if
they're killed by a SIGPIPE when writing to a pipe. Yet pipefail was
considered useful enough to be included in the next POSIX standard.

This implementation remembers the value of the pipefail option when
a pipeline is started, as described as option 1) in

https://www.austingroupbugs.net/view.php?id=789#c4102

Requested by ajacoutot@, ok millert@

some corrections to CDPATH;
from chohag

Fix spelling

Partial revert of rev. 1.151:
Reference the First Edition (1989) of Bolsky/Korn which is about ksh88,
the shell the OpenBSD ksh(1) descends from (via pdksh).
The Second Edition (1995) of the book is about ksh93 which we don't provide.
Pointed out by Andras Farkas on bugs@.

zap trailing whitespace;

Allow string greater/less than than operators to work with test aka [.
Previously they were only recognized in [[ ... ]] expressions. This
changes sh/ksh to be consistent with test(1) as well as shells like
bash and dash. OK jca@ jmc@

Bind ^L (C-l) to clear-screen instead of redraw

Slightly more useful for some, same defaults as bash.
No objection deraadt@ phessler@, ok tb@ kn@ benno@

Tweak the syntax displays to show that the list of words
in "for name in [word ...]; do list; done" can be empty.
In sh(1), clarify what happens in that case.
In ksh(1), clarify how it can happen that the list is never executed.
OK jmc@ tb@

Add clear-screen emacs editing command, currently not bound to a
key by default. The shell will query the terminfo database to
find the escape sequence to clear the screen. OK deraadt@

point readers to the correct place instead of having them
jump around.

suggested and ok jmc@, ok jcm@

remove the alias type='whence -v' and replace it with
a buildin command, that just calls into c_whence().
This makes type look like the buildin in other shells and makes
things like system("'type' 'git'"); work.

With lots of suggestions and feedback from anton@, kn@ and jca@.
ok kn@ jca@

Clarify that the "bind" built-in command only affects Emacs editing
mode, to avoid potential confusion pointed out by andreas dot kahari
at icm dot uu dot se on misc@.
Patch minimally tweaked based on a comment from jmc@.
OK anton@ jca@ jmc@ on the previous version without the tweak.

Document change in the default shell prompts.

Discussed with and ok jmc

Fix some incorrectness related to Emacs editing mode in ksh:

- Keep the order of bindings in sync between the manual and implementation
- Fix wrongly documented bindings in the manual
- Break out commands without a default binding in the manual

ok jmc@ tb@

Implement HISTCONTROL ignoredups & ignorespace features

ignoredups: don't save the current line if it is identical to the last
history line.
ignorespace: don't save the current line if it starts with a space

ok anton@ millert@

Stop exposing the emacs-usemeta option, and warn when trying to set it.

Unused since 2012, to be removed after 6.2. Input from anton@,
ok anton@ millert@

\$ in PS1 gets you the default prompt character, which differs from the
default prompt by not including a space after the character;

diff from scott cheloha

Retire old behavior of requiring root prompt to contain # or \!

Requested by akoshibe and phessler
ok phessler@ anton@ jca@

fix broken cross references; found with mandoc -Tlint

Backout previous due to a bug discovered by zhuk@ that requires some tinkering
and is not an easy fix for now.

Add support for pattern substitution to variables in ksh using a common syntax
borrowed from ksh93.

Survived a ports build performed by naddy@ and encouraged by many.

'^T' is documented to be bounded to transpose-chars while in emacs
mode, not "stuff". While here, remove the poorly described "stuff"
and "stuff" bind functions as well..

In the unlikely event anyone has those in .kshrc, they'll need to
be removed.

ok anton@

Assigning to RANDOM calls srand_deterministic(3), not srand(3),
and simplify the next sentence; triggered by a different patch
from Anton dot Lindqvist at gmail dot com.

OK millert@ tb@

Markup improvements:
1. Mark up reserved words intended to appear at the beginning of command
lines with .Ic.
2. Mark up syntax elements intended to appear in the middle of command
lines with .Cm.

This is not only more systematic, but also makes internal search
commands like :tif<enter>, :tstop<enter>, and :t[[<enter> work in
man(1).

remove the "in the future" notes from the past;
from ray lai

Correct description of the "bind" builtin command.

jmc@ finds the key binding stuff "disgusting" (i can't argue with that)
and doesn't want to comment on the content, but agrees with the wording.

note that ulimit -m is not enforced, spotted by jmc

remove unneccessary .Pp; from jan stary

Move bug description to the BUGS section

OK natano@, jmc@

(I forgot to commit it back in mid-August when it was discussed.)

Allow simple custom completions by creating an array named
"complete_commandname_argnum", for example:

set -A complete_kill_1 -- -9 -HUP -INFO -KILL -TERM

To set completions for the first argument to kill(1). If no complete_*
arrays are present, the normal filename completion is offered.

positive comments from many; man page ok/tweaks jmc; ok tedu

Do not handle echo "`echo \"hi\"`" in POSIX mode differently than in
traditional mode. This aligns ksh's behavior with bash and FreeBSD sh.
The interpretation of the POSIX text is disputed, but it is unlikely
that a change from the traditional behavior was intended.

ok millert@

More ksh POSIX compliance fixes by Martijn Dekker:

This simple patch makes the 'command' builtin POSIX-compliant and
consistent with other current shells. It fixes two things:

a) 'command -v' does not find shell reserved words (a.k.a. keywords).
For instance, 'command -v select' outputs nothing but should output
'select'.

b) 'command -pv' always outputs the path of an external command, even
if 'command -p' would execute a builtin. For instance, 'command -p kill'
executes the 'kill' builtin, as expected, but 'command -pv kill' outputs
'/bin/kill'. The '-v' option is supposed to reflect what would actually
be executed, so 'command -pv kill' should output 'kill'. The -p option
sets the PATH to a default system value before doing the search, but
that has no bearing on the fact that builtins take precedence over
external commands.

The patch fixes both issues for 'command' without affecting the
behaviour of the ksh-specific builtin 'whence', which is handled by the
same C function.

Regression test added to obsd-regress.t.

Issues found and fixed by Martijn Dekker, ok millert@

tweak previous a little;

Make set +o conform with POSIX.

It should be possible to use set +o to save the current shell options,
then modify them at will and later restore them to the saved value.
Therefore the output must not only contain the options which are
currently set, but also the ones that are _not_ set, so a restore
disables them if they are enabled.

from Martijn Dekker (martijn (at) inlv (dot) org)
ok sthen@, tb@

Fix a typo. The command is to get the default system PATH value
is "getconf PATH" not "getconf CS_PATH".

remove CSRG BSD-licensed mknod builtin code which was previously used to
accelerate install/upgrade time MAKEDEV runs. That allows ramdisk ksh
to be pledged, without needing "dpath". We'll solve the performance issues
a different way (soon).
ok otto espie natano tb tobiasu rpe

Remove reference to $ERRNO, which was been unimplemented for >= 19
years.

From Michael Reed. ok halex@

remove special characters; diff from michael reed

Revert previous tweak, pending a more complete rewording. Discussed with
halex@, sthen@, jmc@.

This man page uses "Note that" profusely, which is generally considered
bad writing style. Also, add a paragraph break and split up a
megasentence.

Don't use pdksh for self-reference and remove needless historical
context.

Remove commented-out BUGS section that directs reports to long-extinct
upstream.

"one" -> "one or more"

add missing NAME entries; ok schwarze

zap trailing whitespace;

Try to clarify the fc section a tad; fc apprently stands for "fix command".
Also the "fc -e -" syntax is obsoleted by "fc -s". OK tedu@

'fc -e -' is deprecated in favor of 'fc -s', so update the built-in alias

ok millert@

add missing underscore; from theo buehler

Avoid .Ns right after .Pf, it's pointless.
In some cases, do additional cleanup in the immediate vicinity.

Use "alternation" consistently

ok jmc@

Bind the Delete key (ESC[3~) to delete-char-forward

ok mpi@

from zhuk: paths not beginning dot (as well as slash) use CDPATH;

remove the first comma from constructs like ", and," and ", or,": you can use
"and" and "or" to join sentence clauses, and you can use commas, but both hinders
reading;

remove useless escaping; mandoc warned about some of this

put back some information what the character classes actually mean;
while here, remove the lie that regex(3) character classes would
depend on the locale;
ok jmc@

no more ctype(3);

update the built-ins list:
- "times" is both promoted to posix and special (ooh!)
- "pwd" is promoted to posix, but not special (aah!)

while here, jiggle the text somewhat to clarify that "non-POSIX"
actually meant when posix mode is off, not mandated by posix

joint work with guenther

missing .An macros, based on a patch from Jan Stary <hans at stare dot cz>

update book references;
From: Jan Stary
tweaked a bit by myself

Obvious .Xr fixes, found while testing mandocdb(8).

Remove artificial limit on the max array index.
Adapted from a bitrig commit from Martin Natano.
OK zhuk@

Add a proper suspend builtin that saves/restores the tty and pgrp
as needed instead of an alias that just sends SIGSTOP. Login shells
may be suspended if they are not running in an orphan process group.
OK guenther@ jmc@

Document correct interaction of -e flag with && and ||. OK jmc@

Keep documentation in sync with reality and update binding examples.
This should have been part of the keybinding rewrote when support for
multi-character sequences has been added. Pointed out by mikeb@

remove some uesless Pp;

Document ${var[@]} and ${var[*]}

ok espie@

for arithmetic expressions, note that:
+A parameter that is NULL or unset evaluates to 0.

original diff from robert peichaer
this version from naddy

generalise ksh keybinding system - bindings are no longer limited in
length and need not start with ESC or ^X; patch from marco@ feedback mpi@

make -column lists pretty again;

specifically, rewrite them to permit some markup in the column headers,
and use "Ta" instead of literal tabs; mandoc does not currently match groff
100%, but a mandoc fix may be some time off, and we've gone enough releases
with poorly formatting column lists.

in some cases i have rewritten the lists as -tag, where -column made
little sense.

improve line breaking in SYNOPSIS now that semantics of the .Bk macro
has changed to conform with modern groff releases.

diff for route6d(8) written mostly by schwarze@, with lots of useful
advice from jmc@.

ok jmc@, schwarze@

substitute '~' for $HOME in the \W prompt case; matches bash's behavior.

feedback from Dan Harnett

OK deraadt@

according to ingo, it isn;t a problem to have empty [BDS]q macros, so
revert that part of yesterday's diffs;

some macro cleanup:
- remove some old groff workarounds in SYNOPSIS
- avoid "Op ... Oc" constructions
- avoid empty Sq/Dq
- change some incorrect Xr

When the first argument or arguments of a macro are opening delimiters
(parentheses and/or square brackets), both modern groff and mandoc first
output those leading delimiters as plain text, then start the macro scope
after these opening delimiters. This is similar to printing trailing
punctuation and trailing closing delimiters on a macro line outside and
after the macro scope. For example, ".Sq ( text )" is "(`text')",
not "`(text)'". Thus, we now need to quote leading opening delimiters
when we want them inside the macro scope.
These are the cases in src/bin.

"makes sense" jmc@

ksh.1: add ((, )), ]], and `name' to the list of reserved words for compounds
sh.1: add ]] and `name' to the list; additionally document the commands
`select' (requested by Ross L Richardson) and [[

ok millert sobrado

tweak previous;

unset doesn't fail except for read-only variables and functions

ok guenther@

`b' and `c' arguments to mknod are not optional; from LEVAI Daniel
c_sh.c part from sobrado;

ok millert guenther

Array indices are limited to 0 through 10239 now.

ok jmc deraadt

get rid of overlapping blocks discouraged by mdoc.samples(7);
found by and unbreaking the build with mandoc(1);
"fine" jmc@ sobrado@

fix missing bracket by slightly rewriting; from Alan R. S. Bueno

after some discussion with millert, pull in the character class info
from glob(7) to the shell pages - they have their own glob stuff,
and pointing to glob(7) is just not helpful.

ok millert

Add glob(7) and refer to it. OK jmc@ sobrado@

Add POSIX character class support ([:alpha:] and friends) to ksh globbing.
OK deraadt@ krw@ jmc@ sobrado@

some tweaks for previous that got missed;

Make built-in echo behave according to POSIX when set -o posix is in effect:
the only option is -n, and only one of those is parsed.

diff from Ingo Schwarze
ok otto@ kili@; manpage changes ok jmc@

pass "xerrok" status across the execution call stack to more closely
match what both POSIX and ksh.1 already describe in regards to set
-e/errexit's behavior in determining when to exit from nonzero return
values.

specifically, the truth values tested as operands to `&&' and `||', as
well as the resulting compound expression itself, along with the truth
value resulting from a negated command (i.e. a pipeline prefixed `!'),
should not make the shell exit when -e is in effect.

issue reported by matthieu.
testing matthieu, naddy.
ok miod (earlier version), otto.
man page ok jmc.

documentation tweaks.

ok (some time ago) jmc@

Make ulimit able to get and set multiple limits in a single invocation
like bash and zsh do. Requested by espie@, OK deraadt@

Only process ENV when in interactive mode as SU says. Started by
tedu to fix make -j, SU thing spotted by naddy@; ok millert@ @tedu
jmc@ for the man page bit.

convert to new .Dd format;

typos; from alf schlichting

comment fix; from tamas tevesz

clarifications for OPTIND:

hunk #1 from tamas tevesz;
hunk #2 from otto;

ok otto

simplify the PS1 section somewhat by moving the ! and \! stuff together;
ok otto

document some advice when \$ is part of your prompt;
from otto and myself

Implement \$ and \# expansion for PS1. Whoever thought it a clever
idea to assign a special meaning to "\$" -- two chars that are already
(very) special -- deserves a spanking.

back out last - it was not a command line prompt;
thanks matthias kilian

missing command line prompt;

simplify the description of what a restricted shell is:
sh.1 does not need to talk about rksh;
ksh.1 does not need to talk about rsh;
also references to rpdksh are not really helpful;

from tamas tevesz;
ok jaredy

make the PS1 section a little easier to read;

document root PS1 feature:

from matthias kilian (pr #4924), but slightly tweaked version of his diff;
ok otto

document the procedure for providing a startup file for interactive
(but non-login) shells;

document /etc/ksh.kshrc and sort FILES;

fix + ok otto@

document new max array size

ok otto@

brief description of mknod built-in;
add mknod to list of (regular ksh) built-ins;

ok otto@

Introducing mknod as a built-in. It might be against the unix
philosophy, but in this case it's worth it. ok deraadt@

LINES and \\ in PS1 prompt are implemented, so remove information
which says they are not;

from matthias kilian (pr #4476);
ok otto@

Comment out description for 'ulimit -v' since we don't support
RLIMIT_VMEM / RLIMIT_AS at the moment. Support for those may
go in post-3.8.

more `version' removal;
ok otto@

Document ^V change from version to quote command

add missing detail from IFS example

From PR4220/TAKAHASHI Tamotsu <ttakah@lapis.plala.or.jp>

.Xr script 7

document that $HOME is shown as `~' for PS1 \w substitutions;
diff from niall o'higgins, pr 4132 (slightly reworded);

ok otto@

sort options to echo builtin;

a note about why we have both EDITOR and VISUAL env vars;
ok otto@ jaredy@

document that `emacs' is the default editing mode, and that as well
as being set explicitly w/ `set', modes can also be set using
EDITOR and VISUAL;

ok jaredy@ otto@

oops, missed a prompt;

use command prompts and literal for commands;

big overhaul of these pages, to make them clearer to read;
various other issues fixed along the way;

lots of stuff from jaredy@ in here too..

note that symlinks in test file are followed, except -h and -L
(just like test(1) does);

from paul de weerd;

Introduce POSIX hex and octal (0x... and 0...) constants in arithmetic
expressions. Work by Matthias Kilian, based on an old diff by myself.
Note: MAKEDEV should be updated. Tested by many, thanks.

ok millert@ deraadt@

typo;

although the description for the `comment' bind was in the page,
the bind itself (^[#) was missing;

typo;

typo;

These man pages are not derived from UCB and are actually public domain.
Verified with aaron@ who did the conversion from man -> mdoc.

Our times(3) just calls getrusage(2) and gettimeofday(2), converting seconds
to ticks. Since ksh needs things in seconds it then converted them back.
Avoid the silliness and use the getrusage(2) and gettimeofday(2) directly.
With man page help from jmc@

- document IFS behavior changes
- grammar, punctuation, xref, mdoc tweaks
- sort keywords
- indent example displays
- minimize against diffing between sh.1tbl
- document set -o sh
- tweak the displays of some editing and builtin commands
- rework some operator descriptions

ok jmc

escape punctuation;

csh-style ! history completion, which can be activated by using
set -o csh-history (off by default)
this is not feature complete, and likely never will be. no ^
and ! has some oddities ... ksh's internal history stuff has got
some very odd behaviours that are rather nasty
ok various developers

minor tweaks; ok jmc

add a more complex PS1 example, which does not use the new backslash
sequences;

from naddy@, via deraadt@

correct my description of \[ and \]
found by (and ok) deraadt@

document PS1 expansions;

incorrect markup was messing the description of IFS. i've rearranged it
slightly so that no one (me, most likely) falls for it again.

also, make sh(1)'s IFS description a little more consistent.

HISTSIZE cranked to 500

fix spacing after period at EOL. from ray@cyth.net

Document ulimit -v; mickey@ OK

- mdoc and macro fixes and simplification
- punctuation and grammar fixes
- show prompts for example shell commands

help and ok jmc

Document new behavior of $RANDOM. ok deraadt@

typo;
from ray at cyth dot net (PR 3762);

missing commas;

reversed word -> reserved word;
from Karl O. Pinc

escape full stops where necessary to kill erroneous double spaces;
softly, softly, catchy monkey...

`Ns' implies `No', so `Ns No' -> `Ns'; (even simpler in adduser(8))
discussed with todd@

escape full stops properly to avoid double spaces;

- revert a typo introduced when this page was mdoc'ed
- use .Sq rather than .Dq for single letters

note that `&&' and `||' are left-associative;
from vax at carolina dot rr com (NetBSD PR 23814);

agreed (in principal) millert@ deraadt@

typo; ok jmc@

standard section order;

macros with too many args;

escapes in the wrong place;

escape punctuation;
ok deraadt@

In emacs editing mode, ksh by default interprets a set 8th bit as
meta prefix, i.e. all characters with the top bit set (>= 0x80) are
taken as commands and cannot be entered literally.

Introduce a new shell option, emacs-usemeta, that allows to toggle
this behavior. The default is the traditional behavior; to enter
8-bit characters use "set +o emacs-usemeta".

ok fgsch@, henning@

redrew -> redraw;
from Andrey Matveev.

- convert some lists to displays
- .Ql -> .Sq where necessary
- remove some .Pp's before lists
- add missing .El
- remove some wrong macros

liternal -> literal;

Remove the advertising clause in the UCB license which Berkeley
rescinded 22 July 1999. Proofed by myself and Theo.

typos;

suggestions, corrections and ok millert@

typos;

from NetBSD (Soren Jacobsen)
ok millert@

better document how getopts handles options;
plus typos;
from J.A. Neitzel;

Fixes PR 3141;
ok millert@

fix Xr refs; frisco@blackant.net

The 'newgrp' alias is explicitly disabled in our ksh, so do not mention
it in the manpage.

ok millert@

Typo; addresses pr/2038 from tim@robbins.dropbear.id.au

OpenBSD doesn't have RLIMIT_VMEM or RLIMIT_SWAP, so we don't
have ulimit -v or -w either.
ok millert@

fix a typo; tim@robbins.dropbear.id.au

Typo; heko@iki.fi

Documentation fixes per PR 1979 by Kaspar Toomik.

more .%A handling; bk@rt.fm

Various fixes; mpech@prosoft.org.lv

Insert some missing commas.

Typo (introduced by me in mdoc conversion of Rev 1.15); d@

typo; bk@rt.fm

Where we talk about $$, also say mention that PIDs are not safe to use for
generating temporary file names, and point people to mktemp(1) instead.

fix another, similar typo

typo

Sometimes it is confusing when a sentence begins with lowercase; d@

Try to synchronize the ksh(1) and sh(1) man pages somewhat.

Remove hard sentence breaks.

Better mdoc formatting.

'braceexapnd' -> 'braceexpand'

Minor formatting nits.

Formatting fix.

cetain -> certain; kwesterback@home.com

Turn <tab> completion on by default for people used to bash, tcsh, etc...

update man page to pdksh v5.2.14

mdoc'ify syntax of the more complicated commands; when I converted this man
page originally, I didn't know how to do such funky things

remove trailing spaces; missed from a previous commit

patches from pdksh 5.2.13.11

Fixed a bunch of typos, grammar errors, and unclear wording in the sh
man page, and the corresponding problems in the ksh page.

use the hyphenated form "command-line" in the case of a compound adjective;
pjanzen@

repairs

be more conservative with the .Dq macro; in particular, use .Sq macros most of
the time instead when quoting 1-2 characters (makes it much easier to read);
based on a patch received awhile ago from d@

``that that'' -> ``than that''; nash@mcs.net

Finally, new and improved mandoc ksh/sh man pages. This took hours, and hours,
and hours... Someone please read it and find any typos I may have missed.

obliveous -> oblivious; d@

Updates from pdksh-unstable-5.2.13.7. Most notable change is:
trap: exit traps now executed in subshells (without explicit exit call).
See the Changelog for a full list of changes.

Bug fixes from pdksh-unstable-5.2.13.4, including "official" versions of
some that we had already fixed locally.
o typeset -f FUNC doesn't print follows command (and expression) substitutions.
o when re-allocating memory, too much may be copied from old memory.
o set -o printed some options sans names.
o emacs mode: <esc>. in very fist command causes core dump.
o pdksh dumps core after a cd command.
o typeset -i reports on array elements that have no value
(at&t ksh reports on array base name - no index).
o ulimit -ctn unlimittttted kills shell (resource exceeded).
o ". /dev/null" says access denied.
o flag field in aliases incorrectly changed (all flags set instead of
clearing ISSET) in exec.c(flushcom).
o ${#array[*]} prints largest index instead of number of (set) elements
in an array (ksh88 does the former).
o sys_siglist[] doesn't always have NSIG non-null entries...

preceeding -> preceding

More man page fixes. Particularly fix uses of it's/its, affect/effect,
then/than and such. Other miscellaneous problems fixed across these source
files.

more man page fixes; aaron@ug.cs.dal.ca

fix spelling of separate; aaron@ug.cs.dal.ca

pdksh-5.2.13 + local changes

do not nest RS/RE

Document what sh/ksh does when the cd command matches a non empty path
in the CDPATH environment variable.

Don't document ulimit flags that don't exist

sync

Update to 5.2.12.

Move ksh.1 to ksh.1tbl, fixing all of it's problems in one massive paradigm
shift.

Or something like that.

reformat some multiline markup, which had been causing an unwanted
space (specifically in the "alias", "readonly" and "typeset"
commands);

from josiah frentsos

== in [[ does pattern matching as well

OK millert

man pages: add missing commas between subordinate and main clauses

jmc@ dislikes a comma before "then" in a conditional, so leave those
untouched.

ok jmc@

shell scripts should use getopts instead of getopt

Add a prominent deprecation notice to getopt.1.
Add examples of the getopts idiom to sh.1 and ksh.1.

Requested by and ok espie@, ok jmc@

groff complains about the word "An" in an Rs/Re block, believing it a macro,
so escape it;

Add support for ^R (redraw) in insert mode too.
From gotroyb127, OK tb@

use a journal reference instead of cstr when possible

Add some references, most of these were removed when we stopped building
and installing USD/SMM/PSD docs.

jmc@ agrees with the direction, ok millert@ on an earlier diff