| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
It is implemented by the code in usr.bin/mandoc since OpenBSD 5.7.
|
|
|
|
|
|
|
|
|
| |
possible. Annotate <sys/param.h> lines with their current reasons. Switch
to PATH_MAX, NGROUPS_MAX, HOST_NAME_MAX+1, LOGIN_NAME_MAX, etc. Change
MIN() and MAX() to local definitions of MINIMUM() and MAXIMUM() where
sensible to avoid pulling in the pollution. These are the files confirmed
through binary verification.
ok guenther, millert, doug (helped with the verification protocol)
|
|
|
|
|
|
|
|
| |
from optarg and from literal strings - for clarity and to make sure
that future code changes do not attempt to write to them.
No functional change.
ok jca@
|
|
|
|
|
|
|
|
| |
specified "more" as (MAN)PAGER. Document the default pager used while
here. Discussed with schwarze@, ok schwarze@ millert@, jasper@ agrees.
Users can get the previous behaviour by specifying eg.
export MANPAGER="more -s".
|
|
|
|
| |
ok schwarze@ sthen@
|
|
|
|
|
|
|
| |
it's useless and causes out-of-bounds array access later.
While here, remove two unused local variables and two unused struct members.
From Tobias Ulmer <tobiasu at tmux dot org> on tech@, thanks.
"off to you" deraadt@
|
|
|
|
|
|
|
| |
get a pointer to the next element before freeing the previous one,
do not access free'd pointers. Problem found the hard way
by fgsch@ running "man -a x" with malloc.conf -> S on amd64.
ok deraadt@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
such that we don't display them twice.
To be able to do that path check, we have to brace-expand the search
path up front - of course, for -a and -w only, so it doesn't slow
us down in the standard case.
As a free bonus, let -wa print all the filenames without looking
at the timestamps. In the past, -w implied -a, so that's not
introducing any incompatibility.
Issue originally reported and fix OK'd by guenther@,
and deraadt@ agrees with the semantics.
|
|
|
|
|
|
| |
are available in the same tree, only show the newer of the two.
So far, this only handles the case without -a;
i need to look at the case with -a next.
|
|
|
|
|
| |
to be deleted at termination), and then make the signal handler race safe.
ok nicm
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* when combining -m $path with -s $section, -s was ignored
* when combining -m with -s, ONLY $path was searched, like with -M
* when combining -M (or $MANPATH) with -s, -M was ignored
Instead, when combining -m, -M or $MANPATH with -s, let's simply look into
$path/{cat,man}$section{/$arch,} first (for -m) or only (for -M or $MANPATH).
Do not change the behaviour in other cases.
This required a rewrite of the whole path handling,
as a bonus also making the code easier to understand.
Hopefully, it helps to fix www.openbsd.org/cgi-bin/man.cgi properly.
Jason convinced me that changing the manual is not really needed.
Regression tests will follow when the tree unlocks completely.
"you are on the right track" beck@
tested and agreed by jmc@
|
|
|
|
|
|
|
| |
unmaintainable). these days, people use source. these id's do not provide
any benefit, and do hurt the small install media
(the 33,000 line diff is essentially mechanical)
ok with the idea millert, ok dms
|
|
|
|
|
|
|
| |
that MD page headers use mixed case; tidy up the man page regarding this;
diff from Ingo Schwarze
ok sobrado millert krw
|
| |
|
|
|
|
|
|
|
|
| |
and synopsis of commands
lots of good ideas by jmc@
ok jmc@
|
|
|
|
| |
jaredy@
|
| |
|
|
|
|
|
|
| |
Fix from christos NetBSD.
ok otto@, millert@ (earlier version)
|
| |
|
|
|
|
| |
simple as well. ok hshoexer@ jaredy@ henning@
|
|
|
|
| |
in" deraadt@
|
|
|
|
|
|
| |
- use more suitable macros where necessary
- a little grammar
- sort options and sync usage() (last part checked by henning)
|
|
|
|
| |
Closes PR 3563
|
| |
|
|
|
|
| |
rescinded 22 July 1999. Proofed by myself and Theo.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
each non-option shall be treated as arguments to option '\1'.
BSD getopt match '-' in optstring with a '-' on the command line.
This is used to support deprecated options like "su -" that would
otherwise prevent the use of getopt().
Resolving this simply requires that the leading '-' be moved somewhere
else (I moved it to the end of optstring) since position within
optstring is not meaningful.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
use memcpy() not memmove() when we don't care about overlapping segments
use puts(foo) instead of print("%s\n", foo)
|
| |
|
| |
|
| |
|
|
|
|
|
| |
that is, use PAGER rather than "more" if it is defined and non-null, not
just if it is defined.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
man -s 4 -S alpha scc
|
| |
|
| |
|
| |
|
|
|
|
| |
as the rest, and not in reverse.
|
| |
|
|
|