Add support for `-i random' for seeking to a random position.
Updated manual with new examples.

Add support for negative `-i num' values
(for seeking to the back-end of a series)

+ cmb(1,3): Fix broken handling of -i num' when given -k -1'
+ cmb(3): Remove a stray newline
+ cmb(3): Add show_empty option for including the empty-set
+ cmb(1): Add `-e' option for showing the empty-set

Bugs already fixed for next update:
+ -i NUM was broken with -k -1

Update to address feedback from adridg

In D16132#380314, @adridg wrote:

Comments in general not tied to a specific line of code:

• The -c and -i flags ask for a deterministic order of the combinations. If I use -c 4 -i 1 I expect 4 combinations (whatever they may be). If I use -c 3 -i 2 I expect the last three of those four combinations. So the ordering must be consistent. It is, in your implementation, but the documentation doesn't guarantee that.

cmb(1): Add missing -n option to usage and give a more informative usage
cmb(3): Fix style issue of single-line for-loops
cmb(3): API changes to cmb_print(1) and struct cmb_config

Address feedback and minor improvements:
+ Move static variables out of header (adridg)
+ Remove "All rights reserved" from copyrights (0mp)
+ Add EXAMPLES section to cmb(1) manual (0mp)
+ Add missing $FreeBSD$ (0mp)
+ Update cmb(1,3) HISTORY section [s/12.0/13.0/]
+ Improved error checking for 64-bit integer overflow
+ Fix copy-paste error in cmb(3) manual (/#endif/d)
+ Use errno(2) instead of errx(3) in cmb(3) library
+ Sync with upstream version 0.7 of [lib]cmb
+ Fix style(9) issue of single-line if-statements
+ Add missing $FrauBSD$ keyword to cmb(1) manual
+ Remove redundant RCS keywords from cmb(1)

In D16132#379579, @0mp wrote:

Also, is there a pattern for the use of Qq, Dq and Ql (instead of Dq Li)?

If no additional comments, I'll move ahead without adding the date (which I could not figure out how to add)

Remove WARNS=6 (this is the default)

Make integer overflow with -t print -1 under MK_OPENSSL=no

Update patch for 13.0, bug fix, and libname edits

In D17739#379451, @dteske wrote:

In D17739#379438, @imp wrote:

In D17739#379433, @dteske wrote:

In D17739#379046, @jmg wrote:

oh, don't forget to bump date!

There's no date to bump?

Add one :)

This manual is not written in standard mdoc from what I can see. It's like it was written by martians.

It doesn't have any of the normal anchors I am familiar with.

I'll just shoehorn a date in here somewhere and hope it works. If it does, I'll hoist it up for review.

In D17739#379438, @imp wrote:

In D17739#379433, @dteske wrote:

In D17739#379046, @jmg wrote:

oh, don't forget to bump date!

There's no date to bump?

Add one :)

In D17739#379439, @imp wrote:

Looks good to me...

Not sure about the \f syntax, but other places use it...

In D17739#379046, @jmg wrote:

oh, don't forget to bump date!

Do you have a commit bit or do you need me to ferry this in?

Looks good.

Rebase following r339413 and add patch for rc.conf(5) manual

Fix wrong URL in pkg-descr and update to 0.4.1 (and convert to pkg-plist)

Address feedback from mat@ and update to 0.3.1

Thanks @lme for checking for potential conflict.

Update to 0.1.1

Make `portlint -A' happy

I just thought of something ...

Provide a new offering

Take

This patch is a good effort. Let's keep building upon it and get it ready for when head is thawed on Sep 23. Thank you for resurrecting this topic.

The patch is showing usr.sbin/certctl/trustctl.sh -- is this a phab issue or did the file not get renamed in the patch?

Template suggestion because file is pretty large

Hope this helps

Corrections

Thank you so much for your work on this. It is very much appreciated.

I would just like to comment for the record that I wish there was more time before the 12.0 release so that I could have created dwatch profiles to simplify tapping into these new hooks. I don't have enough time to develop that before 12.0 is completely frozen.

Please run igor -Dgpxy on the manual changes. OK to ignore any nits it complains about that you didn't introduce yourself.

Additional comments following an update

Ping. Is the GNUmakefiles that are an issue? I gues I can host those in my forked OS, but I sincerely believe the other changes are good because it allows people to take the source code from FreeBSD and compile it without modification (assuming they have something like the included GNUmakefiles)

Ping. Reasons for having this in base are similar to basic reasoning for jot, seq, and other argument processing/generating utilities. cmb fills a gap in UNIX that should have been filled decades ago but nobody gave the effort (I suspect because it is a very difficult problem to solve; and given the complexity required, I can absolutely see why nobody has ever BSD-licensed such a tool).

In D16377#347691, @tuexen wrote:

In D16377#347679, @dteske wrote:

Can you expand your patch to include the creation of a corresponding dwatch module? It's easy ...

cd cddl/usr.sbin/dwatch/libexec
sed -e 's/udp/&lite/g;s/length/coverage/g' udp > udplite
echo "$( awk '(line=$1)||1;NF==2&&sub(/udp/,"&lite");/^LINKS/&&gsub(/udp/,"&lite"){_[n++]=$0}/udplite-send/{for(x=0;x<n;)print _[x++]}' Makefile )" > Makefile

Not that easy:

tuexen@head:~/head/cddl/usr.sbin/dwatch/libexec % echo "$( awk '(line=$1)||1;NF==2&&sub(/udp/,"&lite");/^LINKS/&&gsub(/udp/,"&lite"){_[n++]=$0}/udplite-send/{for(x=0;x<n;)print _[x++]}' Makefile )" > Makefile
Illegal variable name.

Can you expand your patch to include the creation of a corresponding dwatch module? It's easy ...

Worth mentioning, I tested on stable/11 (for MFC consideration).

Testing was successful. Everything worked as expected. Cheers.

Aside from testing with what comes with FreeBSD base, I also wanted to make sure that this change would be handled well by my DTrace network visualizer (sysutils/dwatch-gource in ports).