svn commit: r505908 - in head/sysutils: . netevent netevent/files

[Jan Beich]

Alexey Dokuchaev <danfe at freebsd.org> writes:

> On Fri, Jul 05, 2019 at 07:08:08PM +0000, Jan Beich wrote:
>
>> New Revision: 505908
>> URL: https://svnweb.freebsd.org/changeset/ports/505908
>> 
>> Log:
>>   sysutils/netevent: add new port
>
> It would be nice to briefly describe it, even in the commit log.

Those who bother reading commit logs also read actual diffs.
Copy-pasting pkg-descr won't answer "why" the port was added or provide
more input about "what" was done.

A better commit message can only be written by whoever made the change.
Doing so requires a DVCS like Git or if provided separately. Obviously,
the reviewer can still improve commit message by fixing wording/typos,
adjust formatting and maybe elaborate some confusing points.

>> +COMMENT=	Tool for sharing uinput devices with other machines
>
> What's "uinput"?  'U' and 'I' letters are adjacent on my keyboard, is
> this a typo?

evdev(4) created from userland.

https://www.kernel.org/doc/html/latest/input/uinput.html

>> +OPTIONS_SUB=		MANPAGES
>
> That's the same as OPTIONS_SUB=yes.  I also like the idea of putting
> specific options on the OPTIONS_SUB list, but someone on portmgr@ had
> told me they won't do this.

portlint doesn't complain, so stylistically anything can replace "yes".
This time maintainer wanted to annotate why OPTIONS_SUB is needed.
As a committer I don't see the point of enforcing my (or your) bias.

>> Added: head/sysutils/netevent/pkg-plist
>> @@ -0,0 +1,3 @@
>> +bin/netevent
>> +%%MANPAGES%%man/man1/netevent.1.gz
>> +%%PORTEXAMPLES%%%%EXAMPLESDIR%%/simple.ne2
>
> Three-line pkg-plist is a good candidate for PLIST_FILES/PORTDOCS/
> PORTEXAMPLES.

Assuming pkg-plist won't grow in future. As changing required maintainer
approval but the benefit was subjective I've dropped the idea.