On Wed, Dec 21, 2022 at 10:43:18AM -0500, Paul R. Tagliamonte wrote:
> On Tue, Dec 20, 2022 at 8:27 PM Paul R. Tagliamonte <paul...@gmail.com> wrote:
> >
> > Heyya tech@,
> >
> > Please keep me on cc, I'm not subscribed
>
> Please keep me on cc. The last message wasn't sent to me, so my In-Reply-To
> is going to be wrong. I'm not subscribed to tech@.
yep, sorry, i didn;t spot that note
> From the web:
> > some of the relevant flags are already documented in route(8) ("Routes
> > have associated flags...). the entire list is documented in route(4),
> > but you have to explicitly ask for it (man 4 route)
>
> AFAICT none of these documents which flag "h" maps to, for example.
>
no, but netstat(8) does.
> > and again the flags with detail in netstat(8) ("The mapping between
> > letters and flags is...").
>
> I will admit I'm not smart enough to think to check netstat(8) when
> looking at route(8) output, but that's a fair point.
>
> I understand netstat.8 documenting flags defined in usr.bin/netstat/show.c,
> but is the review here that we should instruct users in route.8 to look up
> the flags coming from sbin/route/show.c in netstat.8 which documents
> usr.bin/netstat/show.c, not sbin/route/show.c ?
>
> I am very sympathetic to the argument that duplicating documentation
> is bad, and can result in maintenance burden or out of date docs, but
> surely people would be more likely to update a manpage in the same
> directory as the file?
>
> I'm OK with this going NOTABUG WONTFIX; I did find the right mappings,
> but I just had to go to the source repo to find it, so I guess
> selfishly I've got
> the knowledge I needed. I was just trying to fix a doc bug when I had state
> in memory, since I know I'd appreciate that as a fellow distro maintainer.
>
> paultag
>
i don;t know why the text is where it is. maybe some of the network
people can say whether the placement makes sense or not.
jmc
