On (07/20/07 17:03), Peter Memishian wrote:
> 
> I'm a bit confused by the example here.  So far, all of the show-*
> commands produce tabular output.  Among other benefits, this makes
> it easy to produce machine-parseable output via the `-p' option.
> I'm not sure how that would work with the output above.
> 
> -- 
> meem

In retrospect, the line with "flow control" is redundant since this
just reflects the info in the lines with "pause" and "asm_pause".

However, if you want to put the link state and transceiver info
into columns there's no meaningful way to define "capable",
"adv" and "lp adv" values for the link state and tranceiver address.

My intent in coming up with  "show-ether" was to produce something that was
human-readable, not machine readable (recall the discussion about the
sea of properties and reducing it to something sensible to humans). If
we focus on machine-readability, then we lose some human readability.

We could have something like this:

         -----------------------------------------------------------------
                     current     capable     advertised     link-partner
                                                            advertised
         -----------------------------------------------------------------
         autoneg
         speed
         duplex                  .......
         rem_fault
         pause                   .......
         asm_pause
         state
         link_clock
         xcvr addr      1          --             --           --
         xcvr id        0          --             --           --
         xcvr type     10Mb/s      --             --           --
         ------------------------------------------------------------------

i.e., a lot of redundant dashes, but it conforms to a tabular
output. Is that preferable?

[I like the other output myself, but don't feel too strongly either way].

--Sowmini


Reply via email to