On Fri, Oct 30, 2009 at 05:37:13PM -0700, Glenn Skinner wrote: > Date: Fri, 30 Oct 2009 16:53:51 -0700 > From: Renee Danson Sommerfeld <renee.sommerfeld at sun.com> > Subject: Re: 2009/577 [Network Auto-Magic (NWAM) Phase 1 Updates] > > On Thu, Oct 22, 2009 at 05:04:04PM -0700, Glenn Skinner wrote: > > ... > > > > 4. Add '-V' option to nwamcfg's 'get' subcommand > > Default output from the get command is > > 'propname=propval'. For ease of scripting, the -V > > option outputs the value only. > > > > Example: > > % nwamcfg "select wlan Asbury; get priority" > > priority 0 > > % nwamcfg "select wlan Asbury; get -V priority" > > 0 > > > > The example lacks the "=" character. Is it supposed to be > > there, or not? > > No, it is not intended to be there; the example is correct. > > Ok. (You might want to revise the "'propname=propval'" phrasing in > the description above; it certainly led me to expect to see "=" in the > output of the first example, and I suspect others will find it > confusing as well.)
Yipes, I hadn't even noticed that. I guess that would explain why you were asking! Yes, I should remove that '=' from the text. > > Also, what stability level does this output have? (If you > > intend it to be scriptable, it ought to be Committed...) > > The stability level is addressed in the following item, which > you ask about next. > > > 5. Clarification of command-line output stability > > PSARC 2008/532 incorrectly defined command-line output > > stability to be Uncommitted; it should be Volatile. > > > > See above. You probably want to call out subcommands that are > > intended to produce scriptable output and specify their output > > stability separately. > > We have an RFE to further address scriptability in the command > output, and when we do that, will specify appropriate stability. > > This particular change (the addition of the -V option) was added > specifically because the project is delivering some scripts (smf > service methods) which benefitted from this option. > > So if you've added "-V" specifically to emit scriptable output, why > not declare the corresponding output format to be Committed now rather > than waiting? (Not a big deal, I just don't understand the motivation > for waiting.) I suppose the motivation is simply to wait until we are committing to a more coherent set of scriptable options. -renee
