On Fri, Apr 26, 2013 at 08:07:42AM -0400, Ryan Kavanagh wrote:
> Hi Jason,
> 
> On Tuesday, April 23, 2013 at 08:08:59 +0059, Jason McIntyre wrote:
> > i've committed this, but used the text from freebsd's man page. it was
> > very similar anyway, but reads a wee bit better.
> 
> > +.Sh EXIT STATUS
> > +The
> > +.Nm
> > +utility exits with one of the following values:
> > +.It 0
> > +Notification is enabled.
> > +.It 1
> > +Notification is disabled.
> 
> Thanks for looking at my patch. In that case, I would recommend
> tweaking my patch to make it read a bit better rather than applying
> FreeBSD's patch since my "was _ at invocation time" bit is crucial to
> understanding the exit statuses. FreeBSD's patch doesn't tell you
> based on what state it returns an exit status---before or after it
> makes the change---when in fact, the exit status is based on what the
> value was before any change was made.
> 
> For example, if notification was disabled,
> 
>     biff y
> 
> would return 1, which might be confusing seeing that the FreeBSD
> manpage says it returns 0 if "Notification is enabled". Similarly, if
> notification was enabled,
> 
>     biff n
> 
> would return 0, and just
> 
>     biff
> 
> returns 0/1 if enabled/disabled respectively.
> 
> I hope this helps clarify the wordiness of the initial patch.
> 
> Best wishes,
> Ryan
> 

hmm. you're right that i failed to see that. but i'm surprised it works
that way. surely if i run "biff y" i want to see that it has
successfully enabled mail notification, and not whether it was
previously enabled. and if i just want to check if it's currently on or off,
i just run biff w/o flags.

so, i'll make the amendment, but posting this in case someone can
clarify.

jmc

Reply via email to