On 04/25/12 09:07, Philip Brown wrote:
Alan Coopersmith wrote:
On 04/23/12 10:02 PM, Philip Brown wrote:
It wouldnt be a big deal, if order of these modifiers was not
significant.
You really need to stop thinking of them as modifiers - in the
current design
(already admitted to be not the most intuitive) they are ordered
fields in
which you can omit empty/wildcard fields using a non-ambiguous
format. You
can no more reorder them than you can the fields in an IPv6 address with
omitted 0-valued segments - ::1 and 1:: are very different addresses.
Forget the terminology; I dont care if you call them "modifiers", or
"fish-bone-handles".
The fixed order is more or less fine if it's consistent. My recent
email pointed out that consistency is broken, when you say "use this
order for input. but we're going to use a different order for output".
order of input fields, and default display order of output fields,
need to be consistent with each other, or it's hostile to the user.
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
The order of the output fields was determined by the previous version of
search, which the current implementation attempted to match. The order
of the input fields is based on containment (a package contains an
action, an action contains a type). I'm sorry this doesn't meet with
your expectations. Perhaps, the next time we work on search, we'll look
at changing the output format.
Brock
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
