Re: iSCSI: Text/Operational keys: booleans and ranges

["Julian Satran" <Julian_Satran@il.ibm.com>]

Luben,

For the booleans - 1 and 0 are also integers. The parsing of strings is already mandated (for keys and string values) so no harm in letting them be text.

For ranges - I think you have a good point and if I don't hear anything against it I will introduce it in 12 (I would suggest a colon or a double-colon instead of dash).

Regards,
Julo

Luben Tuikov <luben@splentec.com> Sent by: luben@ns.splentec.com 12-03-02 01:07 Please respond to iSCSI

To:        iSCSI <ips@ece.cmu.edu>, Julian Satran/Haifa/IBM@IBMIL         cc:                 Subject:        iSCSI: Text/Operational keys: booleans and ranges

1. Is it possible to make all Boolean Text/Operational keys's
value range be ``0'' and ``1'', rather than the current
``Yes'', ``No''.

Since those are _Boolean_ values, what will be the objection to
this? 

A user interface will probably display ``Yes'' and ``No'', but
in the protocol, why burden implementations with string
comparison, case-sensitive on top of this, rather than use
strtol(), this making content check easier (strtol() will detect
range errors, e.g. ``OFMarker=hello'').

2. Why not make integer ranges, like OFMarkInt, use dash,
``-'' ASCII: 0x2d, like ``1-65535'', indicating all values in the interval.

The reason is that in the future there may be a text/operational key which can take only a set of
distinct integer values, but no range, e.g. ``SomeBit=0,4,8,16'', That is ``SomeBit'' is still an
integer variable but in negotiation it can choose between certain disctinct values. Furthermore the
standard text/operational code which many of you already have written will work in this situation,
only add atoi()/strtol() at the end...

Comments?

-- 
Luben