Comments on: AND truncation of series
Sometimes I just need to ask...
Do you see the value in the truncation of this operation:
>> #{010203} and #{0202}
== #{0002}
(Note that the result is the length of the shorter argument.)
However:
>> #{010203} or #{0202}
== #{030203}
leaves it full length.
There may have been some logical reason in the past, but if their was, I don't see it in the docs. So, unless someone objects, AND will be modified to produce the same length as OR.
This is Curecode bug #1013.
6 Comments Comments:
Brian Hawley
3-Jul-2009 2:33:46 |
One possibility wold be to pretend that the short binary was 00-padded to the length of the longer one and then AND it:
>> #{010203} and #{0202}
== #{000200}
>> (#{010203} and #{0202}) = (#{010203} and #{020200})
== true
This would be consistent with the behavior of OR:
>> #{010203} or #{0202}
== #{030203}
>> (#{010203} or #{0202}) = (#{010203} or #{020200})
== true
FF-padding would not be consistent with OR:
>> #{010203} and #{0202}
== #{000203}
>> (#{010203} and #{0202}) = (#{010203} and #{0202FF})
== true | Pijoter
3-Jul-2009 3:20:22 |
Short binary will be 00-padded from left or right side?
For now (view 2.6):
>> to-integer (#{FFFF} and #{00});== 255
>> (to-integer #{FFFF}) and (to-integer #{00});== 0
| Sunanda
3-Jul-2009 3:23:46 |
zero padding would also be consistent with existing (R2 and
1.2.3 and 1.2.3.4.5
== 1.2.3.0.0> | Gregg Irwin
3-Jul-2009 8:55:31 |
Zero padding makes sense to me. | Brian Hawley
3-Jul-2009 16:10:59 |
Piotr: "Short binary will be 00-padded from left or right side?"
Padded on the right side. | Carl Sassenrath
6-Jul-2009 13:20:50 |
This change has been made. |
Post a Comment:
You can post a comment here. Keep it on-topic.
|
