This was the setup I have used in the past. The advantage over MC/N , MC/A is
that
it is more flexible if you want to be very specific, or if you want to allow
some
non alphanumerics as well, since you choose which characters are allowed to
stay.
George
-Original Message-
From:
In message
f07cf5651001211306h41458f3bkf3a2452d48f1d...@mail.gmail.com, Richard
Lewis rbl...@gmail.com writes
I would probably start by making an index on the Part Number, based on an
I-descriptor with an expression like this:
CONVERT(OCONV(OCONV(PartNumber,'MC/N'),'MC/A'),'',PartNumber)
I would probably start by making an index on the Part Number, based on an
I-descriptor with an expression like this:
CONVERT(OCONV(OCONV(PartNumber,'MC/N'),'MC/A'),'',PartNumber)
where PartNumber is a D-pointer to the raw data field that is the Part
Number. This will strip out all the
On Unidata the MCB conversion returns alphanumerics only.
-Kevin (mobile)
On Jan 21, 2010, at 2:06 PM, Richard Lewis rbl...@gmail.com wrote:
I would probably start by making an index on the Part Number, based
on an
I-descriptor with an expression like this:
