From: "Robert A. Rosenberg" <[email protected]>
Sent: Wednesday, 31 August 2011 1:02 PM
At 14:46 +1000 on 08/30/2011, robin wrote about Re: EDIT instruction
-- idiosyncrasies:
A trivial 'solution' is to have a some leading zeros in the
decimal number being converted.
Or have the target field of the instruction start one byte before the
actual target position and prefix the number to be converted with a
non-zero
That may not be possible when the number already has
a digit in the maximum position.
It also requires an instruction to place this digit in position,
or, a byte with '90'x in it to precede the source.
What I suggested requires only that a byte before the
source have '00'x in it.
Of course, other methods are possible, and one would be to
OI 'F0' to the leading digit byte.
Or, as has already been pointed out, use UNPK.
(the mask
mask? Pattern?
also needs to be extended to support the padding
non-zero). This forces the actual first byte of the number to be
displayed not blank filled.
