[Date Prev][Date Next][Subject Prev][Subject Next][ Date Index][ Subject Index]

Re: Change Invisible problems with BX



Reply to note from Harry Binswanger  Mon, 19 Apr
2004 13:31:39 -0400

Harry:

> I used ITER to test speed with and without display off (XD). I
> looked at the U2 code for ITER and it doesn't appear to change
> the state of XD/DO that is set manually (by my entering FUNC XD
> on the command line before invoking ITER). Is that correct?

You mean func DX, to freeze the display, not XD.

But no, your statement is not correct! To my surprise, ITER _does_
change the state of DX|DO: there is a func DX in ITER that freezes
the display before starting to execute the subject code. The
reason, as I recall is that, without it, some PRompt messages in the
subject code don't display correctly. However, in principle it's
wrong; ITER shouldn't change the display state, because -- it's
obvious, isn't it? -- freezing the display does, often, materially
increase program speed (although DX may be contraindicated in some
critical situations).

I've deleted DX (a single instance) from my working copy of ITER.
Why don't you do the same (issue LH after doing so, to
reLOAD U2), and let me know if you see any adverse consequences.
There shouldn't be any, and you'll now be able to test DX vs. DO
simply by inserting or omitting DX in your subject code.

> >func GH immediately before BX sortQ2 .
>
> Yes, that works here, too. How on earth did you come up with
> that? If I knew your thinking, I could I apply that the next
> time I have some BX trouble (if ever).

There was very little thinking involved. I observed that the pre-
SORT DeFine procedure forces the cursor into the text area, and
asked what would happen if the cursor were up top. Trial and error.

--
Carl Distefano
cld@xxxxxxxx