Subject: | |
From: | |
Reply To: | |
Date: | Sun, 6 Sep 1998 20:09:59 -0600 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
Newman, Kevin: wrote:
>
> I'm getting ready to write an upshift and downshift routine,
[snip]
> I am between
> decisions on calling CI and UPS, DWNS or setting up two arrays,
> searching for the position of the character in one, then looking up the
> value in the other, or finding the decimal value of the character, then
> adding the appropriate offset and decoding it back to a character. Any
> input as the best approach and why would be appreciated.
An earlier reply already mentioned functions using COBOL 85.
If you don't have COBOL (mentioned in an earlier reply), you might be able
to use the "ctranslate" intrinsic.
I don't know if there is currently a built in table for up-shifting and
downshifting. When I used to work with "classic" HP3000, I wrote my own
tables, except for ASCII and EBCDIC.
The length of the table supplied to "ctranslate" is normally 256 bytes.
But, if you can guarantee that no byte in the input string will have a
value exceeding 127, you can get away with supplying a translation table of
128 bytes.
-----
"In the middle of the pack", using the computer of "Alley Opps".
attached e-mail address is false, to avoid SPAM.
|
|
|