In a recent note, "Shmuel Metz (Seymour J.)" said:
> Date: Mon, 3 Oct 2005 13:37:51 -0300
>
> In <[EMAIL PROTECTED]>, on 10/03/2005
> at 07:43 AM, Paul Gilmartin <[log in to unmask]> said:
> >I edited in the comment Now I find the logical NOT symbol is
> >corrupted
>
> What is the hexadecimal value before and after translation? I would
> expect one of ~ (7E) or ¬ (AC) in the PC version.
>
It's the 0xAC. I know you periodically chastise persons who
submit non-USASCII text and fail to indicate the code page.
This ought to apply even more to persons who post code in ASCII
extensions intended to run on an EBCDIC system and fail to
indicate both the ASCII and EBCDIC pages needed for translation.
All the tabs I noted were in comments. Relatively harmless.
And, in one of the files, I see:
grep '[\\/¬] *='
/* IF LINE = JLIN & TY_FL /= 'Y' THEN CALL TYPRUN */
IF POS2 /= 0 THEN
IF POSX /= 0 THEN DO
IF TRANSLATE(SUBSTR(FUNC1,1,1)) /= ' ' THEN
IF UID /= USERID() THEN
IF MSG \= ' ' THEN
IF POS3 /= 0 THEN
DO UNTIL LISTRC ¬= 0
... a marvelous stylistic variety: three different representations
of "not equal"; maximizing the probability that on any given
target system at least one will fail.
(I hadn't realized "/=" is an acceptable form; I had known only
of "\=".)
And, for the individual who sent me privately the two files
as attachments, still containing the non-USASCII character,
('Content-Type: text/plain; charset="iso-8859-1"'), but still
with no indication of the EBCDIC code page to use:
Thanks. However it wasn't I who wanted them; I was merely
sympathizing and volunteering analysis for others who were
having problems.
EBCDIC is a curse!
Thanks again,
gil
--
StorageTek
INFORMATION made POWERFUL
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
