Jul 86 Letters
Volume Number: 2
Issue Number: 7
Column Tag: Letters
SYSTEM 3.1.1 BUG
David Smith
All three versions of the new ROMs have a bug in the resource manager that
causes resources to be mashed together. This bug shows up in funny Scrapbook
insertions (two or more resources run together) and in Pagemaker 1.2 document
trashing. Aldus has gone to the trouble of sending a postcard to all owners (a great
service, which we encourage and support!) warning them of the problem. System file
3.1.1, which was supposed to fix bugs in 3.1, which was supposed to fix bugs in 3.0,
does not patch the ROM code correctly. BUT, I am assured by Apple that System 3.2,
which fixes 3.1.1, which fixed...and so forth, DOES INDEED FIX THIS BUG. So, stop using
3.1.1 at once and move to system 3.2, until further notice. (The foregoing information
subject to change at any instant.)
Try Omnis 3
Timothy Standing
San Francisco, CA
Thanks for an outrageously great programming rag. I just received the May issue
this morning and have spent the day reading most of it. It is the only Mac rag I read
now. The rest I just skim to see what new products are being advertised.
I am writing in response to the editorial on pages 5 and 6. I have encountered
similar problems trying to print mailing labels and wanted to share my solution with
you. First, if you are going to print labels in the printer, try using label sheets made
for copiers. They have a tendency to get jammed much less frequently. Second, try
printing labels and then copying them onto labels. This gets around the whole problem. I
admit this is not an elegant solution, but it is reliable. If you only have to do it once
every month, it is not very onerous. [For 5,000 labels? -Ed.]
I have written a subscription database using Omnis 3 which tracks subscribers.
I had a similar problem printing on labels (I am using the 1" high variety). My
solution was to use the ImageWriterII's resident fonts. They are actually quite
reasonable, and they allow you to sidestep the whole print driver problem. I also had to
get around Omnis 3's problems with using nonproportional fonts, and the only way that
was at all feasible was to use the fonts in the ImageWriterII. It is funny using some of
the most sophisticated computer equipment, one had to take a giant step backwards in
order to get anything to work correctly. Thanks again for a great magazine.
TML BUG WORKAROUND
Dr. Christopher Dunn
Wollongon, Australia
I have just received my first two copies of MacTutor and I am pleased to say my
subscription is worth it. It is hard to see how I have managed over the last two years of
rummaging around in a Mac without it.
I don't know if the following has been mentioned in these pages before, but I spent
several hours trying to debug "normal" code that wouldn't behave in TML Pascal. I was
trying to filter out lower case characters from input and convert them to upper case. I
thought the routine was fairly "standard" Pascal. Some characters would behave, but
others would not.
Try the following two programs using TML Pascal (1.11). The program testA
will not work as one expects (but does work in Turbo Pascal and Mac Pascal). It seems
the TML has trouble recognising the elements of a straight set assignment as "if s[i] in
['a'..'z']". Even assigning individual elements to the set by doing "if s[i] in ['a','b','c',]
will still result in an incorrect result. However, if one assigns a variable to that set as
in program testB, the routine works as it should.
Program testA(input, output); { This program will bomb}
var s:string[80];
i:integer;
begin
write('give me a sentence: ');
readln(s);
for i:=1 to length(s) do
if s[i] in ['a'..'z'] then s[i]:=chr(ord(s[i])-32); {upper}
writeln(s);
for i:=1 to length(s) do
if s[i] in ['A'..'Z'] then s[i]:=chr(ord(s[i])+32); {lower}
writeln(s);
readln;
end.
Program testB(input, output); {This program will work}
var s:string[80];
i:integer;
lower, upper: set of char;
begin
lower:=['a'..'z'];
upper:=['A'..'Z'];