You don't know what you're talking about. Obviously my program uses
find-first find-next interrupts. How else would it build the drive's
directory structure and find files in all folders?
The mode that creates a reference and allows the user to later run a
check against that reference for alterations in date-time and file
length is clearly a form of integrity checking.
Since you insist, then let me rephrase my original question: What is the point
in real-dumb "integrity checking" in an AV scenario?
Exactly what I use. One of the calls is INT 21h service 71A6h
"Windows 95 - Long Filename - Get file info by handle".
Does not work in plain DOS.
You should have said "thank you" on my post
<
[email protected]> and use it to quietly fix and
correct your program's deficiencies, point by point. Instead, you get yourself
entangled deeper and deeper with ridicule excuses and claims and further expose
ignorance and pretentiousness.
The selection of that particular service was a bad choice and shows poor
engineering judgement. There is more than one way to do what you need, with DOS
services, uniquely. It just takes a little more work and brains. Besides, your
waving with "long filename" is a false excuse and eyewash for the simple reason
that the rest of your program, particularly the I/O functions which are DOS,
cannot handle long filename path-strings.
It is compiled as a DOS program fer gawdsakes. WTF do you think QB 4.5
is?
Control you temper, silly, and clean your source from services that are
exclusive to Windows. There are many ways to achieve the worthless
functionality of your program, based on DOS services, uniquely.
[...]
The real reason is a lack of interest and a reluctance to get my hands
on a NT based OS just for testing purposes. I just plain don't want
one (at least right now) for any purpose.
You are bullshitting again and I gave you the way how to graciously retreat from
wriggling. You don't need an NT based OS for testing that a program will work
under NT / W2K and XP. It suffices that a program runs under pure DOS to assure
that it will run under all Windows versions DOS box, NT based inclusive (with
just a couple of exceptions, but they do not concern you, only if your program
was data recovery). That simple!
[...]
What total crap. It's far simpler to design for the keyboard than for
a mouse. I just don't buy the idea that you must make provisions for
both.
Eyewash. Is that how you made decisions throughout your engineering career?
Are you sure you weren't fired?
[...]
You speak of crap while writing total crap. Anyone who know anything
about programming knows that I must necessarily have error handling
all over the program for all kinds of conditions or it wouldn't work
smoothly at all.
Typical to ignorance is to not know what you ignore. Here is how to fix that
error handling deficiency: First, learn what "exit code" and exit procedure
are. Then, add an exit procedure to your code, that will handle exit codes last
thing before exiting the program. The exit procedure is where to call the error
messages. You could then add a test in your program body to see if report files
can be created and set an appropriate exit code if not, and terminate. The exit
procedure would then fetch an error message that would say something like:
"Cannot write report to <specify drive or path here>. Program aborted!"
You don't have to be an engineer to realize that your program is simply
unacceptable the way it's implemented and that your claims are false,
unprofessional, and lack intellectual honesty.
[...]
Anyway, back on topic. I doubt if anyone is much interested in this
useless exchange.
On the contrary. From the e-mails I get, seems that many are interested in the
thread.
Regards, Zvi
