There seems to be a string comparison problem that is 
related to the length of the strings being compared.  If a 
string is longer than another string it seems to always 
evaluate to being > the shorter string regardless of the 
starting characters.  In my example below I compare the 
string "AB" to the string "Z" and would expect "AB" to be 
< "Z" from a sorting/comparison standpoint but as you can 
see from the code below that is not the case.  My example 
is comparing literal strings but I discovered the problem 
using variables and just sent you this simplfied case to 
reproduce the error.  The trace snippet follows the code 
also.

NOTE: This error is not critical at this point because I 
just wrote a RightPad function and normalize the strings to 
the same length before comparing them, but this does add 
overhead and it would be nice to not have to remember to do 
that in the future.

Thanks, Mark


<TESEC
 

This has been fixed, available in this afternoon's build.

It was an optimization to rapidly find text that isn't 
equal, but as you have pointed-out, it then is providing 
results that are useless for an alphanumeric comparison.

Wow, I am really impressed with your rapid response to 
problems.  I always try to isolate and send a simplified 
easily reproducable demonstration of an error and I have 
done this on different occasions with other companies and 
even when I have a clear and simple case where I can 
demonstrate a bug I often get the run around.  And when 
they finally acknowledge that it is a problem then the 
resolution is often not in the forseeable future.  It has 
been a pleasure working with Screensurfer support.  Please 
pass this comment along to your manager or supervisor to 
let them know about the excellent service.

Thanks again,

Mark Horan

Thanks for the compliments.  The main reason for our 
responsiveness is that we don't rely on an escalation 
system--all support and development staff are on the e-mail 
distribution list for new tickets--if a problem is 
obviously something only a developer can handle, he jumps 
on it and avoids hassles for everyone.

A much better system than phone support!

Anyway, new build is now available with fixes for tickets 
765 and 766--

ftp://www.screensurfer.com/t3_0_1E.zip

Could we have those same changes applied to 2.0.11G since 
that is what we are currently using?  We haven't tested or 
upgraded to version 3 yet and would like to not introduce 
too many new variables.

Thanks, Mark

Sorry, forgot that you were V2...will let you know when 
it's ready...

OK, ftp://www.screensurfer.com/T2_0_11H.zip has the fixes 
for FindText, large literal assignment, and text comparison 
of strings with different lengths...