Page 2 of 2
Re: Bug in string handling in SMSQ/E
Posted: Mon Jan 15, 2024 11:08 am
by NormanDunbar
Re: Bug in string handling in SMSQ/E
Posted: Mon Jan 15, 2024 4:40 pm
by Derek_Stewart
Hi,
I assumed that performing a reverse string slice was was not allowed, though it is not stated thst this is not allowed. I used a common sense approach to the issue.
The string length in podition zero, only causes problems when converting other BASIC programmes to S*BASIC. But if one knows about this it is not a problem.
Re: Bug in string handling in SMSQ/E
Posted: Fri Aug 02, 2024 2:55 pm
by NormanDunbar
I had a chat, via email, with Wolfgang on this matter and he is of the opinion that:
- Debugging SuperBASIC is an absolute pain!
- a$(2 to 1) or a$(1 to 0) giving an empty string should possible result in an error instead.
- Using negative numbers, he thinks that the code treats numbers as unsigned, so that a$(1 to -1) is being read as a$(1 to $FFFF) which is as per the results, so makes sense.
Given his last point above, I did try to find it in the source code to see if it matches what he thinks, and for the life of me, I couldn't find it! I suspect this is why he made his first point!
Cheers,
Norm.