Strings passed to a DECLARE LIBRARY function are not NUL terminated

[qb64-phoenix-edition]

Strings passed to a DECLARE LIBRARY function are not NUL terminated

[hackwrench]

Sometimes I pass raw data as a string and whatever process adds the null can add an unwanted byte or unwanted time to my program. I'm not sure if programs that don't expect it to be there can have some idea it is there.
Most of the time it's graphics, palette or sound data.
Anyways the process of adding it can affect performance.

[mkilgore]

Can you clarify what your concern is? This would not impact the length of the string, only the actual char * buffer passed to the library function, and it would be added at the time of the call and thus invisible from the QB64 code (this only applies to external calls to DECLARE LIBRARY functions).

If you're currently passing a string along with the length to a DECLARE LIBRARY function then this has no impact you, because the length you'll send stays the same and the NUL character would be the character past the end of the length you sent to the function.

In terms of performance I don't think that's a significant concern here (especially compared to the number of buggy programs I've seen the don't add the NUL byte), but also qbs could definitely be modified to simply maintain the NUL byte itself, which would avoid any real performance penalty for having the NUL byte on the end.