On Jun 29, 2018, at 4:36 PM, Keith Medcalf <kmedc...@dessus.com> wrote:
>
> All of the issues raised are "application" problems, not database problems.
Computers are here to aid humans, not the other way around.
> Clearly if you retrieved a value from the database and want to use it as an
> index you have to do bounds checking.
Why? I told the DBMS that the values in that column will be unsigned integers,
yet it accepted a non-integer for storage and then yielded a negative value on
retrieval.
This code will yield a complaint from a sufficiently on-the-ball C compiler:
unsigned foo = external_function();
if (foo >= 0) do_happy_path();
It will rightly complain that the condition is always true.
> The ability to forsee that the world may not be entirely as you expect is the
> root of the difference between a mere coder and a professional software
> programmer.
Therefore, all of the bugs written in C that we can attribute to language
design issues were perpetrated by mere coders. No true Scotsman^Wprogrammer
would ever make such a mistake. Compiler diagnostics are for the weak. TODAY
IS A GOOD DAY TO WRITE SOFTWARE.
_______________________________________________
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
