Re: OT: A bit of Sudoku

[Adam Beneschan <adambeneschan@gmail.com>]

On Friday, June 6, 2014 12:51:47 AM UTC-7, Dmitry A. Kazakov wrote:

> > Otherwise, someone could look at a subprogram that is called in between,
> > and never guess that the subprogram might not complete normally (A calls
> > B, B calls C, C raises an exception that gets passed over B's head back to
> > A; a programmer trying to read B might not suspect that B may not complete
> > in a non-error situation.)  In other words, keep such usages as localized
> > as possible.
> 
> Rather, let Ada 2X formalize exception contracts and leave checking the
> contracts to the compiler.
> 
> The scenario you described is the valid and very useful use case of
> exceptions. The reason is that, likely, C and B could not handle the state
> when the exception was raised and propagated because in order to be able to
> handle it, they must be redesigned completely. They must have the
> information hidden from them for very good reason, affect program states
> possibly unrelated to them etc. I.e. it would make the design tightly
> coupled and fragile. The idea of exception is postponing handling of some
> [exceptional] states up to a context where handling were possible, e.g. in
> A. That context can be and usually is totally unknown to the designer of
> the code raising the exception.

I'm not sure we're on the same page.  What you're describing sounds more like the "normal" use case for exceptions, as I understand it (i.e. for "exceptional" conditions); but I thought the discussion was about something different, namely using exceptions as sort of a stack-unwinding "goto".  Even with an explicit contract, it still seems like using exceptions in this way, for non-exceptional situations, is going to impair readability if its use isn't kept within a small, local area of the source.  This is just my instinct, though, and I'm finding it difficult to explain why.  

                               -- Adam