[bitc-dev] Opinions wanted: Infix shift operators

[wren ng thornton]

Jonathan S. Shapiro wrote:
> On Fri, Aug 6, 2010 at 8:31 PM, Ben Karel <eschew at gmail.com> wrote:
> 
>> Remove '\n' from the set of ignorable-whitespace tokens...
> 
> Personally, I think that's worse than making ';' mandatory or adopting the
> Haskell Layout rules. The essential point, I think, is that we're
> introducing an end-of-statement notion and debating how visually intrusive
> it should be. Something along these lines might be the right thing to do.
> 
> In fact, in the interests of honesty I should admit that if I had understood
> the Layout rules earlier, I might have adopted Haskell syntax from the
> beginning. Without them, the language is horribly ambiguous to parse, and
> not knowing what to look for at the time I failed to delve into the
> specification deeply enough.

To what extent is it desirable/necessary that the grammar allows 
expressions as statements?

If we had a syntax for marking expressions that should stand alone as 
statements (instead of allowing expressions to be statements by a unary 
rule in the grammar), then you could remove the ambiguity by looking for 
the extra syntax.

-- 
Live well,
~wren