Re: Pike 8.0 RC1

On Tue, Oct 28, 2014 at 7:18 PM, Stephen R. van den Berg <[email protected]> wrote:
> Zenothing wrote:
>>o Added new syntax for literal-string constants
>
>>  #{, #( and #[ starts a literal string, and it is ended by the corresponding
>>  end marker: #}, #) and #] respectively.
>
> I agree that it's a pity that this way you only have three fixed ones.
> SQL solved this differently.
>
> I'd indeed suggest that the SQL way is more elegant.  It would amount
> to something like the following Pike equivalent:
>
> #{ is terminated by #}
> #abc{ is terminated by #abc}
>
> etc.  All other rules stay the same.  It will support unlimited nesting.
> Supporting #( and #[ as well is in a similar manner of course is possible.

Presumably anything that's actually a preprocessor command would be
illegal for that - it'd be awkward if you go #string{    #string}.

The question is, do we actually need more than three? Is the
flexibility all that useful?

>>o Added the '?:' operator for compatibility with other C-like languages. It
>>is identical to '||' in pike.
>
> I guess a?:b is shorthand for a?a:b.  I'm curious though, which languages
> already have it?  I tried to google for it, but that turnes out to be
> difficult.

Yes, it is. It's a GCC extension:

http://scottmoonen.com/2008/05/27/gcc-binary-conditional/

Googled that as "binary question colon", not one of the most common
search terms I am sure!

ChrisA