On Monday 15 September 2008 07:56:13 pm Robert Dodier wrote:
> Sets are represented internally as set(1, 2, 3). The conversion from the
> so-called matchfix notation {1, 2, 3} to set(1, 2, 3) happens in
> evaluation, but the expected results in a test script are not evaluated.
> So the comparison is between two expressions with different operators.
>
> The code to translate {} to set() can be executed by the parser, so that
> it happens before evaluation or simplification. Other operators are handled
> that way so it makes sense that {} should be also. I'll take a look at
> that.
I'll have read about it... don't understand matchfix yet.
I think I ran into some similar problems with some other
tests not involving sets, maybe I can find them. Anyway, a
satisfactory workaround so far has been 'is ( a = b)' where
a evaluates to b.
Thanks,
John