[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]

Re: Table.test

I just rememberd this, maybe it'll help...

michael writes:
> markm writes:
> > Yes, it would be nice to have a more "semantic" compare that could,
> > e.g., ignore the ordering of Set elements.  The only idea I have here
> > is to have Set::printOn print the elements in a canonical sorted
> > order.  This would be similar in philosophy to Michael's hex_point
> > function. 
> I don't see any other easy way, either.
> 	michael

When the output is line oriented, there is a quick way to get a first
order aproximation if it is identical, even if blocks of the text are
scrambled relative to each other.

Simply 'sort' both files before running them through diff!  If they
fail, at least you know where to look, and if it passes it gives a
somewhat better feeling of security than pure eye-balling.

Naturally many things might interfere and make a test like this
useless, but, as I said, it's only a first check.