2011年1月26日 07:30:09 UTC+9 Sean Bolton
<sean@smbolton.com>:
Hmm, we're not there yet. Your patch 5.1.4 yields the correct results
for these two cases:
Thank you to point out my mistake. I forgot that 'nil' is equivalence
to 'false' in the logical operation.
assert((((1 or false) and true) or false) == true)
assert((((1 or 2) and true) or false) == true)
assert(((nil and true) or false and true) == false)
assert((((nil and true) or false) and true) == false)
I fixed the patch to clear these cases. Is it enough test case for this problem?
Ryota Hirose