Re: RFC: implementation of package pools
Anthony Towns wrote:
Eray, if you want to help, help. If you want to randomly pester people,
do it to a lecturer who's actually paid to answer your questions.
I'm trying to figure out if the code should be improved, isn't that help?
The above translations are simple and automatable, the simplified form
you've decided to use may be equivalent, but it's not particularly
relevant to either the code or the original problem.
Yes, "a & (~a | b)" is the same as "a & b". No, that's not helpful in the
slightest for this problem
No, I just showed how the translation should go. If that's hardwired to
the code,
which might be IIRC, of course it can remain that way.
Anyhow, you haven't written your opinion about the second potential problem
with the code. Does it converge to exhaustive search in the worst case?
I think
you should know that since you've designed the code. I don't know if you
read
all of my previous mail which I tried to put only correct things and
helpful comments.
I believe that even if you haven't perhaps you might want to check the
last part
of that mail. It's summarized as follows:
We should make sure that the search is complete. That's correctness
requirement.
Performance can be left as future work because most sentences are easy.
That's my review of the code which you requested. Still, it should be
clarified
whether the search if complete, then a fix may be in order. If there's
anything
else I can help, let me know. Especially data structure / algo. stuff.
Thanks,
__
Eray Ozkural
Reply to: