This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: new extension: boyer-moore search algorithm variants


Benjamin Kosnik wrote:

I guess I must have missed this in my mailer confusion. I don't mean to
be dim, so perhaps you can restate or provide a pointer to me about
where this was discussed?


Recently, maybe in private email with Dhruv. Anyway, there isn't much to add, the issue is trivial, the algorithm obviously isn't viable if the range of values is too big.

Why a new trait at all then, and not just sizeof(T) == 1? Isn't there
another compile-time way to do this that would sidestep this issue
entirely?


I don't know, sizeof came to my mind, but I'm not sure it guarantees that we have the right, very simple, properties for the type, I think PODness is already too much, but I'd like to be wrong, because in that case is_pod + sizeof would do the job.

Paolo.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]