"Arkadiy Vertleyb" <vertleyb@hotmail.com> wrote in message news:cqhqvc$f40$1@sea.gmane.org...
"Andy Little" <andy@servocomm.freeserve.co.uk> wrote
in upgrading the documentation. Among the bones of contention in a review would be the type deduction mechanism, which is a major part of the library,eg a substitute for current lack of decltype, auto typeof etc. ( I dont think it will be possible to do units really cleanly until C++ has this in place in the language).
I remember you once considered using the BOOST_TYPEOF facility. Have you decided against using it after all?
No I simply havent tried it out with pqs. I apologise as I said that I would try it in pqs out some months ago. I will try to spend some time on it and get back to you. However there are various aspects to type deduction and I dont expect BOOST_TYPEOF to cover all of them. I think it is more of a user mechanism. rather than an implementation mechanism. ( Well for me as macros cause me error finding problems... maybe this is a user issue too.) pqs requires a fair amount of template machinery in the implementation ( and I think some might be needed in pqs even with mods to the language). regards Andy little