> First thing to note is that you leave out "typename" all over the > place. Try testing your examples with GCC-3.2.
Yeah, I noticed. I had the same problem when porting my own code from Intel to GCC. Strange that the strongly-conforming Intel compiler doesn't issue a warning for it. Fixed in the WIKI. > Second thing: you can save yourself lots of typenames by taking > advantage of the way MPL lets you pass an uninstantiated metafunction, > without getting its nested ::type member, wherever a bool_c<...> is > expected: I did not know this. I do remember a previous MPL discussion about automatic application of ::type but that was in a different context. I updated the WIKI, but I also added a question there about this. > Oh, your foo example seems to have too many "integral_c<"s in it: Fixed. > One more hint you should use as an item: "Use inheritance to forward > metafunction results". So: Awesome! I hadn't thought of this one before. Thanks!! And added to the WIKI. Small question: is struct inheritance public by default, or did you forget a 'public' there? > Lastly, I know that Aleksey will argue with me about this, but I have > a strong preference for "class" rather than "typename" in template > parameter lists. Haha, I did not add an item to the WIKI, because I don't think it's good to promote something if even the MPL itself doesn't use it. However, you did convince me, mostly because it allows me to do easier global-search-and-replaces for typename to BOOST_DEDUCED_TYPENAME :) hehe. Thanks, Jaap Suter _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
