> People seem to reimplement C++ compilers, > despite there being an enormous amount of complex just parsing it
Realistically though, how many implementations are libre and complete enough to actually build a C++ compiler? On 13 September 2013 10:51, Niko Matsakis <[email protected]> wrote: > People seem to reimplement C++ compilers, despite there being an > enormous amount of complex just parsing it... that said, the trickiest > and least specified part of the type checker right now is probably the > type inferencing algorithm, which I hope we can overhaul for something > that is clearer or more easily specified. > > > Niko > > On Wed, Sep 11, 2013 at 08:50:39AM -0400, Corey Richardson wrote: >> On Wed, Sep 11, 2013 at 3:27 AM, Gregory Maxwell <[email protected]> wrote: >> > Presumably before rust reaches a point of earth-shaking importance >> > there will be an second implementation which can compile the first >> > compiler, thus permitting this solution. :) >> >> I wouldn't be so sure of that. There's a ton of trickiness around the >> type and borrow checker and everything else, and I'm not sure anyone >> would *want* to reimplement it. What gain would there be? >> _______________________________________________ >> Rust-dev mailing list >> [email protected] >> https://mail.mozilla.org/listinfo/rust-dev > _______________________________________________ > Rust-dev mailing list > [email protected] > https://mail.mozilla.org/listinfo/rust-dev _______________________________________________ Rust-dev mailing list [email protected] https://mail.mozilla.org/listinfo/rust-dev
