[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Tinycc-devel] OT (Re: modern c++ compiler written in C)

From: sylware
Subject: Re: [Tinycc-devel] OT (Re: modern c++ compiler written in C)
Date: Sat, 16 May 2015 13:30:09 +0000
User-agent: Mutt/1.5.23 (2014-03-12)

--- Begin Message ---
On Fri, May 15, 2015 at 09:53:23PM +0200, address@hidden wrote:
> On Fri, May 15, 2015 at 07:32:02PM +0000, address@hidden wrote:
> > > Historical research - yes. Any practical application -
> > > only at one's own risk, left to the discretion of AT&T to wipe one out
> > > of existence or not.
> > 
> > Due to this legal issue, cfront is not worth it.
> > 
> > Any other alternative?
> cfront showed us a usable C++ dialect and also a proof that such a
> dialect can be efficiently implemented by a translator to C.
> This means that the effort to recreate such a translator from scratch
> - would be useful, as a compact and extremely fast compiler
>   for a practical subset of C++
> - is realistically doable
> This is my feeling about the best or the only possible way to
> combine many of the attractive features of tcc and C++/OOP.
> An easier but much less certain way would be to try to talk to someone at
> AT&T - this would be certainly a lot more efficient, _if_ it could work.

Due to this legal issue, and since there is no alternative, there is only one
reasonable choice: write a new "modern c++" to C89/99 translator from scratch
(can drive anybody sane to madness).

Without that, since critical apps uses "modern c++", and with the loop "modern
c++ compiler" compiling only with "modern c++ compiler", tinycc and any other
c-only toolchains are doomed on the medium term.

I would rather have those c++ apps ported to C and get rid of that insanity
which is c++ for good from my systems since the trade-off between its
complexity and what "features" it brings would be a definitive no-no (most of
the time, there is no significant benefits to get away from the kernel
language)... but unfortunately it seems less reasonnable than getting such a


--- End Message ---

reply via email to

[Prev in Thread] Current Thread [Next in Thread]