This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC 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: Converting GCC to compilation with C++


Ben Elliston wrote:
Robert Dewar <dewar@gnat.com> writes:

By bootstrapping the compiler via a subset-of-C++ front-end that
wouldn't bootstrap if you deviated outside of that subset?  ;-)

You are assuming that the set of restrictions can be formulated in a manner that is easily checkable.

After all, we also have a coding standard requiring clear readable
code, but unfortunately we do not have a "subset-of-C" front end
that enforces this restriction :-)

As an example of this difficulty, consider a (reasonable) rule
that says that functions must from a conceptual point of view
be side effect free.

That's not easy to check in a way that allows side effects that
do not violate the "conceptual" part of this rule. For example,
memoizing tables, or a counter that counts the number of times
a function is called.

Ben




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