RE: C++ intermediate representation.

I said:


> >As someone who has attempted to write a complete formal grammar
> >specification for C++ without jimmying and ad hockery -- I attest to
> >the fact that it's one nasty language to parse -- but a great test of
> >a grammar formalism!


Greg Comeau replied (in small part):


> If nothing else, a lot of fun :) After all, programming should be
> enjoyable, right?


Yes, certainly, and maybe even challenging. Hard to parse languages
(such as C++ and Perl) beat the heck out of a parsing engine. I'm glad
for languages like C++ and Perl when it comes to the challenges they
present to parsing technology. They push the envelope and convince me
that parsing is not the solved problem so many seem to assume it is.


Enhancements made specifically to deal with C++/Perl parsing issues
came in useful with parsing genetic pseudoknots in linear time, for
example. In 2000, a very small subset of C++, when parsed with an
adaptive engine, did so in exponential time, and now does it in linear
time, with a smaller constant than the LR(1) parser with hand
optimized lexical analyzer that someone sent me as a benchmark against
which to measure test runs. The optimizations made to the engine as a
result of that C++ parsing bench mark have had an overall effect
across the system.


It's either all a lot of fun -- or some of us are just quite
unimaginative in how we elect to pass our idle hours. ;-)


--
Chev. Quinn Tyler Jackson
http://members.shaw.ca/qjackson/

Return to the comp.compilers page.
Search the comp.compilers archives again.