1 Sep
2007
1 Sep
'07
6:31 p.m.
David Abrahams wrote:
It has often seemed to me that it might make sense to parse C++ nondeterministically, just to avoid some of these issues. The number of real instances of ambiguity is probably pretty small.
Anyway, I should probably take this over to the LLVM list...
Agreed, and there has been done some work on this already: http://www.computing.surrey.ac.uk/research/dsrg/fog/, which seems to be promising. Parsing here doesn't have to deal with ambiguities (the parser understands some superset of C++). These are resolved later during semantic analysis. Regards Hartmut