Robert Roessler:
I would rather that non-legal constructions in C/C++ not appear to be something they aren't. ;)
OK, committed, but you'll have to turn that on with lexer.cpp.allow.dollars=0 now. To make implementing this easier there is a new SetOfCharacters class that checks by indexing into a boolean array. It can cope with including or excluding everything over a particular value (normally 0x80) and should handle Unicode character values > 255 if given those. The class is currently just in LexCPP.cxx but if it works well it will be moved into a header for other lexers to use. I wouldn't mind squeezing an extra bit of speed out of it by using C++'s template specialization to compile a version that knows whether values past the size position should be assumed true or false so Contains becomes (val < size) && bset[val] or (val >= size) || bset[val], assuming these are faster than (val < size) ? bset[val] : valueAfter. Neil _______________________________________________ Scintilla-interest mailing list [email protected] http://mailman.lyra.org/mailman/listinfo/scintilla-interest
