erik.pilkington added a comment. > Looks like this demangler's design is similar to my demangler for Microsoft > name mangling scheme (https://reviews.llvm.org/D34667). Is that a > coincidence? Both demanglers create AST, stringize it using > print_left/print_right (I named them write_pre/write_post), and use custom > memory allocator. Looks like both demangler can share more code once both > land.
Yep, that was coincidental. I glanced at your patch and I think we could end up sharing some code here, which would be really neat. > Do you think you can avoid STL containers so that your demangler don't need > any destructors? I observed that that makes my demangler faster. The AST doesn't have destruction now for this reason, I used a bump pointer to allocate the AST. Looks like your patch followed this strategy too! https://reviews.llvm.org/D35159 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
