evelez7 wrote: I've rewritten a large part of the initial implementation idea. The biggest change is that I've separated the Markdown AST, parser, and HTML generation (which doesn't exist here yet). The parser is independent of the AST. The AST owns the arena and string saver. The motivation behind the separation is that a single parser object could be reused for different ASTs in the future or several ASTs could be parsed in parallel. Then, these ASTs could be fed to different engines and not just HTML.
There's still a lot of work to be done here but I thought some initial feedback on the design direction could be helpful, especially the problems with arena allocations and memory leaks that I had last time. https://github.com/llvm/llvm-project/pull/155887 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
