Cool, thanks. I haven't looked deeply into it, but from a quick glance it seems to want to transform the input.

For my use case I need a parse tree, with specific byte (well, character really) offsets. For example, I need to know where the link started, not that there is now a a link with text T and href H.

My program (and Macchiato I assume) doesn't actually transform the source to an output format, we add attributes to the source text (similar to RTF).