Right now, a new pandoc process is created for every single structured comment section, which is very inefficient. If we corral the sections first, and pipe them through one pandoc process, it would probably be a lot better. But then there would have to be some way to mark which sections came from where, so we could parse out the individual section from the pandoc output.
Right now, a new pandoc process is created for every single structured comment section, which is very inefficient. If we corral the sections first, and pipe them through one pandoc process, it would probably be a lot better. But then there would have to be some way to mark which sections came from where, so we could parse out the individual section from the pandoc output.