Currently, all tags have to be listed in the Tags.rakumod file - however, there are very obvious examples of missing tags - most HTML5 tags, for example.
I failed to find anything authoritative about HTML tags, here are some possibilities and considerations:
- run a development test (under xt/) against some scraped site that contained all the tags supposedly, to see if there are missing ones
- programmatically generate tag classes during the build phase
- ditch the idea of separate class per tag
- I tend to like the idea that matching tags could have extra operations that mixed tags couldn't, still I wonder how practical that would be
Currently, all tags have to be listed in the Tags.rakumod file - however, there are very obvious examples of missing tags - most HTML5 tags, for example.
I failed to find anything authoritative about HTML tags, here are some possibilities and considerations: