You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, html pages are simply fed to mustache. md pages are first fed to unified, which runs them through a series of plugins, and then the output html is fed to mustache.
This can create weird behaviors. For example, if a page is written in md, then its heading will get automatic ids for in-document references. If the same page is in html, they won't.
Potential solutions:
do nothing, this is fine
run html pages through unified in a slightly different pipeline (rehype-parse 👉 rehype plugins 👉 rehype-stringify). The remark plugins wouldn't take effect (and I don't think they should, if something can be done on either the mdast or the hast it should be done on the hast) but the rehype plugins would.
decide that ALL pages should be markdown. html should be only for templating.
what about templates that have separate content "slots"?
The text was updated successfully, but these errors were encountered:
Currently, html pages are simply fed to mustache. md pages are first fed to unified, which runs them through a series of plugins, and then the output html is fed to mustache.
This can create weird behaviors. For example, if a page is written in md, then its heading will get automatic ids for in-document references. If the same page is in html, they won't.
Potential solutions:
The text was updated successfully, but these errors were encountered: