plgg-md

A Markdown-to-typed-data parser, built from scratch on plgg. A layout-marker frontmatter splitter and a block tokenizer for the plggpress Markdown subset produce an immutable Box-union AST — Result, never a throw. It underpins plggpress content parsing.

Why it exists

plggpress needs Markdown as data it can render through plgg-view, not as an opaque HTML string. plgg-md parses source text into a typed tree so the render step is a pure fold:

plgg ── plgg-md ── plggpress
          └── renders to plgg-view Html

How it's organized

• Frontmatter — the layout-marker splitter that peels a page's frontmatter off its body.
• Block — the block tokenizer that turns the body into the Box-union AST (headings, lists, fenced code, …).
• Inline — inline-span tokenizing within a block.
• Render — folds the AST into a plgg-view Html tree.

Because the AST is a Box union and every step returns a Result, malformed input is a value to handle, not an exception. The exact node types and the tokenizer's grammar live in the plgg-md source.