markdowny

A Markdown parser with strict error handling, generating an AST that can be easily adjusted and converted to text or HTML. It supports some additional syntax:

• spans between [ and ];
• divs without block elements (such as div and p) between [[ and ]];
• divs with block elements (such as div and p) between [[[ and ]]];
• literals between [[[[ and ]]]];
• setting styles/classes/ids etc with {#id; style: value; .classname}.
• HTML tags such as div with a newline after are assumed to contain block elements. HTML tags without a newline after are assumed to not contain block elements.
• !![description](image-src) is a block level image (not wrapped in <p>)

This library also has a resolve_markdown() function that does a number of things:

• makes sections around headings (h1,h2,..), and make links to them;
• highlight source code (if syntect feature is enabled);
• converts math to mathml (if latex2mathml feature is enabled);
• creates figures (start blockquote with figure_type #reference_key: caption, figure type can be listing/figure/etc, #reference_key can be used in links);
• resolves section and figure counters (start headings and figures with ?, if link body contains ?, replace by actual number).

Neovim LSP config

vim.api.nvim_create_autocmd('FileType', {
  pattern = 'markdown',
  callback = function(args)
    vim.lsp.start({
      cmd = { 'markdowny-lsp' },
      name = 'markdowny',
    })
  end,
})