npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

comark-knap

v0.1.0

Published

Comark plugin that renders knap templates (variables, filters, if/for logic) before parsing

Readme

comark-knap

A Comark plugin that renders knap templates before parsing. Write Markdown with knap variables, filters and if / for logic, hand it to Comark with your data, and get a regular Comark document out.

---
title: "{{ name | title }}"
---

# {{ title }}

{% if tags %}Tags: {{ tags | join:", " }}{% endif %}

{% for post in posts %}
- [{{ post.title }}]({{ post.url }})
{% endfor %}

{{ stats | table }}
import { parseMarkdown } from 'comark'
import knap from 'comark-knap'

const tree = await parseMarkdown(content, {
  plugins: [
    knap({
      variables: {
        name: 'an imported note',
        tags: ['reading', 'reference'],
        posts: [{ title: 'One', url: '/one' }],
        stats: [{ metric: 'users', value: 1200 }],
      },
    }),
  ],
})

tree.frontmatter // { title: 'An Imported Note' }
tree.nodes       // h1, p, ul, table … — plain Comark nodes
tree.meta.knap   // { errors: [], warnings: [] }

Install

pnpm add comark-knap

comark is a peer dependency (>= 0.4); knap is bundled as a dependency.

How it works

Comark plugins have a pre hook that runs on the raw Markdown string before tokenization. This plugin passes that string through knap's engine and hands the rendered Markdown back to Comark, so:

  • Every knap feature works: chained filters, if / elseif / else, for, set, {# comments #}, nullish fallbacks, whitespace control, and all standard filters, including the Markdown producing ones (table, list, callout, yaml_property, …).
  • The output is ordinary Markdown, so it flows through Comark's component syntax, attributes, code highlighting and every renderer unchanged.
  • Templating happens once, at parse time. The rendered document is static; nothing is re-evaluated when the renderer's data changes.

Compared to Comark's binding plugin

Both use {{ … }}, but they answer different questions.

| | comark-knap | comark/plugins/binding | | --- | --- | --- | | When it runs | Parse time (pre hook) | Render time, in every renderer | | Input | Any value you pass as variables (plus frontmatter) | frontmatter, data, meta, props namespaces | | Expressions | Full knap: filters, comparisons, and / or, ?? | A dot-path and an optional \|\| default | | Logic | {% if %}, {% elseif %}, {% for %}, {% set %} | ::if{:value="…"} component (+ ::for on main) | | Output | Markdown text, then parsed | binding / if nodes resolved by the renderer | | Reactive to data changes | No | Yes (Vue, React, Svelte, Angular) | | Round-trips through renderMarkdown | No (already rendered) | Yes, via the Binding handler |

Use knap when the data is known when you parse (build pipelines, importers, CLI output, "turn this JSON into a note"). Use binding when values come from the running app and must update live.

Using both

Set keepUnresolved and register knap() before binding(). Paths rooted at a name the template does not know (not a variable, frontmatter key, for iterator or set target) are left in the source as {{ name }}, and the binding plugin turns them into binding nodes for the renderer. Pass a list such as ['data', 'props'] to name the roots explicitly instead:

const tree = await parseMarkdown('# {{ title | upper }}\n\nHello {{ data.user.name }}!', {
  plugins: [knap({ variables: { title: 'Docs' }, keepUnresolved: true }), binding()],
})
// ['h1', { id: 'docs' }, 'DOCS'],
// ['p', {}, 'Hello ', ['binding', { ':value': 'data.user.name' }], '!']

Two things to keep in mind when mixing them:

  • knap parses || as boolean or, so the binding default syntax {{ path || fallback }} is evaluated by knap (usually to false). Use knap's {{ path ?? "fallback" }} instead, or keep such paths out of knap's reach.
  • Filters still apply to kept variables: {{ data.x | upper }} becomes {{ DATA.X }}.
  • Conditions on kept variables ({% if data.flag %}) see the placeholder string, which is truthy. Decide such branches with ::if at render time.

Frontmatter

By default (frontmatter: true) the plugin:

  1. exposes the document's frontmatter keys as top-level template variables (explicit variables win on conflicts), and
  2. runs the template over the frontmatter block itself, then re-reads it so tree.frontmatter and the body both see the rendered values.

Since Comark 0.7 the built-in frontmatter plugin parses the YAML block before user plugins run. Unquoted knap tags are not valid YAML (title: {{ x }} is a YAML flow mapping), so:

  • Quoted values work out of the box: title: "{{ name | title }}".

  • For knap-style frontmatter ({{ year | yaml_property:"year" }}, block output, …), register Comark's own frontmatter plugin after knap(). Comark moves it behind yours, so knap sees the raw block first:

    import frontmatter from 'comark/plugins/frontmatter'
    
    await parseMarkdown(source, {
      plugins: [knap({ variables }), frontmatter()],
    })

If the rendered frontmatter is not valid YAML, the original block is kept and a RENDER_ERROR is reported (see below). Pass frontmatter: false to leave the block alone and hide it from the template.

Errors and warnings

knap reports structured diagnostics (code, message, line, column). The plugin collects them under tree.meta.knap:

tree.meta.knap
// {
//   errors:   [{ code: 'PARSE_ERROR', message: 'Missing closing }}', line: 3, column: 12 }],
//   warnings: [{ code: 'INVALID_FILTER_INPUT', filter: 'date', … }],
// }

When a template has errors, its source is left untouched so Comark still parses the document. That makes the plugin safe in streaming mode, where a {% if %} may not be closed yet. Set strict: true to throw a TemplateRenderError instead.

Options

| Option | Type | Default | Description | | --- | --- | --- | --- | | variables | TemplateVariables \| (ctx) => TemplateVariables | {} | Values for the template. A function receives { frontmatter, markdown } and may be async. | | frontmatter | boolean | true | Expose frontmatter as variables and template the frontmatter block. | | filters | FilterRegistry | standardFilters | Filter registry for the engine. Spread standardFilters to add your own. | | engine | TemplateEngine | — | Bring your own createEngine() result; overrides filters, limits, allowRegex. | | context | TContext | — | Host context forwarded to custom filters and resolveVariable. | | resolveVariable | VariableResolver | — | Resolve variables missing from variables and frontmatter, lazily and possibly async. | | keepUnresolved | boolean \| string[] | false | Leave variables rooted at unknown names (or at the listed names) as {{ name }} for a render-time layer such as binding. | | strict | boolean | false | Throw on template errors instead of reporting them in tree.meta.knap. | | limits | RenderLimits | knap defaults | Forwarded to createEngine. | | allowRegex | boolean | true | Forwarded to createEngine; false makes split / replace literal. | | enabled | boolean | true | Set to false to skip the plugin. |

createEngine, standardFilters, TemplateRenderError and the knap types are re-exported for convenience.

Custom filters

import knap, { standardFilters } from 'comark-knap'
import type { TemplateFilter } from 'comark-knap'

const shout: TemplateFilter = (value) => `${value}!`

knap({ filters: { ...standardFilters, shout } })

DOM-dependent filters (html_to_json, remove_html) live in knap/html and can be spread in the same way where a DOM is available.

Caveats

  • knap is a text template engine and does not know Markdown: {{ … }} inside fenced code blocks is rendered too. Escape it with knap's {# #} comments or a {% set %} literal when you need the raw syntax in output.
  • Unrendered knap tags left in the document (after an error) meet Comark's attribute syntax: text{% if x %} is parsed as an attribute block. Check tree.meta.knap.errors when output looks off.
  • Rendering is limited by knap's defaults (template size, output size, work operations). Tune them with limits.

Development

pnpm install
pnpm play      # run playground/index.ts in Node
pnpm play:nuxt # docs + playground site in the browser (playground/)
pnpm test      # vitest
pnpm build     # tsdown → dist/

The Nuxt site serves this README at / and a playground at /play, both server-rendered and prerendered. The playground renders a knap template with @comark/nuxt, shows the Markdown knap generated, and keeps a {{ data.* }} line for the binding plugin so you can see both layers side by side.

Every push and pull request publishes a preview build to pkg.pr.new, so a branch can be tried with pnpm add https://pkg.pr.new/comark-knap@<sha> before it is released.

Release

pnpm release

release-it runs the type check and tests, bumps the version, tags, creates the GitHub release and publishes to npm. Run it from a clean main checkout with GITHUB_TOKEN set and an npm login.

License

MIT