MarkdownDB is a javascript library that turns markdown files into structured queryable databaase (SQL-based and simple JSON). It helps you build rich markdown-powered sites easily and reliably. Specifically it:

• Parses your markdown files to extract structured data (frontmatter, tags etc) and builds a queryable index either in JSON files or a local SQLite database
• Provides a lightweight javascript API for querying the index and using the data files into your application

• Index a folder of files - create a db index given a folder of markdown and other files
  • Command line tool for indexing: Create a markdowndb (index) on the command line v0.1
  • SQL(ite) index v0.2
  • JSON index v0.6
  • BONUS Index multiple folders (with support for configuring e.g. prefixing in some way e.g. i have all my blog files in this separate folder over here)
  • Configuration for Including/Excluding Files in the folder

Extract structured data like:

• Frontmatter metadata: Extract markdown frontmatter and add in a metadata field
  • deal with casting types e.g. string, number so that we can query in useful ways e.g. find me all blog posts before date X
• Tags: Extracts tags in markdown pages
  • Extract tags in frontmatter v0.1
  • Extract tags in body like #abc v0.5
• Links: links between files like [hello](abc.md) or wikilink style [[xyz]] so we can compute backlinks or deadlinks etc (see #4) v0.2
• Tasks: extract tasks like this - [ ] this is a task (See obsidian data view) v0.4

Data enhancement and validation

• Computed fields: add new metadata properties based on existing metadata e.g. a slug field computed from title field; or, adding a title based on the first h1 heading in a doc; or, a type field based on the folder of the file (e.g. these are blog posts). cf https://www.contentlayer.dev/docs/reference/source-files/define-document-type#computedfields.
• 🚧 Data validation and Document Types: validate metadata against a schema/type so that I know the data in the database is "valid" #55
  • BYOT (bring your own types): i want to create my own types ... so that when i get an object out it is cast to the right typescript type

For example, your blog posts. Each file can have a YAML frontmatter header with metadata like title, date, tags, etc.

---
title: My first blog post
date: 2021-01-01
tags: [a, b, c]
author: John Doe
---

# My first blog post

This is my first blog post.
I'm using MarkdownDB to manage my blog posts.

Use the npm mddb package to index Markdown files into an SQLite database. This will create a markdown.db file in the current directory. You can preview it with any SQLite viewer, e.g. https://sqlitebrowser.org/.

# npx mddb <path-to-folder-with-your-md-files>
npx mddb ./blog

E.g. get all the files with with tag a.

SELECT files.*
FROM files
INNER JOIN file_tags ON files._id = file_tags.file
WHERE file_tags.tag = 'a'

Use our Node API to query your data for your blog, wiki, docs, digital garden, or anything you want!

Install mddb package in your project:

npm install mddb

Now, once the data is in the database, you can add the following script to your project (e.g. in /lib folder). It will allow you to establish a single connection to the database and use it across you app.

// @/lib/mddb.mjs
import { MarkdownDB } from "mddb";

const dbPath = "markdown.db";

const client = new MarkdownDB({
  client: "sqlite3",
  connection: {
    filename: dbPath,
  },
});

const clientPromise = client.init();

export default clientPromise;

Now, you can import it across your project to query the database, e.g.:

import clientPromise from "@/lib/mddb";

const mddb = await clientPromise;
const blogs = await mddb.getFiles({
  folder: "blog",
  extensions: ["md", "mdx"],
});

This feature helps you define functions that compute additional fields you want to include.

Next, define a function that computes the additional field you want to include. In this example, we have a function named addTitle that extracts the title from the first heading in the AST (Abstract Syntax Tree) of a Markdown file.

const addTitle = (fileInfo, ast) => {
  // Find the first header node in the AST
  const headerNode = ast.children.find((node) => node.type === "heading");

  // Extract the text content from the header node
  const title = headerNode
    ? headerNode.children.map((child) => child.value).join("")
    : "";

  // Add the title property to the fileInfo
  fileInfo.title = title;
};

Now, use the client.indexFolder method to scan and index the folder containing your Markdown files. Pass the addTitle function in the computedFields option array to include the computed title in the database.

client.indexFolder(folderPath: "PATH_TO_FOLDER", customConfig: { computedFields: [addTitle] });

• Implement computed fields to dynamically calculate values based on specified logic or dependencies.
• Specify the patterns for including or excluding files in MarkdownDB.

Here's an example markdowndb.config.js with custom configurations:

export default {
  computedFields: [
    (fileInfo, ast) => {
      // Your custom logic here
    },
  ],
  include: ["docs/**/*.md"], // Include only files matching this pattern
  exclude: ["drafts/**/*.md"], // Exclude those files matching this pattern
};

{
  "name": "my-mddb-app",
  "scripts": {
    ...
    "mddb": "mddb <path-to-your-content-folder>",
    "prebuild": "npm run mddb"
  },
  ...
}

For example, in your Next.js project's pages, you could do:

// @/pages/blog/index.js
import React from "react";
import clientPromise from "@/lib/mddb.mjs";

export default function Blog({ blogs }) {
  return (
    <div>
      <h1>Blog</h1>
      <ul>
        {blogs.map((blog) => (
          <li key={blog.id}>
            <a href={blog.url_path}>{blog.title}</a>
          </li>
        ))}
      </ul>
    </div>
  );
}

export const getStaticProps = async () => {
  const mddb = await clientPromise;
  // get all files that are not marked as draft in the frontmatter
  const blogFiles = await mddb.getFiles({
    frontmatter: {
      draft: false,
    },
  });

  const blogsList = blogFiles.map(({ metadata, url_path }) => ({
    ...metadata,
    url_path,
  }));

  return {
    props: {
      blogs: blogsList,
    },
  };
};

Retrieve a file by URL path:

mddb.getFileByUrl("urlPath");

Currently used file path -> url resolver function:

const defaultFilePathToUrl = (filePath: string) => {
  let url = filePath
    .replace(/\.(mdx|md)/, "") // remove file extension
    .replace(/\\/g, "/") // replace windows backslash with forward slash
    .replace(/(\/)?index$/, ""); // remove index at the end for index.md files
  url = url.length > 0 ? url : "/"; // for home page
  return encodeURI(url);
};

🚧 The resolver function will be configurable in the future.

Retrieve a file by it's database ID:

mddb.getFileByUrl("fileID");

Get all indexed files:

mddb.getFiles();

By file types:

You can specify type of the document in its frontmatter. You can then get all the files of this type, e.g. all blog type documents.

mddb.getFiles({ filetypes: ["blog", "article"] }); // files of either blog or article type

By tags:

mddb.getFiles({ tags: ["tag1", "tag2"] }); // files tagged with either tag1 or tag2

By file extensions:

mddb.getFiles({ extensions: ["mdx", "md"] }); // all md and mdx files

By frontmatter fields:

You can query by multiple frontmatter fields at once.

At them moment, only exact matches are supported. However, false values do not need to be set explicitly. I.e. if you set draft: true on some blog posts and want to get all the posts that are not drafts, you don't have to explicitly set draft: false on them.

mddb.getFiles({
  frontmatter: {
    key1: "value1",
    key2: true,
    key3: 123,
    key4: ["a", "b", "c"], // this will match exactly ["a", "b", "c"]
  },
});

By folder:

Get all files in a subfolder (path relative to your content folder).

mddb.getFiles({ folder: "path" });

Combined conditions:

mddb.getFiles({ tags: ["tag1"], filetypes: ["blog"], extensions: ["md"] });

Retrieve all tags:

mddb.getTags();

Get links (forward or backward) related to a file:

mddb.getLinks({ fileId: "ID", direction: "forward" });

graph TD

markdown --remark-parse--> st[syntax tree]
st --extract features--> jsobj1[TS Object eg. File plus Metadata plus Tags plus Links]
jsobj1 --computing--> jsobj[TS Objects]
jsobj --convert to sql--> sqlite[SQLite markdown.db]
jsobj --write to disk--> json[JSON on disk in .markdowndb folder]
jsobj --tests--> testoutput[Test results]