Markdown info loader

Beta. Use on your own risk!

Develop a static blog? You can get all required information about markdown file (includes title, YAML metadata, author, editor and last update time) without loading whole markdown contents (or with it).

Output

Result of loader is object with advanced file information.

* - optional property.

• resource Information about target resource:
  • path Absolute path to file;
  • localPath Path relative to the project root;
  • hashName Unique hash number between 0 and 4294967295 (generated from localPath);
• meta* Document metadata:
  • heading* Document heading;
  • ...* Front matter metadata (see here and here)
• git* Information about file git state:
  • commits* File commits info;
    • initial* Initial commit info;
    • last* Last commit info;
    • all* All commits info;
• module* File module;

For example:

{
  resource: {
    path: '~/projects/library/src/items/hamlet.md',
    localPath: '/src/items/hamlet.md',
    hashName: 4294967295,
  },
  meta: {
    heading: 'Hamlet',
    author: 'William Shakespeare',
    year: '1600'
  },
  git: {
    commits: {
      initial: {
        an: 'admin',
        ae: '[email protected]'
        at: 1524365103,
      },
      last: {
        an: 'moderator',
        ae: '[email protected]',
        at: 1524365204,
      },
    }
  },
  module: Function
}

Options

parse.heading (bool)

Default: true

Parse markdown primary heading.

For example:

Hamlet
--

To be or not to be

Will give:

{
  meta: {
    heading: 'Hamlet'
  }
}

parse.frontMatter (bool)

Default: true

Parses front matter data. Result will be assigned in to meta property.

For example:

---
author: William Shakespeare
year: 1600
---
Hamlet
--

To be or not to be

Will give:

{
  meta: {
    author: 'William Shakespeare',
    year: '1600',
  }
}

More ditails here and here.

git (bool|object)

Default: true

Use git CLI to collect advanced information about file, such as author name, email, and create date;

Ensure that git is installed on machine on which your project builds and project root directory contains .git folder!

By the defaults, you'll get information (author name, author email and date) about initial commit and last commit.

{
  git: {
    commits: {
      initial: {
        at: "1522742493",
        ae: "[email protected]",
        an: "Ivan Ivanov"
      },
      last: {
        at: "1523235676",
        ae: "[email protected]",
        an: "Petr Petrov"
      }
    },
  }
  ...
}

But if you prefer to configure git collector in advanced mode, you should define git option as an object. That object may have next properties:

• placeholders (array) Default: [an, ae, at] Array of placeholders, supported by git log --pretty=format;
• initial (bool) Default: true Get properties of initial commit;
• last (bool) Default: true Get properties of last commit;
• all (bool) Default: false Get all commits (heavy);
• formatSep (string) Separator of placeholders (by default the separator is quite unique, but you can define your delimiter to avoid collisions)

Custom configuration will be deep mixed with the default configuration, thus shape like { all: false } will be correct.

importSource (object)

Use another loader to load source (will be placed into source property)

Thus you can at the same time load both file's metadata and its content.

{
  test   : /\.md$/,
  exclude: /node_modules/,
  use    : {
    loader : `markdown-info-loader`,
    options: {
      git: {
        last: false,
      },
      importSource: [
        `raw-loader`,
      ],
    },
  },
}

{
  heading: 'My article 1',
  commits: {
    initial: {
      at: "1522742493",
      ae: "[email protected]",
      an: "Ivan Ivanov"
    }
  },
  source: 'My article\n==Hello, world'
}

But this option is most convenient if you'd prefer to get file's metadata immediately, but show its content only at some special route. For this, you may use bundle-loader as an advanced loader in the option importSource. And in this case, metadata will be used for creating lightweight list of available markdown files, at the same time the source can be loaded asynchronously.

{
  test   : /\.md$/,
  exclude: /node_modules/,
  use    : {
    loader : `markdown-info-loader`,
    options: {
      importSource: [
        `bundle-loader`,
        `markdown-loader`,
      ],
    },
  },
}

{
  heading: 'My article 1',
  source: function bundle
}

parse.commonmark (bool)

Default: true

Enables remark-parse commonmark option (see ditails)

plugins (bool)

Default: undefined

Array of functions, which will be called in loader context.

The plugin is a function, that is called in loader context and has parameters info (result object) and ast (parsed from markdown). The result of a plugin can be a javascript code, which will be added between import block and result block.

License

MIT

Author

Vladimir Kalmykov [email protected]