metalsmith-pure-text

A Metalsmith plugin that extracts plain text without HTML tags, Markdown syntax, etc.
It acts as a wrapper for textract and uses its capabilities for handling different file formats.

• Author: Sascha Zarhuber
• GitHub: @saschazar21
• Twitter: @saschazar
• Source: https://github.com/saschazar21/metalsmith-pure-text
• Issues: https://github.com/saschazar21/metalsmith-pure-text/issues
• Releases: https://github.com/saschazar21/metalsmith-pure-text/releases

Prerequisites

• Node.js >= v6
• Metalsmith

Installation

npm install metalsmith-pure-text

Options

The plugin takes an object containing different textract options - please look them up at textracts GitHub repository.

Additionally, some custom options are possible:

encoding

It can be either base64 or html encoded, to provide a sane string without any unwanted special characters:

{
  "encoding": "base64"      // ...or "html" for HTML-encoded strings
}

pattern

It can be configured to filter specific file suffixes by providing a pattern option:

{
  "pattern": "**/*.html"    // string for one pattern only, array for multiple patterns
}

upperCase/lowerCase

The outcome may be modified to contain only uppercase or lowercase letters. This can be done by adding only one key to the configuration:

{
  // either:
  "lowerCase": true
  // or:
  "upperCase": true
}

Usage

Usage is possible via JavaScript API and via Metalsmith CLI:

JavaScript API

Pass the plugin to the Metalsmith instance using its .use() function:

const metalsmith = require('metalsmith');
const pureText = require('metalsmith-pure-text');

metalsmith.use(pureText({
  pattern: ['**/*.html'],   // The globbing pattern you want to use. Single pattern also in array.
  preserveLineBreaks: true, // textract option: preserve line breaks in extracted text.
  // (...)
}));

Metalsmith CLI

Similar to the JavaScript API, the plugin may be used as follows:

{
  "plugins": {
    "metalsmith-pure-text": {
      "pattern": ["**/*.html"],
      "preserveLineBreaks": true
    }
  }
}

Results

After the text extraction is done, the outcome is stored in the text property of the respective file object, so one may be using it in templates similar like this (handlebars-example):

So you want to see the pure text contents of your file?
Here they are:

<pre>
  {{text}}
</pre>

Issues

Please report any bugs or issues to the issues section.

Contribution

Contributors welcome!
Please fork this repository, open a pull request and drop me a line on twitter.

Credits

• @dbashford for textract
• @sindresorhus for multimatch

License

MIT

Milestones

• Support caching
• Handle image data

Changelog

• v1.2.0 - Added encoding functionality to encode the outcome either in base64 or html-compatible strings.
• v1.1.0 - Added lowerCase/upperCase functionality, removed whitespace from text outcome.
• v1.0.0 - Initial version