https://github.com/conversation/upmark

A HTML to Markdown converter.
https://github.com/conversation/upmark

converter html markdown peg ruby

Last synced: about 2 months ago
JSON representation

A HTML to Markdown converter.

• Host: GitHub
• URL: https://github.com/conversation/upmark
• Owner: conversation
• License: mit
• Created: 2011-09-21T01:09:29.000Z (about 13 years ago)
• Default Branch: main
• Last Pushed: 2024-04-21T22:53:16.000Z (5 months ago)
• Last Synced: 2024-07-20T10:50:32.188Z (2 months ago)
• Topics: converter, html, markdown, peg, ruby
• Language: Ruby
• Homepage: Upmark has the skills to convert your HTML to Markdown.
• Size: 114 KB
• Stars: 28
• Watchers: 14
• Forks: 5
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGELOG.md
  • License: LICENSE.md

Awesome Lists containing this project

README

        
# Upmark

A HTML to Markdown converter.

## Installation

    > gem install upmark

## Usage

In ruby:

```ruby

require "upmark"

html = "
messenger bag skateboard
"

markdown = Upmark.convert(html)

puts markdown

```


From the command-line:

    > upmark foo.html

You can also pipe poorly formatted HTML documents through `tidy` before piping them into `upmark`:

    > cat bar.html | tidy -asxhtml -indent -quiet --show-errors 0 --show-warnings 0 --show-body-only 1 --wrap 0 | upmark

## Features

Upmark will convert the following (arbitrarily nested) HTML elements to Markdown:

* `strong`

* `em`

* `p`

* `a`

* `h1`, `h2`, `h3`, `h4`, `h5`, `h6`

* `ul`

* `ol`

* `br`

It will also pass through block and span-level HTML elements (e.g. `table`, `div`, `span`, etc) which aren't used by Markdown.

## How it works

Upmark defines a parsing expression grammar (PEG) using the very awesome [Parslet](https://github.com/kschiess/parslet/) gem. This PEG is then used to convert HTML into Markdown in 4 steps:

1. Parse the XHTML into an abstract syntax tree (AST).

2. Normalize the AST into a nested hash of HTML elements.

3. Mark the block and span-level subtrees which should be ignored (`table`, `div`, `span`, etc).

4. Convert the AST leaves into Markdown.