TypeScript Markdown parser

Downloads in past


1051.1.34 years ago6 years agoMinified + gzip package size for flavormark in KB


Based on commonmark.js
FlavorMark is a Markdown parser, made in TypeScript, with emphasis placed on modularizing element parsing logic, making it easy(-ish) to add/remove syntax for your favorite Markdown flavors.


There are a lot of classes with the word Parser in them.
  • Parser

The main parser, parses block elements. Then, parses inline content.
  • BlockParser

Each block element will generally be parsed by one BlockParser.
  • InlineContentParser

Parses inline content, using InlineParser.
  • InlineParser

Each inline element will generally be parsed by one InlineParser.
  • DelimitedInlineParser

This parser makes it easier to parse inline elements that behave like emphasis elements.
  • DelimitedInlineSubParser

If using DelimitedInlineParser, each sub-parser will parse one emphasis-like element. (Emphasis, smart quotes, superscript, strikethrough, etc.)


  1. Instantiate all BlockParser classes and add them to a BlockParserCollection.
  2. Instantiate all InlineParser classes and add them to an array.
  3. Instantiate an InlineContentParser and pass it the array of InlineParser.
  4. Instantiate a Parser and pass it the BlockParserCollection, and InlineContentParser.
  5. Call Parser.parse(str).

See the tests in src/test for more examples.


Each Node constructor is used to uniquely identify BlockParser, and HtmlSubRenderer instances.
When trying to parse a block, instanceof checks are used to find the BlockParser associated with a Node. The first matching BlockParser in the array is used.
For this reason, each unique block type should have its own Node sub-class.
See LICENSE file (copy-pasted from commonmark.js). Also, since this is just a glorified port of commonmark.js, I guess the BSD-license carries over. I have no idea how this works. I don't really care, personally. I just don't want to step on anyone's toes.