a new, tiny, and fast, markdown parser
written in TypeScript under the unified umbrella
Something like remark (markdown processor powered by plugins based on
unified), but on a lower level: a lexer (in nerdy terms
We’re launching micromark as just an idea. The first line of code still needs to be written. But we imagine it to be:
- small in file size, max 10 kB minzipped, and tiny in memory use
- fast in speed, compared to existing parsers on real world documents
- safe to use, it should safely work on untrusted content by default
- compliant to CommonMark but extendible for GFM (GitHub Flavored Markdown), MDX, etc.
- complete, in that it should give access to all info in the source document
But it’s not:
- something that creates HTML and the like: other projects use micromark for that
- something that creates a syntax tree: remark will use it to do just that
How it’ll be used
micromark will be part of unified. But it will likely not be something you’d directly interact with, unless you’re interested in working on parsers, but it will make high-level tooling better.
Evolving unified shouldn’t just be about new high-level features but also about rethinking core mechanisms. That’s where micromark comes in, so unified can build the most friendly, secure, fast, and extensive bridges between content formats.