majas

Markdown and JSON Are Similar.
All data is hierarchical if you look at it long enough.

majas is a format-agnostic structured data converter. It lets you translate between hierarchical formats like Markdown, JSON, and directory trees — not by trying to preserve syntax, but by capturing the underlying shape.

Whether you're trying to break up a massive Markdown doc, reverse-engineer a messy JSON export, or treat your filesystem like a structured document — majas gives you tools to traverse, transform, and rematerialize hierarchical data.

Philosophy

• Structure > Syntax
  Markdown, JSON, XML, folders — all express trees. majas extracts the hierarchy and leaves the quirks behind.

• Not lossless, not perfect, not magic
  majas makes opinionated conversions, focused on exploration and manipulation, not round-trip fidelity.

• One simple internal model
  All formats are parsed into a common IR (intermediate representation): a minimal tree of nodes with optional titles, contents, and children.

• CLI-first, API-ready
  The core logic is a clean TypeScript library. The CLI is just a friendly wrapper.

Installation

npm install -g majas
# or just use via npx
npx majas ...

Usage

majas [-i] [-4 from] [-2 to] [-o output] [-ioption...] [-ooption] [FILE]

If no input file is provided, data is read from stdin.

Output is always written to stdout unless --out is specified.

Options

| Flag | Alias | Description | | ----------------- | ----- | ----------------------------------------------------------------------------------------------------------------------- | | --from <format> | -4 | Source format. Optional if file extension is unambiguous. | | --to <format> | -2 | Target format. If absent, Majas simply outputs the resolved input format and no conversion is performed. | | --in-* <value> | -i* | Format-specific input options (e.g. --in-indent 4, --in-encoding ascii) | | --out-* <value> | -o* | Format-specific output options | | --out <file> | -o | Output file (defaults to stdout) | | --help [FORMAT] | -h | Print usage help (including list of supported formats). If format is specified: explain format, print available options | | --infer | -i | Infer input format from file extension; treat --from as a fallback on ambiguous input | | --version | -V | Show program version. | | --help | | Show help. |

Examples

Convert a Markdown doc into a directory of smaller .md files

majas notes.md --from md --to fs

Flatten a deeply nested JSON structure into Markdown

majas data.json --from json --to md

Turn a folder of text files into a single Markdown document

majas ./docs --from fs --to md

Convert from file (format inferred)

majas -2 json mydoc.md > output.json

Convert from stdin (must specify --from)

majas -4 xml -2 markdown < input.xml

With format-specific options

majas -4 json -2 markdown --out-table-style compact input.json

Retrieve the inferred format of a file

$ majas --infer input.xml
XML

How it works

• -4 means you're using majas "for" a format.
• -2 means you're converting "to" a format.
• If --from is not provided, majas will try to infer the input format from the file extension. Ambiguity is an error.
• If you pass stdin, --from is required.
• All input is parsed into a generic intermediate representation (IR).
• All output is generated from that IR — so conversion is always format → IR → format.

Notes

• Majas is format-agnostic, language-agnostic, and sometimes even logic-agnostic.
• It doesn't care what you're doing. But it knows all data is a tree.
• Be careful: using majas means accepting the truth of its name.

Changelog

1.1.1

• --infer is now implicit when --from is not present
• Fixed errors related to the handling of missing config values

1.1.0

• Moved @types/fs-extra from dependencies to devDependencies

1.0.0

Initial release

Markdown and JSON are similar. Everything else is detail.