LMY: is Metä(l) YAML :-{}

README.md

LMY: is Metä(l) YAML :-{}

Literate YAML with mustaches

Goals

• write YAML...
  • or Markdown
  • or JSON
• add mustache-style template tags...
• load multiple files as data sources...
  • ...of many types...
    • .json
      • .jsonld
      • .ipynb
    • .ini/.cfg
    • .yml
    • .csv
  • ...into optional aliases
  • ...from a path in the files
• ...get out JSON
  • ...or YAML
  • ...or Markdown
  • ...or HTML
  • ...or a JS app

Show me the LMY: Command Line

Any valid YML (and by extension JSON) is valid LMY:

# data.lmy
{% for lang in ["py", "js"]%}
{{ lang }}:
  name: {{ data[lang].name }}
  version: {{ data[lang].version }}
{% endfor %}

...and any number of data files can be loaded into the available data context.

lmy data.lmy py:setup.cfg#/metadata js:package.json

By default, lmy will make JSON...

{"py": {"name": "py-lmy", "version": "0.1.0"}, "js": {"name": "js-lmy", "version": "0.1.0"}}

Make it pretty:

lmy data.lmy --pretty py:setup.cfg#/metadata js:package.json

{
  "py": {
    "name": "py-lmy",
    "version": "0.1.0"
  },
  "js": {
    "name": "js-lmy",
    "version": "0.1.0"
  }
}

...or, get the intermediate YAML, which is already pretty...

lmy data.lmy --yaml py:setup.cfg#/metadata js:package.json

py:
  name: py-lmy
  version: 0.1.0
js:
  name: js-lmy
  version: 0.1.0

Add -o or --output to any of the above to write files.

Think More About CLI

• multiple input files...
  • join?
  • merge?
  • split?
  • stream?
• multiple output files...
  • split into documents on YAML document streams?
• meta LMY
  • .lmy.lmy?
    • document vs data expansion?
  • How to escape without being awful?

Literate LMY

If the intent of a document is more important than the data, # comments are unnecessarily verbose. In the style of literate CoffeeScript, an .lmy.md file will be processed as a templated markdown file, and

# The LMYsphere

    {% for lang in ["py", "js"]%}
    {{ lang }}:
      name: {{ get[lang].name }}
      version: {{ data[lang].version }}
    {% endfor %}

LMY is about data, so the default will generate JSON...

lmy README.lmy.md py:setup.cfg#/metadata js:package.json

same as above :{)

...or can be expanded to template-free Markdown with embedded YAML...

lmy README.lmy.md --markdown py:setup.cfg#/metadata js:package.json

# The LMYsphere

    py:
      name: py-lmy
      version: 0.1.0
    js:
      name: js-lmy
      version: 0.1.0

...or go all the way to HTML, leaving the data as expanded, indented YAML...

lmy README.lmy.md --html py:setup.cfg#/metadata js:package.json

too long

...with the added option of generating nested tables, a la console.table...

lmy README.lmy.md --html --pretty=tables py:setup.cfg#/metadata js:package.json

too long to show

Think more about HTML

• different table representations

LMY apps

Some LMY can make a standalone JavaScript application

Prior Art

• salt uses Jinja and YAML to define states
• metadown uses Markdown and YAML to author CoffeeScript
• literate CoffeeScript

The LMYsphere

• JavaScript
  • js-lmy: a LMY processor for node and the browser
  • codemirror-lmy: a CodeMirror mode
  • prosemiror-lmy: a ProseMirror mode
  • ace-lmy: an Ace mode
• Python
  • py-lmy: a LMY processor for python
  • a pygments mode
• Editors
  • atom-lmy: LMY syntax highlighting for atom
  • atom-linter-lmy: LMY