Yatom

Beautiful HTML/XHTML/XML using YAML.

Yatom is a YAML to markup compiler supporting XML and HTML as output formats.

HTML and its siblings are not friendly at all, their very convoluted and often unreadable nature (due its parent: SGML) cannot be overtaken regardless on how much effort is put on both indentation and formatting. And even that requires further postprocessing to avoid the a huge size overhead.

In some way, Yatom shares the same concept as jedi/pugjs and HAML, but unless being yaml-inspired^TM, Yatom is absolutely pure YAML, completely language agnostic, without extra stuff, requirements or incompatibilities.

YAML in, HTML out, simple.

Usage

Simple HTML5 page using Yatom.

import yatom

source = '''
  doctype: html5
  html:
    head:
      title: My Yatom page!
    body:
      h2: Yatom is awesome
      p: >-
        Now, you can code your page with
        beautiful markup without worrying
        about inefficient HTML output nor
        erratic linebreak behavior thanks
        to Yatom.
  '''
print(
  yatom.HTMLProcessor
    .from_source(source)
    .render()
  )

<!DOCTYPE html>
<html><head><title>My Yatom page!</title></head><body><h2>Yatom is awesome</h2><p>Now, you can code your page with beautiful markup without worrying about inefficient HTML output nor erratic linebreak behavior thanks to Yatom.</p></body></html>

The Yatom syntax

Yatom uses regular YAML, but that doesn't mean documents can of arbitrary shape.

The YAML document structure is defined by the target language, but mostly all follow the following rules:

• YAML root must be a mapping, as mappings define the document structure.
• If a non-mapping value is is encountered, it is treated as text.

HTML/XHTML/XML

The YAML syntax defining an entire HTML document is quite simple. If you already know HTML you can start writing Yatom templates with very few rules:

• Mapping keys are tag names, with one exception:
  • Key doctype is handled differently, its accepted values are listed here:
    • html4-strict
    • html4-transitional
    • html4-frameset
    • html5
    • xhtml1-strict
    • xhtml1-transitional
    • xhtml1-frameset
    • xhtml11
• Mapping string values are treated as text content.
• Dot-prefixed keys are tag attributes and must appear before any other sibling keys, with some exceptions:
  • .text for inline text strings (escaped)
  • .literal for unescaped inlined text strings (allowing inline markup)
  • .cdata for XML CDATA tags
  • .comment for HTML comments

In addition to previous rules, attributes support nesting, with the following rules:

• When style value is a mapping, properties are treated as CSS properties, and nested mapping keys are joined with dashes (-).
• When class value is a mapping or list, its properties are treated as different classes (the dot is prefixed). It its a mapping its hierarchy (mapping or array) is combined with dashes (-).
• If any other attribute value is a mapping or list, its hierarchy is combined with dashes (-).

And as a bonus, and only if required by doctype, few tags provide sane defaults:

• Both html4 and xhtml:
  • style element's attribute type defaults to text/css.
  • script element's attribute type defaults to application/javascript.
• Only xhtml:
  • html element has both default lang and xmlns.
  • Both style and script element's content is automatically wrapped with CDATA tags.

Semantic HTML rules are applied.

Examples

Simple example for HTML5.

import yatom

source = '''
  doctype: html5
  html:
    head:
      title: my page
    body:
      h2: my page
      p: |
        multiline
        text
      p:
        .text: mixed
        span:
          .style:
            color: red
          .text: tags
        .text: and
        strong: text
  '''
print(
  yatom.HTMLProcessor
    .from_source(source)
    .render()
  )

<!DOCTYPE html>
<html><head><title>my page</title></head><body><h2>my page</h2><p>multiline
text
</p><p>mixed<span style="color:red">tags</span>and<strong>text</strong></p></body></html>

More advanced XHTML4 example.

import yatom

source = '''
  doctype: xhtml11
  html:
    head:
      title: my page
      script: window.alert('<hello world>')
    body:
      .data:
        something: 1
        other: 2
      .class:
        - simple:
          - nested
        - other
      .style:
        padding:
          top: 2em
          bottom: 2em
          left: 25%
          right: 25%
      p: some simple text
  '''
print(
  yatom.HTMLProcessor
    .from_source(source)
    .render()
  )

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"><head><title>my page</title><script type="application/javascript"><![CDATA[window.alert('<hello world>')]]></script></head><body data-something="1" data-other="2" class="simple-nested other" style="padding-top:2em;padding-bottom:2em;padding-left:25%;padding-right:25%"><p>some simple text</p></body></html>

Roadmap

• Pretty print
• Pluggable templating engines or logic
• Drop python2 for good