design.md

Goals

• Generally match MSXML2.DOMDocument
• Utilize Dictionary and Collection to add no new classes
• Xml = ConvertToXml(ParseXml(Xml))

Parsing Components

1. prolog: <? ... ?>
2. doctype: <!doctype ... [...]>
3. documentElement: Root element
4. Element:

#document (Element)
  prolog: (Element)
  doctype: (Element)
  documentElement: (Element)
  nodeName: "#document"
  childNodes:
  - (prolog)
  - (doctype)
  - (documentElement)
  attributes: (empty)
  text: ""
  xml: "..."

Element

• nodeName String
• childNodes Collection
• attributes Dictionary
• text String
• xml String

<messages name="Tim"><message id="1">A</message><message id="2">B</message></messages>
^         ^    ^    ^ ^              ^ ^         ^                ^         ^
a         b    c    d e              f g         h                i         j

parseElement

• a: <, nodeName = "messages"
• b: non->, attribute, key = "name"
• c: =, value = "Tim"
• d: >, close opening tag, check for childNodes/text/void next
• e: < + non-/, opening tag, childNodes + parseElement
• f: non-<, text = "A"
• g: </, closing tag, find > and exit parseElement
• h: < + non-/, another opening tag, childNodes + parseElement
• i: </, closing tag, find > and exit parseElement
• j: </, closing tag, find > and exit original parseElement

Look for space -> attributes
Look for > -> end of opening

If void element, look for immediate </...> or <
Otherwise:
  Look for immediate < -> childNodes -> parseElement
  Look for ... -> text
  Look for immediate </ -> close element

TODO Handle comments

parseAttribute -> Array(Key, Value)

createElement(nodeName, childNodes, attributes, text, xml) -> Dictionary

Helper for loading values into Dictionary

Process

1. Parse prolog into Element
2. Parse doctype into Element
3. Use parseElement starting after doctype for documentElement
4. Create #document element and add prolog, doctype, and documentElement