Minimal Javascript Xml Parser. 38 lines. Associative arrays.

	Minimal Javascript Xml Parser Webneurons Web 3

Webneurons Web 3 Home

Back
Page titles A-Z index

Main Pages

Original Posting
Original Site
Newsgroup Post
Web 3
Wrong Approach
XML

Reference

Links
Contact us

Webneurons Web 3 - The main goal of webneurons is to develop a better way of programming computers. Demonstration of a simple webneuron network. Also includes a minimal javascript XML parser of 37 lines.

Minimal Javascript Xml Parser. 38 line XML parser producing a resultant array of associative and normal arrays from the xml text input.

XML - XMLHttpRequest, Minimal javascript xml parser, and the Ajax system of xmlHttpRequest, DOM and style sheets, for adjusting web pages without reloading.

Tag blocks: text with start tag and end tag. Two main types:-
1. Section block: contains sub-blocks, e.g. "<a> <r>r1</r><s>s1</s> </a>"
2. Data block: no sub-blocks, e.g. "<b>b1</b>"

Each block becomes a property of R, e.g. R.a, R.b, sub-blocks R.a[0].r etc.

Each block also appears in R as a string. Two types:-
1. section block example: R.a[0]._value_=' <r>r1</r> <s>s1</s> '
2. data block example: R.b[0].value='b1'.
Uses _name_ and _value_ properties in section blocks to reduce conflicts with other property names. Uses name and value properties for data blocks since there can't be conflicts.

E.g. R=parseXml("<a> <r>r1</r><s>s1</s> </a> <a> <r>r2</r><t>t2</t> </a> <b>b1</b>") gives:-

R.a[0]._name_='a'      R.a[0]._value_=' <r>r1</r> <s>s1</s> '
R.a[0].r[0].name='r'   R.a[0].r[0].value='r1'
R.a[0].s[0].name='s'   R.a[0].s[0].value='s1'
R.a[1]._name_='a'      R.a[1]._value_=' <r>r2</r> <t>t2</t> '
R.a[1].r[0].name='r'   R.a[1].r[0].value='r2'
R.a[1].t[0].name='t'   R.a[1].t[0].value='t2'
R.b[0].name='b'        R.b[0].value='b1'

or in another view R is:-
    
a = [] length = 2
  0 = [] length = 0
    _name_ = "a"
    _value_ = " <r>r1</r> <s>s1</s> "
    r = [] length = 1
      0 = [] length = 0
        name = "r"
        value = "r1"
    s = [] length = 1
      0 = [] length = 0
        name = "s"
        value = "s1"
  1 = [] length = 0
    _name_ = "a"
    _value_ = " <r>r2</r> <t>t2</t> "
    r = [] length = 1
      0 = [] length = 0
        name = "r"
        value = "r2"
    t = [] length = 1
      0 = [] length = 0
        name = "t"
        value = "t2"
b = [] length = 1
  0 = [] length = 0
    name = "b"
    value = "b1"

function parseXml(xmlStr) {
// 4 lines of code
var Xml=[]
Xml._value_=xmlStr
return subTree(Xml)}

function subTree(Node) {
// 34 lines of code
// Block: text with tag start tag and end tag 
// Node: combined associative and non associative array at the current depth
var lt, gt, tagName, tagNameEnd, endTag, endLt, cursor=0, block, val, ix
block=Node._value_
while( (lt=block.indexOf('<',cursor))!=-1 ) {                                         
   // Get the next block on the this level
   gt=block.indexOf('>', lt)
   if(gt==-1) {
      alert('Block aborted. Maybe no end tag ' +endTag+ '\n\n' +block)
      break
   }
   tagNameEnd=block.indexOf(' ',lt) // look for space so can ignore attributes
   if(tagNameEnd<lt+2 || tagNameEnd>gt) tagNameEnd=gt 
   tagName=block.substring(lt+1, tagNameEnd) 
   endTag='</'+tagName+'>'
   endLt=block.indexOf(endTag, gt)
   if(endLt==-1) {
      alert('Block aborted. Maybe no end tag:\n\n' +block)
      break
   }   
   val=block.substring(gt+1, endLt)    // Got the next block, i.e. val
   Node[tagName] = Node[tagName] || []
   ix=Node[tagName].length
   Node[tagName].push([])
   if(val.indexOf('<')!=-1) {    
      Node[tagName][ix]._name_=tagName // obscure names less conflicts.
      Node[tagName][ix]._value_=val
      subTree(Node[tagName][ix])       // New sub block. Recursive.
   }     
   else {
      Node[tagName][ix].name=tagName   
      Node[tagName][ix].value=val
   }
   cursor=endLt+endTag.length                     
}
return Node}