Saturday, March 24, 2012

xml-motor ~ what it is; how & why should you use it

xml-motor ~ what it is; why & how you should use it

Download this article as pdf on what,why,how
http://speakerdeck.com/u/abhishekkr/p/xml-motor-whatwhyhow-this-xml-parsing-rubygem#


or read it all here.....

Late 2011, I started with a new rubygem project for parsing xml, html content.
  @Rubygems: http://rubygems.org/gems/xml-motor
  @GitHub     : https://github.com/abhishekkr/rubygem_xml_motor

Just created it to test out my work at compact, quick & easy xml-parsing algorithm... can see that
  @Slideshare: http://www.slideshare.net/AbhishekKr/xmlmotor

So, currently this is a non-native, completely independent less-than-250 ruby-LOC available as a simple rubygem to be require-d and use in an easy freehand notation and match with any node attributes.

Current Features:
  • Has a single method access to parse require xml nodes from content or file. Use it only if you are gonna parse that xml-content once. For using same xml-content more than once, follow the 3-way step mentioned in examples.
  • It doesn't depend on presence of any other system library, purely non-native.
  • It parses broken or corrupted xml/html content correctly, just for the content it have.
  • Can parse results on looking for node-names, attributes of node or both.
  • Uses free-freehand notation to retrieve desired xml nodes
    if your xml looks like,
    '<library>...
      <book> <title>ABC</title> <author>CBA</author> </book>...
      <book>
        <title>XYZ</title>
         <authors> <author>XY</author><author>YZ</author> </authors></book>...
    </library>'

    and you look for 'book.author',
    then, you'll get back ['CBA', 'XY', 'YZ'];
    what that means is the child-node could be at any depth in the parent-node.
  • Default return mode is without the tags, there is a switch to get the nodes.
    as you'd have seen in above example:
    'CBA' gets sent by default, not 'CBA'
  • To filter your nodes on the basis of attributes, single or multiple attributes can be provided.
  • These attribute searches can be combined up with freehand node name searches.
  • Readme (a bit weird): https://raw.github.com/abhishekkr/rubygem_xml_motor/master/README


Features To Come:
  • Work on making it more performance efficient.
  • Limit over result-nodes retrieved from start/end of matching nodes.
  • Multi-node attribute-based filter for a hierarchical node search.
  • Add dev-knows CSS Selector, it's already present using attribute based search... just need to add a mapping method.


EXAMPLES of usage:
example code to try: https://github.com/abhishekkr/axml-motor/tree/master/ruby/examples
  • say, you have an xml file 'dummy.xml', with data as
    <dummy>
      <ummy>    <mmy class="sys">non-native</mmy>  </ummy>
      <ummy>
        <mmy class="sys">      <my class="sys" id="mem">compact</my>    </mmy>
      </ummy>
      <mmy type="user">    <my class="usage">easy</my>  </mmy></dummy>
  • its available at rubygems.org, install it as
      $ gem install xml-motor
  • include it in your ruby code,
      #!/usr/bin/env ruby
      require 'xml-motor'
  • get the XML Filename and/or XML data available
      fyl = File.join(File.expand_path(File.dirname __FILE__),'dummy.xml')
      xml = File.open(fyl,'r'){|fr| fr.read }
  • One-time XML-Parsing directly from file
      XMLMotor.get_node_from_file(fyl, 'ummy.mmy', 'class="sys"')
         Result: ["non-native", "\n      compact\n    "]
  • One-time XML-Parsing directly from content
      XMLMotor.get_node_from_content xml, 'dummy.my', 'class="usage"'
         Result: ["easy"]
      
  • Way to go for XML-Parsing for xml node searches
      xsplit = XMLMotor.splitter xml
      xtags  = XMLMotor.indexify xsplit


      [] just normal node name based freehand notation to search:

        XMLMotor.xmldata xsplit, xtags, 'dummy.my'
        Result: ["compact", "easy"]
      [] searching for values of required nodes filtered by attribute:
        XMLMotor.xmldata xsplit, xtags, nil, 'class="usage"'
        Result: ["easy"]

      [] searching for values of required nodes filtered by freehand tag-name notation & attribute:

        XMLMotor.xmldata xsplit, xtags, 'dummy.my', 'class="usage"'
        Result: ["easy"]

      [] searching for values of required nodes filtered by freehand tag-name notation & multiple attributes:

        XMLMotor.xmldata xsplit, xtags, 'dummy.my', ['class="sys"', 'id="mem"']
        Result: ["compact"]

1 comment:

  1. With EasyHits4U you can earn free advertising credits by visiting other website-ads from a account base of over 1,200,000 accounts. Earn advertising credits fast with a 1:1 Exchange Ratio.

    ReplyDelete