libxml-bindings

libxml-bindings is a light set of methods and bolt-ons which aren’t maintained by the core libxml ruby library. These methods aim to provide a more convenient API interface which is provided and documented separately, but actually mixed in to extend the original LibXML::classes. Using these methods should significantly reduce the lines of code needed to perform the most common operations of accessing and manipulating an xml document structure.

For a full list of methods, please refer to the RDoc Documentation and also the Libxml-Ruby RDocs

For benkchmarks / performance comparison see cfis.savagexi.com/2008/07/16/resurrecting-libxml-ruby

Installation

gem sources -a gems.github.com gem install dreamcat4-libxml-bindings

Getting started

You can call to_xmldoc on any xml string to convert it into an XML::Document

>> s = ‘<foo id=“1”><author>p. bogle</author><bar>content</bar><bar>cont2</bar></foo>’ >> doc = s.to_xmldoc

The node method returns the first Node matching the given xpath

>> doc.node => <author>p. bogle</author>

The nodes method returns an array of Nodes matching the given xpath

>> doc.nodes => [<bar>content</bar>, <bar>content2</bar>]

nodes[] can be called with a block to iterate through each of the matching nodes

>> doc.nodes do |bar| puts bar.xpath; end /foo/bar /foo/bar

You can call node[] on another node to iterate and search within a smaller context of the document

>> foo = doc.node >> foo.node => <author>p. bogle</author> >> foo.nodes.each {|node| puts node.inner_xml} => “content2” => “cont2”

For more information about XPath syntax, please see www.w3schools.com/XPath/xpath_syntax.asp

Namespace Stripping

The handling of default namespaces in libxml-ruby is extremely awkward and cumbersome as it requires passing along an array of namespace strings with every find() method call. It also represents ambiguity concerning the href of the default namespace.

Suppose you had a namespaced XML source with xmlns:= directives like this >> document.to_xml <?xml version=“1.0” encoding=“UTF-8”?> <feed xmlns=“www.w3.org/2005/Atom” xmlns:openSearch=“a9.com/-/spec/opensearchrss/1.0/” xmlns:gContact=“schemas.google.com/contact/2008" xmlns:gd=”schemas.google.com/g/2005“> <title type="text">Phil Bogle’s Contacts</title> … </feed>

With libxml-bindings its possible to do:

>> document.strip!

And be left with plain xml which can be parsed more easily and without the need for specifying any confusing namespace directives

>> document.to_xml <?xml version=“1.0” encoding=“UTF-8”?> <feed> <title type="text">Phil Bogle’s Contacts</title> … </feed> >> document.node.inner_xml => “Phil Bogle’s Contacts”

After manipulating your data model, a default namespace can later be re-applied on the top-level node. However its generally not recommended to use more than one namespace within the same xml document.

Copyright

Contribution, Credits

This project was started on the top of Phil Bogle’s libxml_helper.rb. Most methods have been renamed to represent the api styling of the libxml project. The coding examples above are all adapted from Phil’s original explanatory texts.