This is just a tiny set of extension methods that I've been copy/pasting into lots of my projects lately.
I decided that it was finally time to separate these extension methods out into their own file/library and write specs for them.
Now that I have this tiny library, I will add more tests and functionality to it, as needed.
I write lots of jQuery. I'm used to being able to say things like $('ul#nav li a').attr('foo', 'bar')
to set the foo attribute equal to bar for every link inside of an <li> inside of the <ul id="nav"> element.
When I work with a System.Xml.XmlDocument, I want to be able to use a similar chainable (fluent) interface.
Sample XML:
<?xml version="1.0" encoding="utf-8"?>
<dogs>
<dog name="Rover" breed="Golden Retriever">
<toys>
<toy>Tennis Ball</toy>
<toy>Kong</toy>
</toys>
</dog>
<dog name="Snoopy" breed="Beagle">
<toys>
<toy>Charlie Brown's Football</toy>
</toys>
</dog>
</dogs>Sample usage:
using FluentXml;
// Note, you do NOT need to use FluentXmlDocument ... this all works with a regular XmlDocument.
var doc = FluentXmlDocument.FromFile("dogs.xml");
>> doc.Node("dog").Attr("name");
"Rover"
>> doc.Node("dog toy").Text();
"Tennis Ball"
>> doc.Nodes("dog")[0].Attrs();
{{ "name", "Rover" }, { "breed", "Golden Retriever" }}
>> doc.Nodes("dog")[1].Attrs();
{{ "name", "Snoopy" }, { "breed", "Beagle" }}
>> doc.ToXml();
"<?xml version=\"1.0\" encoding=\"utf-8\"?>
<dogs>
<dog name=\"Rover\" breed=\"Golden Retriever\">
<toys>
<toy>Tennis Ball</toy>
<toy>Kong</toy>
</toys>
</dog>
<dog name=\"Snoopy\" breed=\"Beagle\">
<toys>
<toy>Charlie Brown's Football</toy>
</toys>
</dog>
</dogs>"
// Passing a second argument to Attr() sets the attribute value (just like with jQuery)
doc.Node("dog").Attr("name", "Changed!");
>> doc.ToXml();
"<?xml version=\"1.0\" encoding=\"utf-8\"?>
<dogs>
<dog name=\"Changed!\" breed=\"Golden Retriever\">
<toys>
<toy>Tennis Ball</toy>
<toy>Kong</toy>
</toys>
</dog>
<dog name=\"Snoopy\" breed=\"Beagle\">
<toys>
<toy>Charlie Brown's Football</toy>
</toys>
</dog>
</dogs>"
// Passing an argument to Text() sets the text of this node (just like with jQuery)
doc.Node("toy").Text("new toy text");
>> doc.ToXml();
"<?xml version=\"1.0\" encoding=\"utf-8\"?>
<dogs>
<dog name=\"Changed!\" breed=\"Golden Retriever\">
<toys>
<toy>new toy text</toy>
<toy>Kong</toy>
</toys>
</dog>
<dog name=\"Snoopy\" breed=\"Beagle\">
<toys>
<toy>Charlie Brown's Football</toy>
</toys>
</dog>
</dogs>"
// You can also pass in a lambda to Node() or Nodes() to find a node that matches some arbitrary condition
>> doc.Node(n => n.Attr("name") != null && n.Attr("name").StartsWith("S")).Attr("name");
"Snoopy"Pretty much, yeah. It has lots of safe null handling so I can safely say Node("foo").Node("bar").Node("whatever") and not worry about nulls.
To see more, just have a look at the source. It's a tiny thing, but ... it works for me!
FluentXml is released under the MIT license.