SiftXml is a lightweight xml to java mapper parsed by kXML2 and using a minimal DSL inspired by the Go stdlib encoding/xml package. Regardless of the number of nested nodes, it allows you to sift out only the information you're interested in without having to design complex models.
Mapping is done by calling new SiftXml().parse(InputStream, Class);.
Additional options can be configured before calling .parse() as so:
new SiftXml()
.setFeature(XmlPullParser.FEATURE_PROCESS_DOCDECL, true)
.setFeature(XmlPullParser.FEATURE_VALIDATION, true)
.parse(InputStream, Class);
repositories {
ivy {
url 'https://raw.github.com/thesmythgroup/SiftXml/master/repo/'
}
}
dependencies {
compile "org.tsg.siftxml:siftxml:0.1-SNAPSHOT"
}
<root>
<user>
<name>John Doe</name>
<email>john@doe.com</email>
</user>
</root>
@Xml("root > user")
public class User {
public String name;
public String email;
}
User user = new SiftXml().parse(in, User.class);
<root>
<user>
<name>John Doe</name>
<email>john@doe.com</email>
</user>
<user>
<name>Jane Roe</name>
<email>jane@roe.com</email>
</user>
</root>
@Xml("root")
public class Root {
@Xml("user > name")
public String[] names;
@Xml("user > email")
public String[] emails;
}
Root root = new SiftXml().parse(in, Root.class);
// the User class from the basic example above also works
User[] users = new SiftXml()parse(in, User[].class);
Inner classes should be declared static so they are made available to the mapper.
The type of a field is used when setting the value. Any class with a constructor that accepts a single String argument can be used. As an example, if a node's text was 1234 and a model's member was of type Integer, Integer.parseInt would be used when setting the member's value.