Extensible Markup Language (XML) is an internet standard that is used to express and convert data. In order to find the necessary information out of XML documents, you need a search system for XML documents. In this research, we have developed a search system that can find documents that matches the structure and content of a given XML document, making the best use of XML structure. Search metrics take account of the similarity in tag names, tag values, and the structure of tags. After a search, the system displays the ranked results in the order of aggregate similarity. Three methods of query are provided: keyword search which is conventional; search with tag names and their values; and search with XML documents. These three methods enable users to choose the method that best suits their preference, resulting in the increase of the usefulness of the system.