<join> Examples

The following example is a simple join of two variables based on a single condition. For clarity, the data for the variables is shown directly. The joincondition attribute contains an XPath expression that defines the single condition:

<variable name="movies" type="document">
<movies>
<movie id="Stargate" ... />
<movie id="Lassie" ... />
...
</movies>
</variable>
<variable name="reviews" type="document">
<reviews>
<review>
<title>Encounters of the Third Kind</title>
<rating>4</rating>
...
</review>
<review>
<title>Stargate</title>
<rating>4.5</rating>
...
</review>
...
</reviews>
</variable>
<join outputvariable="$joinResult"
joincondition="$movies/movies/movie/@id =
$reviews/reviews/review/movie/title"/>

Each condition consists of an XPath expression, identifying a node for one item, a comparison operator and another XPath expression, identify the node for a related item to be joined. You can also use XPath functions in these XPath expressions.

Note:

In earlier versions, XPath expression within the join condition did not have to contain the root node, such as $movies/movie/@id instead of $movies/movies/movie/@id. The previous syntax is no longer supported and will fail.

You can use any of the basic math comparison operators (= , !=, <, <=, >, and >=). For string comparisons, you can also use ~ to express a partial match relationship. Comparisons with ~ are not case sensitive.

You must use XML escaping with operators that include < or > characters. See XML Escaping in URLs and Expressions for more information.

The following example would join an item from the location variable to items in the companyProfiles variable as long as the location city name is contained within the company profile address:

You can combine two or more conditions in the joincondition attribute using the logical XPath keywords and or or. You cannot use parentheses or other grouping symbols to define precedence, however. The joincondition attribute only supports very straightforward joins.

The <select> element is an optional child for <join>. It allows you to define the structure of the joined items to return and selectively choose which nodes to include. If you omit <select> all the nodes from the join data are included in the output.

Define the structure for each row or item in the result as literal XML in the body of <select>. This structure must contain:

This result structure also contains mashup expressions that determine what joined data will be selected to fill the result. See Dynamic Mashup Expressions for more information.

Note:

Previous versions of Presto required you to wrap CDATA sections around the contents of <select>. CDATA is a deprecated technique. Instead, declare a result namespace on <mashup> and use this namespace to identify the output.

In the next example, the output variable will contain a <res:recommendations> element with a repeating set of <res:movie> children - the repeating items. Each <res:movie> contains a <res:movietitle> child with data from the variable named movies and <res:rating> and <res:comment> children with data from the variable named reviews.

<join outputvariable="$joinResult"
joincondition="$movies/movie/@id = $reviews/review/movie/title">
<select name="res:recommendations">
<res:movie>
<res:movietitle>{$movies/title}</res:movietitle>
<res:rating>{$reviews/rating}</res:rating>
<res:comment>{$reviews/comment}</res:comment>
</res:movie>
</select>
</join>