XPath Expressions

Overview

DataPipeline's query engine uses a subset of XPath 1.0 location paths notation to identify field values and demarcate records in XML, JSON, and Java objects.

Axis Specifiers

Axis	Abbreviated Syntax	Supported	Examples
ancestor
ancestor-or-self
attribute	@	yes	@lang or attribute::lang
child		yes	title or child::title
descendant		yes
descendant-or-self	//	yes	//book or /descendant-or-self::book/
following
following-sibling
namespace
parent	..
preceding
preceding-sibling
self	.	yes

Node Tests

comment(), text(), processing-instruction(), node() are all supported.

Predicates, Functions and Operators

Not supported.

Special Symbols

DataPipeline supports escaping symbols that have special meaning in XPath. For example:

• If you have a node name that looks like Price (£), you can include it in an XPath location path by wrapping the name in backticks (also known as backquotes, grave accents, or graves): addField("COL_1", "/object/table/array/object/`Price (£)`").
• If the node name includes a backtick, such as Price `£`, you can escape them by doubling each backtick: addField("COL_1", "/object/table/array/object/Price ``£``").

Wildcards

DataPipeline supports wildcards inside of XPath expressions using asterisks. For example, you can match nodes that starts with Price by using: addField("COL_1", "/object/table/array/object/Price*").

• When wildcards are in play, the node test becomes case insensitive allowing you to use: addField("COL_1", "/object/table/array/object/PRICE (*)") or addField("COL_1", "/object/table/array/object/price (*)").
• Wildcards can be placed inside backticks if they include symbols: addField("COL_1", "/object/table/array/object/`Price (*)`").
• If the node name includes asterisks (Price *£*), you can escape them by doubling each asterisks: addField("COL_1", "/object/table/array/object/`Price **£**`").