XPath with RegEx Search Field Expression Plugin
The XPath with RegEx plugin extracts values from XML in the message body in two steps: an XPath 1.0 expression selects an element, and a regular expression extracts the part of its value you need. Use it when the value you want is only part of an element, such as a file name at the end of a path.
- ✅ Cut a file name, an ID, or a code out of a longer XML value
- ✅ Treat sub XML as a string lets the RegEx read the markup of a sub-tree, including element names
- ✅ RegEx groups returns only the capturing groups you choose
When to use it
| Your data | Use |
|---|---|
| Plain XML in the message body; you want element or attribute values as they are | XPath |
| Only part of an XML value (a file name in a path, an ID in a formatted string), or tag names inside an XML sub-tree | XPath with RegEx (this page) |
| XML stored as escaped text inside an element of another XML message | XPath on Wrapped XPath |
| XML stored in a context property instead of the body | XPath on Message Context |
| XPath 2.0, or XPath combined with other functions (decode Base64 first, concatenate, compare) | Formula with xPath2 |
For non-XML bodies, use RegEx or RegEx with capturing groups on the whole message; for the other plugins, see Choose a plugin.
Expression syntax
selects the element"] B --> C["RegEx to style value
extracts from its value"] C --> D["fa:fa-magnifying-glass-plus Unique values"] style A fill:#2678BD,color:#fff style B fill:#E47528,color:#fff style C fill:#9E59BB,color:#fff style D fill:#93d16d,color:#000
XPath runs first (orange) on the body (blue); the RegEx (purple) runs on each selected value; the matches become the Search Field values (green).
The plugin shows these fields:
| Field | What goes in it |
|---|---|
| XPath | An XPath 1.0 expression that selects the element, written from the root without a leading / (Orders/Header/FileName). Element names match without namespace prefixes; the supported syntax is listed in Allowed XPath expressions. |
| Treat sub XML as a string | Off: the RegEx runs on the element's text value. On: the RegEx runs on everything from the selected XPath as a string, markup included, so it can match tags and attributes. |
| RegEx to style value | The regular expression that extracts data from the XPath value. Despite the field label, it extracts; it doesn't format the value. |
| RegEx groups | Which capturing groups to return: group numbers or named groups, comma-separated (1,3 or value1,value2). Leave it empty to take all groups. |
| Global | On: keep matching after the first match (Don't return after first match). Off: return only the first match. |
Duplicates are collapsed into unique values (How expressions run). Pick the plugin in the Expression Type list when you add or manage a Search Field, and check the result in the Test Expression tab (Test Your Expression).
Selecting XPath with RegEx in the Expression Type list.
Examples
Example 1: Get the file name from a full file path
The header holds the full path of the received file; the Search Field should hold only the file name.
Input
<ns0:Orders xmlns:ns0="Common.Schemas/Nodinite/1.0">
<Header>
<FileName>\\nodinitesrv01\ftp\public\INT001\Order\OrderFile_123.xml</FileName>
</Header>
<Order>
<Id>101</Id>
<Amount>1000</Amount>
<City>Karlstad</City>
</Order>
</ns0:Orders>
Expression
| Field | Value |
|---|---|
| XPath | Orders/Header/FileName |
| Treat sub XML as a string | Off |
| RegEx to style value | ([^\\]+$) |
Result
OrderFile_123.xml
XPath returns \\nodinitesrv01\ftp\public\INT001\Order\OrderFile_123.xml; the RegEx matches the characters after the last backslash. The whole match and group 1 are the same here.
XPath Orders/Header/FileName plus RegEx ([^\\]+$) returns one unique match, OrderFile_123.xml.
Example 2: List the element names inside a sub-tree
Input
<Envelope>
<Any>
<Nodes>
<node1 id="1">
<node2/>
</node1>
</Nodes>
</Any>
</Envelope>
Expression
| Field | Value |
|---|---|
| XPath | Envelope/Any/Nodes |
| Treat sub XML as a string | On |
| RegEx to style value | <([a-z0-9]{0,}) |
| RegEx groups | 1 |
| Global | On |
Result
node1
node2
Nodes contains only elements, so its text value is empty: without Treat sub XML as a string the RegEx has nothing to match. With it on, the RegEx reads the markup. Group 1 returns the name after each <, and Global keeps matching after the first one. The closing tag </node1> gives an empty group, which doesn't appear in the result.
Treat sub XML as a string, RegEx group 1, and Global turned on: 2 unique matches, node1 and node2.
Limitations
- The XPath step uses the same forward-only reader as the XPath plugin, so only the syntax in Allowed XPath expressions works.
- The RegEx runs on the value the XPath selects, never on the rest of the message.
An unsupported expression, or one that matches nothing, gives this result in the test:
No match: the test reports a warning instead of values.