Absolute Path
Starts selection at the XML document root.
/Report/<NodeCollection>/<Node>When to use
Use when the target location is fixed regardless of current context.
Notes
Absolute paths break if the root structure changes.
Phase 4 · Reference Material
XPath Library
Phase 4 extends the XPath Library with XPath 1.0 navigation, filtering, duplicate detection, and text functions.
DuplicatesFilteringLogicMath and CountsNavigationText Functions
Relative Path
Starts from the current context node.
Detail/<FieldName>When to use
Use inside Transformer actions where the current node is already selected.
Notes
Know your current node before writing the expression.
Parent Node
Selects the parent of the current node.
..When to use
Use when a field-level match needs to act on the parent record.
Notes
Avoid climbing too far without confirming the XML structure.
Previous Sibling
Selects prior sibling Detail nodes.
preceding-sibling::DetailWhen to use
Useful for duplicate detection and comparing current row to earlier rows.
Notes
Only checks siblings at the same level.
Next Sibling
Selects following sibling Detail nodes.
following-sibling::DetailWhen to use
Useful when comparing a current line to later lines.
Notes
Can be expensive on large XML if overused.
First Node
Selects the first matching node.
[position() = 1]When to use
Use when only the first occurrence should remain or be processed.
Notes
[1] is shorter but context can matter.
Not First Node
Selects every matching node except the first.
[position() > 1]When to use
Use for removing repeated nodes when position is enough.
Notes
Do not write position()=!1; that is invalid XPath syntax.
Last Node
Selects the last matching node.
[last()]When to use
Use when totals or final records should be processed.
Notes
last() is based on the current node set.
Duplicate Based on Previous Sibling Field
Finds rows where the current field value already appeared earlier.
Detail[FIELD_ID = preceding-sibling::Detail/FIELD_ID]When to use
Use for deleting duplicate Detail records while keeping the first occurrence.
Notes
Only works for duplicates under the same parent.
Normalize Space
Trims and collapses internal whitespace.
normalize-space(FIELD_NAME)When to use
Use before comparing values that may contain extra spaces.
Notes
normalize-space() returns a string, not a node.
Contains
Checks whether a field contains text.
contains(FIELD_NAME, "ABC")When to use
Use for partial text matching.
Notes
Case-sensitive in XPath 1.0.
Starts With
Checks whether a field starts with text.
starts-with(FIELD_NAME, "ABC")When to use
Use for prefix checks such as account, item, or document patterns.
Notes
Case-sensitive in XPath 1.0.
Ends With Workaround
XPath 1.0 workaround for suffix matching.
substring(FIELD_NAME, string-length(FIELD_NAME) - string-length("ABC") + 1) = "ABC"When to use
Use because XPath 1.0 has no ends-with() function.
Notes
Fails if you forget the + 1 offset.
String Length
Returns the number of characters in a string.
string-length(FIELD_NAME)When to use
Use for testing blank values or fixed-length fields.
Notes
Whitespace counts unless normalized first.
Count Nodes
Counts matching child nodes.
count(Detail)When to use
Use in conditions that depend on number of records.
Notes
Make sure the path is relative to the right node.
Not Empty
Checks that a field has content.
normalize-space(FIELD_NAME) != ""When to use
Use instead of FIELD_NAME != "" when spaces are possible.
Notes
Avoid treating whitespace as meaningful data.
AND Condition
Requires both conditions to be true.
FIELD_A != "" and FIELD_B = "Yes"When to use
Use for precise filtering.
Notes
Use lowercase and/or in XPath.
OR Condition
Allows either condition to be true.
FIELD_A = "A" or FIELD_A = "B"When to use
Use for multiple accepted values.
Notes
Parentheses help readability for complex logic.
Related Topics
Phase 5 adds related-topic guidance to reduce duplicate entries and make reusable patterns easier to find.
RedactedReviewedCanonical