XPath Standard Functions

The Publisher supports the following XPath standard functions. In addition, there are the Publisher-specific layout functions and user-defined functions.

Strings

concat(<value>,<value>, ...)

Create a new text by concatenating the values.

contains(<haystack>,<needle>)

True if haystack contains needle. Example: contains('bana','na') returns true().

starts-with(<string>, <string>)

True if the first string starts with the second.

ends-with(<string>, <string>)

True if the first string ends with the second.

substring(<input>,<start>[,<length>])

Return a part of the string. substring('Goldfarb', 5, 3) returns far. start can be negative (counts from the end).

substring-before(<string>,<string>)

Return the part of the first string before the second. Example: substring-before("tattoo", "attoo") returns "t".

substring-after(<string>,<string>)

Return the part of the first string after the second. Example: substring-after("tattoo", "tat") returns "too".

string-length(<string>)

Return the length of the string. Multi-byte UTF-8 sequences count as one.

string-join(<sequence>, <separator>)

Join the elements of the sequence with the separator.

normalize-space(<text>)

Remove leading and trailing whitespace. Multiple spaces and newlines are replaced by a single space.

upper-case(<text>)

Convert text to uppercase. upper-case('text') returns TEXT.

lower-case(<text>)

Convert text to lowercase. lower-case('Text') returns text.

translate(<input>,<from>,<to>)

Replace all characters in input found in from with the corresponding character in to. Example: translate('banana','an','uo') returns buouou. If to is shorter than from, extra characters are deleted.

matches(<text>,<regexp>[,<flags>])

True if the text matches the regular expression. Flags: s, i, m (see XPath spec). Example: matches("banana", "^(.a)+$") returns true.

replace(<input>,<regexp>,<replacement>)

Replace all matches of the regular expression with the replacement text. Example: replace('banana', 'a', 'o') returns bonono. With groups: replace('W151TBH','^[A-Z]([0-9]+)[A-Z]+$', '$1') returns 151.

tokenize(<input>,<regexp>)

Split the input by the regular expression into a sequence of strings. Example: tokenize("Go home, Jack!", "\W+") returns the sequence "Go", "home", "Jack", "".

string(<sequence>)

Return the text value of the sequence.

codepoints-to-string(<codepoints>)

Convert a sequence of codepoints to a string.

string-to-codepoints(<string>)

Convert a string to a sequence of codepoints.

Numbers

number(<value>)

Convert the value to a number. Returns NaN if conversion fails.

abs(<number>)

Return the absolute value. abs(-1.34) returns 1.34.

ceiling(<number>)

Round up to the next integer. ceiling(1.34) returns 2.

floor(<number>)

Round down to the next integer. floor(1.7) returns 1.

round(<number>[,<decimals>])

Round to the given number of decimal places (default: 0).

round-half-to-even(<number>[,<decimals>])

Banker’s rounding: at .5, round to the nearest even number.

format-number(<number>, <pattern>)

Format the number according to the pattern. Example: format-number(12345.67, '#,##0.00') returns 12,345.67.

max(<values>)

Return the maximum. max(1.1, 2.2, 3.3) returns 3.3.

min(<values>)

Return the minimum. min(1.1, 2.2, 3.3) returns 1.1.

Boolean Functions

true()

Return true.

false()

Return false.

not(<value>)

Negate the boolean value. not(true()) returns false().

boolean(<sequence>)

Return the effective boolean value.

empty(<sequence>)

True if the sequence is empty (e.g. a non-existing attribute or element).

Sequences and Nodes

count(<nodes>)

Count child elements with the given name. Example: count(article).

position()

Return the position of the current node.

last()

Return the number of same-named sibling elements.

distinct-values(<sequence>)

Return unique values. distinct-values((1,2,2,3)) returns (1,2,3).

reverse(<sequence>)

Reverse the order of the sequence.

local-name()

Return the name of the current node (without namespace).

name()

Return the name of the current node (with namespace prefix).

namespace-uri([<nodeset>])

Return the namespace URI of the first node.

root(<element>)

Return the root element.

doc(<filename>)

Open the file and return its content as an XML tree.

unparsed-text(<filename>)

Return the file content as uninterpreted text.