The <collations> element

The collations element in the configuration file defines a number of collations that can be preloaded into the configuration. It contains zero or more child collation elements as detailed below.

Each collation element may have the following attributes:

Attribute

Applies to

Values

Effect

uri

HE PE EE

The collation URI (mandatory)

An absolute URI used to identify the collation in queries and stylesheets.

alphanumeric

HE PE EE

yes|no|codepoint

Yes indicates that a sequence of digits within a string is read as a number, for example "test8.xml" precedes "test10.xml".

case-order

HE PE EE

upper-first|lower-first|#default

Indicates whether upper-case characters should precede or follow their lower-case equivalents.

class

HE PE EE

Java class implementing Collator, StringCollator, or Comparator

Class used to perform string comparisons.

decomposition

HE PE EE

none|standard|full

Determines whether Unicode normalization should be applied to strings before comparison (Java platform only).

ignore-case

HE PE EE

yes|no

Yes indicates that upper-case and lower-case are equivalent.

ignore-modifiers

HE PE EE

yes|no

Yes indicates that accents and other modifiers are ignored.

ignore-symbols

HE PE EE

yes|no

Yes indicates that punctuation symbols are ignored (.NET only).

ignore-width

HE PE EE

yes|no

Yes indicates that width variations between characters are ignored.

lang

HE PE EE

Language code, e.g. en-US

Language supported by the collation.

rules

HE PE EE

Rules in Java RuleBasedCollator format

Detailed rules for ordering of characters.

strength

HE PE EE

primary|secondary|tertiary|identical

The strength of the collation. A stronger collation takes more details of the character into account, e.g. accents and case.