Ratel

SRX and Java

If you are using an Okapi Tool after the M9 release, you should be using the wiki online help:
http://www.opentag.com/okapi/wiki/index.php?title=SRX_and_Java

The SRX 2.0 standard is based on the ICU regular expression notation.

Ratel uses Java's regular expressions to implement SRX. One of the reasons for this is because ICU4J (ICU for Java) does not provide support of ICU regular expressions.

As of version 1.6 Java does not have support for some of the Unicode-enabled features as described in ICU. For example in Java "\w" means "[a-zA-Z_0-9]" not "[\p{Ll}\p{Lu}\p{Lt}\p{Lo}\p{Nd}]" like in ICU. Some ICU features can be replaced by an equivalent expression in Java, but some other features simply cannot be implemented in Java.

The following table shows the ICU and Java differences. The yellow entries denote a case where the ICU expression needs to be mapped to a Java equivalent (sometimes a complex one), and the red entries indictate the cases where the ICU expression cannot be mapped in Java.

ICU Meta Character	Java Equivalent	ICU Description
`\a`	same	Match a BELL, `\u0007`
`\A`	same	Match at the beginning of the input. Differs from `^` in that `\A` will not match after a new line within the input.
`\b`, outside of a set	`\b` exists but does not have exactly the same behavior.	Match if the current position is a word boundary. Boundaries occur at the transitions betweem word (`\w`) and non-word (`\W`) characters, with combining marks ignored. And the option UREGEX_UWORD is assumed to be NOT set (default).
`\b`, within a set	`\b` is invalid when within a set. Use `\u0008` instead.	Match a BACKSPACE, \u0008.
`\B`	`\B` exists but does not have exactly the same behavior.	Match if the current position is not a word boundary. And the option UREGEX_UWORD is assumed to be NOT set (default).
`\cX`	same	Match a control-X character.
`\d`	`\d` exists but is ASCII based. Use `[\p{Nd}]` instead.	Match any character with the Unicode General Category of Nd (Number, Decimal Digit.)
`\D`	`\D` exists but is ASCII based. Use `[^\p{Nd}]` instead.	Match any character that is not a decimal digit.
`\e`	same	Match an ESCAPE, `\u001B`.
`\E`	same	Terminates a `\Q` ... `\E` quoted sequence.
`\f`	same	Match a FORM FEED, `\u000C`.
`\G`	same	Match if the current position is at the end of the previous match.
`\n`	same	Match a LINE FEED, `\u000A`.
`\N{UNICODE CHARACTER NAME}`	Does not exists	Match the named character.
`\p{UNICODE PROPERTY NAME}`	same	Match any character with the specified Unicode Property.
`\P{UNICODE PROPERTY NAME}`	same	Match any character not having the specified Unicode Property.
`\Q`	same	Quotes all following characters until `\E`.
`\r`	same	Match a CARRIAGE RETURN, `\u000D`.
`\s`	`\s` exists but is ASCII based (it matches `[ \t\n\x0B\f\r]`) Use `[\t\n\f\r\p{Z}]` instead.	Match a white space character. White space is defined as `[\t\n\f\r\p{Z}]`.
`\S`	`\S` exists but is ASCII based Use `[^\t\n\f\r\p{Z}]` instead.	Match a non-white space character.
`\t`	same	Match a HORIZONTAL TABULATION, `\u0009`.
`\uhhhh`	same	Match the character with the hex value `hhhh`.
`\Uhhhhhhhh`	Does not exist	Match the character with the hex value `hhhhhhhh`. Exactly eight hex digits must be provided, even though the largest Unicode code point is `\U0010ffff`.
`\w`	`\w` exists but is ASCII based. Use `[\p{Ll}\p{Lu}\p{Lt}\p{Lo}\p{Nd}]` instead.	Match a word character. Word characters are `[\p{Ll}\p{Lu}\p{Lt}\p{Lo}\p{Nd}]`.
`\W`	`\W` exists but is ASCII based Use `[^\p{Ll}\p{Lu}\p{Lt}\p{Lo}\p{Nd}]` instead.	Match a non-word character.
`\x{hhhh}`	Does not exists Use `\uhhhh` instead.	Match the character with hex value `hhhh`
`\xhh`	same	Match the character with two digit hex value `hh`
`\X`	Does not exists	Match a Grapheme Cluster.
`\Z`	same	Match if the current position is at the end of input, but before the final line terminator, if one exists.
`\z`	same	Match if the current position is at the end of input.
`\0nnn`	same	Match the character with octal value `nnn`.
`\n`	same	Back Reference. Match whatever the nth capturing group matched. n must be >1 and < total number of capture groups in the pattern.
`[pattern]`	same	Match any one character from the set. See UnicodeSet for a full description of what may appear in the pattern.
`.`	same	Match any character.
`^`	same	Match at the beginning of a line.
`$`	same	Match at the end of a line.
`\`	same	Quotes the following character. Characters that must be quoted to be treated as literals are `* ? + [ ( ) { } ^ $ \| \ . /`