12
Using Regular Expressions With Oracle Database

This chapter introduces regular expression support for Oracle Database. This chapter covers the following topics:

• What are Regular Expressions?
• Oracle Database Regular Expression Support
• Oracle Database SQL Functions for Regular Expressions
• Metacharacters Supported in Regular Expressions
• Constructing Regular Expressions

  See Also: Oracle Database SQL Reference for additional details on Oracle Database SQL functions for regular expressions Oracle Database Globalization Support Guide for details on using SQL regular expression functions in a multilingual environment Mastering Regular Expressions published by O'Reilly & Associates, Inc.

What are Regular Expressions?

Regular expressions specify patterns to search for in string data using standardized syntax conventions. A regular expression can specify complex patterns of character sequences. For example, the following regular expression:

a(b|c)d

searches for the pattern: 'a', followed by either 'b' or 'c', then followed by 'd'. This regular expression matches both 'abd' and 'acd'.

A regular expression is specified using two types of characters:

• Metacharacters--operators that specify algorithms for performing the search.
• Literals--the actual characters to search for.

Examples of regular expression syntax are given later in this chapter.

Oracle Database Regular Expression Support

Oracle Database implements regular expression support compliant with the POSIX Extended Regular Expression (ERE) specification.

Regular expression support is implemented with a set of Oracle Database SQL functions that allow you to search and manipulate string data. You can use these functions in any environment where Oracle Database SQL is used. See "Oracle Database SQL Functions for Regular Expressions" later in this chapter for more information.

Oracle Database supports a set of common metacharacters used in regular expressions. The behavior of supported metacharacters and related features is described in "Metacharacters Supported in Regular Expressions".

Oracle Database SQL Functions for Regular Expressions

The database provides a set of SQL functions that allow you to search and manipulate strings using regular expressions. You can use these functions on any datatype that holds character data such as CHAR, NCHAR, CLOB, NCLOB, NVARCHAR2, and VARCHAR2.

A regular expression must be enclosed or wrapped between single quotes. Doing so, ensures that the entire expression is interpreted by the SQL function and can improve the readability of your code.

Table 12-1 gives a brief description of each regular expression function.

Table 12-1   SQL Regular Expression Functions

SQL Function	Description
REGEXP_LIKE	This function searches a character column for a pattern. Use this function in the WHERE clause of a query to return rows matching the regular expression you specify. See the Oracle Database SQL Reference for syntax details on the REGEXP_LIKE function.
REGEXP_REPLACE	This function searches for a pattern in a character column and replaces each occurrence of that pattern with the pattern you specify. See the Oracle Database SQL Reference for syntax details on the REGEXP_REPLACE function.
REGEXP_INSTR	This function searches a string for a given occurrence of a regular expression pattern. You specify which occurrence you want to find and the start position to search from. This function returns an integer indicating the position in the string where the match is found. See the Oracle Database SQL Reference for syntax details on the REGEXP_INSTR function.
REGEXP_SUBSTR	This function returns the actual substring matching the regular expression pattern you specify. See the Oracle Database SQL Reference for syntax details on the REGEXP_SUBSTR function.

Metacharacters Supported in Regular Expressions

Table 12-2 lists the metacharacters supported for use in regular expressions passed to SQL regular expression functions. Details on the matching behavior of these metacharacters is given in "Constructing Regular Expressions".

Table 12-2   Metacharacters Supported in Regular Expressions

Metacharacter Syntax	Operator Name	Description
.	Any Character -- Dot	Matches any character
+	One or More -- Plus Quantifier	Matches one or more occurrences of the preceding subexpression
?	Zero or One -- Question Mark Quantifier	Matches zero or one occurrence of the preceding subexpression
*	Zero or More -- Star Quantifier	Matches zero or more occurrences of the preceding subexpression
{m}	Interval--Exact Count	Matches exactlym occurrences of the preceding subexpression
{m,}	Interval--At Least Count	Matches at least m occurrences of the preceding subexpression
{m,n}	Interval--Between Count	Matches at least m, but not more than n occurrences of the preceding subexpression
[ ... ]	Matching Character List	Matches any character in list ...
[^ ... ]	Non-Matching Character List	Matches any character not in list ...
|	Or	'a|b' matches character 'a' or 'b'.
( ... )	Subexpression or Grouping	Treat expression ... as a unit. The subexpression can be a string of literals or a complex expression containing operators.
\n	Backreference	Matches the nth preceding subexpression, where n is an integer from 1 to 9.
\	Escape Character	Treat the subsequent metacharacter in the expression as a literal.
^	Beginning of Line Anchor	Match the subsequent expression only when it occurs at the beginning of a line.
$	End of Line Anchor	Match the preceding expression only when it occurs at the end of a line.
[:class:]	POSIX Character Class	Match any character belonging to the specified character class. Can be used inside any list expression.
[.element.]	POSIX Collating Sequence	Specifies a collating sequence to use in the regular expression. The element you use must be a defined collating sequence, in the current locale.
[=character=]	POSIX Character Equivalence Class	Match characters having the same base character as the character you specify.

Constructing Regular Expressions

This section discusses construction of regular expressions.

Basic String Matching with Regular Expressions

The simplest match that you can perform with regular expressions is the basic string match. For this type of match, the regular expression is a string of literals with no metacharacters. For example, to find the sequence 'abc', you specify the regular expression:

abc

Regular Expression Operations on Subexpressions

As mentioned earlier, regular expressions are constructed using metacharacters and literals. Metacharacters that operate on a single literal, such as '+' and '?' can also operate on a sequence of literals or on a whole expression. To do so, you use the grouping operator to enclose the sequence or subexpression. See "Subexpression" for more information on grouping.

Regular Expression Operator and Metacharacter Usage

This section gives usage examples for each supported metacharacter or regular expression operator.

Match Any Character--Dot

The dot operator '.' matches any single character in the current character set. For example, to find the sequence--'a', followed by any character, followed by 'c'--use the expression:

a.c

This expression matches all of the following sequences:

abc
adc
a1c
a&c

The expression does not match:

abb

One or More--Plus

The one or more operator '+' matches one or more occurrences of the preceding expression. For example, to find one or more occurrences of the character 'a', you use the regular expression:

a+

This expression matches all of the following:

a
aa
aaa

The expression does not match:

bbb

Zero or One--Question Mark Operator

The question mark matches zero or one--and only one--occurrence of the preceding character or subexpression. You can think of this operator as specifying an expression that is optional in the source text.

For example, to find--'a', optionally followed by 'b', then followed by 'c'--you use the following regular expression:

ab?c

This expression matches:

abc
ac

The expression does not match:

adc
abbc

Zero or More--Star

The zero or more operator '*', matches zero or more occurrences of the preceding character or subexpression. For example, to find--'a', followed by zero or more occurrences of 'b', then followed by 'c'--use the regular expression:

ab*c

This expression matches all of the following sequences:

ac
abc
abbc
abbbbc

The expression does not match:

adc

Interval--Exact Count

The exact-count interval operator is specified with a single digit enclosed in braces. You use this operator to search for an exact number of occurrences of the preceding character or subexpression.

For example, to find where 'a' occurs exactly 5 times, you specify the regular expression:

a{5}

This expression matches:

aaaaa

The expression does not match:

aaaa

Interval--At Least Count

You use the at-least-count interval operator to search for a specified number of occurrences, or more, of the preceding character or subexpression. For example, to find where 'a' occurs at least 3 times, you use the regular expression:

a{3,}

This expression matches all of the following:

aaa
aaaaa

The expression does not match:

aa

Interval--Between Count

You use the between-count interval operator to search for a number of occurrences within a specified range. For example, to find where 'a' occurs at least 3 times and no more than 5 times, you use the following regular expression:

a{3,5}

This expression matches all of the following sequences:

aaa
aaaa
aaaaa

The expression does not match:

aa

Matching Character List

You use the matching character list to search for an occurrence of any character in a list. For example, to find either 'a', 'b', or 'c' use the following regular expression:

[abc]

This expression matches the first character in each of the following strings:

at
bet
cot

The expression does not match:

def

The following regular expression operators are allowed within the character list, any other metacharacters included in a character list lose their special meaning (are treated as literals):

• Range operator '-'
• POSIX character class [::]
• POSIX collating sequence [. .]
• POSIX character equivalence class [= =]

Non-Matching Character List

Use the non-matching character list to specify characters that you do not want to match. Characters that are not in the non-matching character list are returned as a match. For example, to exclude the characters 'a', 'b', and 'c' from your search results, use the following regular expression:

[^abc]

This expression matches characters 'd' and 'g' in the following strings:

abcdef
ghi

The expression does not match:

abc

As with the matching character list, the following regular expression operators are allowed within the non-matching character list (any other metacharacters included in a character list are ignored):

• Range operator '-'
• POSIX character class [::]
• POSIX collating sequence [. .]
• POSIX character equivalence class [= =]

For example, the following regular expression excludes any character between 'a' and 'i' from the search result:

[^a-i]

This expression matches the characters 'j' and 'l' in the following strings:

hijk
lmn

The expression does not match the characters:

abcdefghi

Or

Use the Or operator '|' to specify an alternate expression. For example to match 'a' or 'b', use the following regular expression:

a|b

Subexpression

You can use the subexpression operator to group characters that you want to find as a string or to create a complex expression. For example, to find the optional string 'abc', followed by 'def', use the following regular expression:

(abc)?def

This expression matches strings 'abcdef' and 'def' in the following strings:

abcdefghi
defghi

The expression does not match the string:

ghi

Backreference

The backreference lets you search for a repeated expression. You specify a backreference with '\n', where n is an integer from 1 to 9 indicating the n^th preceding subexpression in your regular expression.

For example, to find a repeated occurrence of either string 'abc' or 'def', use the following regular expression:

(abc|def)\1

This expression matches the following strings:

abcabc
defdef

The expression does not match the following strings:

abcdef
abc

The backreference counts subexpressions from left to right starting with the opening parenthesis of each preceding subexpression.

The backreference lets you search for a repeated string without knowing the actual string ahead of time. For example, the regular expression:

^(.*)\1$ 

matches a line consisting of two adjacent appearances of the same string.

Escape Character

Use the escape character '\' to search for a character that is normally treated as a metacharacter. For example to search for the '+' character, use the following regular expression:

\+

This expression matches the plus character '+' in the following string:

abc+def

The expression does not match any characters in the string:

abcdef

Beginning of Line Anchor

Use the beginning of line anchor ^ to search for an expression that occurs only at the beginning of a line. For example, to find an occurrence of the string def at the beginning of a line, use the expression:

^def

This expression matches def in the string:

defghi

The expression does not match def in the following string:

abcdef

End of Line Anchor

The end of line anchor metacharacter '$' lets you search for an expression that occurs only at the end of a line. For example, to find an occurrence of def that occurs at the end of a line, use the following expression:

def$

This expression matches def in the string:

abcdef

The expression does not match def in the following string:

defghi

POSIX Character Class

The POSIX character class operator lets you search for an expression within a character list that is a member of a specific POSIX Character Class. You can use this operator to search for characters with specific formatting such as uppercase characters, or you can search for special characters such as digits or punctuation characters. The full set of POSIX character classes is supported.

To use this operator, specify the expression using the syntax [:class:] where class is the name of the POSIX character class to search for. For example, to search for one or more consecutive uppercase characters, use the following regular expression:

[[:upper:]]+

This expression matches 'DEF' in the string:

abcDEFghi

The expression does not return a match for the following string:

abcdefghi

Note that the character class must occur within a character list, so the character class is always nested within the brackets for the character list in the regular expression.

POSIX Collating Sequence

The POSIX collating sequence element operator [. .] lets you use a collating sequence in your regular expression. The element you specify must be a defined collating sequence in the current locale.

This operator lets you use a multicharacter collating sequence in your regular expression where only one character would otherwise be allowed. For example, you can use this operator to ensure that the collating sequence 'ch', when defined in a locale such as Spanish, is treated as one character in operations that depend on the ordering of characters.

To use the collating sequence operator, specify [.element.] where element is the collating sequence you want to find. You can use any collating sequence that is defined in the current locale including single-character elements as well as multicharacter elements.

For example, to find the collating sequence 'ch', use the following regular expression:

[[.ch.]]

This expression matches the sequence 'ch' in the following string:

chabc

The expression does not match the following string:

cdefg

You can use the collating sequence operator in any regular expression where collation is needed. For example, to specify the range from 'a' to 'ch', you can use the following expression:

[a-[.ch.]]

POSIX Character Equivalence Class

Use the POSIX character equivalence class operator to search for characters in the current locale that are equivalent. For example, to find the Spanish character 'ñ' as well as 'n'.

To use this operator, specify [=character=], to find all characters that are members of the same character equivalence class as the specified character.

For example, the following regular expression could be used to search for characters equivalent to 'n' in a Spanish locale:

[[=n=]]

This expression matches both 'N' and 'ñ' in the following string:

El Niño