9.4. String Functions and Operators

This section describes functions and operators for examining and manipulating string values. Strings in this context include values of the types character, character varying, and text. Unless otherwise noted, all of the functions listed below work on all of these types, but be wary of potential effects of automatic space-padding when using the character type. Some functions also exist natively for the bit-string types.

SQL defines some string functions that use key words, rather than commas, to separate arguments. Details are in Table 9.8. PostgreSQL also provides versions of these functions that use the regular function invocation syntax (see Table 9.9).

Note

Before PostgreSQL 8.3, these functions would silently accept values of several non-string data types as well, due to the presence of implicit coercions from those data types to text. Those coercions have been removed because they frequently caused surprising behaviors. However, the string concatenation operator (||) still accepts non-string input, so long as at least one input is of a string type, as shown in Table 9.8. For other cases, insert an explicit coercion to text if you need to duplicate the previous behavior.

Table 9.8. SQL String Functions and Operators

FunctionReturn TypeDescriptionExampleResult
string || stringtext String concatenation 'Post' || 'greSQL'PostgreSQL
string || non-string or non-string || stringtext String concatenation with one non-string input 'Value: ' || 42Value: 42
bit_length(string)intNumber of bits in stringbit_length('jose')32
char_length(string) or character_length(string)int Number of characters in string char_length('jose')4
lower(string)textConvert string to lower caselower('TOM')tom
octet_length(string)intNumber of bytes in stringoctet_length('jose')4
overlay(string placing string from int [for int])text Replace substring overlay('Txxxxas' placing 'hom' from 2 for 4)Thomas
position(substring in string)intLocation of specified substringposition('om' in 'Thomas')3
substring(string [from int] [for int])text Extract substring substring('Thomas' from 2 for 3)hom
substring(string from pattern)text Extract substring matching POSIX regular expression. See Section 9.7 for more information on pattern matching. substring('Thomas' from '...$')mas
substring(string from pattern for escape)text Extract substring matching SQL regular expression. See Section 9.7 for more information on pattern matching. substring('Thomas' from '%#"o_a#"_' for '#')oma
trim([leading | trailing | both] [characters] from string)text Remove the longest string containing only characters from characters (a space by default) from the start, end, or both ends (both is the default) of stringtrim(both 'xyz' from 'yxTomxx')Tom
trim([leading | trailing | both] [from] string [, characters] )text Non-standard syntax for trim()trim(both from 'yxTomxx', 'xyz')Tom
upper(string)textConvert string to upper caseupper('tom')TOM

Additional string manipulation functions are available and are listed in Table 9.9. Some of them are used internally to implement the SQL-standard string functions listed in Table 9.8.

Table 9.9. Other String Functions

FunctionReturn TypeDescriptionExampleResult
ascii(string)intASCII code of the first character of the argument. For UTF8 returns the Unicode code point of the character. For other multibyte encodings, the argument must be an ASCII character. ascii('x')120
btrim(string text [, characters text])text Remove the longest string consisting only of characters in characters (a space by default) from the start and end of stringbtrim('xyxtrimyyx', 'xyz')trim
chr(int)text Character with the given code. For UTF8 the argument is treated as a Unicode code point. For other multibyte encodings the argument must designate an ASCII character. The NULL (0) character is not allowed because text data types cannot store such bytes. chr(65)A
concat(str "any" [, str "any" [, ...] ])text Concatenate the text representations of all the arguments. NULL arguments are ignored. concat('abcde', 2, NULL, 22)abcde222
concat_ws(sep text, str "any" [, str "any" [, ...] ])text Concatenate all but the first argument with separators. The first argument is used as the separator string. NULL arguments are ignored. concat_ws(',', 'abcde', 2, NULL, 22)abcde,2,22
convert(string bytea, src_encoding name, dest_encoding name)bytea Convert string to dest_encoding. The original encoding is specified by src_encoding. The string must be valid in this encoding. Conversions can be defined by CREATE CONVERSION. Also there are some predefined conversions. See Table 9.10 for available conversions. convert('text_in_utf8', 'UTF8', 'LATIN1')text_in_utf8 represented in Latin-1 encoding (ISO 8859-1)
convert_from(string bytea, src_encoding name)text Convert string to the database encoding. The original encoding is specified by src_encoding. The string must be valid in this encoding. convert_from('text_in_utf8', 'UTF8')text_in_utf8 represented in the current database encoding
convert_to(string text, dest_encoding name)bytea Convert string to dest_encoding. convert_to('some text', 'UTF8')some text represented in the UTF8 encoding
decode(string text, format text)bytea Decode binary data from textual representation in string. Options for format are same as in encode. decode('MTIzAAE=', 'base64')\x3132330001
encode(data bytea, format text)text Encode binary data into a textual representation. Supported formats are: base64, hex, escape. escape converts zero bytes and high-bit-set bytes to octal sequences (\nnn) and doubles backslashes. encode('123\000\001', 'base64')MTIzAAE=
format(formatstr text [, formatarg "any" [, ...] ])text Format arguments according to a format string. This function is similar to the C function sprintf. See Section 9.4.1. format('Hello %s, %1$s', 'World')Hello World, World
initcap(string)text Convert the first letter of each word to upper case and the rest to lower case. Words are sequences of alphanumeric characters separated by non-alphanumeric characters. initcap('hi THOMAS')Hi Thomas
left(str text, n int)text Return first n characters in the string. When n is negative, return all but last |n| characters. left('abcde', 2)ab
length(string)int Number of characters in stringlength('jose')4
length(string bytea, encoding name )int Number of characters in string in the given encoding. The string must be valid in this encoding. length('jose', 'UTF8')4
lpad(string text, length int [, fill text])text Fill up the string to length length by prepending the characters fill (a space by default). If the string is already longer than length then it is truncated (on the right). lpad('hi', 5, 'xy')xyxhi
ltrim(string text [, characters text])text Remove the longest string containing only characters from characters (a space by default) from the start of stringltrim('zzzytest', 'xyz')test
md5(string)text Calculates the MD5 hash of string, returning the result in hexadecimal md5('abc')900150983cd24fb0 d6963f7d28e17f72
parse_ident(qualified_identifier text [, strictmode boolean DEFAULT true ] )text[] Split qualified_identifier into an array of identifiers, removing any quoting of individual identifiers. By default, extra characters after the last identifier are considered an error; but if the second parameter is false, then such extra characters are ignored. (This behavior is useful for parsing names for objects like functions.) Note that this function does not truncate over-length identifiers. If you want truncation you can cast the result to name[]. parse_ident('"SomeSchema".someTable'){SomeSchema,sometable}
pg_client_encoding()name Current client encoding name pg_client_encoding()SQL_ASCII
quote_ident(string text)text Return the given string suitably quoted to be used as an identifier in an SQL statement string. Quotes are added only if necessary (i.e., if the string contains non-identifier characters or would be case-folded). Embedded quotes are properly doubled. See also Example 43.1. quote_ident('Foo bar')"Foo bar"
quote_literal(string text)text Return the given string suitably quoted to be used as a string literal in an SQL statement string. Embedded single-quotes and backslashes are properly doubled. Note that quote_literal returns null on null input; if the argument might be null, quote_nullable is often more suitable. See also Example 43.1. quote_literal(E'O\'Reilly')'O''Reilly'
quote_literal(value anyelement)text Coerce the given value to text and then quote it as a literal. Embedded single-quotes and backslashes are properly doubled. quote_literal(42.5)'42.5'
quote_nullable(string text)text Return the given string suitably quoted to be used as a string literal in an SQL statement string; or, if the argument is null, return NULL. Embedded single-quotes and backslashes are properly doubled. See also Example 43.1. quote_nullable(NULL)NULL
quote_nullable(value anyelement)text Coerce the given value to text and then quote it as a literal; or, if the argument is null, return NULL. Embedded single-quotes and backslashes are properly doubled. quote_nullable(42.5)'42.5'
regexp_match(string text, pattern text [, flags text])text[] Return captured substring(s) resulting from the first match of a POSIX regular expression to the string. See Section 9.7.3 for more information. regexp_match('foobarbequebaz', '(bar)(beque)'){bar,beque}
regexp_matches(string text, pattern text [, flags text])setof text[] Return captured substring(s) resulting from matching a POSIX regular expression to the string. See Section 9.7.3 for more information. regexp_matches('foobarbequebaz', 'ba.', 'g'){bar}

{baz} (2 rows)

regexp_replace(string text, pattern text, replacement text [, flags text])text Replace substring(s) matching a POSIX regular expression. See Section 9.7.3 for more information. regexp_replace('Thomas', '.[mN]a.', 'M')ThM
regexp_split_to_array(string text, pattern text [, flags text ])text[] Split string using a POSIX regular expression as the delimiter. See Section 9.7.3 for more information. regexp_split_to_array('hello world', '\s+'){hello,world}
regexp_split_to_table(string text, pattern text [, flags text])setof text Split string using a POSIX regular expression as the delimiter. See Section 9.7.3 for more information. regexp_split_to_table('hello world', '\s+')hello

world (2 rows)

repeat(string text, number int)textRepeat string the specified number of timesrepeat('Pg', 4)PgPgPgPg
replace(string text, from text, to text)textReplace all occurrences in string of substring from with substring toreplace('abcdefabcdef', 'cd', 'XX')abXXefabXXef
reverse(str)text Return reversed string. reverse('abcde')edcba
right(str text, n int)text Return last n characters in the string. When n is negative, return all but first |n| characters. right('abcde', 2)de
rpad(string text, length int [, fill text])text Fill up the string to length length by appending the characters fill (a space by default). If the string is already longer than length then it is truncated. rpad('hi', 5, 'xy')hixyx
rtrim(string text [, characters text])text Remove the longest string containing only characters from characters (a space by default) from the end of stringrtrim('testxxzx', 'xyz')test
split_part(string text, delimiter text, field int)textSplit string on delimiter and return the given field (counting from one) split_part('abc~@~def~@~ghi', '~@~', 2)def
strpos(string, substring)int Location of specified substring (same as position(substring in string), but note the reversed argument order) strpos('high', 'ig')2
substr(string, from [, count])text Extract substring (same as substring(string from from for count)) substr('alphabet', 3, 2)ph
starts_with(string, prefix)bool Returns true if string starts with prefix. starts_with('alphabet', 'alph')t
to_ascii(string text [, encoding text])text Convert string to ASCII from another encoding (only supports conversion from LATIN1, LATIN2, LATIN9, and WIN1250 encodings) to_ascii('Karel')Karel
to_hex(number int or bigint)textConvert number to its equivalent hexadecimal representation to_hex(2147483647)7fffffff
translate(string text, from text, to text)text Any character in string that matches a character in the from set is replaced by the corresponding character in the to set. If from is longer than to, occurrences of the extra characters in from are removed. translate('12345', '143', 'ax')a2x5

The concat, concat_ws and format functions are variadic, so it is possible to pass the values to be concatenated or formatted as an array marked with the VARIADIC keyword (see Section 38.5.5). The array's elements are treated as if they were separate ordinary arguments to the function. If the variadic array argument is NULL, concat and concat_ws return NULL, but format treats a NULL as a zero-element array.

See also the aggregate function string_agg in Section 9.20.

Table 9.10. Built-in Conversions

Conversion Name [a]Source EncodingDestination Encoding
ascii_to_micSQL_ASCIIMULE_INTERNAL
ascii_to_utf8SQL_ASCIIUTF8
big5_to_euc_twBIG5EUC_TW
big5_to_micBIG5MULE_INTERNAL
big5_to_utf8BIG5UTF8
euc_cn_to_micEUC_CNMULE_INTERNAL
euc_cn_to_utf8EUC_CNUTF8
euc_jp_to_micEUC_JPMULE_INTERNAL
euc_jp_to_sjisEUC_JPSJIS
euc_jp_to_utf8EUC_JPUTF8
euc_kr_to_micEUC_KRMULE_INTERNAL
euc_kr_to_utf8EUC_KRUTF8
euc_tw_to_big5EUC_TWBIG5
euc_tw_to_micEUC_TWMULE_INTERNAL
euc_tw_to_utf8EUC_TWUTF8
gb18030_to_utf8GB18030UTF8
gbk_to_utf8GBKUTF8
iso_8859_10_to_utf8LATIN6UTF8
iso_8859_13_to_utf8LATIN7UTF8
iso_8859_14_to_utf8LATIN8UTF8
iso_8859_15_to_utf8LATIN9UTF8
iso_8859_16_to_utf8LATIN10UTF8
iso_8859_1_to_micLATIN1MULE_INTERNAL
iso_8859_1_to_utf8LATIN1UTF8
iso_8859_2_to_micLATIN2MULE_INTERNAL
iso_8859_2_to_utf8LATIN2UTF8
iso_8859_2_to_windows_1250LATIN2WIN1250
iso_8859_3_to_micLATIN3MULE_INTERNAL
iso_8859_3_to_utf8LATIN3UTF8
iso_8859_4_to_micLATIN4MULE_INTERNAL
iso_8859_4_to_utf8LATIN4UTF8
iso_8859_5_to_koi8_rISO_8859_5KOI8R
iso_8859_5_to_micISO_8859_5MULE_INTERNAL
iso_8859_5_to_utf8ISO_8859_5UTF8
iso_8859_5_to_windows_1251ISO_8859_5WIN1251
iso_8859_5_to_windows_866ISO_8859_5WIN866
iso_8859_6_to_utf8ISO_8859_6UTF8
iso_8859_7_to_utf8ISO_8859_7UTF8
iso_8859_8_to_utf8ISO_8859_8UTF8
iso_8859_9_to_utf8LATIN5UTF8
johab_to_utf8JOHABUTF8
koi8_r_to_iso_8859_5KOI8RISO_8859_5
koi8_r_to_micKOI8RMULE_INTERNAL
koi8_r_to_utf8KOI8RUTF8
koi8_r_to_windows_1251KOI8RWIN1251
koi8_r_to_windows_866KOI8RWIN866
koi8_u_to_utf8KOI8UUTF8
mic_to_asciiMULE_INTERNALSQL_ASCII
mic_to_big5MULE_INTERNALBIG5
mic_to_euc_cnMULE_INTERNALEUC_CN
mic_to_euc_jpMULE_INTERNALEUC_JP
mic_to_euc_krMULE_INTERNALEUC_KR
mic_to_euc_twMULE_INTERNALEUC_TW
mic_to_iso_8859_1MULE_INTERNALLATIN1
mic_to_iso_8859_2MULE_INTERNALLATIN2
mic_to_iso_8859_3MULE_INTERNALLATIN3
mic_to_iso_8859_4MULE_INTERNALLATIN4
mic_to_iso_8859_5MULE_INTERNALISO_8859_5
mic_to_koi8_rMULE_INTERNALKOI8R
mic_to_sjisMULE_INTERNALSJIS
mic_to_windows_1250MULE_INTERNALWIN1250
mic_to_windows_1251MULE_INTERNALWIN1251
mic_to_windows_866MULE_INTERNALWIN866
sjis_to_euc_jpSJISEUC_JP
sjis_to_micSJISMULE_INTERNAL
sjis_to_utf8SJISUTF8
tcvn_to_utf8WIN1258UTF8
uhc_to_utf8UHCUTF8
utf8_to_asciiUTF8SQL_ASCII
utf8_to_big5UTF8BIG5
utf8_to_euc_cnUTF8EUC_CN
utf8_to_euc_jpUTF8EUC_JP
utf8_to_euc_krUTF8EUC_KR
utf8_to_euc_twUTF8EUC_TW
utf8_to_gb18030UTF8GB18030
utf8_to_gbkUTF8GBK
utf8_to_iso_8859_1UTF8LATIN1
utf8_to_iso_8859_10UTF8LATIN6
utf8_to_iso_8859_13UTF8LATIN7
utf8_to_iso_8859_14UTF8LATIN8
utf8_to_iso_8859_15UTF8LATIN9
utf8_to_iso_8859_16UTF8LATIN10
utf8_to_iso_8859_2UTF8LATIN2
utf8_to_iso_8859_3UTF8LATIN3
utf8_to_iso_8859_4UTF8LATIN4
utf8_to_iso_8859_5UTF8ISO_8859_5
utf8_to_iso_8859_6UTF8ISO_8859_6
utf8_to_iso_8859_7UTF8ISO_8859_7
utf8_to_iso_8859_8UTF8ISO_8859_8
utf8_to_iso_8859_9UTF8LATIN5
utf8_to_johabUTF8JOHAB
utf8_to_koi8_rUTF8KOI8R
utf8_to_koi8_uUTF8KOI8U
utf8_to_sjisUTF8SJIS
utf8_to_tcvnUTF8WIN1258
utf8_to_uhcUTF8UHC
utf8_to_windows_1250UTF8WIN1250
utf8_to_windows_1251UTF8WIN1251
utf8_to_windows_1252UTF8WIN1252
utf8_to_windows_1253UTF8WIN1253
utf8_to_windows_1254UTF8WIN1254
utf8_to_windows_1255UTF8WIN1255
utf8_to_windows_1256UTF8WIN1256
utf8_to_windows_1257UTF8WIN1257
utf8_to_windows_866UTF8WIN866
utf8_to_windows_874UTF8WIN874
windows_1250_to_iso_8859_2WIN1250LATIN2
windows_1250_to_micWIN1250MULE_INTERNAL
windows_1250_to_utf8WIN1250UTF8
windows_1251_to_iso_8859_5WIN1251ISO_8859_5
windows_1251_to_koi8_rWIN1251KOI8R
windows_1251_to_micWIN1251MULE_INTERNAL
windows_1251_to_utf8WIN1251UTF8
windows_1251_to_windows_866WIN1251WIN866
windows_1252_to_utf8WIN1252UTF8
windows_1256_to_utf8WIN1256UTF8
windows_866_to_iso_8859_5WIN866ISO_8859_5
windows_866_to_koi8_rWIN866KOI8R
windows_866_to_micWIN866MULE_INTERNAL
windows_866_to_utf8WIN866UTF8
windows_866_to_windows_1251WIN866WIN
windows_874_to_utf8WIN874UTF8
euc_jis_2004_to_utf8EUC_JIS_2004UTF8
utf8_to_euc_jis_2004UTF8EUC_JIS_2004
shift_jis_2004_to_utf8SHIFT_JIS_2004UTF8
utf8_to_shift_jis_2004UTF8SHIFT_JIS_2004
euc_jis_2004_to_shift_jis_2004EUC_JIS_2004SHIFT_JIS_2004
shift_jis_2004_to_euc_jis_2004SHIFT_JIS_2004EUC_JIS_2004

[a] The conversion names follow a standard naming scheme: The official name of the source encoding with all non-alphanumeric characters replaced by underscores, followed by _to_, followed by the similarly processed destination encoding name. Therefore, the names might deviate from the customary encoding names.


9.4.1. format

The function format produces output formatted according to a format string, in a style similar to the C function sprintf.

format(formatstr text [, formatarg "any" [, ...] ])

formatstr is a format string that specifies how the result should be formatted. Text in the format string is copied directly to the result, except where format specifiers are used. Format specifiers act as placeholders in the string, defining how subsequent function arguments should be formatted and inserted into the result. Each formatarg argument is converted to text according to the usual output rules for its data type, and then formatted and inserted into the result string according to the format specifier(s).

Format specifiers are introduced by a % character and have the form

%[position][flags][width]type

where the component fields are:

position (optional)

A string of the form n$ where n is the index of the argument to print. Index 1 means the first argument after formatstr. If the position is omitted, the default is to use the next argument in sequence.

flags (optional)

Additional options controlling how the format specifier's output is formatted. Currently the only supported flag is a minus sign (-) which will cause the format specifier's output to be left-justified. This has no effect unless the width field is also specified.

width (optional)

Specifies the minimum number of characters to use to display the format specifier's output. The output is padded on the left or right (depending on the - flag) with spaces as needed to fill the width. A too-small width does not cause truncation of the output, but is simply ignored. The width may be specified using any of the following: a positive integer; an asterisk (*) to use the next function argument as the width; or a string of the form *n$ to use the nth function argument as the width.

If the width comes from a function argument, that argument is consumed before the argument that is used for the format specifier's value. If the width argument is negative, the result is left aligned (as if the - flag had been specified) within a field of length abs(width).

type (required)

The type of format conversion to use to produce the format specifier's output. The following types are supported:

  • s formats the argument value as a simple string. A null value is treated as an empty string.

  • I treats the argument value as an SQL identifier, double-quoting it if necessary. It is an error for the value to be null (equivalent to quote_ident).

  • L quotes the argument value as an SQL literal. A null value is displayed as the string NULL, without quotes (equivalent to quote_nullable).

In addition to the format specifiers described above, the special sequence %% may be used to output a literal % character.

Here are some examples of the basic format conversions:

SELECT format('Hello %s', 'World');
Result: Hello World

SELECT format('Testing %s, %s, %s, %%', 'one', 'two', 'three');
Result: Testing one, two, three, %

SELECT format('INSERT INTO %I VALUES(%L)', 'Foo bar', E'O\'Reilly');
Result: INSERT INTO "Foo bar" VALUES('O''Reilly')

SELECT format('INSERT INTO %I VALUES(%L)', 'locations', 'C:\Program Files');
Result: INSERT INTO locations VALUES('C:\Program Files')

Here are examples using width fields and the - flag:

SELECT format('|%10s|', 'foo');
Result: |       foo|

SELECT format('|%-10s|', 'foo');
Result: |foo       |

SELECT format('|%*s|', 10, 'foo');
Result: |       foo|

SELECT format('|%*s|', -10, 'foo');
Result: |foo       |

SELECT format('|%-*s|', 10, 'foo');
Result: |foo       |

SELECT format('|%-*s|', -10, 'foo');
Result: |foo       |

These examples show use of position fields:

SELECT format('Testing %3$s, %2$s, %1$s', 'one', 'two', 'three');
Result: Testing three, two, one

SELECT format('|%*2$s|', 'foo', 10, 'bar');
Result: |       bar|

SELECT format('|%1$*2$s|', 'foo', 10, 'bar');
Result: |       foo|

Unlike the standard C function sprintf, PostgreSQL's format function allows format specifiers with and without position fields to be mixed in the same format string. A format specifier without a position field always uses the next argument after the last argument consumed. In addition, the format function does not require all function arguments to be used in the format string. For example:

SELECT format('Testing %3$s, %2$s, %s', 'one', 'two', 'three');
Result: Testing three, two, three

The %I and %L format specifiers are particularly useful for safely constructing dynamic SQL statements. See Example 43.1.