1 <?xml version="1.0" encoding="UTF-8" standalone="no"?>
2 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>F.48. unaccent — a text search dictionary which removes diacritics</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="tsm-system-time.html" title="F.47. tsm_system_time — the SYSTEM_TIME sampling method for TABLESAMPLE" /><link rel="next" href="uuid-ossp.html" title="F.49. uuid-ossp — a UUID generator" /></head><body id="docContent" class="container-fluid col-10"><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">F.48. unaccent — a text search dictionary which removes diacritics</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="tsm-system-time.html" title="F.47. tsm_system_time — the SYSTEM_TIME sampling method for TABLESAMPLE">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="contrib.html" title="Appendix F. Additional Supplied Modules and Extensions">Up</a></td><th width="60%" align="center">Appendix F. Additional Supplied Modules and Extensions</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 18.0 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="uuid-ossp.html" title="F.49. uuid-ossp — a UUID generator">Next</a></td></tr></table><hr /></div><div class="sect1" id="UNACCENT"><div class="titlepage"><div><div><h2 class="title" style="clear: both">F.48. unaccent — a text search dictionary which removes diacritics <a href="#UNACCENT" class="id_link">#</a></h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="unaccent.html#UNACCENT-CONFIGURATION">F.48.1. Configuration</a></span></dt><dt><span class="sect2"><a href="unaccent.html#UNACCENT-USAGE">F.48.2. Usage</a></span></dt><dt><span class="sect2"><a href="unaccent.html#UNACCENT-FUNCTIONS">F.48.3. Functions</a></span></dt></dl></div><a id="id-1.11.7.58.2" class="indexterm"></a><p>
3 <code class="filename">unaccent</code> is a text search dictionary that removes accents
4 (diacritic signs) from lexemes.
5 It's a filtering dictionary, which means its output is
6 always passed to the next dictionary (if any), unlike the normal
7 behavior of dictionaries. This allows accent-insensitive processing
10 The current implementation of <code class="filename">unaccent</code> cannot be used as a
11 normalizing dictionary for the <code class="filename">thesaurus</code> dictionary.
13 This module is considered <span class="quote">“<span class="quote">trusted</span>”</span>, that is, it can be
14 installed by non-superusers who have <code class="literal">CREATE</code> privilege
15 on the current database.
16 </p><div class="sect2" id="UNACCENT-CONFIGURATION"><div class="titlepage"><div><div><h3 class="title">F.48.1. Configuration <a href="#UNACCENT-CONFIGURATION" class="id_link">#</a></h3></div></div></div><p>
17 An <code class="literal">unaccent</code> dictionary accepts the following options:
18 </p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem"><p>
19 <code class="literal">RULES</code> is the base name of the file containing the list of
20 translation rules. This file must be stored in
21 <code class="filename">$SHAREDIR/tsearch_data/</code> (where <code class="literal">$SHAREDIR</code> means
22 the <span class="productname">PostgreSQL</span> installation's shared-data directory).
23 Its name must end in <code class="literal">.rules</code> (which is not to be included in
24 the <code class="literal">RULES</code> parameter).
25 </p></li></ul></div><p>
26 The rules file has the following format:
27 </p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem"><p>
28 Each line represents one translation rule, consisting of a character with
29 accent followed by a character without accent. The first is translated
30 into the second. For example,
31 </p><pre class="programlisting">
40 The two characters must be separated by whitespace, and any leading or
41 trailing whitespace on a line is ignored.
42 </p></li><li class="listitem"><p>
43 Alternatively, if only one character is given on a line, instances of
44 that character are deleted; this is useful in languages where accents
45 are represented by separate characters.
46 </p></li><li class="listitem"><p>
47 Actually, each <span class="quote">“<span class="quote">character</span>”</span> can be any string not containing
48 whitespace, so <code class="filename">unaccent</code> dictionaries could be used for
49 other sorts of substring substitutions besides diacritic removal.
50 </p></li><li class="listitem"><p>
51 Some characters, like numeric symbols, may require whitespaces in their
52 translation rule. It is possible to use double quotes around the translated
53 characters in this case. A double quote needs to be escaped with a second
54 double quote when including one in the translated character. For example:
55 </p><pre class="programlisting">
62 </p></li><li class="listitem"><p>
63 As with other <span class="productname">PostgreSQL</span> text search configuration files,
64 the rules file must be stored in UTF-8 encoding. The data is
65 automatically translated into the current database's encoding when
66 loaded. Any lines containing untranslatable characters are silently
67 ignored, so that rules files can contain rules that are not applicable in
69 </p></li></ul></div><p>
70 A more complete example, which is directly useful for most European
71 languages, can be found in <code class="filename">unaccent.rules</code>, which is installed
72 in <code class="filename">$SHAREDIR/tsearch_data/</code> when the <code class="filename">unaccent</code>
73 module is installed. This rules file translates characters with accents
74 to the same characters without accents, and it also expands ligatures
75 into the equivalent series of simple characters (for example, Æ to
77 </p></div><div class="sect2" id="UNACCENT-USAGE"><div class="titlepage"><div><div><h3 class="title">F.48.2. Usage <a href="#UNACCENT-USAGE" class="id_link">#</a></h3></div></div></div><p>
78 Installing the <code class="literal">unaccent</code> extension creates a text
79 search template <code class="literal">unaccent</code> and a dictionary <code class="literal">unaccent</code>
80 based on it. The <code class="literal">unaccent</code> dictionary has the default
81 parameter setting <code class="literal">RULES='unaccent'</code>, which makes it immediately
82 usable with the standard <code class="filename">unaccent.rules</code> file.
83 If you wish, you can alter the parameter, for example
85 </p><pre class="programlisting">
86 mydb=# ALTER TEXT SEARCH DICTIONARY unaccent (RULES='my_rules');
89 or create new dictionaries based on the template.
91 To test the dictionary, you can try:
92 </p><pre class="programlisting">
93 mydb=# select ts_lexize('unaccent','Hôtel');
100 Here is an example showing how to insert the
101 <code class="filename">unaccent</code> dictionary into a text search configuration:
102 </p><pre class="programlisting">
103 mydb=# CREATE TEXT SEARCH CONFIGURATION fr ( COPY = french );
104 mydb=# ALTER TEXT SEARCH CONFIGURATION fr
105 ALTER MAPPING FOR hword, hword_part, word
106 WITH unaccent, french_stem;
107 mydb=# select to_tsvector('fr','Hôtels de la Mer');
113 mydb=# select to_tsvector('fr','Hôtel de la Mer') @@ to_tsquery('fr','Hotels');
119 mydb=# select ts_headline('fr','Hôtel de la Mer',to_tsquery('fr','Hotels'));
121 ------------------------
122 <b>Hôtel</b> de la Mer
125 </p></div><div class="sect2" id="UNACCENT-FUNCTIONS"><div class="titlepage"><div><div><h3 class="title">F.48.3. Functions <a href="#UNACCENT-FUNCTIONS" class="id_link">#</a></h3></div></div></div><p>
126 The <code class="function">unaccent()</code> function removes accents (diacritic signs) from
127 a given string. Basically, it's a wrapper around
128 <code class="filename">unaccent</code>-type dictionaries, but it can be used outside normal
129 text search contexts.
130 </p><a id="id-1.11.7.58.8.3" class="indexterm"></a><pre class="synopsis">
131 unaccent([<span class="optional"><em class="replaceable"><code>dictionary</code></em> <code class="type">regdictionary</code>, </span>] <em class="replaceable"><code>string</code></em> <code class="type">text</code>) returns <code class="type">text</code>
133 If the <em class="replaceable"><code>dictionary</code></em> argument is
134 omitted, the text search dictionary named <code class="literal">unaccent</code> and
135 appearing in the same schema as the <code class="function">unaccent()</code>
136 function itself is used.
139 </p><pre class="programlisting">
140 SELECT unaccent('unaccent', 'Hôtel');
141 SELECT unaccent('Hôtel');
143 </p></div></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="tsm-system-time.html" title="F.47. tsm_system_time — the SYSTEM_TIME sampling method for TABLESAMPLE">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="contrib.html" title="Appendix F. Additional Supplied Modules and Extensions">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="uuid-ossp.html" title="F.49. uuid-ossp — a UUID generator">Next</a></td></tr><tr><td width="40%" align="left" valign="top">F.47. tsm_system_time —
144 the <code class="literal">SYSTEM_TIME</code> sampling method for <code class="literal">TABLESAMPLE</code> </td><td width="20%" align="center"><a accesskey="h" href="index.html" title="PostgreSQL 18.0 Documentation">Home</a></td><td width="40%" align="right" valign="top"> F.49. uuid-ossp — a UUID generator</td></tr></table></div></body></html>