path: root/doc/src/sgml/ref/create_sequence.sgml

<!--
doc/src/sgml/ref/create_sequence.sgml
PostgreSQL documentation
-->

<refentry id="sql-createsequence">
 <indexterm zone="sql-createsequence">
  <primary>CREATE SEQUENCE</primary>
 </indexterm>

 <refmeta>
  <refentrytitle>CREATE SEQUENCE</refentrytitle>
  <manvolnum>7</manvolnum>
  <refmiscinfo>SQL - Language Statements</refmiscinfo>
 </refmeta>

 <refnamediv>
  <refname>CREATE SEQUENCE</refname>
  <refpurpose>define a new sequence generator</refpurpose>
 </refnamediv>

 <refsynopsisdiv>
<synopsis>
CREATE [ { TEMPORARY | TEMP } | UNLOGGED ] SEQUENCE [ IF NOT EXISTS ] <replaceable class="parameter">name</replaceable>
    [ AS <replaceable class="parameter">data_type</replaceable> ]
    [ INCREMENT [ BY ] <replaceable class="parameter">increment</replaceable> ]
    [ MINVALUE <replaceable class="parameter">minvalue</replaceable> | NO MINVALUE ] [ MAXVALUE <replaceable class="parameter">maxvalue</replaceable> | NO MAXVALUE ]
    [ START [ WITH ] <replaceable class="parameter">start</replaceable> ] [ CACHE <replaceable class="parameter">cache</replaceable> ] [ [ NO ] CYCLE ]
    [ OWNED BY { <replaceable class="parameter">table_name</replaceable>.<replaceable class="parameter">column_name</replaceable> | NONE } ]
</synopsis>
 </refsynopsisdiv>

 <refsect1>
  <title>Description</title>

  <para>
   <command>CREATE SEQUENCE</command> creates a new sequence number
   generator.  This involves creating and initializing a new special
   single-row table with the name <replaceable
   class="parameter">name</replaceable>.  The generator will be
   owned by the user issuing the command.
  </para>

  <para>
   If a schema name is given then the sequence is created in the
   specified schema.  Otherwise it is created in the current schema.
   Temporary sequences exist in a special schema, so a schema name cannot be
   given when creating a temporary sequence.
   The sequence name must be distinct from the name of any other relation
   (table, sequence, index, view, materialized view, or foreign table) in
   the same schema.
  </para>

  <para>
   After a sequence is created, you use the functions
   <function>nextval</function>,
   <function>currval</function>, and
   <function>setval</function>
   to operate on the sequence.  These functions are documented in
   <xref linkend="functions-sequence"/>.
  </para>

  <para>
   Although you cannot update a sequence directly, you can use a query like:

<programlisting>
SELECT * FROM <replaceable>name</replaceable>;
</programlisting>

   to examine the parameters and current state of a sequence.  In particular,
   the <literal>last_value</literal> field of the sequence shows the last value
   allocated by any session.  (Of course, this value might be obsolete
   by the time it's printed, if other sessions are actively doing
   <function>nextval</function> calls.)
  </para>
 </refsect1>

 <refsect1>
  <title>Parameters</title>

  <variablelist>
   <varlistentry>
    <term><literal>TEMPORARY</literal> or <literal>TEMP</literal></term>
    <listitem>
     <para>
      If specified, the sequence object is created only for this
      session, and is automatically dropped on session exit.  Existing
      permanent sequences with the same name are not visible (in this
      session) while the temporary sequence exists, unless they are
      referenced with schema-qualified names.
     </para>
    </listitem>
   </varlistentry>

   <varlistentry>
    <term><literal>UNLOGGED</literal></term>
    <listitem>
     <para>
      If specified, the sequence is created as an unlogged sequence.  Changes
      to unlogged sequences are not written to the write-ahead log.  They are
      not crash-safe: an unlogged sequence is automatically reset to its
      initial state after a crash or unclean shutdown.  Unlogged sequences are
      also not replicated to standby servers.
     </para>

     <para>
      Unlike unlogged tables, unlogged sequences do not offer a significant
      performance advantage.  This option is mainly intended for sequences
      associated with unlogged tables via identity columns or serial columns.
      In those cases, it usually wouldn't make sense to have the sequence
      WAL-logged and replicated but not its associated table.
     </para>
    </listitem>
   </varlistentry>

   <varlistentry>
    <term><literal>IF NOT EXISTS</literal></term>
    <listitem>
     <para>
      Do not throw an error if a relation with the same name already exists.
      A notice is issued in this case. Note that there is no guarantee that
      the existing relation is anything like the sequence that would have
      been created &mdash; it might not even be a sequence.
     </para>
    </listitem>
   </varlistentry>

   <varlistentry>
    <term><replaceable class="parameter">name</replaceable></term>
    <listitem>
     <para>
      The name (optionally schema-qualified) of the sequence to be created.
     </para>
    </listitem>
   </varlistentry>

   <varlistentry>
    <term><replaceable class="parameter">data_type</replaceable></term>
    <listitem>
     <para>
      The optional
      clause <literal>AS <replaceable class="parameter">data_type</replaceable></literal>
      specifies the data type of the sequence.  Valid types are
      <literal>smallint</literal>, <literal>integer</literal>,
      and <literal>bigint</literal>.  <literal>bigint</literal> is the
      default.  The data type determines the default minimum and maximum
      values of the sequence.
     </para>
    </listitem>
   </varlistentry>

   <varlistentry>
    <term><replaceable class="parameter">increment</replaceable></term>
    <listitem>
     <para>
      The optional clause <literal>INCREMENT BY <replaceable
      class="parameter">increment</replaceable></literal> specifies
      which value is added to the current sequence value to create a
      new value.  A positive value will make an ascending sequence, a
      negative one a descending sequence.  The default value is 1.
     </para>
    </listitem>
   </varlistentry>

   <varlistentry>
    <term><replaceable class="parameter">minvalue</replaceable></term>
    <term><literal>NO MINVALUE</literal></term>
    <listitem>
     <para>
      The optional clause <literal>MINVALUE <replaceable
      class="parameter">minvalue</replaceable></literal> determines
      the minimum value a sequence can generate. If this clause is not
      supplied or <option>NO MINVALUE</option> is specified, then
      defaults will be used.  The default for an ascending sequence is 1.  The
      default for a descending sequence is the minimum value of the data type.
     </para>
    </listitem>
   </varlistentry>

   <varlistentry>
    <term><replaceable class="parameter">maxvalue</replaceable></term>
    <term><literal>NO MAXVALUE</literal></term>
    <listitem>
     <para>
      The optional clause <literal>MAXVALUE <replaceable
      class="parameter">maxvalue</replaceable></literal> determines
      the maximum value for the sequence. If this clause is not
      supplied or <option>NO MAXVALUE</option> is specified, then
      default values will be used.  The default for an ascending sequence is
      the maximum value of the data type.  The default for a descending
      sequence is -1.
     </para>
    </listitem>
   </varlistentry>

   <varlistentry>
    <term><replaceable class="parameter">start</replaceable></term>
    <listitem>
     <para>
      The optional clause <literal>START WITH <replaceable
      class="parameter">start</replaceable> </literal> allows the
      sequence to begin anywhere.  The default starting value is
      <replaceable class="parameter">minvalue</replaceable> for
      ascending sequences and <replaceable
      class="parameter">maxvalue</replaceable> for descending ones.
     </para>
    </listitem>
   </varlistentry>

   <varlistentry>
    <term><replaceable class="parameter">cache</replaceable></term>
    <listitem>
     <para>
      The optional clause <literal>CACHE <replaceable
      class="parameter">cache</replaceable></literal> specifies how
      many sequence numbers are to be preallocated and stored in
      memory for faster access. The minimum value is 1 (only one value
      can be generated at a time, i.e., no cache), and this is also the
      default.
     </para>
    </listitem>
   </varlistentry>

   <varlistentry>
    <term><literal>CYCLE</literal></term>
    <term><literal>NO CYCLE</literal></term>
    <listitem>
     <para>
      The <literal>CYCLE</literal> option allows the sequence to wrap
      around when the <replaceable
      class="parameter">maxvalue</replaceable> or <replaceable
      class="parameter">minvalue</replaceable> has been reached by an
      ascending or descending sequence respectively. If the limit is
      reached, the next number generated will be the <replaceable
      class="parameter">minvalue</replaceable> or <replaceable
      class="parameter">maxvalue</replaceable>, respectively.
     </para>

     <para>
      If <literal>NO CYCLE</literal> is specified, any calls to
      <function>nextval</function> after the sequence has reached its
      maximum value will return an error.  If neither
      <literal>CYCLE</literal> or <literal>NO CYCLE</literal> are
      specified, <literal>NO CYCLE</literal> is the default.
     </para>
    </listitem>
   </varlistentry>

   <varlistentry>
    <term><literal>OWNED BY</literal> <replaceable class="parameter">table_name</replaceable>.<replaceable class="parameter">column_name</replaceable></term>
    <term><literal>OWNED BY NONE</literal></term>
    <listitem>
     <para>
      The <literal>OWNED BY</literal> option causes the sequence to be
      associated with a specific table column, such that if that column
      (or its whole table) is dropped, the sequence will be automatically
      dropped as well.  The specified table must have the same owner and be in
      the same schema as the sequence.
      <literal>OWNED BY NONE</literal>, the default, specifies that there
      is no such association.
     </para>
    </listitem>
   </varlistentry>
  </variablelist>
 </refsect1>

 <refsect1>
  <title>Notes</title>

  <para>
   Use <command>DROP SEQUENCE</command> to remove a sequence.
  </para>

  <para>
   Sequences are based on <type>bigint</type> arithmetic, so the range
   cannot exceed the range of an eight-byte integer
   (-9223372036854775808 to 9223372036854775807).
  </para>

  <para>
   Because <function>nextval</function> and <function>setval</function> calls are never
   rolled back, sequence objects cannot be used if <quote>gapless</quote>
   assignment of sequence numbers is needed.  It is possible to build
   gapless assignment by using exclusive locking of a table containing a
   counter; but this solution is much more expensive than sequence
   objects, especially if many transactions need sequence numbers
   concurrently.
  </para>

  <para>
   Unexpected results might be obtained if a <replaceable
   class="parameter">cache</replaceable> setting greater than one is
   used for a sequence object that will be used concurrently by
   multiple sessions.  Each session will allocate and cache successive
   sequence values during one access to the sequence object and
   increase the sequence object's <literal>last_value</literal> accordingly.
   Then, the next <replaceable class="parameter">cache</replaceable>-1
   uses of <function>nextval</function> within that session simply return the
   preallocated values without touching the sequence object.  So, any
   numbers allocated but not used within a session will be lost when
   that session ends, resulting in <quote>holes</quote> in the
   sequence.
  </para>

  <para>
   Furthermore, although multiple sessions are guaranteed to allocate
   distinct sequence values, the values might be generated out of
   sequence when all the sessions are considered.  For example, with
   a <replaceable class="parameter">cache</replaceable> setting of 10,
   session A might reserve values 1..10 and return
   <function>nextval</function>=1, then session B might reserve values
   11..20 and return <function>nextval</function>=11 before session A
   has generated <function>nextval</function>=2.  Thus, with a
   <replaceable class="parameter">cache</replaceable> setting of one
   it is safe to assume that <function>nextval</function> values are generated
   sequentially; with a <replaceable
   class="parameter">cache</replaceable> setting greater than one you
   should only assume that the <function>nextval</function> values are all
   distinct, not that they are generated purely sequentially.  Also,
   <literal>last_value</literal> will reflect the latest value reserved by
   any session, whether or not it has yet been returned by
   <function>nextval</function>.
  </para>

  <para>
   Another consideration is that a <function>setval</function> executed on
   such a sequence will not be noticed by other sessions until they
   have used up any preallocated values they have cached.
  </para>
 </refsect1>

 <refsect1>
  <title>Examples</title>

  <para>
   Create an ascending sequence called <literal>serial</literal>, starting at 101:
<programlisting>
CREATE SEQUENCE serial START 101;
</programlisting>
  </para>

  <para>
   Select the next number from this sequence:
<programlisting>
SELECT nextval('serial');

 nextval
---------
     101
</programlisting>
  </para>

  <para>
   Select the next number from this sequence:
<programlisting>
SELECT nextval('serial');

 nextval
---------
     102
</programlisting>
  </para>

  <para>
   Use this sequence in an <command>INSERT</command> command:
<programlisting>
INSERT INTO distributors VALUES (nextval('serial'), 'nothing');
</programlisting>
  </para>

  <para>
   Update the sequence value after a <command>COPY FROM</command>:
<programlisting>
BEGIN;
COPY distributors FROM 'input_file';
SELECT setval('serial', max(id)) FROM distributors;
END;
</programlisting></para>
 </refsect1>

 <refsect1>
  <title>Compatibility</title>

  <para>
   <command>CREATE SEQUENCE</command> conforms to the <acronym>SQL</acronym>
   standard, with the following exceptions:
   <itemizedlist>
    <listitem>
     <para>
      Obtaining the next value is done using the <function>nextval()</function>
      function instead of the standard's <command>NEXT VALUE FOR</command>
      expression.
     </para>
    </listitem>
    <listitem>
     <para>
      The <literal>OWNED BY</literal> clause is a <productname>PostgreSQL</productname>
      extension.
     </para>
    </listitem>
   </itemizedlist></para>
 </refsect1>

 <refsect1>
  <title>See Also</title>

  <simplelist type="inline">
   <member><xref linkend="sql-altersequence"/></member>
   <member><xref linkend="sql-dropsequence"/></member>
  </simplelist>
 </refsect1>

</refentry>