postgresql/doc/src/sgml/ref/set_role.sgml

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

160 lines
5.6 KiB
Plaintext
Raw Normal View History

<!--
2010-09-20 22:08:53 +02:00
doc/src/sgml/ref/set_role.sgml
PostgreSQL documentation
-->
<refentry id="sql-set-role">
<indexterm zone="sql-set-role">
<primary>SET ROLE</primary>
</indexterm>
<refmeta>
<refentrytitle>SET ROLE</refentrytitle>
<manvolnum>7</manvolnum>
<refmiscinfo>SQL - Language Statements</refmiscinfo>
</refmeta>
<refnamediv>
<refname>SET ROLE</refname>
<refpurpose>set the current user identifier of the current session</refpurpose>
</refnamediv>
<refsynopsisdiv>
<synopsis>
SET [ SESSION | LOCAL ] ROLE <replaceable class="parameter">role_name</replaceable>
SET [ SESSION | LOCAL ] ROLE NONE
RESET ROLE
</synopsis>
</refsynopsisdiv>
<refsect1>
<title>Description</title>
<para>
This command sets the current user
identifier of the current SQL session to be <replaceable
class="parameter">role_name</replaceable>. The role name can be
written as either an identifier or a string literal.
After <command>SET ROLE</command>, permissions checking for SQL commands
is carried out as though the named role were the one that had logged
in originally. Note that <command>SET ROLE</command> and
<command>SET SESSION AUTHORIZATION</command> are exceptions; permissions
checks for those continue to use the current session user and the initial
session user (the <firstterm>authenticated user</firstterm>), respectively.
</para>
<para>
The current session user must have the <literal>SET</literal> option for the
specified <replaceable class="parameter">role_name</replaceable>, either
directly or indirectly via a chain of memberships with the
<literal>SET</literal> option.
(If the session user is a superuser, any role can be selected.)
</para>
<para>
The <literal>SESSION</literal> and <literal>LOCAL</literal> modifiers act the same
Improve <xref> vs. <command> formatting in the documentation SQL commands are generally marked up as <command>, except when a link to a reference page is used using <xref>. But the latter doesn't create monospace markup, so this looks strange especially when a paragraph contains a mix of links and non-links. We considered putting <command> in the <refentrytitle> on the target side, but that creates some formatting side effects elsewhere. Generally, it seems safer to solve this on the link source side. We can't put the <xref> inside the <command>; the DTD doesn't allow this. DocBook 5 would allow the <command> to have the linkend attribute itself, but we are not there yet. So to solve this for now, convert the <xref>s to <link> plus <command>. This gives the correct look and also gives some more flexibility what we can put into the link text (e.g., subcommands or other clauses). In the future, these could then be converted to DocBook 5 style. I haven't converted absolutely all xrefs to SQL command reference pages, only those where we care about the appearance of the link text or where it was otherwise appropriate to make the appearance match a bit better. Also in some cases, the links where repetitive, so in those cases the links where just removed and replaced by a plain <command>. In cases where we just want the link and don't specifically care about the generated link text (typically phrased "for further information see <xref ...>") the xref is kept. Reported-by: Dagfinn Ilmari Mannsåker <ilmari@ilmari.org> Discussion: https://www.postgresql.org/message-id/flat/87o8pco34z.fsf@wibble.ilmari.org
2020-10-03 16:16:51 +02:00
as for the regular <link linkend="sql-set"><command>SET</command></link>
command.
</para>
<para>
<literal>SET ROLE NONE</literal> sets the current user identifier to the
current session user identifier, as returned by
<function>session_user</function>. <literal>RESET ROLE</literal> sets the
current user identifier to the connection-time setting specified by the
<link linkend="libpq-connect-options">command-line options</link>,
<link linkend="sql-alterrole"><command>ALTER ROLE</command></link>, or
<link linkend="sql-alterdatabase"><command>ALTER DATABASE</command></link>,
if any such settings exist. Otherwise, <literal>RESET ROLE</literal> sets
the current user identifier to the current session user identifier. These
forms can be executed by any user.
</para>
</refsect1>
<refsect1>
<title>Notes</title>
<para>
Using this command, it is possible to either add privileges or restrict
one's privileges. If the session user role has been granted memberships
<literal>WITH INHERIT TRUE</literal>, it automatically has all the
privileges of every such role. In this case, <command>SET ROLE</command>
effectively drops all the privileges except for those which the target role
directly possesses or inherits. On the other hand, if the session user role
has been granted memberships <literal>WITH INHERIT FALSE</literal>, the
privileges of the granted roles can't be accessed by default. However, if
the role was granted <literal>WITH SET TRUE</literal>, the
session user can use <command>SET ROLE</command> to drop the privileges
assigned directly to the session user and instead acquire the privileges
available to the named role. If the role was granted <literal>WITH INHERIT
FALSE, SET FALSE</literal> then the privileges of that role cannot be
exercised either with or without <literal>SET ROLE</literal>.
</para>
<para>
<command>SET ROLE</command> has effects comparable to
Improve <xref> vs. <command> formatting in the documentation SQL commands are generally marked up as <command>, except when a link to a reference page is used using <xref>. But the latter doesn't create monospace markup, so this looks strange especially when a paragraph contains a mix of links and non-links. We considered putting <command> in the <refentrytitle> on the target side, but that creates some formatting side effects elsewhere. Generally, it seems safer to solve this on the link source side. We can't put the <xref> inside the <command>; the DTD doesn't allow this. DocBook 5 would allow the <command> to have the linkend attribute itself, but we are not there yet. So to solve this for now, convert the <xref>s to <link> plus <command>. This gives the correct look and also gives some more flexibility what we can put into the link text (e.g., subcommands or other clauses). In the future, these could then be converted to DocBook 5 style. I haven't converted absolutely all xrefs to SQL command reference pages, only those where we care about the appearance of the link text or where it was otherwise appropriate to make the appearance match a bit better. Also in some cases, the links where repetitive, so in those cases the links where just removed and replaced by a plain <command>. In cases where we just want the link and don't specifically care about the generated link text (typically phrased "for further information see <xref ...>") the xref is kept. Reported-by: Dagfinn Ilmari Mannsåker <ilmari@ilmari.org> Discussion: https://www.postgresql.org/message-id/flat/87o8pco34z.fsf@wibble.ilmari.org
2020-10-03 16:16:51 +02:00
<link linkend="sql-set-session-authorization"><command>SET SESSION AUTHORIZATION</command></link>, but the privilege
checks involved are quite different. Also,
<command>SET SESSION AUTHORIZATION</command> determines which roles are
allowable for later <command>SET ROLE</command> commands, whereas changing
roles with <command>SET ROLE</command> does not change the set of roles
allowed to a later <command>SET ROLE</command>.
</para>
<para>
<command>SET ROLE</command> does not process session variables as specified by
Improve <xref> vs. <command> formatting in the documentation SQL commands are generally marked up as <command>, except when a link to a reference page is used using <xref>. But the latter doesn't create monospace markup, so this looks strange especially when a paragraph contains a mix of links and non-links. We considered putting <command> in the <refentrytitle> on the target side, but that creates some formatting side effects elsewhere. Generally, it seems safer to solve this on the link source side. We can't put the <xref> inside the <command>; the DTD doesn't allow this. DocBook 5 would allow the <command> to have the linkend attribute itself, but we are not there yet. So to solve this for now, convert the <xref>s to <link> plus <command>. This gives the correct look and also gives some more flexibility what we can put into the link text (e.g., subcommands or other clauses). In the future, these could then be converted to DocBook 5 style. I haven't converted absolutely all xrefs to SQL command reference pages, only those where we care about the appearance of the link text or where it was otherwise appropriate to make the appearance match a bit better. Also in some cases, the links where repetitive, so in those cases the links where just removed and replaced by a plain <command>. In cases where we just want the link and don't specifically care about the generated link text (typically phrased "for further information see <xref ...>") the xref is kept. Reported-by: Dagfinn Ilmari Mannsåker <ilmari@ilmari.org> Discussion: https://www.postgresql.org/message-id/flat/87o8pco34z.fsf@wibble.ilmari.org
2020-10-03 16:16:51 +02:00
the role's <link linkend="sql-alterrole"><command>ALTER ROLE</command></link> settings; this only happens during
login.
</para>
<para>
<command>SET ROLE</command> cannot be used within a
<literal>SECURITY DEFINER</literal> function.
</para>
</refsect1>
<refsect1>
<title>Examples</title>
<programlisting>
SELECT SESSION_USER, CURRENT_USER;
session_user | current_user
--------------+--------------
peter | peter
SET ROLE 'paul';
SELECT SESSION_USER, CURRENT_USER;
session_user | current_user
--------------+--------------
peter | paul
</programlisting>
</refsect1>
<refsect1>
<title>Compatibility</title>
<para>
<productname>PostgreSQL</productname>
allows identifier syntax (<literal>"<replaceable>rolename</replaceable>"</literal>), while
the SQL standard requires the role name to be written as a string
literal. SQL does not allow this command during a transaction;
<productname>PostgreSQL</productname> does not make this
restriction because there is no reason to.
The <literal>SESSION</literal> and <literal>LOCAL</literal> modifiers are a
<productname>PostgreSQL</productname> extension, as is the
<literal>RESET</literal> syntax.
</para>
</refsect1>
<refsect1>
<title>See Also</title>
<simplelist type="inline">
<member><xref linkend="sql-set-session-authorization"/></member>
</simplelist>
</refsect1>
</refentry>