From 3b6afdd7f9d38c33f21e5211ccbbcd3514312372 Mon Sep 17 00:00:00 2001
From: Neil Conway <neilc@samurai.com>
Date: Fri, 11 May 2007 19:40:08 +0000
Subject: [PATCH] Improvements to the SGML docs for TRUNCATE and CLUSTER.
---
doc/src/sgml/ref/cluster.sgml | 4 +--
doc/src/sgml/ref/truncate.sgml | 61 ++++++++++++++++++----------------
2 files changed, 34 insertions(+), 31 deletions(-)
diff --git a/doc/src/sgml/ref/cluster.sgml b/doc/src/sgml/ref/cluster.sgml
index acb2468f7ba..c181019b255 100644
--- a/doc/src/sgml/ref/cluster.sgml
+++ b/doc/src/sgml/ref/cluster.sgml
@@ -1,5 +1,5 @@
<!--
-$PostgreSQL: pgsql/doc/src/sgml/ref/cluster.sgml,v 1.42 2007/04/08 02:07:35 tgl Exp $
+$PostgreSQL: pgsql/doc/src/sgml/ref/cluster.sgml,v 1.43 2007/05/11 19:40:07 neilc Exp $
PostgreSQL documentation
-->
@@ -45,7 +45,7 @@ CLUSTER
not clustered. That is, no attempt is made to store new or
updated rows according to their index order. (If one wishes, one can
periodically recluster by issuing the command again. Also, setting
- the table's FILLFACTOR storage parameter to less than 100% can aid
+ the table's <literal>FILLFACTOR</literal> storage parameter to less than 100% can aid
in preserving cluster ordering during updates, since updated rows
are preferentially kept on the same page.)
</para>
diff --git a/doc/src/sgml/ref/truncate.sgml b/doc/src/sgml/ref/truncate.sgml
index 271d16af03e..3dca068b457 100644
--- a/doc/src/sgml/ref/truncate.sgml
+++ b/doc/src/sgml/ref/truncate.sgml
@@ -1,5 +1,5 @@
<!--
-$PostgreSQL: pgsql/doc/src/sgml/ref/truncate.sgml,v 1.23 2007/04/07 17:12:15 tgl Exp $
+$PostgreSQL: pgsql/doc/src/sgml/ref/truncate.sgml,v 1.24 2007/05/11 19:40:08 neilc Exp $
PostgreSQL documentation
-->
@@ -31,9 +31,9 @@ TRUNCATE [ TABLE ] <replaceable class="PARAMETER">name</replaceable> [, ...] [ C
<command>TRUNCATE</command> quickly removes all rows from a set of
tables. It has the same effect as an unqualified
<command>DELETE</command> on each table, but since it does not actually
- scan the tables it is faster; furthermore it reclaims disk space
- immediately, rather than requiring a subsequent vacuum operation.
- This is most useful on large tables.
+ scan the tables it is faster. Furthermore, it reclaims disk space
+ immediately, rather than requiring a subsequent <command>VACUUM</command>
+ operation. This is most useful on large tables.
</para>
</refsect1>
@@ -86,35 +86,38 @@ TRUNCATE [ TABLE ] <replaceable class="PARAMETER">name</replaceable> [, ...] [ C
in the same command. Checking validity in such cases would require table
scans, and the whole point is not to do one. The <literal>CASCADE</>
option can be used to automatically include all dependent tables —
- but be very careful when using this option, else you might lose data you
+ but be very careful when using this option, or else you might lose data you
did not intend to!
</para>
<para>
- <command>TRUNCATE</> will not run any user-defined <literal>ON
- DELETE</literal> triggers that might exist for the tables.
+ <command>TRUNCATE</> will not run any <literal>ON DELETE</literal>
+ triggers that might exist for the tables.
</para>
- <para>
- <command>TRUNCATE</> is not MVCC-safe (see <xref linkend="mvcc">
- for general information about MVCC). After truncation, the table
- will appear empty to all concurrent transactions, even if they are
- using a snapshot taken before the truncation occurred. This will
- only be an issue for a transaction that did not touch the table
- before the truncation started — any transaction that has done
- so would hold at least <literal>ACCESS SHARE</literal> lock,
- which would block
- <command>TRUNCATE</> until that transaction completes. So
- truncation will not cause any apparent inconsistency in the table
- contents for successive queries on the same table, but it could
- cause visible inconsistency between the contents of the
- truncated table and other tables.
- </para>
-
- <para>
- <command>TRUNCATE</> is transaction-safe, however: the truncation
- will roll back if the surrounding transaction does not commit.
- </para>
+ <warning>
+ <para>
+ <command>TRUNCATE</> is not MVCC-safe (see <xref linkend="mvcc">
+ for general information about MVCC). After truncation, the table
+ will appear empty to all concurrent transactions, even if they
+ are using a snapshot taken before the truncation occurred. This
+ will only be an issue for a transaction that did not access the
+ truncated table before the truncation happened — any
+ transaction that has done so would hold at least an
+ <literal>ACCESS SHARE</literal> lock, which would block
+ <command>TRUNCATE</> until that transaction completes. So
+ truncation will not cause any apparent inconsistency in the table
+ contents for successive queries on the same table, but it could
+ cause visible inconsistency between the contents of the truncated
+ table and other tables in the database.
+ </para>
+
+ <para>
+ <command>TRUNCATE</> is transaction-safe, however: the truncation
+ will be safely rolled back if the surrounding transaction does not
+ commit.
+ </para>
+ </warning>
</refsect1>
<refsect1>
@@ -124,13 +127,13 @@ TRUNCATE [ TABLE ] <replaceable class="PARAMETER">name</replaceable> [, ...] [ C
Truncate the tables <literal>bigtable</literal> and <literal>fattable</literal>:
<programlisting>
-TRUNCATE TABLE bigtable, fattable;
+TRUNCATE bigtable, fattable;
</programlisting>
</para>
<para>
Truncate the table <literal>othertable</literal>, and cascade to any tables
- that are referencing <literal>othertable</literal> via foreign-key
+ that reference <literal>othertable</literal> via foreign-key
constraints:
<programlisting>
--
GitLab