doc'd <timestamp use-db="true|false"/>

git-svn-id: https://svn.jboss.org/repos/hibernate/trunk/Hibernate3/doc@7738 1b8cb986-b30d-0410-93ca-fae66ebed9b2
This commit is contained in:
Steve Ebersole 2005-08-03 20:17:00 +00:00
parent 207f54a546
commit d3c6b9fd42
1 changed files with 18 additions and 2 deletions

View File

@ -1133,7 +1133,7 @@
</para>
</sect2>
<sect2 id="mapping-declaration-timestamp">
<sect2 id="mapping-declaration-timestamp" revision="1" >
<title>timestamp (optional)</title>
<para>
@ -1149,6 +1149,7 @@
<area id="timestamp2" coords="3 70" />
<area id="timestamp3" coords="4 70" />
<area id="timestamp4" coords="5 70" />
<area id="timestamp5" coords="6 70" />
</areaspec>
<programlisting><![CDATA[<timestamp
column="timestamp_column"
@ -1156,6 +1157,7 @@
access="field|property|ClassName"
unsaved-value="null|undefined"
node="element-name|@attribute-name|element/@attribute|."
use-db="true|false"
/>]]></programlisting>
<calloutlist>
<callout arearefs="timestamp1">
@ -1186,12 +1188,26 @@
property value should be used.)
</para>
</callout>
<callout arearefs="timestamp5">
<para>
<literal>use-db</literal> (optional - defaults to <literal>false</literal>):
Should Hibernate use JVM-based timestamp values, or database-based
timestamp values? Database-based timestamps incur an overhead because
Hibernate must hit the database in order to determine the "next value",
but will be safer for use in clustered environments. Note also, that not
all <literal>Dialect</literal>s are known to support retreiving of the
database's current timestamp, while others might be unsafe for usage
due to lack of precision (Oracle 8 for example).
</para>
</callout>
</calloutlist>
</programlistingco>
<para>
Note that <literal>&lt;timestamp&gt;</literal> is equivalent to
<literal>&lt;version type="timestamp"&gt;</literal>.
<literal>&lt;version type="timestamp"&gt;</literal>. And
<literal>&lt;timestamp use-db="true"&gt;</literal> is equivalent to
<literal>&lt;version type="dbtimestamp"&gt;</literal>
</para>
</sect2>