<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Ales Justin wrote:
<blockquote cite="mid:4A29AF10.5000902@gmail.com" type="cite">All of
this already exists. ;-)
<br>
<br>
<blockquote type="cite">* You should have a AnnotationScanner
deployer that scans jars for annotations and puts an annotation
database somewhere in memory that other deployers can reference. (This
might already exist, I don't know).
<br>
</blockquote>
<br>
It is called AnnotationEnvironment, and it's part of the deployment's
attachments.
<br>
<br>
It's already used to help Metadata get annotation info.
<br>
But how it's done in Metadata is bad/wrong.
<br>
<br>
<blockquote type="cite">* If a file META-INF/no.scan exists in a jar,
then don't scan the file. This allows libraries to exclude themselves
from scanning.
<br>
<br>
* If a file META-INF/scan.only exists it will contain a newline
delimited list of class names to scan. Only scan those class names.
<br>
</blockquote>
<br>
This is called jboss-scanning.xml.
<br>
- <a class="moz-txt-link-freetext" href="http://www.jboss.org/community/wiki/JBoss5custommetadatafiles">http://www.jboss.org/community/wiki/JBoss5custommetadatafiles</a>
<br>
</blockquote>
I hacked my local trunk (6) build to use jboss-scanning.xml and don't
see much of an impact on boot performance. Prior to the
jboss-scanning.xml changes, we were scanning about 600 classes for
annotations. After the changes, we scan zero classes for annotations
(as per trace logging in o.j.d.p.a.GenericAnnotationResourceVisitor).
My local timings are:<br>
<br>
Before the change:<br>
<br>
<meta http-equiv="CONTENT-TYPE" content="text/html; charset=utf-8">
<title></title>
<meta name="GENERATOR" content="OpenOffice.org 2.3 (Linux)">
<style type="text/css">
<!--
@page { size: 8.5in 11in; margin: 0.79in }
P { margin-bottom: 0.08in }
-->
</style>
<p><span style="text-decoration: none;"> Started in
35s:304ms</span></p>
<p><span style="text-decoration: none;"> Started in 33s:229ms</span></p>
<p><span style="text-decoration: none;"> Started in 34s:445ms</span></p>
<p>After adding jboss-scanning.xml:<br>
</p>
<p>
<meta http-equiv="CONTENT-TYPE" content="text/html; charset=utf-8">
<title></title>
<meta name="GENERATOR" content="OpenOffice.org 2.3 (Linux)">
<style type="text/css">
<!--
@page { size: 8.5in 11in; margin: 0.79in }
P { margin-bottom: 0.08in }
-->
</style>
<p><span style="text-decoration: none;"> Started in
33s:324ms</span></p>
<p><span style="text-decoration: none;"> Started in
34s:163ms</span></p>
<p><span style="text-decoration: none;"> Started in
33s:957ms</span></p>
</p>
<br>
<br>
<blockquote cite="mid:4A29AF10.5000902@gmail.com" type="cite"><br>
We already exclude all of our boot.
<br>
See metadata-deployer-jboss-beans.xml:
<br>
-
<a class="moz-txt-link-freetext" href="http://anonsvn.jboss.org/repos/jbossas/branches/Branch_5_x/server/src/etc/deployers/metadata-deployer-jboss-beans.xml">http://anonsvn.jboss.org/repos/jbossas/branches/Branch_5_x/server/src/etc/deployers/metadata-deployer-jboss-beans.xml</a>
<br>
</blockquote>
<br>
FYI, JBAS-7025 is about excluding our boot with separate
jboss-scanning.xml files, instead of using a central list
(metadata-deployer-jboss-beans.xml).<br>
<br>
<blockquote cite="mid:4A29AF10.5000902@gmail.com" type="cite">_______________________________________________
<br>
jboss-development mailing list
<br>
<a class="moz-txt-link-abbreviated" href="mailto:jboss-development@lists.jboss.org">jboss-development@lists.jboss.org</a>
<br>
<a class="moz-txt-link-freetext" href="https://lists.jboss.org/mailman/listinfo/jboss-development">https://lists.jboss.org/mailman/listinfo/jboss-development</a>
<br>
</blockquote>
<br>
</body>
</html>