Author: painter
Date: Thu Oct 11 04:35:08 2018
New Revision: 1843532
URL: http://svn.apache.org/viewvc?rev=1843532&view=rev
Log:
add doc on updated input encoding valve
Modified:
turbine/core/trunk/xdocs/howto/migrate-from-4_0-howto.xml
Modified: turbine/core/trunk/xdocs/howto/migrate-from-4_0-howto.xml
URL:
http://svn.apache.org/viewvc/turbine/core/trunk/xdocs/howto/migrate-from-4_0-howto.xml?rev=1843532&r1=1843531&r2=1843532&view=diff
==============================================================================
--- turbine/core/trunk/xdocs/howto/migrate-from-4_0-howto.xml (original)
+++ turbine/core/trunk/xdocs/howto/migrate-from-4_0-howto.xml Thu Oct 11
04:35:08 2018
@@ -38,43 +38,142 @@
</p>
</section>
+
<section name="Updating configuration">
-<p>
- TurbineResources.properties have changed making it less verbose
- to point to the log4j config file.
-
- <ul>
- <li>Remove the WEB-INF/conf/ prefix from your log4j config file
location</li>
- </ul>
-</p>
-</section>
+ <subsection name="Log4j changes">
+ <p>
+ TurbineResources.properties have changed making it less
verbose
+ to point to the log4j config file.
+
+ Old config line:
+<source>
+<![CDATA[
+ # -------------------------------------------------------------------
+ #
+ # L O G 4 J - L O G G I N G
+ #
+ # -------------------------------------------------------------------
+
+ log4j.file = WEB-INF/conf/log4j.properties
+
+]]>
+ </source>
+
+ New config line:
+<source>
+<![CDATA[
+ # -------------------------------------------------------------------
+ #
+ # L O G 4 J - L O G G I N G
+ #
+ # -------------------------------------------------------------------
+
+ log4j.file = log4j.properties
+]]>
+</source>
-<section name="Migrating file upload to Parts">
+ </p>
-<p>
- In turbine-4.0.1 and prior, file uploads were processed through the
- data.getParameters().getFileItem("file_field_name") method and returned
- a FileItem object.
-</p>
+ </subsection>
-<p>
- With Turbine-5.0, the framework is now using Java servlet 3.1.0.
- As such, you will need to migrate this code using the
- new Part object from the servlet spec. This actually saves you some
- time since you don't have to convert the FileItem to a byte array and
- then into an InputStream for processing.. you auto-magically get an
- getInputStream() method on your javax.servlet.http.Part object to then
- do as you please...
-</p>
+ <subsection name="Setting character encoding">
+ <p>
+ The default character encoding typically should be set
+ by your servlet container. For example, in Tomcat you
can
+ update the server.xml in the following way to force URI
connections
+ to be encoded in UTF-8.
+ </p>
+
+<source>
+<![CDATA[
+ <!-- Define a non-SSL HTTP/1.1 Connector on port 8080 -->
+ <Connector port="8080" maxHttpHeaderSize="8192"
+ maxThreads="150" minSpareThreads="25" maxSpareThreads="75"
+ enableLookups="false" redirectPort="8443" acceptCount="100"
+ connectionTimeout="20000" disableUploadTimeout="true"
+ URIEncoding="UTF-8"
+ />
+]]>
+</source>
+
+ <p>
+ However, earlier versions of Apache Turbine allowed you
to override this
+ value by setting the input.encoding property in your
TurbineResources.properties file.
+ </p>
+
+ <p>
+ With Turbine 5.0, you can still accomplish this same
behavior, but you need
+ to make sure that you have added the encoding valve to
the turbine-classic-pipeline.xml
+ to enable it.
+ </p>
+
+ <p>
+ In TurbineResources.properties, you can set the
following parameter:
+ </p>
<source>
<![CDATA[
+ input.encoding = UTF-8
+]]>
+</source>
+
+ <p>
+ But it will not be picked up unless you edit the
pipeline valves
+ to read as follows. Note that ORDER does matter if you
are not
+ seeing the behavior you expected.
+ </p>
+
+<source>
+<![CDATA[
+ <pipeline name="default">
+ <valves>
+
+ <!-- Use this valve to enable use of the input.encoding
+ parameter defined in your TurbineResources.properties file
-->
+ <valve>org.apache.turbine.pipeline.DefaultSetEncodingValve</valve>
+
+ <valve>org.apache.turbine.pipeline.DetermineActionValve</valve>
+ <valve>org.apache.turbine.pipeline.DetermineTargetValve</valve>
+
<valve>org.apache.turbine.pipeline.DefaultSessionTimeoutValve</valve>
+ <valve>org.apache.turbine.pipeline.DefaultLoginValve</valve>
+
<valve>org.apache.turbine.pipeline.DefaultSessionValidationValve</valve>
+ <valve>org.apache.turbine.pipeline.DefaultACLCreationValve</valve>
+ <valve>org.apache.turbine.pipeline.ExecutePageValve</valve>
+ <valve>org.apache.turbine.pipeline.CleanUpValve</valve>
+
<valve>org.apache.turbine.pipeline.DetermineRedirectRequestedValve</valve>
+ </valves>
+ </pipeline>
+]]>
+</source>
+
+ </subsection>
+</section>
+
+
+<section name="Migrating file upload to Parts">
+
+ <p>
+ In turbine-4.0.1 and prior, file uploads were processed through
the
+ data.getParameters().getFileItem("file_field_name") method and
returned
+ a FileItem object.
+ </p>
+
+ <p>
+ With Turbine-5.0, the framework is now using Java servlet 3.1.0.
+ As such, you will need to migrate this code using the
+ new Part object from the servlet spec. This actually saves you
some
+ time since you don't have to convert the FileItem to a byte
array and
+ then into an InputStream for processing.. you auto-magically
get an
+ getInputStream() method on your javax.servlet.http.Part object
to then
+ do as you please...
+ </p>
+<source>
+<![CDATA[
// all file items are now parts
Part fileItem = data.getParameters().getPart("file");
-
if (fileItem != null) {
InputStream is = fileItem.getInputStream();
@@ -88,26 +187,27 @@
}
} catch (IOException e) {
+ // handle exception here
e.printStackTrace();
} finally {
try {
if (is != null)
is.close();
} catch (Exception ex) {
-
+ // handle exception here
}
}
}
]]>
-</source>
-
-<p>
- And if you really do need a byte array (for example to store the
- contents as a binary object in the database), you can do this using the
- following method calls.
-</p>
-
+ </source>
+
+ <p>
+ And if you really do need a byte array (for example to store the
+ contents as a binary object in the database), you can do this
using the
+ following method calls.
+ </p>
+
<source>
<