JODE decompiler for Eclipse

Installation

Go to Help -> Software Updates -> Find and Install. Select the “Search for new features to install” option. Press the “New Remote Site…” button and enter “Technoetic” for the name and “http://www.technoetic.com/eclipse/update” for the URL. Press OK, and install the Jode plugin from the subsequent list of features.

Java Util Logging in WebSphere Application Server

Use Java Util Logging in WebSphere Application Server


1. Create a Logger
Logger.getLogger("LoggerName");

Once this function is called, the logger is initialized.

2. logger name choosen

Generally, use package to generate the logger.
For large application, it is better to create in group or component.

such as com.package.COMMAND, com.package.USER, com.package.INVENTORY, com.package.ORDERS.
3. Level
For logging, use error or warning level.
For tracing, use fine, finner.

4. Handler
Do not need specify the handler explicitly. Leave WAS server to handle it.

5. Initialization
Here is the key point and the beauty of WAS.
At server startup, use Servlet initialization method to initialize all the loggers.
This will allow you to view all the logger components at WAS admin console.
Therefore, you have the full ability to control to turn on /off on specific tracing component.

6. Turn on/off tracing component
For the large application, you can always turn on/off tracing at run time without restarting the servers.
At WAS admin console, you can view the tracing components.
Use Runtime tab to turn it on/off for quick problem determination.

Configure Tomcat - Server.xml

<Context path="/testing" docBase="D:/temp/temp/mypath" reloadable="true"></Context>
<Context path="/test" docBase="E:/workspace/soft/myproject/WebContent" reloadable="true"></Context>

</Host>


</Engine>


</Service>


</Server>

Compose an email

Send an email:
Refer to infocenter : "Examples: Outbound messaging system interface".

To compose a jsp file, you need add to VIEWREG table.
Notice that, the deviceId is not -1, it should be the one used for email message. such as -3. Check database for detail.
Also, the interfacename and classname are different as regular viewreg.

Example to insert a new view:
-----------------------------
insert into viewreg (
VIEWNAME, DEVICEFMT_ID, STOREENT_ID, INTERFACENAME, CLASSNAME, PROPERTIES)values(
'OrderErrorNotificationView',-3, 0, 'com.ibm.commerce.messaging.viewcommands.MessagingViewCommand',
'com.ibm.commerce.messaging.viewcommands.MessagingViewCommandImpl', 'docname=OrderErrorNotification.jsp&storeDir=no')

Kiana on the fly

Hashtable to iterator

Hashtable aTable = new Hashtable();
Set set = aTable.entrySet();
Iterator it = set.iterator();
while(it.hasNext()){
java.util.Map.Entry e = (java.util.Map.Entry) it.next();
String key = e.getKey();
String value = e.getValue();
}

BigDecimal error:

As API indicates, it is "immutable".
If you do following, it is wrong:

BigDecimal a = new BigDecimal("0");
BigDecimal b = new BigDecimal("1");
a.add(b); // wrong

a will not be changed.

Notice that add() method return BigDecimal object, which value is "this+val",
then this should be:
a = a.add(b);

What are inside "Message mappers" in WCS 561?

Snapshot of runtime object contents:

CommandContext
|_ adapter
|_ commandProperties - CommandProperty // Contains all properties
|_ sessionContext - CredentialsSpecifiedProgramAdapterSessionContextImpl
|_ commanProperty - CommandProperty // same as above commandProperties


commandProperties and commandProperty are pointing to same object, which contains:
-commandName
-executionProperties (Contains properties where their attributes are specified as 'FieldInfo=CONTROL'
in template file)
-requestProperties (same as getRequestProperties())


From infocenter, it says:
"Control
The name-value pair will be put into a "messageProperty" which contains control information
for the command, such as USERID or PASSWORD
"
I guess, here should be executionProperties instead of messageProperty

=================================================================
Quote from infocenter:
FieldInfo
Indicate the TypedProperty into which the name value pair should be placed. Data is the default.
If you want to put the name value pair into more than one TypedProperty, you must specify more
than one of the values listed below, separated by a comma:

Data
The name-value pair will be put into the a commandProperty object which contains arguments
for the command.
Control
The name-value pair will be put into a messageProperty which contains control information
for the command, such as USERID or PASSWORD
Command
The name-value pair is used to determine which command should be called. The generated
name-value pairs are used in the CommandMapping element of the TemplateDocument element.

XML over HTTP at WCS

XML over HTTP at WCS

WebSphere Commerce can receive inbound XML messages over HTTP.
From info center, it says:
---
The following steps illustrate the overall flow of an XML over HTTP request:

1. An external system sends an XML message to WebSphere Commerce over HTTP for example, http://host_name/webapp/wcs/stores/servlet/.
2. The request is mapped to the Program Adapter.
3. The Program Adapter passes the XML request to the appropriate message mapper.
4. The message mapper converts the XML request into a CommandProperty object and passes it back to the Program Adapter.
5. The Program Adapter prepares the command for execution and passes it to the WebController for execution.
6. The Program Adapter generates the proper XML response and returns the XML response to the external system that made the request.
---

So, how to create the request?

1. Use Java stand alone class to use socket function.
Here is my focus,if you want to send request through SSL channel, then you can initialize the socket with port 8000.
You have to establish the connection by creating the socket with port 80 first,
then, specify the port name following the HOST at header:

Example for header:

POST /webapp/wcs/tools/servlet HTTP/1.1
Host: hostname:8000
Content-Type: text/xml
Content-Length: 507

<?xml version="1.0" encoding="UTF-8"?>CONTENT...




2. Ajax solution
Seems fancy, but ...
Because Ajax does not support cross domain request,
it loses the point to use it. Why do we need a internal request (XMLHttpRequest)?

However, it seems the performance is better than socket.

The project was not built due to "Problems encountered while deleting resources."

Delete all the .class files of the project and refresh, to do a clean build .

JSTL date format

if I have a time in String format, and would like to output by using JSTL;
then this is what to do:

First, parse String to Date type:

<fmt:parseDate value="${someDateString}" pattern="yyyy-MM-dd HH:mm:SS" var="expireDate"></fmt:parseDate>

Second, output it with fmt:formatDate:
<fmt:formatDate value="${expireDate}" pattern="yyyy-MM-dd" />

The specified formatting pattern must use the pattern syntax specified by java.text.SimpleDateFormat.

Examples

The following examples show how date and time patterns are interpreted in
the U.S. locale. The given date and time are 2001-07-04 12:08:56 local time
in the U.S. Pacific Time time zone.

Date and Time Pattern	Result
"yyyy.MM.dd G 'at' HH:mm:ss z"	2001.07.04 AD at 12:08:56 PDT
"EEE, MMM d, ''yy"	Wed, Jul 4, '01
"h:mm a"	12:08 PM
"hh 'o''clock' a, zzzz"	12 o'clock PM, Pacific Daylight Time
"K:mm a, z"	0:08 PM, PDT
"yyyyy.MMMMM.dd GGG hh:mm aaa"	02001.July.04 AD 12:08 PM
"EEE, d MMM yyyy HH:mm:ss Z"	Wed, 4 Jul 2001 12:08:56 -0700
"yyMMddHHmmssZ"	010704120856-0700

Scheduler job

To create a scheduler job under site level, just associate this job with "RunAlways",
example of association:
|SCHCMD_ID | CHKCMD_ID | CHKCMD.DISPLAYNAME | PATHINFO |
|-1 | -1 | RunAlways | ReturnCreditAndCloseScan |
|-2 | 12 | CheckReleaseToFulfillment | ReleaseToFulfillment |
|-3 | 13 | CheckRAReallocate | RAReallocate |
|-4 | 14 | CheckReleaseExpiredAllocations | ReleaseExpiredAllocations |
|-5 | 15 | CheckProcessBackorders | ProcessBackorders |
|-6 | -1 | RunAlways | BalancePayment |
|-7 | -1 | RunAlways | PayCleanup |
|-8 | -1 | RunAlways | PaymentSummaryGenerate |
|-9 | -1 | RunAlways | SendEmailActivity |
|-10 | -1 | RunAlways | CheckForBouncedEmail |
|-11 | -1 | RunAlways | RetryBouncedEmail |
|-12 | -1 | RunAlways | NotifyOnOrderProcessFailure |


Those not associate with -1, are store level jobs. There are checking commands to filter the jobs.

example to add a new scheduler at site level,

"
INSERT INTO SCHCMD (SCHCMD_ID,STOREENT_ID,PATHINFO) VALUES( (select min (schcmd_id)-1 from schcmd),0,'URL_COMMAND');

INSERT INTO CHKARRANG(CHKCMD_ID,SCHCMD_ID) VALUES(-1,(SELECT SCHCMD_ID FROM SCHCMD WHERE PATHINFO='URL_COMMAND'));
"

Access control

How to create a new set of access control:
1. Create a new role. At orgadminconsole page, create a new role. The role name can be any string.
2. Create a new user, or choose existing user, to set this role. Then this user will have this role's access right.
3. Create a new access group if it doesnot exist.
Use default user access group policy xml file to create a new one. Do a query to find out the access group Id first, to ensure there is no conflict with current group.

At orgadminconsole, set this access group with the created role above.

4. Create a new access control policy xml file. The default access control xml can be a good template.
5. Load policies.
6. At orgadminconsole, subscribe the new policy for this organization or parent org.
7. At adminconsole, refresh registry to active the changes.

Adding payment methods in WC

Adding the payment method through POLICY table:
Add new policy for the new payment method in POLICY,
add new DoPaymentPolicyCmd,
ProcessOrder will invoke DoPaymentCmd,
DoPaymentCmd will invoke DoPaymentPolicyCmd,
for differnt policy, different impl will be invoked (by registering cmd class at POLICYCMD table).

you can implement your own policy cmd,customized code in DoPaymentPolicyCmd.
for each payment method, you can create a policycmd impl for it.

Need modify this to detail later.

RSS feed

First thought building RSS is the complicated work.
After a little work on it, I found that it is pretty simple and cool.

1. All about XML
RSS feed is a XML file. Anyone can simply write a XML file in RSS standard, it will become a feed.

2. Choose a RSS version.
There are many different version. Still need time to merge different standards.
Anyway, RSS 2.0 is a simple one and easy to understand.

3. Create a RSS feed
Use any tool to generate a xml file like:

<rss version="2.0">

    <channel>
        <title>Allan's Blogger</title>
        <link>http://allanxu.blogspot.com
        <description>Allan's blogger feed </description>
        <language>en-ca</language>
        <image>
            <title>Logo</title>  
            <url>http://my.com/logo.gif</url>
            <link>http://allanxu.blogspot.com</link>
        </image>

        <item>
             <title>Hello world</title>  
            <link>http://allanxu.blogspot.com</link>
            <description> Hello ! </description>
        </item>

    </channel>
</rss>

If you need value pairs inside the URL link, use escape characters. like & to replace "&". Otherwise, most feed reader wont parse it properly.

4. Publish
Deploy your page on your server. Well, yes, we need a server.
Record the link.

5. Test
Download a few RSS readers, to add your RSS feed link.
Or use IE/Firefox to open it.
Note: Firefox 2.0 has the function to render the RSS.
IE/Firefox 1.x only display the xml content.
To solve this problem, we can define a xsl to translate it. Or just open any other RSS feed, and copy their xsl link, if there is any.

Caching tips

Caching

1. Installing cachemonitor:

For cluster env, create one virtual host with more ports depends on how many app servers you have.
For each app server, at web container transport chains, define a new chain to associate the given port.
Then, each port monitor each app server.

Install dynache monitor .ear file from InstallableApp folder, accept default settings.
Map modules to application servers: Select both cluster and web server.

Re-generate plugin-cfg.xml file, and propagate it.
Restart the web server.

http(s)://<hostname>:<portN>/cachemonitor

2. cachespec.xml
For caching jsp, if there is no any Id to put in cache-in section,
use :

<class>servlet</class>
<name>com.ibm.commerce.server.RequestServlet.class</name>
<property name="save-attributes">false</property>
<property name="store-cookies">false</property>
<cache-id>
<component id="" type="pathinfo">
<required>true</required>
<value>/MY_VIEW_COMMAND</value>
</component>

Frist testing

I thought I might need a place to store my development tips, experience ...
Now, here it is.