Code maintainence for SQR

Tuesday, September 30, 2008, 05:07 AM

In order to fix code many a times one needs to search for terms, function names, defect/case ID's etc in SQR/SQC/INC (henceforth referred to as SQR files) files.

TextPad allows one to search thorough folders containing files having certain terms (e.g a defect ID or a case number) which is suspected to be related to a bug. However this process is often very slow as TextPad needs to open each and every file one by one and parse through it.

If there was a way by which we could index all terms in SQR files and store them upfront the amount of time required to search problem areas in SQR's and fix them could be drastically reduced.

Fortunately there are two free tools one could use to index these files. The first is an open source tool called Lucene and the second is Microsoft's Indexing Service that comes by default with the Windows XP operating system.

Lucene though very powerful needs knowledge of Java to build an application which would fit our needs.

Microsoft Indexing Server on the other hand is simpler to configure. The service can be accessed through Control Panel --> Administrative Tools --> Computer Maintainence. All one needs to do is create a new Catalog and add the directories which have the SQR files. The indexing service keeps a track of any changes made to the files and updates the indexes automatically. However, one needs an IFilter implementation to read SQR files even though SQR files are as good as text files. There is one workaround though. Change the file extension from .SQR to .SQR.TXT.

If you want to search on the index, all one needs to do is query the catalog and presto you can search through all the code that resides in your SQR files. Only those files which have the term would show up in the search results.

To take it one step further one could write a program to use this index and highlight terms in the file.

more

Java and Excel

Sunday, September 28, 2008, 03:02 PM

Interfacing Java with Microsoft Excel has always been a pain for software engineers.

The Apache foundation came out wit POI a Java API To Access Microsoft Format Files but it had its own limitations and the learning curve was a bit steep.

While looking for other ready made libraries I came across "jexcel". Its a simple library which can do quite a bit for its size. Its very easy to use and can manipulate existing excel files having formatted information as well as create new files with custom information and formats based on pure coding or a template.

If your existing libraries or custom developed tool does not do its job or does not do its job well I highly recommend giving this a shot.

Also check out http://jexcelapi.sourceforge.net/ and http://www.andykhan.com/jexcelapi/tutorial.html for a quick tutorial on how to use it.

more

XAMPP a boon for PHP/MySql developers

Tuesday, September 23, 2008, 04:22 AM

An web development requires a server configuration.

If the configuration also involves a database server it adds up to the complexity of getting one cofigured to work with your web server.

XAMPP to the rescue.

XAMPP supports MySql, PHP as well as Perl. Its similar to the Wamp Server but I personally had some problems with Wamp. XAMPP on the other hand was pretty simple and worked without any problems. I use it very often when I need to continue with my PHP development pet projects and carry it along in my thumb drive wherever I go.

more

Making PHP work with Lighty (LightTPD)

Sunday, September 21, 2008, 07:21 AM

Yes I know there are numerous articles and forums out there which talk about this.

Why am I writing about this again?

That's because there is still some doubt out there on how to get PHP to work with Lighty.

In this blog post I would be providing the configuration files as well for Lighty and PHP whcih is missing in all the forms and general discussions out there. This would help folks get started faster.

You can download the files here:- PHP.ini, lighttpd-inc.conf

The PHP version I am using is 5.2.6 and the Lighty version is 1.4.19-1 which I downloaded from WLMP project. The good thing is they have already complied LightTPD to work with windows.

The link to download LightTPD is http://en.wlmp-project.net/downloads.php.

The idea is to start PHP in fastCGI mode. That way you are opening a port that PHP listens to and whenever Lighty gets a PHP request it just forwards the same to be executed as a PHP script.

The important part of the Lighty configuration file is mod_fastcgi which exists in server.modules. Remove the hash(#) before mod_fastcgi in the config file.

The next important part is:-

fastcgi.server             = ( ".php" =>
                               ( "localhost" =>
                                 (
                                   "host" => "127.0.0.1",
				   "port" => 1879,
				   "docroot" => "c:/LightTPD/htdocs/"
                                 )
                               )
                            )

What's happening here is that I am asking Lighty to forward all php requests on to port 1879 on the host ip 127.0.0.1 which is a localhost ip. Also what's important is you mention the docroot location of where the php files would be located. (This part is important.)

Lets talk about the php.ini file for now.

Here are the important parts of the file:-

Force Redirect should be set to enabled for security reasons.

cgi.force_redirect = 1

Fix Path info should also be turned on as is spoken of in the Lighty documentation

cgi.fix_pathinfo=1

Once you have setup your configuration files what you need to do is make sure you do not have any other webservers running on port 81. The reason I use port 81 is because of IIS. IIS is very greedy when it comes to port 80 and does not allow one to change the listening port.

An easy way to find out if any severs are listening on port 81 is to run the command netstat -a and look for any entries for port 81. I am assuming that you have not yet started Lighty.

Once we have confirmed that nothing is listening on port 81 we can fire up Lighty and type in http://localhost:81. If you get a page telling you about Lighty and stuff you have completed the first hurdle of getting a basic http server running with the new configuration file.

Once again assuming you have completed the previous step type php-cgi.exe -b 127.0.0.1:1879 through the command line. Of course before running this command I am assuming that you have changed he path to the location where the php home directory resides. Also make sure that you have php-cgi.exe in the home directory. If you do not have it don't fret. Just download the zip version of the PHP installation and extract php-cgi to the folder.

Once you are done create a file having the function phpinfo() in a file ending with .php in the htdocs folder.

If everything has gone well you should see all the details of your php server on typing the name of the php file after the localhost url.

Hope this helps.

more

Object Types in PeopleSoft

Tuesday, April 22, 2008, 12:34 AM

If you are wondering what Object Types are they are nothing but your usual objects that you create through the application designer or the PIA.

You might be wondering okay so why am I writing a blog on this.

Well here's why. Have you ever wondered how PeopleSoft stores the records, fields, pages, componets etc in the database. In an earlier blog I had mentioned about where PeopleSoft stores all your projects and objects that belong to your project.

Object Types in the peoplesoft are stored as numbers where each number stands for a certain object.

After snooping around in the database as well as google I have finally managed to collate a list of almost all the object types that are generally used.

The list is available here.

more

Application Engine and Command Line

Tuesday, April 22, 2008, 12:05 AM

More often than not I had situations where I needed to literally step through the application engine looking at the sections and steps that got exected under differnet conditions just to confirm if the logic applied to build the Application Engine (AE) was right or wrong.

So here are the steps and parameters I use to do just.

First, you need to set PSAE.exe which actually initiates your AE in your environment variables. I am assuming you are using Windows XP.

How would you set the path to point to the folder which houses PSAE?

1. Right Click on My Computer on your desktop and then click on Properties.

2. Click on the Advanced tab.

3. Click on Environment Variables.

4. Under System Variables look for Path.

5. Double Click that entry and append (I said append not overwrite) the folder path which has PSAE. On my system I had the executable under " C:\PT8.46\bin\client\winx86 ".

Now that you have everything setup to debug your application engine open the command prompt.

Type psae and press enter. If you get a message like :
Invalid command line argument list. Usage:
psae -CT <dbtype>
-CS <server>
-CD <database name>
-CO <oprid>
-CP <oprpswd>
-R <run control id>
-AI <program id>
-I <process instance>
-DEBUG <Y|N>
-DR <Y|N>
-TRACE <value>
-DBFLAGS <value>
-TOOLSTRACESQL <value>
-TOOLSTRACEPC <value>
-OT <output type>
-OF <output format>
-FP <file path>
or
psae <parmfile>
-(RE)CONVERT <product>.<appl id> (e.g. AE.TESTAPPL)

Congaratulations you have just finished the first hurdle.

If you notice the error you get on the screen is quite intuitive.

Of all the options mentioned -CT, -CD, -CO, -CP, -R, and -AI are mandatory to get your application engine working.

Note: The order of these parameters is immaterial.

If you are trying to restart an application which had abended for some reason by supplying the correct UserID, pwd, run control id and process instance you can rerun the application engine.

If you want to step through each and every step that the AE would execute in realtime just plug in the -Debug option.

Once you are in the Debug mode and have initiated your application engine to know what choices you have just type '?' for help.

Hope this helps.

more

Application Engine and Run Control ID's

Monday, April 21, 2008, 09:44 AM

Many a times I have heard people new to PeopleSoft say that one cannot run an application engine using the same Run control ID. Nothing can be further from the truth.

The reason why an application engine often does not run with the same run control id is that a previous instance of the same application engine might have gone to error with the same run control id.

If at all this happens and you still want to run the application engine it can be done through the command line if all else fails.

more

PeopleSoft Project Secrets

Monday, April 21, 2008, 08:59 AM

I had always wondered how does peoplesoft store projects created through the application designer.

Here I was with an interesting problem. So how did I go about finding out which are the tables where PeopleSoft stores all its projects?

Well, in my current assignment almost all computers have a local instance of PeopleSoft Vanilla installed. This vanilla instance runs on an SQL Server Database. SQL Server...yuppie good news for me. Why you might ask?

SQL Server has its own profiler tool which can be used to watch all database level accesses that happen in realtime.

So here's what I did to find out which tables are populated when one saves a project or accesses a project using the application designer.

I started the profiler. Opened the application designer. Opened an existing project and bang I got my answers straight away.

The tables which are of interest to us are:-

PSPROJECTDEFN

PSPROJECTITEM

PSPROJECTMSG

So what do each of these tables store.

PSPROJECTDEFN - seems to store the project name and description of the project.

PSPROJECTITEM - seems to store the individual objects in the project.

PSPROJECTMSG - seems to store a short description of the very same objects which are present in the previous table.

In my next post I will highlight some of the object types that I managed to reverse Engineer from various sources including the net and by joining the tables PSPROJECTITEM and PSPROJECTMSG.

Till then hold on to your horses.

more

Hello PeopleSoft

Monday, April 21, 2008, 08:41 AM

In this section I will be sharing some gyan that I have about this beautiful ERP package called PeopleSoft.

For all those who are new to ERP and do not know what it is, ERP stands for Enterprise Resource Planning. Its a packaged application that lets you manage your business/organization in a much more organized manner. Unlike a custom application where one needs to start from scratch a packaged applicaton can be used right out of the box (well almost) with some configuration changes.

A packaged application allows some level of customization and PeopleSoft is no exception. I have even heard my clients and many other folks in the ERP business saying quite openly that PeopleSoft is the easiest and fastest ERP which can be customized and that's what makes it a hot ERP.

Enough talk. I will be splitting this topic into multiple categories so watch out for updates. If any of you would like to request for some topics do let me know so that I can add that as well...that is provided I know enough about the topis to write about it.

more

SQR Pad

Thursday, April 17, 2008, 03:13 AM

Well I finally got my hands on a beautiful application called CodePress. It has its very own rich text editor which you can configure quite easily.

Its quite a powerful editor. To get a sneak preview of my editor click on SQR Pad.

Do keep in mind that this is work under progress and I might add more features to it as and when time permits. In the mean time try it out.

Oh I almost forgot..if incase you want to copy the code written by you you would have to click the button at the bottom of the page in order to reformat the code so that it can be copied.

Enjoy.

more

IFrame to build rich text client

Friday, December 14, 2007, 12:02 AM

Today I found a cool site which helps you build a simple rich text client based WYSIWYG HTML editor.

I had been researching for some time on IFRAME's and did find some good material but this particular site IFrame based editor was just excellent.

Go ahead and have a look at this site.

Look out for my next project which is going to be an SQR editor using IFrames.

more

Apache HTTP Server (2.2.4) + Tomcat 5.5.23 + SSL

Friday, August 10, 2007, 02:27 AM

Tomcat with Apache HTTP server is a very common configuration for getting a J2EE application up and running. Though Tomcat is not a complete J2EE server(does not support EJB's) it is a very stable J2EE web server. It can also act as a standalone web server without the need for a proxy server like Apache's HTTP server.

I personally however feel that is a better J2EE server out there in the open source market namely Resin but Tomcat has one major advantage which Resin does not have which is proper support and docuemntation. This is what makes it more popular than Resin which beats Tomcat on performance grounds.

Enough gyan on servers. Lets get our hands dirty in setting up Tomcat with Apache Http server having SSL for encription and security.

We will tackle the setup in two stages:-

1. Get the HTTP server working with Tomcat so that it can re-route the jsp and servlet requests to Tomcat.

2. Get SSL working with Apache's HTTP server.

Before we start doing anything make sure you have all the necessary software.

My software setup had the was as follows:-

Windows 2003 Server

Apache HTTP server 2.2.4 with SSL

Apache Tomcat 5.5.23

mod_jk

The reason I chose Apache with SSL so that I need not download and install the ssl module seperately.

mod_jk would be responsible for forwarding the requests from the HTTP server to the J2EE server(Tomcat).

Download three other files which should serve as referneces for your setup :-

workers.properties

httpd.conf

httpd-ssl.conf

The files workers.properties and httpd.conf have the necessary info redirect servlet requests to the J2EE server through HTTP server.

The line LoadModule jk_module modules/mod_jk.so enables the jk module which is a tomcat connector to the apache http server. You would need to download the jk module file which is specific to your Apache HTTP server version from any of the mirrors from the link mentioned above for mod_jk and rename the file to mod_jk.so.

The sample workers.properties file contains just one worker which would handle the servlet requests. The lines

JkWorkersFile "C:/Program Files/Apache Software Foundation/Apache2.2/conf/workers.properties" and

JkMount /ARNT|/* worker1

tell the http server what needs to go where. /ARNT|/* tells the URL patterns which need to be redirected to Tomcat. Have a look at the file httpd.conf to know what goes where.

Try accessing any JSP now through the http server that is on port 80 to see if the request goes through.

Now its time to setup SSL.

Before proceeding make sure you have the certificate and the private key for implementing SSL.

Since we have installed a version of apache with SSL we need not install the ssl module seperately however we need to uncomment the line LoadModule ssl_module modules/mod_ssl.so in the http.conf file to tell apache that it needs to lead that module.

Open the file httpd-ssl.conf which you downloaded for reference and go to the line

SSLCertificateFile "C:/Program Files/Apache Software Foundation/Apache2.2/conf/10.186.133.143.crt".

This line tells Apache where the certificate is stored.

SSLCertificateKeyFile "C:/Program Files/Apache Software Foundation/Apache2.2/conf/10.186.133.143.key"

tells apache where the private key is stored.

And you are done. Restart Apache and tomcat after any change to the configuration file so that your changes can take effect.

Hope it works. If not post your comments and I will try to help you out.

more

Database accessing local files using CLR integration

Thursday, August 9, 2007, 03:34 AM

Given below is a way to access the file contents of a database server using the latest feature of SQL Server 2005 called CLR integration.

The theory and source code can be accessed from http://blogs.msdn.com/sqlclr/archive/20 ... 15034.aspx.

The compilation is quite staraight forward and all you need to do is create a dll file from the C# code by creating a Class Library.

You may face some problems giving the assembly external_access.

The error maybe something like

Msg 10327, Level 14, State 1, Line 1
CREATE ASSEMBLY for assembly 'TVF' failed because assembly 'TVF' is not authorized for PERMISSION_SET = EXTERNAL_ACCESS. The assembly is authorized when either of the following is true: the database owner (DBO) has EXTERNAL ACCESS ASSEMBLY permission and the database has the TRUSTWORTHY database property on; or the assembly is signed with a certificate or an asymmetric key that has a corresponding login with EXTERNAL ACCESS ASSEMBLY permission. If you have restored or attached this database, make sure the database owner is mapped to the correct login on this server. If not, use sp_changedbowner to fix the problem.

In my case I was executing the statements as an administrator hence the database owner had the EXTERNAL ACCESS ASSEMBLY permission so all I needed to do was alter the database to make it trustworthy by executing the command

alter database {database name} set trustworthy on

You may also be required to set the database compatibility level to the highest value using :-

sp_dbcmptlevel {database name} , 90

Note :- Make sure that you set the Enable CLR Integration option in the Surface Area Configuration to true.
Note :- Make sure that you set the Enable CLR Integration option in the Surface Area Configuration to true.

more

Split a field in an SQL table

Wednesday, July 4, 2007, 11:43 PM

Many a times there are situations when you want to split a string in a field based on a seperator like a comma(",") or maybe a semicolon(";"). Here's an SQL function which would allow you to do just that.

CREATE FUNCTION [dbo].[Split_with] (@sep varchar(max), @s varchar(max))

RETURNS @mytab table (rowid int,keys varchar(max))

AS

[more]BEGIN

WITH Pieces(pn, start, stop) AS (

SELECT 1, 1, CAST(CHARINDEX(@sep, @s) AS INT)

UNION ALL

SELECT pn + 1, stop + 1, CAST(CHARINDEX(@sep, @s, stop + 1) AS INT)

FROM Pieces

WHERE stop > 0

)

INSERT @mytab

SELECT pn,

SUBSTRING(@s, start, CASE WHEN stop > 0 THEN stop-start ELSE 512 END) AS s

FROM Pieces

OPTION (MAXRECURSION 10000)

RETURN

END

To make the function work you need to call it like:-

SELECT * FROM Split_With(',',(SELECT {field} FROM {table} Term where {filter a single row} ))

Ofcourse the function will work on a single field at a time. All you need to do a recurse through the entire table using a cursor or a while loop.

Hope this helps whoever needs it.

more

Create An Executable JAR

Sunday, May 27, 2007, 09:26 AM

All JAR files contain something called a manifest file which holds information Java wants to know. One piece of information a manifest file may contain is the name of a class that will be run if the JAR file is executed.

The first thing you must do is create a text file that lists the "main" class - the class that has the main method you want executed when the JAR is executed. Let's say that Three from the above example has the main method I want executed. I create a text file called "mainClass.txt" with the following text:

Main-Class: Three

IMPORTANT: the text file only needs the one line of text for this purpose. However, the file must end with a blank line or this will not work, ie the file has two lines in it - the second one is empty. Note too the class is called "Three" and not "Three.java" (the file containing the source code) or "Three.class" (the file containing the byte codes). If your class file is in a package hierarchy, you must use the fully qualified name of the class (eg "myPackage.MyClass").

I then run the jar utility with this command line:

jar cmf mainClass.txt example.jar *.class

With this line, I told jar to create a JAR file (option c) with modifications to the manifest file (option m) as specified within mainClass.txt, naming the JAR file (option f) as example.jar and including everything that matches the pattern *Class

Source:- http://neptune.netcomp.monash.edu.au/Ja ... to/jar.htm

more

NSE Stock Updater

Sunday, May 27, 2007, 08:30 AM

Download daily (bhavcopy) NSE quotes

Here's a small program which downloads daily NSE quotes or in other words the bhavcopy directly from the NSE site (www.nse-india.com).

Download NseStockUpdater here.

Just type:-

java -jar nsestockupdater.jar

If you want the bhavcopy for a particular date then just type:-

java -jar nsestockupdater.jar 22-05-2007

Note that the date should be in dd-mm-yyyy format.

The output file is created in the same folder where the .jar file exists.

The source code and the class files are in the jar file.

Most of the code has been reused from other sources as I am a very lazy programmer. :)

Let me know if you face any problems.

more

List Zip file contents

Wednesday, April 18, 2007, 09:54 AM

try {
// Open the ZIP file
ZipFile zf = new ZipFile("filename.zip");

// Enumerate each entry
for (Enumeration entries = zf.entries(); entries.hasMoreElements();) {
// Get the entry name
String zipEntryName = ((ZipEntry)entries.nextElement()).getName();
}
} catch (IOException e) {
}

more

Java and Zip files

Wednesday, April 18, 2007, 09:52 AM

Unzipping Files with java.util.zip.ZipFile

The Java core packages include jave.util.zip, which is useful for creating and reading compressed files or archives in the ZIP and GZIP formats. This saves you the time of creating custom file archive formats for your applications.

by Daniel F. Savarese

The Java core packages include jave.util.zip, which is useful for creating and reading compressed files or archives in the ZIP and GZIP formats. ZIP files form the basis for the JAR file format, so it was natural to include classes for manipulating ZIP files as part of the standard Java APIs. This functionality is enormously useful, because it saves you the time of creating custom file archive formats for your applications. You can use the ZipFile [more]class to read entries in a ZIP file. The entries are represented by instances of the ZipEntry class, which contains information such as the size, name, compression method, and timestamp of the entry. ZipFile.entries() will return the entries as an Enumeration which you can step through to access all of the entries. Once you have found an entry of interest, ZipFile.getInputStream(ZipEntry) will produce an InputStream, from which the archived entry can be read.

The following program demonstrates how to use the ZipFile class to create a simple unzip utility. It simply cycles through all of the entries in the file and copies each entry to a new file by reading its input stream. If an entry is a directory, it is created. The example is not robust and is only intended to demonstrate the ZipFile class. Obvious additions would include checking if a file already exists before creating it, prompting the user if it should be overwritten, as well as other error checking. View the example>:

import java.io.*;
                                import java.util.*;
                                import java.util.zip.*;

                                public class Unzip {
                            public static final void copyInputStream(InputStream in, OutputStream out)
                            throws IOException
                            {
byte[] buffer = new byte[1024];
                            int len;

                            while((len = in.read(buffer)) >= 0)
                            out.write(buffer, 0, len);

                            in.close();
                            out.close();
                            }

                            public static final void main(String[] args) {
                            Enumeration entries;
                            ZipFile zipFile;

                            if(args.length != 1) {
                            System.err.println("Usage: Unzip zipfile");
                            return;
                            }

                            try {
                            zipFile = new ZipFile(args[0]);

                            entries = zipFile.entries();

                            while(entries.hasMoreElements()) {
                            ZipEntry entry = (ZipEntry)entries.nextElement();

                            if(entry.isDirectory()) {
                            // Assume directories are stored parents first then children.
          System.err.println("Extracting directory: " + entry.getName());
          // This is not robust, just for demonstration purposes.
          (new File(entry.getName())).mkdir();
          continue;
        }

        System.err.println("Extracting file: " + entry.getName());
        copyInputStream(zipFile.getInputStream(entry),
           new BufferedOutputStream(new FileOutputStream(entry.getName())));
      }

      zipFile.close();
    } catch (IOException ioe) {
      System.err.println("Unhandled exception:");
      ioe.printStackTrace();
      return;
    }
  }

}

more