Projects @ nioto.org

I recently open a project on java.net where files will be hosted for jdbadmin ( jdbadmin ).
I began committing code, but don’t expect a version before next month. More I use java.sql package, more I like it
The first version will be v0.1, and allow browsing a database description and data.

All data from database structure can be retrieve from the JDBC driver (SUN made a good job on java.sql package specifications), I already have a custom interface to allow SELECT maker very user-friendly. The Taglibrary are very easy to code. One of problems is to convert java.sql.Types retrieve from jdbc to something more readable , something that the database uses. The database used for development will be mysql 3.23&4.0, postgreql 7.4 and Hsql 1.7.2 .
The aplication will be a webapp, which is the easier for the user.

In my last job, I wrote a web log parser in Java , which reports are XML files instead of common HTML ( see http://jxla.nvdcms.org/ ).
Someone, from the apache jakarta project, wrote me about it telling that web logs are so ‘old fashion’, so 90’s , and a better tool will be to parse any log format ( /var/log/syslog, log file from log4j, … ). Well, I didn’t reality understand the need of such tool at the time. But after modifying JXLA to store data to database instead of XML files, allowing my boss to view the log analize ‘in real time’ ( marketing operation in progress ), I understood.
Using a simple reg-exp converter to simplify configuration of file format (like http://jxla.nvdcms.org/en/conf-file.xml [logfiles] ), and using the package java.lang.reflect to fill fields, allow the user to only code an interface representing the log line data.

Since more than two years, I was thinking of writing a phpmyadmin-like in Java, but I had not time to code. Now that I’m unemployed, the project is on my mind again.
Using the JDBC capabilities to offer a web application that allow to view/modify data to any database which has a good JDBC driver, and the JSP tag libraries to display such data.
An interesting feature will be to export database structure in SQL for another database (example: oracle to postgres).

I’m looking for a crawler/searchengine for some personal websites. Something easy to use, easy to request via Perl, PHP or Java and maybe Python. It will be able to manage multiple web sites, allow to search throw only one site and, if possible, use a mysql database.

I already know :

• Nutch
• AspSeek
• ht://Dig
• SWISH-E
• Fluffy Search
• mnoGoSearch

But :

• Nutch is ’still in development’
• ht://Dig is not multi websites
• Fluffy search indexes only local files.

For now, I will investigate more on mnoGoSearch, on of the most complete.

My Openbrick’s hard drive feels very bad, so I move all websites to this server which was waiting since 3 months, and I feel very happy to not have lost sensistive data.