NetBeans and data consistancy guarantee without large DBMS (simpletransactional database / ACID support)

[Tonny Iversen]

Hello. I haven't used NetBeans yet and I haven't programmed much Java (I've programmed more in C++), but I have a question:

Can I get a data consistancy guarantee when writing to a file from Java if I use the NetBeans IDE package without using a large DBMS?

I mean, does Java standard classes or some NetBeans packages support some kind of simple transactional database / ACID support or similar? Or maybe some good third-party packages can do this (although the best would be if Java itself can "help me out" here)? I'm going to program for several operating systems and different platforms (standard PC's, but also pretty low performing mobile platforms), so I do not want to use MySQL, PostgreSQL or similar "large" DBMS's.

Tonny

[Thomas Kellerer]

Tonny Iversen wrote on 13.11.2007 23:29:

Hello. I haven't used NetBeans yet and I haven't programmed much Java (I've
programmed more in C++), but I have a question:

Can I get a data consistancy guarantee when writing to a file from Java if I
use the NetBeans IDE package without using a large DBMS?

I mean, does Java standard classes or some NetBeans packages support some
kind of simple transactional database / ACID support or similar? Or maybe
some good third-party packages can do this (although the best would be if
Java itself can "help me out" here)? I'm going to program for several
operating systems and different platforms (standard PC's, but also pretty low
performing mobile platforms), so I do not want to use MySQL, PostgreSQL or
similar "large" DBMS's.

If you want a transactional relational DBMS then look at one of the Java based
RDBMS. There are three major players in this field:

- H2 Database (www.h2database.com)
- Apache Derby (http://db.apache.org/derby/)
- HSQLDB (www.hsqldb.org)

Apache Derby comes packaged with the JDK under the name "JavaDB", so this might
qualify as a "Java standard".

All three offer an embedded mode where no additional server is required, but all
of them have some limitations when it comes to very large datasets.

Thomas