On Tue, Jun 8, 2010 at 6:57 AM, Max Vlasov <max.vla...@gmail.com> wrote:
> You didn't mention your language/development tools. It would help also. > The application is written primarily in Python 2.6, using the sqlalchemy package to talk to the database. Parts of the program are written in C and loaded as DLLs using Python's ctypes module. Some of the DLLs are build using MINGW gcc and others are built with Microsoft Visual C. The user who experienced the problem is running Windows XP. > I also suggest to implement artificial tests like constantly emulating user > actions in your program. Once it helped me to find the cause of the same > corruption error. It would be some dev-only menu item that activates a loop > constantly reading randomly and writing randomly until specially > interrupted. The GUI part of my application actually lives in a completely separate process. It communicates with the backend over a socket, so automated testing is straightforward and part of my development process. Up until now, I have not been looking for database corruption specifically however. > While the program reads and writes you check PRAGMA > integrity_check; from time to time and stop if result shows corruption. > If you can't reproduce the problem on your machine you can send specially > prepared version to the user and ask him to perform the similar actions Thanks. I will add the PRAGMA integrity_check to my test process. -- Daniel Stutzbach, Ph.D. President, Stutzbach Enterprises, LLC <http://stutzbachenterprises.com> _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
