The database structure of OpenSER should be reviewed, streamlined and
made consistent across the tables.
Today, the database appears to be a combination of OpenSER, SEMS and SERWeb.
For example, the column indicating the last time a row was modified has
three different names in the three tables below:
subscriber table: datetime_modified
grp: last_modified
usr_preferences: modified
There have recently been some database changes that appeared to move in
the "clean-up" direction. For example, the "nickname" field was removed
from the subscriber table. There are many other examples of tables and
columns that are not needed by OpenSER to operate.
I realize that the requested database changes will "break" SEMS and
SERWeb, but in my opinion, support for these external applications
should not be in the "core" database.
Perhaps the idea of a "core" database and a set of "mods to the core
database" for SEMS or SERWeb support could be used to retain compatibility.
From a "performance" standpoint, a clean and efficient database
structure is as important as good coding (imho).
Regards,
Norm
_______________________________________________
Devel mailing list
[email protected]
http://openser.org/cgi-bin/mailman/listinfo/devel
