I'm trying to come up with an efficient way to do the following task, but I
haven't found the correct join syntax to do it.
I have table "users" for keeping a list of users I need to check logins for.
It contains the following fields:
id,Full-Name,User-Name
I have table called "logs" that contains the actual radius log data. The
three fields I am concerned with are:
User-Name,Date,Time
I have gotten thus far almost what I want with the following query.
SELECT users."User-Name",max(logs."Date") as login_date,max(logs."Time") as
login_time FROM logs where logs."User-Name"=users."User-Name" GROUP BY
users."User-Name";
This gives me the User-Name, date, and time of people WHO HAVE LOGGED IN.
I'm looking to have a query that returns all of the users in the first
table, including those who have not logged in. I have an external perl
script that generates a nice html report for the manager. I have a script
that parses the raw .csv files, but I'm trying to move it entirely to
postgres, including if possible stored-procedures in plperl. I already have
a perl script that imports the raw log files into the logs table.
Am I on the right track?
Thanks,
Stephen Sill II