The membership data is stored in multiple tables within the database "ffii" on vic (postgresql):

== asoc ==

 * uid, character varying(8): aktiv id
 * org, character varying(8): organisation, default 'ffii'
 * asolvl, numeric(1,0): membership status, -1 (unsubscribed) to 3 (active member), not necessarily correct
 * asorem , text, ???
 * asolack, character varying(16), acknowledgement code of membership

== asolack ==

??, no permissions

== mail ==

 * uid,  character varying(8) 
 * mailref, ?? 
 * mailuser, part of email before @
 * mailhost, part of email after @

== padr ==
 * uid, character varying(8) 
 * adrref, character varying(8), ?? 
 * land, character varying(2), 2 letter code for country (belgium = be, netherlands = nl)
 * plz, text, zip code 
 * urb, text, city 
 * str, text, street name and number 

== pnom ==
 * uid    | character varying(8) 
 * pntyp  | character varying(2) | ??
 * pnom, full name

== teln ==
 * uid    | character varying(8) | 
 * telref | character varying(8) | ??
 * tel    | text                 | telephone number
 * telrem | text | ??

== pass ==
 * uid    | character varying(8)  |
 * pass   | character varying(16) | hash (what kind of?) of password

== Some useful queries ==

Get member info from country BE:

psql ffii

ffii=#  \o /home/dietvu/ffii.csv; 

ffii=# select distinct * from mail left outer join teln on (teln.uid = mail.uid) inner join pnom on (pnom.uid = mail.uid) inner join padr on (padr.uid = mail.uid) left outer join asoc on ( asoc.uid = mail.uid) where "land" = 'be';

email only:
ffii=# select distinct mailuser, mailhost from mail left outer join teln on (teln.uid = mail.uid) inner join pnom on (pnom.uid = mail.uid) inner join padr on (padr.uid = mail.uid) left outer join asoc on ( asoc.uid = mail.uid) where "land" = 'be';



== Observations during manual correction ==

A lot of double records, only difference is organisation (nothing/ffii/ael) or norm/subs status for mailing lists.