In order to upgrade a pre ACS-3.1 ticket database you need to do some
manual work.

The basic sequence is to do:

@ticket-save 
@ticket-nuke-old
@ticket
@ticket-data
@ticket-map

Unfortunately, since there are as many local variants of the ticket 
system as there are installed versions of the ACS it is not always
this straight-forward.

The two places that manual intervention may be needed are: 

ticket-data.sql -- need to make sure that the "code_long" values inserted
                   into ticket_codes_i match those on your system (you can
                   add or change values here without creating any problems since this 
                   is now data driven).

ticket-map.sql -- since users email addresses were stored with tickets
                  it is necessary to make sure they all map correctly
                  if you want to proserve ownership of tickets.

Once you have run ticket-save.sql there is a full copy of the original
data in TICKET_(blah)_S and you can do the cycle 

@ticket
@ticket-data
@ticket-map 

If something goes wrong, you can nuke the new data-model via 

@ticket-nuke 

edit the ticket-data and ticket-map files and try again.

Note that you will get some errors about the sequence
ticket_issue_id_sequence already existing.  This is because it is 
never dropped since we go to great length to preserve the original 
Ticket ID numbers.

