Howdy 
I've recently started writing my Oracle SQL scripts in ANSI - I find it much cleaner, much easier to read, much easier to make sure joins aren't missed, allows you to use multiple outer joins against the same table and isn't proprietary. My colleagues, however, all use regular Oracle joins - and are adamant that the old-school method is faster and better.
Which do you guys use?
ANSI:
Oracle:
I've recently started writing my Oracle SQL scripts in ANSI - I find it much cleaner, much easier to read, much easier to make sure joins aren't missed, allows you to use multiple outer joins against the same table and isn't proprietary. My colleagues, however, all use regular Oracle joins - and are adamant that the old-school method is faster and better.
Which do you guys use?
ANSI:
Code:
cursor cs_jobs is
select w.id wid,
w.ext_sys_ref ext,
o.name opn,
g.name gpn,
nvl(h.date_time,p.date_time) app,
u.user_id||' - '||u.name usr
from assign.worklist w
left join assign.possibledates p on p.worklist_id = w.id
left join assign.users u on u.id = w.user_id
left join assign.scheduledetail d on d.worklist_id = w.id
left join assign.scheduleheader h on h.id = d.schedule_id
left join assign.operatives o on o.id = h.operative_id
left join assign.operativegroups g on g.id = o.group_id
where trunc(w.date_booked) = trunc(sysdate)
and w.location_postcode like '&&outcode'||'%';Oracle:
Code:
cursor cs_jobs is
select w.id wid,
w.ext_sys_ref ext,
o.name opn,
g.name gpn,
nvl(h.date_time,p.date_time) app,
u.user_id||' - '||u.name usr
from assign.operativegroups g,
assign.operatives o,
assign.scheduleheader h,
assign.scheduledetail d,
assign.users u,
assign.possibledates p,
assign.worklist w
where trunc(w.date_booked) = trunc(sysdate)
and w.location_postcode like '&&outcode'||'%'
and p.worklist_id(+) = w.id
and u.id(+) = w.user_id
and d.worklist_id(+) = w.id
and h.id(+) = d.schedule_id
and o.id(+) = h.operative_id
and g.id(+) = o.group_id;