Duplicate Results in Oracle SQL Plus - sql

Question

When I run the script below, I keep getting duplicate results, even when using distinct.
select distinct
a.SDT, a.fNo, b.IDType, b.pNo, b.pfName, b.plName, b.PDoB, b.Street, b.City, c.Phone
from Scheduled_Flight a, Passenger b, pass_Phone c
where fNo = '0000021'
and
a.SDT = '08-sep-2017 17:30';
I am new to SQL and any help would be much appreciated into solving this issue.

Answer 1

"I keep getting duplicate results, even when using distinct"
You are not getting duplicates in your result set. Rather you have a Cartesian product which is a combination of ONE flight, THREE passengers and THREE phone numbers. Each record in the set is unique so distinct doesn't have any affect.
The problem is you have no join conditions in your from clause. There should be a column on passenger which is the foreign key on flight, and a column on pass_phone which is the foreign key on passenger.
It is easy to fix: you just need to join the tables. Assuming your data model is consistent, your query should look like this (and you don't need DISTINCT):
select a.SDT, a.fNo, b.IDType, b.pNo, b.pfName, b.plName, b.PDoB, b.Street, b.City,c.Phone
from Scheduled_Flight a
join Passenger b on b.fNo = a.fNo
join pass_Phone c on c.pNo = b.bNo
where a.fNo = '0000021'
and a.SDT = '08-sep-2017 17:30';
However, I notice that in your version of the query you didn't prefix fNo. That makes me think you don't have a column of that name on passenger (otherwise the query would have failed on ORA-00918: column ambiguously defined). So, either the foreign key columns are named differently or you haven't got them.
"Is it possible to specify only the date without the time?"
Yep. Use an ANSI date literal e.g. date '2017-09-08'
"Is it possible to specify only the date without the time to still produce results from the database?"
That depends on the how the data is stored. Oracle dates are stored with a time element. If no time is specified (or the time element is truncated) then the time element defaults to midnight. This often catches beginners out, for instance because the pseudo-column sysdate returns the current date and time, not just the current date.
So, if you know the dates are stored in your table without a time element you can do this:
where a.sdt = date '2017-09-08'
But if you don't know that, you can truncate ...
where trunc(a.sdt) = date '2017-09-08'
or test for a range
where a.sdt >= date '2017-09-08'
and a.sdt < date '2017-09-09'
"How come the following code is still producing duplicate results?
select distinct r.sNo, r.tCode, s.fNo, s.SDT
from Airplane r, Scheduled_Flight s
where SDT >= SYSDATE -1;
The airplane attribute cannot have the s.SDT attribute."
Without seeing the output I can't be sure but I would bet that this query does not produce duplicate records either. What you have is a product combining all your AIRPLANE records with all your FLIGHT records matching the sdt filter.
This is another data modelling problem. Of course aeroplanes don't have a flight time: one aeroplane makes many flights. But it makes perfect sense for a flight to be assigned to a plane. In fact that's crucial to ensuring that you don't have more flights than you have planes to fly them, and that one plane isn't planned to take off from London for Madrid at a time when it's planned to be half-way to Hong Kong.
You really should use the ANSI 92 syntax, as I showed in my answer to your previous posted code. The explicit joins not only make it easier to understand the query but they prevent mistakes like this. The fact that you apparently don't have any candidate columns to make the join immediately highlights the flaw in the data model.
select distinct r.sNo, r.tCode, s.fNo, s.SDT
from Airplane r
INNER JOIN Scheduled_Flight s ON ????
where SDT >= SYSDATE -1;

Answer 2

i don't see any rows which are duplicated, if you compare every column of each row, each row is uniquely identified, since you are doing cartesian product you are getting multiple records. but each rows are unique to each other.