SQL - Comparing two rows and two columns

Question

I'm studying SQL and can't seem to find an answer to this exercise.

Exercise: For all cases where the same reviewer rated the same movie twice and gave it a higher rating the second time, return the reviewer's name and the title of the movie.

I don't know how to compare 2 rows and then get the higher rating.

The tables' schemas are:

• Movie ( mID, title, year, director ) English: There is a movie with ID number mID, a title, a release year, and a director.
• Reviewer ( rID, name ) English: The reviewer with ID number rID has a certain name.
• Rating ( rID, mID, stars, ratingDate ) English: The reviewer rID gave the movie mIDa number of stars rating (1-5) on a certain ratingDate.*

Researching here in the forum I've got as far as to this point:

select *
from rating a
join Reviewer rv on rv.rid = a.rid
where 1 < (select COUNT(*) from rating b
            where b.rid = a.rid and b.mid = a.mid)

I'd be glad to be given also an explanation of the code. Since even the code above is making me really confused.

/* Create the schema for our tables */
create table Movie(mID int, title text, year int, director text);
create table Reviewer(rID int, name text);
create table Rating(rID int, mID int, stars int, ratingDate date);

/* Populate the tables with our data */
insert into Movie values(101, 'Gone with the Wind', 1939, 'Victor Fleming');
insert into Movie values(102, 'Star Wars', 1977, 'George Lucas');
insert into Movie values(103, 'The Sound of Music', 1965, 'Robert Wise');
insert into Movie values(104, 'E.T.', 1982, 'Steven Spielberg');
insert into Movie values(105, 'Titanic', 1997, 'James Cameron');
insert into Movie values(106, 'Snow White', 1937, null);
insert into Movie values(107, 'Avatar', 2009, 'James Cameron');
insert into Movie values(108, 'Raiders of the Lost Ark', 1981, 'Steven Spielberg');

insert into Reviewer values(201, 'Sarah Martinez');
insert into Reviewer values(202, 'Daniel Lewis');
insert into Reviewer values(203, 'Brittany Harris');
insert into Reviewer values(204, 'Mike Anderson');
insert into Reviewer values(205, 'Chris Jackson');
insert into Reviewer values(206, 'Elizabeth Thomas');
insert into Reviewer values(207, 'James Cameron');
insert into Reviewer values(208, 'Ashley White');

insert into Rating values(201, 101, 2, '2011-01-22');
insert into Rating values(201, 101, 4, '2011-01-27');
insert into Rating values(202, 106, 4, null);
insert into Rating values(203, 103, 2, '2011-01-20');
insert into Rating values(203, 108, 4, '2011-01-12');
insert into Rating values(203, 108, 2, '2011-01-30');
insert into Rating values(204, 101, 3, '2011-01-09');
insert into Rating values(205, 103, 3, '2011-01-27');
insert into Rating values(205, 104, 2, '2011-01-22');
insert into Rating values(205, 108, 4, null);
insert into Rating values(206, 107, 3, '2011-01-15');
insert into Rating values(206, 106, 5, '2011-01-19');
insert into Rating values(207, 107, 5, '2011-01-20');
insert into Rating values(208, 104, 3, '2011-01-02');

Answer 1

something like that should work (they are other ways, too)

SELECT rev.name, m.title
FROM Reviewer rev
INNER JOIN Rating r1 on r1.rID = rev.rID
INNER JOIN Rating r2 on r2.rID = rev.rID and r2.mID = r1.mID
INNER JOIN Movie m on m.mID = r1.mID
WHERE r2.ratingDate > r1.ratingDate and r2.stars > r1.stars 

or you can do all in join (instead of WHERE clause) in this case

SELECT rev.name, m.title
FROM Reviewer rev
INNER JOIN Rating r1 on r1.rID = rev.rID
INNER JOIN Rating r2 
  on r2.rID = rev.rID 
  and r2.mID = r1.mID
  and r2.ratingDate > r1.ratingDate
  and r2.stars > r1.stars
INNER JOIN Movie m on m.mID = r1.mID

SqlFiddle (with your sample datas)

Explanation : I suppose you know the JOIN syntax, so

The trick is to join Rating two times. Then the WHERE part checks if there's exist a line where one of the rating (from same reviewer on same movie) has a bigger ratingDate and more stars. Which checks : "gave it a higher rating the second time".

Then we just group by reviewerName and movie title (this part is to avoid duplicates if we have 3 reviews, the second having more stars than the first, and the third more than the second) : with your sample datas, the GROUP BY is not needed, but...

Answer 2

Start with getting all the reviewers who reviewed exactly twice:

select rid
from rating r
group by rid
having count(*) = 2

Now the question is: are they the same or is the second larger? To do this, join back in the ratings, but also include the two dates:

from (select rid, min(ratingdate) as minratingdate, max(ratingdate) as maxratingdate
      from rating r
      group by rid
      having count(*) = 2
     ) twotimes join
     rating r1
     on r1.rid = twotimes.rid and r1.ratingdate = twotimes.minratingdate join
     rating r2
     on r2.rid = twotimes.rid and r2.ratingdate = twotimes.maxratingdate

This brings in the information about the two reviews. You can finish the query from here.

Answer 3

You can use GROUP BY and HAVING :

SELECT m.mId, m.Title, r.Name, ra.stars
FROM Movie m 
   JOIN (SELECT mId, rId, MAX(stars) stars
        FROM Rating 
        GROUP BY mId, rId
        HAVING COUNT(*) > 1) ra ON m.mId = ra.mId
   JOIN Reviewer r ON ra.rId = r.rId
GROUP BY m.mId, m.Title, r.Name, ra.stars

This will return you any movie that has multiple reviews from the same reviewer with the highest number of stars.

Here is the SQL Fiddle for testing.

Good luck.