Basic Sql: Selecting The Same Column Multiple Times in One Query, When Each Occurrence Is Dependent on Different Where Clause

basic sql : selecting the same column multiple times in one query, when each occurrence is dependent on different where clause

You should use 3 queries. It will be a lot faster with proper indexing when self joins. Additionally it will be more readable.

If you would like one query call, it might be this :)

SELECT
(SELECT x FROM table WHERE y=1) AS x1,
(SELECT x FROM table WHERE y=2) AS x2,
(SELECT x FROM table WHERE y=3) AS x3

How to select a single row multiple times in PostgreSql

You can cross join against generate_series(1,4), which will return a table containing the numbers 1 to 4:

SELECT mytable.*
FROM mytable 
CROSS JOIN generate_series(1,4) as x
WHERE id=1

For each row in your original result set, there will be one copy with 1 next to it, one with 2, and so on.

How to get multiple counts with one SQL query?

You can use a CASE statement with an aggregate function. This is basically the same thing as a PIVOT function in some RDBMS:

SELECT distributor_id,
    count(*) AS total,
    sum(case when level = 'exec' then 1 else 0 end) AS ExecCount,
    sum(case when level = 'personal' then 1 else 0 end) AS PersonalCount
FROM yourtable
GROUP BY distributor_id

Selecting most recent and specific version in each group of records, for multiple groups

To get only latest revisions:

SELECT * from t t1
WHERE t1.rev = 
  (SELECT max(rev) FROM t t2 WHERE t2.id = t1.id)

To get a specific revision, in this case 1 (and if an item doesn't have the revision yet the next smallest revision):

SELECT * from foo t1
WHERE t1.rev = 
  (SELECT max(rev) 
   FROM foo t2 
   WHERE t2.id = t1.id
   AND t2.rev <= 1)

It might not be the most efficient way to do this, but right now I cannot figure a better way to do this.

Select rows with same id but different value in another column

This ought to do it:

SELECT *
FROM YourTable
WHERE ARIDNR IN (
    SELECT ARIDNR
    FROM YourTable
    GROUP BY ARIDNR
    HAVING COUNT(*) > 1
)

The idea is to use the inner query to identify the records which have a ARIDNR value that occurs 1+ times in the data, then get all columns from the same table based on that set of values.

Retrieving multiple data from db and grouping them

Well, queries produce a rectangular result set, not a hierarchical one, but something like this would work:

SELECT Color, Make
FROM Cars
ORDER By Color, Make

Then when you display the results you can show the Color only when it changes.

SQL query return data from multiple tables

Part 1 - Joins and Unions

This answer covers:

Part 1
- Joining two or more tables using an inner join (See the wikipedia entry for additional info)
- How to use a union query
- Left and Right Outer Joins (this stackOverflow answer is excellent to describe types of joins)
- Intersect queries (and how to reproduce them if your database doesn't support them) - this is a function of SQL-Server (see info) and part of the reason I wrote this whole thing in the first place.
Part 2
- Subqueries - what they are, where they can be used and what to watch out for
- Cartesian joins AKA - Oh, the misery!

There are a number of ways to retrieve data from multiple tables in a database. In this answer, I will be using ANSI-92 join syntax. This may be different to a number of other tutorials out there which use the older ANSI-89 syntax (and if you are used to 89, may seem much less intuitive - but all I can say is to try it) as it is much easier to understand when the queries start getting more complex. Why use it? Is there a performance gain? The short answer is no, but it is easier to read once you get used to it. It is easier to read queries written by other folks using this syntax.

I am also going to use the concept of a small caryard which has a database to keep track of what cars it has available. The owner has hired you as his IT Computer guy and expects you to be able to drop him the data that he asks for at the drop of a hat.

I have made a number of lookup tables that will be used by the final table. This will give us a reasonable model to work from. To start off, I will be running my queries against an example database that has the following structure. I will try to think of common mistakes that are made when starting out and explain what goes wrong with them - as well as of course showing how to correct them.

The first table is simply a color listing so that we know what colors we have in the car yard.

mysql> create table colors(id int(3) not null auto_increment primary key, 
    -> color varchar(15), paint varchar(10));
Query OK, 0 rows affected (0.01 sec)

mysql> show columns from colors;
+-------+-------------+------+-----+---------+----------------+
| Field | Type        | Null | Key | Default | Extra          |
+-------+-------------+------+-----+---------+----------------+
| id    | int(3)      | NO   | PRI | NULL    | auto_increment |
| color | varchar(15) | YES  |     | NULL    |                |
| paint | varchar(10) | YES  |     | NULL    |                |
+-------+-------------+------+-----+---------+----------------+
3 rows in set (0.01 sec)

mysql> insert into colors (color, paint) values ('Red', 'Metallic'), 
    -> ('Green', 'Gloss'), ('Blue', 'Metallic'), 
    -> ('White' 'Gloss'), ('Black' 'Gloss');
Query OK, 5 rows affected (0.00 sec)
Records: 5  Duplicates: 0  Warnings: 0

mysql> select * from colors;
+----+-------+----------+
| id | color | paint    |
+----+-------+----------+
|  1 | Red   | Metallic |
|  2 | Green | Gloss    |
|  3 | Blue  | Metallic |
|  4 | White | Gloss    |
|  5 | Black | Gloss    |
+----+-------+----------+
5 rows in set (0.00 sec)

The brands table identifies the different brands of the cars out caryard could possibly sell.

mysql> create table brands (id int(3) not null auto_increment primary key, 
    -> brand varchar(15));
Query OK, 0 rows affected (0.01 sec)

mysql> show columns from brands;
+-------+-------------+------+-----+---------+----------------+
| Field | Type        | Null | Key | Default | Extra          |
+-------+-------------+------+-----+---------+----------------+
| id    | int(3)      | NO   | PRI | NULL    | auto_increment |
| brand | varchar(15) | YES  |     | NULL    |                |
+-------+-------------+------+-----+---------+----------------+
2 rows in set (0.01 sec)

mysql> insert into brands (brand) values ('Ford'), ('Toyota'), 
    -> ('Nissan'), ('Smart'), ('BMW');
Query OK, 5 rows affected (0.00 sec)
Records: 5  Duplicates: 0  Warnings: 0

mysql> select * from brands;
+----+--------+
| id | brand  |
+----+--------+
|  1 | Ford   |
|  2 | Toyota |
|  3 | Nissan |
|  4 | Smart  |
|  5 | BMW    |
+----+--------+
5 rows in set (0.00 sec)

The model table will cover off different types of cars, it is going to be simpler for this to use different car types rather than actual car models.

mysql> create table models (id int(3) not null auto_increment primary key, 
    -> model varchar(15));
Query OK, 0 rows affected (0.01 sec)

mysql> show columns from models;
+-------+-------------+------+-----+---------+----------------+
| Field | Type        | Null | Key | Default | Extra          |
+-------+-------------+------+-----+---------+----------------+
| id    | int(3)      | NO   | PRI | NULL    | auto_increment |
| model | varchar(15) | YES  |     | NULL    |                |
+-------+-------------+------+-----+---------+----------------+
2 rows in set (0.00 sec)

mysql> insert into models (model) values ('Sports'), ('Sedan'), ('4WD'), ('Luxury');
Query OK, 4 rows affected (0.00 sec)
Records: 4  Duplicates: 0  Warnings: 0

mysql> select * from models;
+----+--------+
| id | model  |
+----+--------+
|  1 | Sports |
|  2 | Sedan  |
|  3 | 4WD    |
|  4 | Luxury |
+----+--------+
4 rows in set (0.00 sec)

And finally, to tie up all these other tables, the table that ties everything together. The ID field is actually the unique lot number used to identify cars.

mysql> create table cars (id int(3) not null auto_increment primary key, 
    -> color int(3), brand int(3), model int(3));
Query OK, 0 rows affected (0.01 sec)

mysql> show columns from cars;
+-------+--------+------+-----+---------+----------------+
| Field | Type   | Null | Key | Default | Extra          |
+-------+--------+------+-----+---------+----------------+
| id    | int(3) | NO   | PRI | NULL    | auto_increment |
| color | int(3) | YES  |     | NULL    |                |
| brand | int(3) | YES  |     | NULL    |                |
| model | int(3) | YES  |     | NULL    |                |
+-------+--------+------+-----+---------+----------------+
4 rows in set (0.00 sec)

mysql> insert into cars (color, brand, model) values (1,2,1), (3,1,2), (5,3,1), 
    -> (4,4,2), (2,2,3), (3,5,4), (4,1,3), (2,2,1), (5,2,3), (4,5,1);
Query OK, 10 rows affected (0.00 sec)
Records: 10  Duplicates: 0  Warnings: 0

mysql> select * from cars;
+----+-------+-------+-------+
| id | color | brand | model |
+----+-------+-------+-------+
|  1 |     1 |     2 |     1 |
|  2 |     3 |     1 |     2 |
|  3 |     5 |     3 |     1 |
|  4 |     4 |     4 |     2 |
|  5 |     2 |     2 |     3 |
|  6 |     3 |     5 |     4 |
|  7 |     4 |     1 |     3 |
|  8 |     2 |     2 |     1 |
|  9 |     5 |     2 |     3 |
| 10 |     4 |     5 |     1 |
+----+-------+-------+-------+
10 rows in set (0.00 sec)

This will give us enough data (I hope) to cover off the examples below of different types of joins and also give enough data to make them worthwhile.

So getting into the grit of it, the boss wants to know The IDs of all the sports cars he has.

This is a simple two table join. We have a table that identifies the model and the table with the available stock in it. As you can see, the data in the model column of the cars table relates to the models column of the cars table we have. Now, we know that the models table has an ID of 1 for Sports so lets write the join.

select
    ID,
    model
from
    cars
        join models
            on model=ID

So this query looks good right? We have identified the two tables and contain the information we need and use a join that correctly identifies what columns to join on.

ERROR 1052 (23000): Column 'ID' in field list is ambiguous

Oh noes! An error in our first query! Yes, and it is a plum. You see, the query has indeed got the right columns, but some of them exist in both tables, so the database gets confused about what actual column we mean and where. There are two solutions to solve this. The first is nice and simple, we can use tableName.columnName to tell the database exactly what we mean, like this:

select
    cars.ID,
    models.model
from
    cars
        join models
            on cars.model=models.ID

+----+--------+
| ID | model  |
+----+--------+
|  1 | Sports |
|  3 | Sports |
|  8 | Sports |
| 10 | Sports |
|  2 | Sedan  |
|  4 | Sedan  |
|  5 | 4WD    |
|  7 | 4WD    |
|  9 | 4WD    |
|  6 | Luxury |
+----+--------+
10 rows in set (0.00 sec)

The other is probably more often used and is called table aliasing. The tables in this example have nice and short simple names, but typing out something like KPI_DAILY_SALES_BY_DEPARTMENT would probably get old quickly, so a simple way is to nickname the table like this:

select
    a.ID,
    b.model
from
    cars a
        join models b
            on a.model=b.ID

Now, back to the request. As you can see we have the information we need, but we also have information that wasn't asked for, so we need to include a where clause in the statement to only get the Sports cars as was asked. As I prefer the table alias method rather than using the table names over and over, I will stick to it from this point onwards.

Clearly, we need to add a where clause to our query. We can identify Sports cars either by ID=1 or model='Sports'. As the ID is indexed and the primary key (and it happens to be less typing), lets use that in our query.

select
    a.ID,
    b.model
from
    cars a
        join models b
            on a.model=b.ID
where
    b.ID=1

+----+--------+
| ID | model  |
+----+--------+
|  1 | Sports |
|  3 | Sports |
|  8 | Sports |
| 10 | Sports |
+----+--------+
4 rows in set (0.00 sec)

Bingo! The boss is happy. Of course, being a boss and never being happy with what he asked for, he looks at the information, then says I want the colors as well.

Okay, so we have a good part of our query already written, but we need to use a third table which is colors. Now, our main information table cars stores the car color ID and this links back to the colors ID column. So, in a similar manner to the original, we can join a third table:

select
    a.ID,
    b.model
from
    cars a
        join models b
            on a.model=b.ID
        join colors c
            on a.color=c.ID
where
    b.ID=1

+----+--------+
| ID | model  |
+----+--------+
|  1 | Sports |
|  3 | Sports |
|  8 | Sports |
| 10 | Sports |
+----+--------+
4 rows in set (0.00 sec)

Damn, although the table was correctly joined and the related columns were linked, we forgot to pull in the actual information from the new table that we just linked.

select
    a.ID,
    b.model,
    c.color
from
    cars a
        join models b
            on a.model=b.ID
        join colors c
            on a.color=c.ID
where
    b.ID=1

+----+--------+-------+
| ID | model  | color |
+----+--------+-------+
|  1 | Sports | Red   |
|  8 | Sports | Green |
| 10 | Sports | White |
|  3 | Sports | Black |
+----+--------+-------+
4 rows in set (0.00 sec)

Right, that's the boss off our back for a moment. Now, to explain some of this in a little more detail. As you can see, the from clause in our statement links our main table (I often use a table that contains information rather than a lookup or dimension table. The query would work just as well with the tables all switched around, but make less sense when we come back to this query to read it in a few months time, so it is often best to try to write a query that will be nice and easy to understand - lay it out intuitively, use nice indenting so that everything is as clear as it can be. If you go on to teach others, try to instill these characteristics in their queries - especially if you will be troubleshooting them.

It is entirely possible to keep linking more and more tables in this manner.

select
    a.ID,
    b.model,
    c.color
from
    cars a
        join models b
            on a.model=b.ID
        join colors c
            on a.color=c.ID
        join brands d
            on a.brand=d.ID
where
    b.ID=1

While I forgot to include a table where we might want to join more than one column in the join statement, here is an example. If the models table had brand-specific models and therefore also had a column called brand which linked back to the brands table on the ID field, it could be done as this:

select
    a.ID,
    b.model,
    c.color
from
    cars a
        join models b
            on a.model=b.ID
        join colors c
            on a.color=c.ID
        join brands d
            on a.brand=d.ID
            and b.brand=d.ID
where
    b.ID=1

You can see, the query above not only links the joined tables to the main cars table, but also specifies joins between the already joined tables. If this wasn't done, the result is called a cartesian join - which is dba speak for bad. A cartesian join is one where rows are returned because the information doesn't tell the database how to limit the results, so the query returns all the rows that fit the criteria.

So, to give an example of a cartesian join, lets run the following query:

select
    a.ID,
    b.model
from
    cars a
        join models b

+----+--------+
| ID | model  |
+----+--------+
|  1 | Sports |
|  1 | Sedan  |
|  1 | 4WD    |
|  1 | Luxury |
|  2 | Sports |
|  2 | Sedan  |
|  2 | 4WD    |
|  2 | Luxury |
|  3 | Sports |
|  3 | Sedan  |
|  3 | 4WD    |
|  3 | Luxury |
|  4 | Sports |
|  4 | Sedan  |
|  4 | 4WD    |
|  4 | Luxury |
|  5 | Sports |
|  5 | Sedan  |
|  5 | 4WD    |
|  5 | Luxury |
|  6 | Sports |
|  6 | Sedan  |
|  6 | 4WD    |
|  6 | Luxury |
|  7 | Sports |
|  7 | Sedan  |
|  7 | 4WD    |
|  7 | Luxury |
|  8 | Sports |
|  8 | Sedan  |
|  8 | 4WD    |
|  8 | Luxury |
|  9 | Sports |
|  9 | Sedan  |
|  9 | 4WD    |
|  9 | Luxury |
| 10 | Sports |
| 10 | Sedan  |
| 10 | 4WD    |
| 10 | Luxury |
+----+--------+
40 rows in set (0.00 sec)

Good god, that's ugly. However, as far as the database is concerned, it is exactly what was asked for. In the query, we asked for for the ID from cars and the model from models. However, because we didn't specify how to join the tables, the database has matched every row from the first table with every row from the second table.

Okay, so the boss is back, and he wants more information again. I want the same list, but also include 4WDs in it.

This however, gives us a great excuse to look at two different ways to accomplish this. We could add another condition to the where clause like this:

select
    a.ID,
    b.model,
    c.color
from
    cars a
        join models b
            on a.model=b.ID
        join colors c
            on a.color=c.ID
        join brands d
            on a.brand=d.ID
where
    b.ID=1
    or b.ID=3

While the above will work perfectly well, lets look at it differently, this is a great excuse to show how a union query will work.

We know that the following will return all the Sports cars:

select
    a.ID,
    b.model,
    c.color
from
    cars a
        join models b
            on a.model=b.ID
        join colors c
            on a.color=c.ID
        join brands d
            on a.brand=d.ID
where
    b.ID=1

And the following would return all the 4WDs:

select
    a.ID,
    b.model,
    c.color
from
    cars a
        join models b
            on a.model=b.ID
        join colors c
            on a.color=c.ID
        join brands d
            on a.brand=d.ID
where
    b.ID=3

So by adding a union all clause between them, the results of the second query will be appended to the results of the first query.

select
    a.ID,
    b.model,
    c.color
from
    cars a
        join models b
            on a.model=b.ID
        join colors c
            on a.color=c.ID
        join brands d
            on a.brand=d.ID
where
    b.ID=1
union all
select
    a.ID,
    b.model,
    c.color
from
    cars a
        join models b
            on a.model=b.ID
        join colors c
            on a.color=c.ID
        join brands d
            on a.brand=d.ID
where
    b.ID=3

+----+--------+-------+
| ID | model  | color |
+----+--------+-------+
|  1 | Sports | Red   |
|  8 | Sports | Green |
| 10 | Sports | White |
|  3 | Sports | Black |
|  5 | 4WD    | Green |
|  7 | 4WD    | White |
|  9 | 4WD    | Black |
+----+--------+-------+
7 rows in set (0.00 sec)

As you can see, the results of the first query are returned first, followed by the results of the second query.

In this example, it would of course have been much easier to simply use the first query, but union queries can be great for specific cases. They are a great way to return specific results from tables from tables that aren't easily joined together - or for that matter completely unrelated tables. There are a few rules to follow however.

The column types from the first query must match the column types from every other query below.
The names of the columns from the first query will be used to identify the entire set of results.
The number of columns in each query must be the same.

Now, you might be wondering what the difference is between using union and union all. A union query will remove duplicates, while a union all will not. This does mean that there is a small performance hit when using union over union all but the results may be worth it - I won't speculate on that sort of thing in this though.

On this note, it might be worth noting some additional notes here.

If we wanted to order the results, we can use an order by but you can't use the alias anymore. In the query above, appending an order by a.ID would result in an error - as far as the results are concerned, the column is called ID rather than a.ID - even though the same alias has been used in both queries.
We can only have one order by statement, and it must be as the last statement.

For the next examples, I am adding a few extra rows to our tables.

I have added Holden to the brands table.
I have also added a row into cars that has the color value of 12 - which has no reference in the colors table.

Okay, the boss is back again, barking requests out - *I want a count of each brand we carry and the number of cars in it!` - Typical, we just get to an interesting section of our discussion and the boss wants more work.

Rightyo, so the first thing we need to do is get a complete listing of possible brands.

select
    a.brand
from
    brands a

+--------+
| brand  |
+--------+
| Ford   |
| Toyota |
| Nissan |
| Smart  |
| BMW    |
| Holden |
+--------+
6 rows in set (0.00 sec)

Now, when we join this to our cars table we get the following result:

select
    a.brand
from
    brands a
        join cars b
            on a.ID=b.brand
group by
    a.brand

+--------+
| brand  |
+--------+
| BMW    |
| Ford   |
| Nissan |
| Smart  |
| Toyota |
+--------+
5 rows in set (0.00 sec)

Which is of course a problem - we aren't seeing any mention of the lovely Holden brand I added.

This is because a join looks for matching rows in both tables. As there is no data in cars that is of type Holden it isn't returned. This is where we can use an outer join. This will return all the results from one table whether they are matched in the other table or not:

select
    a.brand
from
    brands a
        left outer join cars b
            on a.ID=b.brand
group by
    a.brand

+--------+
| brand  |
+--------+
| BMW    |
| Ford   |
| Holden |
| Nissan |
| Smart  |
| Toyota |
+--------+
6 rows in set (0.00 sec)

Now that we have that, we can add a lovely aggregate function to get a count and get the boss off our backs for a moment.

select
    a.brand,
    count(b.id) as countOfBrand
from
    brands a
        left outer join cars b
            on a.ID=b.brand
group by
    a.brand

+--------+--------------+
| brand  | countOfBrand |
+--------+--------------+
| BMW    |            2 |
| Ford   |            2 |
| Holden |            0 |
| Nissan |            1 |
| Smart  |            1 |
| Toyota |            5 |
+--------+--------------+
6 rows in set (0.00 sec)

And with that, away the boss skulks.

Now, to explain this in some more detail, outer joins can be of the left or right type. The Left or Right defines which table is fully included. A left outer join will include all the rows from the table on the left, while (you guessed it) a right outer join brings all the results from the table on the right into the results.

Some databases will allow a full outer join which will bring back results (whether matched or not) from both tables, but this isn't supported in all databases.

Now, I probably figure at this point in time, you are wondering whether or not you can merge join types in a query - and the answer is yes, you absolutely can.

select
    b.brand,
    c.color,
    count(a.id) as countOfBrand
from
    cars a
        right outer join brands b
            on b.ID=a.brand
        join colors c
            on a.color=c.ID
group by
    a.brand,
    c.color

+--------+-------+--------------+
| brand  | color | countOfBrand |
+--------+-------+--------------+
| Ford   | Blue  |            1 |
| Ford   | White |            1 |
| Toyota | Black |            1 |
| Toyota | Green |            2 |
| Toyota | Red   |            1 |
| Nissan | Black |            1 |
| Smart  | White |            1 |
| BMW    | Blue  |            1 |
| BMW    | White |            1 |
+--------+-------+--------------+
9 rows in set (0.00 sec)

So, why is that not the results that were expected? It is because although we have selected the outer join from cars to brands, it wasn't specified in the join to colors - so that particular join will only bring back results that match in both tables.

Here is the query that would work to get the results that we expected:

select
    a.brand,
    c.color,
    count(b.id) as countOfBrand
from
    brands a
        left outer join cars b
            on a.ID=b.brand
        left outer join colors c
            on b.color=c.ID
group by
    a.brand,
    c.color

+--------+-------+--------------+
| brand  | color | countOfBrand |
+--------+-------+--------------+
| BMW    | Blue  |            1 |
| BMW    | White |            1 |
| Ford   | Blue  |            1 |
| Ford   | White |            1 |
| Holden | NULL  |            0 |
| Nissan | Black |            1 |
| Smart  | White |            1 |
| Toyota | NULL  |            1 |
| Toyota | Black |            1 |
| Toyota | Green |            2 |
| Toyota | Red   |            1 |
+--------+-------+--------------+
11 rows in set (0.00 sec)

As we can see, we have two outer joins in the query and the results are coming through as expected.

Now, how about those other types of joins you ask? What about Intersections?

Well, not all databases support the intersection but pretty much all databases will allow you to create an intersection through a join (or a well structured where statement at the least).

An Intersection is a type of join somewhat similar to a union as described above - but the difference is that it only returns rows of data that are identical (and I do mean identical) between the various individual queries joined by the union. Only rows that are identical in every regard will be returned.

A simple example would be as such:

select
    *
from
    colors
where
    ID>2
intersect
select
    *
from
    colors
where
    id<4

While a normal union query would return all the rows of the table (the first query returning anything over ID>2 and the second anything having ID<4) which would result in a full set, an intersect query would only return the row matching id=3 as it meets both criteria.

Now, if your database doesn't support an intersect query, the above can be easily accomlished with the following query:

select
    a.ID,
    a.color,
    a.paint
from
    colors a
        join colors b
            on a.ID=b.ID
where
    a.ID>2
    and b.ID<4

+----+-------+----------+
| ID | color | paint    |
+----+-------+----------+
|  3 | Blue  | Metallic |
+----+-------+----------+
1 row in set (0.00 sec)

If you wish to perform an intersection across two different tables using a database that doesn't inherently support an intersection query, you will need to create a join on every column of the tables.

Counting DISTINCT over multiple columns

If you are trying to improve performance, you could try creating a persisted computed column on either a hash or concatenated value of the two columns.

Once it is persisted, provided the column is deterministic and you are using "sane" database settings, it can be indexed and / or statistics can be created on it.

I believe a distinct count of the computed column would be equivalent to your query.

Basic Sql: Selecting The Same Column Multiple Times in One Query, When Each Occurrence Is Dependent on Different Where Clause