2019-06-29 16:58:41 +02:00
SELECT statement is used to retrieve records from one or more tables in PostgreSQL.
More about SELECT statement in postgres can be found at: https://www.postgresql.org/docs/11/sql-select.html
2019-07-02 10:31:52 +02:00
Following clauses are supported:
- SELECT(expressions...) - expressions to form output rows of the SELECT statement.
- DISTINCT() - remove all duplicate rows from result set
- FROM(tableSource...) - specifies one or more source tables for the SELECT.
- WHERE(condition) - only rows for which condition returns true will be selected.
- GROUP BY(groupingElement, ...) - will condense into a single row all selected rows that share the same values for the grouped expressions.
- HAVING(condition) - eliminates group rows that do not satisfy the condition
- ORDER BY(orderBy, ...) - causes the result rows to be sorted according to the specified expression(s)
- LIMIT(count) - specifies the maximum number of rows to return
- OFFSET(start) - specifies the number of rows to skip before starting to return rows
- FOR(lockMode) - how SELECT will lock rows as they are obtained from the table
Lock mode can be:
- jet.UPDATE()
- jet.NO_KEY_UPDATE()
- jet.SHARE()
- jet.KEY_SHARE()
Lock mode can be extended with following clauses:
- NOWAIT()
- SKIP_LOCKED()
- UNION(select) / UNION_ALL(select) - computes the set union of the rows returned by the involved SELECT statements
- INTERSECT(select) / INTERSECT_ALL(select) - computes the set intersection of the rows returned by the involved SELECT statements
- EXCEPT(select) / EXCEPT_ALL(select) - computes the set of rows that are in the result of the left SELECT statement but not in the result of the right one
_This list might be extended with feature Jet releases._
### Examples per clause
2019-06-29 16:58:41 +02:00
##### 1. SELECT clause
Sample SELECT clause written in Go:
```
jet.SELECT(
jet.Int(1).ADD(jet.Int(12)).SUB(jet.Int(21)), // arbitrary expression
table.Film.Name, // column
table.Customer.FirstName.CONCAT(table.Customer.LastName).AS("FullName") // alias
)
```
If jet and autogenerated table package, is dot "." imported, above statement will look more like native SQL:
```
SELECT(
Int(1).ADD(Int(12)).SUB(Int(21)), // arbitrary expression
Film.Name, // column
Customer.FirstName.CONCAT(Customer.LastName).AS("FullName") // alias
)
```
*For the most of the wiki dot import is used.*
Above SQL clause written in go will produce following raw SQL:
```
SELECT 1 + 12 - 21,
film.name AS "film.name", --
customer.first_name || customer.last_name AS "FullName"
```
`film.name AS "film.name"` - column names are aliased by default. Alias is used during execution to map row result to
appropriate `model` structure.
##### 2. DISTINCT clause
```
SELECT(Film.Name).
.DISTINCT().
```
Raw SQL:
```
SELECT DISTINCT film.name
```
##### 3. FROM clause
```
Go:
1).FROM(Film)
2).FROM(
Film.
INNER_JOIN(Language, Langauge.LanguageID.EQ(Film.FilmID))
)
SQL:
1)FROM dvds.film
2)FROM Film
INNER JOIN Language ON (Language.LanguageID = Film.FilmID)
```
##### 4. WHERE clause
```
Go:
.WHERE(Film.Length.GT(Int(150)))
SQL:
WHERE film.length > 150
```
##### 5. GROUP BY clause
```
Go:
.GROUP_BY(Film.Length)
SQL:
GROUP BY film.length
```
##### 6. HAVING clause
```
Go:
.HAVING(Film.Length.GT(Int(150))
SQL:
HAVING film.length > 150
```
##### 7. ORDER BY clause
```
Go:
.ORDER_BY(Film.Length)
SQL:
ORDER BY film.length
```
##### 8. LIMIT clause
```
Go:
.LIMIT(11)
SQL:
LIMIT 11
```
##### 9. OFFSET clause
```
Go:
.OFFSET(11)
SQL:
OFFSET 11
```
2019-07-01 13:18:39 +02:00
##### 9. FOR clause
2019-06-29 16:58:41 +02:00
```
Go:
2019-07-01 13:18:39 +02:00
.FOR(jet.NO_KEY_UPDATE().SKIP_LOCKED())
2019-06-29 16:58:41 +02:00
SQL:
2019-07-01 13:18:39 +02:00
FOR NO KEY UPDATE SKIP LOCKED
2019-06-29 16:58:41 +02:00
```
2019-07-02 10:31:52 +02:00
##### 10. Set clauses (UNION, UNION_ALL, INTERSECT, INTERSECT_ALL, EXCEPT, EXCEPT_ALL)
```
Go:
SELECT(Payment.Amount).FROM(Payment)
UNION_ALL(SELECT(Payment.Amount).FROM(Payment))
Sql:
(
(
SELECT payment.amount AS "payment.amount"
FROM dvds.payment
)
UNION
(
SELECT payment.amount AS "payment.amount"
FROM dvds.payment
)
);
```
2019-07-01 13:18:39 +02:00
2019-06-29 16:58:41 +02:00
### Two forms of select statements in Jet
#### 1. Classical select statement
2019-07-02 10:31:52 +02:00
Columns selected are before table source selected
2019-06-29 16:58:41 +02:00
```
SELECT(
Payment.AllColumns,
Customer.AllColumns,
).
FROM(Payment.
INNER_JOIN(Customer, Payment.CustomerID.EQ(Customer.CustomerID))).
ORDER_BY(Payment.PaymentID.ASC()).
LIMIT(30)
```
#### 2. Jet select statement
2019-07-02 10:31:52 +02:00
Table sources are before columns selected. There is no FROM.
2019-06-29 16:58:41 +02:00
```
Payment.
INNER_JOIN(Customer, Payment.CustomerID.EQ(Customer.CustomerID))
SELECT(
Payment.AllColumns,
Customer.AllColumns,
).
ORDER_BY(Payment.PaymentID.ASC()).
LIMIT(30)
```
Second form is added, because sometimes feels more natural to first think about the tables
of interest, and than about the columns.
**Both forms produces exactly the same raw SQL.**
## Sub-queries
How to write SELECT statement with sub-queries?
Sub-queries are composed first:
```
// select film_id and title from film table that have 'R' rating.
rRatingFilms := Film.
SELECT(
Film.FilmID,
Film.Title,
).
WHERE(Film.Rating.EQ(enum.MpaaRating.R)).
AsTable("rFilms")
```
2019-07-04 14:12:59 +02:00
`AsTable("rFilms")` - allows SELECT statements to be seen as source table.
2019-06-29 16:58:41 +02:00
To use sub-query columns in SELECT statement expressions we have to export column from sub-query,
using `From` method.
```
rFilmId := Film.FilmID.From(rRatingFilms) //used for join condition
```
Now we can write:
```
query := Actor.
INNER_JOIN(FilmActor, Actor.ActorID.EQ(FilmActor.FilmID)).
INNER_JOIN(rRatingFilms, FilmActor.FilmID.EQ(rFilmId)).
SELECT(
Actor.AllColumns,
FilmActor.AllColumns,
rRatingFilms.AllColumns(),
)
```
`rRatingFilms.AllColumns(),` - sub-query columns needed for projection can be exported with AllColumns() method.
No need to export each one of them with `From` .
Debug SQL of above example:
```sql
SELECT actor.actor_id AS "actor.actor_id",
actor.first_name AS "actor.first_name",
actor.last_name AS "actor.last_name",
actor.last_update AS "actor.last_update",
film_actor.actor_id AS "film_actor.actor_id",
film_actor.film_id AS "film_actor.film_id",
film_actor.last_update AS "film_actor.last_update",
"rFilms"."film.film_id" AS "film.film_id",
"rFilms"."film.title" AS "film.title",
"rFilms"."film.rating" AS "film.rating"
FROM dvds.actor
INNER JOIN dvds.film_actor ON (actor.actor_id = film_actor.film_id)
INNER JOIN (
SELECT film.film_id AS "film.film_id",
film.title AS "film.title",
film.rating AS "film.rating"
FROM dvds.film
WHERE film.rating = 'R'
) AS "rFilms" ON (film_actor.film_id = "rFilms"."film.film_id");
```
#### What If sub-query projection is not a column?
For instance:
```
customersPayments := Payment.
SELECT(
Payment.CustomerID,
SUMf(Payment.Amount).AS("amount_sum"),
).
GROUP_BY(Payment.CustomerID).
AsTable("customer_payment_sum")
customerId := Payment.CustomerID.From(customersPayments)
```
To export `"amount_sum"` from `customersPayments` sub-query we have to create column first with appropriate type and a name.
Because SUMf produces float expression we will create FloatColumn with name of the column alias `"amount_sum"`
```
amountSum := FloatColumn("amount_sum").From(customersPayments)
```