Skip to content

Latest commit

 

History

History
466 lines (369 loc) · 8.79 KB

sql.md

File metadata and controls

466 lines (369 loc) · 8.79 KB
Raw

Installation MYSQL


Installation

https://www.youtube.com/watch?v=GIRcpjg-3Eg

SELECT, FROM


Outputs all columns from the table customer (automatically sorted by id)

select * from customer

Outputs only the column firstname (no sorting)

select firstname from customer

Outputs 2 columns firstname and lastname

select firstname, lastname from customer

DISTINCT, WHERE, BETWEEN, IN


With distinct the output occurs only one time if the firstname is the same

select distinct firstname from customer

Everything will be outputed cause the id is unique

select distinct id,firstname from customer

Only when firstname and lastname are equal there will be no outputs

select distinct firstname,lastname


Outputs the row with the id = 0

select * from customer
    where id = 0


Outputs all rows where the firstname is Janet

select * from customer
    where firstname = 'Janet'


Outputs all streets which are in the city of Oslo

select street from customer
    where city = 'Oslo'


all IDs >= 40

...where id >= 40

all IDs except 40 (in some DBs you have to use instead !=)

...where id <> 40

everything between 40 and 50

...where id between 40 and 50

some values are checked for id

...where id in (43,44,45)

AND, OR, ORDER BY, ASC, DESC

Outputs all rows where id >= 40 and firstname is Robert

select * from customer
where id >= 40
and firstname = "Robert"


One of the both conditions have to be fullfilled

...where id >= 40 or firstname = "Robert"

combinaton of logical or and and (better to use parantheses)

...where id >= 40 or firstname = "Robert" and lastname = "Fuller"


Outputs rows >=40 sorted ascending for firstname and lastname

select * from customer
where id >= 40
order by firstname, lastname asc


Output in descending order

...order by firstname, lastname desc

INSERT, INTO, VALUES, UPDATE, SET, DELETE

Inserts row in the table with specific values for the columns

insert into customer
values (50, 'James', 'Karsen', '107 Upland Pl.', 'Dallas')


Inserts row with streetname = NULL (this is not working for key-columns like id)

insert into customer (id, firstname, lastname, city)
values (51, 'James', 'Karsen', 'Dallas')


Update the row with the id = 51 regarding street

update customer
set street = '547 Seventh Av.'
where id = 51


Update the strett where firstname James and lastname Karsen

...where firstname = 'James' and lastname = 'Karsen'


Delete row with the id = 51

delete from customer
where id = 51

NULL, LIKE, TOP, LIMIT, ROWNUM, PERCENT


Outputs the first 5 results (depending on the sorting)

select top 5 * from customer


in some DBs also in the following form - oututs the first 10 results

select * from customer
limit 10


and also in this form in some DBs

select * from customer
where rownum <= 10


Outputs 10 top percent of the results

select top 10 percent * from customer


"_" will be replaced with one char - so the output will be done for eg. Oslo or Oxlo

select * from computer
where city like 'O_lo'


Outputs all entries which end with "lo" for the column city

... where city like '%lo'

Outputs all entries which start with 2 chars and then the char "l"

... where city like '__l%'

When there sould be a serach for the percent sign "%" - this has to be masked with %

... where city like '__l\%'


Outputs all rows where lastname is NULL

select * from computer
where lastname is NULL


Outputs all rows where there is some entry in the column lastname

where lastname is NOT NULL

IN

IDs will be selected from the rechnungen-table and will be used for the aboth where-clausel for select *

select * from customer
where id IN
        (select id from rechnungen
        where date <=3)

Tables and Data-Types

CREATE, TABLE, INT, VARCHAR, BINARY, BOOLEAN, VARBINARY, SMALLINT, BIGINT, DECIMAL, NUMERIC, DATE, TIME


Create a new table with name Rechnung

create table Rechnung
    (
    RechnungsID int,
    CustomerID, int,
    Betrag int
    )


Insert a row in the new table Rechnung with specific values

insert into Rechnung values (1,0,50)

Integer

value INT

variable Character

name VARCHAR(255)

Bits

bin BINARY (255)

Bool (TRUE / FALSE)

bool BOOLEAN

variable Bits

VARBINARY(255)

SmallInt

SMALLINT

BigInt

BIGINT

Decimal Number (total,decimal places) - 3 beforce decimal places, 2 decimal places

DECIMAL (5,2)

Numeric Number

NUMERIC (5,2)

Floating Number

FLOAT (10)

Double Precision

DOUBLE PRECISION

Date value

DATE

Time value

TIME

Timestamp value

TIMESTAMP

Primary und Foreign Keys, not null

DROP, UNIQUE, NOT NULL, PRIMARY KEY, FOREIGN KEY, REFERENCES
Drop whole table

drop table Rechnung IF EXISTS


RechnungID is a mandatory field - must be allways available and not empty with NULL - with UNIQUE the ID has to be unique
Define the Primary Key with PRIMARY KEY
Define the Foreign Key with FOREIN KEY (Primary Key in anderer Tabelle) - is referencing to the id in the customer table

create table Rechnung
    (
    RechnungsID int NOT NULL UNIQUE
    CustomerID int NOT NULL
    Betrag int NOT NULL
    PRIMARY KEY(RechnungsID)
    FOREIGN KEY(CustomerID) REFERENCES Customer(ID)
    )

Change Tables, Autoincrement

AUTO_INCREMENT, DEFAULT, IDENTITY, ALTER TABLE, ADD, DROP COLUMN, ALTER COLUMN
With AUTO_INCREMENT the id will be given automatically - starts with 1 and then ascending (e.g. MYSQL)
In some DBs with: RechnungsID int IDENTITY(0,1) (e.g. MSSQL)
In some DBs with: RechnungsID int IDENTITY (e.g. HSQL)
Using DEFAULT - when something is inserted without specific value - 50 will be used as default value

create table Rechnung
    (
    RechnungsID int NOT NULL AUTO_INCREMENT
    CustomerID int NOT NULL,
    Betrag int DEFAULT 50,
    PRIMARY KEY(RechnungsID),
    FOREIGN KEY(CustomerID) REFERENCES Customer(ID)
    )


change table Rechnung

alter table Rechnung

add datum-column with datatype date

add Datum Date

drop column betrag from the table

drop column betrag

change column to VARCHAR(255)

alter column VARCHAR(255)

Joins

INNER JOIN, ON, LEFT JOIN, RIGTH JOIN, FULL JOIN
Outputs firstname + lastname from all customers which exists in the rechnung table

select firstname, lastname from customer
where id in (select customerID from rechnung)


Outputs all attributes from the customers, which have a rechnung (same as aboth - but for all attributes)

select *
from customer
inner join Rechnung
on customer.ID = rechnung.customerID


Outputs all elements from the left table (customer) - where in the table Rechnung something is inside it will be outputed

...left join Rechnung

Outputs alle elements from the right table (Rechnung) - with informations from both tables

...right join Rechnung

Outputs everything from both sides

...full join Rechnung

SQL Built In Funktionen

AVG, COUNT, TOP, FIRST, LIMIT, LAST, UCASE, UPPER, LCASE, LOWER
Outputs the average

select avg(betrag) from rechnung

Returns the count of the rows

select count(rechnungID) from rechnung


Outputs the count of the rechnungIDs where the betrag is >= the average in the rechnungs table

select count(rechnungID) from rechnung
where betrag >= (select AVB(betrag) from rechnung)


Outputs maximum betrag from the table

select max(betrag) from rechnung

Outputs the minimum betrag from the table

select min(betrag) from rechnung

Outputs the sum betrag from the table

select sum(betrag) from rechnung

Outputs the firstname as uppercase - sometimes also upper(firstname)

select ucase(firstname) from customer

Outputs the firstname as lowercase - sometimes also lower(firstname)

select lcase(firstname) from customer

Outputs the length of the string

select len(firstname) from customer

Outputs the actual date + time

select now() from  customer