Lets review all of the Table-related functions in one place.

Create tables in the database

Top create a table in your database use the CREATE TABLE command. Here, we define the schema of a table. We create users table which has two attributes/fields/columns, name and email. Each of these attributes are defined as character strings with max of 128 characters

people=> CREATE TABLE users(
people(> name VARCHAR(128),
people(> email VARCHAR(128)
people(> );
CREATE TABLE
people=>

List tables in the database

To list tables/relations in the database use to he following command:

people=> \dt
Did not find any relations.

if you haven't created any tables in your database yet, you will get "Did not find any relations." message. Otherwise, you will get list of existing tables like what follows:

people=> \dt
       List of relations
 Schema | Name  | Type  | Owner
--------+-------+-------+-------
 public | users | table | pg4e
(1 row)

people=>

Get the schema of a table

Lets suppose we have forgotten what was the schema (the definition of the table and its attributes/columns), and we want to review the schema one more time. In this case we can use the following command:

people=> \d+ users
                                                  Table "public.users"
 Column |          Type          | Collation | Nullable | Default | Storage  | Compression | Stats target | Description
--------+------------------------+-----------+----------+---------+----------+-------------+--------------+-------------
 name   | character varying(128) |           |          |         | extended |             |              |
 email  | character varying(128) |           |          |         | extended |             |              |
Access method: heap

people=>

Basic CRUD commands

INSERT/create row(s) into the table

Suppose we wan to insert several rows/objects into the table. We have to use the following command:

people=> INSERT INTO users (name, email) VALUES ('Peter', 'peter@example.edu');
INSERT 0 1
people=> INSERT INTO users (name, email) VALUES ('Terry', 'terry@example.edu');
INSERT 0 1
people=> INSERT INTO users (name, email) VALUES ('Eliza', 'eliza@example.edu');
INSERT 0 1
people=> INSERT INTO users (name, email) VALUES ('John', 'john@example.edu');
INSERT 0 1

READ/SELECT row(s) from a table

In order to read the rows in a table we should use SELECT command:

people=> SELECT * FROM users;
 name  |       email
-------+-------------------
 Peter | peter@example.edu
 Terry | terry@example.edu
 Eliza | eliza@example.edu
 John  | john@example.edu
(4 rows)

We can include some criteria in the SELECT command as well:

people=> SELECT * FROM users WHERE email='terry@example.edu';
 name  |       email
-------+-------------------
 Terry | terry@example.edu
(1 row)

UPDATE row(s) in a table

In order to update one/more piece(s) of information in one/more row(s), we use UPDATE command like what follows:

people-> UPDATE users SET name='Johnny' WHERE email='john@example.edu';
UPDATE 1

Hint: this command will update any number of rows that meet its criteria.

Lets see how this command changed the last entry of users table:

people=> SELECT * FROM users;
  name  |       email
--------+-------------------
 Peter  | peter@example.edu
 Terry  | terry@example.edu
 Eliza  | eliza@example.edu
 Johnny | john@example.edu
(4 rows)

DELETE row(s) from a table

To remove one/more row(s) from a table, use the DELETE command like what follows:

people=> DELETE FROM users WHERE email='terry@example.edu';
DELETE 1

Hint: this command will delete any number of rows that meet its criteria.

Lets see how this command removed an entry of users table:

people=> SELECT * FROM users;
  name  |       email
--------+-------------------
 Peter  | peter@example.edu
 Eliza  | eliza@example.edu
 Johnny | john@example.edu
(3 rows)

ALTER TABLE

What happens if you missed something in the time of creation of a table in database? Is it over? ABSOLUTELY NOT! SQL lets us change the table schema in the times needed.

When you use ALTER TABLE and change the definition of one field, a SQL-based database applies that change to both the table's schema and to all of the existing data in that table, all while the system is running and functional. So there are a couple of actions like add, change, drop a column that you can do using ALTER TABLE key words. Lets suppose that we have created the following table and we want to make changes to this table.

CREATE TABLE users(
name VARCHAR(128),
email VARCHAR(128)
);

First, lets take a look at the table's schema:

people=# \d+ users
                                                  Table "public.users"
 Column |          Type          | Collation | Nullable | Default | Storage  | Compression | Stats target | Description
--------+------------------------+-----------+----------+---------+----------+-------------+--------------+-------------
 name   | character varying(128) |           |          |         | extended |             |              |
 email  | character varying(128) |           |          |         | extended |             |              |
Access method: heap

Add a column to a table

In order to add a column with " address VARCHAR(128) " definition to the users table, we have to use the following code. You can see the changes into the table's schema

people=# ALTER TABLE users ADD COLUMN address VARCHAR(128);
ALTER TABLE

people=# \d+ users
                                                  Table "public.users"
 Column  |          Type          | Collation | Nullable | Default | Storage  | Compression | Stats target | Description
---------+------------------------+-----------+----------+---------+----------+-------------+--------------+-------------
 name    | character varying(128) |           |          |         | extended |             |              |
 email   | character varying(128) |           |          |         | extended |             |              |
 address | character varying(128) |           |          |         | extended |             |              |
Access method: heap

Change type of a column

To change type of address column from VARCHAR(128) to TEXT, we have to use to following code:

people=# ALTER TABLE users ALTER COLUMN address TYPE TEXT;
ALTER TABLE
people=# \d+ users
                                                  Table "public.users"
 Column  |          Type          | Collation | Nullable | Default | Storage  | Compression | Stats target | Description
---------+------------------------+-----------+----------+---------+----------+-------------+--------------+-------------
 name    | character varying(128) |           |          |         | extended |             |              |
 email   | character varying(128) |           |          |         | extended |             |              |
 address | text                   |           |          |         | extended |             |              |
Access method: heap

Drop a column of a table

In order to drop the address column of users table, we have to use the following code:

people=# ALTER TABLE users DROP address;
ALTER TABLE
people=# \d+ users
                                                  Table "public.users"
 Column |          Type          | Collation | Nullable | Default | Storage  | Compression | Stats target | Description
--------+------------------------+-----------+----------+---------+----------+-------------+--------------+-------------
 name   | character varying(128) |           |          |         | extended |             |              |
 email  | character varying(128) |           |          |         | extended |             |              |
Access method: heap

UPDATE

What happens if we wanted to update content of one/several row(s)? In this case we can you UPDATE key word.

Update content of row(s) in a table

First Lets take a look at the current contents of the users table:

people=# SELECT * FROM users;
  name  |       email
--------+-------------------
 Peter  | peter@example.edu
 Eliza  | eliza@example.edu
 Johnny | john@example.edu
 Terry  | terry@example.edu
(4 rows)

Now, lets change the email address of the user named Johnny, and watch the results:

people=# UPDATE users SET email='johnny@example.edu' WHERE name='johnny';
UPDATE 1
people=# SELECT * FROM users;
  name  |       email
--------+--------------------
 Peter  | peter@example.edu
 Eliza  | eliza@example.edu
 Terry  | terry@example.edu
 Johnny | johnny@example.edu
(4 rows)

Important basic have-to-knows before learning SQL:

1. What are Database?

2. What is a Relational Database?

3. What is SQL (Structured Query Language)?

4. What are the most popular SQL-Based Databases?

   • MySQL:Fast and scalable, Commercial Open Source.

   • Oracle: Commercial, Enterprise-Scale.

   • MariaDB: Open Source.

   • PostgreSQL: Opensource.

   • Some other smaller projects like: SQLite, HSQL, ...

5. What does CRUD (Create, Read/Select, Update, Delete) mean?

   The four basic actions that one can do using databases are:

   • Creating/Inserting Data

   • Read/Select some Data

   • Update Data

   • Delete Data

6. In a Relational Database, what does a Database represent?

   It contains one or more tables.

7. In a Relational Database, what does a Relation/Tuple represent?

   It contains tuples and attributes.

8. In a Relational Database, what does a Tuple/Row represent?

   a set of fields which generally represent an "Object", like a person or a music track

9. In a Relational Database, what does an Attribute/Column/Field represent?

   One pf the possible many elements of data corresponding to the object represented by the row

   

10. What is a Database Schema?

11. What is the SQL Architecture? How SQL commands are run on a SQL Server (regardless of what kind of SQL Client and/or Server you are using)?

    

Difference between PSQL and SQL commands

First of all, lets differentiate between PostgreSQL and SQL commands:

• Any command starting with backslash \ is a PostgreSQL command.

  For example, commands like getting a list of databases \l, getting a list of tables \dt, getting a table schema \d+ <TableName> and so forth are all PostgreSQL commands since they are starting with \

• otherwise, its SQL command.

  For Instance, commands like SELECT FROM <TableName>, DROP TABLE <TableName>, DELETE * FROM <TableName> WHERE <ConditionDescription> and so forth are all SQL command.

• Since here we are using PSQL client, any command that starts with psql mean that we are asking the client to do some task. Therefore, If you are using another client, you should substitute psql part of that command with the name of your PSQL client.

In what follows, I won't explicitly state whether a command is a PSQL or SQL one.

(Dis)Connecting to databases

Connect to the default database

After Installation is completed, connect to the default server using the following commands:

C:\>psql -U postgres
psql (16.1)
WARNING: Console code page (850) differs from Windows code page (1252)
         8-bit characters might not work correctly. See psql reference
         page "Notes for Windows users" for details.
Type "help" for help.

postgres=#

The postgres=# prompt show that we have connected to the "postgres" database, and the # shows that we have connected to this database with a superuser user.

Connect to another specific database

To connect to any database we should use psql command.

-d DatabaseName : The first parameter of this command is "database name"

-U DatabaseUser : The second parameter is name of the user with whom credentials' we want to connect to the database.

-W : push the server to ask for password

C:\>psql -d people -U pg4e -W
Password:
psql (16.1)
WARNING: Console code page (850) differs from Windows code page (1252)
         8-bit characters might not work correctly. See psql reference
         page "Notes for Windows users" for details.
Type "help" for help.

people=>

The people=> prompt show that we have connected to the "people" database, and the > shows that we have connected to this database with a non-superuser user.

Connecting to a database on another host

To connect to any database on another host we should use the following command:

-h <HostAddress> : Specifies Address of the host on which the database is.

-p <PortNumber> : Specifies the port number to connect to that host

-d <DatabaseName> : Specifies "database name"

-U <DatabaseUser> : Specifies name of the user with whom credentials' we want to connect to the database.

-W : push the server to ask for password

C:\>psql -h example.com -p 5432 -d people -U pg4e -W
Password:
psql (16.1)
WARNING: Console code page (850) differs from Windows code page (1252)
         8-bit characters might not work correctly. See psql reference
         page "Notes for Windows users" for details.
Type "help" for help.

people=>

The people=> prompt show that we have connected to the "people" database, and the > shows that we have connected to this database with a non-superuser user.

Disconnect (quit) a session

In order to quit an PostgreSQL session use the following command:

postgres=# \q

Creating and listing database, table and user

Get the database version

Lest get the version of you PostgreSQL database:

SELECT version();

Get list of databases

Now get list of default databases created by the installer while installing the PostgreSQL database:

postgres=# \l
                                                                      List of databases
   Name    |  Owner   | Encoding | Locale Provider |          Collate           |           Ctype            | ICU Locale | ICU Rules |   Access privileges
-----------+----------+----------+-----------------+----------------------------+----------------------------+------------+-----------+-----------------------
 postgres  | postgres | UTF8     | libc            | English_United States.1252 | English_United States.1252 |            |           |
 template0 | postgres | UTF8     | libc            | English_United States.1252 | English_United States.1252 |            |           | =c/postgres          +
           |          |          |                 |                            |                            |            |           | postgres=CTc/postgres
 template1 | postgres | UTF8     | libc            | English_United States.1252 | English_United States.1252 |            |           | =c/postgres          +
           |          |          |                 |                            |                            |            |           | postgres=CTc/postgres
(3 rows)

Create a new Database

Use the following command to create a database named "people" whose owner is "pg4e" user:

postgres=# CREATE DATABASE people WITH OWNER 'pg4e';
CREATE DATABASE

Create a new non-superuser User

Use the Following command to create a user named "pg4e" with password "secret"

postgres=# CREATE USER pg4e WITH PASSWORD 'secret';
CREATE ROLE

Create tables in the database

Top create a table in your database use the CREATE TABLE command. Here, we define the schema of a table. We create users table which has two attributes/fields/columns, name and email. Each of these attributes are defined as character strings with max of 128 characters

people=> CREATE TABLE users(
people(> name VARCHAR(128),
people(> email VARCHAR(128)
people(> );
CREATE TABLE
people=>

List tables in the database

To list tables/relations in the database use to he following command:

people=> \dt
Did not find any relations.

if you haven't created any tables in your database yet, you will get "Did not find any relations." message. Otherwise, you will get list of existing tables like what follows:

people=> \dt
       List of relations
 Schema | Name  | Type  | Owner
--------+-------+-------+-------
 public | users | table | pg4e
(1 row)

people=>

Get the schema of a table

Lets suppose we have forgotten what was the schema (the definition of the table and its attributes/columns), and we want to review the schema one more time. In this case we can use the following command:

people=> \d+ users
                                                  Table "public.users"
 Column |          Type          | Collation | Nullable | Default | Storage  | Compression | Stats target | Description
--------+------------------------+-----------+----------+---------+----------+-------------+--------------+-------------
 name   | character varying(128) |           |          |         | extended |             |              |
 email  | character varying(128) |           |          |         | extended |             |              |
Access method: heap

people=>

Basic CRUD commands

INSERT/create row(s) into the table

Suppose we wan to insert several rows/objects into the table. We have to use the following command:

people=> INSERT INTO users (name, email) VALUES ('Peter', 'peter@example.edu');
INSERT 0 1
people=> INSERT INTO users (name, email) VALUES ('Terry', 'terry@example.edu');
INSERT 0 1
people=> INSERT INTO users (name, email) VALUES ('Eliza', 'eliza@example.edu');
INSERT 0 1
people=> INSERT INTO users (name, email) VALUES ('John', 'john@example.edu');
INSERT 0 1

READ/SELECT row(s) from a table

In order to read the rows in a table we should use SELECT command:

people=> SELECT * FROM users;
 name  |       email
-------+-------------------
 Peter | peter@example.edu
 Terry | terry@example.edu
 Eliza | eliza@example.edu
 John  | john@example.edu
(4 rows)

We can include some criteria in the SELECT command as well:

people=> SELECT * FROM users WHERE email='terry@example.edu';
 name  |       email
-------+-------------------
 Terry | terry@example.edu
(1 row)

UPDATE row(s) in a table

In order to update one/more piece(s) of information in one/more row(s), we use UPDATE command like what follows:

people-> UPDATE users SET name='Johnny' WHERE email='john@example.edu';
UPDATE 1

Hint: this command will update any number of rows that meet its criteria.

Lets see how this command changed the last entry of users table:

people=> SELECT * FROM users;
  name  |       email
--------+-------------------
 Peter  | peter@example.edu
 Terry  | terry@example.edu
 Eliza  | eliza@example.edu
 Johnny | john@example.edu
(4 rows)

DELETE row(s) from a table

To remove one/more row(s) from a table, use the DELETE command like what follows:

people=> DELETE FROM users WHERE email='terry@example.edu';
DELETE 1

Hint: this command will delete any number of rows that meet its criteria.

Lets see how this command removed an entry of users table:

people=> SELECT * FROM users;
  name  |       email
--------+-------------------
 Peter  | peter@example.edu
 Eliza  | eliza@example.edu
 Johnny | john@example.edu
(3 rows)

Sorting

In order to sort list of objects we can use ORDER BY clause in SELECT statements. By default it sort the object by ascending order. However, you can explicitly state the order using ASC or DESC to sort the objects in ascending or descending manner respectively.

In the following example, you can see that at first the objects are not sorted.

people=> SELECT * FROM users;
  name  |       email
--------+-------------------
 Peter  | peter@example.edu
 Eliza  | eliza@example.edu
 Johnny | john@example.edu
(3 rows)

Then, when we use ORDER BY clause, it by default sorts the objects in ascending manner. Also, notice that adding the ASC at the end of the command leads to the same result.

people=> SELECT * FROM users ORDER BY email;
  name  |       email
--------+-------------------
 Eliza  | eliza@example.edu
 Johnny | john@example.edu
 Peter  | peter@example.edu
(3 rows)

people=> SELECT * FROM users ORDER BY email ASC;
  name  |       email
--------+-------------------
 Eliza  | eliza@example.edu
 Johnny | john@example.edu
 Peter  | peter@example.edu
(3 rows)

You can also change the order by stating DESC at the end of the command.

people=> SELECT * FROM users ORDER BY email DESC;
  name  |       email
--------+-------------------
 Peter  | peter@example.edu
 Johnny | john@example.edu
 Eliza  | eliza@example.edu
(3 rows)

Wildcards/Like clause

In order to find specific objects we can use LIKE operator in WHERE clauses. For example, to find all the users in whose name have e, we can use the following command:

people=> SELECT * FROM users WHERE name LIKE '%e%'
;
 name  |       email
-------+-------------------
 Peter | peter@example.edu
 Terry | terry@example.edu
(2 rows)

Keep in mind that as a result of LIKE operator, the WHOLE table will be scanned and the database cannot use indexes to find the records. Therefore, it is a very slow operation.

Paging/skipping objects

You can ask the database to limit the number objects in the list by using LIMIT clause. Also you can ask the database to skip a specific number of objects from the beginning of the list by using OFFSET clause.

It is worth mentioning that WHERE and ORDER BY clauses will be executed before LIMIT and OFFSET clauses. In addition note that OFFSET and LIMIT are both very fast and efficient clauses.

Lets see the initial list of the objects in the users table in both orders.

people=> SELECT * FROM users ORDER BY email ASC;
  name  |       email
--------+-------------------
 Eliza  | eliza@example.edu
 Johnny | john@example.edu
 Peter  | peter@example.edu
 Terry  | terry@example.edu
(4 rows)

people=> SELECT * FROM users ORDER BY email DESC;
  name  |       email
--------+-------------------
 Terry  | terry@example.edu
 Peter  | peter@example.edu
 Johnny | john@example.edu
 Eliza  | eliza@example.edu
(4 rows)

LIMIT clause

people=> SELECT * FROM users ORDER BY email DESC LIMIT 2;
 name  |       email
-------+-------------------
 Terry | terry@example.edu
 Peter | peter@example.edu
(2 rows)

OFFSET clause

OFFSET starts indexing from 0, therefore index 1 will be the second object in the list.

people=> SELECT * FROM users ORDER BY email OFFSET 1 LIMIT 2;
  name  |       email
--------+-------------------
 Johnny | john@example.edu
 Peter  | peter@example.edu
(2 rows)

Counting

In order to count the number of rows in a result of a query, you can use COUNT clause. COUNT is actually faster than WHERE.

people=> SELECT COUNT(*) FROM users;
 count
-------
     4
(1 row)

people=> SELECT COUNT(*) FROM users WHERE name LIKE '%e%';
 count
-------
     2
(1 row)

Well, this is the very core of PostgreSQL and I will continue this topic in the further articles.

P.S. The cover is generated by DALL-E.

1. Briefly look at the data and dataset

   Take a brief look at your data and try to understand the columns. If you have a document describing the rows and columns, it is in your best interest to read it at least shortly to gain an overview of your dataset.

2. Identify the potential questions
   Try to list several questions regarding the dataset and the context you are working with. Of course, while you are trying to answer these initial questions, you probably face other questions and as you go on the whole context becomes more clear and you will find yourself closer to the most important insights of the dataset. So do not let your possibly short list of questions stop you in this step, push yourself to the next steps and dive into the "data exploration" part.

3. Create some simple sheets and find simple insights

   Try not to be so preoccupied with the style and picky with the correct color of your sheets at this stage. There is a very high chance that you change the color palette as you get more insights from the dataset and reach the end of your viz project. Instead, try to create some rudimentary charts and tables based on your initial questions, understand the relations among the columns and find out the probable need to create other calculated variables to provide better insights.

Make the data pop out

Clean your chart to the point that your audience can get the message; The more we omit the cluttered elements in the chart the more our data pops out. While omitting the unnecessary parts strongly depends on the context that you are working with, applying the following list of actions usually makes sense in more of them:

• Assign clear names to your sheets when you are done with them (Even if you are most likely to delete them later in the process). This simple step makes the whole Exploration process easier.

• Remove the unnecessary axes

• Remove unnecessary axis titles

• Shorten Month names (January --> J, February --> F, etc.)

• Remove unnecessary headers

• If needed benefit from filter action(s)

• Ask others to comment on the readability and understandability of your viz (If your audience gets what you are talking about, you've done good work)

• Keep in mind that people from West world read from left to write. Therefore, if it's possible put the most important charts (interaction pattern) in the upper left and the least important ones towards the lower right. We also put

Finally, the course introduces "The top five Data VIZ best practices" which are:

• Know your audience

• Know your data

• Use color purposefully

• Less is more; Simpler is usually better

• Get feedback early and often

Although this course very shortly focuses on the technical part of Tableau, it gives the audience a very great view of what is data visualization as a whole and how to work with the very basic features of Tableau. If a person wants to learn more about the Basic Features of Tableau in a short amount of time, I recommend them to follow the first course of this specialization and then skip to the third course and follow that one as well.

I will continue summarizing the course contents of this specialization.

Best of luck.

Cover credit: BING search engine

SQL Comments:

You can write comments in SQL files in two ways:

1. Single-Line comment:

You can create a single-line comment using two dashes:

2. Block comment:

You can create a block comment starting with /* and ending with */. This is especially helpful when you want to create a comment which has more than one line. for example, it is a good practice to write a block comment at the beginning of your SQL file and specify the file's creator, date of creation and description.

SELECT:

Use SELECT clause to columns you want to appear in the result table.

Alias:

To make your results more readable for yourself and others you can change the way columns' names are displayed in the result. For this purpose, you should use AS keyword, as shown in the following example:

As can be seen in the above example, for using an alias including more than one word we should put that multi-word alias between [ ] or ' '. However, we can bring a single-word alias immediately after AS without using any other special characters.

Sorting with ORDER BY:

We can sort results of our queries based on column contents. In order to do that, we use ORDER BY clause after FROM clause. Hint the following example which sorts the result table based on the contents of the LastName column.

The default mode of ORDER BY clause is ascending order. However, you can change it. You can use ASC for ascending or DESC for descending order. Use either of these two keywords after the column name stated in the ORDER BY clause. For example, to sort the result table based on LasrName descending order we should use the following query:

We can sort the results based on multiple columns. For example:

LIMIT:

You can limit the number of records shown in the result table using LIMIT Clause. In order to do that, you have to use "LIMIT number-of-rows" in your query. For example:

In the above example, number of rows in the result tables is limited to 10. Therefore, the result table will contain the information of the top 10 customers sorted with the specified order in the query.

P.S. The cover image is generated by DALL-E.