How to have a column set to a custom type?

I am new to using SQL and relational databases. I am creating a basic quiz web app where people can create their own quizzes and have all the data stored in a cloud SQL database. All of this works fine but I have a problem now. Here are two types of data:

_question {
  question_id: number;
  title: string;
  prompts: string[];
  correct_answer: number; 
}

_quiz {
  quiz_id: number;
  title: string;
  questions: _question[];
  created_on_epoch: number; 
}

This is all using TypeScript to define the data. I use this as a blueprint and convert it into creating a SQL table with the types above. For example, I have a table in the database created with this command CREATE TABLE quizs(quiz_id SERIAL PRIMARY KEY, title VARCHAR(150), questions {DONT KNOW WHAT TO PUT HERE}, created_on_epoch bigint);

I want to set the column questions in the quizs table to be equal to an array of type _question. Is there any way to do this in SQL or is there a better approach?

It is possible to add a column of type string array to a table in PostgreSQL. To add a column called questions of type string array to a table called my_table, the DDL would look like this:

ALTER TABLE my_table ADD COLUMN questions text[];

Note that the [] after the data type specifies that this is an array data type. The text data type is used to store strings in PostgreSQL

Once the column is added, you can insert values into the array column using the ARRAY constructor. For example:

INSERT INTO my_table (questions) VALUES (ARRAY['question1', 'question2', 'question3']);

To update the quizs table from the questions table, you can use a subquery and the ARRAY constructor. e.g.

UPDATE quizs
SET questions = ARRAY(
  SELECT ROW(question, answer)::question
  FROM questions
  WHERE quiz_id = quizs.id
);

To determine whether an array column string or numeric type, use the pg_typeof() function.

SELECT pg_typeof(questions) FROM my_table;

This will return the data type of the questions column, which will be either text[] for a string array or integer[] for a numeric array.

BUT

I would encourage you to consider a more conventional table structure, using a "normalized" approach, or, perhaps consider use of JSON instead of arrays. Also note that many Relational DBMS systems do not support array data types, but many more will have some JSON support. (& This decision might be influenced by what tech you intend to use as the front-end for your application.)

What you may save now through the simple table structure seen here will probably cost you more later via query complexities.