Numeric and decimal precision in PostgreSQL - Cheat Sheet & Quick Revision

Recall & Review

beginner

What does the NUMERIC(p, s) data type represent in PostgreSQL?

It represents a number with p total digits (precision) and s digits after the decimal point (scale).

beginner

What happens if you insert a number with more decimal places than the scale defined in NUMERIC(p, s)?

PostgreSQL rounds the number to fit the defined scale (number of decimal places).

beginner

How does DECIMAL differ from NUMERIC in PostgreSQL?

In PostgreSQL, DECIMAL is just an alias for NUMERIC. Both behave the same way.

intermediate

What is the default precision and scale if you declare a column as NUMERIC without parameters?

It allows any number of digits with no fixed precision or scale, limited only by the system's maximum storage.

intermediate

Why use NUMERIC or DECIMAL instead of FLOAT for money values?

Because NUMERIC stores exact decimal values without rounding errors, which is important for money calculations.

What does the scale in NUMERIC(p, s) specify?

A Number of digits before the decimal point

B Total number of digits

C Maximum value allowed

D Number of digits after the decimal point

The scale s specifies how many digits appear after the decimal point.

If you define a column as NUMERIC(5, 2), what is the largest number you can store?

A 999.99

B 99999

C 9999.9

D 99.999

With precision 5 and scale 2, you have 3 digits before and 2 digits after the decimal point, so max is 999.99.

What happens if you insert 12.345 into a NUMERIC(5, 2) column?

A It stores 12.345 exactly

B It stores 12.34 without rounding

C It stores 12.35 after rounding

D It causes an error

The value is rounded to 2 decimal places, so 12.345 becomes 12.35.

Which data type is best for storing exact monetary values?

A INTEGER

B NUMERIC

C FLOAT

D TEXT

NUMERIC stores exact decimal values, avoiding rounding errors common with FLOAT.

What is the difference between DECIMAL and NUMERIC in PostgreSQL?

A No difference, they are synonyms

B DECIMAL is faster

C NUMERIC has fixed precision, DECIMAL does not

D DECIMAL stores integers only

In PostgreSQL, DECIMAL and NUMERIC are the same data type.

Explain how precision and scale work in the NUMERIC(p, s) data type.

Think about total digits and decimal places separately.

Why is NUMERIC preferred over FLOAT for financial data?

Consider the importance of exactness in money.

Practice

1. What does the scale represent in a NUMERIC(10, 3) data type in PostgreSQL?

easy

A. The number of digits after the decimal point

B. The total number of digits allowed

C. The number of digits before the decimal point

D. The maximum value allowed

Solution

1. Step 1: Understand NUMERIC precision and scale

   In NUMERIC(precision, scale), precision is total digits, scale is digits after decimal.

2. Step 2: Apply to NUMERIC(10, 3)

   Here, scale = 3 means 3 digits after decimal point.

3. Final Answer:

   The number of digits after the decimal point -> Option A

4. Quick Check:

   Scale = digits after decimal [OK]

Hint: Scale is digits after decimal point in NUMERIC [OK]

Common Mistakes:

• Confusing scale with precision
• Thinking scale is digits before decimal
• Assuming scale limits total digits

2. Which of the following is the correct syntax to define a column named price with 7 digits total and 2 digits after the decimal in PostgreSQL?

easy

A. price DECIMAL(7)

B. price NUMERIC(2,7)

C. price DECIMAL(2,7)

D. price NUMERIC(7,2)

Solution

1. Step 1: Recall NUMERIC/DECIMAL syntax

   Syntax is NUMERIC(precision, scale) where precision >= scale.

2. Step 2: Match precision=7 and scale=2

   price NUMERIC(7,2) uses NUMERIC(7,2) which means 7 total digits, 2 after decimal.

3. Final Answer:

   price NUMERIC(7,2) -> Option D

4. Quick Check:

   NUMERIC(precision, scale) = NUMERIC(7,2) [OK]

Hint: Precision first, scale second in NUMERIC/DECIMAL [OK]

Common Mistakes:

• Swapping precision and scale
• Using scale larger than precision
• Omitting scale when needed

3. Given the table products(price NUMERIC(5,2)), what will be the stored value after inserting INSERT INTO products(price) VALUES (123.4567);?

medium

A. 123.45

B. 123.46

C. 123.4567

D. Error: value too precise

Solution

1. Step 1: Understand NUMERIC(5,2) limits

   Precision 5 means total 5 digits, scale 2 means 2 digits after decimal.

2. Step 2: Round input 123.4567 to 2 decimal places

   123.4567 rounds to 123.46 to fit scale 2.

3. Final Answer:

   123.46 -> Option B

4. Quick Check:

   Rounded to 2 decimals = 123.46 [OK]

Hint: NUMERIC rounds to scale digits on insert [OK]

Common Mistakes:

• Expecting truncation instead of rounding
• Assuming error on extra decimals
• Ignoring precision limits

4. You defined a column as amount NUMERIC(4,2). You try to insert 999.99 but get an error. What is the problem?

medium

A. NUMERIC cannot store decimal values

B. The scale 2 is too small for the value

C. The value exceeds the total precision of 4 digits

D. Syntax error in column definition

Solution

1. Step 1: Calculate digits in 999.99

   999.99 has 5 digits total (3 before decimal + 2 after).

2. Step 2: Compare with NUMERIC(4,2)

   NUMERIC(4,2) allows only 4 digits total, so 999.99 is too large.

3. Final Answer:

   The value exceeds the total precision of 4 digits -> Option C

4. Quick Check:

   Value digits > precision causes error [OK]

Hint: Total digits must fit precision, not just decimals [OK]

Common Mistakes:

• Thinking scale limits integer part
• Believing NUMERIC disallows decimals
• Ignoring precision limits

5. You want to store monetary values up to 999999.999 with exact precision in PostgreSQL. Which column definition is best to avoid rounding errors and allow this range?

hard

A. amount NUMERIC(9,3)

B. amount NUMERIC(6,3)

C. amount DECIMAL(9,2)

D. amount FLOAT(9)

Solution

1. Step 1: Determine needed precision and scale

   Value 999999.999 has 6 digits before decimal, 3 after, total 9 digits.

2. Step 2: Choose NUMERIC with precision 9 and scale 3

   NUMERIC(9,3) stores exactly 9 digits total with 3 decimals, matching requirement.

3. Step 3: Avoid FLOAT for exact money values

   FLOAT is approximate, may cause rounding errors.

4. Final Answer:

   amount NUMERIC(9,3) -> Option A

5. Quick Check:

   Exact money storage needs NUMERIC(9,3) [OK]

Hint: Use NUMERIC with precision = digits total, scale = decimals [OK]

Common Mistakes:

• Using FLOAT for money causing rounding errors
• Choosing too small precision or scale
• Confusing DECIMAL scale with precision