What data types does YAML support? How to use them correctly?

YAML supports multiple data types, and understanding these types is crucial for correctly writing and parsing YAML files.

YAML Data Type Classification

1. Scalar Types

Strings

• Plain strings: No quotes needed
• Quoted strings: Single or double quotes
• Multi-line strings: Use | or >

yaml
# Plain string
name: John Doe
# Single-quoted string (no escape for special characters)
message: 'Hello\nWorld'
# Double-quoted string (escape special characters)
greeting: "Hello\nWorld"
# Multi-line string (preserve line breaks)
description: |
  This is a
  multi-line
  string
# Multi-line string (fold line breaks)
summary: >
  This is a folded
  string that
  becomes one line

Numbers

• Integers: Decimal, octal (0o), hexadecimal (0x)
• Floats: Support scientific notation

yaml
integer: 42
octal: 0o52
hex: 0x2A
float: 3.14
scientific: 1.23e4
negative: -42

Booleans

• Multiple representations supported

yaml
true_value: true
false_value: false
yes: yes
no: no
on: on
off: off

Null

• Multiple representations

yaml
empty: null
none: ~
empty2:

2. Collection Types

Lists/Arrays

• Represented with hyphen -
• Inline notation supported

yaml
# Standard list
fruits:
  - apple
  - banana
  - orange

# Inline list
colors: [red, green, blue]

# Nested list
matrix:
  - [1, 2, 3]
  - [4, 5, 6]
  - [7, 8, 9]

Maps/Dictionaries

• Represented with key-value pairs
• Inline notation supported

yaml
# Standard map
person:
  name: Alice
  age: 30
  city: Beijing

# Inline map
config: {host: localhost, port: 8080}

# Nested map
server:
  database:
    host: db.example.com
    port: 5432
  cache:
    type: redis
    ttl: 3600

3. Complex Types

Mixed Types (Combining Lists and Maps)

yaml
users:
  - name: Bob
    age: 25
    skills: [Python, JavaScript]
  - name: Carol
    age: 28
    skills: [Java, Go]

Custom Types

yaml
# Use !! tag to specify type
timestamp: !!timestamp 2024-01-01T00:00:00Z
binary: !!binary |
  SGVsbG8gV29ybGQ=

Type Inference Rules

YAML parsers automatically infer types based on value format:

1. Numbers: Pure numeric sequences
2. Booleans: true/false, yes/no, on/off
3. Null: null, ~, empty string
4. Strings: All other cases

Type Conversion Techniques

Force Type Specification

yaml
# Use quotes to force string
port: "8080"  # String, not number

# Use !! tag to specify type
age: !!int "25"  # Force convert to integer

Special Character Handling

yaml
# Strings with special characters need quotes
path: "/usr/local/bin"
regex: "\\d+"

Common Errors

1. Type confusion: Expecting string but getting number
2. Boolean misinterpretation: yes/no interpreted as booleans
3. Date format: Some date formats auto-converted to timestamps
4. Improper quote usage: Causing escape characters to fail

Best Practices

1. Use quotes for values that explicitly need to be strings
2. Use explicit type annotations for configuration items
3. Avoid using yes/no as string values
4. Use YAML Schema for type validation
5. Maintain type consistency, don't mix different representations