Indentation

Like Python, Sharpy uses indentation instead of curly braces to demarcate blocks.

Indentation Rules

• Exactly 4 spaces per indentation level (enforced)
• Tabs are not allowed for indentation
• Mixed spaces and tabs cause a lexical error
• Indentation must be consistent within a file

Lexer Edge Cases

The following rules govern how the lexer handles indentation in special cases:

Scenario	Behavior
Blank lines	Do not affect indentation tracking. May have any amount of whitespace or none.
Comment-only lines	Do not require indentation matching. A # comment at column 0 inside an indented block is valid.
Trailing whitespace	Ignored. Does not affect indentation calculations.
Tab characters	Lexer error: "Tab characters are not permitted for indentation. Use 4 spaces per level."
Inconsistent indentation	Lexer error: "Indentation must be a multiple of 4 spaces. Found N spaces."

def example():
    x = 1           # 4 spaces - valid
# Comment at column 0 is OK
                    # Blank line above - OK

    if x > 0:
        y = 2       # 8 spaces - valid
    z = 3       # Tab character - ERROR

        w = 4       # 10 spaces - ERROR: not multiple of 4

Indentation Stack Behavior:

The lexer maintains a stack of indentation levels:

1. Stack starts with [0] (column 0)
2. When indentation increases by exactly 4 spaces, push new level and emit INDENT
3. When indentation decreases, pop levels and emit DEDENT for each popped level
4. At EOF, emit DEDENT for each level remaining on stack (except 0)

def foo():          # Stack: [0] → [0, 4], emit INDENT
    if True:        # Stack: [0, 4] → [0, 4, 8], emit INDENT
        x = 1
    y = 2           # Stack: [0, 4, 8] → [0, 4], emit DEDENT
z = 3               # Stack: [0, 4] → [0], emit DEDENT
# EOF               # Stack: [0], nothing more to emit

Implementation - 🔄 Lowered - The lexer tracks indentation levels via an indentation stack, emitting INDENT/DEDENT tokens. These are converted to C# braces { } during code generation.