Extended Numeric Literals

# Binary literals
binary = 0b1010        # 10 in decimal
flags = 0b1111_0000

# Hexadecimal literals
hex_value = 0xFF       # 255 in decimal
color = 0x001122

# Octal literals
permissions = 0o755    # 493 in decimal

# Scientific notation (e or E)
avogadro = 6.022e23
planck = 6.626e-34
large = 1E20

Output format. Printing a float uses Python's layout, not .NET's: lowercase e, an explicit exponent sign, and at least two exponent digits (1e+20, 1e-05). Writing a value as 0.d1…dn × 10^decpt, a float prints positionally when -4 < decpt <= 16 and exponentially otherwise — CPython's rule, so 1e16 prints as 1e+16 rather than 10000000000000000.0 (#1204). The digits themselves are .NET's shortest round-trip form, which already matches CPython's.

float32 uses the same renderer with its own threshold, -4 < decpt <= 9: CPython has no float32, so there is no parity answer to copy, and 9 is the point past which positional layout would print more digits than a single carries.

Implementation: - Binary/Hex: ✅ Native - Direct C# support (C# 7.0+) - Octal: 🔄 Lowered - Converted to decimal at compile time - Scientific: ✅ Native - Direct C# support