码位

What is a Code Point?

A code point is the fundamental unit of the Unicode standard: a unique integer assigned to a single character, symbol, or abstract entity. Code points are written in the format U+XXXX where the Xs are hexadecimal digits — for example, U+0041 for the Latin capital letter A, or U+1F600 for the grinning face emoji 😀.

The Unicode code space spans from U+0000 to U+10FFFF, providing 1,114,112 possible positions. Not every position is occupied — as of Unicode 16.0, approximately 154,998 are assigned. The rest are either unassigned, reserved, or permanently set aside as noncharacters.

Anatomy of a Code Point Notation

U+1F600
│ │────┘
│  └── Hexadecimal value (6 digits for supplementary, 4 for BMP)
└───── "U+" prefix (Unicode notation)

Decimal equivalent: 128,512
Binary: 1 1111 0110 0000 0000

The "U+" prefix is a notation convention; it is not part of the value itself. When code points appear in source code or data, they use encoding-specific escape sequences instead:

Code Points vs Characters

A code point is not always identical to what a user perceives as a "character" (called a grapheme cluster). Consider:

é  can be represented as:
  U+00E9  (LATIN SMALL LETTER E WITH ACUTE — single code point)
  U+0065 + U+0301  (e + combining acute accent — two code points)

🏳️‍🌈  (rainbow flag) is:
  U+1F3F3 + U+FE0F + U+200D + U+1F308  (four code points, one visible character)

This distinction matters for string length calculations, cursor movement, and text editing.

Ranges and Planes

Code points are organized into 17 planes of 65,536 values each:

• Plane 0 (U+0000–U+FFFF): Basic Multilingual Plane — most everyday characters
• Plane 1 (U+10000–U+1FFFF): Supplementary Multilingual Plane — historic scripts, emoji
• Plane 2 (U+20000–U+2FFFF): Supplementary Ideographic Plane — CJK extension ideographs
• Planes 3–13: Mostly unassigned
• Plane 14 (U+E0000–U+EFFFF): Tags (language tags, variation selectors)
• Planes 15–16 (U+F0000–U+10FFFF): Private Use Areas

Common Pitfalls

String length confusion: In many languages, length returns the number of code units (bytes or 16-bit words), not code points, and neither matches the number of visible characters.

s = "😀"
len(s)          # Python 3: 1 (correct — one code point)
len(s.encode()) # 4 (bytes in UTF-8)

"😀".length       // 2 (UTF-16 surrogate pair!)
[..."😀"].length  // 1 (iterating by code point)

BMP assumption: Legacy code that assumes all characters fit in 16 bits (a BMP-only assumption) breaks on emoji, historic scripts, and rare CJK extensions.

Quick Facts