# Code points: what does length count?

[Concepts explained](README.md)

A **code point** is a numbered position in the Unicode catalogue. For example, `A` has the label `U+0041`. The `U+` notation writes a Unicode number using hexadecimal, a number system with sixteen digits. You do not need to calculate these labels to use text in Pliro.

## One shape can have several parts

Some visible characters combine several code points. An accented letter can be a ready-made letter, or a plain letter followed by a combining accent. A **grapheme cluster** is a group that behaves like one visible character. Emoji can form such groups too. See the [Unicode explanation of combining characters](https://www.unicode.org/faq/char_combmark.html).

Pliro's `length`, text indexing, and text loops work with code points, not grapheme clusters or UTF-8 bytes.

```pliro
# language: en
say length("🐢")
say length("👍🏽")
```

The answers are `1` and `2`. The second example contains a thumbs-up code point and a skin-tone modifier. It may appear as one picture on your screen.

## Why this matters

If you limit a player name to ten code points, that is not always ten visible letters. Cutting text at an arbitrary code-point position can separate an accent or an emoji component. Pliro does not currently provide a built-in grapheme-cluster counter.

Keep beginner examples simple, but do not assume that everybody's name has one code point per visible letter. For a challenge, compare `length("A")` with `length("AA")`, then explain why byte size answers a different question.

[UTF-8](utf-8.md) · [Indexing](../syntax/indexing.md)