Graphemeclusters

A grapheme cluster is the user-perceived character unit in text. In Unicode, grapheme clusters are defined as extended grapheme clusters (EGC) by the Unicode Standard Annex 29, which specifies how a sequence of one or more code points is treated as a single display character.

A cluster typically consists of a base character followed by one or more combining marks, but it can also include sequences that render as a single symbol. Examples include emoji sequences joined with a zero-width joiner to form a single pictographic character, or a base letter with combining accents. Emoji can also involve modifiers such as skin tone and sequences that render as a single composite symbol, including flags formed by regional indicator symbols.

Grapheme cluster boundaries are determined by the rules in UAX #29. Correct segmentation is essential for user interfaces and text processing because cursor movement, selection, counting, searching, and rendering should operate on clusters rather than individual code points. Normalization can alter code point sequences yet does not change the perceptual unit; therefore, clustering rules are applied after or across normalization in many implementations.

Examples help illustrate the concept. The character "e" followed by a combining acute accent U+0301 forms the same visible character as the precomposed é (U+00E9) in many cases, but they are different code point sequences and belong to the same grapheme cluster when displayed. A sequence like 👩🏻‍💻 demonstrates how base emoji, modifiers, and a zero-width joiner can produce a single cluster that appears as one symbol.

Applications and libraries implement grapheme clustering to support accurate text rendering, editing, and processing across scripts and platforms.