LIBGRAPHEME(7) - Miscellaneous Information Manual

NAME

libgrapheme - unicode string library

SYNOPSIS

#include <grapheme.h>

DESCRIPTION

The libgrapheme library provides functions to properly handle Unicode strings according to the Unicode specification in regard to character, word, sentence and line segmentation and case detection and conversion.

Unicode strings are made up of user-perceived characters (so-called "grapheme clusters", see MOTIVATION) that are composed of one or more Unicode codepoints, which in turn are encoded in one or more bytes in an encoding like UTF-8.

There is a widespread misconception that it was enough to simply determine codepoints in a string and treat them as user-perceived characters to be Unicode compliant. While this may work in some cases, this assumption quickly breaks, especially for non-Western languages and decomposed Unicode strings where user-perceived characters are usually represented using multiple codepoints.

Despite this complicated multilevel structure of Unicode strings, libgrapheme provides methods to work with them at the byte-level (i.e. UTF-8 'char' arrays) while also offering codepoint-level methods. Additionally, it is a "freestanding" library (see ISO/IEC 9899:1999 section 4.6) and thus does not depend on a standard library. This makes it easy to use in bare metal environments.

Every documented function's manual page provides a self-contained example illustrating the possible usage.

STANDARDS

libgrapheme is compliant with the Unicode 15.0.0 specification.

MOTIVATION

The idea behind every character encoding scheme like ASCII or Unicode is to express abstract characters (which can be thought of as shapes making up a written language). ASCII for instance, which comprises the range 0 to 127, assigns the number 65 (0x41) to the abstract character 'A'. This number is called a "codepoint", and all codepoints of an encoding make up its so-called "code space".

Unicode's code space is much larger, ranging from 0 to 0x10FFFF, but its first 128 codepoints are identical to ASCII's. The additional code points are needed as Unicode's goal is to express all writing systems of the world. To give an example, the abstract character 'Ä' is not expressable in ASCII, given no ASCII codepoint has been assigned to it. It can be expressed in Unicode, though, with the codepoint 196 (0xC4).

One may assume that this process is straightfoward, but as more and more codepoints were assigned to abstract characters, the Unicode Consortium (that defines the Unicode standard) was facing a problem: Many (mostly non-European) languages have such a large amount of abstract characters that it would exhaust the available Unicode code space if one tried to assign a codepoint to each abstract character. The solution to that problem is best introduced with an example: Consider the abstract character 'Ǟ', which is 'A' with an umlaut and a macron added to it. In this sense, one can consider 'Ǟ' as a two-fold modification (namely "add umlaut" and "add macron") of the "base character" 'A'.

The Unicode Consortium adapted this idea by assigning codepoints to modifications. For example, the codepoint 0x308 represents adding an umlaut and 0x304 represents adding a macron, and thus, the codepoint sequence "0x41 0x308 0x304", namely the base character 'A' followed by the umlaut and macron modifiers, represents the abstract character 'Ǟ'. As a side-note, the single codepoint 0x1DE was also assigned to 'Ǟ', which is a good example for the fact that there can be multiple representations of a single abstract character in Unicode.

Expressing a single abstract character with multiple codepoints solved the code space exhaustion-problem, and the concept has been greatly expanded since its first introduction (emojis, joiners, etc.). A sequence (which can also have the length 1) of codepoints that belong together this way and represents an abstract character is called a "grapheme cluster".

In many applications it is necessary to count the number of user-perceived characters, i.e. grapheme clusters, in a string. A good example for this is a terminal text editor, which needs to properly align characters on a grid. This is pretty simple with ASCII-strings, where you just count the number of bytes (as each byte is a codepoint and each codepoint is a grapheme cluster). With Unicode-strings, it is a common mistake to simply adapt the ASCII-approach and count the number of code points. This is wrong, as, for example, the sequence "0x41 0x308 0x304", while made up of 3 codepoints, is a single grapheme cluster and represents the user-perceived character 'Ǟ'.

The proper way to segment a string into user-perceived characters is to segment it into its grapheme clusters by applying the Unicode grapheme cluster breaking algorithm (UAX #29). It is based on a complex ruleset and lookup-tables and determines if a grapheme cluster ends or is continued between two codepoints. Libraries like ICU and libunistring, which also offer this functionality, are often bloated, not correct, difficult to use or not reasonably statically linkable.

Analogously, the standard provides algorithms to separate strings by words, sentences and lines, convert cases and compare strings. The motivation behind libgrapheme is to make unicode handling suck less and abide by the UNIX philosophy.

AUTHORS

Laslo Hunhold (dev@frign.de)