Modifier

Partager via

Character and data encoding

  • Article

Computers have no intrinsic knowledge of characters used in writing systems. To represent and store the characters used in writing systems, computers use character sets that are mappings of characters to numerical values. The process of assigning characters to numerical values is called character encoding.

While there are many character encoding standards, one of the first in common use was ASCII (American Standard Code for Information Interchange). The original ASCII standard used a 7-bit character encoding system, representing 128 unique characters. Most modern computers work with and store binary numbers in groups of 8 bits. Early Windows versions used an 8-bit character encoding system and these versions of Windows support multiple 8-bit code pages, enabling Windows releases in several languages.

To maintain compatibility with ASCII and to support a limited compatibility between 8-bit character sets, the first 127 characters of many code pages represent the ASCII characters. It's usually the upper 128 code points (values 128-255) where 8-bit code pages differ considerably. 8-bit character sets are commonly termed single-byte character sets (SBCS). SBCS and their corresponding code pages allowed support for a broad range of scripts and languages, including Cyrillic, Greek, Arabic, and Vietnamese. Here are the characters corresponding to E0-EF (224-239) for several code pages:

Code page Script/Language/Region E0 E1 E2 E3 E4 E5 E6 E7 E8 E9 EA EB EC ED EE EF
Windows-1250 Central European ŕ á â ă ä ĺ ć ç č é ę ë ě í î ď
Windows-1251 Cyrillic а б в г д е ж з и й к л м н о п
Windows-1252 Western European à á â ã ä å æ ç è é ê ë ì í î ï
Windows-1253 Greek ΰ α β γ δ ε ζ η θ ι κ λ μ ν ξ ο
Windows-1254 Turkish à á â ã ä å æ ç è é ê ë ì í î ï
Windows-1255 Hebrew א ב ג ד ה ו ז ח ט י ך כ ל ם מ ן
Windows-1256 Arabic à ل â م ن ه و ç è é ê ë ى ي î ï
Windows-1257 Baltic ą į ā ć ä å ę ē č é ź ė ģ ķ ī ļ
Windows-1258 Vietnamese à á â ă ä å æ ç è é ê ë ◌́ í î ï

Writing systems for languages such as Chinese, Japanese, and Korean use more than 256 characters, so a different scheme needed to be developed to overcome the 256-character limit of SBCS. Double Byte Character Sets (DBCS) and, more generally, Multi-Byte Character Sets (MBCS) were developed to extend the SBCS design. In a DBCS or an MBCS, many characters are represented by a pair of bytes (thus double-byte) or a sequence of 3 or more bytes. For programming awareness, a set of points are set aside to represent the first byte of the set and aren't valued unless they're immediately followed by a defined second (or third) byte in a specific range. When working with MBCS, you have to write code that treats these varying-length sequences of bytes as one character. This scheme still doesn't necessarily allow for the combining of Japanese and Chinese in the same data stream, because depending on the code page, the same code points could represent different characters for the different languages.

To allow for the storage of characters that support different writing systems in the same data stream, Unicode was created. Unicode can represent over a million characters. Using Unicode allows for easier creation of world-ready code because you don’t have to worry about which code page you're addressing, or whether you have to group character points to represent one character.

While Unicode-encoded data streams are common, you shouldn't assume that all data streams are encoded using the Unicode standard. When you're working with .NET, Java, and similar frameworks and languages, it's best practice to:

  1. Identify the encoding of incoming data
  2. Transform incoming data from any non-Unicode encoding to Unicode
  3. Process the data as Unicode
  4. Convert back to the original encoding as necessary on output