Declares the source file text encoding. Requires a backend Prolog compiler supporting the chosen encoding. When used, this directive must be the first term in the source file in the first line. This directive is also supported in files included in a main file or in a dynamically created entity using include/1 directives.

The encoding used in a source file (and, in the case of a Unicode encoding, any BOM present) will be used for the intermediate Prolog file generated by the compiler. Logtalk uses the encoding names specified by IANA. In those cases where a preferred MIME name alias is specified, the alias is used instead. Examples includes 'US-ASCII', 'ISO-8859-1', 'ISO-8859-2', 'ISO-8859-15', 'UCS-2', 'UCS-2LE', 'UCS-2BE', 'UTF-8', 'UTF-16', 'UTF-16LE', 'UTF-16BE', 'UTF-32', 'UTF-32LE', 'UTF-32BE', 'Shift_JIS', and 'EUC-JP'. When writing portable code that cannot be expressed using ASCII, 'UTF-8' is the most commonly supported Unicode encoding.

The backend Prolog compiler adapter files define a table that translates between the Logtalk and Prolog specific atoms that represent each supported encoding. The encoding_directive read-only flag can be used to find if a backend supports this directive and how.

Template and modes



:- encoding('UTF-8').