A line in the source file must be terminated by line feed character (U+000A, UTF-8: 0x0A, typical for Unix, Linux, Android, Mac OS X, BSD, and other operating systems), a carriage return (U+000D, UTF-8: 0x0D typical for Mac OS till Version 9 and other operating systems), or a carriage return immediately followed by a line feed (typical for Windows operating systems).

The character encoding of the source files can either be specified in the import settings or can be automatically detected by Redbex.

The character encoding detection built into Redbex is based on a port of the Universal Charset Detector of Mozilla. This charset encoding detection is able to detect the character encodings shown below

Detection of the character encoding is however in many cases non-deterministic, i.e. the character encoding of a file or a series of bytes can only be detected with a certain probability.

•UTF-8

•UTF-16 (BE and LE)

•UTF-32 (BE and LE)

•windows-1252 (mostly equivalent to iso8859-1)

•windows-1251 and ISO-8859-5 (cyrillic)

•windows-1253 and ISO-8859-7 (greek)

•windows-1255 (logical hebrew. Includes ISO-8859-8-I and most of x-mac-hebrew)

•ISO-8859-8 (visual hebrew)

•Big-5

•gb18030 (superset of gb2312)

•HZ-GB-2312

•Shift-JIS

•EUC-KR, EUC-JP, EUC-TW

•ISO-2022-JP, ISO-2022-KR, ISO-2022-CN

•KOI8-R

•x-mac-cyrillic

•IBM855 and IBM866

•X-ISO-10646-UCS-4-3412 and X-ISO-10646-UCS-4-2413 (unusual BOM)

•ASCII