enca

character encoding detection and conversion

TLDR

Detect encoding of file

$ enca [file.txt]

Detect with language hint

$ enca -L [czech] [file.txt]

Convert encoding

$ enca -x [UTF-8] [file.txt]

Detect and show confidence

$ enca -d [file.txt]

Process multiple files

$ enca -L [russian] [*.txt]

enca (Extremely Naive Charset Analyser) detects character encodings of text files using language-based statistical heuristics, and can convert files between encodings by piping through iconv. It is particularly strong on legacy 8-bit charsets used for Slavic and Central/Eastern European languages (ISO-8859-2/5, KOI8-R, CP1250/1251, Mazovia, T.61, ...), where simpler tools like file -i struggle.Detection works best with a language hint passed via -L; without it, enca falls back to a general profile and may return ambiguous matches. The output is one detected encoding per file by default, or extended information with -d or -v. Conversion is performed in place with -x ENCODING, which calls iconv under the hood; pair it with --cstocs for transliteration when the target charset lacks specific characters.

PARAMETERS

FILES

Files to analyze.

-L LANGUAGE

Hint language for detection.

-x ENCODING

Convert to specified encoding.

-d

Show detailed detection info.

-g, --guess

Output best guess only.

-i, --info

Show available encodings.

--help

Display help information.

INSTALL

sudo apt install enca

sudo dnf install enca

sudo pacman -S enca

sudo apk add enca

sudo zypper install enca

brew install enca

nix profile install nixpkgs#enca

CAVEATS

Detection is heuristic, not deterministic. Short files may be ambiguous. Works best with specific language hints. Some encodings indistinguishable.

HISTORY

enca was developed for handling the encoding diversity in Central/Eastern European computing, where many incompatible character sets were historically used for the same languages.