recode

TLDR

Convert a file from Latin-1 to UTF-8

$ recode latin1..utf8 [file]

Convert from UTF-8 to Latin-1

$ recode utf8..latin1 [file]

Convert HTML entities to UTF-8

$ recode html..utf8 [file]

Convert UTF-8 to HTML entities

$ recode utf8..html [file]

Convert file and save to new file (preserving original)

$ recode latin1..utf8 < [input] > [output]

List all available charsets

$ recode -l

Convert with verbose output

$ recode -v latin1..utf8 [file]

SYNOPSIS

recode [options] request [file...]

recode converts text files between different character encodings and can also translate escape sequences. It supports over 300 charsets including ASCII, Latin series, UTF-8, UTF-16, and various national encodings.
The conversion request uses the format source..destination where both parts specify charsets. Characters that cannot be represented in the destination charset are handled according to the mode (strict, force, or default lossy conversion).
Beyond charset conversion, recode handles surface transformations like CRLF line endings, Base64, quoted-printable, HTML/XML entities, and various escape sequences.
Multiple files can be processed, with each file converted in-place. Use shell redirection to preserve originals.

PARAMETERS

request

Conversion specification in format: source..destination

-l, --list

List all known charsets and surfaces

-f, --force

Force conversion even if information loss may occur

-s, --strict

Strict mode; abort on untranslatable characters

-d, --diacritics-only

Convert only diacritics, leave other characters

-c, --colstrings

Enable column-oriented strings mode

-g, --graphics

Show graphical conversion summary

-v, --verbose

Verbose output showing conversion details

-q, --quiet

Suppress warnings and progress messages

-h, --help

Display help message

COMMON CHARSETS

utf8, utf-8: Unicode UTF-8
latin1, iso-8859-1: Western European
ascii, us-ascii: 7-bit ASCII
html: HTML numeric entities
utf16: Unicode UTF-16
cp1252: Windows Western European

CAVEATS

In-place conversion modifies files directly. Always backup files or use redirection to preserve originals when uncertain about the conversion.
Conversions between incompatible charsets may lose information. Use -s (strict) to detect and abort on problematic characters, or -f (force) to proceed anyway.
HTML entity conversion (html charset) handles numeric entities. Named entities like & require additional processing.

HISTORY

Recode was originally written by Fran├ºois Pinard starting in 1990 and was one of the first comprehensive charset conversion tools for Unix. It became a GNU project and remains useful for batch processing text encoding conversions.