Encodings

UTF-8 by default

Ruby source and strings default to UTF-8, so most languages work with no setup. Every string knows its encoding and you can ask for it.

Problems appear only at the borders: files, network, and old tools that speak other encodings. Language background lives on the Ruby overview.

s = "Hello" puts s.encoding

Magic comments

A magic comment on the first line sets the source encoding for that file. It matters on old setups and for files saved in legacy encodings.

Modern editors save UTF-8 already, so you can often skip this line entirely.

# encoding: utf-8 puts "café".encoding

Convert with encode

encode returns the same text in a new encoding. Name the target encoding as a string and Ruby transcodes every character.

Conversions fail loudly on characters the target cannot hold, unless you pick a fallback strategy.

s = "café" wide = s.encode("UTF-16") puts wide.encoding

Scrub bad bytes

Bytes from broken files may be illegal in UTF-8 and crash later code. scrub replaces every bad byte with a placeholder and returns clean text.

Write one such file first, then read it back and scrub it. Pass your own replacement string when the default mark confuses readers.

File.binwrite("notes.txt", "caf\xE9") raw = File.read("notes.txt") safe = raw.scrub puts safe

Next: Time and date

Article author: Arthur Isaev