UTF-8 encoding not work with gets method in Ruby

Setting the file encoding using the "magic" comment on top of the file only specifies the encoding of your source code in the file (that is: the encoding of string literals created directly from the parser in your code).

Ruby knows two other default encodings:

• the external encoding - this specifies the default encoding of data read from external sources (such as the console, opened files, network sockets, ...)
• the internal encoding - data read from external sources will be transformed into the default internal encoding after reading to ensure you can use compatible encodings everywhere (this is not used by default, the external encoding is thus preserved).

In your case, you have not set the external encoding. On Windows and with Ruby before version 3.0, Ruby assumes the local console encoding of your Windows installation here (such as cp850 in Western Europe).

When Ruby reads your String, it assumes it to be in cp850 encoding (or whatever your default encoding is) while you likely provide utf-8 encoded data. As spoon as you start to operate on this incorrectly encoded data, you will get errors similar to the one you have seen there.

Thus, to be able to correctly read data you need to either provide it with an encoding matching your shell encoding, or you need to tell Ruby which encoding it should assume there.

If you are providing UTF-8 encoded data, you can set the expected encoding using the -E switch when invoking ruby, e.g.:

ruby -E utf-8 your_program.rb

You can also set this in an environment variable of your Windows shell using

set RUBYOPT=-Eutf-8

In Ruby 3.0, the default external encoding on Windows was changed so that it now defaults to UTF-8 on Windows, similar to other platforms. See https://bugs.ruby-lang.org/issues/16604 for details.