`split': invalid byte sequence in US-ASCII (ArgumentError) on ripper.rb

Metadata

node -v
  v14.5.0

ruby -v
  ruby 2.7.1p83 (2020-03ruby 2.7.1p83 (2020-03-31 revision a0c7c23c9c) [x86_64-darwin19]

uname -a
  Darwin mbp 19.6.0 Darwin Kernel Version 19.6.0: Sun Jul  5 00:43:10 PDT 2020; root:xnu-6153.141.1~9/RELEASE_X86_64 x86_64

cat package.json
  {
    "devDependencies": {
      "@prettier/plugin-ruby": "^0.19.0",
      "prettier": "^2.0.5"
    }
  }

Visual Studio Code
  Version: 1.47.0
  Commit: d5e9aa0227e057a60c82568bf31c04730dc15dcd
  Date: 2020-07-09T08:01:54.115Z (1 wk ago)
  Electron: 7.3.2
  Chrome: 78.0.3904.130
  Node.js: 12.8.1
  V8: 7.8.279.23-electron.0
  OS: Darwin x64 19.6.0

https://marketplace.visualstudio.com/items?itemName=esbenp.prettier-vscode
  v5.1.3

Input

# あ 

      d=[30644250780,9003106878,
  30636278846,66641217692,4501790980,
671_24_603036,131_61973916,66_606629_920,
 30642677916,30643069058];a,s=[],$*[0]
    s.each_byte{|b|a<<("%036b"%d[b.
       chr.to_i]).scan(/\d{6}/)}
        a.transpose.each{ |a|
          a.join.each_byte{\
           |i|print i==49?\
             ($*[1]||"#")\
               :32.chr}
                 puts
                  }

class RipperJS < Ripper
  attr_reader :source, :lines, :__end__

  def initialize(source, *args)
    super(source, *args)

    @source = source # <= source.encoding
    @lines = source.split("\n")
    @__end__ = nil
  end

It seems if ran through VSCode, source.encoding would be recognized as “US-ASCII.” And I get an error like below.

ripper.rb:25:in `split': invalid byte sequence in US-ASCII (ArgumentError)

When I run bundle exec rbprettier --write on the same file, source.encoding is “UTF-8”, and the format is done correctly.

[kddnewton]

@stmichael that’s the ticket! Thank you for finding that. Because of that investigation, I was finally able to track it down to:

LC_ALL="en_US.US-ASCII" ruby -e 'p "ä"'

That will reliably fail. I will add something that will detect the default encoding and handle it appropriately.

[stmichael]

I’m having the same issue but in a different setup. I don’t use any editor integration, just plain command line. My setup:

OS: Debian Language: JS (NodeJS v12.18.3) Plugin-Ruby: 0.19.0

Running yarn prettier -l '**/*.rb' in my project prints a whole bunch of errors identical to the ones @github0013 mentioned.

Error: /app/node_modules/@prettier/plugin-ruby/src/ripper.rb:25:in `split': invalid byte sequence in US-ASCII (ArgumentError)

I can’t count them all, there’s too many.

To track the issue I came up with a minimal example. Create a file test.rb with content

'ä'

That’s a valid Ruby program although it doesn’t make any sense. Running yarn prettier -l test.rb will produce the very same issue.

I’m not an expert on character encodings. In my understanding since Ruby 2.0 the default encoding is UTF-8. Then why does this plugin read the files in US-ASCII?