How do I rescue "invalid Unicode codepoint" in Ruby?

Question

s = "\u{d800}" produces an error, test.rb:1: invalid Unicode codepoint, as expected. However,

begin
  s = "\u{d800}"
rescue Exception
  puts "oh no"
end

does the same thing, producing the error rather than an output of oh no. Is there a built-in way to rescue this, or would I have to hardcode valid ranges of Unicode codepoints to check validity?

I can confirm `SyntaxError` in 2.5.1 unfortunately a `SyntaxError` cannot be rescued in that fashion the only way you could rescue this error would be to "eval" that code e.g. `begin; eval('"\u{d800}"'); rescue SyntaxError; 'oh no';end;` but I would not recommend this. Can you explain a bit further how this is happening? — engineersmnky, Jun 25 '18 at 17:10
Unrelated, but never rescue `Exception`. Always rescue something more specific https://stackoverflow.com/questions/10048173/why-is-it-bad-style-to-rescue-exception-e-in-ruby — Max, Jun 25 '18 at 17:49

Max · Accepted Answer · 2018-06-25T18:09:59.033

1

You can test if a number is a valid unicode codepoint with chr

0xd800.chr('UTF-8')
# RangeError (invalid codepoint 0xd800 in UTF-8)

edited Jun 25 '18 at 18:09

answered Jun 25 '18 at 17:50

Max

21,123
5
49
71

2

Alternatively, without raising an exception: `[0xd800].pack('U').valid_encoding?` – cremno Jun 26 '18 at 04:43

How do I rescue "invalid Unicode codepoint" in Ruby?

1 Answers1