Bug #8151: Duplicate character class warning

• Status: Closed
• Assignee: naruse (Yui NARUSE)
• Target version: -
• ruby -v: ruby 1.9.3p374 (2013-01-15 revision 38858) [x86_64-darwin12.2.1]

[tenderlovemaking (Aaron Patterson)]

I get a duplicate character class warning, but I think it's a bug. Here is the example code:

def embed exp, depth
  return exp if depth == 0
  embed(/#{exp}/, depth - 1)
end

3.times { |i|
  puts "DEPTH #{i + 1}"
  embed(/[a-z\u{7b}-\u{7d}]/, i + 1)
}

At depth = 1, there is no warning, but greater than 1, I get a duplicate character class warning. I don't think the character class overlaps, so there should never be a warning.

[mame (Yusuke Endoh)]

• Status changed from Open to Assigned
• Assignee set to akr (Akira Tanaka)

Interestingly, this seems to be a bug of Regexp#to_s, not a regexp creation.

re = /(?:[\u{33}])/
p re #=> /(?:[\u{33}])/

puts re.to_s
#=> warning: character class has duplicated range: /[\u{33}]/
#=> (?-mix:[\u{33}])

As I recall, akr created round trip to_s.

--
Yusuke Endoh [e-mail]

[akr (Akira Tanaka)]

• Assignee changed from akr (Akira Tanaka) to naruse (Yui NARUSE)

The warning is proposed by [ruby-core:24593] [Feature #1831] (2009-07-29) and committed by naruse.

It is seven years later than my Regexp#to_s [ruby-dev:16951] (2002-04-18)

It seems this issue is better to assign naruse.

[naruse (Yui NARUSE)]

• Status changed from Assigned to Closed
• % Done changed from 0 to 100

This issue was solved with changeset r40063.
Aaron, thank you for reporting this issue.
Your contribution to Ruby is greatly appreciated.
May Ruby be with you.

---

• re.c (rb_reg_to_s): suppress duplicated charclass warning.
  Regexp#to_s suppress extra its whole regexp options by calling
  onig_new with its source, but it doesn't call rb_reg_preprocess.
  Therefore its Unicode escapes (\u{XXXX}) are given as is,
  and it may cause duplicated charclass warning for example
  "[\u{33}]" (3 is duplicated) or "[\u{a}\u{b}]" (u is duplicated).
  [ruby-core:53649] [Bug #8151]