From: "akr (Akira Tanaka)" <akr@...>
Date: 2012-04-05T13:43:21+09:00
Subject: [ruby-core:44141] [ruby-trunk - Bug #6258] String#succ has suprising behavior for "\u1036" (MYANMAR SIGN ANUSVARA), producing "\u1000" instead of "\u1037"


Issue #6258 has been updated by akr (Akira Tanaka).


"\u1036".succ is "\u1000\u1000", not a single character.

  % ruby -ve 'puts "\u1036".succ.dump'
  ruby 2.0.0dev (2012-03-16 trunk 35049) [x86_64-linux]
  "\u{1000}\u{1000}"

It is similar that "z".succ is "aa".

It is because U+1000 to U+1036 are alphabet characters and
U+0fff and U+1037 is not.

  % ruby -e '0xfff.upto(0x1037) {|c| p ["%x" % c, /[[:alpha:]]/ =~ c.chr("UTF-8")] }'
  ["fff", nil]
  ["1000", 0]
  ...
  ["1036", 0]
  ["1037", nil]

What I'm not sure is U+1036 is alphabet or not.
I think nurse-san or martin-sensei is appropriate for this matter.

----------------------------------------
Bug #6258: String#succ has suprising behavior for "\u1036" (MYANMAR SIGN ANUSVARA), producing "\u1000" instead of "\u1037"
https://bugs.ruby-lang.org/issues/6258#change-25664

Author: dbenhur (Devin Ben-Hur)
Status: Assigned
Priority: Normal
Assignee: akr (Akira Tanaka)
Category: M17N
Target version: 
ruby -v: ruby 1.9.3p125, ruby 1.9.2p180,


"\u1036".succ.ord.to_s(16) # => "1000"

Discovered when investigating StackOverflow question http://stackoverflow.com/questions/10020230/anomalous-behavior-while-comparing-a-unicode-character-to-a-unicode-character-range

Range#=== ultimately invokes String#upto which uses String#succ

("\u1036".."\u1037").to_a.map{|c| c.ord.to_s(16)}
=> ["1036"]  # expected ["1036","1037"]

Also once #succ! proceeds past U+1036 it continues to produce U+1000 indefinitely

irb(main):115:0> c = "\u1036"
=> "���"
irb(main):116:0> c.ord.to_s(16)
=> "1035"
irb(main):117:0> c.succ!.ord.to_s(16)
=> "1036"
irb(main):118:0> c.succ!.ord.to_s(16)
=> "1000"
irb(main):119:0> c.succ!.ord.to_s(16)
=> "1000"

But if one starts naturally at U+1000 #succ! increments as expected
irb(main):001:0> c = "\u1000"
=> "���"
irb(main):002:0> c.ord.to_s(16)
=> "1000"
irb(main):003:0> c.succ!.ord.to_s(16)
=> "1001"
irb(main):004:0> c.succ!.ord.to_s(16)
=> "1002"



-- 
http://bugs.ruby-lang.org/