Number to UTF-conversion question

Rails
1

All, Sorry for cross-posting but I am stuck on this thing for a quite
some time:

I have a variable x = 1046

How can I convert into UTF-8 character? x.chr does not work for it.

Basically I need to put x in a string as UTF-8 character to display on a
page.

Regards,

  • newB
2

On Sep 29, 2008, at 7:04 PM, Simon Watson wrote:

page.

Regards,

  • newB

Do you mean to say that x holds a Unicode code point? If that’s the
case (since ASCII is a subset of Unicode, x.to_s => “1046” is
trivial), then you can use something like this code I wrote a while
back:
irb> (“U+”+(‘0’*4+x.to_s(16))[-4,4]).to_utf8
=> “\320\226”

Of course, you could hide most of that in an Integer#to_utf8 method.

-Rob

Rob B. http://agileconsultingllc.com
[email protected]

-- ruby --

class String

For a string that matches /(?i:U+?|\u)?\d{4}/, return a

suitable UTF-8

string for that character.

def to_utf8
case point = self.match(/[[:xdigit:]]{4}/)[0].to_i(16)
when 0…0x7f
point.chr
when 0x80…0x07ff
x = point & 0b111111
point >>= 6
y = point
“#{(0xC0 | y).chr}#{(0x80 | x).chr}”
when 0x0800…0xFFFF
x = point & 0b111111
point >>= 6
y = point & 0b111111
point >>= 6
z = point
“#{(0xE0 | z).chr}#{(0x80 | y).chr}#{(0x80 | x).chr}”
when 0x10000…0x10FFFF
raise NotImplementedError, “UTF-8 four byte sequences not yet
supported”
else
raise ArgumentError, “Values above U+10FFFF are not supported”
end
end
end

if FILE == $0
require ‘test/unit’
class UnicodeHelperTest < Test::Unit::TestCase
def test_ascii
assert_equal ‘!’, “U+0021”.to_utf8, ‘EXCLAMATION MARK’
assert_equal ‘A’, “U+0041”.to_utf8, ‘UPPERCASE LETTER A’
assert_equal ‘-’, “U+002D”.to_utf8, ‘HYPHEN-MINUS’
assert_equal ‘~’, “U+007E”.to_utf8, ‘TILDE’

   assert_equal '!', "0021".to_utf8, 'EXCLAMATION MARK'
   assert_equal 'A', "0041".to_utf8, 'UPPERCASE LETTER A'
   assert_equal '-', "002D".to_utf8, 'HYPHEN-MINUS'
   assert_equal '~', "007E".to_utf8, 'TILDE'

   assert_equal '!', "\\u0021".to_utf8, 'EXCLAMATION MARK'
   assert_equal 'A', "\\u0041".to_utf8, 'UPPERCASE LETTER A'
   assert_equal '-', "\\u002D".to_utf8, 'HYPHEN-MINUS'
   assert_equal '~', "\\u007E".to_utf8, 'TILDE'
 end
 def test_hi_bit_ascii
   assert_equal "\xC2\x80", "U+0080".to_utf8, "C-cedilla"
   assert_equal "\xC2\xA4", "U+00A4".to_utf8, "Spanish n-tilde"
 end
 def test_general_punctuation
   assert_equal "\342\200\220", "U+2010".to_utf8, "HYPHEN"
   assert_equal "\342\200\221", "U+2011".to_utf8, "NON-BREAKING

HYPHEN"
assert_equal “\342\200\222”, “U+2012”.to_utf8, “FIGURE DASH”
assert_equal “\342\200\223”, “U+2013”.to_utf8, “EN DASH”
assert_equal “\342\200\224”, “U+2014”.to_utf8, “EM DASH”
assert_equal “\342\200\225”, “U+2015”.to_utf8, “QUOTATION DASH”
end
end
end
END

3

On 30 Sep 2008, at 14:00, Rob B. wrote:

Do you mean to say that x holds a Unicode code point? If that’s the
case (since ASCII is a subset of Unicode, x.to_s => “1046” is
trivial), then you can use something like this code I wrote a while
back:
irb> (“U+”+(‘0’*4+x.to_s(16))[-4,4]).to_utf8
=> “\320\226”

Depend on what you are doing,

[1046].pack(‘U’)

may also be appropriate.

Fred