Module: ActiveSupport::Multibyte::Unicode

Extended by:
Unicode
Included in:
Unicode
Defined in:
lib/active_support/multibyte/unicode.rb

Constant Summary collapse

UNICODE_VERSION =

The Unicode version that is supported by the implementation

RbConfig::CONFIG["UNICODE_VERSION"]

Instance Method Summary collapse

Instance Method Details

#compose(codepoints) ⇒ Object

Compose decomposed characters to the composed form.



33
34
35
# File 'lib/active_support/multibyte/unicode.rb', line 33

def compose(codepoints)
  codepoints.pack("U*").unicode_normalize(:nfc).codepoints
end

#decompose(type, codepoints) ⇒ Object

Decompose composed characters to the decomposed form.



24
25
26
27
28
29
30
# File 'lib/active_support/multibyte/unicode.rb', line 24

def decompose(type, codepoints)
  if type == :compatibility
    codepoints.pack("U*").unicode_normalize(:nfkd).codepoints
  else
    codepoints.pack("U*").unicode_normalize(:nfd).codepoints
  end
end

#default_normalization_formObject



11
12
13
14
15
# File 'lib/active_support/multibyte/unicode.rb', line 11

def default_normalization_form
  ActiveSupport::Deprecation.warn(
    "ActiveSupport::Multibyte::Unicode.default_normalization_form is deprecated and will be removed in Rails 7.0."
  )
end

#default_normalization_form=(_) ⇒ Object



17
18
19
20
21
# File 'lib/active_support/multibyte/unicode.rb', line 17

def default_normalization_form=(_)
  ActiveSupport::Deprecation.warn(
    "ActiveSupport::Multibyte::Unicode.default_normalization_form= is deprecated and will be removed in Rails 7.0."
  )
end

#tidy_bytes(string, force = false) ⇒ Object

Replaces all ISO-8859-1 or CP1252 characters by their UTF-8 equivalent resulting in a valid UTF-8 string.

Passing true will forcibly tidy all bytes, assuming that the string's encoding is entirely CP1252 or ISO-8859-1.



44
45
46
47
48
# File 'lib/active_support/multibyte/unicode.rb', line 44

def tidy_bytes(string, force = false)
  return string if string.empty? || string.ascii_only?
  return recode_windows1252_chars(string) if force
  string.scrub { |bad| recode_windows1252_chars(bad) }
end