Module: Sprockets::Utils

Defined in:
lib/sprockets/utils.rb

Overview

`Utils`, we didn't know where else to put it!

Constant Summary collapse

UTF8_BOM_PATTERN =

Define UTF-8 and UTF-16 BOM pattern matchers. Avoid using a Regexp literal to prevent syntax errors in other interpreters.

Regexp.new("\\A\\xEF\\xBB\\xBF")
UTF16_BOM_PATTERN =
Regexp.new("\\A(\\xFE\\xFF|\\xFF\\xFE)")

Class Method Summary collapse

Class Method Details

.normalize_extension(extension) ⇒ Object

Prepends a leading “.” to an extension if its missing.

normalize_extension("js")
# => ".js"

normalize_extension(".css")
# => ".css"

60
61
62
63
64
65
66
67
# File 'lib/sprockets/utils.rb', line 60

def self.normalize_extension(extension)
  extension = extension.to_s
  if extension[/^\./]
    extension
  else
    ".#{extension}"
  end
end

.read_unicode(pathname) ⇒ Object


11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
# File 'lib/sprockets/utils.rb', line 11

def self.read_unicode(pathname, external_encoding = Encoding.default_external)
  pathname.open("r:#{external_encoding}") do |f|
    f.read.tap do |data|
      # Eager validate the file's encoding. In most cases we
      # expect it to be UTF-8 unless `default_external` is set to
      # something else. An error is usually raised if the file is
      # saved as UTF-16 when we expected UTF-8.
      if !data.valid_encoding?
        raise EncodingError, "#{pathname} has a invalid " +
          "#{data.encoding} byte sequence"

      # If the file is UTF-8 and theres a BOM, strip it for safe concatenation.
      elsif data.encoding.name == "UTF-8" && data =~ UTF8_BOM_PATTERN
        data.sub!(UTF8_BOM_PATTERN, "")
      end
    end
  end
end