ruby/lib/base64.rb

107 строки
3.3 KiB
Ruby
Исходник Обычный вид История

#
# = base64.rb: methods for base64-encoding and -decoding strings
#
# The Base64 module provides for the encoding (#encode64, #strict_encode64,
# #urlsafe_encode64) and decoding (#decode64, #strict_decode64,
# #urlsafe_decode64) of binary data using a Base64 representation.
#
# == Example
#
# A simple encoding and decoding.
#
# require "base64"
#
# enc = Base64.encode64('Send reinforcements')
# # -> "U2VuZCByZWluZm9yY2VtZW50cw==\n"
# plain = Base64.decode64(enc)
# # -> "Send reinforcements"
#
# The purpose of using base64 to encode data is that it translates any
# binary data into purely printable characters.
module Base64
module_function
# Returns the Base64-encoded version of +bin+.
# This method complies with RFC 2045.
# Line feeds are added to every 60 encoded characters.
#
# require 'base64'
# Base64.encode64("Now is the time for all good coders\nto learn Ruby")
#
# <i>Generates:</i>
#
# Tm93IGlzIHRoZSB0aW1lIGZvciBhbGwgZ29vZCBjb2RlcnMKdG8gbGVhcm4g
# UnVieQ==
def encode64(bin)
[bin].pack("m")
end
# Returns the Base64-decoded version of +str+.
# This method complies with RFC 2045.
# Characters outside the base alphabet are ignored.
#
# require 'base64'
# str = 'VGhpcyBpcyBsaW5lIG9uZQpUaGlzIG' +
# 'lzIGxpbmUgdHdvClRoaXMgaXMgbGlu' +
# 'ZSB0aHJlZQpBbmQgc28gb24uLi4K'
# puts Base64.decode64(str)
#
# <i>Generates:</i>
#
# This is line one
# This is line two
# This is line three
# And so on...
def decode64(str)
str.unpack("m").first
end
# Returns the Base64-encoded version of +bin+.
# This method complies with RFC 4648.
# No line feeds are added.
def strict_encode64(bin)
[bin].pack("m0")
end
# Returns the Base64-decoded version of +str+.
# This method complies with RFC 4648.
# ArgumentError is raised if +str+ is incorrectly padded or contains
# non-alphabet characters. Note that CR or LF are also rejected.
def strict_decode64(str)
str.unpack("m0").first
end
# Returns the Base64-encoded version of +bin+.
# This method complies with ``Base 64 Encoding with URL and Filename Safe
# Alphabet'' in RFC 4648.
# The alphabet uses '-' instead of '+' and '_' instead of '/'.
# Note that the result can still contain '='.
# You can remove the padding by setting "padding" as false.
def urlsafe_encode64(bin, padding: true)
str = strict_encode64(bin).tr("+/", "-_")
str = str.delete("=") unless padding
str
end
# Returns the Base64-decoded version of +str+.
# This method complies with ``Base 64 Encoding with URL and Filename Safe
# Alphabet'' in RFC 4648.
# The alphabet uses '-' instead of '+' and '_' instead of '/'.
#
# The padding character is optional.
# This method accepts both correctly-padded and unpadded input.
# Note that it still rejects incorrectly-padded input.
def urlsafe_decode64(str)
# NOTE: RFC 4648 does say nothing about unpadded input, but says that
# "the excess pad characters MAY also be ignored", so it is inferred that
# unpadded input is also acceptable.
str = str.tr("-_", "+/")
if !str.end_with?("=") && str.length % 4 != 0
str = str.ljust((str.length + 3) & ~3, "=")
end
strict_decode64(str)
end
end