idnx 0.0.1 → 0.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 931f92a4d37e74baad51a046058aa31aa0054563936e40e451f77eecd8eaf1b1
-  data.tar.gz: bed5e05b59d801504c2110d851723e09b302dc4a48169334ddebb4a493ed07d1
+  metadata.gz: 7e9eef708b35b383a3270b44ca3670e72d92bd3c29e89ecf2df9ba0ded94a144
+  data.tar.gz: 4c08ebc4dc52aa8ccf6fe5b1c53c17e7eb2a6e07d11c7d689f049d757d6de379
 SHA512:
-  metadata.gz: 905cf486978554c95bb7c7b912338439cb4acd431bb50d11cbe6afaee8047a9b53e28f522b4986873817ca8c76523eab7a5b784109f877001846228e7212ebfa
-  data.tar.gz: c94fd02aaf52386076a9a022f922d1962c1ce813b82a403f35f0d50fd4c68bb719a7def9a401a34a1aca62f8ac8443695e9dbadba5ed7bf8316459ea0b001a16
+  metadata.gz: 960e68d02443bf62b55290187f1785e16b9d16005107ce3b3f7db4ee016c1c91dd867de57d04f8cb92db18081a7b01499ac20fa8082f76ba884e8702806ff8ea
+  data.tar.gz: 0be4f68aa311208b5652e2773192487cec450e6ff604d04a1d72f662136309928bd405a095f36eb4dc6f9180dafa10d1f90fe4f847225dd6288b5a2d4aeab29f

data/LICENSE.txt

@@ -191,7 +191,7 @@
    limitations under the License.
 
 
-* lib/httpx/domain_name.rb
+* lib/idnx/ruby.rb
 
 This file is derived from the implementation of punycode available at
 here:

data/README.md

@@ -1,5 +1,8 @@
 # Idnx
 
+[![CI](https://github.com/HoneyryderChuck/idnx/actions/workflows/test.yml/badge.svg)](https://github.com/HoneyryderChuck/idnx/actions/workflows/test.yml)
+
+
 
 `idnx` provides a Ruby API for decoding Internationalized domain names into Punycode.
 
@@ -7,6 +10,7 @@ It provides multi-platform support by using the most approriate strategy based o
 
 * It uses (and requires the installation of) [libidn2](https://github.com/libidn/libidn2) in Linux / MacOS;
 * It uses [the appropriate winnls APIs](https://docs.microsoft.com/en-us/windows/win32/api/winnls/nf-winnls-idntoascii) in Windows;
+* It falls back to a pure ruby Punycode 2003 implementation;
 
 ## Installation
 
@@ -41,6 +45,16 @@ require "idnx"
 Idnx.to_punycode("bücher.de") #=> "xn--bcher-kva.de"
 ```
 
+## Ruby Support Policy
+
+This library supports at least ruby 2.4 .It also supports both JRuby and Truffleruby.
+
+## Known Issues
+
+### JRuby on MacOS
+
+`idnx` won't work in MacOS when using JRuby 9.2 or lower, due to jruby FFI not having the same path lookup logic than it's counterpart for CRuby, thereby not finding `brew`-installed `libidn2`. This has been fixed since JRuby 9.3 .
+
 ## Development
 
 If you want to contribute, fork this project, and submit changes via a PR on github.

data/lib/idnx/idn2.rb

@@ -21,12 +21,12 @@ module Idnx
     IDN2_NONTRANSITIONAL = 8
 
     FLAGS = if Gem::Version.new(VERSION) >= Gem::Version.new("0.14.0")
-      IDN2_NFC_INPUT | IDN2_NONTRANSITIONAL
-    else
-      IDN2_NFC_INPUT
-    end
+              IDN2_NFC_INPUT | IDN2_NONTRANSITIONAL
+            else
+              IDN2_NFC_INPUT
+            end
 
-    attach_function :idn2_lookup_ul, [:string, :pointer, :int], :int
+    attach_function :idn2_lookup_ul, %i[string pointer int], :int
     attach_function :idn2_strerror, [:int], :string
     attach_function :idn2_free, [:pointer], :void
 
@@ -36,9 +36,7 @@ module Idnx
       string_ptr = FFI::MemoryPointer.new(:pointer)
       result = idn2_lookup_ul(hostname, string_ptr, FLAGS)
 
-      if result != IDN2_OK
-        result = idn2_lookup_ul(hostname, string_ptr, IDN2_TRANSITIONAL)
-      end
+      result = idn2_lookup_ul(hostname, string_ptr, IDN2_TRANSITIONAL) if result != IDN2_OK
 
       if result != IDN2_OK
         string_ptr.free

data/lib/idnx/ruby.rb

@@ -0,0 +1,299 @@
+# frozen_string_literal: true
+
+module Idnx
+  module Ruby
+    module_function
+
+    def lookup(hostname)
+      Punycode.encode_hostname(hostname)
+    end
+
+    # :nocov:
+    # -*- coding: utf-8 -*-
+    #--
+    # punycode.rb - PunyCode encoder for the Domain Name library
+    #
+    # Copyright (C) 2011-2017 Akinori MUSHA, All rights reserved.
+    #
+    # Ported from puny.c, a part of VeriSign XCode (encode/decode) IDN
+    # Library.
+    #
+    # Copyright (C) 2000-2002 Verisign Inc., All rights reserved.
+    #
+    # Redistribution and use in source and binary forms, with or
+    # without modification, are permitted provided that the following
+    # conditions are met:
+    #
+    #  1) Redistributions of source code must retain the above copyright
+    #     notice, this list of conditions and the following disclaimer.
+    #
+    #  2) Redistributions in binary form must reproduce the above copyright
+    #     notice, this list of conditions and the following disclaimer in
+    #     the documentation and/or other materials provided with the
+    #     distribution.
+    #
+    #  3) Neither the name of the VeriSign Inc. nor the names of its
+    #     contributors may be used to endorse or promote products derived
+    #     from this software without specific prior written permission.
+    #
+    # THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+    # "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+    # LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
+    # FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
+    # COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
+    # INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+    # BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
+    # OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
+    # AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+    # LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
+    # ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+    # POSSIBILITY OF SUCH DAMAGE.
+    #
+    # This software is licensed under the BSD open source license. For more
+    # information visit www.opensource.org.
+    #
+    # Authors:
+    #  John Colosi (VeriSign)
+    #  Srikanth Veeramachaneni (VeriSign)
+    #  Nagesh Chigurupati (Verisign)
+    #  Praveen Srinivasan(Verisign)
+    #++
+    module Punycode
+      BASE = 36
+      TMIN = 1
+      TMAX = 26
+      SKEW = 38
+      DAMP = 700
+      INITIAL_BIAS = 72
+      INITIAL_N = 0x80
+      DELIMITER = "-"
+
+      MAXINT = (1 << 32) - 1
+
+      LOBASE = BASE - TMIN
+      CUTOFF = LOBASE * TMAX / 2
+
+      RE_NONBASIC = /[^\x00-\x7f]/.freeze
+
+      # Returns the numeric value of a basic code point (for use in
+      # representing integers) in the range 0 to base-1, or nil if cp
+      # is does not represent a value.
+      DECODE_DIGIT = {}.tap do |map|
+        # ASCII A..Z map to 0..25
+        # ASCII a..z map to 0..25
+        (0..25).each { |i| map[65 + i] = map[97 + i] = i }
+        # ASCII 0..9 map to 26..35
+        (26..35).each { |i| map[22 + i] = i }
+      end
+
+      # Returns the basic code point whose value (when used for
+      # representing integers) is d, which must be in the range 0 to
+      # BASE-1.  The lowercase form is used unless flag is true, in
+      # which case the uppercase form is used.  The behavior is
+      # undefined if flag is nonzero and digit d has no uppercase
+      # form.
+      ENCODE_DIGIT = proc { |d, flag|
+        (d + 22 + (d < 26 ? 75 : 0) - (flag ? (1 << 5) : 0)).chr
+        #  0..25 map to ASCII a..z or A..Z
+        # 26..35 map to ASCII 0..9
+      }
+
+      DOT = "."
+      PREFIX = "xn--"
+
+      # Most errors we raise are basically kind of ArgumentError.
+      class ArgumentError < ::ArgumentError; end
+      class BufferOverflowError < ArgumentError; end
+
+      module_function
+
+      # Encode a +string+ in Punycode
+      def encode(string)
+        input = string.unpack("U*")
+        output = +""
+
+        # Initialize the state
+        n = INITIAL_N
+        delta = 0
+        bias = INITIAL_BIAS
+
+        # Handle the basic code points
+        input.each { |cp| output << cp.chr if cp < 0x80 }
+
+        h = b = output.length
+
+        # h is the number of code points that have been handled, b is the
+        # number of basic code points, and out is the number of characters
+        # that have been output.
+
+        output << DELIMITER if b > 0
+
+        # Main encoding loop
+
+        while h < input.length
+          # All non-basic code points < n have been handled already.  Find
+          # the next larger one
+
+          m = MAXINT
+          input.each do |cp|
+            m = cp if (n...m) === cp
+          end
+
+          # Increase delta enough to advance the decoder's <n,i> state to
+          # <m,0>, but guard against overflow
+
+          delta += (m - n) * (h + 1)
+          raise BufferOverflowError if delta > MAXINT
+
+          n = m
+
+          input.each do |cp|
+            # AMC-ACE-Z can use this simplified version instead
+            if cp < n
+              delta += 1
+              raise BufferOverflowError if delta > MAXINT
+            elsif cp == n
+              # Represent delta as a generalized variable-length integer
+              q = delta
+              k = BASE
+              loop do
+                t = k <= bias ? TMIN : k - bias >= TMAX ? TMAX : k - bias
+                break if q < t
+
+                q, r = (q - t).divmod(BASE - t)
+                output << ENCODE_DIGIT[t + r, false]
+                k += BASE
+              end
+
+              output << ENCODE_DIGIT[q, false]
+
+              # Adapt the bias
+              delta = h == b ? delta / DAMP : delta >> 1
+              delta += delta / (h + 1)
+              bias = 0
+              while delta > CUTOFF
+                delta /= LOBASE
+                bias += BASE
+              end
+              bias += (LOBASE + 1) * delta / (delta + SKEW)
+
+              delta = 0
+              h += 1
+            end
+          end
+
+          delta += 1
+          n += 1
+        end
+
+        output
+      end
+
+      # Encode a hostname using IDN/Punycode algorithms
+      def encode_hostname(hostname)
+        hostname.match(RE_NONBASIC) || (return hostname)
+
+        hostname.split(DOT).map do |name|
+          if name.match(RE_NONBASIC)
+            PREFIX + encode(name)
+          else
+            name
+          end
+        end.join(DOT)
+      end
+
+      # Decode a +string+ encoded in Punycode
+      def decode(string)
+        # Initialize the state
+        n = INITIAL_N
+        i = 0
+        bias = INITIAL_BIAS
+
+        if j = string.rindex(DELIMITER)
+          b = string[0...j]
+
+          b.match(RE_NONBASIC) &&
+            raise(ArgumentError, "Illegal character is found in basic part: #{string.inspect}")
+
+          # Handle the basic code points
+
+          output = b.unpack("U*")
+          u = string[(j + 1)..-1]
+        else
+          output = []
+          u = string
+        end
+
+        # Main decoding loop: Start just after the last delimiter if any
+        # basic code points were copied; start at the beginning
+        # otherwise.
+
+        input = u.unpack("C*")
+        input_length = input.length
+        h = 0
+        out = output.length
+
+        while h < input_length
+          # Decode a generalized variable-length integer into delta,
+          # which gets added to i.  The overflow checking is easier
+          # if we increase i as we go, then subtract off its starting
+          # value at the end to obtain delta.
+
+          oldi = i
+          w = 1
+          k = BASE
+
+          loop do
+            (digit = DECODE_DIGIT[input[h]]) ||
+              raise(ArgumentError, "Illegal character is found in non-basic part: #{string.inspect}")
+            h += 1
+            i += digit * w
+            raise BufferOverflowError if i > MAXINT
+
+            t = k <= bias ? TMIN : k - bias >= TMAX ? TMAX : k - bias
+            break if digit < t
+
+            w *= BASE - t
+            raise BufferOverflowError if w > MAXINT
+
+            k += BASE
+            (h < input_length) || raise(ArgumentError, "Malformed input given: #{string.inspect}")
+          end
+
+          # Adapt the bias
+          delta = oldi == 0 ? i / DAMP : (i - oldi) >> 1
+          delta += delta / (out + 1)
+          bias = 0
+          while delta > CUTOFF
+            delta /= LOBASE
+            bias += BASE
+          end
+          bias += (LOBASE + 1) * delta / (delta + SKEW)
+
+          # i was supposed to wrap around from out+1 to 0, incrementing
+          # n each time, so we'll fix that now:
+
+          q, i = i.divmod(out + 1)
+          n += q
+          raise BufferOverflowError if n > MAXINT
+
+          # Insert n at position i of the output:
+
+          output[i, 0] = n
+
+          out += 1
+          i += 1
+        end
+        output.pack("U*")
+      end
+
+      # Decode a hostname using IDN/Punycode algorithms
+      def decode_hostname(hostname)
+        hostname.gsub(/(\A|#{Regexp.quote(DOT)})#{Regexp.quote(PREFIX)}([^#{Regexp.quote(DOT)}]*)/o) do
+          Regexp.last_match(1) << decode(Regexp.last_match(2))
+        end
+      end
+    end
+    # :nocov:
+  end
+end

data/lib/idnx/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Idnx
-  VERSION = "0.0.1"
+  VERSION = "0.1.0"
 end

data/lib/idnx/windows.rb

@@ -45,9 +45,9 @@ module Idnx
     #   LPCCH                              lpDefaultChar,
     #   LPBOOL                             lpUsedDefaultChar
     # );
-    attach_function :IdnToAscii, [:uint, :pointer, :int, :pointer, :int], :int
-    attach_function :MultiByteToWideChar, [:uint, :uint, :string, :int, :pointer, :int], :int
-    attach_function :WideCharToMultiByte, [:uint, :uint, :pointer, :int, :pointer, :int, :pointer, :pointer], :int
+    attach_function :IdnToAscii, %i[uint pointer int pointer int], :int
+    attach_function :MultiByteToWideChar, %i[uint uint string int pointer int], :int
+    attach_function :WideCharToMultiByte, %i[uint uint pointer int pointer int pointer pointer], :int
 
     module_function
 
@@ -55,6 +55,7 @@ module Idnx
       # turn to wchar
       wchar_len = MultiByteToWideChar(WIN32OLE::CP_UTF8, MB_ERR_INVALID_CHARS, hostname, -1, nil, 0)
       raise Error, "Failed to convert \"#{hostname}\" to wchar" if wchar_len.zero?
+
       wchar_ptr = FFI::MemoryPointer.new(:wchar_t, wchar_len)
       wchar_len = MultiByteToWideChar(WIN32OLE::CP_UTF8, 0, hostname, -1, wchar_ptr, wchar_len)
       raise Error, "Failed to convert \"#{hostname}\" to wchar" if wchar_len.zero?
@@ -64,27 +65,27 @@ module Idnx
       punycode_len = IdnToAscii(0, wchar_ptr, -1, punycode, IDN_MAX_LENGTH)
       wchar_ptr.free
 
-      if punycode_len == 0
+      if punycode_len.zero?
         last_error = FFI::LastError.error
 
         # operation completed successfully, hostname is not an IDN
         # return hostname if last_error == 0
 
         message = case last_error
-        when ERROR_INSUFFICIENT_BUFFER
-          "The supplied buffer size was not large enough, or it was incorrectly set to NULL"
-        when ERROR_INVALID_FLAGS
-          "The values supplied for flags were not valid"
-        when ERROR_INVALID_NAME
-          "An invalid name was supplied to the function"
-        when ERROR_INVALID_PARAMETER
-          "Any of the parameter values was invalid"
-        when ERROR_NO_UNICODE_TRANSLATION
-          "An invalid Unicode was found in a string"
-        else
-          "Failed to convert \"#{hostname}\"; (error: #{last_error})" \
-            "\n\nhttps://docs.microsoft.com/en-us/windows/win32/debug/system-error-codes#system-error-codes-1"
-        end
+                  when ERROR_INSUFFICIENT_BUFFER
+                    "The supplied buffer size was not large enough, or it was incorrectly set to NULL"
+                  when ERROR_INVALID_FLAGS
+                    "The values supplied for flags were not valid"
+                  when ERROR_INVALID_NAME
+                    "An invalid name was supplied to the function"
+                  when ERROR_INVALID_PARAMETER
+                    "Any of the parameter values was invalid"
+                  when ERROR_NO_UNICODE_TRANSLATION
+                    "An invalid Unicode was found in a string"
+                  else
+                    "Failed to convert \"#{hostname}\"; (error: #{last_error})" \
+                    "\n\nhttps://docs.microsoft.com/en-us/windows/win32/debug/system-error-codes#system-error-codes-1"
+                  end
         punycode.free
         raise Error, message
       end
@@ -92,9 +93,11 @@ module Idnx
       # turn to unicode
       unicode_len = WideCharToMultiByte(WIN32OLE::CP_UTF8, 0, punycode, -1, nil, 0, nil, nil)
       raise Error, "Failed to convert \"#{hostname}\" to utf8" if unicode_len.zero?
+
       utf8_ptr = FFI::MemoryPointer.new(:char, unicode_len)
       unicode_len = WideCharToMultiByte(WIN32OLE::CP_UTF8, 0, punycode, -1, utf8_ptr, unicode_len, nil, nil)
       raise Error, "Failed to convert \"#{hostname}\" to utf8" if unicode_len.zero?
+
       unicode = utf8_ptr.read_string(utf8_ptr.size)
       unicode.strip! # remove null-byte
     end

data/lib/idnx.rb

@@ -4,7 +4,7 @@ require_relative "idnx/version"
 require "ffi"
 
 module Idnx
-  Error = Class.new(StandardError)
+  class Error < StandardError; end
 
   module_function
 
@@ -16,5 +16,11 @@ end
 if FFI::Platform.windows?
   require "idnx/windows"
 else
-  require "idnx/idn2"
+  begin
+    require "idnx/idn2"
+  rescue LoadError
+    # fallback to pure ruby punycode 2003 implementation
+    require "idnx/ruby"
+    Idnx::Lib = Idnx::Ruby
+  end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: idnx
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.1.0
 platform: ruby
 authors:
 - HoneyryderChuck
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2021-06-11 00:00:00.000000000 Z
+date: 2021-09-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ffi
@@ -39,6 +39,7 @@ files:
 - README.md
 - lib/idnx.rb
 - lib/idnx/idn2.rb
+- lib/idnx/ruby.rb
 - lib/idnx/version.rb
 - lib/idnx/windows.rb
 homepage: https://github.com/honeyryderchuck/idnx