addressable 2.8.1 → 2.8.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 3ddda72232f6aef9f6f4311c2855f1b3fea9acc80f51c4e5e90bd23820b0d74e
-  data.tar.gz: 88f208cb2d73dec64663e6e3c1710dc08b188402937038fefce6a2d47debc97d
+  metadata.gz: 5b7d4ea3b683b3b7c719ae54919505e8e98823b39e54cd68d3579319b61d9824
+  data.tar.gz: d09b61eba34544826aec64909b848b53549ae64059e09cc5f462991a59fd2465
 SHA512:
-  metadata.gz: 4e1b0c83fc2f2e54cba6804f3840c8bb29f0d73259979d5bf678ebd5d32257b91585d9b71a780321427835d10f98b9c07969267878f0b032c53d031c30202a4c
-  data.tar.gz: 35abc3652c8ff92032411e4daad7133b7c4649aff4a1a25fe622cf1c9b661d56f096a1084392e053788baea79f551a9d4a448d16c9e61e23bf4d9692a6728bf3
+  metadata.gz: 37927e878581e256a98ac619723a90f7b7a2d5b655ef613872a4f963ffdccabc6f0b761b65673fc6929509b0f783b48460b2643ac33c2d076bc5234e4126573b
+  data.tar.gz: 3afd1d272e41f97959bc8b34188b8d9fc419fcc01af9ea8a955d3ee2e6f57e3417d5ec6c58378e6b29fedb15e948ace7c080cd2c3739d6e24a4dd04eb3fa3f70

data/CHANGELOG.md

@@ -1,13 +1,42 @@
+# Addressable 2.8.5
+- Fix thread safety issue with encoding tables ([#515])
+- Define URI::NONE as a module to avoid serialization issues ([#509])
+- Fix YAML serialization ([#508])
+
+[#508]: https://github.com/sporkmonger/addressable/pull/508
+[#509]: https://github.com/sporkmonger/addressable/pull/509
+[#515]: https://github.com/sporkmonger/addressable/pull/515
+
+# Addressable 2.8.4
+- Restore `Addressable::IDNA.unicode_normalize_kc` as a deprecated method ([#504])
+
+[#504]: https://github.com/sporkmonger/addressable/pull/504
+
+# Addressable 2.8.3
+- Fix template expand level 2 hash support for non-string objects ([#499], [#498])
+
+[#499]: https://github.com/sporkmonger/addressable/pull/499
+[#498]: https://github.com/sporkmonger/addressable/pull/498
+
+# Addressable 2.8.2
+- Improve cache hits and JIT friendliness ([#486](https://github.com/sporkmonger/addressable/pull/486))
+- Improve code style and test coverage ([#482](https://github.com/sporkmonger/addressable/pull/482))
+- Ensure reset of deferred validation ([#481](https://github.com/sporkmonger/addressable/pull/481))
+- Resolve normalization differences between `IDNA::Native` and `IDNA::Pure` ([#408](https://github.com/sporkmonger/addressable/issues/408), [#492])
+- Remove redundant colon in `Addressable::URI::CharacterClasses::AUTHORITY` regex ([#438](https://github.com/sporkmonger/addressable/pull/438)) (accidentally reverted by [#449] merge but [added back](https://github.com/sporkmonger/addressable/pull/492#discussion_r1105125280) in [#492])
+
+[#492]: https://github.com/sporkmonger/addressable/pull/492
+
 # Addressable 2.8.1
 - refactor `Addressable::URI.normalize_path` to address linter offenses ([#430](https://github.com/sporkmonger/addressable/pull/430))
-- remove redundant colon in `Addressable::URI::CharacterClasses::AUTHORITY` regex ([#438](https://github.com/sporkmonger/addressable/pull/438))
 - update gemspec to reflect supported Ruby versions ([#466], [#464], [#463])
 - compatibility w/ public_suffix 5.x ([#466], [#465], [#460])
 - fixes "invalid byte sequence in UTF-8" exception when unencoding URLs containing non UTF-8 characters ([#459](https://github.com/sporkmonger/addressable/pull/459))
-- `Ractor` compatibility ([#449](https://github.com/sporkmonger/addressable/pull/449))
+- `Ractor` compatibility ([#449])
 - use the whole string instead of a single line for template match ([#431](https://github.com/sporkmonger/addressable/pull/431))
 - force UTF-8 encoding only if needed ([#341](https://github.com/sporkmonger/addressable/pull/341))
 
+[#449]: https://github.com/sporkmonger/addressable/pull/449
 [#460]: https://github.com/sporkmonger/addressable/pull/460
 [#463]: https://github.com/sporkmonger/addressable/pull/463
 [#464]: https://github.com/sporkmonger/addressable/pull/464

data/Rakefile

@@ -20,11 +20,17 @@ additionally provides extensive support for IRIs and URI templates.
 TEXT
 
 PKG_FILES = FileList[
-    "lib/**/*", "spec/**/*", "vendor/**/*", "data/**/*",
-    "tasks/**/*",
-    "[A-Z]*", "Rakefile"
-].exclude(/pkg/).exclude(/database\.yml/).
-  exclude(/Gemfile\.lock/).exclude(/[_\.]git$/)
+    "data/**/*",
+    "lib/**/*.rb",
+    "spec/**/*.rb",
+    "tasks/**/*.rake",
+    "addressable.gemspec",
+    "CHANGELOG.md",
+    "Gemfile",
+    "LICENSE.txt",
+    "README.md",
+    "Rakefile",
+]
 
 task :default => "spec"
 

data/addressable.gemspec

@@ -0,0 +1,28 @@
+# -*- encoding: utf-8 -*-
+# stub: addressable 2.8.5 ruby lib
+
+Gem::Specification.new do |s|
+  s.name = "addressable".freeze
+  s.version = "2.8.5"
+
+  s.required_rubygems_version = Gem::Requirement.new(">= 0".freeze) if s.respond_to? :required_rubygems_version=
+  s.metadata = { "changelog_uri" => "https://github.com/sporkmonger/addressable/blob/main/CHANGELOG.md" } if s.respond_to? :metadata=
+  s.require_paths = ["lib".freeze]
+  s.authors = ["Bob Aman".freeze]
+  s.date = "2023-08-03"
+  s.description = "Addressable is an alternative implementation to the URI implementation that is\npart of Ruby's standard library. It is flexible, offers heuristic parsing, and\nadditionally provides extensive support for IRIs and URI templates.\n".freeze
+  s.email = "bob@sporkmonger.com".freeze
+  s.extra_rdoc_files = ["README.md".freeze]
+  s.files = ["CHANGELOG.md".freeze, "Gemfile".freeze, "LICENSE.txt".freeze, "README.md".freeze, "Rakefile".freeze, "addressable.gemspec".freeze, "data/unicode.data".freeze, "lib/addressable.rb".freeze, "lib/addressable/idna.rb".freeze, "lib/addressable/idna/native.rb".freeze, "lib/addressable/idna/pure.rb".freeze, "lib/addressable/template.rb".freeze, "lib/addressable/uri.rb".freeze, "lib/addressable/version.rb".freeze, "spec/addressable/idna_spec.rb".freeze, "spec/addressable/net_http_compat_spec.rb".freeze, "spec/addressable/security_spec.rb".freeze, "spec/addressable/template_spec.rb".freeze, "spec/addressable/uri_spec.rb".freeze, "spec/spec_helper.rb".freeze, "tasks/clobber.rake".freeze, "tasks/gem.rake".freeze, "tasks/git.rake".freeze, "tasks/metrics.rake".freeze, "tasks/profile.rake".freeze, "tasks/rspec.rake".freeze, "tasks/yard.rake".freeze]
+  s.homepage = "https://github.com/sporkmonger/addressable".freeze
+  s.licenses = ["Apache-2.0".freeze]
+  s.rdoc_options = ["--main".freeze, "README.md".freeze]
+  s.required_ruby_version = Gem::Requirement.new(">= 2.2".freeze)
+  s.rubygems_version = "3.4.18".freeze
+  s.summary = "URI Implementation".freeze
+
+  s.specification_version = 4
+
+  s.add_runtime_dependency(%q<public_suffix>.freeze, [">= 2.0.2", "< 6.0"])
+  s.add_development_dependency(%q<bundler>.freeze, [">= 1.0", "< 3.0"])
+end

data/lib/addressable/idna/native.rb

@@ -29,8 +29,14 @@ module Addressable
        IDN::Punycode.decode(value.to_s)
      end
 
-    def self.unicode_normalize_kc(value)
-      IDN::Stringprep.nfkc_normalize(value.to_s)
+    class << self
+      # @deprecated Use {String#unicode_normalize(:nfkc)} instead
+      def unicode_normalize_kc(value)
+        value.to_s.unicode_normalize(:nfkc)
+      end
+
+      extend Gem::Deprecate
+      deprecate :unicode_normalize_kc, "String#unicode_normalize(:nfkc)", 2023, 4
     end
 
     def self.to_ascii(value)

data/lib/addressable/idna/pure.rb

@@ -66,7 +66,7 @@ module Addressable
     # domain name as described in RFC 3490.
     def self.to_ascii(input)
       input = input.to_s unless input.is_a?(String)
-      input = input.dup
+      input = input.dup.force_encoding(Encoding::UTF_8).unicode_normalize(:nfkc)
       if input.respond_to?(:force_encoding)
         input.force_encoding(Encoding::ASCII_8BIT)
       end
@@ -77,7 +77,7 @@ module Addressable
             part.force_encoding(Encoding::ASCII_8BIT)
           end
           if part =~ UTF8_REGEX && part =~ UTF8_REGEX_MULTIBYTE
-            ACE_PREFIX + punycode_encode(unicode_normalize_kc(part))
+            ACE_PREFIX + punycode_encode(part)
           else
             part
           end
@@ -112,13 +112,14 @@ module Addressable
       output
     end
 
-    # Unicode normalization form KC.
-    def self.unicode_normalize_kc(input)
-      input = input.to_s unless input.is_a?(String)
-      unpacked = input.unpack("U*")
-      unpacked =
-        unicode_compose(unicode_sort_canonical(unicode_decompose(unpacked)))
-      return unpacked.pack("U*")
+    class << self
+      # @deprecated Use {String#unicode_normalize(:nfkc)} instead
+      def unicode_normalize_kc(value)
+        value.to_s.unicode_normalize(:nfkc)
+      end
+
+      extend Gem::Deprecate
+      deprecate :unicode_normalize_kc, "String#unicode_normalize(:nfkc)", 2023, 4
     end
 
     ##
@@ -136,164 +137,6 @@ module Addressable
     end
     private_class_method :unicode_downcase
 
-    def self.unicode_compose(unpacked)
-      unpacked_result = []
-      length = unpacked.length
-
-      return unpacked if length == 0
-
-      starter = unpacked[0]
-      starter_cc = lookup_unicode_combining_class(starter)
-      starter_cc = 256 if starter_cc != 0
-      for i in 1...length
-        ch = unpacked[i]
-
-        if (starter_cc == 0 &&
-            (composite = unicode_compose_pair(starter, ch)) != nil)
-          starter = composite
-        else
-          unpacked_result << starter
-          starter = ch
-        end
-      end
-      unpacked_result << starter
-      return unpacked_result
-    end
-    private_class_method :unicode_compose
-
-    def self.unicode_compose_pair(ch_one, ch_two)
-      if ch_one >= HANGUL_LBASE && ch_one < HANGUL_LBASE + HANGUL_LCOUNT &&
-          ch_two >= HANGUL_VBASE && ch_two < HANGUL_VBASE + HANGUL_VCOUNT
-        # Hangul L + V
-        return HANGUL_SBASE + (
-          (ch_one - HANGUL_LBASE) * HANGUL_VCOUNT + (ch_two - HANGUL_VBASE)
-        ) * HANGUL_TCOUNT
-      elsif ch_one >= HANGUL_SBASE &&
-          ch_one < HANGUL_SBASE + HANGUL_SCOUNT &&
-          (ch_one - HANGUL_SBASE) % HANGUL_TCOUNT == 0 &&
-          ch_two >= HANGUL_TBASE && ch_two < HANGUL_TBASE + HANGUL_TCOUNT
-           # Hangul LV + T
-        return ch_one + (ch_two - HANGUL_TBASE)
-      end
-
-      p = []
-
-      ucs4_to_utf8(ch_one, p)
-      ucs4_to_utf8(ch_two, p)
-
-      return lookup_unicode_composition(p)
-    end
-    private_class_method :unicode_compose_pair
-
-    def self.ucs4_to_utf8(char, buffer)
-      if char < 128
-        buffer << char
-      elsif char < 2048
-        buffer << (char >> 6 | 192)
-        buffer << (char & 63 | 128)
-      elsif char < 0x10000
-        buffer << (char >> 12 | 224)
-        buffer << (char >> 6 & 63 | 128)
-        buffer << (char & 63 | 128)
-      elsif char < 0x200000
-        buffer << (char >> 18 | 240)
-        buffer << (char >> 12 & 63 | 128)
-        buffer << (char >> 6 & 63 | 128)
-        buffer << (char & 63 | 128)
-      elsif char < 0x4000000
-        buffer << (char >> 24 | 248)
-        buffer << (char >> 18 & 63 | 128)
-        buffer << (char >> 12 & 63 | 128)
-        buffer << (char >> 6 & 63 | 128)
-        buffer << (char & 63 | 128)
-      elsif char < 0x80000000
-        buffer << (char >> 30 | 252)
-        buffer << (char >> 24 & 63 | 128)
-        buffer << (char >> 18 & 63 | 128)
-        buffer << (char >> 12 & 63 | 128)
-        buffer << (char >> 6 & 63 | 128)
-        buffer << (char & 63 | 128)
-      end
-    end
-    private_class_method :ucs4_to_utf8
-
-    def self.unicode_sort_canonical(unpacked)
-      unpacked = unpacked.dup
-      i = 1
-      length = unpacked.length
-
-      return unpacked if length < 2
-
-      while i < length
-        last = unpacked[i-1]
-        ch = unpacked[i]
-        last_cc = lookup_unicode_combining_class(last)
-        cc = lookup_unicode_combining_class(ch)
-        if cc != 0 && last_cc != 0 && last_cc > cc
-          unpacked[i] = last
-          unpacked[i-1] = ch
-          i -= 1 if i > 1
-        else
-          i += 1
-        end
-      end
-      return unpacked
-    end
-    private_class_method :unicode_sort_canonical
-
-    def self.unicode_decompose(unpacked)
-      unpacked_result = []
-      for cp in unpacked
-        if cp >= HANGUL_SBASE && cp < HANGUL_SBASE + HANGUL_SCOUNT
-          l, v, t = unicode_decompose_hangul(cp)
-          unpacked_result << l
-          unpacked_result << v if v
-          unpacked_result << t if t
-        else
-          dc = lookup_unicode_compatibility(cp)
-          unless dc
-            unpacked_result << cp
-          else
-            unpacked_result.concat(unicode_decompose(dc.unpack("U*")))
-          end
-        end
-      end
-      return unpacked_result
-    end
-    private_class_method :unicode_decompose
-
-    def self.unicode_decompose_hangul(codepoint)
-      sindex = codepoint - HANGUL_SBASE;
-      if sindex < 0 || sindex >= HANGUL_SCOUNT
-        l = codepoint
-        v = t = nil
-        return l, v, t
-      end
-      l = HANGUL_LBASE + sindex / HANGUL_NCOUNT
-      v = HANGUL_VBASE + (sindex % HANGUL_NCOUNT) / HANGUL_TCOUNT
-      t = HANGUL_TBASE + sindex % HANGUL_TCOUNT
-      if t == HANGUL_TBASE
-        t = nil
-      end
-      return l, v, t
-    end
-    private_class_method :unicode_decompose_hangul
-
-    def self.lookup_unicode_combining_class(codepoint)
-      codepoint_data = UNICODE_DATA[codepoint]
-      (codepoint_data ?
-        (codepoint_data[UNICODE_DATA_COMBINING_CLASS] || 0) :
-        0)
-    end
-    private_class_method :lookup_unicode_combining_class
-
-    def self.lookup_unicode_compatibility(codepoint)
-      codepoint_data = UNICODE_DATA[codepoint]
-      (codepoint_data ?
-        codepoint_data[UNICODE_DATA_COMPATIBILITY] : nil)
-    end
-    private_class_method :lookup_unicode_compatibility
-
     def self.lookup_unicode_lowercase(codepoint)
       codepoint_data = UNICODE_DATA[codepoint]
       (codepoint_data ?
@@ -302,21 +145,6 @@ module Addressable
     end
     private_class_method :lookup_unicode_lowercase
 
-    def self.lookup_unicode_composition(unpacked)
-      return COMPOSITION_TABLE[unpacked]
-    end
-    private_class_method :lookup_unicode_composition
-
-    HANGUL_SBASE =  0xac00
-    HANGUL_LBASE =  0x1100
-    HANGUL_LCOUNT = 19
-    HANGUL_VBASE =  0x1161
-    HANGUL_VCOUNT = 21
-    HANGUL_TBASE =  0x11a7
-    HANGUL_TCOUNT = 28
-    HANGUL_NCOUNT = HANGUL_VCOUNT * HANGUL_TCOUNT # 588
-    HANGUL_SCOUNT = HANGUL_LCOUNT * HANGUL_NCOUNT # 11172
-
     UNICODE_DATA_COMBINING_CLASS = 0
     UNICODE_DATA_EXCLUSION = 1
     UNICODE_DATA_CANONICAL = 2

data/lib/addressable/template.rb

@@ -892,25 +892,24 @@ module Addressable
     # operator.
     #
     # @param [Hash, Array, String] value
-    #   Normalizes keys and values with IDNA#unicode_normalize_kc
+    #   Normalizes unicode keys and values with String#unicode_normalize (NFC)
     #
     # @return [Hash, Array, String] The normalized values
     def normalize_value(value)
-      unless value.is_a?(Hash)
-        value = value.respond_to?(:to_ary) ? value.to_ary : value.to_str
-      end
-
       # Handle unicode normalization
-      if value.kind_of?(Array)
-        value.map! { |val| Addressable::IDNA.unicode_normalize_kc(val) }
+      if value.respond_to?(:to_ary)
+        value.to_ary.map! { |val| normalize_value(val) }
       elsif value.kind_of?(Hash)
         value = value.inject({}) { |acc, (k, v)|
-          acc[Addressable::IDNA.unicode_normalize_kc(k)] =
-            Addressable::IDNA.unicode_normalize_kc(v)
+          acc[normalize_value(k)] = normalize_value(v)
           acc
         }
       else
-        value = Addressable::IDNA.unicode_normalize_kc(value)
+        value = value.to_s if !value.kind_of?(String)
+        if value.encoding != Encoding::UTF_8
+          value = value.dup.force_encoding(Encoding::UTF_8)
+        end
+        value = value.unicode_normalize(:nfc)
       end
       value
     end