remove workarounds for Ruby < 2.3

- remove constants for differentiating between Ruby versions
- remove fallback assignment for RUBY_ENGINE constant
- remove rubygems require
- use unicode character properties in regular expressions (e.g., \p{Word})
- remove COERCE_ENCODING constant
- always set FORCE_ENCODING constant to true unless default external encoding is not UTF-8
- remove workaround for computing line length
- use lines method instead of lines.entries to get lines for string
- always normalize lines
- check if JRuby class is defined instead of checking for JRuby engine
- remove global RUBY_ENGINE_JRUBY constant
- move RUBY_ENGINE_OPAL constant inside Asciidoctor module
- remove magic encoding comments (since default script encoding is UTF-8)
- remove workaround when printing encoding information
- remove fallback assignments for passthrough escape characters
- remove Ruby core backports (only leave Asciidoctor-specific extensions)
- use __dir__ instead of File.dirname __FILE__
- use Float#truncate to truncate column widths
- override Float#truncate on Ruby 2.3 to implement precision argument
- remove concurrent/hash backport
- simplify class_for_name helper
- don't call to_sym on constants returned from constants method
- remove unnecessary aliasing of regexp match variables; use magic variable instead (e.g., m = $~)
- remove workaround for sub_specialchars to accomodate gsub with map argument
- remove comment about %:z placeholder for strftime
- don't call .to_a on return value of Hash#keys
- get home directory using Dir.home
- remove exceptions and exclusions for Ruby < 2 in test suite
- update list of supported Ruby engines and versions in READMEs
- update ruby versions in run-tests.sh

75 files changed, 340 insertions, 630 deletions

diff --git a/CHANGELOG.adoc b/CHANGELOG.adoc
index bdc4b15b..37b2c2aa 100644
--- a/CHANGELOG.adoc
+++ b/CHANGELOG.adoc
@@ -18,6 +18,7 @@ For a detailed view of what has changed, refer to the {uri-repo}/commits/master[
 Enhancements::
 
   * add support for start attribute when using prettify to highlight source blocks with line numbering enabled
+  * drop support for Ruby < 2.3 and JRuby < 9.1 and remove workarounds (#2764)
 
 Bug Fixes::
 
diff --git a/README-de.adoc b/README-de.adoc
index 5a34646b..6f5117ac 100644
--- a/README-de.adoc
+++ b/README-de.adoc
@@ -139,9 +139,8 @@ Es wird für die AsciiDoc Vorschau-Erweiterungen für Chrome, Atom, Brackets und
 
 Asciidoctor arbeitet unter Linux, MacOS und Windows und benötigt eine der folgenden Implementierungen von {uri-ruby}[Ruby]:
 
-* Ruby (MRI/CRuby 1.8.7 - 2.6)
-* JRuby (1.7 in Ruby 1.8 und 1.9 modes, 9000)
-* Rubinius 2.2.x
+* CRuby (aka MRI) 2.3 - 2.6
+* JRuby 9.1 - 9.2
 * Opal (JavaScript)
 
 [CAUTION]
diff --git a/README-fr.adoc b/README-fr.adoc
index 6dc663ac..a4d2878e 100644
--- a/README-fr.adoc
+++ b/README-fr.adoc
@@ -136,9 +136,8 @@ Asciidoctor.js est utilisé pour faire fonctionner les extensions AsciiDoc Previ
 
 Asciidoctor fonctionne sur Linux, macOS et Windows et requiert une des implémentations suivantes de {uri-ruby}[Ruby] :
 
-* Ruby (MRI/CRuby 1.8.7 - 2.6)
-* JRuby (1.7 en mode Ruby 1.8 et 1.9, 9000)
-* Rubinius 2.2.x
+* CRuby (aka MRI) 2.3 - 2.6
+* JRuby 9.1 - 9.2
 * Opal (JavaScript)
 
 [CAUTION]
diff --git a/README-jp.adoc b/README-jp.adoc
index 616cb942..ce1e612d 100644
--- a/README-jp.adoc
+++ b/README-jp.adoc
@@ -120,9 +120,8 @@ AsciidoctorはHTML5出力をスタイルするため, モダンで,  {uri-founda
 
 AsciidoctorはLinux, OS X (Mac)とWindowsで動き, 下記の {uri-ruby}[Ruby]実装の一つを必要とします.
 
-* MRI (Ruby 1.8.7, 1.9.3, 2.0, 2.1, 2.2 & 2.3)
-* JRuby (1.7 in Ruby 1.8 and 1.9 modes, 9000)
-* Rubinius 2.2.x
+* CRuby (aka MRI) 2.3 - 2.6
+* JRuby 9.1 - 9.2
 * Opal (JavaScript)
 
 [CAUTION]
diff --git a/README-zh_CN.adoc b/README-zh_CN.adoc
index e882910a..4bc0827d 100644
--- a/README-zh_CN.adoc
+++ b/README-zh_CN.adoc
@@ -122,9 +122,8 @@ Asciidoctor 还提供了一个基于 {uri-foundation}[Foundation] 的现代化�
 
 Asciidoctor 可以运行在 Linux，OSX (Mac) 和 Windows 系统，但需要安装下面任意一个 {uri-ruby}[Ruby] 环境去实现：
 
-* MRI (Ruby 1.8.7, 1.9.3, 2.0, 2.1, 2.2 & 2.3)
-* JRuby (1.7 in Ruby 1.8 and 1.9 modes, 9000)
-* Rubinius 2.2.x
+* CRuby (aka MRI) 2.3 - 2.6
+* JRuby 9.1 - 9.2
 * Opal (JavaScript)
 
 我们欢迎你来帮助在这些以及其他平台测试 Asciidoctor。
diff --git a/README.adoc b/README.adoc
index 858edb09..12e9f400 100644
--- a/README.adoc
+++ b/README.adoc
@@ -142,9 +142,8 @@ It's used to power the AsciiDoc preview extensions for Chrome, Atom, Brackets an
 
 Asciidoctor works on Linux, macOS and Windows and requires one of the following implementations of {uri-ruby}[Ruby]:
 
-* Ruby (MRI/CRuby 1.8.7 - 2.6)
-* JRuby (1.7 in Ruby 1.8 and 1.9 modes, 9000)
-* Rubinius 2.2.x
+* CRuby (aka MRI) 2.3 - 2.6
+* JRuby 9.1 - 9.2
 * Opal (JavaScript)
 
 [CAUTION]
diff --git a/asciidoctor.gemspec b/asciidoctor.gemspec
index d9195f85..37c77ac8 100644
--- a/asciidoctor.gemspec
+++ b/asciidoctor.gemspec
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 begin
   require File.expand_path '../lib/asciidoctor/version', __FILE__
 rescue LoadError
diff --git a/bin/asciidoctor b/bin/asciidoctor
index ee394296..fa0b32e2 100755
--- a/bin/asciidoctor
+++ b/bin/asciidoctor
@@ -1,7 +1,5 @@
 #!/usr/bin/env ruby
 
-require 'rubygems' unless defined? Gem
-
 if File.exist?(asciidoctor = (File.expand_path '../../lib/asciidoctor', __FILE__))
   require asciidoctor
 else
diff --git a/bin/asciidoctor-safe b/bin/asciidoctor-safe
index 9a70f233..42053271 100755
--- a/bin/asciidoctor-safe
+++ b/bin/asciidoctor-safe
@@ -1,7 +1,5 @@
 #!/usr/bin/env ruby
 
-require 'rubygems' unless defined? Gem
-
 if File.exist?(asciidoctor = (File.expand_path '../../lib/asciidoctor', __FILE__))
   require asciidoctor
 else
diff --git a/features/step_definitions.rb b/features/step_definitions.rb
index a843aae1..f1ec3823 100644
--- a/features/step_definitions.rb
+++ b/features/step_definitions.rb
@@ -1,14 +1,10 @@
-# encoding: UTF-8
 ASCIIDOCTOR_PROJECT_DIR = File.dirname File.dirname(__FILE__)
+ASCIIDOCTOR_LIB_DIR = ENV['ASCIIDOCTOR_LIB_DIR'] || File.join(ASCIIDOCTOR_PROJECT_DIR, 'lib')
 Dir.chdir ASCIIDOCTOR_PROJECT_DIR
 
-if RUBY_VERSION < '1.9'
-  require 'rubygems'
-end
-
 require 'simplecov' if ENV['COVERAGE'] == 'true'
 
-require File.join(ASCIIDOCTOR_PROJECT_DIR, 'lib', 'asciidoctor')
+require File.join(ASCIIDOCTOR_LIB_DIR, 'asciidoctor')
 
 require 'rspec/expectations'
 require 'tilt'
diff --git a/lib/asciidoctor.rb b/lib/asciidoctor.rb
index 743c44e6..dee90838 100644
--- a/lib/asciidoctor.rb
+++ b/lib/asciidoctor.rb
@@ -1,10 +1,3 @@
-# encoding: UTF-8
-RUBY_ENGINE = 'unknown' unless defined? RUBY_ENGINE
-RUBY_ENGINE_OPAL = (RUBY_ENGINE == 'opal')
-RUBY_ENGINE_JRUBY = (RUBY_ENGINE == 'jruby')
-RUBY_MIN_VERSION_1_9 = (RUBY_VERSION >= '1.9')
-RUBY_MIN_VERSION_2 = (RUBY_VERSION >= '2')
-
 require 'set'
 
 # NOTE RUBY_ENGINE == 'opal' conditional blocks like this are filtered by the Opal preprocessor
@@ -20,7 +13,7 @@ else
 end
 
 # ideally we should use require_relative instead of modifying the LOAD_PATH
-$:.unshift File.dirname __FILE__
+$:.unshift __dir__
 
 require 'asciidoctor/logging'
 
@@ -54,12 +47,10 @@ require 'asciidoctor/logging'
 #   Asciidoctor.convert_file 'sample.adoc', :template_dir => 'path/to/templates'
 #
 module Asciidoctor
-
-  # alias the RUBY_ENGINE constant inside the Asciidoctor namespace
-  RUBY_ENGINE = ::RUBY_ENGINE
+  # alias the RUBY_ENGINE constant inside the Asciidoctor namespace and define a precomputed alias for runtime
+  RUBY_ENGINE_OPAL = (RUBY_ENGINE = ::RUBY_ENGINE) == 'opal'
 
   module SafeMode
-
     # A safe mode level that disables any of the security features enforced
     # by Asciidoctor (Ruby is still subject to its own restrictions).
     UNSAFE = 0;
@@ -185,7 +176,7 @@ module Asciidoctor
   end
 
   # The absolute root path of the Asciidoctor RubyGem
-  ROOT_PATH = ::File.dirname ::File.dirname ::File.expand_path __FILE__
+  ROOT_PATH = ::File.dirname ::File.absolute_path __dir__
 
   # The absolute lib path of the Asciidoctor RubyGem
   #LIB_PATH = ::File.join ROOT_PATH, 'lib'
@@ -194,30 +185,19 @@ module Asciidoctor
   DATA_PATH = ::File.join ROOT_PATH, 'data'
 
   # The user's home directory, as best we can determine it
-  # NOTE not using infix rescue for performance reasons, see: https://github.com/jruby/jruby/issues/1816
-  begin
-    USER_HOME = ::Dir.home
-  rescue
-    USER_HOME = ::ENV['HOME'] || ::Dir.pwd
-  end
+  USER_HOME = ::Dir.home
 
-  # Flag to indicate whether encoding can be coerced to UTF-8
+  # Flag to indicate whether encoding should be forced to UTF-8
   # _All_ input data must be force encoded to UTF-8 if Encoding.default_external is *not* UTF-8
   # Addresses failures performing string operations that are reported as "invalid byte sequence in US-ASCII"
-  # Ruby 1.8 doesn't seem to experience this problem (perhaps because it isn't validating the encodings)
-  COERCE_ENCODING = !::RUBY_ENGINE_OPAL && ::RUBY_MIN_VERSION_1_9
-
-  # Flag to indicate whether encoding of external strings needs to be forced to UTF-8
-  FORCE_ENCODING = COERCE_ENCODING && ::Encoding.default_external != ::Encoding::UTF_8
+  # Opal doesn't require this since all strings are already encoding internally as UTF-16
+  FORCE_ENCODING = ::Encoding.default_external != ::Encoding::UTF_8
 
   # Byte arrays for UTF-* Byte Order Marks
   BOM_BYTES_UTF_8 = [0xef, 0xbb, 0xbf]
   BOM_BYTES_UTF_16LE = [0xff, 0xfe]
   BOM_BYTES_UTF_16BE = [0xfe, 0xff]
 
-  # Flag to indicate that line length should be calculated using a unicode mode hint
-  FORCE_UNICODE_LINE_LENGTH = !::RUBY_MIN_VERSION_1_9
-
   # The endline character used for output; stored in constant table as an optimization
   LF = EOL = "\n"
 
@@ -393,29 +373,10 @@ module Asciidoctor
       # CC_ANY is any character except newlines
       CC_ANY = '.'
       CC_EOL = '$'
-      # character classes for the Regexp engine in Ruby >= 2 (Ruby 1.9 supports \p{} but has problems w/ encoding)
-      if ::RUBY_MIN_VERSION_2
-        CC_ALPHA = CG_ALPHA = '\p{Alpha}'
-        CC_ALNUM = CG_ALNUM = '\p{Alnum}'
-        CG_BLANK = '\p{Blank}'
-        CC_WORD  = CG_WORD = '\p{Word}'
-      # character classes for the Regexp engine in Ruby < 2
-      else
-        CC_ALPHA = '[:alpha:]'
-        CG_ALPHA = '[[:alpha:]]'
-        CC_ALNUM = '[:alnum:]'
-        CG_ALNUM = '[[:alnum:]]'
-        if ::RUBY_MIN_VERSION_1_9
-          CG_BLANK = '[[:blank:]]'
-          CC_WORD = '[:word:]'
-          CG_WORD = '[[:word:]]'
-        else
-          # NOTE Ruby 1.8 cannot match word characters beyond the ASCII range; if you need this feature, upgrade!
-          CG_BLANK = '[ \t]'
-          CC_WORD = '[:alnum:]_'
-          CG_WORD = '[[:alnum:]_]'
-        end
-      end
+      CC_ALPHA = CG_ALPHA = '\p{Alpha}'
+      CC_ALNUM = CG_ALNUM = '\p{Alnum}'
+      CG_BLANK = '\p{Blank}'
+      CC_WORD  = CG_WORD = '\p{Word}'
     end
 
     ## Document header
@@ -1093,7 +1054,7 @@ module Asciidoctor
     #   one two	 three   four
     #   five	six
     #
-    # TODO change to /(?<!\\)[ \t\n]+/ after dropping support for Ruby 1.8.7
+    # TODO change to /(?<!\\)[ \t\n]+/ once lookbehind assertions are implemented in all modern browsers
     SpaceDelimiterRx = /([^\\])[ \t\n]+/
 
     # Matches a + or - modifier in a subs list
@@ -1109,12 +1070,6 @@ module Asciidoctor
     #
     TrailingDigitsRx = /\d+$/
 
-    # Matches any character with multibyte support explicitly enabled (length of multibyte char = 1)
-    #
-    unless RUBY_ENGINE == 'opal'
-      UnicodeCharScanRx = /./u if FORCE_UNICODE_LINE_LENGTH
-    end
-
     # Detects strings that resemble URIs.
     #
     # Examples
@@ -1239,9 +1194,7 @@ module Asciidoctor
   quote_subs = nil
   compat_quote_subs = nil
 
-  # NOTE in Ruby 1.8.7, [^\\] does not match start of line,
-  # so we need to match it explicitly
-  # order is significant
+  # NOTE order of replacements is significant
   REPLACEMENTS = [
     # (C)
     [/\\?\(C\)/, '&#169;', :none],
@@ -1249,8 +1202,8 @@ module Asciidoctor
     [/\\?\(R\)/, '&#174;', :none],
     # (TM)
     [/\\?\(TM\)/, '&#8482;', :none],
-    # foo -- bar
-    # FIXME this drops the endline if it appears at end of line
+    # foo -- bar (where either space character can be a newline)
+    # NOTE this necessarily drops the newline if it appears at end of line
     [/(^|\n| |\\)--( |\n|$)/, '&#8201;&#8212;&#8201;', :none],
     # foo--bar
     [/(#{CG_WORD})\\?--(?=#{CG_WORD})/, '&#8212;&#8203;', :leading],
@@ -1300,7 +1253,7 @@ module Asciidoctor
 
     if !(attrs = options[:attributes])
       attrs = {}
-    elsif ::Hash === attrs || (::RUBY_ENGINE_JRUBY && ::Java::JavaUtil::Map === attrs)
+    elsif ::Hash === attrs || ((defined? ::Java::JavaUtil::Map) && ::Java::JavaUtil::Map === attrs)
       attrs = attrs.dup
     elsif ::Array === attrs
       attrs, attrs_arr = {}, attrs
@@ -1341,7 +1294,6 @@ module Asciidoctor
         attrs['docyear'] ||= input_mtime.year.to_s
       end
       # %Z is OS dependent and may contain characters that aren't UTF-8 encoded (see asciidoctor#2770 and asciidoctor.js#23)
-      # Ruby 1.8 doesn't support %:z
       doctime = (attrs['doctime'] ||= input_mtime.strftime %(%T #{input_mtime.utc_offset == 0 ? 'UTC' : '%z'}))
       attrs['docdatetime'] = %(#{docdate} #{doctime})
     elsif input.respond_to? :readlines
@@ -1353,7 +1305,7 @@ module Asciidoctor
       end
       lines = input.readlines
     elsif ::String === input
-      lines = ::RUBY_MIN_VERSION_2 ? input.lines : input.each_line.to_a
+      lines = input.lines
     elsif ::Array === input
       lines = input.drop 0
     else
diff --git a/lib/asciidoctor/abstract_block.rb b/lib/asciidoctor/abstract_block.rb
index 2c47814c..58d8178b 100644
--- a/lib/asciidoctor/abstract_block.rb
+++ b/lib/asciidoctor/abstract_block.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 class AbstractBlock < AbstractNode
   # Public: Get the Array of Asciidoctor::AbstractBlock sub-blocks for this block
diff --git a/lib/asciidoctor/abstract_node.rb b/lib/asciidoctor/abstract_node.rb
index 66524ed3..b3d6f2cb 100644
--- a/lib/asciidoctor/abstract_node.rb
+++ b/lib/asciidoctor/abstract_node.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: An abstract base class that provides state and methods for managing a
 # node of AsciiDoc content. The state and methods on this class are common to
@@ -404,7 +403,7 @@ class AbstractNode
       # caching requires the open-uri-cached gem to be installed
       # processing will be automatically aborted if these libraries can't be opened
       Helpers.require_library 'open-uri/cached', 'open-uri-cached'
-    elsif !::RUBY_ENGINE_OPAL
+    elsif !RUBY_ENGINE_OPAL
       # autoload open-uri
       ::OpenURI
     end
@@ -510,7 +509,6 @@ class AbstractNode
     opts = { :warn_on_failure => (opts != false) } unless ::Hash === opts
     if ::File.readable? path
       if opts[:normalize]
-        # NOTE Opal does not yet support File#readlines
         (Helpers.normalize_lines_array ::File.open(path, 'rb') {|f| f.each_line.to_a }).join LF
       else
         # QUESTION should we chomp or rstrip content?
@@ -546,7 +544,6 @@ class AbstractNode
         Helpers.require_library 'open-uri/cached', 'open-uri-cached' if doc.attr? 'cache-uri'
         begin
           if opts[:normalize]
-            # NOTE Opal does not yet support File#readlines
             (Helpers.normalize_lines_array ::OpenURI.open_uri(target) {|f| f.each_line.to_a }).join LF
           else
             ::OpenURI.open_uri(target) {|f| f.read }
diff --git a/lib/asciidoctor/attribute_list.rb b/lib/asciidoctor/attribute_list.rb
index 06c39010..66c85016 100644
--- a/lib/asciidoctor/attribute_list.rb
+++ b/lib/asciidoctor/attribute_list.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: Handles parsing AsciiDoc attribute lists into a Hash of key/value
 # pairs. By default, attributes must each be separated by a comma and quotes
diff --git a/lib/asciidoctor/block.rb b/lib/asciidoctor/block.rb
index b0cc5462..b3057cbb 100644
--- a/lib/asciidoctor/block.rb
+++ b/lib/asciidoctor/block.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: Methods for managing blocks of Asciidoc content in a section.
 #
diff --git a/lib/asciidoctor/callouts.rb b/lib/asciidoctor/callouts.rb
index b3eacec4..30f1044e 100644
--- a/lib/asciidoctor/callouts.rb
+++ b/lib/asciidoctor/callouts.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: Maintains a catalog of callouts and their associations.
 class Callouts
diff --git a/lib/asciidoctor/cli/invoker.rb b/lib/asciidoctor/cli/invoker.rb
index 5dd3c240..4bf998a3 100644
--- a/lib/asciidoctor/cli/invoker.rb
+++ b/lib/asciidoctor/cli/invoker.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
   module Cli
     # Public Invocation class for starting Asciidoctor via CLI
diff --git a/lib/asciidoctor/cli/options.rb b/lib/asciidoctor/cli/options.rb
index 4a3c09d9..a254ff35 100644
--- a/lib/asciidoctor/cli/options.rb
+++ b/lib/asciidoctor/cli/options.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
   module Cli
     FS = '/'
@@ -286,14 +285,10 @@ Example: asciidoctor -b html5 source.asciidoc
 
       def print_version os = $stdout
         os.puts %(Asciidoctor #{::Asciidoctor::VERSION} [https://asciidoctor.org])
-        if ::RUBY_MIN_VERSION_1_9
-          encoding_info = { 'lc' => 'locale', 'fs' => 'filesystem', 'in' => 'internal', 'ex' => 'external' }.map do |k, v|
-            %(#{k}:#{v == 'internal' ? (::File.open(__FILE__) {|f| f.getc }).encoding : (::Encoding.find v)})
-          end
-          os.puts %(Runtime Environment (#{::RUBY_DESCRIPTION}) (#{encoding_info.join ' '}))
-        else
-          os.puts %(Runtime Environment (#{::RUBY_DESCRIPTION}))
+        encoding_info = { 'lc' => 'locale', 'fs' => 'filesystem', 'in' => 'internal', 'ex' => 'external' }.map do |k, v|
+          %(#{k}:#{v == 'internal' ? (::File.open(__FILE__) {|f| f.getc }).encoding : (::Encoding.find v)})
         end
+        os.puts %(Runtime Environment (#{::RUBY_DESCRIPTION}) (#{encoding_info.join ' '}))
         0
       end
     end
diff --git a/lib/asciidoctor/converter.rb b/lib/asciidoctor/converter.rb
index d5528ab0..ff58dc8d 100644
--- a/lib/asciidoctor/converter.rb
+++ b/lib/asciidoctor/converter.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
   # A base module for defining converters that can be used to convert {AbstractNode}
   # objects in a parsed AsciiDoc document to a backend format such as HTML or
diff --git a/lib/asciidoctor/converter/base.rb b/lib/asciidoctor/converter/base.rb
index ffd9dfa0..83036e30 100644
--- a/lib/asciidoctor/converter/base.rb
+++ b/lib/asciidoctor/converter/base.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
   module Converter; end # required for Opal
 
diff --git a/lib/asciidoctor/converter/composite.rb b/lib/asciidoctor/converter/composite.rb
index 94d2973f..d1d6ce65 100644
--- a/lib/asciidoctor/converter/composite.rb
+++ b/lib/asciidoctor/converter/composite.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
   # A {Converter} implementation that delegates to the chain of {Converter}
   # objects passed to the constructor. Selects the first {Converter} that
diff --git a/lib/asciidoctor/converter/docbook45.rb b/lib/asciidoctor/converter/docbook45.rb
index 6d6f0f80..de44ff44 100644
--- a/lib/asciidoctor/converter/docbook45.rb
+++ b/lib/asciidoctor/converter/docbook45.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 require 'asciidoctor/converter/docbook5'
 
 module Asciidoctor
diff --git a/lib/asciidoctor/converter/docbook5.rb b/lib/asciidoctor/converter/docbook5.rb
index 83fad601..7e2a9327 100644
--- a/lib/asciidoctor/converter/docbook5.rb
+++ b/lib/asciidoctor/converter/docbook5.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
   # A built-in {Converter} implementation that generates DocBook 5 output
   # similar to the docbook45 backend from AsciiDoc Python, but migrated to the
diff --git a/lib/asciidoctor/converter/factory.rb b/lib/asciidoctor/converter/factory.rb
index c557ac0f..d182892d 100644
--- a/lib/asciidoctor/converter/factory.rb
+++ b/lib/asciidoctor/converter/factory.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
   module Converter
     # A factory for instantiating converters that are used to convert a
@@ -51,9 +50,7 @@ module Asciidoctor
             if initialize_singleton
               # FIXME this assignment itself may not be thread safe; may need to use a helper here
               @__default__ ||= begin
-                unless defined? ::Concurrent::Hash
-                  require ::RUBY_MIN_VERSION_1_9 ? 'concurrent/hash' : 'asciidoctor/core_ext/1.8.7/concurrent/hash'
-                end
+                require 'concurrent/hash' unless defined? ::Concurrent::Hash.new
                 new ::Concurrent::Hash.new
               rescue ::LoadError
                 include Logging unless include? Logging
diff --git a/lib/asciidoctor/converter/html5.rb b/lib/asciidoctor/converter/html5.rb
index 2857b67c..8f7a1ae3 100644
--- a/lib/asciidoctor/converter/html5.rb
+++ b/lib/asciidoctor/converter/html5.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
   # A built-in {Converter} implementation that generates HTML 5 output
   # consistent with the html5 backend from AsciiDoc Python.
diff --git a/lib/asciidoctor/converter/template.rb b/lib/asciidoctor/converter/template.rb
index 1619566e..dfed5c35 100644
--- a/lib/asciidoctor/converter/template.rb
+++ b/lib/asciidoctor/converter/template.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
   # A {Converter} implementation that uses templates composed in template
   # languages supported by {https://github.com/rtomayko/tilt Tilt} to convert
@@ -34,9 +33,7 @@ module Asciidoctor
     }
 
     begin
-      unless defined? ::Concurrent::Hash
-        require ::RUBY_MIN_VERSION_1_9 ? 'concurrent/hash' : 'asciidoctor/core_ext/1.8.7/concurrent/hash'
-      end
+      require 'concurrent/hash' unless defined? ::Concurrent::Hash
       @caches = { :scans => ::Concurrent::Hash.new, :templates => ::Concurrent::Hash.new }
     rescue ::LoadError
       @caches = { :scans => {}, :templates => {} }
diff --git a/lib/asciidoctor/core_ext.rb b/lib/asciidoctor/core_ext.rb
index 9edb7b0a..ccf77f29 100644
--- a/lib/asciidoctor/core_ext.rb
+++ b/lib/asciidoctor/core_ext.rb
@@ -1,18 +1,4 @@
+require 'asciidoctor/core_ext/float/truncate'
 require 'asciidoctor/core_ext/nil_or_empty'
 require 'asciidoctor/core_ext/regexp/is_match'
-if RUBY_MIN_VERSION_1_9
-  require 'asciidoctor/core_ext/string/limit_bytesize'
-  if RUBY_ENGINE == 'opal'
-    require 'asciidoctor/core_ext/1.8.7/io/binread'
-    require 'asciidoctor/core_ext/1.8.7/io/write'
-  end
-elsif RUBY_ENGINE != 'opal'
-  require 'asciidoctor/core_ext/1.8.7/base64/strict_encode64'
-  require 'asciidoctor/core_ext/1.8.7/hash/key'
-  require 'asciidoctor/core_ext/1.8.7/io/binread'
-  require 'asciidoctor/core_ext/1.8.7/io/write'
-  require 'asciidoctor/core_ext/1.8.7/string/chr'
-  require 'asciidoctor/core_ext/1.8.7/string/limit_bytesize'
-  require 'asciidoctor/core_ext/1.8.7/symbol/empty'
-  require 'asciidoctor/core_ext/1.8.7/symbol/length'
-end
+require 'asciidoctor/core_ext/string/limit_bytesize'
diff --git a/lib/asciidoctor/core_ext/1.8.7/base64/strict_encode64.rb b/lib/asciidoctor/core_ext/1.8.7/base64/strict_encode64.rb
deleted file mode 100644
index 4ad928e1..00000000
--- a/lib/asciidoctor/core_ext/1.8.7/base64/strict_encode64.rb
+++ /dev/null
@@ -1,6 +0,0 @@
-# Educate Ruby 1.8.7 about the Base64#strict_encode64 method.
-module Base64
-  def strict_encode64 bin
-    (self.encode64 bin).delete %(\n)
-  end
-end
diff --git a/lib/asciidoctor/core_ext/1.8.7/concurrent/hash.rb b/lib/asciidoctor/core_ext/1.8.7/concurrent/hash.rb
deleted file mode 100644
index 7c9b11a4..00000000
--- a/lib/asciidoctor/core_ext/1.8.7/concurrent/hash.rb
+++ /dev/null
@@ -1,5 +0,0 @@
-require 'thread_safe'
-
-module Concurrent
-  Hash = ::ThreadSafe::Cache
-end
diff --git a/lib/asciidoctor/core_ext/1.8.7/hash/key.rb b/lib/asciidoctor/core_ext/1.8.7/hash/key.rb
deleted file mode 100644
index ffe32d02..00000000
--- a/lib/asciidoctor/core_ext/1.8.7/hash/key.rb
+++ /dev/null
@@ -1,4 +0,0 @@
-# Educate Ruby 1.8.7 about the Hash#key method.
-class Hash
-  alias key index
-end
diff --git a/lib/asciidoctor/core_ext/1.8.7/io/binread.rb b/lib/asciidoctor/core_ext/1.8.7/io/binread.rb
deleted file mode 100644
index 40dffa6c..00000000
--- a/lib/asciidoctor/core_ext/1.8.7/io/binread.rb
+++ /dev/null
@@ -1,6 +0,0 @@
-def IO.binread name, length = nil, offset = 0
-  File.open name, 'rb' do |f|
-    f.seek offset unless offset == 0
-    length ? (f.read length) : f.read
-  end
-end unless IO.respond_to? :binread
diff --git a/lib/asciidoctor/core_ext/1.8.7/io/write.rb b/lib/asciidoctor/core_ext/1.8.7/io/write.rb
deleted file mode 100644
index 864fd9c7..00000000
--- a/lib/asciidoctor/core_ext/1.8.7/io/write.rb
+++ /dev/null
@@ -1,5 +0,0 @@
-def IO.write name, string, offset = 0, opts = nil
-  File.open name, 'w' do |f|
-    f.write string
-  end
-end unless IO.respond_to? :write
diff --git a/lib/asciidoctor/core_ext/1.8.7/string/chr.rb b/lib/asciidoctor/core_ext/1.8.7/string/chr.rb
deleted file mode 100644
index 7c349163..00000000
--- a/lib/asciidoctor/core_ext/1.8.7/string/chr.rb
+++ /dev/null
@@ -1,6 +0,0 @@
-# Educate Ruby 1.8.7 about the String#chr method.
-class String
-  def chr
-    slice 0, 1
-  end unless method_defined? :chr
-end
diff --git a/lib/asciidoctor/core_ext/1.8.7/string/limit_bytesize.rb b/lib/asciidoctor/core_ext/1.8.7/string/limit_bytesize.rb
deleted file mode 100644
index 691922ce..00000000
--- a/lib/asciidoctor/core_ext/1.8.7/string/limit_bytesize.rb
+++ /dev/null
@@ -1,29 +0,0 @@
-if RUBY_ENGINE_JRUBY
-  class String
-    # Safely truncate the string to the specified number of bytes.
-    # If a multibyte char gets split, the dangling fragment is removed.
-    def limit_bytesize size
-      return self unless size < bytesize
-      result = (unpack %(a#{size}))[0]
-      begin
-        result.unpack 'U*'
-      rescue ::ArgumentError
-        result.chop!
-        retry
-      end
-      result
-    end unless method_defined? :limit_bytesize
-  end
-else
-  class String
-    ValidTrailingCharRx = /.$/u
-    # Safely truncate the string to the specified number of bytes.
-    # If a multibyte char gets split, the dangling fragment is removed.
-    def limit_bytesize size
-      return self unless size < bytesize
-      result = (unpack %(a#{size}))[0]
-      result.chop! until result.empty? || (ValidTrailingCharRx.match? result)
-      result
-    end unless method_defined? :limit_bytesize
-  end
-end
diff --git a/lib/asciidoctor/core_ext/1.8.7/symbol/empty.rb b/lib/asciidoctor/core_ext/1.8.7/symbol/empty.rb
deleted file mode 100644
index 9a37445b..00000000
--- a/lib/asciidoctor/core_ext/1.8.7/symbol/empty.rb
+++ /dev/null
@@ -1,6 +0,0 @@
-# Educate Ruby 1.8.7 about the Symbol#empty? method.
-class Symbol
-  def empty?
-    to_s.empty?
-  end unless method_defined? :empty?
-end
diff --git a/lib/asciidoctor/core_ext/1.8.7/symbol/length.rb b/lib/asciidoctor/core_ext/1.8.7/symbol/length.rb
deleted file mode 100644
index 372a624e..00000000
--- a/lib/asciidoctor/core_ext/1.8.7/symbol/length.rb
+++ /dev/null
@@ -1,6 +0,0 @@
-# Educate Ruby 1.8.7 about the Symbol#empty? and Symbol#length methods.
-class Symbol
-  def length
-    to_s.length
-  end unless method_defined? :length
-end
diff --git a/lib/asciidoctor/core_ext/float/truncate.rb b/lib/asciidoctor/core_ext/float/truncate.rb
new file mode 100644
index 00000000..256d22d3
--- /dev/null
+++ b/lib/asciidoctor/core_ext/float/truncate.rb
@@ -0,0 +1,19 @@
+if (Float.instance_method :truncate).arity == 0
+  Float.prepend(Module.new do
+    def truncate *args
+      if args.length == 1
+        if (precision = Integer args.shift) == 0
+          super
+        elsif precision > 0
+          precision_factor = 10.0 ** precision
+          (self * precision_factor).to_i / precision_factor
+        else
+          precision_factor = 10 ** precision.abs
+          (self / precision_factor).to_i * precision_factor
+        end
+      else
+        super
+      end
+    end
+  end)
+end
diff --git a/lib/asciidoctor/document.rb b/lib/asciidoctor/document.rb
index 1ae8d7d1..5c548663 100644
--- a/lib/asciidoctor/document.rb
+++ b/lib/asciidoctor/document.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: The Document class represents a parsed AsciiDoc document.
 #
@@ -317,9 +316,8 @@ class Document < AbstractBlock
         # be permissive in case API user wants to define new levels
         @safe = safe_mode
       else
-        # NOTE: not using infix rescue for performance reasons, see https://github.com/jruby/jruby/issues/1816
         begin
-          @safe = SafeMode.value_for_name safe_mode.to_s
+          @safe = SafeMode.value_for_name safe_mode
         rescue
           @safe = SafeMode::SECURE
         end
@@ -504,7 +502,6 @@ class Document < AbstractBlock
         localyear = (attrs['localyear'] ||= now.year.to_s)
       end
       # %Z is OS dependent and may contain characters that aren't UTF-8 encoded (see asciidoctor#2770 and asciidoctor.js#23)
-      # Ruby 1.8 doesn't support %:z
       localtime = (attrs['localtime'] ||= now.strftime %(%T #{now.utc_offset == 0 ? 'UTC' : '%z'}))
       attrs['localdatetime'] ||= %(#{localdate} #{localtime})
 
@@ -522,7 +519,7 @@ class Document < AbstractBlock
       if initialize_extensions
         if (ext_registry = options[:extension_registry])
           # QUESTION should we warn if the value type of this option is not a registry
-          if Extensions::Registry === ext_registry || (::RUBY_ENGINE_JRUBY &&
+          if Extensions::Registry === ext_registry || ((defined? ::AsciidoctorJ::Extensions::ExtensionRegistry) &&
               ::AsciidoctorJ::Extensions::ExtensionRegistry === ext_registry)
             @extensions = ext_registry.activate self
           end
@@ -1152,7 +1149,7 @@ class Document < AbstractBlock
       converter_opts[:safe] = @safe
     end
     if (converter = @options[:converter])
-      converter_factory = Converter::Factory.new ::Hash[backend, converter]
+      converter_factory = Converter::Factory.new backend => converter
     else
       converter_factory = Converter::Factory.default false
     end
@@ -1223,7 +1220,7 @@ class Document < AbstractBlock
           # ensure there's a trailing endline
           target.write LF
         end
-      elsif COERCE_ENCODING
+      elsif FORCE_ENCODING
         ::IO.write target, output, :encoding => ::Encoding::UTF_8
       else
         ::IO.write target, output
diff --git a/lib/asciidoctor/extensions.rb b/lib/asciidoctor/extensions.rb
index 4c0bed86..70347d9c 100644
--- a/lib/asciidoctor/extensions.rb
+++ b/lib/asciidoctor/extensions.rb
@@ -1,7 +1,6 @@
 # NOTE .to_s hides require from Opal
 require 'asciidoctor'.to_s unless defined? Asciidoctor.load
 
-# encoding: UTF-8
 module Asciidoctor
 # Extensions provide a way to participate in the parsing and converting
 # phases of the AsciiDoc processor or extend the AsciiDoc syntax.
@@ -66,14 +65,12 @@ module Extensions
       #
       # Returns nothing
       def use_dsl
-        if self.name.nil_or_empty?
-          # NOTE contants(false) doesn't exist in Ruby 1.8.7
-          #include const_get :DSL if constants(false).include? :DSL
-          include const_get :DSL if constants.include? :DSL
-        else
-          # NOTE contants(false) doesn't exist in Ruby 1.8.7
-          #extend const_get :DSL if constants(false).include? :DSL
-          extend const_get :DSL if constants.include? :DSL
+        if constants.include? :DSL
+          if self.name.nil_or_empty?
+            include const_get :DSL
+          else
+            extend const_get :DSL
+          end
         end
       end
       alias extend_dsl use_dsl
@@ -1331,7 +1328,6 @@ module Extensions
         #class << processor
         #  include_dsl
         #end
-        # NOTE kind_class.contants(false) doesn't exist in Ruby 1.8.7
         processor.extend kind_class.const_get :DSL if kind_class.constants.include? :DSL
         processor.instance_exec(&block)
         processor.freeze
@@ -1377,7 +1373,6 @@ module Extensions
         #class << processor
         #  include_dsl
         #end
-        # NOTE kind_class.contants(false) doesn't exist in Ruby 1.8.7
         processor.extend kind_class.const_get :DSL if kind_class.constants.include? :DSL
         if block.arity == 1
           yield processor
@@ -1548,35 +1543,12 @@ module Extensions
     # Public: Resolves the Class object for the qualified name.
     #
     # Returns Class
-    if ::RUBY_MIN_VERSION_2
-      def class_for_name qualified_name
-        resolved = ::Object.const_get qualified_name, false
-        raise unless ::Class === resolved
-        resolved
-      rescue
-        raise ::NameError, %(Could not resolve class for name: #{qualified_name})
-      end
-    elsif ::RUBY_MIN_VERSION_1_9
-      def class_for_name qualified_name
-        resolved = (qualified_name.split '::').reduce ::Object do |current, name|
-          name.empty? ? current : (current.const_get name, false)
-        end
-        raise unless ::Class === resolved
-        resolved
-      rescue
-        raise ::NameError, %(Could not resolve class for name: #{qualified_name})
-      end
-    else
-      def class_for_name qualified_name
-        resolved = (qualified_name.split '::').reduce ::Object do |current, name|
-          # NOTE on Ruby 1.8, const_defined? only checks for constant in current scope
-          name.empty? ? current : ((current.const_defined? name) ? (current.const_get name) : raise)
-        end
-        raise unless ::Class === resolved
-        resolved
-      rescue
-        raise ::NameError, %(Could not resolve class for name: #{qualified_name})
-      end
+    def class_for_name qualified_name
+      resolved = ::Object.const_get qualified_name, false
+      raise unless ::Class === resolved
+      resolved
+    rescue
+      raise ::NameError, %(Could not resolve class for name: #{qualified_name})
     end
   end
 end
diff --git a/lib/asciidoctor/helpers.rb b/lib/asciidoctor/helpers.rb
index b62645df..b4ac9006 100644
--- a/lib/asciidoctor/helpers.rb
+++ b/lib/asciidoctor/helpers.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 module Helpers
   # Internal: Require the specified library using Kernel#require.
@@ -65,27 +64,18 @@ module Helpers
   # returns a String Array of normalized lines
   def self.normalize_lines_array data
     return data if data.empty?
-
-    leading_bytes = (first_line = data[0]).unpack 'C3'
-    if COERCE_ENCODING
-      utf8 = ::Encoding::UTF_8
-      if (leading_2_bytes = leading_bytes.slice 0, 2) == BOM_BYTES_UTF_16LE
-        # HACK Ruby messes up trailing whitespace on UTF-16LE, so reencode whole document first
-        data = data.join
-        return (((data.force_encoding ::Encoding::UTF_16LE).slice 1, data.length).encode utf8).each_line.map {|line| line.rstrip }
-      elsif leading_2_bytes == BOM_BYTES_UTF_16BE
-        data[0] = (first_line.force_encoding ::Encoding::UTF_16BE).slice 1, first_line.length
-        return data.map {|line| ((line.force_encoding ::Encoding::UTF_16BE).encode utf8).rstrip }
-      elsif leading_bytes == BOM_BYTES_UTF_8
-        data[0] = (first_line.force_encoding utf8).slice 1, first_line.length
-      end
-
-      data.map {|line| line.encoding == utf8 ? line.rstrip : (line.force_encoding utf8).rstrip }
-    else
-      # Ruby 1.8 has no built-in re-encoding, so no point in removing the UTF-16 BOMs
-      data[0] = first_line.slice 3, first_line.length if leading_bytes == BOM_BYTES_UTF_8
-      data.map {|line| line.rstrip }
+    utf8 = ::Encoding::UTF_8
+    if (leading_2_bytes = (leading_bytes = (first_line = data[0]).unpack 'C3').slice 0, 2) == BOM_BYTES_UTF_16LE
+      # HACK Ruby messes up trailing whitespace on UTF-16LE, so reencode whole document first
+      data = data.join
+      return (((data.force_encoding ::Encoding::UTF_16LE).slice 1, data.length).encode utf8).lines.map {|line| line.rstrip }
+    elsif leading_2_bytes == BOM_BYTES_UTF_16BE
+      data[0] = (first_line.force_encoding ::Encoding::UTF_16BE).slice 1, first_line.length
+      return data.map {|line| ((line.force_encoding ::Encoding::UTF_16BE).encode utf8).rstrip }
+    elsif leading_bytes == BOM_BYTES_UTF_8
+      data[0] = (first_line.force_encoding utf8).slice 1, first_line.length
     end
+    data.map {|line| line.encoding == utf8 ? line.rstrip : (line.force_encoding utf8).rstrip }
   end
 
   # Public: Normalize the String and split into lines to prepare them for parsing
@@ -101,24 +91,17 @@ module Helpers
   # returns a String Array of normalized lines
   def self.normalize_lines_from_string data
     return [] if data.nil_or_empty?
-
-    leading_bytes = data.unpack 'C3'
-    if COERCE_ENCODING
-      utf8 = ::Encoding::UTF_8
-      if (leading_2_bytes = leading_bytes.slice 0, 2) == BOM_BYTES_UTF_16LE
-        data = ((data.force_encoding ::Encoding::UTF_16LE).slice 1, data.length).encode utf8
-      elsif leading_2_bytes == BOM_BYTES_UTF_16BE
-        data = ((data.force_encoding ::Encoding::UTF_16BE).slice 1, data.length).encode utf8
-      elsif leading_bytes == BOM_BYTES_UTF_8
-        data = data.encoding == utf8 ? (data.slice 1, data.length) : ((data.force_encoding utf8).slice 1, data.length)
-      else
-        data = data.force_encoding utf8 unless data.encoding == utf8
-      end
+    utf8 = ::Encoding::UTF_8
+    if (leading_2_bytes = (leading_bytes = data.unpack 'C3').slice 0, 2) == BOM_BYTES_UTF_16LE
+      data = ((data.force_encoding ::Encoding::UTF_16LE).slice 1, data.length).encode utf8
+    elsif leading_2_bytes == BOM_BYTES_UTF_16BE
+      data = ((data.force_encoding ::Encoding::UTF_16BE).slice 1, data.length).encode utf8
+    elsif leading_bytes == BOM_BYTES_UTF_8
+      data = data.encoding == utf8 ? (data.slice 1, data.length) : ((data.force_encoding utf8).slice 1, data.length)
     else
-      # Ruby 1.8 has no built-in re-encoding, so no point in removing the UTF-16 BOMs
-      data = data.slice 3, data.length if leading_bytes == BOM_BYTES_UTF_8
+      data = data.force_encoding utf8 unless data.encoding == utf8
     end
-    data.each_line.map {|line| line.rstrip }
+    data.lines.map {|line| line.rstrip }
   end
 
   # Public: Efficiently checks whether the specified String resembles a URI
diff --git a/lib/asciidoctor/inline.rb b/lib/asciidoctor/inline.rb
index 13082570..c0f0e6fe 100644
--- a/lib/asciidoctor/inline.rb
+++ b/lib/asciidoctor/inline.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: Methods for managing inline elements in AsciiDoc block
 class Inline < AbstractNode
diff --git a/lib/asciidoctor/list.rb b/lib/asciidoctor/list.rb
index 1e98e883..bac283c6 100644
--- a/lib/asciidoctor/list.rb
+++ b/lib/asciidoctor/list.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: Methods for managing AsciiDoc lists (ordered, unordered and description lists)
 class List < AbstractBlock
diff --git a/lib/asciidoctor/logging.rb b/lib/asciidoctor/logging.rb
index 3d9efadc..ded8cf9e 100644
--- a/lib/asciidoctor/logging.rb
+++ b/lib/asciidoctor/logging.rb
@@ -34,8 +34,7 @@ class Logger < ::Logger
 end
 
 class MemoryLogger < ::Logger
-  # NOTE Ruby 1.8.7 returns constants as strings instead of symbols
-  SEVERITY_LABELS = ::Hash[Severity.constants.map {|c| [(Severity.const_get c), c.to_sym] }]
+  SEVERITY_LABELS = ::Hash[Severity.constants.map {|c| [(Severity.const_get c), c] }]
 
   attr_reader :messages
 
@@ -95,7 +94,7 @@ module LoggerManager
     def memoize_logger
       class << self
         alias_method :logger, :logger
-        if RUBY_ENGINE == 'opal'
+        if RUBY_ENGINE_OPAL
           define_method :logger do @logger end
         else
           attr_reader :logger
diff --git a/lib/asciidoctor/parser.rb b/lib/asciidoctor/parser.rb
index 0a4a0579..d65b0e33 100644
--- a/lib/asciidoctor/parser.rb
+++ b/lib/asciidoctor/parser.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: Methods to parse lines of AsciiDoc into an object hierarchy
 # representing the structure of the document. All methods are class methods and
@@ -1115,9 +1114,7 @@ class Parser
     found = false
     autonum = 0
     text.scan(CalloutScanRx) {
-      # lead with assignments for Ruby 1.8.7 compat
-      captured, num = $&, $2
-      document.callouts.register num == '.' ? (autonum += 1).to_s : num unless captured.start_with? '\\'
+      document.callouts.register $2 == '.' ? (autonum += 1).to_s : $2 unless $&.start_with? '\\'
       # we have to mark as found even if it's escaped so it can be unescaped
       found = true
     } if text.include? '<'
@@ -1152,8 +1149,6 @@ class Parser
   # Returns nothing
   def self.catalog_inline_anchors text, block, document, reader
     text.scan(InlineAnchorScanRx) do
-      # alias match for Ruby 1.8.7 compat
-      m = $~
       if (id = $1)
         if (reftext = $2)
           next if (reftext.include? ATTR_REF_HEAD) && (reftext = document.sub_attributes reftext).empty?
@@ -1167,7 +1162,7 @@ class Parser
       end
       unless document.register :refs, [id, (Inline.new block, :anchor, reftext, :type => :ref, :id => id), reftext]
         location = reader.cursor_at_mark
-        if (offset = (m.pre_match.count LF) + ((m[0].start_with? LF) ? 1 : 0)) > 0
+        if (offset = ($`.count LF) + (($&.start_with? LF) ? 1 : 0)) > 0
           (location = location.dup).advance offset
         end
         logger.warn message_with_context %(id assigned to anchor already in use: #{id}), :source_location => location
@@ -1704,7 +1699,7 @@ class Parser
   def self.setext_section_title? line1, line2
     if (level = SETEXT_SECTION_LEVELS[line2_ch1 = line2.chr]) &&
         line2_ch1 * (line2_len = line2.length) == line2 && SetextSectionTitleRx.match?(line1) &&
-        (line_length(line1) - line2_len).abs < 2
+        (line1.length - line2_len).abs < 2
       level
     end
   end
@@ -1766,7 +1761,7 @@ class Parser
     elsif Compliance.underline_style_section_titles && (line2 = reader.peek_line(true)) &&
         (sect_level = SETEXT_SECTION_LEVELS[line2_ch1 = line2.chr]) &&
         line2_ch1 * (line2_len = line2.length) == line2 && (sect_title = SetextSectionTitleRx =~ line1 && $1) &&
-        (line_length(line1) - line2_len).abs < 2
+        (line1.length - line2_len).abs < 2
       atx = false
       if sect_title.end_with?(']]') && InlineSectionAnchorRx =~ sect_title && !$1 # escaped
         sect_title, sect_id, sect_reftext = (sect_title.slice 0, sect_title.length - $&.length), $2, $3
@@ -1779,21 +1774,6 @@ class Parser
     [sect_id, sect_reftext, sect_title, sect_level, atx]
   end
 
-  # Public: Calculate the number of unicode characters in the line, excluding the endline
-  #
-  # line - the String to calculate
-  #
-  # returns the number of unicode characters in the line
-  if FORCE_UNICODE_LINE_LENGTH
-    def self.line_length(line)
-      line.scan(UnicodeCharScanRx).size
-    end
-  else
-    def self.line_length(line)
-      line.length
-    end
-  end
-
   # Public: Consume and parse the two header lines (line 1 = author info, line 2 = revision info).
   #
   # Returns the Hash of header metadata. If a Document object is supplied, the metadata
diff --git a/lib/asciidoctor/path_resolver.rb b/lib/asciidoctor/path_resolver.rb
index ea8eaacb..b3c27a39 100644
--- a/lib/asciidoctor/path_resolver.rb
+++ b/lib/asciidoctor/path_resolver.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: Handles all operations for resolving, cleaning and joining paths.
 # This class includes operations for handling both web paths (request URIs) and
diff --git a/lib/asciidoctor/reader.rb b/lib/asciidoctor/reader.rb
index a3f3a11f..8036ac20 100644
--- a/lib/asciidoctor/reader.rb
+++ b/lib/asciidoctor/reader.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: Methods for retrieving lines from AsciiDoc source files
 class Reader
@@ -807,16 +806,14 @@ class PreprocessorReader < Reader
         # don't honor match if it doesn't meet this criteria
         return false unless no_target && EvalExpressionRx =~ text.strip
 
-        # NOTE save values eagerly for Ruby 1.8.7 compat
-        lhs, op, rhs = $1, $2, $3
-        lhs = resolve_expr_val lhs
-        rhs = resolve_expr_val rhs
+        lhs = resolve_expr_val $1
+        rhs = resolve_expr_val $3
 
         # regex enforces a restricted set of math-related operations
-        if op == '!='
+        if $2 == '!='
           skip = lhs.send :==, rhs
         else
-          skip = !(lhs.send op.to_sym, rhs)
+          skip = !(lhs.send $2.to_sym, rhs)
         end
       end
     end
@@ -972,11 +969,11 @@ class PreprocessorReader < Reader
         begin
           open(inc_path, 'rb') do |f|
             dbl_co, dbl_sb = '::', '[]'
-            encoding = ::Encoding::UTF_8 if COERCE_ENCODING
+            utf8 = ::Encoding::UTF_8
             f.each_line do |l|
               inc_lineno += 1
               # must force encoding since we're performing String operations on line
-              l.force_encoding encoding if encoding
+              l.force_encoding utf8
               if (l.include? dbl_co) && (l.include? dbl_sb) && TagDirectiveRx =~ l
                 this_tag = $2
                 if $1 # end tag
@@ -1016,7 +1013,7 @@ class PreprocessorReader < Reader
             logger.warn message_with_context %(detected unclosed tag '#{tag_name}' starting at line #{tag_lineno} of include #{target_type}: #{inc_path}), :source_location => cursor, :include_location => (create_include_cursor inc_path, expanded_target, tag_lineno)
           end
         end
-        unless (missing_tags = inc_tags.keys.to_a - tags_used.to_a).empty?
+        unless (missing_tags = inc_tags.keys - tags_used.to_a).empty?
           logger.warn message_with_context %(tag#{missing_tags.size > 1 ? 's' : ''} '#{missing_tags.join ', '}' not found in include #{target_type}: #{inc_path}), :source_location => cursor
         end
         shift
@@ -1066,7 +1063,7 @@ class PreprocessorReader < Reader
         # caching requires the open-uri-cached gem to be installed
         # processing will be automatically aborted if these libraries can't be opened
         Helpers.require_library 'open-uri/cached', 'open-uri-cached' unless defined? ::OpenURI::Cache
-      elsif !::RUBY_ENGINE_OPAL
+      elsif !RUBY_ENGINE_OPAL
         # autoload open-uri
         ::OpenURI
       end
@@ -1110,7 +1107,7 @@ class PreprocessorReader < Reader
       # NOTE if file is not a string, assume it's a URI
       if ::String === file
         @dir = ::File.dirname file
-      elsif ::RUBY_ENGINE_OPAL
+      elsif RUBY_ENGINE_OPAL
         @dir = ::URI.parse ::File.dirname(file = file.to_s)
       else
         # NOTE this intentionally throws an error if URI has no path
@@ -1170,7 +1167,7 @@ class PreprocessorReader < Reader
   def create_include_cursor file, path, lineno
     if ::String === file
       dir = ::File.dirname file
-    elsif ::RUBY_ENGINE_OPAL
+    elsif RUBY_ENGINE_OPAL
       dir = ::File.dirname(file = file.to_s)
     else
       dir = (dir = ::File.dirname file.path) == '' ? '/' : dir
diff --git a/lib/asciidoctor/section.rb b/lib/asciidoctor/section.rb
index b5da4826..ebf2d752 100644
--- a/lib/asciidoctor/section.rb
+++ b/lib/asciidoctor/section.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: Methods for managing sections of AsciiDoc content in a document.
 # The section responds as an Array of content blocks by delegating
diff --git a/lib/asciidoctor/stylesheets.rb b/lib/asciidoctor/stylesheets.rb
index 482e9900..03e039c4 100644
--- a/lib/asciidoctor/stylesheets.rb
+++ b/lib/asciidoctor/stylesheets.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # A utility class for working with the built-in stylesheets.
 #--
diff --git a/lib/asciidoctor/substitutors.rb b/lib/asciidoctor/substitutors.rb
index 02e88f4f..7f642941 100644
--- a/lib/asciidoctor/substitutors.rb
+++ b/lib/asciidoctor/substitutors.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: Methods to perform substitutions on lines of AsciiDoc text. This module
 # is intented to be mixed-in to Section and Block to provide operations for performing
@@ -43,24 +42,17 @@ module Substitutors
 
   SUB_HIGHLIGHT = ['coderay', 'pygments']
 
-  if ::RUBY_MIN_VERSION_1_9
-    CAN = %(\u0018)
-    DEL = %(\u007f)
+  CAN = %(\u0018)
+  DEL = %(\u007f)
 
-    # Delimiters and matchers for the passthrough placeholder
-    # See http://www.aivosto.com/vbtips/control-characters.html#listabout for characters to use
+  # Delimiters and matchers for the passthrough placeholder
+  # See http://www.aivosto.com/vbtips/control-characters.html#listabout for characters to use
 
-    # SPA, start of guarded protected area (\u0096)
-    PASS_START = %(\u0096)
+  # SPA, start of guarded protected area (\u0096)
+  PASS_START = %(\u0096)
 
-    # EPA, end of guarded protected area (\u0097)
-    PASS_END = %(\u0097)
-  else
-    CAN = 24.chr
-    DEL = 127.chr
-    PASS_START = 150.chr
-    PASS_END = 151.chr
-  end
+  # EPA, end of guarded protected area (\u0097)
+  PASS_END = %(\u0097)
 
   # match passthrough slot
   PassSlotRx = /#{PASS_START}(\d+)#{PASS_END}/
@@ -177,28 +169,24 @@ module Substitutors
     compat_mode = @document.compat_mode
     passes = @passthroughs
     text = text.gsub(InlinePassMacroRx) {
-      # alias match for Ruby 1.8.7 compat
-      m = $~
       preceding = nil
 
-      if (boundary = m[4]) # $$, ++, or +++
+      if (boundary = $4) # $$, ++, or +++
         # skip ++ in compat mode, handled as normal quoted text
         if compat_mode && boundary == '++'
-          next m[2] ?
-              %(#{m[1]}[#{m[2]}]#{m[3]}++#{extract_passthroughs m[5]}++) :
-              %(#{m[1]}#{m[3]}++#{extract_passthroughs m[5]}++)
+          next $2 ? %(#{$1}[#{$2}]#{$3}++#{extract_passthroughs $5}++) : %(#{$1}#{$3}++#{extract_passthroughs $5}++)
         end
 
-        attributes = m[2]
-        escape_count = m[3].length
-        content = m[5]
+        attributes = $2
+        escape_count = $3.length
+        content = $5
         old_behavior = false
 
         if attributes
           if escape_count > 0
             # NOTE we don't look for nested unconstrained pass macros
-            next %(#{m[1]}[#{attributes}]#{RS * (escape_count - 1)}#{boundary}#{m[5]}#{boundary})
-          elsif m[1] == RS
+            next %(#{$1}[#{attributes}]#{RS * (escape_count - 1)}#{boundary}#{$5}#{boundary})
+          elsif $1 == RS
             preceding = %([#{attributes}])
             attributes = nil
           else
@@ -210,7 +198,7 @@ module Substitutors
           end
         elsif escape_count > 0
           # NOTE we don't look for nested unconstrained pass macros
-          next %(#{RS * (escape_count - 1)}#{boundary}#{m[5]}#{boundary})
+          next %(#{RS * (escape_count - 1)}#{boundary}#{$5}#{boundary})
         end
         subs = (boundary == '+++' ? [] : BASIC_SUBS)
 
@@ -225,12 +213,12 @@ module Substitutors
           passes[pass_key] = {:text => content, :subs => subs}
         end
       else # pass:[]
-        if m[6] == RS
+        if $6 == RS
           # NOTE we don't look for nested pass:[] macros
-          next m[0].slice 1, m[0].length
+          next $&.slice 1, $&.length
         end
 
-        passes[pass_key = passes.size] = {:text => (unescape_brackets m[8]), :subs => (m[7] ? (resolve_pass_subs m[7]) : nil)}
+        passes[pass_key = passes.size] = {:text => (unescape_brackets $8), :subs => ($7 ? (resolve_pass_subs $7) : nil)}
       end
 
       %(#{preceding}#{PASS_START}#{pass_key}#{PASS_END})
@@ -238,13 +226,11 @@ module Substitutors
 
     pass_inline_char1, pass_inline_char2, pass_inline_rx = InlinePassRx[compat_mode]
     text = text.gsub(pass_inline_rx) {
-      # alias match for Ruby 1.8.7 compat
-      m = $~
-      preceding = m[1]
-      attributes = m[2]
-      escape_mark = RS if (quoted_text = m[3]).start_with? RS
-      format_mark = m[4]
-      content = m[5]
+      preceding = $1
+      attributes = $2
+      escape_mark = RS if (quoted_text = $3).start_with? RS
+      format_mark = $4
+      content = $5
 
       if compat_mode
         old_behavior = true
@@ -297,18 +283,16 @@ module Substitutors
 
     # NOTE we need to do the stem in a subsequent step to allow it to be escaped by the former
     text = text.gsub(InlineStemMacroRx) {
-      # alias match for Ruby 1.8.7 compat
-      m = $~
       # honor the escape
       if $&.start_with? RS
-        next m[0].slice 1, m[0].length
+        next $&.slice 1, $&.length
       end
 
-      if (type = m[1].to_sym) == :stem
+      if (type = $1.to_sym) == :stem
         type = STEM_TYPE_ALIASES[@document.attributes['stem']].to_sym
       end
-      content = unescape_brackets m[3]
-      subs = m[2] ? (resolve_pass_subs m[2]) : ((@document.basebackend? 'html') ? BASIC_SUBS : nil)
+      content = unescape_brackets $3
+      subs = $2 ? (resolve_pass_subs $2) : ((@document.basebackend? 'html') ? BASIC_SUBS : nil)
       passes[pass_key = passes.size] = {:text => content, :subs => subs, :type => type}
       %(#{PASS_START}#{pass_key}#{PASS_END})
     } if (text.include? ':') && ((text.include? 'stem:') || (text.include? 'math:'))
@@ -420,14 +404,8 @@ module Substitutors
   # text - The String text to process.
   #
   # returns The String text with special characters replaced.
-  if ::RUBY_MIN_VERSION_1_9
-    def sub_specialchars text
-      (text.include? '<') || (text.include? '&') || (text.include? '>') ? (text.gsub SpecialCharsRx, SpecialCharsTr) : text
-    end
-  else
-    def sub_specialchars text
-      (text.include? '<') || (text.include? '&') || (text.include? '>') ? (text.gsub(SpecialCharsRx) { SpecialCharsTr[$&] }) : text
-    end
+  def sub_specialchars text
+    (text.include? '<') || (text.include? '&') || (text.include? '>') ? (text.gsub SpecialCharsRx, SpecialCharsTr) : text
   end
   alias sub_specialcharacters sub_specialchars
 
@@ -573,16 +551,13 @@ module Substitutors
 
       if found_macroish && (text.include? 'menu:')
         text = text.gsub(InlineMenuMacroRx) {
-          # alias match for Ruby 1.8.7 compat
-          m = $~
           # honor the escape
           if $&.start_with? RS
-            next m[0].slice 1, m[0].length
+            next $&.slice 1, $&.length
           end
 
-          menu, items = m[1], m[2]
-
-          if items
+          menu = $1
+          if (items = $2)
             items = items.gsub ESC_R_SB, R_SB if items.include? R_SB
             if (delim = items.include?('&gt;') ? '&gt;' : (items.include?(',') ? ',' : nil))
               submenus = items.split(delim).map {|it| it.strip }
@@ -600,16 +575,12 @@ module Substitutors
 
       if (text.include? '"') && (text.include? '&gt;')
         text = text.gsub(InlineMenuRx) {
-          # alias match for Ruby 1.8.7 compat
-          m = $~
           # honor the escape
           if $&.start_with? RS
-            next m[0].slice 1, m[0].length
+            next $&.slice 1, $&.length
           end
 
-          input = m[1]
-
-          menu, *submenus = input.split('&gt;').map {|it| it.strip }
+          menu, *submenus = $1.split('&gt;').map {|it| it.strip }
           menuitem = submenus.pop
           Inline.new(self, :menu, nil, :attributes => {'menu' => menu, 'submenus' => submenus, 'menuitem' => menuitem}).convert
         }
@@ -621,17 +592,15 @@ module Substitutors
     if (extensions = doc.extensions) && extensions.inline_macros? # && found_macroish
       extensions.inline_macros.each do |extension|
         text = text.gsub(extension.instance.regexp) {
-          # alias match for Ruby 1.8.7 compat
-          m = $~
           # honor the escape
           if $&.start_with? RS
-            next m[0].slice 1, m[0].length
+            next $&.slice 1, $&.length
           end
 
-          if (m.names rescue []).empty?
-            target, content, extconf = m[1], m[2], extension.config
+          if ($~.names rescue []).empty?
+            target, content, extconf = $1, $2, extension.config
           else
-            target, content, extconf = (m[:target] rescue nil), (m[:content] rescue nil), extension.config
+            target, content, extconf = ($~[:target] rescue nil), ($~[:content] rescue nil), extension.config
           end
           attributes = (attributes = extconf[:default_attrs]) ? attributes.dup : {}
           if content.nil_or_empty?
@@ -656,21 +625,19 @@ module Substitutors
     if found_macroish && ((text.include? 'image:') || (text.include? 'icon:'))
       # image:filename.png[Alt Text]
       text = text.gsub(InlineImageMacroRx) {
-        # alias match for Ruby 1.8.7 compat
-        m = $~
         # honor the escape
-        if (captured = $&).start_with? RS
-          next captured.slice 1, captured.length
-        elsif captured.start_with? 'icon:'
+        if $&.start_with? RS
+          next $&.slice 1, $&.length
+        elsif $&.start_with? 'icon:'
           type, posattrs = 'icon', ['size']
         else
           type, posattrs = 'image', ['alt', 'width', 'height']
         end
-        if (target = m[1]).include? ATTR_REF_HEAD
+        if (target = $1).include? ATTR_REF_HEAD
           # TODO remove this special case once titles use normal substitution order
           target = sub_attributes target
         end
-        attrs = parse_attributes m[2], posattrs, :unescape_input => true
+        attrs = parse_attributes $2, posattrs, :unescape_input => true
         doc.register :images, [target, (attrs['imagesdir'] = doc_attrs['imagesdir'])] unless type == 'icon'
         attrs['alt'] ||= (attrs['default-alt'] = Helpers.basename(target, true).tr('_-', ' '))
         Inline.new(self, :image, nil, :type => type, :target => target, :attributes => attrs).convert
@@ -683,13 +650,12 @@ module Substitutors
       # ((Tigers))
       # indexterm2:[Tigers]
       text = text.gsub(InlineIndextermMacroRx) {
-        captured = $&
         case $1
         when 'indexterm'
           text = $2
           # honor the escape
-          if captured.start_with? RS
-            next captured.slice 1, captured.length
+          if $&.start_with? RS
+            next $&.slice 1, $&.length
           end
           # indexterm:[Tigers,Big cats]
           terms = split_simple_csv normalize_string text, true
@@ -698,8 +664,8 @@ module Substitutors
         when 'indexterm2'
           text = $2
           # honor the escape
-          if captured.start_with? RS
-            next captured.slice 1, captured.length
+          if $&.start_with? RS
+            next $&.slice 1, $&.length
           end
           # indexterm2:[Tigers]
           term = normalize_string text, true
@@ -708,13 +674,13 @@ module Substitutors
         else
           text = $3
           # honor the escape
-          if captured.start_with? RS
+          if $&.start_with? RS
             # escape concealed index term, but process nested flow index term
             if (text.start_with? '(') && (text.end_with? ')')
               text = text.slice 1, text.length - 2
               visible, before, after = true, '(', ')'
             else
-              next captured.slice 1, captured.length
+              next $&.slice 1, $&.length
             end
           else
             visible = true
@@ -747,21 +713,19 @@ module Substitutors
     if found_colon && (text.include? '://')
       # inline urls, target[text] (optionally prefixed with link: and optionally surrounded by <>)
       text = text.gsub(InlineLinkRx) {
-        # alias match for Ruby 1.8.7 compat
-        m = $~
         # honor the escape
         if (target = $2).start_with? RS
-          next %(#{m[1]}#{target.slice 1, target.length}#{m[3]})
+          next %(#{$1}#{target.slice 1, target.length}#{$3})
         end
         # NOTE if text is non-nil, then we've matched a formal macro (i.e., trailing square brackets)
-        prefix, text, suffix = m[1], (macro = m[3]) || '', ''
+        captured, prefix, text, suffix = $&, $1, (macro = $3) || '', ''
         if prefix == 'link:'
           if macro
             prefix = ''
           else
             # invalid macro syntax (link: prefix w/o trailing square brackets)
             # we probably shouldn't even get here...our regex is doing too much
-            next m[0]
+            next captured
           end
         end
         unless macro || UriTerminatorRx !~ target
@@ -794,7 +758,7 @@ module Substitutors
             end
           end
           # NOTE handle case when remaining target is a URI scheme (e.g., http://)
-          return m[0] if target.end_with? '://'
+          return captured if target.end_with? '://'
         end
 
         attrs, link_opts = nil, { :type => :link }
@@ -843,15 +807,18 @@ module Substitutors
     if found_macroish && ((text.include? 'link:') || (text.include? 'mailto:'))
       # inline link macros, link:target[text]
       text = text.gsub(InlineLinkMacroRx) {
-        # alias match for Ruby 1.8.7 compat
-        m = $~
         # honor the escape
         if $&.start_with? RS
-          next m[0].slice 1, m[0].length
+          next $&.slice 1, $&.length
+        end
+        if (mailto = $1)
+          target = %(mailto:#{$2})
+          mailto_text = $2
+        else
+          target = $2
         end
-        target = (mailto = m[1]) ? %(mailto:#{m[2]}) : m[2]
         attrs, link_opts = nil, { :type => :link }
-        unless (text = m[3]).empty?
+        unless (text = $3).empty?
           text = text.gsub ESC_R_SB, R_SB if text.include? R_SB
           if mailto
             if !doc.compat_mode && (text.include? ',')
@@ -892,7 +859,7 @@ module Substitutors
         if text.empty?
           # mailto is a special case, already processed
           if mailto
-            text = m[2]
+            text = mailto_text
           else
             if doc_attrs.key? 'hide-uri-scheme'
               if (text = target.sub UriSniffRx, '').empty?
@@ -918,31 +885,28 @@ module Substitutors
 
     if text.include? '@'
       text = text.gsub(InlineEmailRx) {
-        address, tip = $&, $1
-        if tip
-          next (tip == RS ? (address.slice 1, address.length) : address)
+        if $1
+          next ($1 == RS ? ($&.slice 1, $&.length) : $&)
         end
 
-        target = %(mailto:#{address})
+        target = %(mailto:#{$&})
         # QUESTION should this be registered as an e-mail address?
         doc.register(:links, target)
 
-        Inline.new(self, :anchor, address, :type => :link, :target => target).convert
+        Inline.new(self, :anchor, $&, :type => :link, :target => target).convert
       }
     end
 
     if found_macroish && (text.include? 'tnote')
       text = text.gsub(InlineFootnoteMacroRx) {
-        # alias match for Ruby 1.8.7 compat
-        m = $~
-        # honor the escape
         if $&.start_with? RS
-          next m[0].slice 1, m[0].length
+          next $&.slice 1, $&.length
         end
-        if m[1] # footnoteref (legacy)
-          id, text = (m[3] || '').split(',', 2)
+        if $1 # footnoteref (legacy)
+          id, text = ($3 || '').split(',', 2)
         else
-          id, text = m[2], m[3]
+          id = $2
+          text = $3
         end
         if id
           if text
@@ -967,7 +931,7 @@ module Substitutors
           doc.register(:footnotes, Document::Footnote.new(index, id, text))
           type = target = nil
         else
-          next m[0]
+          next $&
         end
         Inline.new(self, :footnote, text, :attributes => {'index' => index}, :id => id, :target => target, :type => type).convert
       }
@@ -1011,20 +975,18 @@ module Substitutors
   def sub_inline_xrefs(content, found = nil)
     if ((found ? found[:macroish] : (content.include? '[')) && (content.include? 'xref:')) || ((content.include? '&') && (content.include? 'lt;&'))
       content = content.gsub(InlineXrefMacroRx) {
-        # alias match for Ruby 1.8.7 compat
-        m = $~
         # honor the escape
         if $&.start_with? RS
-          next m[0].slice 1, m[0].length
+          next $&.slice 1, $&.length
         end
         attrs, doc = {}, @document
-        if (refid = m[1])
+        if (refid = $1)
           refid, text = refid.split ',', 2
           text = text.lstrip if text
         else
           macro = true
-          refid = m[2]
-          if (text = m[3])
+          refid = $2
+          if (text = $3)
             text = text.gsub ESC_R_SB, R_SB if text.include? R_SB
             # NOTE if an equal sign (=) is present, parse text as attributes
             text = ((AttributeList.new text, self).parse_into attrs)[1] if !doc.compat_mode && (text.include? '=')
diff --git a/lib/asciidoctor/table.rb b/lib/asciidoctor/table.rb
index a4b61412..acc9eb89 100644
--- a/lib/asciidoctor/table.rb
+++ b/lib/asciidoctor/table.rb
@@ -1,10 +1,9 @@
-# encoding: UTF-8
 module Asciidoctor
 # Public: Methods and constants for managing AsciiDoc table content in a document.
 # It supports all three of AsciiDoc's table formats: psv, dsv and csv.
 class Table < AbstractBlock
-  # multipler / divisor for tuning precision of calculated result
-  DEFAULT_PRECISION_FACTOR = 10000.0
+  # precision of column widths
+  DEFAULT_PRECISION = 4
 
   # Public: A data object that encapsulates the collection of rows (head, foot, body) for a table
   class Rows
@@ -111,7 +110,7 @@ class Table < AbstractBlock
   #
   # returns nothing
   def assign_column_widths width_base = nil, autowidth_cols = nil
-    pf = DEFAULT_PRECISION_FACTOR
+    precision = DEFAULT_PRECISION
     total_width = col_pcwidth = 0
 
     if width_base
@@ -120,32 +119,22 @@ class Table < AbstractBlock
           autowidth = 0
           logger.warn %(total column width must not exceed 100% when using autowidth columns; got #{width_base}%)
         else
-          autowidth = ((100.0 - width_base) / autowidth_cols.size * pf).to_i / pf
+          autowidth = ((100.0 - width_base) / autowidth_cols.size).truncate precision
           autowidth = autowidth.to_i if autowidth.to_i == autowidth
           width_base = 100
         end
         autowidth_attrs = { 'width' => autowidth, 'autowidth-option' => '' }
         autowidth_cols.each {|col| col.update_attributes autowidth_attrs }
       end
-      @columns.each {|col| total_width += (col_pcwidth = col.assign_width nil, width_base, pf) }
+      @columns.each {|col| total_width += (col_pcwidth = col.assign_width nil, width_base, precision) }
     else
-      col_pcwidth = ((100 * pf / @columns.size).to_i) / pf
-      # or...
-      #col_pcwidth = (100.0 / @columns.size).truncate 4
+      col_pcwidth = (100.0 / @columns.size).truncate precision
       col_pcwidth = col_pcwidth.to_i if col_pcwidth.to_i == col_pcwidth
-      @columns.each {|col| total_width += col.assign_width col_pcwidth }
+      @columns.each {|col| total_width += col.assign_width col_pcwidth, nil, precision }
     end
 
     # donate balance, if any, to final column (using half up rounding)
-    unless total_width == 100
-      @columns[-1].assign_width(((100 - total_width + col_pcwidth) * pf).round / pf)
-      # or (manual half up rounding)...
-      #numerator = (raw_numerator = (100 - total_width + col_pcwidth) * pf).to_i
-      #numerator += 1 if raw_numerator >= numerator + 0.5
-      #@columns[-1].assign_width numerator / pf
-      # or...
-      #@columns[-1].assign_width((100 - total_width + col_pcwidth).round 4)
-    end
+    @columns[-1].assign_width(((100 - total_width + col_pcwidth).round precision), nil, precision) unless total_width == 100
 
     nil
   end
@@ -201,11 +190,9 @@ class Table::Column < AbstractNode
   # This method assigns the colpcwidth and colabswidth attributes.
   #
   # returns the resolved colpcwidth value
-  def assign_width col_pcwidth, width_base = nil, pf = 10000.0
+  def assign_width col_pcwidth, width_base, precision
     if width_base
-      col_pcwidth = ((@attributes['width'].to_f / width_base) * 100 * pf).to_i / pf
-      # or...
-      #col_pcwidth = (@attributes['width'].to_f * 100.0 / width_base).truncate 4
+      col_pcwidth = (@attributes['width'].to_f * 100.0 / width_base).truncate precision
       col_pcwidth = col_pcwidth.to_i if col_pcwidth.to_i == col_pcwidth
     end
     @attributes['colpcwidth'] = col_pcwidth
diff --git a/lib/asciidoctor/timings.rb b/lib/asciidoctor/timings.rb
index 4a93a908..1d53f514 100644
--- a/lib/asciidoctor/timings.rb
+++ b/lib/asciidoctor/timings.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 module Asciidoctor
   class Timings
     def initialize
diff --git a/run-tests.sh b/run-tests.sh
index 1c6e3c65..5c61da1a 100755
--- a/run-tests.sh
+++ b/run-tests.sh
@@ -4,7 +4,7 @@
 # This script will execute against all supported Ruby versions if "all" is the first argument to the script.
 
 if [ "$1" = "all" ]; then
-  rvm 1.8@asciidoctor-dev,jruby@asciidoctor-dev,rbx@asciidoctor-dev,1.9@asciidoctor-dev,2.0@asciidoctor-dev,2.1@asciidoctor-dev "do" ./run-tests.sh
+  rvm 2.3@asciidoctor-dev,2.4@asciidoctor-dev,2.5@asciidoctor-dev,2.6@asciidoctor-dev,jruby-9.1@asciidoctor-dev,jruby-9.2@asciidoctor-dev "do" ./run-tests.sh
 else
   rake test:all > /tmp/asciidoctor-test-results.txt 2>&1; cat /tmp/asciidoctor-test-results.txt
 fi
diff --git a/test/api_test.rb b/test/api_test.rb
index f68831d9..a5e2d30e 100644
--- a/test/api_test.rb
+++ b/test/api_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -24,6 +23,39 @@ context 'API' do
       assert_equal '.asciidoc', doc.attr('docfilesuffix')
     end
 
+    test 'should coerce encoding of file to UTF-8' do
+      old_external = Encoding.default_external
+      old_internal = Encoding.default_internal
+      old_verbose = $VERBOSE
+      begin
+        $VERBOSE = nil # disable warnings since we have to modify constants
+        input_path = fixture_path 'encoding.asciidoc'
+        Encoding.default_external = Encoding.default_internal = Encoding::IBM437
+        output = Asciidoctor.convert_file input_path, :to_file => false, :safe => :safe
+        assert_equal Encoding::UTF_8, output.encoding
+        assert_includes output, 'Romé'
+      ensure
+        Encoding.default_external = old_external
+        Encoding.default_internal = old_internal
+        $VERBOSE = old_verbose
+      end
+    end
+
+    test 'should not load file with unrecognized encoding' do
+      begin
+        tmp_input = Tempfile.new %w(test- .adoc), encoding: Encoding::IBM437
+        # NOTE using a character whose code differs between UTF-8 and IBM437
+        tmp_input.write %(ƒ\n)
+        tmp_input.close
+        exception = assert_raises ArgumentError do
+          Asciidoctor.load_file tmp_input.path, :safe => :safe
+        end
+        assert_match(/Failed to load AsciiDoc document - invalid byte sequence in UTF-8/, exception.message)
+      ensure
+        tmp_input.close!
+      end
+    end
+
     test 'should not load invalid file' do
       sample_input_path = fixture_path('hello-asciidoctor.pdf')
       exception = assert_raises ArgumentError do
@@ -31,8 +63,33 @@ context 'API' do
       end
       assert_match(/Failed to load AsciiDoc document/, exception.message)
       # verify we have the correct backtrace (should be in at least first 5 lines)
-      assert_match((RUBY_ENGINE == 'rbx' ? /parser\.rb/ : /helpers\.rb/), exception.backtrace[0..4].join("\n"))
-    end if RUBY_MIN_VERSION_1_9
+      assert_match(/helpers\.rb/, exception.backtrace[0..4].join("\n"))
+    end
+
+    test 'should convert filename that contains non-ASCII characters independent of default encodings' do
+      old_external = Encoding.default_external
+      old_internal = Encoding.default_internal
+      old_verbose = $VERBOSE
+      begin
+        $VERBOSE = nil # disable warnings since we have to modify constants
+        tmp_input = Tempfile.new %w(test-ＵＴＦ８- .adoc)
+        tmp_input.write %(ＵＴＦ８\n)
+        tmp_input.close
+        tmp_output = tmp_input.path.sub '.adoc', '.html'
+        Asciidoctor.convert_file tmp_input.path, :safe => :safe, :attributes => 'linkcss !copycss'
+        assert File.exist? tmp_output
+        output = File.read tmp_output, mode: 'rb', encoding: 'utf-8:utf-8'
+        assert_equal ::Encoding::UTF_8, output.encoding
+        refute_empty output
+        assert_includes output, 'ＵＴＦ８'
+      ensure
+        tmp_input.close!
+        FileUtils.rm tmp_output
+        Encoding.default_external = old_external
+        Encoding.default_internal = old_internal
+        $VERBOSE = old_verbose
+      end
+    end
 
     test 'should load input IO' do
       input = StringIO.new(<<-EOS)
@@ -67,7 +124,7 @@ Document Title
 
 preamble
       EOS
-      doc = Asciidoctor.load(input.lines.entries, :safe => Asciidoctor::SafeMode::SAFE)
+      doc = Asciidoctor.load(input.lines, :safe => Asciidoctor::SafeMode::SAFE)
       assert_equal 'Document Title', doc.doctitle
       refute doc.attr?('docfile')
       assert_equal doc.base_dir, doc.attr('docdir')
diff --git a/test/attribute_list_test.rb b/test/attribute_list_test.rb
index 40454a64..74f389a7 100644
--- a/test/attribute_list_test.rb
+++ b/test/attribute_list_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
diff --git a/test/attributes_test.rb b/test/attributes_test.rb
index cc63812f..eb491465 100644
--- a/test/attributes_test.rb
+++ b/test/attributes_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -59,8 +58,7 @@ context 'Attributes' do
         result = convert_string_to_embedded str
         assert_includes result, %(<p>#{value}</p>)
       end
-
-    end if ::RUBY_MIN_VERSION_1_9
+    end
 
     test 'creates an attribute by fusing a legacy multi-line value' do
       str = <<-EOS
@@ -245,11 +243,7 @@ content
 {imagesdir}
 EOS
       output = convert_inline_string input, :safe => :safe
-      if RUBY_VERSION >= '1.9'
-        assert_equal %(#{Dir.home}/etc/images), output
-      else
-        assert_equal %(#{ENV['HOME']}/etc/images), output
-      end
+      assert_equal %(#{Asciidoctor::USER_HOME}/etc/images), output
     end
 
     test 'user-home attribute resolves to . if safe mode is SERVER or greater' do
diff --git a/test/blocks_test.rb b/test/blocks_test.rb
index effcee5a..b4ca21ac 100644
--- a/test/blocks_test.rb
+++ b/test/blocks_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -927,9 +926,9 @@ EOS
         assert_xpath '//pre', output, 1
         assert_xpath '//pre/text()', output, 1
         text = xmlnodes_at_xpath('//pre/text()', output, 1).text
-        lines = text.lines.entries
+        lines = text.lines
         assert_equal 5, lines.size
-        expected = "line one\n\nline two\n\nline three".lines.entries
+        expected = "line one\n\nline two\n\nline three".lines
         assert_equal expected, lines
         blank_lines = output.scan(/\n[ \t]*\n/).size
         assert blank_lines >= 2
@@ -952,9 +951,9 @@ EOS
         assert_xpath '//pre/code', output, 1
         assert_xpath '//pre/code/text()', output, 1
         text = xmlnodes_at_xpath('//pre/code/text()', output, 1).text
-        lines = text.lines.entries
+        lines = text.lines
         assert_equal 5, lines.size
-        expected = "line one\n\nline two\n\nline three".lines.entries
+        expected = "line one\n\nline two\n\nline three".lines
         assert_equal expected, lines
         blank_lines = output.scan(/\n[ \t]*\n/).size
         assert blank_lines >= 2
@@ -979,9 +978,9 @@ EOS
         assert_xpath '//*[@class="verseblock"]/pre', output, 1
         assert_xpath '//*[@class="verseblock"]/pre/text()', output, 1
         text = xmlnodes_at_xpath('//*[@class="verseblock"]/pre/text()', output, 1).text
-        lines = text.lines.entries
+        lines = text.lines
         assert_equal 5, lines.size
-        expected = "line one\n\nline two\n\nline three".lines.entries
+        expected = "line one\n\nline two\n\nline three".lines
         assert_equal expected, lines
         blank_lines = output.scan(/\n[ \t]*\n/).size
         assert blank_lines >= 2
diff --git a/test/converter_test.rb b/test/converter_test.rb
index cd61c4e1..69a99d30 100644
--- a/test/converter_test.rb
+++ b/test/converter_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
diff --git a/test/document_test.rb b/test/document_test.rb
index 87e70e52..fbfdec80 100644
--- a/test/document_test.rb
+++ b/test/document_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -275,17 +274,13 @@ content
     test 'should force encoding of docinfo files to UTF-8' do
       sample_input_path = fixture_path('basic.asciidoc')
 
-      if RUBY_VERSION >= '1.9'
-        default_external_old = Encoding.default_external
-        force_encoding_old = Asciidoctor::FORCE_ENCODING
-        verbose_old = $VERBOSE
-      end
+      default_external_old = Encoding.default_external
+      force_encoding_old = Asciidoctor::FORCE_ENCODING
+      verbose_old = $VERBOSE
       begin
-        if RUBY_VERSION >= '1.9'
-          $VERBOSE = nil # disable warnings since we have to modify constants
-          Encoding.default_external = 'US-ASCII'
-          Asciidoctor::FORCE_ENCODING = true
-        end
+        $VERBOSE = nil # disable warnings since we have to modify constants
+        Encoding.default_external = 'ISO-8859-1'
+        Asciidoctor::FORCE_ENCODING = true
         output = Asciidoctor.convert_file sample_input_path, :to_file => false,
             :header_footer => true, :backend => 'docbook', :safe => Asciidoctor::SafeMode::SERVER, :attributes => {'docinfo2' => ''}
         refute_empty output
@@ -294,11 +289,9 @@ content
         assert_xpath '//xmlns:edition[text()="1.0"]', output, 1 # verifies substitutions are performed
         assert_css 'copyright', output, 1
       ensure
-        if RUBY_VERSION >= '1.9'
-          Encoding.default_external = default_external_old
-          Asciidoctor::FORCE_ENCODING = force_encoding_old
-          $VERBOSE = verbose_old
-        end
+        Encoding.default_external = default_external_old
+        Asciidoctor::FORCE_ENCODING = force_encoding_old
+        $VERBOSE = verbose_old
       end
     end
 
diff --git a/test/extensions_test.rb b/test/extensions_test.rb
index 37798554..da4a5936 100644
--- a/test/extensions_test.rb
+++ b/test/extensions_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -141,32 +140,13 @@ class TemperatureMacro < Asciidoctor::Extensions::InlineMacroProcessor; use_dsl
     c = target.to_f
     case units
     when 'C'
-      %(#{round_with_precision c, precision} &#176;C)
+      %(#{c.round precision} &#176;C)
     when 'F'
-      %(#{round_with_precision c * 1.8 + 32, precision} &#176;F)
+      %(#{(c * 1.8 + 32).round precision} &#176;F)
     else
       raise ::ArgumentError, %(Unknown temperature units: #{units})
     end
   end
-
-  if (::Numeric.instance_method :round).arity == 0
-    def round_with_precision value, precision = 0
-      if precision == 0
-        value.round
-      else
-        factor = 10 ** precision
-        if precision < 0
-          (value * factor).round.div factor
-        else
-          (value * factor).round.fdiv factor
-        end
-      end
-    end
-  else
-    def round_with_precision value, precision = 0
-      value.round precision
-    end
-  end
 end
 
 class MetaRobotsDocinfoProcessor < Asciidoctor::Extensions::DocinfoProcessor
diff --git a/test/invoker_test.rb b/test/invoker_test.rb
index 060d04ab..b9d23324 100644
--- a/test/invoker_test.rb
+++ b/test/invoker_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -130,7 +129,7 @@ context 'Invoker' do
     ensure
       FileUtils.rm_f sample_inpath
     end
-  end if RUBY_MIN_VERSION_1_9 && !windows?
+  end unless windows?
 
   test 'should allow docdir to be specified when input is a string' do
     expected_docdir = fixturedir
diff --git a/test/links_test.rb b/test/links_test.rb
index b4dee7dd..750c73fd 100644
--- a/test/links_test.rb
+++ b/test/links_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -158,7 +157,7 @@ context 'Links' do
 
   test 'qualified url following no-break space' do
     assert_xpath '//a[@href="http://asciidoc.org"][text()="AsciiDoc"]', convert_string(%(#{[0xa0].pack 'U1'}http://asciidoc.org[AsciiDoc] project page.)), 1
-  end if ::RUBY_MIN_VERSION_1_9
+  end
 
   test 'qualified url following smart apostrophe' do
     output = convert_string_to_embedded("l&#8217;http://www.irit.fr[IRIT]")
diff --git a/test/lists_test.rb b/test/lists_test.rb
index a779ef8a..1bfa4ff0 100644
--- a/test/lists_test.rb
+++ b/test/lists_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -239,7 +238,7 @@ second wrapped line
       assert_css 'ul', output, 1
       assert_css 'ul li', output, 2
       # NOTE for some reason, we're getting an extra line after the indented line
-      lines = xmlnodes_at_xpath('(//ul/li)[1]/p', output, 1).text.gsub(/\n[[:space:]]*\n/, "\n").lines.entries
+      lines = xmlnodes_at_xpath('(//ul/li)[1]/p', output, 1).text.gsub(/\n[[:space:]]*\n/, "\n").lines
       assert_equal 3, lines.size
       assert_equal 'list item 1', lines[0].chomp
       assert_equal '  // not line comment', lines[1].chomp
@@ -539,7 +538,7 @@ List
       output = convert_string input
       assert_xpath '//ul', output, 1
       assert_xpath '//ul/li', output, 3
-    end if ::RUBY_MIN_VERSION_1_9
+    end
 
     test 'indented asterisk elements using tabs' do
       input = <<-EOS
@@ -996,7 +995,7 @@ List
       output = convert_string input
       assert_xpath '//ul', output, 0
       assert_includes output, '•'
-    end if ::RUBY_MIN_VERSION_1_9
+    end
 
     test "nested ordered elements (2)" do
       input = <<-EOS
diff --git a/test/logger_test.rb b/test/logger_test.rb
index 3f921b0e..fe1be778 100644
--- a/test/logger_test.rb
+++ b/test/logger_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
diff --git a/test/manpage_test.rb b/test/manpage_test.rb
index 0d56c40b..8102450b 100644
--- a/test/manpage_test.rb
+++ b/test/manpage_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -175,7 +174,7 @@ Does stuff.
 
 (C) & (R) are translated to character references, but not the &.)
       output = Asciidoctor.convert input, :backend => :manpage
-      assert_equal '\\(co & \\(rg are translated to character references, but not the &.', output.lines.entries.last.chomp
+      assert_equal '\\(co & \\(rg are translated to character references, but not the &.', output.lines.last.chomp
     end
 
     test 'should replace em dashes' do
@@ -194,7 +193,7 @@ go--to)
 
 .)
       output = Asciidoctor.convert input, :backend => :manpage
-      assert_equal '\&.', output.lines.entries.last.chomp
+      assert_equal '\&.', output.lines.last.chomp
     end
 
     test 'should escape raw macro' do
@@ -205,7 +204,7 @@ AAA this line of text should be show
 BBB this line and the one above it should be visible)
 
       output = Asciidoctor.convert input, :backend => :manpage
-      assert_equal '\&.if 1 .nx', output.lines.entries[-2].chomp
+      assert_equal '\&.if 1 .nx', output.lines[-2].chomp
     end
 
     test 'should normalize whitespace in a paragraph' do
@@ -261,7 +260,7 @@ Describe this thing.)
 
 "`hello`" '`goodbye`' *strong* _weak_ `even`)
       output = Asciidoctor.convert input, :backend => :manpage
-      assert_equal '\(lqhello\(rq \(oqgoodbye\(cq \fBstrong\fP \fIweak\fP \f(CReven\fP', output.lines.entries.last.chomp
+      assert_equal '\(lqhello\(rq \(oqgoodbye\(cq \fBstrong\fP \fIweak\fP \f(CReven\fP', output.lines.last.chomp
     end
 
     test 'should escape backslashes in content' do
@@ -270,7 +269,7 @@ Describe this thing.)
 \\.foo \\ bar\\
 baz)
       output = Asciidoctor.convert input, :backend => :manpage
-      assert_equal '\(rs.foo \(rs bar\(rs', output.lines.entries[-2].chomp
+      assert_equal '\(rs.foo \(rs bar\(rs', output.lines[-2].chomp
     end
 
     test 'should escape literal escape sequence' do
@@ -289,7 +288,7 @@ After break.)
       output = Asciidoctor.convert input, :backend => :manpage
       assert_equal 'Before break.
 .br
-After break.', output.lines.entries[-3..-1].join
+After break.', output.lines[-3..-1].join
     end
   end
 
@@ -303,7 +302,7 @@ http://asciidoc.org[AsciiDoc])
       assert_equal '.sp
 First paragraph.
 .sp
-.URL "http://asciidoc.org" "AsciiDoc" ""', output.lines.entries[-4..-1].join
+.URL "http://asciidoc.org" "AsciiDoc" ""', output.lines[-4..-1].join
     end
 
     test 'should not swallow content following URL' do
@@ -312,7 +311,7 @@ First paragraph.
 http://asciidoc.org[AsciiDoc] can be used to create man pages.)
       output = Asciidoctor.convert input, :backend => :manpage
       assert_equal '.URL "http://asciidoc.org" "AsciiDoc" " "
-can be used to create man pages.', output.lines.entries[-2..-1].join
+can be used to create man pages.', output.lines[-2..-1].join
     end
 
     test 'should pass adjacent character as final argument of URL macro' do
@@ -321,7 +320,7 @@ can be used to create man pages.', output.lines.entries[-2..-1].join
 This is http://asciidoc.org[AsciiDoc].)
       output = Asciidoctor.convert input, :backend => :manpage
       assert_equal 'This is \c
-.URL "http://asciidoc.org" "AsciiDoc" "."', output.lines.entries[-2..-1].join
+.URL "http://asciidoc.org" "AsciiDoc" "."', output.lines[-2..-1].join
     end
 
     test 'should pass adjacent character as final argument of URL macro and move trailing content to next line' do
@@ -331,7 +330,7 @@ This is http://asciidoc.org[AsciiDoc], which can be used to write content.)
       output = Asciidoctor.convert input, :backend => :manpage
       assert_equal 'This is \c
 .URL "http://asciidoc.org" "AsciiDoc" ","
-which can be used to write content.', output.lines.entries[-3..-1].join
+which can be used to write content.', output.lines[-3..-1].join
     end
 
     test 'should not leave blank lines between URLs on contiguous lines of input' do
@@ -351,7 +350,7 @@ The corresponding implementations are
 .URL "http://cmucl.org" "CMUCL" ","
 .URL "http://ecls.sf.net" "ECL" ","
 and \c
-.URL "http://sbcl.sf.net" "SBCL" "."', output.lines.entries[-8..-1].join
+.URL "http://sbcl.sf.net" "SBCL" "."', output.lines[-8..-1].join
     end
 
     test 'should not leave blank lines between URLs on same line of input' do
@@ -366,7 +365,7 @@ The corresponding implementations are \c
 .URL "http://cmucl.org" "CMUCL" ","
 .URL "http://ecls.sf.net" "ECL" ","
 and
-.URL "http://sbcl.sf.net" "SBCL" "."', output.lines.entries[-8..-1].join
+.URL "http://sbcl.sf.net" "SBCL" "."', output.lines[-8..-1].join
     end
 
     test 'should not insert space between link and non-whitespace characters surrounding it' do
@@ -377,7 +376,7 @@ Please search |link:http://discuss.asciidoctor.org[the forums]| before asking.)
       assert_equal '.sp
 Please search |\c
 .URL "http://discuss.asciidoctor.org" "the forums" "|"
-before asking.', output.lines.entries[-4..-1].join
+before asking.', output.lines[-4..-1].join
     end
 
     test 'should be able to use monospaced text inside a link' do
@@ -388,7 +387,7 @@ Enter the link:cat[`cat`] command.)
       assert_equal '.sp
 Enter the \c
 .URL "cat" "\f(CRcat\fP" " "
-command.', output.lines.entries[-4..-1].join
+command.', output.lines[-4..-1].join
     end
   end
 
@@ -402,7 +401,7 @@ mailto:doc@example.org[Contact the doc])
       assert_equal '.sp
 First paragraph.
 .sp
-.MTO "doc\\(atexample.org" "Contact the doc" ""', output.lines.entries[-4..-1].join
+.MTO "doc\\(atexample.org" "Contact the doc" ""', output.lines[-4..-1].join
     end
 
     test 'should set text of MTO macro to blank for implicit email' do
diff --git a/test/options_test.rb b/test/options_test.rb
index a2629983..4dd31ce6 100644
--- a/test/options_test.rb
+++ b/test/options_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
diff --git a/test/paragraphs_test.rb b/test/paragraphs_test.rb
index 62b9cade..35b5eda6 100644
--- a/test/paragraphs_test.rb
+++ b/test/paragraphs_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
diff --git a/test/parser_test.rb b/test/parser_test.rb
index 2e96bef1..59c67ec6 100644
--- a/test/parser_test.rb
+++ b/test/parser_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -298,7 +297,7 @@ context "Parser" do
     assert_equal 'Stéphane', metadata['firstname']
     assert_equal 'Brontë', metadata['lastname']
     assert_equal 'SB', metadata['authorinitials']
-  end if ::RUBY_MIN_VERSION_1_9
+  end
 
   test 'parse ideographic author names' do
     metadata, _ = parse_header_metadata '李 四 <si.li@example.com>'
@@ -310,7 +309,7 @@ context "Parser" do
     assert_equal '四', metadata['lastname']
     assert_equal 'si.li@example.com', metadata['email']
     assert_equal '李四', metadata['authorinitials']
-  end if ::RUBY_MIN_VERSION_1_9
+  end
 
   test "parse author condenses whitespace" do
     metadata, _ = parse_header_metadata '   Stuart       Rackham     <founder@asciidoc.org>'
@@ -710,7 +709,7 @@ devtmpfs                3.9G       0     3.9G     0%    /dev
 
     expected = input
 
-    lines = input.lines.entries
+    lines = input.lines
     Asciidoctor::Parser.adjust_indentation! lines, -1
     assert_equal expected, lines.join
   end
diff --git a/test/paths_test.rb b/test/paths_test.rb
index 459705d5..1340dc76 100644
--- a/test/paths_test.rb
+++ b/test/paths_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
diff --git a/test/preamble_test.rb b/test/preamble_test.rb
index d9b8127a..3f76edc8 100644
--- a/test/preamble_test.rb
+++ b/test/preamble_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
diff --git a/test/reader_test.rb b/test/reader_test.rb
index 0c71228c..4364275d 100644
--- a/test/reader_test.rb
+++ b/test/reader_test.rb
@@ -1,11 +1,10 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
 end
 
 class ReaderTest < Minitest::Test
-  DIRNAME = File.expand_path File.dirname __FILE__
+  DIRNAME = ASCIIDOCTOR_TEST_DIR
 
   SAMPLE_DATA = <<-EOS.chomp.split(::Asciidoctor::LF)
 first line
@@ -40,34 +39,32 @@ third line
         assert_equal SAMPLE_DATA, reader.lines
       end
 
-      if Asciidoctor::COERCE_ENCODING
-        test 'should encode UTF-16LE string to UTF-8 when BOM is found' do
-          data = "\ufeff#{SAMPLE_DATA.join ::Asciidoctor::LF}".encode('UTF-16LE').force_encoding('UTF-8')
-          reader = Asciidoctor::Reader.new data, nil, :normalize => true
-          assert_equal 'f', reader.lines.first.chr
-          assert_equal SAMPLE_DATA, reader.lines
-        end
+      test 'should encode UTF-16LE string to UTF-8 when BOM is found' do
+        data = "\ufeff#{SAMPLE_DATA.join ::Asciidoctor::LF}".encode('UTF-16LE').force_encoding('UTF-8')
+        reader = Asciidoctor::Reader.new data, nil, :normalize => true
+        assert_equal 'f', reader.lines.first.chr
+        assert_equal SAMPLE_DATA, reader.lines
+      end
 
-        test 'should encode UTF-16LE string array to UTF-8 when BOM is found' do
-          data = "\ufeff#{SAMPLE_DATA.join ::Asciidoctor::LF}".encode('UTF-16LE').force_encoding('UTF-8').lines.to_a
-          reader = Asciidoctor::Reader.new data, nil, :normalize => true
-          assert_equal 'f', reader.lines.first.chr
-          assert_equal SAMPLE_DATA, reader.lines
-        end
+      test 'should encode UTF-16LE string array to UTF-8 when BOM is found' do
+        data = "\ufeff#{SAMPLE_DATA.join ::Asciidoctor::LF}".encode('UTF-16LE').force_encoding('UTF-8').lines.to_a
+        reader = Asciidoctor::Reader.new data, nil, :normalize => true
+        assert_equal 'f', reader.lines.first.chr
+        assert_equal SAMPLE_DATA, reader.lines
+      end
 
-        test 'should encode UTF-16BE string to UTF-8 when BOM is found' do
-          data = "\ufeff#{SAMPLE_DATA.join ::Asciidoctor::LF}".encode('UTF-16BE').force_encoding('UTF-8')
-          reader = Asciidoctor::Reader.new data, nil, :normalize => true
-          assert_equal 'f', reader.lines.first.chr
-          assert_equal SAMPLE_DATA, reader.lines
-        end
+      test 'should encode UTF-16BE string to UTF-8 when BOM is found' do
+        data = "\ufeff#{SAMPLE_DATA.join ::Asciidoctor::LF}".encode('UTF-16BE').force_encoding('UTF-8')
+        reader = Asciidoctor::Reader.new data, nil, :normalize => true
+        assert_equal 'f', reader.lines.first.chr
+        assert_equal SAMPLE_DATA, reader.lines
+      end
 
-        test 'should encode UTF-16BE string array to UTF-8 when BOM is found' do
-          data = "\ufeff#{SAMPLE_DATA.join ::Asciidoctor::LF}".encode('UTF-16BE').force_encoding('UTF-8').lines.to_a
-          reader = Asciidoctor::Reader.new data, nil, :normalize => true
-          assert_equal 'f', reader.lines.first.chr
-          assert_equal SAMPLE_DATA, reader.lines
-        end
+      test 'should encode UTF-16BE string array to UTF-8 when BOM is found' do
+        data = "\ufeff#{SAMPLE_DATA.join ::Asciidoctor::LF}".encode('UTF-16BE').force_encoding('UTF-8').lines.to_a
+        reader = Asciidoctor::Reader.new data, nil, :normalize => true
+        assert_equal 'f', reader.lines.first.chr
+        assert_equal SAMPLE_DATA, reader.lines
       end
     end
 
@@ -269,7 +266,7 @@ third line
 
     context 'Read lines until' do
       test 'Read lines until until end' do
-        lines = <<-EOS.lines.entries
+        lines = <<-EOS.lines
 This is one paragraph.
 
 This is another paragraph.
@@ -284,7 +281,7 @@ This is another paragraph.
       end
 
       test 'Read lines until until blank line' do
-        lines = <<-EOS.lines.entries
+        lines = <<-EOS.lines
 This is one paragraph.
 
 This is another paragraph.
@@ -1207,8 +1204,7 @@ include::fixtures/include-file.asciidoc[tags=no-such-tag-b;no-such-tag-a]
 
         using_memory_logger do |logger|
           convert_string_to_embedded input, :safe => :safe, :base_dir => DIRNAME
-          # NOTE Ruby 1.8 swaps the order of the list for some silly reason
-          expected_tags = ::RUBY_MIN_VERSION_1_9 ? 'no-such-tag-b, no-such-tag-a' : 'no-such-tag-a, no-such-tag-b'
+          expected_tags = 'no-such-tag-b, no-such-tag-a'
           assert_message logger, :WARN, %(~<stdin>: line 2: tags '#{expected_tags}' not found in include file), Hash
         end
       end
diff --git a/test/sections_test.rb b/test/sections_test.rb
index 513baed3..98f5cbc1 100644
--- a/test/sections_test.rb
+++ b/test/sections_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -594,11 +593,7 @@ endif::[]
 == Asciidoctor in 中文
       EOS
       output = convert_string input
-      if ::RUBY_MIN_VERSION_1_9
-        assert_xpath '//h2[@id="_asciidoctor_in_中文"][text()="Asciidoctor in 中文"]', output
-      else
-        assert_xpath '//h2[@id="_asciidoctor_in"][text()="Asciidoctor in 中文"]', output
-      end
+      assert_xpath '//h2[@id="_asciidoctor_in_中文"][text()="Asciidoctor in 中文"]', output
     end
 
     test 'with only multibyte characters' do
@@ -607,7 +602,7 @@ endif::[]
       EOS
       output = convert_string_to_embedded input
       assert_xpath '//h2[@id="_视图"][text()="视图"]', output
-    end if ::RUBY_MIN_VERSION_1_9
+    end
 
     test 'multiline syntax with only multibyte characters' do
       input = <<-EOS
@@ -624,7 +619,7 @@ content
       output = convert_string_to_embedded input
       assert_xpath '//h2[@id="_视图"][text()="视图"]', output
       assert_xpath '//h2[@id="_连接器"][text()="连接器"]', output
-    end if ::RUBY_MIN_VERSION_1_9
+    end
   end
 
   context "level 2" do
diff --git a/test/substitutions_test.rb b/test/substitutions_test.rb
index 65d9752d..2e4f6f37 100644
--- a/test/substitutions_test.rb
+++ b/test/substitutions_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -228,7 +227,7 @@ context 'Substitutions' do
     test 'constrained strong string containing an asterisk and multibyte word chars' do
       para = block_from_string(%q{*黑*眼圈*})
       assert_equal '<strong>黑*眼圈</strong>', para.sub_quotes(para.source)
-    end if ::RUBY_MIN_VERSION_1_9
+    end
 
     test 'single-line constrained quote variation emphasized string' do
       para = block_from_string(%q{_a few emphasized words_})
@@ -1486,12 +1485,12 @@ EOS
       test 'should process menu macro with items containing multibyte characters' do
         para = block_from_string('menu:视图[放大, 重置]', :attributes => {'experimental' => ''})
         assert_equal %q{<span class="menuseq"><b class="menu">视图</b>&#160;<b class="caret">&#8250;</b> <b class="submenu">放大</b>&#160;<b class="caret">&#8250;</b> <b class="menuitem">重置</b></span>}, para.sub_macros(para.source)
-      end if ::RUBY_MIN_VERSION_1_9
+      end
 
       test 'should process inline menu with items containing multibyte characters' do
         para = block_from_string('"视图 &gt; 放大 &gt; 重置"', :attributes => {'experimental' => ''})
         assert_equal %q{<span class="menuseq"><b class="menu">视图</b>&#160;<b class="caret">&#8250;</b> <b class="submenu">放大</b>&#160;<b class="caret">&#8250;</b> <b class="menuitem">重置</b></span>}, para.sub_macros(para.source)
-      end if ::RUBY_MIN_VERSION_1_9
+      end
 
       test 'should process a menu macro with a target that begins with a character reference' do
         para = block_from_string('menu:&#8942;[More Tools, Extensions]', :attributes => {'experimental' => ''})
@@ -1879,7 +1878,7 @@ foo&#8201;&#8212;&#8201;'
       para = block_from_string %(富--巴)
       expected = '富&#8212;&#8203;巴'
       assert_equal expected, para.sub_replacements(para.source)
-    end if ::RUBY_MIN_VERSION_1_9
+    end
 
     test 'replaces marks' do
       para = block_from_string '(C) (R) (TM) \(C) \(R) \(TM)'
diff --git a/test/tables_test.rb b/test/tables_test.rb
index 6e8a8715..816f30a9 100644
--- a/test/tables_test.rb
+++ b/test/tables_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'
@@ -1101,10 +1100,10 @@ I wouldn't have it any other way.
       assert_css 'table > tbody > tr > td:nth-child(1).halign-left.valign-top > p.tableblock', output, 7
       assert_css 'table > tbody > tr > td:nth-child(2).halign-center.valign-top > div.verse', output, 1
       verse = xmlnodes_at_css 'table > tbody > tr > td:nth-child(2).halign-center.valign-top > div.verse', output, 1
-      assert_equal 26, verse.text.lines.entries.size
+      assert_equal 26, verse.text.lines.size
       assert_css 'table > tbody > tr > td:nth-child(3).halign-center.valign-top > div.literal > pre', output, 1
       literal = xmlnodes_at_css 'table > tbody > tr > td:nth-child(3).halign-center.valign-top > div.literal > pre', output, 1
-      assert_equal 26, literal.text.lines.entries.size
+      assert_equal 26, literal.text.lines.size
     end
 
     test 'should strip trailing endline when splitting paragraphs' do
diff --git a/test/test_helper.rb b/test/test_helper.rb
index 720b9472..87a16a60 100644
--- a/test/test_helper.rb
+++ b/test/test_helper.rb
@@ -1,13 +1,8 @@
-# encoding: UTF-8
 ASCIIDOCTOR_TEST_DIR = File.expand_path File.dirname __FILE__
 ASCIIDOCTOR_PROJECT_DIR = File.dirname ASCIIDOCTOR_TEST_DIR
 ASCIIDOCTOR_LIB_DIR = ENV['ASCIIDOCTOR_LIB_DIR'] || File.join(ASCIIDOCTOR_PROJECT_DIR, 'lib')
 Dir.chdir ASCIIDOCTOR_PROJECT_DIR
 
-if RUBY_VERSION < '1.9'
-  require 'rubygems'
-end
-
 require 'simplecov' if ENV['COVERAGE'] == 'true'
 
 require File.join(ASCIIDOCTOR_LIB_DIR, 'asciidoctor')
@@ -182,9 +177,9 @@ class Minitest::Test
   def document_from_string(src, opts = {})
     assign_default_test_options opts
     if opts[:parse]
-      (Asciidoctor::Document.new src.lines.entries, opts).parse
+      (Asciidoctor::Document.new src.lines, opts).parse
     else
-      Asciidoctor::Document.new src.lines.entries, opts
+      Asciidoctor::Document.new src.lines, opts
     end
   end
 
@@ -301,8 +296,7 @@ class Minitest::Test
   end
 
   def resolve_localhost
-    (RUBY_VERSION < '1.9' || RUBY_ENGINE == 'rbx') ? Socket.gethostname :
-        Socket.ip_address_list.find {|addr| addr.ipv4? }.ip_address
+    Socket.ip_address_list.find {|addr| addr.ipv4? }.ip_address
   end
 
   def using_memory_logger
diff --git a/test/text_test.rb b/test/text_test.rb
index e3f0d21f..894d3e85 100644
--- a/test/text_test.rb
+++ b/test/text_test.rb
@@ -1,4 +1,3 @@
-# encoding: UTF-8
 unless defined? ASCIIDOCTOR_PROJECT_DIR
   $: << File.dirname(__FILE__); $:.uniq!
   require 'test_helper'