Merge branch 'master' into bash-scanner

18 files changed, 241 insertions, 37 deletions

diff --git a/lib/coderay/encoders/debug.rb b/lib/coderay/encoders/debug.rb
index 95d6138..c03d3fb 100644
--- a/lib/coderay/encoders/debug.rb
+++ b/lib/coderay/encoders/debug.rb
@@ -24,11 +24,12 @@ module Encoders
     end
     
     def text_token text, kind
+      raise 'empty token' if $CODERAY_DEBUG && text.empty?
       if kind == :space
         @out << text
       else
         # TODO: Escape (
-        text = text.gsub(/[)\\]/, '\\\\\0')  # escape ) and \
+        text = text.gsub(/[)\\]/, '\\\\\0') if text.index(/[)\\]/)
         @out << kind.to_s << '(' << text << ')'
       end
     end
diff --git a/lib/coderay/encoders/html.rb b/lib/coderay/encoders/html.rb
index 635a4d8..0fd1317 100644
--- a/lib/coderay/encoders/html.rb
+++ b/lib/coderay/encoders/html.rb
@@ -286,8 +286,7 @@ module Encoders
     
     def end_group kind
       if $CODERAY_DEBUG && (@opened.empty? || @opened.last != kind)
-        warn 'Malformed token stream: Trying to close a token (%p) ' \
-          'that is not open. Open are: %p.' % [kind, @opened[1..-1]]
+        warn 'Malformed token stream: Trying to close a token group (%p) that is not open. Open are: %p.' % [kind, @opened[1..-1]]
       end
       if @opened.pop
         @out << '</span>'
@@ -312,8 +311,7 @@ module Encoders
     
     def end_line kind
       if $CODERAY_DEBUG && (@opened.empty? || @opened.last != kind)
-        warn 'Malformed token stream: Trying to close a line (%p) ' \
-          'that is not open. Open are: %p.' % [kind, @opened[1..-1]]
+        warn 'Malformed token stream: Trying to close a line (%p) that is not open. Open are: %p.' % [kind, @opened[1..-1]]
       end
       if @opened.pop
         @out << '</span>'
diff --git a/lib/coderay/helpers/file_type.rb b/lib/coderay/helpers/file_type.rb
index 637001b..a5d83ff 100644
--- a/lib/coderay/helpers/file_type.rb
+++ b/lib/coderay/helpers/file_type.rb
@@ -99,6 +99,7 @@ module CodeRay
       'mab'      => :ruby,
       'pas'      => :delphi,
       'patch'    => :diff,
+      'phtml'    => :php,
       'php'      => :php,
       'php3'     => :php,
       'php4'     => :php,
@@ -116,10 +117,10 @@ module CodeRay
       'rpdf'     => :ruby,
       'ru'       => :ruby,
       'rxml'     => :ruby,
-      # 'sch'      => :scheme,
+      'sass'     => :sass,
       'sql'      => :sql,
-      # 'ss'       => :scheme,
       'tmproj'   => :xml,
+      'xaml'     => :xml,
       'xhtml'    => :html,
       'xml'      => :xml,
       'yaml'     => :yaml,
diff --git a/lib/coderay/helpers/plugin.rb b/lib/coderay/helpers/plugin.rb
index 137c1ab..dd4e830 100644
--- a/lib/coderay/helpers/plugin.rb
+++ b/lib/coderay/helpers/plugin.rb
@@ -180,7 +180,6 @@ module CodeRay
         rescue LoadError => boom
           if @plugin_map_loaded
             if h.has_key?(:default)
-              warn '%p could not load plugin %p; falling back to %p' % [self, id, h[:default]]
               h[:default]
             else
               raise PluginNotFound, '%p could not load plugin %p: %s' % [self, id, boom]
diff --git a/lib/coderay/scanners/c.rb b/lib/coderay/scanners/c.rb
index 8d24b99..84b6e8e 100644
--- a/lib/coderay/scanners/c.rb
+++ b/lib/coderay/scanners/c.rb
@@ -148,7 +148,7 @@ module Scanners
             encoder.text_token match, :char
           elsif match = scan(/ \\ | $ /x)
             encoder.end_group :string
-            encoder.text_token match, :error
+            encoder.text_token match, :error unless match.empty?
             state = :initial
             label_expected = false
           else
diff --git a/lib/coderay/scanners/cpp.rb b/lib/coderay/scanners/cpp.rb
index 9da62f4..e61f56f 100644
--- a/lib/coderay/scanners/cpp.rb
+++ b/lib/coderay/scanners/cpp.rb
@@ -160,7 +160,7 @@ module Scanners
             encoder.text_token match, :char
           elsif match = scan(/ \\ | $ /x)
             encoder.end_group :string
-            encoder.text_token match, :error
+            encoder.text_token match, :error unless match.empty?
             state = :initial
             label_expected = false
           else
diff --git a/lib/coderay/scanners/css.rb b/lib/coderay/scanners/css.rb
index 7b731ef..003eed6 100644
--- a/lib/coderay/scanners/css.rb
+++ b/lib/coderay/scanners/css.rb
@@ -15,19 +15,17 @@ module Scanners
     
     module RE  # :nodoc:
       Hex = /[0-9a-fA-F]/
-      Unicode = /\\#{Hex}{1,6}(?:\r\n|\s)?/ # differs from standard because it allows uppercase hex too
-      Escape = /#{Unicode}|\\[^\r\n\f0-9a-fA-F]/
-      NMChar = /[-_a-zA-Z0-9]|#{Escape}/
-      NMStart = /[_a-zA-Z]|#{Escape}/
-      NL = /\r\n|\r|\n|\f/
-      String1 = /"(?:[^\n\r\f\\"]|\\#{NL}|#{Escape})*"?/  # TODO: buggy regexp
-      String2 = /'(?:[^\n\r\f\\']|\\#{NL}|#{Escape})*'?/  # TODO: buggy regexp
+      Unicode = /\\#{Hex}{1,6}\b/ # differs from standard because it allows uppercase hex too
+      Escape = /#{Unicode}|\\[^\n0-9a-fA-F]/
+      NMChar = /[-_a-zA-Z0-9]/
+      NMStart = /[_a-zA-Z]/
+      String1 = /"(?:[^\n\\"]+|\\\n|#{Escape})*"?/  # TODO: buggy regexp
+      String2 = /'(?:[^\n\\']+|\\\n|#{Escape})*'?/  # TODO: buggy regexp
       String = /#{String1}|#{String2}/
       
       HexColor = /#(?:#{Hex}{6}|#{Hex}{3})/
-      Color = /#{HexColor}/
       
-      Num = /-?(?:[0-9]+|[0-9]*\.[0-9]+)/
+      Num = /-?(?:[0-9]*\.[0-9]+|[0-9]+)/
       Name = /#{NMChar}+/
       Ident = /-?#{NMStart}#{NMChar}*/
       AtKeyword = /@#{Ident}/
@@ -35,16 +33,15 @@ module Scanners
       
       reldimensions = %w[em ex px]
       absdimensions = %w[in cm mm pt pc]
-      Unit = Regexp.union(*(reldimensions + absdimensions + %w[s]))
+      Unit = Regexp.union(*(reldimensions + absdimensions + %w[s dpi dppx deg]))
       
       Dimension = /#{Num}#{Unit}/
       
-      Comment = %r! /\* (?: .*? \*/ | .* ) !mx
-      Function = /(?:url|alpha|attr|counters?)\((?:[^)\n\r\f]|\\\))*\)?/
+      Function = /(?:url|alpha|attr|counters?)\((?:[^)\n]|\\\))*\)?/
       
-      Id = /##{Name}/
+      Id = /(?!#{HexColor}\b(?!-))##{Name}/
       Class = /\.#{Name}/
-      PseudoClass = /:#{Name}/
+      PseudoClass = /::?#{Ident}/
       AttributeSelector = /\[[^\]]*\]?/
     end
     
@@ -52,7 +49,7 @@ module Scanners
     
     def setup
       @state = :initial
-      @value_expected = nil
+      @value_expected = false
     end
     
     def scan_tokens encoder, options
@@ -158,7 +155,7 @@ module Scanners
         elsif match = scan(/(?: #{RE::Dimension} | #{RE::Percentage} | #{RE::Num} )/ox)
           encoder.text_token match, :float
           
-        elsif match = scan(/#{RE::Color}/o)
+        elsif match = scan(/#{RE::HexColor}/o)
           encoder.text_token match, :color
           
         elsif match = scan(/! *important/)
@@ -170,7 +167,7 @@ module Scanners
         elsif match = scan(RE::AtKeyword)
           encoder.text_token match, :directive
           
-        elsif match = scan(/ [+>:;,.=()\/] /x)
+        elsif match = scan(/ [+>~:;,.=()\/] /x)
           if match == ':'
             value_expected = true
           elsif match == ';'
diff --git a/lib/coderay/scanners/diff.rb b/lib/coderay/scanners/diff.rb
index 38efaf4..af0f755 100644
--- a/lib/coderay/scanners/diff.rb
+++ b/lib/coderay/scanners/diff.rb
@@ -45,7 +45,7 @@ module Scanners
           if match = scan(/--- |\+\+\+ |=+|_+/)
             encoder.begin_line line_kind = :head
             encoder.text_token match, :head
-            if match = scan(/.*?(?=$|[\t\n\x00]|  \(revision)/)
+            if match = scan(/[^\x00\n]+?(?=$|[\t\n]|  \(revision)/)
               encoder.text_token match, :filename
               if options[:highlight_code] && match != '/dev/null'
                 file_type = CodeRay::FileType.fetch(match, :text)
diff --git a/lib/coderay/scanners/java.rb b/lib/coderay/scanners/java.rb
index c1490ac..b282864 100644
--- a/lib/coderay/scanners/java.rb
+++ b/lib/coderay/scanners/java.rb
@@ -147,7 +147,7 @@ module Scanners
           elsif match = scan(/ \\ | $ /x)
             encoder.end_group state
             state = :initial
-            encoder.text_token match, :error
+            encoder.text_token match, :error unless match.empty?
           else
             raise_inspect "else case \" reached; %p not handled." % peek(1), encoder
           end
diff --git a/lib/coderay/scanners/java_script.rb b/lib/coderay/scanners/java_script.rb
index 43ecb18..92e3dfa 100644
--- a/lib/coderay/scanners/java_script.rb
+++ b/lib/coderay/scanners/java_script.rb
@@ -175,11 +175,11 @@ module Scanners
             encoder.text_token match, :content
           elsif match = scan(/ \\ | $ /x)
             encoder.end_group state
-            encoder.text_token match, :error
+            encoder.text_token match, :error unless match.empty?
             key_expected = value_expected = false
             state = :initial
           else
-            raise_inspect "else case \" reached; %p not handled." % peek(1), encoder
+            raise_inspect "else case #{string_delimiter} reached; %p not handled." % peek(1), encoder
           end
           
         else
diff --git a/lib/coderay/scanners/json.rb b/lib/coderay/scanners/json.rb
index 0c90c34..4e0f462 100644
--- a/lib/coderay/scanners/json.rb
+++ b/lib/coderay/scanners/json.rb
@@ -70,7 +70,7 @@ module Scanners
             encoder.text_token match, :content
           elsif match = scan(/ \\ | $ /x)
             encoder.end_group state
-            encoder.text_token match, :error
+            encoder.text_token match, :error unless match.empty?
             state = :initial
           else
             raise_inspect "else case \" reached; %p not handled." % peek(1), encoder
diff --git a/lib/coderay/scanners/python.rb b/lib/coderay/scanners/python.rb
index cbdbbdb..a9492ab 100644
--- a/lib/coderay/scanners/python.rb
+++ b/lib/coderay/scanners/python.rb
@@ -133,7 +133,7 @@ module Scanners
           elsif match = scan(/ \\ | $ /x)
             encoder.end_group string_type
             string_type = nil
-            encoder.text_token match, :error
+            encoder.text_token match, :error unless match.empty?
             state = :initial
           else
             raise_inspect "else case \" reached; %p not handled." % peek(1), encoder, state
diff --git a/lib/coderay/scanners/sass.rb b/lib/coderay/scanners/sass.rb
new file mode 100644
index 0000000..0eb2caa
--- /dev/null
+++ b/lib/coderay/scanners/sass.rb
@@ -0,0 +1,207 @@
+module CodeRay
+module Scanners
+  
+  # A scanner for Sass.
+  class Sass < CSS
+    
+    register_for :sass
+    file_extension 'sass'
+    
+    SASS_FUNCTION = /(?:inline-image|linear-gradient|color-stops|mix|lighten|darken|rotate|image-url|image-width|image-height|sprite-url|sprite-path|sprite-file|sprite-map|sprite-position|sprite|unquote|join|round|ceil|floor|nth)/
+    
+    STRING_CONTENT_PATTERN = {
+      "'" => /(?:[^\n\'\#]+|\\\n|#{RE::Escape}|#(?!\{))+/,
+      '"' => /(?:[^\n\"\#]+|\\\n|#{RE::Escape}|#(?!\{))+/,
+    }
+    
+  protected
+    
+    def setup
+      @state = :initial
+    end
+    
+    def scan_tokens encoder, options
+      states = Array(options[:state] || @state)
+      string_delimiter = nil
+      
+      until eos?
+        
+        if match = scan(/\s+/)
+          encoder.text_token match, :space
+          value_expected = false if match.index(/\n/)
+          
+        elsif states.last == :sass_inline && (match = scan(/\}/))
+          encoder.text_token match, :inline_delimiter
+          encoder.end_group :inline
+          states.pop
+        
+        elsif case states.last
+          when :initial, :media, :sass_inline
+            if match = scan(/(?>#{RE::Ident})(?!\()/ox)
+              encoder.text_token match, value_expected ? :value : (check(/.*:/) ? :key : :type)
+              next
+            elsif !value_expected && (match = scan(/\*/))
+              encoder.text_token match, :type
+              next
+            elsif match = scan(RE::Class)
+              encoder.text_token match, :class
+              next
+            elsif match = scan(RE::Id)
+              encoder.text_token match, :constant
+              next
+            elsif match = scan(RE::PseudoClass)
+              encoder.text_token match, :pseudo_class
+              next
+            elsif match = scan(RE::AttributeSelector)
+              # TODO: Improve highlighting inside of attribute selectors.
+              encoder.text_token match[0,1], :operator
+              encoder.text_token match[1..-2], :attribute_name if match.size > 2
+              encoder.text_token match[-1,1], :operator if match[-1] == ?]
+              next
+            elsif match = scan(/(\=|@mixin +)#{RE::Ident}/o)
+              encoder.text_token match, :function
+              next
+            elsif match = scan(/@media/)
+              encoder.text_token match, :directive
+              # states.push :media_before_name
+              next
+            end
+          
+          when :block
+            if match = scan(/(?>#{RE::Ident})(?!\()/ox)
+              if value_expected
+                encoder.text_token match, :value
+              else
+                encoder.text_token match, :key
+              end
+              next
+            end
+            
+          when :string
+            if match = scan(STRING_CONTENT_PATTERN[string_delimiter])
+              encoder.text_token match, :content
+            elsif match = scan(/['"]/)
+              encoder.text_token match, :delimiter
+              encoder.end_group :string
+              string_delimiter = nil
+              states.pop
+            elsif match = scan(/#\{/)
+              encoder.begin_group :inline
+              encoder.text_token match, :inline_delimiter
+              states.push :sass_inline
+            elsif match = scan(/ \\ | $ /x)
+              encoder.end_group state
+              encoder.text_token match, :error unless match.empty?
+              states.pop
+            else
+              raise_inspect "else case #{string_delimiter} reached; %p not handled." % peek(1), encoder
+            end
+          
+          else
+            #:nocov:
+            raise_inspect 'Unknown state', encoder
+            #:nocov:
+            
+          end
+          
+        elsif match = scan(/\$#{RE::Ident}/o)
+          encoder.text_token match, :variable
+          next
+        
+        elsif match = scan(/&/)
+          encoder.text_token match, :local_variable
+          
+        elsif match = scan(/\+#{RE::Ident}/o)
+          encoder.text_token match, :include
+          value_expected = true
+          
+        elsif match = scan(/\/\*(?:.*?\*\/|.*)|\/\/.*/)
+          encoder.text_token match, :comment
+          
+        elsif match = scan(/#\{/)
+          encoder.begin_group :inline
+          encoder.text_token match, :inline_delimiter
+          states.push :sass_inline
+          
+        elsif match = scan(/\{/)
+          value_expected = false
+          encoder.text_token match, :operator
+          states.push :block
+          
+        elsif match = scan(/\}/)
+          value_expected = false
+          encoder.text_token match, :operator
+          if states.last == :block || states.last == :media
+            states.pop
+          end
+          
+        elsif match = scan(/['"]/)
+          encoder.begin_group :string
+          string_delimiter = match
+          encoder.text_token match, :delimiter
+          if states.include? :sass_inline
+            content = scan_until(/(?=#{string_delimiter}|\}|\z)/)
+            encoder.text_token content, :content unless content.empty?
+            encoder.text_token string_delimiter, :delimiter if scan(/#{string_delimiter}/)
+            encoder.end_group :string
+          else
+            states.push :string
+          end
+          
+        elsif match = scan(/#{SASS_FUNCTION}/o)
+          encoder.text_token match, :predefined
+          
+        elsif match = scan(/#{RE::Function}/o)
+          encoder.begin_group :function
+          start = match[/^[-\w]+\(/]
+          encoder.text_token start, :delimiter
+          if match[-1] == ?)
+            encoder.text_token match[start.size..-2], :content
+            encoder.text_token ')', :delimiter
+          else
+            encoder.text_token match[start.size..-1], :content
+          end
+          encoder.end_group :function
+          
+        elsif match = scan(/(?: #{RE::Dimension} | #{RE::Percentage} | #{RE::Num} )/ox)
+          encoder.text_token match, :float
+          
+        elsif match = scan(/#{RE::HexColor}/o)
+          encoder.text_token match, :color
+          
+        elsif match = scan(/! *(?:important|optional)/)
+          encoder.text_token match, :important
+          
+        elsif match = scan(/(?:rgb|hsl)a?\([^()\n]*\)?/)
+          encoder.text_token match, :color
+          
+        elsif match = scan(/@else if\b|#{RE::AtKeyword}/)
+          encoder.text_token match, :directive
+          value_expected = true
+          
+        elsif match = scan(/ == | != | [-+*\/>~:;,.=()] /x)
+          if match == ':'
+            value_expected = true
+          elsif match == ';'
+            value_expected = false
+          end
+          encoder.text_token match, :operator
+          
+        else
+          encoder.text_token getch, :error
+          
+        end
+        
+      end
+      
+      if options[:keep_state]
+        @state = states
+      end
+      
+      encoder
+    end
+    
+  end
+  
+end
+end
diff --git a/lib/coderay/scanners/sql.rb b/lib/coderay/scanners/sql.rb
index bcbffd5..b757278 100644
--- a/lib/coderay/scanners/sql.rb
+++ b/lib/coderay/scanners/sql.rb
@@ -148,7 +148,7 @@ module CodeRay module Scanners
               encoder.text_token string_content, :content
               string_content = ''
             end
-            encoder.text_token match, :error
+            encoder.text_token match, :error unless match.empty?
             state = :initial
           else
             raise "else case \" reached; %p not handled." % peek(1), encoder
diff --git a/lib/coderay/styles/alpha.rb b/lib/coderay/styles/alpha.rb
index 8506d10..1f073b6 100644
--- a/lib/coderay/styles/alpha.rb
+++ b/lib/coderay/styles/alpha.rb
@@ -78,6 +78,7 @@ table.CodeRay td { padding: 2px 4px; vertical-align: top; }
 .exception { color:#C00; font-weight:bold }
 .float { color:#60E }
 .function { color:#06B; font-weight:bold }
+.function .delimiter { color:#024; font-weight:bold }
 .global-variable { color:#d70 }
 .hex { color:#02b }
 .imaginary { color:#f00 }
@@ -86,6 +87,7 @@ table.CodeRay td { padding: 2px 4px; vertical-align: top; }
 .inline-delimiter { font-weight: bold; color: #666 }
 .instance-variable { color:#33B }
 .integer  { color:#00D }
+.important { color:#D00 }
 .key .char { color: #60f }
 .key .delimiter { color: #404 }
 .key { color: #606 }
diff --git a/lib/coderay/token_kinds.rb b/lib/coderay/token_kinds.rb
index 3b8d07e..8db8530 100755
--- a/lib/coderay/token_kinds.rb
+++ b/lib/coderay/token_kinds.rb
@@ -85,6 +85,4 @@ module CodeRay
   TokenKinds[:method]    = TokenKinds[:function]
   TokenKinds[:escape]    = TokenKinds[:delimiter]
   TokenKinds[:docstring] = TokenKinds[:comment]
-  
-  TokenKinds.freeze
 end
diff --git a/lib/coderay/tokens.rb b/lib/coderay/tokens.rb
index c747017..6957d69 100644
--- a/lib/coderay/tokens.rb
+++ b/lib/coderay/tokens.rb
@@ -93,6 +93,7 @@ module CodeRay
     # This method is used by @Scanner#tokenize@ when called with an Array
     # of source strings. The Diff encoder uses it for inline highlighting.
     def split_into_parts *sizes
+      return Array.new(sizes.size) { Tokens.new } if size == 2 && first == ''
       parts = []
       opened = []
       content = nil
diff --git a/lib/coderay/version.rb b/lib/coderay/version.rb
index bfb5f24..4b4f085 100644
--- a/lib/coderay/version.rb
+++ b/lib/coderay/version.rb
@@ -1,3 +1,3 @@
 module CodeRay
-  VERSION = '1.0.9'
+  VERSION = '1.1.0'
 end