llm-shell 0.9.2 → 0.10.0

data/lib/llm/shell/internal/coderay/lib/coderay/scanners/groovy.rb

@@ -0,0 +1,268 @@
+module CodeRay
+module Scanners
+  
+  load :java
+  
+  # Scanner for Groovy.
+  class Groovy < Java
+    
+    register_for :groovy
+    
+    # TODO: check list of keywords
+    GROOVY_KEYWORDS = %w[
+      as assert def in
+    ]  # :nodoc:
+    KEYWORDS_EXPECTING_VALUE = WordList.new.add %w[
+      case instanceof new return throw typeof while as assert in
+    ]  # :nodoc:
+    GROOVY_MAGIC_VARIABLES = %w[ it ]  # :nodoc:
+    
+    IDENT_KIND = Java::IDENT_KIND.dup.
+      add(GROOVY_KEYWORDS, :keyword).
+      add(GROOVY_MAGIC_VARIABLES, :local_variable)  # :nodoc:
+    
+    ESCAPE = / [bfnrtv$\n\\'"] | x[a-fA-F0-9]{1,2} | [0-7]{1,3} /x  # :nodoc:
+    UNICODE_ESCAPE = / u[a-fA-F0-9]{4} /x  # :nodoc: no 4-byte unicode chars? U[a-fA-F0-9]{8}
+    REGEXP_ESCAPE = / [bfnrtv\n\\'"] | x[a-fA-F0-9]{1,2} | [0-7]{1,3} | \d | [bBdDsSwW\/] /x  # :nodoc:
+    
+    # TODO: interpretation inside ', ", /
+    STRING_CONTENT_PATTERN = {
+      "'" => /(?>\\[^\\'\n]+|[^\\'\n]+)+/,
+      '"' => /[^\\$"\n]+/,
+      "'''" => /(?>[^\\']+|'(?!''))+/,
+      '"""' => /(?>[^\\$"]+|"(?!""))+/,
+      '/' => /[^\\$\/\n]+/,
+    }  # :nodoc:
+    
+  protected
+    
+    def setup
+      @state = :initial
+    end
+    
+    def scan_tokens encoder, options
+      state = options[:state] || @state
+      inline_block_stack = []
+      inline_block_paren_depth = nil
+      string_delimiter = nil
+      import_clause = class_name_follows = last_token = after_def = false
+      value_expected = true
+      
+      until eos?
+        
+        case state
+        
+        when :initial
+          
+          if match = scan(/ \s+ | \\\n /x)
+            encoder.text_token match, :space
+            if match.index ?\n
+              import_clause = after_def = false
+              value_expected = true unless value_expected
+            end
+            next
+          
+          elsif match = scan(%r! // [^\n\\]* (?: \\. [^\n\\]* )* | /\* (?: .*? \*/ | .* ) !mx)
+            value_expected = true
+            after_def = false
+            encoder.text_token match, :comment
+          
+          elsif bol? && match = scan(/ \#!.* /x)
+            encoder.text_token match, :doctype
+          
+          elsif import_clause && match = scan(/ (?!as) #{IDENT} (?: \. #{IDENT} )* (?: \.\* )? /ox)
+            after_def = value_expected = false
+            encoder.text_token match, :include
+          
+          elsif match = scan(/ #{IDENT} | \[\] /ox)
+            kind = IDENT_KIND[match]
+            value_expected = (kind == :keyword) && KEYWORDS_EXPECTING_VALUE[match]
+            if last_token == '.'
+              kind = :ident
+            elsif class_name_follows
+              kind = :class
+              class_name_follows = false
+            elsif after_def && check(/\s*[({]/)
+              kind = :method
+              after_def = false
+            elsif kind == :ident && last_token != '?' && check(/:/)
+              kind = :key
+            else
+              class_name_follows = true if match == 'class' || (import_clause && match == 'as')
+              import_clause = match == 'import'
+              after_def = true if match == 'def'
+            end
+            encoder.text_token match, kind
+          
+          elsif match = scan(/;/)
+            import_clause = after_def = false
+            value_expected = true
+            encoder.text_token match, :operator
+          
+          elsif match = scan(/\{/)
+            class_name_follows = after_def = false
+            value_expected = true
+            encoder.text_token match, :operator
+            if !inline_block_stack.empty?
+              inline_block_paren_depth += 1
+            end
+          
+          # TODO: ~'...', ~"..." and ~/.../ style regexps
+          elsif match = scan(/ \.\.<? | \*?\.(?!\d)@? | \.& | \?:? | [,?:(\[] | -[->] | \+\+ |
+              && | \|\| | \*\*=? | ==?~ | <=?>? | [-+*%^~&|>=!]=? | <<<?=? | >>>?=? /x)
+            value_expected = true
+            value_expected = :regexp if match == '~'
+            after_def = false
+            encoder.text_token match, :operator
+          
+          elsif match = scan(/ [)\]}] /x)
+            value_expected = after_def = false
+            if !inline_block_stack.empty? && match == '}'
+              inline_block_paren_depth -= 1
+              if inline_block_paren_depth == 0  # closing brace of inline block reached
+                encoder.text_token match, :inline_delimiter
+                encoder.end_group :inline
+                state, string_delimiter, inline_block_paren_depth = inline_block_stack.pop
+                next
+              end
+            end
+            encoder.text_token match, :operator
+          
+          elsif check(/[\d.]/)
+            after_def = value_expected = false
+            if match = scan(/0[xX][0-9A-Fa-f]+/)
+              encoder.text_token match, :hex
+            elsif match = scan(/(?>0[0-7]+)(?![89.eEfF])/)
+              encoder.text_token match, :octal
+            elsif match = scan(/\d+[fFdD]|\d*\.\d+(?:[eE][+-]?\d+)?[fFdD]?|\d+[eE][+-]?\d+[fFdD]?/)
+              encoder.text_token match, :float
+            elsif match = scan(/\d+[lLgG]?/)
+              encoder.text_token match, :integer
+            end
+            
+          elsif match = scan(/'''|"""/)
+            after_def = value_expected = false
+            state = :multiline_string
+            encoder.begin_group :string
+            string_delimiter = match
+            encoder.text_token match, :delimiter
+            
+          # TODO: record.'name' syntax
+          elsif match = scan(/["']/)
+            after_def = value_expected = false
+            state = match == '/' ? :regexp : :string
+            encoder.begin_group state
+            string_delimiter = match
+            encoder.text_token match, :delimiter
+            
+          elsif value_expected && match = scan(/\//)
+            after_def = value_expected = false
+            encoder.begin_group :regexp
+            state = :regexp
+            string_delimiter = '/'
+            encoder.text_token match, :delimiter
+            
+          elsif match = scan(/ @ #{IDENT} /ox)
+            after_def = value_expected = false
+            encoder.text_token match, :annotation
+            
+          elsif match = scan(/\//)
+            after_def = false
+            value_expected = true
+            encoder.text_token match, :operator
+            
+          else
+            encoder.text_token getch, :error
+            
+          end
+          
+        when :string, :regexp, :multiline_string
+          if match = scan(STRING_CONTENT_PATTERN[string_delimiter])
+            encoder.text_token match, :content
+            
+          elsif match = scan(state == :multiline_string ? /'''|"""/ : /["'\/]/)
+            encoder.text_token match, :delimiter
+            if state == :regexp
+              # TODO: regexp modifiers? s, m, x, i?
+              modifiers = scan(/[ix]+/)
+              encoder.text_token modifiers, :modifier if modifiers && !modifiers.empty?
+            end
+            state = :string if state == :multiline_string
+            encoder.end_group state
+            string_delimiter = nil
+            after_def = value_expected = false
+            state = :initial
+            next
+            
+          elsif (state == :string || state == :multiline_string) &&
+              (match = scan(/ \\ (?: #{ESCAPE} | #{UNICODE_ESCAPE} ) /mox))
+            if string_delimiter[0] == ?' && !(match == "\\\\" || match == "\\'")
+              encoder.text_token match, :content
+            else
+              encoder.text_token match, :char
+            end
+          elsif state == :regexp && match = scan(/ \\ (?: #{REGEXP_ESCAPE} | #{UNICODE_ESCAPE} ) /mox)
+            encoder.text_token match, :char
+            
+          elsif match = scan(/ \$ #{IDENT} /mox)
+            encoder.begin_group :inline
+            encoder.text_token '$', :inline_delimiter
+            match = match[1..-1]
+            encoder.text_token match, IDENT_KIND[match]
+            encoder.end_group :inline
+            next
+          elsif match = scan(/ \$ \{ /x)
+            encoder.begin_group :inline
+            encoder.text_token match, :inline_delimiter
+            inline_block_stack << [state, string_delimiter, inline_block_paren_depth]
+            inline_block_paren_depth = 1
+            state = :initial
+            next
+            
+          elsif match = scan(/ \$ /mx)
+            encoder.text_token match, :content
+            
+          elsif match = scan(/ \\. /mx)
+            encoder.text_token match, :content  # TODO: Shouldn't this be :error?
+            
+          elsif match = scan(/ \\ | \n /x)
+            encoder.end_group state == :regexp ? :regexp : :string
+            encoder.text_token match, :error
+            after_def = value_expected = false
+            state = :initial
+            
+          else
+            raise_inspect "else case \" reached; %p not handled." % peek(1), encoder
+            
+          end
+          
+        else
+          raise_inspect 'Unknown state', encoder
+          
+        end
+        
+        last_token = match unless [:space, :comment, :doctype].include? kind
+        
+      end
+      
+      if [:multiline_string, :string, :regexp].include? state
+        encoder.end_group state == :regexp ? :regexp : :string
+      end
+      
+      if options[:keep_state]
+        @state = state
+      end
+      
+      until inline_block_stack.empty?
+        state, = *inline_block_stack.pop
+        encoder.end_group :inline
+        encoder.end_group state == :regexp ? :regexp : :string
+      end
+      
+      encoder
+    end
+    
+  end
+  
+end
+end

data/lib/llm/shell/internal/coderay/lib/coderay/scanners/haml.rb

@@ -0,0 +1,168 @@
+module CodeRay
+module Scanners
+  
+  load :ruby
+  load :html
+  load :java_script
+  
+  class HAML < Scanner
+    
+    register_for :haml
+    title 'HAML Template'
+    
+    KINDS_NOT_LOC = HTML::KINDS_NOT_LOC
+    
+  protected
+    
+    def setup
+      super
+      @ruby_scanner          = CodeRay.scanner :ruby, :tokens => @tokens, :keep_tokens => true
+      @embedded_ruby_scanner = CodeRay.scanner :ruby, :tokens => @tokens, :keep_tokens => true, :state => @ruby_scanner.interpreted_string_state
+      @html_scanner          = CodeRay.scanner :html, :tokens => @tokens, :keep_tokens => true
+    end
+    
+    def scan_tokens encoder, options
+      
+      match = nil
+      code = ''
+      
+      until eos?
+        
+        if bol?
+          if match = scan(/!!!.*/)
+            encoder.text_token match, :doctype
+            next
+          end
+          
+          if match = scan(/(?>( *)(\/(?!\[if)|-\#|:javascript|:ruby|:\w+) *)(?=\n)/)
+            encoder.text_token match, :comment
+            
+            code = self[2]
+            if match = scan(/(?:\n+#{self[1]} .*)+/)
+              case code
+              when '/', '-#'
+                encoder.text_token match, :comment
+              when ':javascript'
+                # TODO: recognize #{...} snippets inside JavaScript
+                @java_script_scanner ||= CodeRay.scanner :java_script, :tokens => @tokens, :keep_tokens => true
+                @java_script_scanner.tokenize match, :tokens => encoder
+              when ':ruby'
+                @ruby_scanner.tokenize match, :tokens => encoder
+              when /:\w+/
+                encoder.text_token match, :comment
+              else
+                raise 'else-case reached: %p' % [code]
+              end
+            end
+          end
+          
+          if match = scan(/ +/)
+            encoder.text_token match, :space
+          end
+          
+          if match = scan(/\/.*/)
+            encoder.text_token match, :comment
+            next
+          end
+          
+          if match = scan(/\\/)
+            encoder.text_token match, :plain
+            if match = scan(/.+/)
+              @html_scanner.tokenize match, :tokens => encoder
+            end
+            next
+          end
+          
+          tag = false
+          
+          if match = scan(/%[-\w:]+\/?/)
+            encoder.text_token match, :tag
+            # if match = scan(/( +)(.+)/)
+            #   encoder.text_token self[1], :space
+            #   @embedded_ruby_scanner.tokenize self[2], :tokens => encoder
+            # end
+            tag = true
+          end
+          
+          while match = scan(/([.#])[-\w]*\w/)
+            encoder.text_token match, self[1] == '#' ? :constant : :class
+            tag = true
+          end
+          
+          if tag && match = scan(/(\()([^)]+)?(\))?/)
+            # TODO: recognize title=@title, class="widget_#{@widget.number}"
+            encoder.text_token self[1], :plain
+            @html_scanner.tokenize self[2], :tokens => encoder, :state => :attribute if self[2]
+            encoder.text_token self[3], :plain if self[3]
+          end
+          
+          if tag && match = scan(/\{/)
+            encoder.text_token match, :plain
+            
+            code = ''
+            level = 1
+            while true
+              code << scan(/([^\{\},\n]|, *\n?)*/)
+              case match = getch
+              when '{'
+                level += 1
+                code << match
+              when '}'
+                level -= 1
+                if level > 0
+                  code << match
+                else
+                  break
+                end
+              when "\n", ",", nil
+                break
+              end
+            end
+            @ruby_scanner.tokenize code, :tokens => encoder unless code.empty?
+            
+            encoder.text_token match, :plain if match
+          end
+          
+          if tag && match = scan(/(\[)([^\]\n]+)?(\])?/)
+            encoder.text_token self[1], :plain
+            @ruby_scanner.tokenize self[2], :tokens => encoder if self[2]
+            encoder.text_token self[3], :plain if self[3]
+          end
+          
+          if tag && match = scan(/\//)
+            encoder.text_token match, :tag
+          end
+          
+          if scan(/(>?<?[-=]|[&!]=|(& |!)|~)( *)([^,\n\|]+(?:(, *|\|(?=.|\n.*\|$))\n?[^,\n\|]*)*)?/)
+            encoder.text_token self[1] + self[3], :plain
+            if self[4]
+              if self[2]
+                @embedded_ruby_scanner.tokenize self[4], :tokens => encoder
+              else
+                @ruby_scanner.tokenize self[4], :tokens => encoder
+              end
+            end
+          elsif match = scan(/((?:<|><?)(?![!?\/\w]))?(.+)?/)
+            encoder.text_token self[1], :plain if self[1]
+            # TODO: recognize #{...} snippets
+            @html_scanner.tokenize self[2], :tokens => encoder if self[2]
+          end
+          
+        elsif match = scan(/.+/)
+          @html_scanner.tokenize match, :tokens => encoder
+          
+        end
+        
+        if match = scan(/\n/)
+          encoder.text_token match, :space
+        end
+      end
+      
+      encoder
+      
+    end
+    
+  end
+  
+end
+end

data/lib/llm/shell/internal/coderay/lib/coderay/scanners/html.rb

@@ -0,0 +1,275 @@
+module CodeRay
+module Scanners
+  
+  # HTML Scanner
+  # 
+  # Alias: +xhtml+
+  # 
+  # See also: Scanners::XML
+  class HTML < Scanner
+    
+    register_for :html
+    
+    KINDS_NOT_LOC = [
+      :comment, :doctype, :preprocessor,
+      :tag, :attribute_name, :operator,
+      :attribute_value, :string,
+      :plain, :entity, :error,
+    ]  # :nodoc:
+    
+    EVENT_ATTRIBUTES = %w(
+      onabort onafterprint onbeforeprint onbeforeunload onblur oncanplay
+      oncanplaythrough onchange onclick oncontextmenu oncuechange ondblclick
+      ondrag ondragdrop ondragend ondragenter ondragleave ondragover
+      ondragstart ondrop ondurationchange onemptied onended onerror onfocus
+      onformchange onforminput onhashchange oninput oninvalid onkeydown
+      onkeypress onkeyup onload onloadeddata onloadedmetadata onloadstart
+      onmessage onmousedown onmousemove onmouseout onmouseover onmouseup
+      onmousewheel onmove onoffline ononline onpagehide onpageshow onpause
+      onplay onplaying onpopstate onprogress onratechange onreadystatechange
+      onredo onreset onresize onscroll onseeked onseeking onselect onshow
+      onstalled onstorage onsubmit onsuspend ontimeupdate onundo onunload
+      onvolumechange onwaiting
+    )
+    
+    IN_ATTRIBUTE = WordList::CaseIgnoring.new(nil).
+      add(EVENT_ATTRIBUTES, :script).
+      add(['style'], :style)
+    
+    ATTR_NAME = /[\w.:-]+/  # :nodoc:
+    TAG_END = /\/?>/  # :nodoc:
+    HEX = /[0-9a-fA-F]/  # :nodoc:
+    ENTITY = /
+      &
+      (?:
+        \w+
+      |
+        \#
+        (?:
+          \d+
+        |
+          x#{HEX}+
+        )
+      )
+      ;
+    /ox  # :nodoc:
+    
+    PLAIN_STRING_CONTENT = {
+      "'" => /[^&'>\n]+/,
+      '"' => /[^&">\n]+/,
+    }  # :nodoc:
+    
+    def reset
+      super
+      @state = :initial
+      @plain_string_content = nil
+    end
+    
+  protected
+    
+    def setup
+      @state = :initial
+      @plain_string_content = nil
+      @in_tag = nil
+    end
+    
+    def scan_java_script encoder, code
+      if code && !code.empty?
+        @java_script_scanner ||= Scanners::JavaScript.new '', :keep_tokens => true
+        @java_script_scanner.tokenize code, :tokens => encoder
+      end
+    end
+    
+    def scan_css encoder, code, state = [:initial]
+      if code && !code.empty?
+        @css_scanner ||= Scanners::CSS.new '', :keep_tokens => true
+        @css_scanner.tokenize code, :tokens => encoder, :state => state
+      end
+    end
+    
+    def scan_tokens encoder, options
+      state = options[:state] || @state
+      plain_string_content = @plain_string_content
+      in_tag = @in_tag
+      in_attribute = nil
+      
+      encoder.begin_group :string if state == :attribute_value_string
+      
+      until eos?
+        
+        if state != :in_special_tag && match = scan(/\s+/m)
+          encoder.text_token match, :space
+          
+        else
+          
+          case state
+          
+          when :initial
+            if match = scan(/<!\[CDATA\[/)
+              encoder.text_token match, :inline_delimiter
+              if match = scan(/.*?\]\]>/m)
+                encoder.text_token match[0..-4], :plain
+                encoder.text_token ']]>', :inline_delimiter
+              elsif match = scan(/.+/)
+                encoder.text_token match, :error
+              end
+            elsif match = scan(/<!--(?:.*?-->|.*)/m)
+              encoder.text_token match, :comment
+            elsif match = scan(/<!(\w+)(?:.*?>|.*)|\]>/m)
+              encoder.text_token match, :doctype
+            elsif match = scan(/<\?xml(?:.*?\?>|.*)/m)
+              encoder.text_token match, :preprocessor
+            elsif match = scan(/<\?(?:.*?\?>|.*)/m)
+              encoder.text_token match, :comment
+            elsif match = scan(/<\/[-\w.:]*>?/m)
+              in_tag = nil
+              encoder.text_token match, :tag
+            elsif match = scan(/<(?:(script|style)|[-\w.:]+)(>)?/m)
+              encoder.text_token match, :tag
+              in_tag = self[1]
+              if self[2]
+                state = :in_special_tag if in_tag
+              else
+                state = :attribute
+              end
+            elsif match = scan(/[^<>&]+/)
+              encoder.text_token match, :plain
+            elsif match = scan(/#{ENTITY}/ox)
+              encoder.text_token match, :entity
+            elsif match = scan(/[<>&]/)
+              in_tag = nil
+              encoder.text_token match, :error
+            else
+              raise_inspect '[BUG] else-case reached with state %p' % [state], encoder
+            end
+            
+          when :attribute
+            if match = scan(/#{TAG_END}/o)
+              encoder.text_token match, :tag
+              in_attribute = nil
+              if in_tag
+                state = :in_special_tag
+              else
+                state = :initial
+              end
+            elsif match = scan(/#{ATTR_NAME}/o)
+              in_attribute = IN_ATTRIBUTE[match]
+              encoder.text_token match, :attribute_name
+              state = :attribute_equal
+            else
+              in_tag = nil
+              encoder.text_token getch, :error
+            end
+            
+          when :attribute_equal
+            if match = scan(/=/)  #/
+              encoder.text_token match, :operator
+              state = :attribute_value
+            else
+              state = :attribute
+              next
+            end
+            
+          when :attribute_value
+            if match = scan(/#{ATTR_NAME}/o)
+              encoder.text_token match, :attribute_value
+              state = :attribute
+            elsif match = scan(/["']/)
+              if in_attribute == :script || in_attribute == :style
+                encoder.begin_group :string
+                encoder.text_token match, :delimiter
+                if scan(/javascript:[ \t]*/)
+                  encoder.text_token matched, :comment
+                end
+                code = scan_until(match == '"' ? /(?="|\z)/ : /(?='|\z)/)
+                if in_attribute == :script
+                  scan_java_script encoder, code
+                else
+                  scan_css encoder, code, [:block]
+                end
+                match = scan(/["']/)
+                encoder.text_token match, :delimiter if match
+                encoder.end_group :string
+                state = :attribute
+                in_attribute = nil
+              else
+                encoder.begin_group :string
+                state = :attribute_value_string
+                plain_string_content = PLAIN_STRING_CONTENT[match]
+                encoder.text_token match, :delimiter
+              end
+            elsif match = scan(/#{TAG_END}/o)
+              encoder.text_token match, :tag
+              state = :initial
+            else
+              encoder.text_token getch, :error
+            end
+            
+          when :attribute_value_string
+            if match = scan(plain_string_content)
+              encoder.text_token match, :content
+            elsif match = scan(/['"]/)
+              encoder.text_token match, :delimiter
+              encoder.end_group :string
+              state = :attribute
+            elsif match = scan(/#{ENTITY}/ox)
+              encoder.text_token match, :entity
+            elsif match = scan(/&/)
+              encoder.text_token match, :content
+            elsif match = scan(/[\n>]/)
+              encoder.end_group :string
+              state = :initial
+              encoder.text_token match, :error
+            end
+            
+          when :in_special_tag
+            case in_tag
+            when 'script', 'style'
+              encoder.text_token match, :space if match = scan(/[ \t]*\n/)
+              if scan(/(\s*<!--)(?:(.*?)(-->)|(.*))/m)
+                code = self[2] || self[4]
+                closing = self[3]
+                encoder.text_token self[1], :comment
+              else
+                code = scan_until(/(?=(?:\n\s*)?<\/#{in_tag}>)|\z/)
+                closing = false
+              end
+              unless code.empty?
+                encoder.begin_group :inline
+                if in_tag == 'script'
+                  scan_java_script encoder, code
+                else
+                  scan_css encoder, code
+                end
+                encoder.end_group :inline
+              end
+              encoder.text_token closing, :comment if closing
+              state = :initial
+            else
+              raise 'unknown special tag: %p' % [in_tag]
+            end
+            
+          else
+            raise_inspect 'Unknown state: %p' % [state], encoder
+            
+          end
+          
+        end
+        
+      end
+      
+      if options[:keep_state]
+        @state = state
+        @plain_string_content = plain_string_content
+        @in_tag = in_tag
+      end
+      
+      encoder.end_group :string if state == :attribute_value_string
+      
+      encoder
+    end
+    
+  end
+  
+end
+end