halyard-puppet-lint 1.2.0

data/lib/puppet-lint/data.rb

@@ -0,0 +1,521 @@
+require 'singleton'
+require 'set'
+
+# Public: A singleton class storing all the information about the manifest
+# being analysed.
+class PuppetLint::Data
+  include Singleton
+
+  class << self
+    # Internal: Get/Set the full expanded path to the manifest file being
+    # checked.
+    attr_reader :path, :fullpath, :filename
+
+    # Internal: Get/Set the raw manifest data, split by \n.
+    attr_accessor :manifest_lines
+
+    # Internal: Store the tokenised manifest.
+    #
+    # tokens - The Array of PuppetLint::Lexer::Token objects to store.
+    #
+    # Returns nothing.
+    def tokens=(tokens)
+      @tokens = tokens
+      @title_tokens = nil
+      @resource_indexes = nil
+      @class_indexes = nil
+      @defined_type_indexes = nil
+      @function_indexes = nil
+      @array_indexes = nil
+      @hash_indexes = nil
+      @defaults_indexes = nil
+    end
+
+    # Public: Get the tokenised manifest.
+    #
+    # Returns an Array of PuppetLint::Lexer::Token objects.
+    def tokens
+      calling_method = begin
+        caller[0][/`.*'/][1..-2]
+      rescue NoMethodError
+        caller[1][/`.*'/][1..-2]
+      end
+
+      if calling_method == 'check'
+        @tokens.dup
+      else
+        @tokens
+      end
+    end
+
+    # Internal: Store the path to the manifest file and populate fullpath and
+    # filename.
+    #
+    # val - The path to the file as a String.
+    #
+    # Returns nothing.
+    def path=(val)
+      @path = val
+      if val.nil?
+        @fullpath = nil
+        @filename = nil
+      else
+        @fullpath = File.expand_path(val, ENV['PWD'])
+        @filename = File.basename(val)
+      end
+    end
+
+    # Internal: Retrieve a list of tokens that represent resource titles
+    #
+    # Returns an Array of PuppetLint::Lexer::Token objects.
+    def title_tokens
+      @title_tokens ||= Proc.new do
+        result = []
+        tokens.each_index do |token_idx|
+          if tokens[token_idx].type == :COLON
+            # gather a list of tokens that are resource titles
+            if tokens[token_idx-1].type == :RBRACK
+              array_start_idx = tokens.rindex { |r|
+                r.type == :LBRACK
+              }
+              title_array_tokens = tokens[(array_start_idx + 1)..(token_idx - 2)]
+              result += title_array_tokens.select { |token|
+                {:STRING => true, :NAME => true}.include? token.type
+              }
+            else
+              next_token = tokens[token_idx].next_code_token
+              if next_token.type != :LBRACE
+                result << tokens[token_idx - 1]
+              end
+            end
+          end
+        end
+        result
+      end.call
+    end
+
+    # Internal: Calculate the positions of all resource declarations within the
+    # tokenised manifest. These positions only point to the content of the
+    # resource declarations, they do not include resource types or titles.
+    #
+    # Returns an Array of Hashes, each containing:
+    #   :start - An Integer position in the `tokens` Array pointing to the
+    #            first Token of a resource declaration.
+    #   :end   - An Integer position in the `tokens` Array pointing to the last
+    #            Token of a resource declaration.
+    def resource_indexes
+      @resource_indexes ||= Proc.new do
+        result = []
+        tokens.each_index do |token_idx|
+          if tokens[token_idx].type == :COLON
+            next_token = tokens[token_idx].next_code_token
+            depth = 1
+            if next_token && next_token.type != :LBRACE
+              tokens[(token_idx + 1)..-1].each_index do |idx|
+                real_idx = token_idx + idx + 1
+                if tokens[real_idx].type == :LBRACE
+                  depth += 1
+                elsif {:SEMIC => true, :RBRACE => true}.include? tokens[real_idx].type
+                  unless tokens[real_idx].type == :SEMIC && depth > 1
+                    depth -= 1
+                    if depth == 0
+                      result << {
+                        :start        => token_idx + 1,
+                        :end          => real_idx,
+                        :tokens       => tokens[(token_idx + 1)..real_idx],
+                        :type         => find_resource_type_token(token_idx),
+                        :param_tokens => find_resource_param_tokens(tokens[(token_idx + 1)..real_idx]),
+                      }
+                      break
+                    end
+                  end
+                end
+              end
+            end
+          end
+        end
+        result
+      end.call
+    end
+
+    # Internal: Find the Token representing the type of a resource definition.
+    #
+    # index - The Integer pointing to the start of the resource in the `tokens`
+    #         array.
+    #
+    # Returns a Token object.
+    def find_resource_type_token(index)
+      tokens[tokens[0..index].rindex { |token| token.type == :LBRACE }].prev_code_token
+    end
+
+    # Internal: Find all the Token objects representing the parameter names in
+    # a resource definition.
+    #
+    # resource_tokens - An Array of Token objects that comprise the resource
+    #                   definition.
+    #
+    # Returns an Array of Token objects.
+    def find_resource_param_tokens(resource_tokens)
+      resource_tokens.select { |token|
+        token.type == :NAME && token.next_code_token.type == :FARROW
+      }
+    end
+
+    # Internal: Calculate the positions of all class definitions within the
+    # `tokens` Array.
+    #
+    # Returns an Array of Hashes, each containing:
+    #   :start  - An Integer position in the `tokens` Array pointing to the
+    #             first Token of a class definition.
+    #   :end    - An Integer position in the `tokens` Array pointing to the last
+    #             Token of a class definition.
+    #   :tokens - An Array consisting of all the Token objects that make up the
+    #             class definition.
+    def class_indexes
+      @class_indexes ||= definition_indexes(:CLASS)
+    end
+
+    # Internal: Calculate the positions of all defined type definitions within
+    # the `tokens` Array.
+    #
+    # Returns an Array of Hashes, each containing:
+    #   :start  - An Integer position in the `tokens` Array pointing to the
+    #             first Token of a defined type definition.
+    #   :end    - An Integer position in the `tokens` Array pointing to the last
+    #             Token of a defined type definition.
+    #   :tokens - An Array consisting of all the Token objects that make up the
+    #             defined type.
+    def defined_type_indexes
+      @defined_type_indexes ||= definition_indexes(:DEFINE)
+    end
+
+    # Internal: Calculate the positions of all node definitions within the
+    # `tokens` Array.
+    #
+    # Returns an Array of Hashes, each containing:
+    #   :start  - An Integer position in the `tokens` Array pointing to the
+    #             first Token of a defined type definition.
+    #   :end    - An Integer position in the `tokens` Array pointing to the last
+    #             Token of a defined type definition.
+    #   :tokens - An Array consisting of all the Token objects that make up the
+    #             defined type.
+    def node_indexes
+      @node_indexes ||= definition_indexes(:NODE)
+    end
+
+    # Internal: Calculate the positions of all the specified defintion types
+    # within the `tokens` Array.
+    #
+    # Returns an Array of Hashes, each containing:
+    #   :start  - An Integer position in the `tokens` Array pointing to the
+    #             first Token of a definition.
+    #   :end    - An Integer position in the `tokens` Array pointing to the last
+    #             Token of a definition.
+    #   :tokens - An Array consisting of all the Token objects that make up the
+    #             definition.
+    def definition_indexes(type)
+      result = []
+      tokens.each_with_index do |token, i|
+        if token.type == type
+          brace_depth = 0
+          paren_depth = 0
+          in_params = false
+          inherited_class = nil
+          tokens[i+1..-1].each_with_index do |definition_token, j|
+            case definition_token.type
+            when :INHERITS
+              inherited_class = definition_token.next_code_token
+            when :LPAREN
+              in_params = true if paren_depth == 0
+              paren_depth += 1
+            when :RPAREN
+              in_params = false if paren_depth == 1
+              paren_depth -= 1
+            when :LBRACE
+              brace_depth += 1
+            when :RBRACE
+              brace_depth -= 1
+              if brace_depth == 0 && !in_params
+                if token.next_code_token.type != :LBRACE
+                  result << {
+                    :start           => i,
+                    :end             => i + j + 1,
+                    :tokens          => tokens[i..(i + j + 1)],
+                    :param_tokens    => param_tokens(tokens[i..(i + j + 1)]),
+                    :type            => type,
+                    :name_token      => token.next_code_token,
+                    :inherited_token => inherited_class,
+                  }
+                  break
+                end
+              end
+            end
+          end
+        end
+      end
+      result
+    end
+
+    # Internal: Calculate the positions of all function calls within
+    # `tokens` Array.
+    #
+    # Returns an Array of Hashes, each containing:
+    #   :start  - An Integer position in the `tokens` Array pointing to the
+    #             first Token of a function call
+    #   :end    - An Integer position in the `tokens` Array pointing to the last
+    #             Token of a function call
+    #   :tokens - An Array consisting of all the Token objects that make up the
+    #             function call.
+    def function_indexes
+      @function_indexes ||= Proc.new do
+        functions = []
+        tokens.each_with_index do |token, token_idx|
+          if token.type == :NAME && \
+            (token_idx == 0 || (token_idx == 1 && tokens[0].type == :WHITESPACE) || token.prev_token.type == :NEWLINE || token.prev_token.type == :INDENT || \
+              # function in a function
+              (token.prev_code_token && token.prev_code_token.type == :LPAREN))
+
+              # Hash key
+              next if token.next_code_token && token.next_code_token.type == :FARROW
+
+              level = 0
+              real_idx = 0
+              in_paren = false
+              tokens[token_idx+1..-1].each_with_index do |cur_token, cur_token_idx|
+                break if level == 0 && in_paren
+                break if level == 0 && cur_token.type == :NEWLINE
+
+                if cur_token.type == :LPAREN
+                  level += 1
+                  in_paren = true
+                end
+                level -= 1 if cur_token.type == :RPAREN
+                real_idx = token_idx + 1 + cur_token_idx
+              end
+
+              functions << {
+                :start  => token_idx,
+                :end    => real_idx,
+                :tokens => tokens[token_idx..real_idx],
+              }         
+          end
+        end
+        functions
+      end.call
+    end
+
+    # Internal: Calculate the positions of all array values within
+    # `tokens` Array.
+    #
+    # Returns an Array of Hashes, each containing:
+    #   :start  - An Integer position in the `tokens` Array pointing to the
+    #             first Token of an array value
+    #   :end    - An Integer position in the `tokens` Array pointing to the last
+    #             Token of an array value
+    #   :tokens - An Array consisting of all the Token objects that make up the
+    #             array value.
+    def array_indexes
+      @array_indexes ||= Proc.new do
+        arrays = []
+        tokens.each_with_index do |token, token_idx|
+          if token.type == :LBRACK
+            real_idx = 0
+            tokens[token_idx+1..-1].each_with_index do |cur_token, cur_token_idx|
+              real_idx = token_idx + 1 + cur_token_idx
+              break if cur_token.type == :RBRACK
+            end
+
+            # Ignore resource references
+            next if token.prev_code_token && \
+              token.prev_code_token.type == :CLASSREF
+            arrays << {
+              :start  => token_idx,
+              :end    => real_idx,
+              :tokens => tokens[token_idx..real_idx],
+            }
+          end
+        end
+        arrays
+      end.call
+    end
+
+    # Internal: Calculate the positions of all hash values within
+    # `tokens` Array.
+    #
+    # Returns an Array of Hashes, each containing:
+    #   :start  - An Integer position in the `tokens` Array pointing to the
+    #             first Token of an hash value
+    #   :end    - An Integer position in the `tokens` Array pointing to the last
+    #             Token of an hash value
+    #   :tokens - An Array consisting of all the Token objects that make up the
+    #             hash value.
+    def hash_indexes
+      @hash_indexes ||= Proc.new do
+        hashes = []
+        tokens.each_with_index do |token, token_idx|
+          next unless token.prev_code_token
+          next unless [:EQUALS, :ISEQUAL, :FARROW, :LPAREN].include? token.prev_code_token.type
+          if token.type == :LBRACE
+            level = 0
+            real_idx = 0
+            tokens[token_idx+1..-1].each_with_index do |cur_token, cur_token_idx|
+              real_idx = token_idx + 1 + cur_token_idx
+
+              level += 1 if cur_token.type == :LBRACE
+              level -= 1 if cur_token.type == :RBRACE
+              break if level < 0
+            end
+
+            hashes << {
+              :start  => token_idx,
+              :end    => real_idx,
+              :tokens => tokens[token_idx..real_idx],
+            }
+          end
+        end
+        hashes
+      end.call
+    end
+
+    # Internal: Calculate the positions of all defaults declarations within
+    # `tokens` Array.
+    #
+    # Returns an Array of Hashes, each containing:
+    #   :start  - An Integer position in the `tokens` Array pointing to the
+    #             first Token of the defaults declaration
+    #   :end    - An Integer position in the `tokens` Array pointing to the last
+    #             Token of the defaults declaration
+    #   :tokens - An Array consisting of all the Token objects that make up the
+    #             defaults declaration.
+    def defaults_indexes
+      @defaults_indexes ||= Proc.new do
+        defaults = []
+        tokens.each_with_index do |token, token_idx|
+          if token.type == :CLASSREF && token.next_code_token && \
+            token.next_code_token.type == :LBRACE
+            real_idx = 0
+
+            tokens[token_idx+1..-1].each_with_index do |cur_token, cur_token_idx|
+              real_idx = token_idx + 1 + cur_token_idx
+              break if cur_token.type == :RBRACE
+            end
+
+            defaults << {
+              :start  => token_idx,
+              :end    => real_idx,
+              :tokens => tokens[token_idx..real_idx],
+            }
+          end
+        end
+        defaults
+      end.call
+    end
+
+    # Internal: Finds all the tokens that make up the defined type or class
+    # definition parameters.
+    #
+    # these_tokens - An Array of PuppetLint::Lexer::Token objects that make up
+    #                the defined type or class definition.
+    #
+    # Returns an Array of PuppetLint::Lexer::Token objects or nil if it takes
+    # no parameters.
+    def param_tokens(these_tokens)
+      depth = 0
+      lparen_idx = nil
+      rparen_idx = nil
+
+      these_tokens.each_with_index do |token, i|
+        if token.type == :LPAREN
+          depth += 1
+          lparen_idx = i if depth == 1
+        elsif token.type == :RPAREN
+          depth -= 1
+          if depth == 0
+            rparen_idx = i
+            break
+          end
+        elsif token.type == :LBRACE && depth == 0
+          # no parameters
+          break
+        end
+      end
+
+      if lparen_idx.nil? or rparen_idx.nil?
+        nil
+      else
+        these_tokens[(lparen_idx + 1)..(rparen_idx - 1)]
+      end
+    end
+
+    # Internal: Retrieves a list of token types that are considered to be
+    # formatting tokens (whitespace, newlines, etc).
+    #
+    # Returns an Array of Symbols.
+    def formatting_tokens
+      @formatting_tokens ||= PuppetLint::Lexer::FORMATTING_TOKENS
+    end
+
+    # Internal: Retrieves a Hash of Sets. Each key is a check name Symbol and
+    # the Set of Integers returned lists all the lines that the check results
+    # should be ignored on.
+    #
+    # Returns a Hash of Sets of Integers.
+    def ignore_overrides
+      @ignore_overrides ||= {}
+    end
+
+    # Internal: Parses all COMMENT, MLCOMMENT and SLASH_COMMENT tokens looking
+    # for control comments (comments that enable or disable checks). Builds the
+    # contents of the `ignore_overrides` hash.
+    #
+    # Returns nothing.
+    def parse_control_comments
+      @ignore_overrides.each_key { |check| @ignore_overrides[check].clear }
+      control_re = /\A(lint:\S+)(\s+lint:\S+)*(.*)/
+
+      comment_token_types = Set[:COMMENT, :MLCOMMENT, :SLASH_COMMENT]
+
+      comment_tokens = tokens.select { |token|
+        comment_token_types.include?(token.type)
+      }
+      control_comment_tokens = comment_tokens.select { |token|
+        token.value.strip =~ /\Alint:(ignore:[\w\d]+|endignore)/
+      }
+
+      stack = []
+      control_comment_tokens.each do |token|
+        comment_data = control_re.match(token.value.strip).to_a[1..-1].compact.map(&:strip)
+        if comment_data.last =~ /\Alint:(ignore|endignore)/
+          comment_data << ''
+        end
+        reason = comment_data.pop
+        stack_add = []
+        comment_data.each do |control|
+          split_control = control.split(':')
+          command = split_control[1]
+
+          if command == 'ignore'
+            check = split_control[2].to_sym
+
+            if token.prev_token && !Set[:NEWLINE, :INDENT].include?(token.prev_token.type)
+              # control comment at the end of the line, override applies to
+              # a single line only
+              (ignore_overrides[check] ||= {})[token.line] = reason
+            else
+              stack_add << [token.line, reason, check]
+            end
+          else
+            stack.pop.each do |start|
+              unless start.nil?
+                (start[0]..token.line).each do |i|
+                  (ignore_overrides[start[2]] ||= {})[i] = start[1]
+                end
+              end
+            end
+          end
+        end
+        stack << stack_add unless stack_add.empty?
+      end
+    end
+  end
+end