nanaimo 0.1.0

data/lib/nanaimo/plist.rb

@@ -0,0 +1,33 @@
+module Nanaimo
+  # A Plist.
+  #
+  class Plist
+    # @return [Nanaimo::Object] The root level object in the plist.
+    #
+    attr_accessor :root_object
+
+    # @return [String] The encoding of the plist.
+    #
+    attr_accessor :file_type
+
+    def initialize(root_object = nil, file_type = nil)
+      @root_object = root_object
+      @file_type = file_type
+    end
+
+    def ==(other)
+      return unless other.is_a?(Nanaimo::Plist)
+      file_type == other.file_type && root_object == other.root_object
+    end
+
+    def hash
+      root_object.hash
+    end
+
+    # @return A native Ruby object representation of the plist.
+    #
+    def as_ruby
+      root_object.as_ruby
+    end
+  end
+end

data/lib/nanaimo/reader.rb

@@ -0,0 +1,252 @@
+# frozen-string-literal: true
+module Nanaimo
+  # Transforms plist strings into Plist objects.
+  #
+  class Reader
+    # Raised when attempting to read a plist with an unsupported file format.
+    #
+    class UnsupportedPlistFormatError < Error
+      # @return [Symbol] The unsupported format.
+      #
+      attr_reader :format
+
+      def initialize(format)
+        @format = format
+      end
+
+      def to_s
+        "#{format} plists are currently unsupported"
+      end
+    end
+
+    # Raised when parsing fails.
+    #
+    class ParseError < Error
+      # @return [[Integer, Integer]] The (line, column) offset into the plist
+      #         where the error occurred
+      #
+      attr_accessor :location
+
+      # @return [String] The contents of the plist.
+      #
+      attr_accessor :plist_string
+    end
+
+    # @param plist_contents [String]
+    #
+    # @return [Symbol] The file format of the plist in the given string.
+    #
+    def self.plist_type(plist_contents)
+      case plist_contents
+      when /\Abplist/
+        :binary
+      when /\A<\?xml/
+        :xml
+      else
+        :ascii
+      end
+    end
+
+    # @param file_path [String]
+    #
+    # @return [Plist] A parsed plist from the given file
+    #
+    def self.from_file(file_path)
+      new(File.read(file_path))
+    end
+
+    # @param contents [String] The plist to be parsed
+    #
+    def initialize(contents)
+      @scanner = StringScanner.new(contents)
+    end
+
+    # Parses the contents of the plist
+    #
+    # @return [Plist] The parsed Plist object.
+    #
+    def parse!
+      plist_format = ensure_ascii_plist!
+      read_string_encoding
+      root_object = parse_object
+
+      eat_whitespace!
+      raise_parser_error ParseError, "unrecognized characters #{@scanner.rest.inspect} after parsing" unless @scanner.eos?
+
+      Nanaimo::Plist.new(root_object, plist_format)
+    end
+
+    private
+
+    def ensure_ascii_plist!
+      self.class.plist_type(@scanner.string).tap do |plist_format|
+        raise UnsupportedPlistFormatError, plist_format unless plist_format == :ascii
+      end
+    end
+
+    def read_string_encoding
+      # TODO
+    end
+
+    def parse_object
+      _comment = skip_to_non_space_matching_annotations
+      start_pos = @scanner.pos
+      raise_parser_error ParseError, 'Unexpected eos while parsing' if @scanner.eos?
+      if @scanner.skip(/\{/)
+        parse_dictionary
+      elsif @scanner.skip(/\(/)
+        parse_array
+      elsif @scanner.skip(/</)
+        parse_data
+      elsif quote = @scanner.scan(/['"]/)
+        parse_quotedstring(quote)
+      else
+        parse_string
+      end.tap do |o|
+        o.annotation = skip_to_non_space_matching_annotations
+        Nanaimo.debug { "parsed #{o.inspect} from #{start_pos}..#{@scanner.pos}" }
+      end
+    end
+
+    def parse_string
+      eat_whitespace!
+      unless match = @scanner.scan(%r{[\w/.]+})
+        raise_parser_error ParseError, "not a valid string at index #{@scanner.pos} (char is #{current_character.inspect})"
+      end
+      Nanaimo::String.new(match, nil)
+    end
+
+    def parse_quotedstring(quote)
+      unless string = @scanner.scan(/(?:([^#{quote}\\]|\\.)*)#{quote}/)
+        raise_parser_error ParseError, "unterminated quoted string started at #{@scanner.pos}, expected #{quote} but never found it"
+      end
+      string = Unicode.unquotify_string(string.chomp!(quote))
+      Nanaimo::QuotedString.new(string, nil)
+    end
+
+    def parse_array
+      objects = []
+      until @scanner.eos?
+        eat_whitespace!
+        break if @scanner.skip(/\)/)
+
+        objects << parse_object
+
+        eat_whitespace!
+        break if @scanner.skip(/\)/)
+        unless @scanner.skip(/,/)
+          raise_parser_error ParseError, "Array #{objects} missing ',' in between objects"
+        end
+      end
+
+      Nanaimo::Array.new(objects, nil)
+    end
+
+    def parse_dictionary
+      objects = {}
+      until @scanner.eos?
+        skip_to_non_space_matching_annotations
+        break if @scanner.skip(/}/)
+
+        key = parse_object
+        eat_whitespace!
+        unless @scanner.skip(/=/)
+          raise_parser_error ParseError, "Dictionary missing value after key #{key.inspect} at index #{@scanner.pos}, expected '=' and got #{current_character.inspect}"
+        end
+
+        value = parse_object
+        objects[key] = value
+
+        eat_whitespace!
+        break if @scanner.skip(/}/)
+        unless @scanner.skip(/;/)
+          raise_parser_error ParseError, "Dictionary (#{objects}) missing ';' after key-value pair (#{key} = #{value}) at index #{@scanner.pos} (got #{current_character})"
+        end
+      end
+
+      Nanaimo::Dictionary.new(objects, nil)
+    end
+
+    def parse_data
+      unless data = @scanner.scan(/[\h ]*>/)
+        raise_parser_error ParseError, "Data missing closing '>'"
+      end
+      data.chomp!('>')
+      data.delete!(' ')
+      unless data.size.even?
+        @scanner.unscan
+        raise_parser_error ParseError, 'Data has an uneven number of hex digits'
+      end
+      data = [data].pack('H*')
+      Nanaimo::Data.new(data, nil)
+    end
+
+    def current_character
+      @scanner.peek(1)
+    end
+
+    def read_singleline_comment
+      unless comment = @scanner.scan_until(NEWLINE)
+        raise_parser_error ParseError, "failed to terminate single line comment #{@scanner.rest.inspect}"
+      end
+      comment
+    end
+
+    def eat_whitespace!
+      @scanner.skip(MANY_WHITESPACES)
+    end
+
+    NEWLINE_CHARACTERS = %W(\x0A \x0D \u2028 \u2029).freeze
+    NEWLINE = Regexp.union(*NEWLINE_CHARACTERS)
+
+    WHITESPACE_CHARACTERS = NEWLINE_CHARACTERS + %W(\x09 \x0B \x0C \x20)
+    WHITESPACE = Regexp.union(*WHITESPACE_CHARACTERS)
+
+    MANY_WHITESPACES = /#{WHITESPACE}+/
+
+    def read_multiline_comment
+      unless annotation = @scanner.scan(%r{(?:.+?)(?=\*/)}m)
+        raise_parser_error ParseError, "#{@scanner.rest.inspect} failed to terminate multiline comment"
+      end
+      @scanner.skip(%r{\*/})
+
+      annotation
+    end
+
+    def skip_to_non_space_matching_annotations
+      annotation = ''.freeze
+      until @scanner.eos?
+        eat_whitespace!
+
+        # Comment Detection
+        if @scanner.skip(%r{//})
+          annotation = read_singleline_comment
+          next
+        elsif @scanner.skip(%r{/\*})
+          annotation = read_multiline_comment
+          next
+        end
+
+        eat_whitespace!
+
+        break
+      end
+      annotation
+    end
+
+    def location_in(scanner)
+      pos = scanner.charpos
+      line = scanner.string[0..scanner.charpos].scan(NEWLINE).size + 1
+      column = pos - (scanner.string.rindex(NEWLINE, pos - 1) || -1)
+      [line, column]
+    end
+
+    def raise_parser_error(klass, message)
+      exception = klass.new(message).tap do |error|
+        error.location = location_in(@scanner)
+        error.plist_string = @scanner.string
+      end
+      raise(exception)
+    end
+  end
+end

data/lib/nanaimo/unicode.rb

@@ -0,0 +1,88 @@
+# frozen-string-literal: true
+require 'nanaimo/unicode/next_step_mapping'
+require 'nanaimo/unicode/quote_maps'
+module Nanaimo
+  # @!visibility private
+  #
+  module Unicode
+    class UnsupportedEscapeSequenceError < Error; end
+    class InvalidEscapeSequenceError < Error; end
+
+    module_function
+
+    def quotify_string(string)
+      string.gsub(QUOTE_REGEXP) { |s| QUOTE_MAP[s] }
+    end
+
+    ESCAPE_PREFIXES = %W(
+      0 1 2 3 4 5 6 7 a b f n r t v \n U
+    ).freeze
+
+    OCTAL_DIGITS = (0..7).map(&:to_s).freeze
+
+    # Credit to Samantha Marshall
+    # Taken from https://github.com/samdmarshall/pbPlist/blob/346c29f91f913d35d0e24f6722ec19edb24e5707/pbPlist/StrParse.py#L197
+    # Licensed under https://raw.githubusercontent.com/samdmarshall/pbPlist/blob/346c29f91f913d35d0e24f6722ec19edb24e5707/LICENSE
+    #
+    # Originally from: http://www.opensource.apple.com/source/CF/CF-744.19/CFOldStylePList.c See `getSlashedChar()`
+    def unquotify_string(string)
+      formatted_string = ::String.new
+      extracted_string = string
+      string_length = string.size
+      index = 0
+      while index < string_length
+        if escape_index = extracted_string.index('\\', index)
+          formatted_string << extracted_string[index..escape_index - 1] unless index == escape_index
+          index = escape_index + 1
+          next_char = extracted_string[index]
+          if ESCAPE_PREFIXES.include?(next_char)
+            index += 1
+            if unquoted = UNQUOTE_MAP[next_char]
+              formatted_string << unquoted
+            elsif next_char == 'U'
+              length = 4
+              unicode_numbers = extracted_string[index, length]
+              unless unicode_numbers =~ /\A\h{4}\z/
+                raise InvalidEscapeSequenceError, "Unicode '\\U' escape sequence terminated without 4 following hex characters"
+              end
+              index += length
+              formatted_string << [unicode_numbers.to_i(16)].pack('U')
+            elsif OCTAL_DIGITS.include?(next_char) # https://twitter.com/Catfish_Man/status/658014170055507968
+              octal_string = extracted_string[index - 1, 3]
+              if octal_string =~ /\A[0-7]{3}\z/
+                index += 2
+                code_point = octal_string.to_i(8)
+                unless code_point <= 0x80 || converted = NEXT_STEP_MAPPING[code_point]
+                  raise InvalidEscapeSequenceError, "Invalid octal escape sequence #{octal_string}"
+                end
+                formatted_string << [converted].pack('U')
+              else
+                formatted_string << next_char
+              end
+            else
+              raise UnsupportedEscapeSequenceError, "Failed to handle #{next_char} which is in the list of possible escapes"
+            end
+          else
+            index += 1
+            formatted_string << next_char
+          end
+        else
+          formatted_string << extracted_string[index..-1]
+          index = string_length
+        end
+      end
+      formatted_string
+    end
+
+    XML_STRING_ESCAPES = {
+      '&' => '&amp;',
+      '<' => '&lt;',
+      '>' => '&gt;'
+    }.freeze
+    XML_STRING_ESCAPE_REGEXP = Regexp.union(XML_STRING_ESCAPES.keys)
+
+    def xml_escape_string(string)
+      string.to_s.gsub(XML_STRING_ESCAPE_REGEXP) { |m| XML_STRING_ESCAPES[m] }
+    end
+  end
+end

data/lib/nanaimo/unicode/next_step_mapping.rb

@@ -0,0 +1,136 @@
+# frozen-string-literal: true
+module Nanaimo
+  module Unicode
+    # Taken from http://ftp.unicode.org/Public/MAPPINGS/VENDORS/NEXT/NEXTSTEP.TXT
+    NEXT_STEP_MAPPING = {
+      0x80 => 0x00a0, # NO-BREAK SPACE
+      0x81 => 0x00c0, # LATIN CAPITAL LETTER A WITH GRAVE
+      0x82 => 0x00c1, # LATIN CAPITAL LETTER A WITH ACUTE
+      0x83 => 0x00c2, # LATIN CAPITAL LETTER A WITH CIRCUMFLEX
+      0x84 => 0x00c3, # LATIN CAPITAL LETTER A WITH TILDE
+      0x85 => 0x00c4, # LATIN CAPITAL LETTER A WITH DIAERESIS
+      0x86 => 0x00c5, # LATIN CAPITAL LETTER A WITH RING
+      0x87 => 0x00c7, # LATIN CAPITAL LETTER C WITH CEDILLA
+      0x88 => 0x00c8, # LATIN CAPITAL LETTER E WITH GRAVE
+      0x89 => 0x00c9, # LATIN CAPITAL LETTER E WITH ACUTE
+      0x8a => 0x00ca, # LATIN CAPITAL LETTER E WITH CIRCUMFLEX
+      0x8b => 0x00cb, # LATIN CAPITAL LETTER E WITH DIAERESIS
+      0x8c => 0x00cc, # LATIN CAPITAL LETTER I WITH GRAVE
+      0x8d => 0x00cd, # LATIN CAPITAL LETTER I WITH ACUTE
+      0x8e => 0x00ce, # LATIN CAPITAL LETTER I WITH CIRCUMFLEX
+      0x8f => 0x00cf, # LATIN CAPITAL LETTER I WITH DIAERESIS
+      0x90 => 0x00d0, # LATIN CAPITAL LETTER ETH
+      0x91 => 0x00d1, # LATIN CAPITAL LETTER N WITH TILDE
+      0x92 => 0x00d2, # LATIN CAPITAL LETTER O WITH GRAVE
+      0x93 => 0x00d3, # LATIN CAPITAL LETTER O WITH ACUTE
+      0x94 => 0x00d4, # LATIN CAPITAL LETTER O WITH CIRCUMFLEX
+      0x95 => 0x00d5, # LATIN CAPITAL LETTER O WITH TILDE
+      0x96 => 0x00d6, # LATIN CAPITAL LETTER O WITH DIAERESIS
+      0x97 => 0x00d9, # LATIN CAPITAL LETTER U WITH GRAVE
+      0x98 => 0x00da, # LATIN CAPITAL LETTER U WITH ACUTE
+      0x99 => 0x00db, # LATIN CAPITAL LETTER U WITH CIRCUMFLEX
+      0x9a => 0x00dc, # LATIN CAPITAL LETTER U WITH DIAERESIS
+      0x9b => 0x00dd, # LATIN CAPITAL LETTER Y WITH ACUTE
+      0x9c => 0x00de, # LATIN CAPITAL LETTER THORN
+      0x9d => 0x00b5, # MICRO SIGN 
+      0x9e => 0x00d7, # MULTIPLICATION SIGN
+      0x9f => 0x00f7, # DIVISION SIGN
+      0xa0 => 0x00a9, # COPYRIGHT SIGN
+      0xa1 => 0x00a1, # INVERTED EXCLAMATION MARK
+      0xa2 => 0x00a2, # CENT SIGN
+      0xa3 => 0x00a3, # POUND SIGN
+      0xa4 => 0x2044, # FRACTION SLASH
+      0xa5 => 0x00a5, # YEN SIGN
+      0xa6 => 0x0192, # LATIN SMALL LETTER F WITH HOOK
+      0xa7 => 0x00a7, # SECTION SIGN
+      0xa8 => 0x00a4, # CURRENCY SIGN
+      0xa9 => 0x2019, # RIGHT SINGLE QUOTATION MARK
+      0xaa => 0x201c, # LEFT DOUBLE QUOTATION MARK
+      0xab => 0x00ab, # LEFT-POINTING DOUBLE ANGLE QUOTATION MARK
+      0xac => 0x2039, # SINGLE LEFT-POINTING ANGLE QUOTATION MARK
+      0xad => 0x203a, # SINGLE RIGHT-POINTING ANGLE QUOTATION MARK
+      0xae => 0xfb01, # LATIN SMALL LIGATURE FI
+      0xaf => 0xfb02, # LATIN SMALL LIGATURE FL
+      0xb0 => 0x00ae, # REGISTERED SIGN
+      0xb1 => 0x2013, # EN DASH
+      0xb2 => 0x2020, # DAGGER
+      0xb3 => 0x2021, # DOUBLE DAGGER
+      0xb4 => 0x00b7, # MIDDLE DOT
+      0xb5 => 0x00a6, # BROKEN BAR
+      0xb6 => 0x00b6, # PILCROW SIGN
+      0xb7 => 0x2022, # BULLET
+      0xb8 => 0x201a, # SINGLE LOW-9 QUOTATION MARK
+      0xb9 => 0x201e, # DOUBLE LOW-9 QUOTATION MARK
+      0xba => 0x201d, # RIGHT DOUBLE QUOTATION MARK
+      0xbb => 0x00bb, # RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK
+      0xbc => 0x2026, # HORIZONTAL ELLIPSIS
+      0xbd => 0x2030, # PER MILLE SIGN
+      0xbe => 0x00ac, # NOT SIGN
+      0xbf => 0x00bf, # INVERTED QUESTION MARK
+      0xc0 => 0x00b9, # SUPERSCRIPT ONE
+      0xc1 => 0x02cb, # MODIFIER LETTER GRAVE ACCENT
+      0xc2 => 0x00b4, # ACUTE ACCENT
+      0xc3 => 0x02c6, # MODIFIER LETTER CIRCUMFLEX ACCENT
+      0xc4 => 0x02dc, # SMALL TILDE
+      0xc5 => 0x00af, # MACRON
+      0xc6 => 0x02d8, # BREVE
+      0xc7 => 0x02d9, # DOT ABOVE
+      0xc8 => 0x00a8, # DIAERESIS
+      0xc9 => 0x00b2, # SUPERSCRIPT TWO
+      0xca => 0x02da, # RING ABOVE
+      0xcb => 0x00b8, # CEDILLA
+      0xcc => 0x00b3, # SUPERSCRIPT THREE
+      0xcd => 0x02dd, # DOUBLE ACUTE ACCENT
+      0xce => 0x02db, # OGONEK
+      0xcf => 0x02c7, # CARON
+      0xd0 => 0x2014, # EM DASH
+      0xd1 => 0x00b1, # PLUS-MINUS SIGN
+      0xd2 => 0x00bc, # VULGAR FRACTION ONE QUARTER
+      0xd3 => 0x00bd, # VULGAR FRACTION ONE HALF
+      0xd4 => 0x00be, # VULGAR FRACTION THREE QUARTERS
+      0xd5 => 0x00e0, # LATIN SMALL LETTER A WITH GRAVE
+      0xd6 => 0x00e1, # LATIN SMALL LETTER A WITH ACUTE
+      0xd7 => 0x00e2, # LATIN SMALL LETTER A WITH CIRCUMFLEX
+      0xd8 => 0x00e3, # LATIN SMALL LETTER A WITH TILDE
+      0xd9 => 0x00e4, # LATIN SMALL LETTER A WITH DIAERESIS
+      0xda => 0x00e5, # LATIN SMALL LETTER A WITH RING ABOVE
+      0xdb => 0x00e7, # LATIN SMALL LETTER C WITH CEDILLA
+      0xdc => 0x00e8, # LATIN SMALL LETTER E WITH GRAVE
+      0xdd => 0x00e9, # LATIN SMALL LETTER E WITH ACUTE
+      0xde => 0x00ea, # LATIN SMALL LETTER E WITH CIRCUMFLEX
+      0xdf => 0x00eb, # LATIN SMALL LETTER E WITH DIAERESIS
+      0xe0 => 0x00ec, # LATIN SMALL LETTER I WITH GRAVE
+      0xe1 => 0x00c6, # LATIN CAPITAL LETTER AE
+      0xe2 => 0x00ed, # LATIN SMALL LETTER I WITH ACUTE
+      0xe3 => 0x00aa, # FEMININE ORDINAL INDICATOR
+      0xe4 => 0x00ee, # LATIN SMALL LETTER I WITH CIRCUMFLEX
+      0xe5 => 0x00ef, # LATIN SMALL LETTER I WITH DIAERESIS
+      0xe6 => 0x00f0, # LATIN SMALL LETTER ETH
+      0xe7 => 0x00f1, # LATIN SMALL LETTER N WITH TILDE
+      0xe8 => 0x0141, # LATIN CAPITAL LETTER L WITH STROKE
+      0xe9 => 0x00d8, # LATIN CAPITAL LETTER O WITH STROKE
+      0xea => 0x0152, # LATIN CAPITAL LIGATURE OE
+      0xeb => 0x00ba, # MASCULINE ORDINAL INDICATOR
+      0xec => 0x00f2, # LATIN SMALL LETTER O WITH GRAVE
+      0xed => 0x00f3, # LATIN SMALL LETTER O WITH ACUTE
+      0xee => 0x00f4, # LATIN SMALL LETTER O WITH CIRCUMFLEX
+      0xef => 0x00f5, # LATIN SMALL LETTER O WITH TILDE
+      0xf0 => 0x00f6, # LATIN SMALL LETTER O WITH DIAERESIS
+      0xf1 => 0x00e6, # LATIN SMALL LETTER AE
+      0xf2 => 0x00f9, # LATIN SMALL LETTER U WITH GRAVE
+      0xf3 => 0x00fa, # LATIN SMALL LETTER U WITH ACUTE
+      0xf4 => 0x00fb, # LATIN SMALL LETTER U WITH CIRCUMFLEX
+      0xf5 => 0x0131, # LATIN SMALL LETTER DOTLESS I
+      0xf6 => 0x00fc, # LATIN SMALL LETTER U WITH DIAERESIS
+      0xf7 => 0x00fd, # LATIN SMALL LETTER Y WITH ACUTE
+      0xf8 => 0x0142, # LATIN SMALL LETTER L WITH STROKE
+      0xf9 => 0x00f8, # LATIN SMALL LETTER O WITH STROKE
+      0xfa => 0x0153, # LATIN SMALL LIGATURE OE
+      0xfb => 0x00df, # LATIN SMALL LETTER SHARP S
+      0xfc => 0x00fe, # LATIN SMALL LETTER THORN
+      0xfd => 0x00ff, # LATIN SMALL LETTER Y WITH DIAERESIS
+      0xfe => 0xfffd, # .notdef, REPLACEMENT CHARACTER
+      0xff => 0xfffd, # .notdef, REPLACEMENT CHARACTER
+    }.freeze
+  end
+end