superp-rubyzip 0.1.0

data/lib/zip/entry_set.rb

@@ -0,0 +1,86 @@
+module Zip
+  class EntrySet #:nodoc:all
+    include Enumerable
+    attr_accessor :entry_set, :entry_order
+
+    def initialize(an_enumerable = [])
+      super()
+      @entry_set   = {}
+      @entry_order = []
+      an_enumerable.each { |o| push(o) }
+    end
+
+    def include?(entry)
+      @entry_set.include?(to_key(entry))
+    end
+
+    def find_entry(entry)
+      @entry_set[to_key(entry)]
+    end
+
+    def <<(entry)
+      @entry_order.delete(to_key(entry))
+      @entry_order << to_key(entry)
+      @entry_set[to_key(entry)] = entry
+    end
+
+    alias :push :<<
+
+    def size
+      @entry_set.size
+    end
+
+    alias :length :size
+
+    def delete(entry)
+      if @entry_order.delete(to_key(entry)) && @entry_set.delete(to_key(entry))
+        entry
+      else
+        nil
+      end
+    end
+
+    def each(&block)
+      @entry_order.each do |key|
+        block.call @entry_set[key]
+      end
+    end
+
+    def entries
+      @entry_order.map { |key| @entry_set[key] }
+    end
+
+    # deep clone
+    def dup
+      EntrySet.new(@entry_order.map { |key| @entry_set[key].dup })
+    end
+
+    def ==(other)
+      return false unless other.kind_of?(EntrySet)
+      @entry_set == other.entry_set && @entry_order == other.entry_order
+    end
+
+    def parent(entry)
+      @entry_set[to_key(entry.parent_as_string)]
+    end
+
+    def glob(pattern, flags = ::File::FNM_PATHNAME|::File::FNM_DOTMATCH)
+      entries.map do |entry|
+        next nil unless ::File.fnmatch(pattern, entry.name.chomp('/'), flags)
+        yield(entry) if block_given?
+        entry
+      end.compact
+    end
+
+    protected
+
+    private
+    def to_key(entry)
+      entry.to_s.sub(/\/$/, '')
+    end
+  end
+end
+
+# Copyright (C) 2002, 2003 Thomas Sondergaard
+# rubyzip is free software; you can redistribute it and/or
+# modify it under the terms of the ruby license.

data/lib/zip/errors.rb

@@ -0,0 +1,8 @@
+module Zip
+  class ZipError < StandardError; end
+  class ZipEntryExistsError < ZipError; end
+  class ZipDestinationFileExistsError < ZipError; end
+  class ZipCompressionMethodError < ZipError; end
+  class ZipEntryNameError < ZipError; end
+  class ZipInternalError < ZipError; end
+end

data/lib/zip/extra_field.rb

@@ -0,0 +1,90 @@
+module Zip
+  class ExtraField < Hash
+    ID_MAP = {}
+
+    def initialize(binstr = nil)
+      binstr and merge(binstr)
+    end
+
+    def extra_field_type_exist(binstr, id, len, i)
+      field_name = ID_MAP[id].name
+      if self.member?(field_name)
+        self[field_name].merge(binstr[i, len + 4])
+      else
+        field_obj        = ID_MAP[id].new(binstr[i, len + 4])
+        self[field_name] = field_obj
+      end
+    end
+
+    def extra_field_type_unknown(binstr, len, i)
+      create_unknown_item unless self['Unknown']
+      if !len || len + 4 > binstr[i..-1].bytesize
+        self['Unknown'] << binstr[i..-1]
+        return
+      end
+      self['Unknown'] << binstr[i, len + 4]
+    end
+
+    def create_unknown_item
+      s = ''
+      class << s
+        alias_method :to_c_dir_bin, :to_s
+        alias_method :to_local_bin, :to_s
+      end
+      self['Unknown'] = s
+    end
+
+    def merge(binstr)
+      return if binstr.empty?
+      i = 0
+      while i < binstr.bytesize
+        id  = binstr[i, 2]
+        len = binstr[i + 2, 2].to_s.unpack('v').first
+        if id && ID_MAP.member?(id)
+          extra_field_type_exist(binstr, id, len, i)
+        elsif id
+          create_unknown_item unless self['Unknown']
+          break unless extra_field_type_unknown(binstr, len, i)
+        end
+        i += len + 4
+      end
+    end
+
+    def create(name)
+      unless field_class = ID_MAP.values.find { |k| k.name == name }
+        raise ZipError, "Unknown extra field '#{name}'"
+      end
+      self[name] = field_class.new
+    end
+
+    def to_local_bin
+      self.map { |_, v| v.to_local_bin }.join
+    end
+
+    alias :to_s :to_local_bin
+
+    def to_c_dir_bin
+      self.map { |_, v| v.to_c_dir_bin }.join
+    end
+
+    def c_dir_length
+      to_c_dir_bin.bytesize
+    end
+
+    def local_length
+      to_local_bin.bytesize
+    end
+
+    alias :c_dir_size :c_dir_length
+    alias :local_size :local_length
+    alias :length :local_length
+    alias :size :local_length
+  end
+end
+
+require 'zip/extra_field/generic'
+require 'zip/extra_field/universal_time'
+require 'zip/extra_field/unix'
+# Copyright (C) 2002, 2003 Thomas Sondergaard
+# rubyzip is free software; you can redistribute it and/or
+# modify it under the terms of the ruby license.

data/lib/zip/extra_field/generic.rb

@@ -0,0 +1,43 @@
+module Zip
+  class ExtraField::Generic
+    def self.register_map
+      if self.const_defined?(:HEADER_ID)
+        ::Zip::ExtraField::ID_MAP[self.const_get(:HEADER_ID)] = self
+      end
+    end
+
+    def self.name
+      self.to_s.split("::")[-1]
+    end
+
+    # return field [size, content] or false
+    def initial_parse(binstr)
+      if !binstr
+        # If nil, start with empty.
+        return false
+      elsif binstr[0, 2] != self.class.const_get(:HEADER_ID)
+        $stderr.puts "Warning: weired extra feild header ID. skip parsing"
+        return false
+      end
+      [binstr[2, 2].unpack("v")[0], binstr[4..-1]]
+    end
+
+    def ==(other)
+      return false if self.class != other.class
+      each do |k, v|
+        v != other[k] and return false
+      end
+      true
+    end
+
+    def to_local_bin
+      s = pack_for_local
+      self.class.const_get(:HEADER_ID) + [s.bytesize].pack("v") + s
+    end
+
+    def to_c_dir_bin
+      s = pack_for_c_dir
+      self.class.const_get(:HEADER_ID) + [s.bytesize].pack("v") + s
+    end
+  end
+end

data/lib/zip/extra_field/universal_time.rb

@@ -0,0 +1,47 @@
+module Zip
+  # Info-ZIP Additional timestamp field
+  class ExtraField::UniversalTime < ExtraField::Generic
+    HEADER_ID = "UT"
+    register_map
+
+    def initialize(binstr = nil)
+      @ctime = nil
+      @mtime = nil
+      @atime = nil
+      @flag  = nil
+      binstr and merge(binstr)
+    end
+
+    attr_accessor :atime, :ctime, :mtime, :flag
+
+    def merge(binstr)
+      return if binstr.empty?
+      size, content = initial_parse(binstr)
+      size or return
+      @flag, mtime, atime, ctime = content.unpack("CVVV")
+      mtime and @mtime ||= ::Zip::DOSTime.at(mtime)
+      atime and @atime ||= ::Zip::DOSTime.at(atime)
+      ctime and @ctime ||= ::Zip::DOSTime.at(ctime)
+    end
+
+    def ==(other)
+      @mtime == other.mtime &&
+        @atime == other.atime &&
+        @ctime == other.ctime
+    end
+
+    def pack_for_local
+      s = [@flag].pack("C")
+      @flag & 1 != 0 and s << [@mtime.to_i].pack("V")
+      @flag & 2 != 0 and s << [@atime.to_i].pack("V")
+      @flag & 4 != 0 and s << [@ctime.to_i].pack("V")
+      s
+    end
+
+    def pack_for_c_dir
+      s = [@flag].pack("C")
+      @flag & 1 == 1 and s << [@mtime.to_i].pack("V")
+      s
+    end
+  end
+end

data/lib/zip/extra_field/unix.rb

@@ -0,0 +1,39 @@
+module Zip
+  # Info-ZIP Extra for UNIX uid/gid
+  class ExtraField::IUnix < ExtraField::Generic
+    HEADER_ID = "Ux"
+    register_map
+
+    def initialize(binstr = nil)
+      @uid = 0
+      @gid = 0
+      binstr and merge(binstr)
+    end
+
+    attr_accessor :uid, :gid
+
+    def merge(binstr)
+      return if binstr.empty?
+      size, content = initial_parse(binstr)
+      # size: 0 for central directory. 4 for local header
+      return if (!size || size == 0)
+      uid, gid = content.unpack("vv")
+      @uid     ||= uid
+      @gid     ||= gid
+    end
+
+    def ==(other)
+      @uid == other.uid &&
+        @gid == other.gid
+    end
+
+    def pack_for_local
+      [@uid, @gid].pack("vv")
+    end
+
+    def pack_for_c_dir
+      ""
+    end
+  end
+
+end

data/lib/zip/file.rb

@@ -0,0 +1,419 @@
+module Zip
+  # ZipFile is modeled after java.util.zip.ZipFile from the Java SDK.
+  # The most important methods are those inherited from
+  # ZipCentralDirectory for accessing information about the entries in
+  # the archive and methods such as get_input_stream and
+  # get_output_stream for reading from and writing entries to the
+  # archive. The class includes a few convenience methods such as
+  # #extract for extracting entries to the filesystem, and #remove,
+  # #replace, #rename and #mkdir for making simple modifications to
+  # the archive.
+  #
+  # Modifications to a zip archive are not committed until #commit or
+  # #close is called. The method #open accepts a block following
+  # the pattern from File.open offering a simple way to
+  # automatically close the archive when the block returns.
+  #
+  # The following example opens zip archive <code>my.zip</code>
+  # (creating it if it doesn't exist) and adds an entry
+  # <code>first.txt</code> and a directory entry <code>a_dir</code>
+  # to it.
+  #
+  #   require 'zip/zip'
+  #
+  #   Zip::ZipFile.open("my.zip", Zip::ZipFile::CREATE) {
+  #    |zipfile|
+  #     zipfile.get_output_stream("first.txt") { |f| f.puts "Hello from ZipFile" }
+  #     zipfile.mkdir("a_dir")
+  #   }
+  #
+  # The next example reopens <code>my.zip</code> writes the contents of
+  # <code>first.txt</code> to standard out and deletes the entry from
+  # the archive.
+  #
+  #   require 'zip/zip'
+  #
+  #   Zip::ZipFile.open("my.zip", Zip::ZipFile::CREATE) {
+  #     |zipfile|
+  #     puts zipfile.read("first.txt")
+  #     zipfile.remove("first.txt")
+  #   }
+  #
+  # ZipFileSystem offers an alternative API that emulates ruby's
+  # interface for accessing the filesystem, ie. the File and Dir classes.
+
+  class File < CentralDirectory
+
+    CREATE           = 1
+    SPLIT_SIGNATURE  = 0x08074b50
+    MAX_SEGMENT_SIZE = 3221225472
+    MIN_SEGMENT_SIZE = 65536
+    DATA_BUFFER_SIZE = 8192
+
+    attr_reader :name
+
+    # default -> false
+    attr_accessor :restore_ownership
+    # default -> false
+    attr_accessor :restore_permissions
+    # default -> true
+    attr_accessor :restore_times
+
+    # Opens a zip archive. Pass true as the second parameter to create
+    # a new archive if it doesn't exist already.
+    def initialize(fileName, create = nil, buffer = false)
+      super()
+      @name    = fileName
+      @comment = ""
+      @create = create
+      case
+      when ::File.exists?(fileName) && !buffer
+        @create = nil
+        ::File.open(name, "rb") do |f|
+          read_from_stream(f)
+        end
+      when create
+        @entry_set = EntrySet.new
+      else
+        raise ZipError, "File #{fileName} not found"
+      end
+      @storedEntries       = @entry_set.dup
+      @storedComment       = @comment
+      @restore_ownership   = false
+      @restore_permissions = false
+      @restore_times       = true
+    end
+
+    class << self
+      # Same as #new. If a block is passed the ZipFile object is passed
+      # to the block and is automatically closed afterwards just as with
+      # ruby's builtin File.open method.
+      def open(fileName, create = nil)
+        zf = ::Zip::File.new(fileName, create)
+        if block_given?
+          begin
+            yield zf
+          ensure
+            zf.close
+          end
+        else
+          zf
+        end
+      end
+
+      # Same as #open. But outputs data to a buffer instead of a file
+      def add_buffer
+        zf = ::Zip::File.new('', true, true)
+        yield zf
+        zf.write_buffer
+      end
+
+      # Like #open, but reads zip archive contents from a String or open IO
+      # stream, and outputs data to a buffer.
+      # (This can be used to extract data from a 
+      # downloaded zip archive without first saving it to disk.)
+      def open_buffer(io)
+        zf = ::Zip::File.new('', true, true)
+        if io.is_a? IO
+          zf.read_from_stream(io)
+        elsif io.is_a? String
+          require 'stringio'
+          zf.read_from_stream(StringIO.new(io))
+        else
+          raise "Zip::ZipFile.open_buffer expects an argument of class String or IO. Found: #{io.class}"
+        end
+        yield zf
+        zf.write_buffer
+      end
+
+      # Iterates over the contents of the ZipFile. This is more efficient
+      # than using a ZipInputStream since this methods simply iterates
+      # through the entries in the central directory structure in the archive
+      # whereas ZipInputStream jumps through the entire archive accessing the
+      # local entry headers (which contain the same information as the
+      # central directory).
+      def foreach(aZipFileName, &block)
+        open(aZipFileName) do |zipFile|
+          zipFile.each(&block)
+        end
+      end
+
+      def get_segment_size_for_split(segment_size)
+        case
+        when MIN_SEGMENT_SIZE > segment_size
+          MIN_SEGMENT_SIZE
+        when MAX_SEGMENT_SIZE < segment_size
+          MAX_SEGMENT_SIZE
+        else
+          segment_size
+        end
+      end
+
+      def get_partial_zip_file_name(zip_file_name, partial_zip_file_name)
+        partial_zip_file_name = zip_file_name.sub(/#{::File.basename(zip_file_name)}\z/,
+                                                  partial_zip_file_name + ::File.extname(zip_file_name)) unless partial_zip_file_name.nil?
+        partial_zip_file_name ||= zip_file_name
+        partial_zip_file_name
+      end
+
+      def get_segment_count_for_split(zip_file_size, segment_size)
+        (zip_file_size / segment_size).to_i + (zip_file_size % segment_size == 0 ? 0 : 1)
+      end
+
+      def put_split_signature(szip_file, segment_size)
+        signature_packed = [SPLIT_SIGNATURE].pack('V')
+        szip_file << signature_packed
+        segment_size - signature_packed.size
+      end
+
+      #
+      # TODO: Make the code more understandable
+      #
+      def save_splited_part(zip_file, partial_zip_file_name, zip_file_size, szip_file_index, segment_size, segment_count)
+        ssegment_size = zip_file_size - zip_file.pos
+        ssegment_size = segment_size if ssegment_size > segment_size
+        szip_file_name = "#{partial_zip_file_name}.#{'%03d'%(szip_file_index)}"
+        ::File.open(szip_file_name, 'wb') do |szip_file|
+          if szip_file_index == 1
+            ssegment_size = put_split_signature(szip_file, segment_size)
+          end
+          chunk_bytes = 0
+          until ssegment_size == chunk_bytes || zip_file.eof?
+            segment_bytes_left = ssegment_size - chunk_bytes
+            buffer_size        = segment_bytes_left < DATA_BUFFER_SIZE ? segment_bytes_left : DATA_BUFFER_SIZE
+            chunk              = zip_file.read(buffer_size)
+            chunk_bytes        += buffer_size
+            szip_file << chunk
+            # Info for track splitting
+            yield segment_count, szip_file_index, chunk_bytes, ssegment_size if block_given?
+          end
+        end
+      end
+
+      # Splits an archive into parts with segment size
+      def split(zip_file_name, segment_size = MAX_SEGMENT_SIZE, delete_zip_file = true, partial_zip_file_name = nil)
+        raise ZipError, "File #{zip_file_name} not found" unless ::File.exists?(zip_file_name)
+        raise Errno::ENOENT, zip_file_name unless ::File.readable?(zip_file_name)
+        zip_file_size = ::File.size(zip_file_name)
+        segment_size  = get_segment_size_for_split(segment_size)
+        return if zip_file_size <= segment_size
+        segment_count = get_segment_count_for_split(zip_file_size, segment_size)
+        # Checking for correct zip structure
+        self.open(zip_file_name) {}
+        partial_zip_file_name = get_partial_zip_file_name(zip_file_name, partial_zip_file_name)
+        szip_file_index       = 0
+        ::File.open(zip_file_name, 'rb') do |zip_file|
+          until zip_file.eof?
+            szip_file_index += 1
+            save_splited_part(zip_file, partial_zip_file_name, zip_file_size, szip_file_index, segment_size, segment_count)
+          end
+        end
+        ::File.delete(zip_file_name) if delete_zip_file
+        szip_file_index
+      end
+    end
+
+    # Returns the zip files comment, if it has one
+    attr_accessor :comment
+
+    # Returns an input stream to the specified entry. If a block is passed
+    # the stream object is passed to the block and the stream is automatically
+    # closed afterwards just as with ruby's builtin File.open method.
+    def get_input_stream(entry, &aProc)
+      get_entry(entry).get_input_stream(&aProc)
+    end
+
+    # Returns an output stream to the specified entry. If a block is passed
+    # the stream object is passed to the block and the stream is automatically
+    # closed afterwards just as with ruby's builtin File.open method.
+    def get_output_stream(entry, permissionInt = nil, &aProc)
+      newEntry = entry.kind_of?(Entry) ? entry : Entry.new(@name, entry.to_s)
+      if newEntry.directory?
+        raise ArgumentError,
+              "cannot open stream to directory entry - '#{newEntry}'"
+      end
+      newEntry.unix_perms = permissionInt
+      zipStreamableEntry  = StreamableStream.new(newEntry)
+      @entry_set << zipStreamableEntry
+      zipStreamableEntry.get_output_stream(&aProc)
+    end
+
+    # Returns the name of the zip archive
+    def to_s
+      @name
+    end
+
+    # Returns a string containing the contents of the specified entry
+    def read(entry)
+      get_input_stream(entry) { |is| is.read }
+    end
+
+    # Convenience method for adding the contents of a file to the archive
+    def add(entry, srcPath, &continue_on_exists_proc)
+      continue_on_exists_proc ||= proc { Zip.continue_on_exists_proc }
+      check_entry_exists(entry, continue_on_exists_proc, "add")
+      newEntry = entry.kind_of?(Entry) ? entry : Entry.new(@name, entry.to_s)
+      newEntry.gather_fileinfo_from_srcpath(srcPath)
+      @entry_set << newEntry
+    end
+
+    # Removes the specified entry.
+    def remove(entry)
+      @entry_set.delete(get_entry(entry))
+    end
+
+    # Renames the specified entry.
+    def rename(entry, new_name, &continue_on_exists_proc)
+      foundEntry = get_entry(entry)
+      check_entry_exists(new_name, continue_on_exists_proc, 'rename')
+      @entry_set.delete(foundEntry)
+      foundEntry.name = new_name
+      @entry_set << foundEntry
+    end
+
+    # Replaces the specified entry with the contents of srcPath (from
+    # the file system).
+    def replace(entry, srcPath)
+      check_file(srcPath)
+      remove(entry)
+      add(entry, srcPath)
+    end
+
+    # Extracts entry to file dest_path.
+    def extract(entry, dest_path, &block)
+      block       ||= proc { ::Zip.on_exists_proc }
+      found_entry = get_entry(entry)
+      found_entry.extract(dest_path, &block)
+    end
+
+    # Commits changes that has been made since the previous commit to
+    # the zip archive.
+    def commit
+      return if !commit_required?
+      on_success_replace(name) {
+        |tmpFile|
+        OutputStream.open(tmpFile) {
+          |zos|
+
+          @entry_set.each {
+            |e|
+            e.write_to_zip_output_stream(zos)
+            e.dirty = false
+          }
+          zos.comment = comment
+        }
+        true
+      }
+      initialize(name)
+    end
+
+    # Write buffer write changes to buffer and return
+    def write_buffer
+      buffer = OutputStream.write_buffer do |zos|
+        @entry_set.each { |e| e.write_to_zip_output_stream(zos) }
+        zos.comment = comment
+      end
+      return buffer
+    end
+
+    # Closes the zip file committing any changes that has been made.
+    def close
+      commit
+    end
+
+    # Returns true if any changes has been made to this archive since
+    # the previous commit
+    def commit_required?
+      @entry_set.each do |e|
+        return true if e.dirty
+      end
+      @comment != @storedComment || @entry_set != @storedEntries || @create == File::CREATE
+    end
+
+    # Searches for entry with the specified name. Returns nil if
+    # no entry is found. See also get_entry
+    def find_entry(entry_name)
+      @entry_set.find_entry(entry_name)
+    end
+
+    # Searches for entries given a glob
+    def glob(*args, &block)
+      @entry_set.glob(*args, &block)
+    end
+
+    # Searches for an entry just as find_entry, but throws Errno::ENOENT
+    # if no entry is found.
+    def get_entry(entry)
+      selectedEntry = find_entry(entry)
+      unless selectedEntry
+        raise Errno::ENOENT, entry
+      end
+      selectedEntry.restore_ownership   = @restore_ownership
+      selectedEntry.restore_permissions = @restore_permissions
+      selectedEntry.restore_times       = @restore_times
+      selectedEntry
+    end
+
+    # Creates a directory
+    def mkdir(entryName, permissionInt = 0755)
+      if find_entry(entryName)
+        raise Errno::EEXIST, "File exists - #{entryName}"
+      end
+      entryName = entryName.dup.to_s
+      entryName << '/' unless entryName.end_with?('/')
+      @entry_set << StreamableDirectory.new(@name, entryName, nil, permissionInt)
+    end
+
+    private
+
+    def is_directory(newEntry, srcPath)
+      srcPathIsDirectory = ::File.directory?(srcPath)
+      if newEntry.is_directory && !srcPathIsDirectory
+        raise ArgumentError,
+              "entry name '#{newEntry}' indicates directory entry, but "+
+                "'#{srcPath}' is not a directory"
+      elsif !newEntry.is_directory && srcPathIsDirectory
+        newEntry.name += "/"
+      end
+      newEntry.is_directory && srcPathIsDirectory
+    end
+
+    def check_entry_exists(entryName, continue_on_exists_proc, procedureName)
+      continue_on_exists_proc ||= proc { Zip.continue_on_exists_proc }
+      if @entry_set.include?(entryName)
+        if continue_on_exists_proc.call
+          remove get_entry(entryName)
+        else
+          raise ZipEntryExistsError,
+                procedureName + " failed. Entry #{entryName} already exists"
+        end
+      end
+    end
+
+    def check_file(path)
+      unless ::File.readable?(path)
+        raise Errno::ENOENT, path
+      end
+    end
+
+    def on_success_replace(aFilename)
+      tmpfile     = get_tempfile
+      tmpFilename = tmpfile.path
+      tmpfile.close
+      if yield tmpFilename
+        ::File.rename(tmpFilename, name)
+      end
+    end
+
+    def get_tempfile
+      tempFile = Tempfile.new(::File.basename(name), ::File.dirname(name))
+      tempFile.binmode
+      tempFile
+    end
+
+  end
+end
+
+# Copyright (C) 2002, 2003 Thomas Sondergaard
+# rubyzip is free software; you can redistribute it and/or
+# modify it under the terms of the ruby license.