rubyzip 3.2.1 → 3.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 4f451a3c548c2fb166c19e08e716375d3f27850975e33fad1ef57a09e15fadc0
-  data.tar.gz: 2578620d86942c589aa4b74178d92fa7690a408be4fc1ecdc3285dfcebc7526b
+  metadata.gz: 301afd00cf28573bc4e6b12cd9a7f6dca91dcd260244423ac8dab1c97e4dcbec
+  data.tar.gz: 8c6d46366f1fe3064632b0de1a00a82f44184438ae7febdc3db82407092965f2
 SHA512:
-  metadata.gz: 93d7eaa1d76b58942f116334e1195a88ed7cae1f7bc77493c087464323fe3a1829c04bcfe56beebffe18bb252bb45b08574b65782fa30b5d0d2f25d80241e0d0
-  data.tar.gz: 2dbdc409218799a047fc396eae4f7ba3e9087122aa4dbeefc5f0c0b10953c870680820f7ac086207f9f006118fdd2f26211ebea2414f6f0ea16f13a9ec4745f7
+  metadata.gz: 44a3a3e3176ea832db64716d81ae5dcf2919d975e395012a27ca849f3d6d22a2b9e199f73bae152f2ae2acb62f4ccbce3de33b4ffd26f21935d2ab997f2abb19
+  data.tar.gz: beeec7382ef911feeb9e4403dc6025a02904f74cc31cb9948f180c060b8706aa9b080d0e2bbbcfaae9c89465e0922799169713bba5563d275db3d677dc0962d4

data/Changelog.md

@@ -1,3 +1,22 @@
+# 3.3.0 (2026-05-02)
+
+- Refactor `InputStream` and `AbstractInputStream`.
+ [#661](https://github.com/rubyzip/rubyzip/pull/661)
+
+Tooling/internal:
+
+- Update Actions to use checkout@v5.
+- Add Ruby4.0 to the CI matrix. [#659](https://github.com/rubyzip/rubyzip/pull/659)
+
+# 3.2.2 (2025-11-02)
+
+- Fix reading EOCDs when header signatures are in an Entry payload. [#656](https://github.com/rubyzip/rubyzip/issues/656)
+
+Tooling/internal:
+
+- Stop using macos-13 runners in GitHub Actions.
+- Update YJIT GitHub Actions runners.
+
 # 3.2.1 (2025-10-24)
 
 - Fix `Entry#gather_fileinfo_from_srcpath` error messages. [#654](https://github.com/rubyzip/rubyzip/issues/654)

data/LICENSE.md

@@ -1,6 +1,6 @@
 BSD 2-Clause License
 
-Copyright (c) 2002-2025, The Rubyzip Developers
+Copyright (c) 2002-2026, The Rubyzip Developers
 
 Redistribution and use in source and binary forms, with or without
 modification, are permitted provided that the following conditions are met:

data/lib/zip/central_directory.rb

@@ -143,28 +143,27 @@ module Zip
       zip64_eocd_offset
     end
 
-    def unpack_e_o_c_d(buffer) # :nodoc:
+    # Unpack the EOCD and return a boolean indicating whether this header is
+    # complete without needing Zip64 extensions.
+    def unpack_e_o_c_d(buffer) # :nodoc: # rubocop:disable Naming/PredicateMethod
       _, # END_OF_CD_SIG. We know we have this at this point.
-      num_disk,
-      num_disk_cdir,
-      num_cdir_disk,
-      num_entries,
-      size_in_bytes,
-      cdir_offset,
+      @number_of_this_disk,
+      @number_of_disk_with_start_of_cdir,
+      @total_number_of_entries_in_cdir_on_this_disk,
+      @size,
+      @size_in_bytes,
+      @cdir_offset,
       comment_length = buffer.unpack('VvvvvVVv')
 
-      @number_of_this_disk = num_disk unless num_disk == 0xFFFF
-      @number_of_disk_with_start_of_cdir = num_disk_cdir unless num_disk_cdir == 0xFFFF
-      @total_number_of_entries_in_cdir_on_this_disk = num_cdir_disk unless num_cdir_disk == 0xFFFF
-      @size = num_entries unless num_entries == 0xFFFF
-      @size_in_bytes = size_in_bytes unless size_in_bytes == 0xFFFFFFFF
-      @cdir_offset = cdir_offset unless cdir_offset == 0xFFFFFFFF
-
       @comment = if comment_length.positive?
                    buffer.slice(STATIC_EOCD_SIZE, comment_length)
                  else
                    ''
                  end
+
+      !([@number_of_this_disk, @number_of_disk_with_start_of_cdir,
+         @total_number_of_entries_in_cdir_on_this_disk, @size].any?(0xFFFF) ||
+         @size_in_bytes == 0xFFFFFFFF || @cdir_offset == 0xFFFFFFFF)
     end
 
     def read_central_directory_entries(io) # :nodoc:
@@ -217,30 +216,37 @@ module Zip
       eocd_location = data.rindex([END_OF_CD_SIG].pack('V'))
       raise Error, 'Zip end of central directory signature not found' unless eocd_location
 
-      zip64_eocd_locator = data.rindex([ZIP64_EOCD_LOCATOR_SIG].pack('V'))
-
-      if zip64_eocd_locator
-        zip64_eocd_location = data.rindex([ZIP64_END_OF_CD_SIG].pack('V'))
+      # Parse the EOCD and return if it is complete without Zip64 extensions.
+      return if unpack_e_o_c_d(data.slice(eocd_location..-1))
 
-        zip64_eocd_data =
-          if zip64_eocd_location
-            data.slice(zip64_eocd_location..zip64_eocd_locator)
-          else
-            zip64_eocd_location = unpack_64_eocd_locator(
-              data.slice(zip64_eocd_locator..eocd_location)
-            )
-            unless zip64_eocd_location
-              raise Error, 'Zip64 end of central directory signature not found'
-            end
+      # Need to read in the Zip64 EOCD locator and then the Zip64 EOCD.
+      zip64_eocd_locator = data.rindex([ZIP64_EOCD_LOCATOR_SIG].pack('V'), eocd_location)
+      unless zip64_eocd_locator
+        raise Error, 'Zip64 end of central directory locator signature expected but not found'
+      end
 
-            io.seek(zip64_eocd_location, IO::SEEK_SET)
-            io.read(base_location + zip64_eocd_locator - zip64_eocd_location)
+      # Do we already have the Zip64 EOCD in the data we've read?
+      zip64_eocd_location = data.rindex([ZIP64_END_OF_CD_SIG].pack('V'), zip64_eocd_locator)
+
+      zip64_eocd_data =
+        if zip64_eocd_location
+          # Yes.
+          data.slice(zip64_eocd_location..zip64_eocd_locator)
+        else
+          # No. Read its location from the locator and then read it in.
+          zip64_eocd_location = unpack_64_eocd_locator(
+            data.slice(zip64_eocd_locator..eocd_location)
+          )
+          unless zip64_eocd_location
+            raise Error, 'Zip64 end of central directory signature not found'
           end
 
-        unpack_64_e_o_c_d(zip64_eocd_data)
-      end
+          io.seek(zip64_eocd_location, IO::SEEK_SET)
+          io.read(base_location + zip64_eocd_locator - zip64_eocd_location)
+        end
 
-      unpack_e_o_c_d(data.slice(eocd_location..-1))
+      # Finally, unpack the Zip64 EOCD.
+      unpack_64_e_o_c_d(zip64_eocd_data)
     end
 
     def eocd_data(io)

data/lib/zip/crypto/decrypted_io.rb

@@ -8,13 +8,13 @@ module Zip
       @io = io
       @decrypter = decrypter
       @bytes_remaining = compressed_size
-      @buffer = +''
+      @buffer = +''.b
     end
 
-    def read(length = nil, outbuf = +'')
-      return (length.nil? || length.zero? ? '' : nil) if eof?
+    def read(maxlen = nil)
+      return (maxlen.nil? || maxlen.zero? ? '' : nil) if eof?
 
-      while length.nil? || (@buffer.bytesize < length)
+      while maxlen.nil? || (@buffer.bytesize < maxlen)
         break if input_finished?
 
         @buffer << produce_input
@@ -22,7 +22,7 @@ module Zip
 
       @decrypter.check_integrity!(@io) if input_finished?
 
-      outbuf.replace(@buffer.slice!(0...(length || @buffer.bytesize)))
+      @buffer.slice!(0...(maxlen || @buffer.bytesize))
     end
 
     private

data/lib/zip/entry.rb

@@ -751,21 +751,20 @@ module Zip
 
     def create_file(dest_path, _continue_on_exists_proc = proc { Zip.continue_on_exists_proc })
       if ::File.exist?(dest_path) && !yield(self, dest_path)
-        raise ::Zip::DestinationExistsError, dest_path
+        raise DestinationExistsError, dest_path
       end
 
       ::File.open(dest_path, 'wb') do |os|
         get_input_stream do |is|
           bytes_written = 0
           warned = false
-          buf = +''
-          while (buf = is.sysread(::Zip::Decompressor::CHUNK_SIZE, buf))
+          while (buf = is.sysread(Decompressor::CHUNK_SIZE))
             os << buf
             bytes_written += buf.bytesize
             next unless bytes_written > size && !warned
 
-            error = ::Zip::EntrySizeError.new(self)
-            raise error if ::Zip.validate_entry_sizes
+            error = EntrySizeError.new(self)
+            raise error if Zip.validate_entry_sizes
 
             warn "WARNING: #{error.message}"
             warned = true

data/lib/zip/inflater.rb

@@ -5,20 +5,20 @@ module Zip
     def initialize(*args)
       super
 
-      @buffer = +''
+      @buffer = +''.b
       @zlib_inflater = ::Zlib::Inflate.new(-Zlib::MAX_WBITS)
     end
 
-    def read(length = nil, outbuf = +'')
-      return (length.nil? || length.zero? ? '' : nil) if eof?
+    def read(maxlen = nil)
+      return (maxlen.nil? || maxlen.zero? ? '' : nil) if eof?
 
-      while length.nil? || (@buffer.bytesize < length)
+      while maxlen.nil? || (@buffer.bytesize < maxlen)
         break if input_finished?
 
         @buffer << produce_input
       end
 
-      outbuf.replace(@buffer.slice!(0...(length || @buffer.bytesize)))
+      @buffer.slice!(0...(maxlen || @buffer.bytesize))
     end
 
     def eof?

data/lib/zip/input_stream.rb

@@ -89,9 +89,28 @@ module Zip
       open_entry
     end
 
-    # Modeled after IO.sysread
-    def sysread(length = nil, outbuf = +'')
-      @decompressor.read(length, outbuf)
+    # Modelled after IO#sysread.
+    #
+    # Reads up to maxlen bytes from the stream; returns a string
+    # (either a new string or the given out_string).
+    # Its encoding is the unchanged encoding of out_string, if out_string is
+    # given; ASCII-8BIT, otherwise. Output contains maxlen bytes from the
+    # stream, if available; otherwise contains all available bytes, if any
+    # available; otherwise is an empty string.
+    #
+    # This method should not be used with buffered input stream-reader methods,
+    # such as #read, #readline, #gets.
+    def sysread(maxlen, out_string = nil)
+      return (maxlen.nil? || maxlen.zero? ? '' : nil) if eof?
+
+      output = produce_input(maxlen)
+
+      if out_string.nil?
+        output.force_encoding(Encoding::ASCII_8BIT)
+      else
+        encoding = out_string.encoding
+        out_string.replace(output).force_encoding(encoding)
+      end
     end
 
     # Returns the size of the current entry, or `nil` if there isn't one.
@@ -197,8 +216,8 @@ module Zip
       decompressor_class.new(io, decompressed_size)
     end
 
-    def produce_input # :nodoc:
-      @decompressor.read(CHUNK_SIZE)
+    def produce_input(maxlen = CHUNK_SIZE) # :nodoc:
+      @decompressor.read(maxlen)
     end
 
     def input_finished? # :nodoc:

data/lib/zip/ioextras/abstract_input_stream.rb

@@ -3,126 +3,208 @@
 module Zip
   module IOExtras # :nodoc:
     # Implements many of the convenience methods of IO
-    # such as gets, getc, readline and readlines
+    # such as gets, getc, read, readline and readlines
     # depends on: input_finished?, produce_input and read
-    module AbstractInputStream # :nodoc:
+    module AbstractInputStream
       include Enumerable
       include FakeIO
 
-      def initialize
+      def initialize # :nodoc:
         super
         @lineno        = 0
         @pos           = 0
-        @output_buffer = +''
+        @output_buffer = +''.b
       end
 
+      # Returns (or sets) the current line number in the decompressed
+      # (possibly decrypted) data stream. See the Line Number documentation
+      # for the IO class for more information.
       attr_accessor :lineno
+
+      # Returns the current position (in bytes) in the decompressed (possibly
+      # decrypted) data stream.
       attr_reader :pos
 
-      def read(number_of_bytes = nil, buf = +'')
+      # Reads bytes from the stream decompressed (possibly decrypted) data
+      # stream. If `maxlen` is `nil`, reads all bytes; otherwise, reads up to
+      # `maxlen` bytes. If `maxlen` is zero, returns an empty string.
+      #
+      # Returns a string (either a new string or the given `out_string`)
+      # containing the bytes read. The string's encoding is the unchanged
+      # encoding of `out_string`, if `out_string` is given; `ASCII-8BIT`,
+      # otherwise.
+      def read(maxlen = nil, out_string = nil) # rubocop:disable Metrics/PerceivedComplexity
+        return (maxlen.nil? || maxlen.zero? ? '' : nil) if eof?
+
         tbuf = if @output_buffer.bytesize > 0
-                 if number_of_bytes && number_of_bytes <= @output_buffer.bytesize
-                   @output_buffer.slice!(0, number_of_bytes)
+                 if maxlen && maxlen <= @output_buffer.bytesize
+                   @output_buffer.slice!(0, maxlen)
                  else
-                   number_of_bytes -= @output_buffer.bytesize if number_of_bytes
-                   rbuf = sysread(number_of_bytes, buf)
+                   maxlen -= @output_buffer.bytesize if maxlen
+                   rbuf = produce_input(maxlen)
                    out  = @output_buffer
                    out << rbuf if rbuf
-                   @output_buffer = ''
+                   @output_buffer = +''.b
                    out
                  end
                else
-                 sysread(number_of_bytes, buf)
+                 produce_input(maxlen)
                end
 
         if tbuf.nil? || tbuf.empty?
-          return nil if number_of_bytes&.positive?
+          return nil if maxlen&.positive?
 
           return ''
         end
 
         @pos += tbuf.length
 
-        if buf
-          buf.replace(tbuf)
+        if out_string.nil?
+          tbuf.force_encoding(Encoding::ASCII_8BIT)
         else
-          buf = tbuf
+          encoding = out_string.encoding
+          out_string.replace(tbuf).force_encoding(encoding)
         end
-        buf
       end
 
-      def readlines(a_sep_string = $INPUT_RECORD_SEPARATOR)
-        ret_val = []
-        each_line(a_sep_string) { |line| ret_val << line }
-        ret_val
+      # Reads and returns all remaining lines from the stream. See the Line IO
+      # documentation in the IO class for more information.
+      #
+      # With no arguments given, returns lines as determined by line
+      # separator `$/`, or `nil` if none.
+      #
+      # With only string argument `sep` given, returns lines as
+      # determined by line separator `sep`, or `nil` if none. See the
+      # Line Separator documentation in the IO class for more information.
+      # The two special values for `sep` (`nil` and `""`) are honoured.
+      #
+      # With only integer argument `limit` given, limits the number of bytes
+      # in each line; see the Line Limit documentation in the IO class for more
+      # information.
+      #
+      # With arguments `sep` and `limit` given, combines the two behaviors.
+      #
+      # Optional keyword argument `chomp` specifies whether line separators
+      # are to be omitted.
+      def readlines(sep = $INPUT_RECORD_SEPARATOR, limit = nil, chomp: false)
+        each(sep, limit, chomp: chomp).to_a
       end
 
-      def gets(a_sep_string = $INPUT_RECORD_SEPARATOR, number_of_bytes = nil)
-        @lineno = @lineno.next
+      # Reads and returns a line from the stream. See the Line IO
+      # documentation in the IO class for more information.
+      #
+      # With no arguments given, returns the next line as determined by line
+      # separator `$/`, or `nil` if none.
+      #
+      # With only string argument `sep` given, returns the next line as
+      # determined by line separator `sep`, or `nil` if none. See the
+      # Line Separator documentation in the IO class for more information.
+      # The two special values for `sep` (`nil` and `""`) are honoured.
+      #
+      # With only integer argument `limit` given, limits the number of bytes
+      # in the line; see the Line Limit documentation in the IO class for more
+      # information.
+      #
+      # With arguments `sep` and `limit` given, combines the two behaviors.
+      #
+      # Optional keyword argument `chomp` specifies whether line separators
+      # are to be omitted.
+      def gets(sep = $INPUT_RECORD_SEPARATOR, limit = nil, chomp: false)
+        if sep.nil?
+          return nil if eof?
+
+          @lineno = @lineno.next
+          return read(limit)
+        end
 
-        if number_of_bytes.respond_to?(:to_int)
-          number_of_bytes = number_of_bytes.to_int
-          a_sep_string = a_sep_string.to_str if a_sep_string
-        elsif a_sep_string.respond_to?(:to_int)
-          number_of_bytes = a_sep_string.to_int
-          a_sep_string    = $INPUT_RECORD_SEPARATOR
-        else
-          number_of_bytes = nil
-          a_sep_string = a_sep_string.to_str if a_sep_string
+        if sep.respond_to?(:to_int)
+          limit = sep.to_int
+          sep   = $INPUT_RECORD_SEPARATOR
+        elsif sep&.empty?
+          sep = "#{$INPUT_RECORD_SEPARATOR}#{$INPUT_RECORD_SEPARATOR}"
         end
 
-        return read(number_of_bytes) if a_sep_string.nil?
+        buffer_index = 0
+        while (sep_index = @output_buffer.index(sep, buffer_index)).nil?
+          break if limit && @output_buffer.bytesize >= limit
 
-        a_sep_string = "#{$INPUT_RECORD_SEPARATOR}#{$INPUT_RECORD_SEPARATOR}" if a_sep_string.empty?
+          if input_finished?
+            return nil if @output_buffer.empty?
 
-        buffer_index = 0
-        over_limit   = number_of_bytes && @output_buffer.bytesize >= number_of_bytes
-        while (match_index = @output_buffer.index(a_sep_string, buffer_index)).nil? && !over_limit
-          buffer_index = [buffer_index, @output_buffer.bytesize - a_sep_string.bytesize].max
-          return @output_buffer.empty? ? nil : flush if input_finished?
+            @lineno = @lineno.next
+            @pos += @output_buffer.bytesize
+            return @output_buffer.slice!(0..)
+          end
 
+          buffer_index = [buffer_index, @output_buffer.bytesize - sep.bytesize].max
           @output_buffer << produce_input
-          over_limit = number_of_bytes && @output_buffer.bytesize >= number_of_bytes
         end
-        sep_index = [
-          match_index + a_sep_string.bytesize,
-          number_of_bytes || @output_buffer.bytesize
-        ].min
-        @pos += sep_index
-        @output_buffer.slice!(0...sep_index)
+
+        limit ||= @output_buffer.bytesize
+        cut_index = sep_index ? [sep_index + sep.bytesize, limit].min : limit
+        @lineno = @lineno.next
+        @pos += cut_index
+        chomp ? @output_buffer.slice!(0, cut_index).chomp(sep) : @output_buffer.slice!(0, cut_index)
       end
 
-      def ungetc(byte)
+      def ungetc(byte) # :nodoc:
         @output_buffer = byte.chr + @output_buffer
       end
 
-      def flush
-        ret_val        = @output_buffer
-        @output_buffer = +''
-        ret_val
+      def flush # :nodoc:
+        @output_buffer.slice!(0..)
       end
 
-      def readline(a_sep_string = $INPUT_RECORD_SEPARATOR)
-        ret_val = gets(a_sep_string)
-        raise EOFError unless ret_val
+      # Reads a line as with #gets, but raises `EOFError` if already at
+      # end-of-stream.
+      #
+      # Optional keyword argument `chomp` specifies whether line separators
+      # are to be omitted.
+      def readline(sep = $INPUT_RECORD_SEPARATOR, limit = nil, chomp: false)
+        raise EOFError if eof?
 
-        ret_val
+        gets(sep, limit, chomp: chomp)
       end
 
-      def each_line(a_sep_string = $INPUT_RECORD_SEPARATOR)
-        loop { yield readline(a_sep_string) }
-      rescue EOFError
-        # We just need to catch this; we don't need to handle it.
+      # Calls the block with each remaining line read from the stream.
+      # Does nothing if already at end-of-stream. See the Line IO
+      # documentation in the IO class for more information.
+      #
+      # With no arguments given, reads lines as determined by line separator
+      # `$/`. With only string argument `sep` given, reads lines as determined
+      # by line separator `sep`. See the Line Separator documentation in the
+      # IO class for more information. The two special values for `sep`
+      # (`nil` and `""`) are honoured.
+      #
+      # With only integer argument `limit` given, limits the number of bytes
+      # in each line; see the Line Limit documentation in the IO class for
+      # more information.
+      #
+      # With arguments `sep` and `limit` given, combines the two behaviors.
+      #
+      # Optional keyword argument `chomp` specifies whether line separators
+      # are to be omitted.
+      #
+      # Returns an `Enumerator` if no block is given.
+      def each(sep = $INPUT_RECORD_SEPARATOR, limit = nil, chomp: false)
+        return to_enum(:each, sep, limit, chomp: chomp) unless block_given?
+
+        while (line = gets(sep, limit, chomp: chomp))
+          yield line
+        end
       end
 
-      alias each each_line
+      alias each_line each
 
+      # Returns `true` if the stream is positioned at its end, `false`
+      # otherwise. See Position documentation in the IO class for more
+      # information.
       def eof?
         @output_buffer.empty? && input_finished?
       end
 
       # Alias for compatibility. Remove for version 4.
-      alias eof eof?
+      alias eof eof? # :nodoc:
     end
   end
 end

data/lib/zip/pass_thru_decompressor.rb

@@ -7,15 +7,15 @@ module Zip
       @read_so_far = 0
     end
 
-    def read(length = nil, outbuf = +'')
-      return (length.nil? || length.zero? ? '' : nil) if eof?
+    def read(maxlen = nil)
+      return (maxlen.nil? || maxlen.zero? ? '' : nil) if eof?
 
-      if length.nil? || (@read_so_far + length) > decompressed_size
-        length = decompressed_size - @read_so_far
+      if maxlen.nil? || (@read_so_far + maxlen) > decompressed_size
+        maxlen = decompressed_size - @read_so_far
       end
 
-      @read_so_far += length
-      input_stream.read(length, outbuf)
+      @read_so_far += maxlen
+      input_stream.read(maxlen)
     end
 
     def eof?

data/lib/zip/version.rb

@@ -2,5 +2,5 @@
 
 module Zip
   # The version of the Rubyzip library.
-  VERSION = '3.2.1'
+  VERSION = '3.3.0'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: rubyzip
 version: !ruby/object:Gem::Version
-  version: 3.2.1
+  version: 3.3.0
 platform: ruby
 authors:
 - Robert Haines
@@ -195,9 +195,9 @@ licenses:
 - BSD-2-Clause
 metadata:
   bug_tracker_uri: https://github.com/rubyzip/rubyzip/issues
-  changelog_uri: https://github.com/rubyzip/rubyzip/blob/v3.2.1/Changelog.md
-  documentation_uri: https://www.rubydoc.info/gems/rubyzip/3.2.1
-  source_code_uri: https://github.com/rubyzip/rubyzip/tree/v3.2.1
+  changelog_uri: https://github.com/rubyzip/rubyzip/blob/v3.3.0/Changelog.md
+  documentation_uri: https://www.rubydoc.info/gems/rubyzip/3.3.0
+  source_code_uri: https://github.com/rubyzip/rubyzip/tree/v3.3.0
   wiki_uri: https://github.com/rubyzip/rubyzip/wiki
   rubygems_mfa_required: 'true'
 rdoc_options: []