excavate 0.1.0

data/lib/excavate/extractors/cpio_extractor.rb

@@ -0,0 +1,44 @@
+require_relative "cpio/cpio"
+require_relative "cpio/cpio_old_format"
+
+module Excavate
+  module Extractors
+    class CpioExtractor < Extractor
+      def extract(target)
+        extract_cpio_new_format(target)
+      rescue RuntimeError => e
+        raise unless e.message.start_with?("Invalid magic")
+
+        extract_cpio_old_format(target)
+      end
+
+      private
+
+      def extract_cpio_new_format(target)
+        File.open(@archive, "rb") do |archive_file|
+          CPIO::ASCIIReader.new(archive_file).each do |entry, file|
+            path = File.join(target, entry.name)
+            if entry.directory?
+              FileUtils.mkdir_p(path)
+            else
+              File.write(path, file.read, mode: "wb")
+            end
+          end
+        end
+      end
+
+      def extract_cpio_old_format(target)
+        File.open(@archive, "rb") do |archive_file|
+          CPIO::ArchiveReader.new(archive_file).each_entry do |entry|
+            path = File.expand_path(entry.filename, target)
+            if entry.directory?
+              FileUtils.mkdir_p(path)
+            else
+              File.write(path, entry.data, mode: "wb")
+            end
+          end
+        end
+      end
+    end
+  end
+end

data/lib/excavate/extractors/extractor.rb

@@ -0,0 +1,13 @@
+module Excavate
+  module Extractors
+    class Extractor
+      def initialize(archive)
+        @archive = archive
+      end
+
+      def extract(_target)
+        raise NotImplementedError.new("You must implement this method")
+      end
+    end
+  end
+end

data/lib/excavate/extractors/gzip_extractor.rb

@@ -0,0 +1,13 @@
+module Excavate
+  module Extractors
+    class GzipExtractor < Extractor
+      def extract(target)
+        Zlib::GzipReader.open(@archive) do |gz|
+          basename = File.basename(@archive, ".*")
+          path = File.join(target, basename)
+          File.write(path, gz.read, mode: "wb")
+        end
+      end
+    end
+  end
+end

data/lib/excavate/extractors/ole_extractor.rb

@@ -0,0 +1,66 @@
+require "ole/storage"
+
+module Excavate
+  module Extractors
+    class OleExtractor < Extractor
+      def extract(target)
+        do_extract(target)
+        rename_archives(target)
+      end
+
+      private
+
+      def do_extract(target)
+        reset_filename_lookup
+
+        Ole::Storage.open(@archive) do |ole|
+          children(ole).each do |file|
+            filename = prepare_filename(file)
+            path = File.join(target, filename)
+            content = ole.file.read(file)
+            File.write(path, content, mode: "wb")
+          end
+        end
+      end
+
+      def children(ole)
+        ole.dir.entries(".") - [".", ".."]
+      end
+
+      def reset_filename_lookup
+        @file_lookup = {}
+      end
+
+      def prepare_filename(file)
+        filename = sanitize_filename(file)
+
+        @file_lookup[filename] ||= 0
+        @file_lookup[filename] += 1
+        filename += @file_lookup[filename].to_s if @file_lookup[filename] > 1
+
+        filename
+      end
+
+      def sanitize_filename(filename)
+        filename.strip.tap do |name|
+          # NOTE: File.basename doesn't work right with Windows paths on Unix
+          # get only the filename, not the whole path
+          name.gsub!(/^.*(\\|\/)/, "")
+
+          # Strip out the non-ascii character
+          name.gsub!(/[^0-9A-Za-z.\-]/, "_")
+        end
+      end
+
+      def rename_archives(target)
+        Dir.glob(File.join(target, "**", "*")).each do |file|
+          FileUtils.mv(file, "#{file}.cab") if cab?(file)
+        end
+      end
+
+      def cab?(file)
+        FileMagic.detect(file) == :cab
+      end
+    end
+  end
+end

data/lib/excavate/extractors/rpm_extractor.rb

@@ -0,0 +1,33 @@
+require "arr-pm"
+
+# fix for Ruby 3.0
+unless RPM::File::Header::HEADER_MAGIC == "\x8e\xad\xe8\x01\x00\x00\x00\x00".force_encoding("BINARY")
+  RPM::File::Header.send(:remove_const, "HEADER_MAGIC")
+  RPM::File::Header.const_set("HEADER_MAGIC", "\x8e\xad\xe8\x01\x00\x00\x00\x00".force_encoding("BINARY"))
+end
+
+module Excavate
+  module Extractors
+    class RpmExtractor < Extractor
+      def extract(target)
+        File.open(@archive, "rb") do |file|
+          rpm = RPM::File.new(file)
+          content = rpm.payload.read
+          path = target_path(@archive, rpm.tags, target)
+
+          File.write(path, content, mode: "wb")
+        end
+      end
+
+      private
+
+      def target_path(archive, tags, dir)
+        archive_format = tags[:payloadformat]
+        compression_format = tags[:payloadcompressor] == "gzip" ? "gz" : tags[:payloadcompressor]
+        basename = File.basename(archive, ".*")
+        filename = basename + "." + archive_format + "." + compression_format
+        File.join(dir, filename)
+      end
+    end
+  end
+end

data/lib/excavate/extractors/seven_zip_extractor.rb

@@ -0,0 +1,13 @@
+require "seven_zip_ruby"
+
+module Excavate
+  module Extractors
+    class SevenZipExtractor < Extractor
+      def extract(target)
+        File.open(@archive, "rb") do |file|
+          SevenZipRuby::Reader.extract_all(file, target)
+        end
+      end
+    end
+  end
+end

data/lib/excavate/extractors/tar_extractor.rb

@@ -0,0 +1,29 @@
+require "rubygems/package"
+
+module Excavate
+  module Extractors
+    class TarExtractor < Extractor
+      def extract(target)
+        File.open(@archive, "rb") do |archive_file|
+          Gem::Package::TarReader.new(archive_file) do |tar|
+            tar.each do |tarfile|
+              save_tar_file(tarfile, target)
+            end
+          end
+        end
+      end
+
+      private
+
+      def save_tar_file(file, dir)
+        path = File.join(dir, file.full_name)
+
+        if file.directory?
+          FileUtils.mkdir_p(path)
+        else
+          File.write(path, file.read, mode: "wb")
+        end
+      end
+    end
+  end
+end

data/lib/excavate/extractors/zip_extractor.rb

@@ -0,0 +1,17 @@
+require "zip"
+
+module Excavate
+  module Extractors
+    class ZipExtractor < Extractor
+      def extract(target)
+        Zip::File.open(@archive) do |zip_file|
+          zip_file.each do |entry|
+            path = File.join(target, entry.name)
+            FileUtils.mkdir_p(File.dirname(path))
+            entry.extract(path)
+          end
+        end
+      end
+    end
+  end
+end

data/lib/excavate/file_magic.rb

@@ -0,0 +1,18 @@
+module Excavate
+  class FileMagic
+    def self.detect(path)
+      new(path).detect
+    end
+
+    def initialize(path)
+      @path = path
+    end
+
+    def detect
+      case File.read(@path, 8, mode: "rb")
+      when "MSCF\x00\x00\x00\x00".force_encoding("BINARY")
+        :cab
+      end
+    end
+  end
+end

data/lib/excavate/utils.rb

@@ -0,0 +1,14 @@
+module Excavate
+  module Utils
+    module_function
+
+    def silence_stream(stream)
+      old_stream = stream.dup
+      stream.reopen(RbConfig::CONFIG["host_os"] =~ /mswin|mingw/ ? "NUL:" : "/dev/null") # rubocop:disable Performance/RegexpMatch, Metrics/LineLength
+      stream.sync = true
+      yield
+    ensure
+      stream.reopen(old_stream)
+    end
+  end
+end

data/lib/excavate/version.rb

@@ -0,0 +1,5 @@
+# frozen_string_literal: true
+
+module Excavate
+  VERSION = "0.1.0"
+end

metadata

@@ -0,0 +1,143 @@
+--- !ruby/object:Gem::Specification
+name: excavate
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Ribose Inc.
+autorequire:
+bindir: exe
+cert_chain: []
+date: 2021-02-26 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: arr-pm
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.0'
+- !ruby/object:Gem::Dependency
+  name: libmspack
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.1'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.1'
+- !ruby/object:Gem::Dependency
+  name: ruby-ole
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+- !ruby/object:Gem::Dependency
+  name: rubyzip
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.3'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.3'
+- !ruby/object:Gem::Dependency
+  name: seven_zip_ruby
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+description: Extract nested archives with a single command.
+email:
+- open.source@ribose.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- ".github/workflows/release.yml"
+- ".github/workflows/rspec.yml"
+- ".gitignore"
+- ".rspec"
+- ".rubocop.yml"
+- Gemfile
+- LICENSE.adoc
+- README.adoc
+- Rakefile
+- excavate.gemspec
+- lib/excavate.rb
+- lib/excavate/archive.rb
+- lib/excavate/extractors.rb
+- lib/excavate/extractors/cab_extractor.rb
+- lib/excavate/extractors/cpio/cpio.rb
+- lib/excavate/extractors/cpio/cpio_old_format.rb
+- lib/excavate/extractors/cpio_extractor.rb
+- lib/excavate/extractors/extractor.rb
+- lib/excavate/extractors/gzip_extractor.rb
+- lib/excavate/extractors/ole_extractor.rb
+- lib/excavate/extractors/rpm_extractor.rb
+- lib/excavate/extractors/seven_zip_extractor.rb
+- lib/excavate/extractors/tar_extractor.rb
+- lib/excavate/extractors/zip_extractor.rb
+- lib/excavate/file_magic.rb
+- lib/excavate/utils.rb
+- lib/excavate/version.rb
+homepage: https://github.com/fontist/excavate
+licenses:
+- BSD-3-Clause
+metadata:
+  homepage_uri: https://github.com/fontist/excavate
+  source_code_uri: https://github.com/fontist/excavate
+  changelog_uri: https://github.com/fontist/excavate
+post_install_message:
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.0.3
+signing_key:
+specification_version: 4
+summary: Extract nested archives with a single command.
+test_files: []