iostreams 1.6.2 → 1.7.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: a6dd68508835099ef4c9de7f81a9527927d3225138f0bf5ecd00c44194e11858
-  data.tar.gz: 28535e95ca83a4cf0c522de4cd48889a489b4fcefdc8feb9a7bfe0b70124fd7a
+  metadata.gz: 849ceda63eb30f95762a7c985cd215d424e62afd68ab20776e8d16c188dd6aed
+  data.tar.gz: 8e26af86c40bb673ce36855a7fb30d1c4b401edc3eac0b27a71b9760cfe865dd
 SHA512:
-  metadata.gz: 8e4d38ef41234f62fdcfde1ae3a96fe59203dcfc38562843106b081c7292efffdda36b88707674059a24b0fa2996d4c1dbe0627694e80a9b12d23d47195a53d3
-  data.tar.gz: 23e16854c305542ee0976f570444d7b99cadcb6ec460d83bebc73a65cba9c28e3beba485c9943797b174b80bdd783c216d89d539e8aecdbda62e1c31f017efce
+  metadata.gz: 99318c4c64e0133df57b84429b1c2f9caa064abb1405ace5d55208e41b6bf8bb8fa83a75db8ae46d53753f10d566bab53971d95871ed4011bab4571d31bebe8a
+  data.tar.gz: bfba3a033c753e3fe05f798177b8f3c7ee8f566eabaf9223984fa902b288cb3515154f621a4e97f75b6c5bc31da88f284390c2d82c5015d7682ecf08c2a671d3

data/lib/io_streams/builder.rb

@@ -97,7 +97,9 @@ module IOStreams
     end
 
     def format=(format)
-      raise(ArgumentError, "Invalid format: #{format.inspect}") unless format.nil? || IOStreams::Tabular.registered_formats.include?(format)
+      unless format.nil? || IOStreams::Tabular.registered_formats.include?(format)
+        raise(ArgumentError, "Invalid format: #{format.inspect}")
+      end
 
       @format = format
     end
@@ -106,7 +108,7 @@ module IOStreams
 
     def class_for_stream(type, stream)
       ext = IOStreams.extensions[stream.nil? ? nil : stream.to_sym] ||
-        raise(ArgumentError, "Unknown Stream type: #{stream.inspect}")
+            raise(ArgumentError, "Unknown Stream type: #{stream.inspect}")
       ext.send("#{type}_class") || raise(ArgumentError, "No #{type} registered for Stream type: #{stream.inspect}")
     end
 

data/lib/io_streams/line/reader.rb

@@ -96,17 +96,17 @@ module IOStreams
           while line.count(@embedded_within).odd?
             if eof? || line.length > @buffer_size * 10
               raise(Errors::MalformedDataError.new(
-                "Unbalanced delimited field, delimiter: #{@embedded_within}",
-                initial_line_number
-              ))
+                      "Unbalanced delimited field, delimiter: #{@embedded_within}",
+                      initial_line_number
+                    ))
             end
             line << @delimiter
             next_line = _readline
             if next_line.nil?
               raise(Errors::MalformedDataError.new(
-                "Unbalanced delimited field, delimiter: #{@embedded_within}",
-                initial_line_number
-              ))
+                      "Unbalanced delimited field, delimiter: #{@embedded_within}",
+                      initial_line_number
+                    ))
             end
             line << next_line
           end

data/lib/io_streams/paths/s3.rb

@@ -3,7 +3,7 @@ require "uri"
 module IOStreams
   module Paths
     class S3 < IOStreams::Path
-      attr_reader :bucket_name, :client, :options
+      attr_reader :bucket_name, :options
 
       # Largest file size supported by the S3 copy object api.
       S3_COPY_OBJECT_SIZE_LIMIT = 5 * 1024 * 1024 * 1024
@@ -141,16 +141,17 @@ module IOStreams
 
         @bucket_name = uri.hostname
         key          = uri.path.sub(%r{\A/}, "")
-        if client.is_a?(Hash)
-          client[:access_key_id]     = access_key_id if access_key_id
-          client[:secret_access_key] = secret_access_key if secret_access_key
-          @client                    = ::Aws::S3::Client.new(client)
+
+        if client && !client.is_a?(Hash)
+          @client = client
         else
-          @client = client || ::Aws::S3::Client.new(access_key_id: access_key_id, secret_access_key: secret_access_key)
+          @client_options                     = client.is_a?(Hash) ? client.dup : {}
+          @client_options[:access_key_id]     = access_key_id if access_key_id
+          @client_options[:secret_access_key] = secret_access_key if secret_access_key
         end
-        @options = args
 
-        @options.merge(uri.query) if uri.query
+        @options = args
+        @options.merge!(uri.query.transform_keys(&:to_sym)) if uri.query
 
         super(key)
       end
@@ -190,11 +191,11 @@ module IOStreams
       end
 
       # Make S3 perform direct copies within S3 itself.
-      def copy_to(target_path, convert: true)
-        return super(target_path) if convert || (size.to_i >= S3_COPY_OBJECT_SIZE_LIMIT)
+      def copy_to(target_path, convert: true, **args)
+        return super(target_path, convert: convert, **args) if convert || (size.to_i >= S3_COPY_OBJECT_SIZE_LIMIT)
 
         target = IOStreams.new(target_path)
-        return super(target) unless target.is_a?(self.class)
+        return super(target, convert: convert, **args) unless target.is_a?(self.class)
 
         source_name = ::File.join(bucket_name, path)
         client.copy_object(options.merge(bucket: target.bucket_name, key: target.path, copy_source: source_name))
@@ -202,11 +203,13 @@ module IOStreams
       end
 
       # Make S3 perform direct copies within S3 itself.
-      def copy_from(source_path, convert: true)
-        return super(source_path) if convert
+      def copy_from(source_path, convert: true, **args)
+        return super(source_path, convert: true, **args) if convert
 
         source = IOStreams.new(source_path)
-        return super(source) if !source.is_a?(self.class) || (source.size.to_i >= S3_COPY_OBJECT_SIZE_LIMIT)
+        if !source.is_a?(self.class) || (source.size.to_i >= S3_COPY_OBJECT_SIZE_LIMIT)
+          return super(source, convert: convert, **args)
+        end
 
         source_name = ::File.join(source.bucket_name, source.path)
         client.copy_object(options.merge(bucket: bucket_name, key: path, copy_source: source_name))
@@ -312,6 +315,11 @@ module IOStreams
       def partial_files_visible?
         false
       end
+
+      # Lazy load S3 client since it takes two seconds to create itself!
+      def client
+        @client ||= ::Aws::S3::Client.new(@client_options)
+      end
     end
   end
 end

data/lib/io_streams/paths/sftp.rb

@@ -47,9 +47,23 @@ module IOStreams
       # password: [String]
       #   Password for the user.
       #
-      # **ssh_options
-      #   Any other options supported by ssh_config.
-      #   `man ssh_config` to see all available options.
+      # ssh_options: [Hash]
+      #   - IdentityKey [String]
+      #     The identity key that this client should use to talk to this host.
+      #     Under the covers this value is written to a file and then the file name is passed as `IdentityFile`
+      #   - HostKey [String]
+      #     The expected SSH Host key that is presented by the remote host.
+      #     Instead of storing the host key in the `known_hosts` file, it can be supplied explicity
+      #     using this option.
+      #     Under the covers this value is written to a file and then the file name is passed as `UserKnownHostsFile`
+      #     Notes:
+      #     - It must contain the entire line that would be stored in `known_hosts`,
+      #       including the hostname, ip address, key type and key value. This value is written as-is into a
+      #       "known_hosts" like file and then passed into sftp using the `UserKnownHostsFile` option.
+      #     - The easiest way to generate the required is to use `ssh-keyscan` and then supply that value in this field.
+      #       For example: `ssh-keyscan hostname`
+      #   - Any other options supported by ssh_config.
+      #     `man ssh_config` to see all available options.
       #
       # Examples:
       #
@@ -168,38 +182,36 @@ module IOStreams
       def sftp_download(remote_file_name, local_file_name)
         with_sftp_args do |args|
           Open3.popen2e(*args) do |writer, reader, waith_thr|
-            begin
-              # Give time for remote sftp server to get ready to accept the password.
-              sleep self.class.before_password_wait_seconds
-
-              writer.puts password
-
-              # Give time for password to be processed and stdin to be passed to sftp process.
-              sleep self.class.sshpass_wait_seconds
-
-              writer.puts "get #{remote_file_name} #{local_file_name}"
-              writer.puts "bye"
-              writer.close
-              out = reader.read.chomp
-              unless waith_thr.value.success?
-                raise(
-                  Errors::CommunicationsFailure,
-                  "Download failed calling #{self.class.sftp_bin} via #{self.class.sshpass_bin}: #{out}"
-                )
-              end
-
-              out
-            rescue Errno::EPIPE
-              out = begin
-                      reader.read.chomp
-                    rescue StandardError
-                      nil
-                    end
+            # Give time for remote sftp server to get ready to accept the password.
+            sleep self.class.before_password_wait_seconds
+
+            writer.puts password
+
+            # Give time for password to be processed and stdin to be passed to sftp process.
+            sleep self.class.sshpass_wait_seconds
+
+            writer.puts "get #{remote_file_name} #{local_file_name}"
+            writer.puts "bye"
+            writer.close
+            out = reader.read.chomp
+            unless waith_thr.value.success?
               raise(
                 Errors::CommunicationsFailure,
                 "Download failed calling #{self.class.sftp_bin} via #{self.class.sshpass_bin}: #{out}"
               )
             end
+
+            out
+          rescue Errno::EPIPE
+            out = begin
+              reader.read.chomp
+            rescue StandardError
+              nil
+            end
+            raise(
+              Errors::CommunicationsFailure,
+              "Download failed calling #{self.class.sftp_bin} via #{self.class.sshpass_bin}: #{out}"
+            )
           end
         end
       end
@@ -207,48 +219,64 @@ module IOStreams
       def sftp_upload(local_file_name, remote_file_name)
         with_sftp_args do |args|
           Open3.popen2e(*args) do |writer, reader, waith_thr|
-            begin
-              writer.puts(password) if password
-              # Give time for password to be processed and stdin to be passed to sftp process.
-              sleep self.class.sshpass_wait_seconds
-              writer.puts "put #{local_file_name.inspect} #{remote_file_name.inspect}"
-              writer.puts "bye"
-              writer.close
-              out = reader.read.chomp
-              unless waith_thr.value.success?
-                raise(
-                  Errors::CommunicationsFailure,
-                  "Upload failed calling #{self.class.sftp_bin} via #{self.class.sshpass_bin}: #{out}"
-                )
-              end
-
-              out
-            rescue Errno::EPIPE
-              out = begin
-                      reader.read.chomp
-                    rescue StandardError
-                      nil
-                    end
+            writer.puts(password) if password
+            # Give time for password to be processed and stdin to be passed to sftp process.
+            sleep self.class.sshpass_wait_seconds
+            writer.puts "put #{local_file_name.inspect} #{remote_file_name.inspect}"
+            writer.puts "bye"
+            writer.close
+            out = reader.read.chomp
+            unless waith_thr.value.success?
               raise(
                 Errors::CommunicationsFailure,
                 "Upload failed calling #{self.class.sftp_bin} via #{self.class.sshpass_bin}: #{out}"
               )
             end
+
+            out
+          rescue Errno::EPIPE
+            out = begin
+              reader.read.chomp
+            rescue StandardError
+              nil
+            end
+            raise(
+              Errors::CommunicationsFailure,
+              "Upload failed calling #{self.class.sftp_bin} via #{self.class.sshpass_bin}: #{out}"
+            )
           end
         end
       end
 
       def with_sftp_args
-        return yield sftp_args(ssh_options) unless ssh_options.key?("IdentityKey")
+        return yield sftp_args(ssh_options) if !ssh_options.key?("IdentityKey") && !ssh_options.key?("HostKey")
+
+        with_identity_key(ssh_options.dup) do |options|
+          with_host_key(options) do |options2|
+            yield sftp_args(options2)
+          end
+        end
+      end
+
+      def with_identity_key(options)
+        return yield options unless ssh_options.key?("IdentityKey")
+
+        with_temp_file(options, "IdentityFile", options.delete("IdentityKey")) { yield options }
+      end
+
+      def with_host_key(options)
+        return yield options unless ssh_options.key?("HostKey")
+
+        with_temp_file(options, "UserKnownHostsFile", options.delete("HostKey")) { yield options }
+      end
 
+      def with_temp_file(options, option, value)
         Utils.temp_file_name("iostreams-sftp-args", "key") do |file_name|
-          options = ssh_options.dup
-          key     = options.delete("IdentityKey")
           # sftp requires that private key is only readable by the current user
-          ::File.open(file_name, "wb", 0o600) { |io| io.write(key) }
+          ::File.open(file_name, "wb", 0o600) { |io| io.write(value) }
 
-          options["IdentityFile"] = file_name
-          yield sftp_args(options)
+          options[option] = file_name
+          yield options
         end
       end
 
@@ -277,7 +305,7 @@ module IOStreams
       end
 
       def build_ssh_options
-        options                = ssh_options.dup
+        options = ssh_options.dup
         options[:logger]       ||= logger if defined?(SemanticLogger)
         options[:port]         ||= port
         options[:max_pkt_size] ||= 65_536

data/lib/io_streams/pgp.rb

@@ -48,8 +48,8 @@ module IOStreams
     # See `man gpg` for the remaining options
     def self.generate_key(name:,
                           email:,
-                          comment: nil,
                           passphrase:,
+                          comment: nil,
                           key_type: "RSA",
                           key_length: 4096,
                           subkey_type: "RSA",
@@ -291,10 +291,8 @@ module IOStreams
       version_check
       Open3.popen2e("#{executable} --list-keys --fingerprint --with-colons #{email}") do |_stdin, out, waith_thr|
         output = out.read.chomp
-        unless waith_thr.value.success?
-          unless output =~ /(public key not found|No public key)/i
-            raise(Pgp::Failure, "GPG Failed calling #{executable} to list keys for #{email}: #{output}")
-          end
+        if !waith_thr.value.success? && !(output !~ /(public key not found|No public key)/i)
+          raise(Pgp::Failure, "GPG Failed calling #{executable} to list keys for #{email}: #{output}")
         end
 
         output.each_line do |line|
@@ -336,9 +334,11 @@ module IOStreams
             match[1]
           end
         else
-          return [] if err =~ /(key not found|No (public|secret) key)/i
+          if err !~ /(key not found|No (public|secret) key)/i
+            raise(Pgp::Failure, "GPG Failed calling #{executable} to list keys for #{email || key_id}: #{err}#{out}")
+          end
 
-          raise(Pgp::Failure, "GPG Failed calling #{executable} to list keys for #{email || key_id}: #{err}#{out}")
+          []
         end
       end
     end
@@ -382,10 +382,10 @@ module IOStreams
             key_length: match[3].to_s.to_i,
             key_type:   match[2],
             date:       (begin
-                           Date.parse(match[4].to_s)
-                         rescue StandardError
-                           match[4]
-                         end)
+              Date.parse(match[4].to_s)
+            rescue StandardError
+              match[4]
+            end)
           }
         elsif (match = line.match(%r{(pub|sec)\s+(\d+)(.*)/(\w+)\s+(\d+-\d+-\d+)(\s+(.+)<(.+)>)?}))
           # Matches: pub  2048R/C7F9D9CB 2016-10-26
@@ -396,10 +396,10 @@ module IOStreams
             key_type:   match[3],
             key_id:     match[4],
             date:       (begin
-                           Date.parse(match[5].to_s)
-                         rescue StandardError
-                           match[5]
-                         end)
+              Date.parse(match[5].to_s)
+            rescue StandardError
+              match[5]
+            end)
           }
           # Prior to gpg v2.0.30
           if match[7]

data/lib/io_streams/pgp/writer.rb

@@ -19,8 +19,7 @@ module IOStreams
 
         private
 
-        attr_reader :default_signer_passphrase
-        attr_reader :default_signer
+        attr_reader :default_signer_passphrase, :default_signer
 
         @default_signer_passphrase = nil
         @default_signer            = nil

data/lib/io_streams/stream.rb

@@ -151,6 +151,9 @@ module IOStreams
     #     Whether to apply the stream conversions during the copy.
     #     Default: true
     #
+    #   :mode [:line, :array, :hash]
+    #     When convert is `true` then use this mode to convert the contents of the file.
+    #
     # Examples:
     #
     # # Copy and convert streams based on file extensions
@@ -162,11 +165,17 @@ module IOStreams
     # # Advanced copy with custom stream conversions on source and target.
     # source = IOStreams.path("source_file").stream(encoding: "BINARY")
     # IOStreams.path("target_file.pgp").option(:pgp, passphrase: "hello").copy_from(source)
-    def copy_from(source, convert: true)
+    def copy_from(source, convert: true, mode: nil, **args)
       if convert
         stream = IOStreams.new(source)
-        writer do |target|
-          stream.reader { |src| IO.copy_stream(src, target) }
+        if mode
+          writer(mode, **args) do |target|
+            stream.each(mode) { |row| target << row }
+          end
+        else
+          writer(**args) do |target|
+            stream.reader { |src| IO.copy_stream(src, target) }
+          end
         end
       else
         stream = source.is_a?(Stream) ? source.dup : IOStreams.new(source)
@@ -176,9 +185,9 @@ module IOStreams
       end
     end
 
-    def copy_to(target, convert: true)
+    def copy_to(target, **args)
       target = IOStreams.new(target)
-      target.copy_from(self, convert: convert)
+      target.copy_from(self, **args)
     end
 
     # Set/get the original file_name
@@ -365,8 +374,8 @@ module IOStreams
         IOStreams::Row::Writer.stream(
           io,
           original_file_name: builder.file_name,
-          format: builder.format,
-          format_options: builder.format_options,
+          format:             builder.format,
+          format_options:     builder.format_options,
           **args,
           &block
         )
@@ -380,10 +389,11 @@ module IOStreams
         IOStreams::Record::Writer.stream(
           io,
           original_file_name: builder.file_name,
-          format: builder.format,
-          format_options: builder.format_options,
+          format:             builder.format,
+          format_options:     builder.format_options,
           **args,
-          &block)
+          &block
+        )
       end
     end
   end

data/lib/io_streams/tabular/parser/fixed.rb

@@ -146,7 +146,7 @@ module IOStreams
 
           attr_reader :key, :size, :type, :decimals
 
-          def initialize(key: nil, size:, type: :string, decimals: 2)
+          def initialize(size:, key: nil, type: :string, decimals: 2)
             @key      = key
             @size     = size == :remainder ? -1 : size.to_i
             @type     = type.to_sym

data/lib/io_streams/utils.rb

@@ -28,11 +28,9 @@ module IOStreams
     def self.temp_file_name(basename, extension = "")
       result = nil
       ::Dir::Tmpname.create([basename, extension], IOStreams.temp_dir, max_try: MAX_TEMP_FILE_NAME_ATTEMPTS) do |tmpname|
-        begin
-          result = yield(tmpname)
-        ensure
-          ::File.unlink(tmpname) if ::File.exist?(tmpname)
-        end
+        result = yield(tmpname)
+      ensure
+        ::File.unlink(tmpname) if ::File.exist?(tmpname)
       end
       result
     end

data/lib/io_streams/version.rb

@@ -1,3 +1,3 @@
 module IOStreams
-  VERSION = "1.6.2".freeze
+  VERSION = "1.7.0".freeze
 end

data/lib/iostreams.rb

@@ -23,33 +23,41 @@ module IOStreams
     autoload :Reader, "io_streams/bzip2/reader"
     autoload :Writer, "io_streams/bzip2/writer"
   end
+
   module Encode
     autoload :Reader, "io_streams/encode/reader"
     autoload :Writer, "io_streams/encode/writer"
   end
+
   module Gzip
     autoload :Reader, "io_streams/gzip/reader"
     autoload :Writer, "io_streams/gzip/writer"
   end
+
   module Line
     autoload :Reader, "io_streams/line/reader"
     autoload :Writer, "io_streams/line/writer"
   end
+
   module Record
     autoload :Reader, "io_streams/record/reader"
     autoload :Writer, "io_streams/record/writer"
   end
+
   module Row
     autoload :Reader, "io_streams/row/reader"
     autoload :Writer, "io_streams/row/writer"
   end
+
   module SymmetricEncryption
     autoload :Reader, "io_streams/symmetric_encryption/reader"
     autoload :Writer, "io_streams/symmetric_encryption/writer"
   end
+
   module Xlsx
     autoload :Reader, "io_streams/xlsx/reader"
   end
+
   module Zip
     autoload :Reader, "io_streams/zip/reader"
     autoload :Writer, "io_streams/zip/writer"

data/test/paths/file_test.rb

@@ -126,15 +126,13 @@ module Paths
 
         it "missing source file" do
           IOStreams.temp_file("iostreams_move_test", ".txt") do |temp_file|
-            begin
-              refute temp_file.exist?
-              target = temp_file.directory.join("move_test.txt")
-              assert_raises Errno::ENOENT do
-                temp_file.move_to(target)
-              end
-              refute target.exist?
-              refute temp_file.exist?
+            refute temp_file.exist?
+            target = temp_file.directory.join("move_test.txt")
+            assert_raises Errno::ENOENT do
+              temp_file.move_to(target)
             end
+            refute target.exist?
+            refute temp_file.exist?
           end
         end
 

data/test/paths/sftp_test.rb

@@ -20,7 +20,13 @@ module Paths
       let(:file_name) { File.join(File.dirname(__FILE__), "..", "files", "text file.txt") }
       let(:raw) { File.read(file_name) }
 
-      let(:root_path) { IOStreams::Paths::SFTP.new(url, username: username, password: password) }
+      let(:root_path) do
+        if ENV["SFTP_HOST_KEY"]
+          IOStreams::Paths::SFTP.new(url, username: username, password: password, ssh_options: {"HostKey" => ENV["SFTP_HOST_KEY"]})
+        else
+          IOStreams::Paths::SFTP.new(url, username: username, password: password)
+        end
+      end
 
       let :existing_path do
         path = root_path.join("test.txt")

data/test/stream_test.rb

@@ -45,9 +45,9 @@ class StreamTest < Minitest::Test
       it "reads a zip file" do
         File.open(multiple_zip_file_name, "rb") do |io|
           result = IOStreams::Stream.new(io).
-            file_name(multiple_zip_file_name).
-            option(:zip, entry_file_name: "test.json").
-            read
+                   file_name(multiple_zip_file_name).
+                   option(:zip, entry_file_name: "test.json").
+                   read
           assert_equal contents_test_json, result
         end
       end
@@ -55,8 +55,8 @@ class StreamTest < Minitest::Test
       it "reads a zip file from within a gz file" do
         File.open(zip_gz_file_name, "rb") do |io|
           result = IOStreams::Stream.new(io).
-            file_name(zip_gz_file_name).
-            read
+                   file_name(zip_gz_file_name).
+                   read
           assert_equal contents_test_txt, result
         end
       end

data/test/test_helper.rb

@@ -2,7 +2,6 @@ $LOAD_PATH.unshift File.dirname(__FILE__) + "/../lib"
 
 require "yaml"
 require "minitest/autorun"
-require "minitest/reporters"
 require "iostreams"
 require "amazing_print"
 require "symmetric-encryption"
@@ -10,8 +9,6 @@ require "symmetric-encryption"
 # Since PGP libraries use UTC for Dates
 ENV["TZ"] = "UTC"
 
-Minitest::Reporters.use! Minitest::Reporters::SpecReporter.new
-
 # Test cipher
 SymmetricEncryption.cipher = SymmetricEncryption::Cipher.new(
   cipher_name: "aes-128-cbc",

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: iostreams
 version: !ruby/object:Gem::Version
-  version: 1.6.2
+  version: 1.7.0
 platform: ruby
 authors:
 - Reid Morrison
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2021-05-04 00:00:00.000000000 Z
+date: 2021-06-23 00:00:00.000000000 Z
 dependencies: []
 description:
 email:
@@ -125,7 +125,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: '2.3'
+      version: '2.5'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="