pdi 1.0.1 → 2.0.0.pre.alpha

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 4dc8c8b41b233e1982d4c6164318cf9b4682dfcb335699d4027372edee921f8a
-  data.tar.gz: 6f61afc314e69019464cb4722063f02513a06f493e820661e2f12128af5d6658
+  metadata.gz: 3adae07a9c634b7075a87a47c810c0d4e3c90137ee85a08e01e09038baf990e1
+  data.tar.gz: 9f7ecf0e06ace02def29a7e710ea87d5ed5e2af82282a1dba935fa3628116533
 SHA512:
-  metadata.gz: a540f0ad04dc38d0155c156041e22269155aa2fcad3958ec9a874302d888af4f70afdb69cf88ecebd624457d2fa8b03bba59c9963e5772d86f679088186a59bb
-  data.tar.gz: '085f9c0f9a0e2a3d4487d63a0c3a59ab0f5757e7eace1c757abedd96faf7d9152894e9c41a5c37bafe985a64889d03d5f8de3be72960a0ed9247ced3a14adfed'
+  metadata.gz: 5104d4d97f12a70e7f2d492dee6f8f870c388fea8091bc26d60fd87efb826331479a5efa851de7cbeb278d48aad9368d12cbad7bbb675b9cd9a20baee50f087e
+  data.tar.gz: 3ff6caa7e64806a1f03f918b7a6692e25c7493aa6280c0f9edabc99dd937e8890d76f3e8e0f060584964b993791c990621dc3cd8c908f91b5e81a25ac3e9f5ff

data/CHANGELOG.md

@@ -1,3 +1,13 @@
+# 2.0.0 (May 7th, 2020)
+
+Breaking Changes:
+
+* Standard error and output have been combined into one stream (out).  Early feedback indicated that reading both at the same time was preferable.
+
+Enhancements:
+
+* Added optional `timeout_in_seconds` argument to Pdi::Spoon#initialize.
+
 # 1.0.1 (February 19th, 2020)
 
 Fixes:

data/README.md

@@ -50,6 +50,7 @@ Notes:
 
 * You can also override the names of the scripts using the `kitchen` and `pan` constructor keyword arguments.  The defaults are `kitchen.sh` and `pan.sh`, respectively.
 * For other command line arguments that are not supported first-class in the Options objects below you can utilize the `args` argument when instantiating a `Spoon` instance.
+* Another optional argument is `timeout_in_seconds`.  If set it will ensure the sub-process runs within a given window.  If it times out the sub-process will be terminated and a Timeout::Error will be raised.
 
 ### Executing a Job/Transformation
 

data/lib/pdi.rb

@@ -8,8 +8,10 @@
 #
 
 require 'acts_as_hashable'
+require 'English'
 require 'forwardable'
 require 'open3'
+require 'timeout'
 
 require_relative 'pdi/executor'
 require_relative 'pdi/spoon'

data/lib/pdi/executor.rb

@@ -13,20 +13,42 @@ module Pdi
   # This class is the library's "metal" layer, the one which actually makes the system call and
   # interacts with the operating system (through Ruby's standard library.)
   class Executor
+    attr_reader :timeout_in_seconds
+
+    def initialize(timeout_in_seconds: nil)
+      @timeout_in_seconds = timeout_in_seconds
+
+      freeze
+    end
+
     def run(args)
       args = Array(args).map(&:to_s)
 
-      out, err, status = Open3.capture3(*args)
-
-      Result.new(
-        args: args,
-        status: {
-          code: status.exitstatus,
-          out: out,
-          err: err,
-          pid: status.pid
-        }
-      )
+      IO.popen(args, err: %i[child out]) do |io|
+        begin
+          io_read =
+            if timeout_in_seconds
+              Timeout.timeout(timeout_in_seconds) { io.read }
+            else
+              io.read
+            end
+
+          io.close
+          status = $CHILD_STATUS
+
+          Result.new(
+            args: args,
+            status: {
+              code: status.exitstatus,
+              out: io_read,
+              pid: status.pid
+            }
+          )
+        rescue Timeout::Error => e
+          Process.kill(9, io.pid)
+          raise e
+        end
+      end
     end
   end
 end

data/lib/pdi/executor/result.rb

@@ -18,7 +18,7 @@ module Pdi
 
       attr_reader :args, :status
 
-      def_delegators :status, :code, :out, :err, :pid
+      def_delegators :status, :code, :out, :pid
 
       def initialize(args:, status: {})
         @args   = args

data/lib/pdi/executor/status.rb

@@ -13,12 +13,11 @@ module Pdi
     class Status
       acts_as_hashable
 
-      attr_reader :code, :out, :err, :pid
+      attr_reader :code, :out, :pid
 
-      def initialize(code:, out: '', err: '', pid:)
+      def initialize(code:, out: '', pid:)
         @code = code
         @out  = out
-        @err  = err
         @pid  = pid
 
         freeze

data/lib/pdi/spoon.rb

@@ -30,7 +30,8 @@ module Pdi
       args: [],
       dir:,
       kitchen: DEFAULT_KITCHEN,
-      pan: DEFAULT_PAN
+      pan: DEFAULT_PAN,
+      timeout_in_seconds: nil
     )
       assert_required(:dir, dir)
       assert_required(:kitchen, kitchen)
@@ -40,7 +41,7 @@ module Pdi
       @dir            = File.expand_path(dir.to_s)
       @kitchen        = kitchen.to_s
       @pan            = pan.to_s
-      @executor       = Executor.new
+      @executor       = Executor.new(timeout_in_seconds: timeout_in_seconds)
       @parser         = Parser.new
 
       freeze

data/lib/pdi/version.rb

@@ -8,5 +8,5 @@
 #
 
 module Pdi
-  VERSION = '1.0.1'
+  VERSION = '2.0.0-alpha'
 end

data/spec/mocks/spoon/sleep.sh

@@ -0,0 +1,10 @@
+#!/bin/bash
+
+echo 'std_out'
+echo 'err_out' >&2
+
+sleep $1
+
+echo 'after_sleep'
+
+exit $2

data/spec/pdi/executor_spec.rb

@@ -0,0 +1,52 @@
+# frozen_string_literal: true
+
+#
+# Copyright (c) 2018-present, Blue Marble Payroll, LLC
+#
+# This source code is licensed under the MIT license found in the
+# LICENSE file in the root directory of this source tree.
+#
+
+require 'spec_helper'
+
+describe Pdi::Executor do
+  let(:script)      { File.join('spec', 'mocks', 'spoon', 'sleep.sh') }
+  let(:one_second)  { 1 }
+  let(:return_code) { 33 }
+
+  describe '#run' do
+    context 'with a timeout' do
+      # do not make these too high, bugs could cause the entire script to still be executed without
+      # killing it so these high values could draw out results.
+      let(:thirty_seconds) { 30 }
+
+      subject { described_class.new(timeout_in_seconds: one_second) }
+
+      # This will run a script that will take 30 seconds to process, but by limiting the
+      # timeout using the #run argument, it should raise an error after one second.
+      it 'times out and kills process after 5 seconds' do
+        args = [script, thirty_seconds]
+
+        expect { subject.run(args) }.to raise_error(Timeout::Error)
+      end
+    end
+
+    context 'without a timeout' do
+      it 'returns right exit status as code' do
+        args = [script, one_second, return_code]
+
+        result = subject.run(args)
+
+        expect(result.code).to eq(return_code)
+      end
+
+      it 'returns right standard output and error as out' do
+        args = [script, one_second, return_code]
+
+        result = subject.run(args)
+
+        expect(result.out).to eq("std_out\nerr_out\nafter_sleep\n")
+      end
+    end
+  end
+end

data/spec/pdi/spoon_spec.rb

@@ -14,6 +14,21 @@ describe Pdi::Spoon do
   let(:mocks_dir) { %w[spec mocks spoon] }
   let(:dir)       { File.join(*mocks_dir) }
 
+  describe '#initialize' do
+    it 'sets executor' do
+      timeout_in_seconds = 987
+
+      subject = described_class.new(dir: dir, timeout_in_seconds: timeout_in_seconds)
+
+      # Private/internal testing is not recommended, but I really wanted to ensure
+      # this class is properly configuring the Executor instance, that way I can rely
+      # mainly on the Executor unit tests instead of integration tests at this level.
+      executor = subject.send('executor')
+
+      expect(executor.timeout_in_seconds).to eq(timeout_in_seconds)
+    end
+  end
+
   describe '#run' do
     context 'transformations' do
       let(:options) do
@@ -43,8 +58,7 @@ describe Pdi::Spoon do
 
           result = subject.run(options)
 
-          expect(result.out).to  eq("output to stdout\n")
-          expect(result.err).to  eq("output to sterr\n")
+          expect(result.out).to  eq("output to stdout\noutput to sterr\n")
           expect(result.code).to eq(0)
         end
       end
@@ -97,8 +111,7 @@ describe Pdi::Spoon do
 
           result = subject.run(options)
 
-          expect(result.out).to  eq("output to stdout\n")
-          expect(result.err).to  eq("output to sterr\n")
+          expect(result.out).to  eq("output to stdout\noutput to sterr\n")
           expect(result.code).to eq(0)
         end
       end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: pdi
 version: !ruby/object:Gem::Version
-  version: 1.0.1
+  version: 2.0.0.pre.alpha
 platform: ruby
 authors:
 - Matthew Ruggio
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-02-20 00:00:00.000000000 Z
+date: 2020-05-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: acts_as_hashable
@@ -160,7 +160,9 @@ files:
 - lib/pdi/version.rb
 - pdi.gemspec
 - spec/mocks/spoon/return_code.sh
+- spec/mocks/spoon/sleep.sh
 - spec/mocks/spoon/version.sh
+- spec/pdi/executor_spec.rb
 - spec/pdi/spoon/kitchen_error_spec.rb
 - spec/pdi/spoon/options/arg_spec.rb
 - spec/pdi/spoon/options/param_spec.rb
@@ -183,9 +185,9 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: 2.3.8
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ">="
+  - - ">"
     - !ruby/object:Gem::Version
-      version: '0'
+      version: 1.3.1
 requirements: []
 rubygems_version: 3.0.3
 signing_key: 
@@ -193,7 +195,9 @@ specification_version: 4
 summary: Ruby wrapper for invoking Pentaho Data Integration
 test_files:
 - spec/mocks/spoon/return_code.sh
+- spec/mocks/spoon/sleep.sh
 - spec/mocks/spoon/version.sh
+- spec/pdi/executor_spec.rb
 - spec/pdi/spoon/kitchen_error_spec.rb
 - spec/pdi/spoon/options/arg_spec.rb
 - spec/pdi/spoon/options/param_spec.rb