active_encode 1.1.2 → 1.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: aae1d452186e5101e0e42b4211fd25da561b068caee41d8839f019067309048a
-  data.tar.gz: d54cc90a08b0907885281ada0ad0c8a62aae0806213b36010282e279f85803bc
+  metadata.gz: e8c41200059443abc28d64ef9748086b36a4a1c97537138da24bd1b5016e902d
+  data.tar.gz: 3fec8ac243d8e7ff3c96abaa26b08c23e5c409337c2f63463f465096e29e7cba
 SHA512:
-  metadata.gz: c1080a31e991e8784dcbf9a8b1f7740fedf299a3eb5075d3ff9f665589fe0f73560679f592f3e97925a9424277a6e12a845187501858d520e2564f3e2a078302
-  data.tar.gz: 77f94da2c42eeaff56e9298138b1c04a5a79a7fc7b256bd6fa65ba691cb25b3b216dd1cfbad414ad22f3180aceb934a07518a0a2fb062ce63dc8c6f37a3ab6e1
+  metadata.gz: 5cae3d56d8c068e3cb73e01b4d15d7ba99d305f6c7bd09e7de51def230d247fa3b423d8b32674cadbf66ad7845ce235399b6b53fa068ef305b65bfad1c829f85
+  data.tar.gz: 7265200be814de1969887dd1275a8e89da3e0befe6a92542800c42119f93307f0bf7584a70dcfe8127ef11c7a7a76f5f14fab7b9a5f47ef9408cb474f64d2039

data/.rubocop_todo.yml

@@ -37,6 +37,7 @@ Layout/LineLength:
 Metrics/MethodLength:
   Exclude:
     - 'lib/active_encode/engine_adapters/*'
+    - 'lib/active_encode/engine_adapters/ffmpeg_adapter/*'
     - 'lib/file_locator.rb'
 
 Metrics/PerceivedComplexity:

data/lib/active_encode/engine_adapters/ffmpeg_adapter/cleaner.rb

@@ -0,0 +1,71 @@
+# frozen_string_literal: true
+module ActiveEncode
+  module EngineAdapters
+    class FfmpegAdapter
+      module Cleaner
+        # This method is to clean up files leftover from the ffmpeg encode process.
+        # File names for the pass_through adapter are the same, so this will clean up
+        # pass_through encodes as well.
+        def remove_old_files!(options = {})
+          default_options = {
+            older_than: 2.weeks,
+            no_outputs: ['input_metadata', 'duration_input_metadata', 'error.log', 'exit_status.code', 'progress', 'completed', 'pid', 'output_metadata-*'],
+            outputs: false,
+            all: false
+          }
+          options.reverse_merge!(default_options)
+
+          if options[:all]
+            files = build_file_list(WORK_DIR, "*")
+            directories = remove_files(files, options[:older_than])
+            remove_empty_directories(directories)
+          elsif options[:outputs]
+            output_directories = build_file_list(WORK_DIR, "outputs")
+            remove_child_files(output_directories, options[:older_than])
+            remove_empty_directories(output_directories)
+          else
+            files = []
+            options[:no_outputs].each { |fn| files += build_file_list(WORK_DIR, fn) }
+            remove_files(files, options[:older_than])
+          end
+        end
+
+        def build_file_list(directory, filename)
+          file_path = File.join(directory, "**", filename)
+          # Some of the files generated by the ffmpeg encode seem to be hidden files.
+          # This uses File::FNM_DOTMATCH to include them in the results.
+          Dir.glob(file_path, File::FNM_DOTMATCH)
+        end
+
+        def file_check(path, older_than)
+          File.mtime(path) < DateTime.now - older_than && File.file?(path)
+        end
+
+        def remove_files(files, older_than)
+          files_to_delete = files.select { |f| file_check(f, older_than) }
+          FileUtils.rm(files_to_delete) unless files_to_delete.empty?
+
+          # Return a list of any directories that were included in the files list for further processing.
+          # The files list can include directories such as "/tmp/.." which should not be included in directory list.
+          files.select { |f| File.directory?(f) unless f.end_with?(".") }
+        end
+
+        def remove_empty_directories(directories)
+          directories_to_delete = directories.select { |d| Dir.empty?(d) }
+          non_empty_directories = directories - directories_to_delete
+          directories_to_delete += non_empty_directories.select { |ned| Dir.children(ned) == ["outputs"] && directories_to_delete.include?(File.join(ned, "outputs")) }
+          FileUtils.rmdir(directories_to_delete) unless directories_to_delete.empty?
+        end
+
+        def remove_child_files(directories, older_than)
+          files_to_delete = []
+          directories.each do |d|
+            files = Dir.children(d).select { |ch| file_check(File.join(d, ch), older_than) }
+            files_to_delete += files.collect { |f| File.join(d, f) }
+          end
+          FileUtils.rm(files_to_delete) unless files_to_delete.empty?
+        end
+      end
+    end
+  end
+end

data/lib/active_encode/engine_adapters/ffmpeg_adapter.rb

@@ -1,12 +1,16 @@
 # frozen_string_literal: true
+require 'English'
 require 'fileutils'
 require 'nokogiri'
 require 'shellwords'
 require 'addressable/uri'
+require 'active_encode/engine_adapters/ffmpeg_adapter/cleaner'
 
 module ActiveEncode
   module EngineAdapters
     class FfmpegAdapter
+      extend ActiveEncode::EngineAdapters::FfmpegAdapter::Cleaner
+
       WORK_DIR = ENV["ENCODE_WORK_DIR"] || "encodes" # Should read from config
       MEDIAINFO_PATH = ENV["MEDIAINFO_PATH"] || "mediainfo"
       FFMPEG_PATH = ENV["FFMPEG_PATH"] || "ffmpeg"
@@ -51,18 +55,38 @@ module ActiveEncode
 
         new_encode.input = build_input new_encode
 
-        if new_encode.input.duration.blank?
-          new_encode.state = :failed
-          new_encode.percent_complete = 1
+        # Log error if file is empty or inaccessible
+        if new_encode.input.duration.blank? && new_encode.input.file_size.blank?
+          file_error(new_encode, input_url)
+          return new_encode
+        # If file is not empty, try copying file to generate missing metadata
+        elsif new_encode.input.duration.blank? && new_encode.input.file_size.present?
+
+          # This regex removes the query string from URIs. This is necessary to
+          # properly process files originating from S3 or similar providers.
+          filepath = clean_url.to_s.gsub(/\?.*/, '')
+          copy_url = filepath.gsub(filepath, "#{File.basename(filepath, File.extname(filepath))}_temp#{File.extname(filepath)}")
+          copy_path = working_path(copy_url, new_encode.id)
+
+          # -map 0 sets ffmpeg to copy all available streams.
+          # -c copy sets ffmpeg to copy all codecs
+          # -y automatically overwrites the temp file if one already exists
+          `#{FFMPEG_PATH} -loglevel level+fatal -i \"#{input_url}\" -map 0 -c copy -y \"#{copy_path}\"`
+
+          # If ffmpeg copy fails, log error because file is either not a media file
+          # or the file extension does not match the codecs used to encode the file
+          unless $CHILD_STATUS.success?
+            file_error(new_encode, input_url)
+            return new_encode
+          end
 
-          new_encode.errors = if new_encode.input.file_size.blank?
-                                ["#{input_url} does not exist or is not accessible"]
-                              else
-                                ["Error inspecting input: #{input_url}"]
-                              end
+          # Write the mediainfo output to a separate file to preserve metadata from original file
+          `#{MEDIAINFO_PATH} #{curl_option} --Output=XML --LogFile=#{working_path("duration_input_metadata", new_encode.id)} "#{copy_path}"`
 
-          write_errors new_encode
-          return new_encode
+          File.delete(copy_path)
+
+          # Assign duration to the encode created for the original file.
+          new_encode.input.duration = fixed_duration(working_path("duration_input_metadata", new_encode.id))
         end
 
         new_encode.state = :running
@@ -99,6 +123,7 @@ module ActiveEncode
         encode.output = []
         encode.created_at, encode.updated_at = get_times encode.id
         encode.input = build_input encode
+        encode.input.duration ||= fixed_duration(working_path("duration_input_metadata", encode.id)) if File.exist?(working_path("duration_input_metadata", encode.id))
         encode.percent_complete = calculate_percent_complete encode
 
         pid = get_pid(id)
@@ -302,6 +327,24 @@ module ActiveEncode
       def get_xpath_text(doc, xpath, cast_method)
         doc.xpath(xpath).first&.text&.send(cast_method)
       end
+
+      def fixed_duration(path)
+        get_tech_metadata(path)[:duration]
+      end
+
+      def file_error(new_encode, input_url)
+        new_encode.state = :failed
+        new_encode.percent_complete = 1
+
+        new_encode.errors = if new_encode.input.file_size.blank?
+                              ["#{input_url} does not exist or is not accessible"]
+                            else
+                              ["Error inspecting input: #{input_url}"]
+                            end
+
+        write_errors new_encode
+        new_encode
+      end
     end
   end
 end

data/lib/active_encode/engine_adapters/pass_through_adapter.rb

@@ -15,6 +15,7 @@ module ActiveEncode
     class PassThroughAdapter
       WORK_DIR = ENV["ENCODE_WORK_DIR"] || "encodes" # Should read from config
       MEDIAINFO_PATH = ENV["MEDIAINFO_PATH"] || "mediainfo"
+      FFMPEG_PATH = ENV["FFMPEG_PATH"] || "ffmpeg"
 
       def create(input_url, options = {})
         # Decode file uris for ffmpeg (mediainfo works either way)
@@ -36,18 +37,38 @@ module ActiveEncode
         new_encode.input.id = new_encode.id
         new_encode.created_at, new_encode.updated_at = get_times new_encode.id
 
-        if new_encode.input.duration.blank?
-          new_encode.state = :failed
-          new_encode.percent_complete = 1
-
-          new_encode.errors = if new_encode.input.file_size.blank?
-                                ["#{input_url} does not exist or is not accessible"]
-                              else
-                                ["Error inspecting input: #{input_url}"]
-                              end
-
-          write_errors new_encode
+        # Log error if file is empty or inaccessible
+        if new_encode.input.duration.blank? && new_encode.input.file_size.blank?
+          file_error(new_encode, input_url)
           return new_encode
+        # If file is not empty, try copying file to generate missing metadata
+        elsif new_encode.input.duration.blank? && new_encode.input.file_size.present?
+
+          # This regex removes the query string from URIs. This is necessary to
+          # properly process files originating from S3 or similar providers.
+          filepath = input_url.to_s.gsub(/\?.*/, '')
+          copy_url = filepath.gsub(filepath, "#{File.basename(filepath, File.extname(filepath))}_temp#{File.extname(input_url)}")
+          copy_path = working_path(copy_url, new_encode.id)
+
+          # -map 0 sets ffmpeg to copy all available streams.
+          # -c copy sets ffmpeg to copy all codecs
+          # -y automatically overwrites the temp file if one already exists
+          `#{FFMPEG_PATH} -loglevel level+fatal -i \"#{input_url}\" -map 0 -c copy -y \"#{copy_path}\"`
+
+          # If ffmpeg copy fails, log error because file is either not a media file
+          # or the file extension does not match the codecs used to encode the file
+          unless $CHILD_STATUS.success?
+            file_error(new_encode, input_url)
+            return new_encode
+          end
+
+          # Write the mediainfo output to a separate file to preserve metadata from original file
+          `#{MEDIAINFO_PATH} --Output=XML --LogFile=#{working_path("duration_input_metadata", new_encode.id)} \"#{copy_path}\"`
+
+          File.delete(copy_path)
+
+          # Assign duration to the encode created for the original file.
+          new_encode.input.duration = fixed_duration(working_path("duration_input_metadata", new_encode.id))
         end
 
         # For saving filename to label map used to find the label when building outputs
@@ -86,6 +107,7 @@ module ActiveEncode
         encode.created_at, encode.updated_at = get_times encode.id
         encode.input = build_input encode
         encode.input.id = encode.id
+        encode.input.duration ||= fixed_duration(working_path("duration_input_metadata", encode.id)) if File.exist?(working_path("duration_input_metadata", encode.id))
         encode.output = []
         encode.current_operations = []
 
@@ -231,6 +253,24 @@ module ActiveEncode
       def get_xpath_text(doc, xpath, cast_method)
         doc.xpath(xpath).first&.text&.send(cast_method)
       end
+
+      def fixed_duration(path)
+        get_tech_metadata(path)[:duration]
+      end
+
+      def file_error(new_encode, input_url)
+        new_encode.state = :failed
+        new_encode.percent_complete = 1
+
+        new_encode.errors = if new_encode.input.file_size.blank?
+                              ["#{input_url} does not exist or is not accessible"]
+                            else
+                              ["Error inspecting input: #{input_url}"]
+                            end
+
+        write_errors new_encode
+        new_encode
+      end
     end
   end
 end

data/lib/active_encode/filename_sanitizer.rb

@@ -24,7 +24,9 @@ module ActiveEncode
       when /^file\:\/\/\//
         input_url.to_s.gsub(/file:\/\//, '')
       when /^s3\:\/\//
-        input_url.to_s.gsub(/#{input_url.normalized_site}/, '')
+        input_url.to_s.gsub(/#{Addressable::URI.parse(input_url).normalized_site}/, '')
+      when /^https?:\/\//
+        input_url
       end
     end
   end

data/lib/active_encode/version.rb

@@ -1,4 +1,4 @@
 # frozen_string_literal: true
 module ActiveEncode
-  VERSION = '1.1.2'
+  VERSION = '1.2.0'
 end

data/spec/integration/ffmpeg_adapter_spec.rb

@@ -112,6 +112,62 @@ describe ActiveEncode::EngineAdapters::FfmpegAdapter do
       end
     end
 
+    context "input file format does not match extension" do
+      let(:improper_format_file) { "file://" + Rails.root.join('..', 'spec', 'fixtures', 'file_without_metadata.mp4').to_s }
+      let(:improper_format_job) { ActiveEncode::Base.create(improper_format_file, outputs: [{ label: "low", ffmpeg_opt: "-s 640x480", extension: 'mp4' }]) }
+
+      it "returns the encode with correct error" do
+        expect(improper_format_job.errors).to include("Error inspecting input: #{improper_format_file}")
+        expect(improper_format_job.percent_complete).to be 1
+      end
+    end
+
+    context "input file with missing metadata" do
+      let(:file_without_metadata) { "file://" + Rails.root.join('..', 'spec', 'fixtures', 'file_without_metadata.webm').to_s }
+      let!(:create_without_metadata_job) { ActiveEncode::Base.create(file_without_metadata, outputs: [{ label: "low", ffmpeg_opt: "-s 640x480", extension: 'mp4' }]) }
+      let(:find_without_metadata_job) { ActiveEncode::Base.find create_without_metadata_job.id }
+
+      it "does not have errors" do
+        sleep 2
+        expect(find_without_metadata_job.errors).to be_empty
+      end
+
+      it "has the input technical metadata in a file" do
+        expect(File.read("#{work_dir}/#{create_without_metadata_job.id}/input_metadata")).not_to be_empty
+      end
+
+      it "has the pid in a file" do
+        expect(File.read("#{work_dir}/#{create_without_metadata_job.id}/pid")).not_to be_empty
+      end
+
+      it "assigns the correct duration to the encode" do
+        expect(create_without_metadata_job.input.duration).to eq 68_653
+        expect(find_without_metadata_job.input.duration).to eq 68_653
+      end
+
+      context 'when uri encoded' do
+        let(:file_without_metadata) { Addressable::URI.encode("file://" + Rails.root.join('..', 'spec', 'fixtures', 'file_without_metadata.webm').to_s) }
+
+        it "does not have errors" do
+          sleep 2
+          expect(find_without_metadata_job.errors).to be_empty
+        end
+
+        it "has the input technical metadata in a file" do
+          expect(File.read("#{work_dir}/#{create_without_metadata_job.id}/input_metadata")).not_to be_empty
+        end
+
+        it "has the pid in a file" do
+          expect(File.read("#{work_dir}/#{create_without_metadata_job.id}/pid")).not_to be_empty
+        end
+
+        it "assigns the correct duration to the encode" do
+          expect(create_without_metadata_job.input.duration).to eq 68_653
+          expect(find_without_metadata_job.input.duration).to eq 68_653
+        end
+      end
+    end
+
     context "input filename with spaces" do
       let(:file_with_space) { "file://" + Rails.root.join('..', 'spec', 'fixtures', 'file with space.mp4').to_s }
       let!(:create_space_job) { ActiveEncode::Base.create(file_with_space, outputs: [{ label: "low", ffmpeg_opt: "-s 640x480", extension: 'mp4' }]) }
@@ -372,4 +428,108 @@ describe ActiveEncode::EngineAdapters::FfmpegAdapter do
       expect { running_job.cancel! }.to raise_error(ActiveEncode::CancelError)
     end
   end
+
+  describe "#remove_old_files!" do
+    subject { created_job }
+    # 'exit_status.code' and 'progress' seem to be hidden files so rspec does not see them.
+    # That is why they are not explicitly included in the tests even though they are in the filenames list.
+    # If they were not being deleted they would cause other tests to fail.
+    let(:base_path) { "#{work_dir}/#{subject.id}" }
+    let(:input_metadata_file) { "#{base_path}/input_metadata" }
+    let(:error_log_file) { "#{base_path}/error.log" }
+    let(:pid_file) { "#{base_path}/pid" }
+    let(:exit_status_file) { "#{base_path}/exit_status.code" }
+    let(:progress_file) { "#{base_path}/progress" }
+    let(:pathnames) { [input_metadata_file, error_log_file, pid_file, exit_status_file, progress_file] }
+
+    # There was some flaky behavior with the file creation for created_job that
+    # would cause tests to fail. This ensures the files are created.
+    before :each do
+      FileUtils.touch(pathnames)
+    end
+
+    context ":no_outputs" do
+      it "deletes files created from encode process older than 2 weeks" do
+        # Another measure to give files time to be created.
+        sleep 1
+        travel 3.weeks do
+          expect { described_class.remove_old_files! }
+            .to change { File.exist?(input_metadata_file) }.from(true).to(false)
+            .and change { File.exist?(error_log_file) }.from(true).to(false)
+            .and change { File.exist?(pid_file) }.from(true).to(false)
+            .and not_change { Dir.children("#{work_dir}/#{subject.id}/outputs").count }.from(2)
+        end
+      end
+
+      it "does not delete files younger than 2 weeks" do
+        sleep 1
+        expect { described_class.remove_old_files! }
+          .to not_change { File.exist?(input_metadata_file) }.from(true)
+          .and not_change { File.exist?(error_log_file) }.from(true)
+          .and not_change { File.exist?(pid_file) }.from(true)
+          .and not_change { Dir.children("#{work_dir}/#{subject.id}/outputs").count }.from(2)
+      end
+    end
+
+    context ":outputs" do
+      it "deletes outputs created from encode process older than 2 weeks" do
+        sleep 1
+        travel 3.weeks do
+          expect { described_class.remove_old_files!(outputs: true) }
+            .to not_change { File.exist?(input_metadata_file) }.from(true)
+            .and not_change { File.exist?(error_log_file) }.from(true)
+            .and not_change { File.exist?(pid_file) }.from(true)
+            .and change { Dir.exist?("#{work_dir}/#{subject.id}/outputs") }.from(true).to(false)
+        end
+      end
+
+      it "does not delete outputs younger than 2 weeks" do
+        sleep 1
+        expect { described_class.remove_old_files!(outputs: true) }
+          .to not_change { File.exist?(input_metadata_file) }.from(true)
+          .and not_change { File.exist?(error_log_file) }.from(true)
+          .and not_change { File.exist?(pid_file) }.from(true)
+          .and not_change { Dir.children("#{work_dir}/#{subject.id}/outputs").count }.from(2)
+      end
+
+      it "does not delete outputs directory containing files younger than 2 weeks" do
+        sleep 1
+        travel 3.weeks do
+          allow(File).to receive(:mtime).and_call_original
+          allow(File).to receive(:mtime).with("#{work_dir}/#{subject.id}/outputs/fireworks-low.mp4").and_return(DateTime.now)
+          expect { described_class.remove_old_files!(outputs: true) }
+            .to not_change { Dir.exist?("#{work_dir}/#{subject.id}/outputs") }.from(true)
+          expect(Dir.children("#{work_dir}/#{subject.id}/outputs")).to eq(["fireworks-low.mp4"])
+        end
+      end
+    end
+
+    context ":all" do
+      it "deletes all files and directories older than 2 weeks" do
+        sleep 1
+        travel 3.weeks do
+          expect { described_class.remove_old_files!(all: true) }
+            .to change { Dir.exist?("#{work_dir}/#{subject.id}") }.from(true).to(false)
+        end
+      end
+
+      it "does not delete files and directories younger than 2 weeks" do
+        sleep 1
+        expect { described_class.remove_old_files!(all: true) }
+          .to not_change { Dir.exist?("#{work_dir}/#{subject.id}") }.from(true)
+          .and not_change { Dir.children("#{work_dir}/#{subject.id}").count }
+      end
+
+      it "does not delete directories containing files younger than 2 weeks" do
+        sleep 1
+        travel 3.weeks do
+          allow(File).to receive(:mtime).and_call_original
+          allow(File).to receive(:mtime).with("#{work_dir}/#{subject.id}/input_metadata").and_return(DateTime.now)
+          expect { described_class.remove_old_files!(all: true) }
+            .to not_change { Dir.exist?("#{work_dir}/#{subject.id}") }.from(true)
+          expect(Dir.children("#{work_dir}/#{subject.id}")).to eq(["input_metadata"])
+        end
+      end
+    end
+  end
 end

data/spec/integration/pass_through_adapter_spec.rb

@@ -109,6 +109,54 @@ describe ActiveEncode::EngineAdapters::PassThroughAdapter do
       end
     end
 
+    context "input file format does not match extension" do
+      let(:improper_format_file) { "file://" + Rails.root.join('..', 'spec', 'fixtures', 'file_without_metadata.mp4').to_s }
+      let(:improper_format_job) { ActiveEncode::Base.create(improper_format_file, outputs: [{ label: "low", url: improper_format_file }]) }
+
+      it "returns the encode with correct error" do
+        expect(improper_format_job.errors).to include("Error inspecting input: #{improper_format_file}")
+        expect(improper_format_job.percent_complete).to be 1
+      end
+    end
+
+    context "input file without metadata" do
+      let(:file_without_metadata) { "file://" + Rails.root.join('..', 'spec', 'fixtures', 'file_without_metadata.webm').to_s }
+      let(:file_without_metadata_derivative) { "file://" + Rails.root.join('..', 'spec', 'fixtures', 'file_without_metadata.low.webm').to_s }
+      let(:create_without_metadata_job) { ActiveEncode::Base.create(file_without_metadata, outputs: [{ label: "low", url: file_without_metadata_derivative }]) }
+      let(:find_without_metadata_job) { ActiveEncode::Base.find create_without_metadata_job.id }
+
+      it "does not have errors" do
+        expect(find_without_metadata_job.errors).to be_empty
+      end
+
+      it "has the input technical metadata in a file" do
+        expect(File.read("#{work_dir}/#{create_without_metadata_job.id}/input_metadata")).not_to be_empty
+      end
+
+      it "assigns the correct duration to the encode" do
+        expect(create_without_metadata_job.input.duration).to eq 68_653
+        expect(find_without_metadata_job.input.duration).to eq 68_653
+      end
+
+      context 'when uri encoded' do
+        let(:file_without_metadata) { Addressable::URI.encode("file://" + Rails.root.join('..', 'spec', 'fixtures', 'file_without_metadata.webm').to_s) }
+        let(:file_without_metadata_derivative) { Addressable::URI.encode("file://" + Rails.root.join('..', 'spec', 'fixtures', 'file_without_metadata.low.webm').to_s) }
+
+        it "does not have errors" do
+          expect(find_without_metadata_job.errors).to be_empty
+        end
+
+        it "has the input technical metadata in a file" do
+          expect(File.read("#{work_dir}/#{create_without_metadata_job.id}/input_metadata")).not_to be_empty
+        end
+
+        it "assigns the correct duration to the encode" do
+          expect(create_without_metadata_job.input.duration).to eq 68_653
+          expect(find_without_metadata_job.input.duration).to eq 68_653
+        end
+      end
+    end
+
     context "input filename with spaces" do
       let(:file_with_space) { "file://" + Rails.root.join('..', 'spec', 'fixtures', 'file with space.mp4').to_s }
       let(:file_with_space_derivative) { "file://" + Rails.root.join('..', 'spec', 'fixtures', 'file with space.low.mp4').to_s }

data/spec/rails_helper.rb

@@ -20,4 +20,6 @@ RSpec.configure do |config|
   config.after do |example|
     DatabaseCleaner.clean if example.metadata[:db_clean]
   end
+
+  config.include ActiveSupport::Testing::TimeHelpers
 end

data/spec/spec_helper.rb

@@ -16,3 +16,5 @@ RSpec::Matchers.define :be_the_same_time_as do |expected|
     expect(Time.parse(expected).utc).to eq(Time.parse(actual).utc)
   end
 end
+
+RSpec::Matchers.define_negated_matcher :not_change, :change

data/spec/units/filename_sanitizer_spec.rb

@@ -0,0 +1,20 @@
+# frozen_string_literal: true
+require 'spec_helper'
+require 'addressable'
+
+describe ActiveEncode::FilenameSanitizer do
+  describe '#sanitize_uri' do
+    it 'removes file scheme' do
+      uri = "file:///path/to/file"
+      expect(ActiveEncode.sanitize_uri(uri)).to eq "/path/to/file"
+    end
+    it 'relativizes s3 uris' do
+      uri = "s3://mybucket/guitar.mp4"
+      expect(ActiveEncode.sanitize_uri(uri)).to eq "/guitar.mp4"
+    end
+    it 'does nothing for http(s) uris' do
+      uri = "https://www.googleapis.com/drive/v3/files/1WkWJ12WecI9hX-PmEbuKDGLPK_mN3kYP?alt=media"
+      expect(ActiveEncode.sanitize_uri(uri)).to eq uri
+    end
+  end
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: active_encode
 version: !ruby/object:Gem::Version
-  version: 1.1.2
+  version: 1.2.0
 platform: ruby
 authors:
 - Michael Klein, Chris Colvard, Phuong Dinh
-autorequire:
+autorequire: 
 bindir: bin
 cert_chain: []
-date: 2023-01-30 00:00:00.000000000 Z
+date: 2023-05-10 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rails
@@ -278,6 +278,7 @@ files:
 - lib/active_encode/engine_adapters.rb
 - lib/active_encode/engine_adapters/elastic_transcoder_adapter.rb
 - lib/active_encode/engine_adapters/ffmpeg_adapter.rb
+- lib/active_encode/engine_adapters/ffmpeg_adapter/cleaner.rb
 - lib/active_encode/engine_adapters/matterhorn_adapter.rb
 - lib/active_encode/engine_adapters/media_convert_adapter.rb
 - lib/active_encode/engine_adapters/media_convert_output.rb
@@ -364,6 +365,9 @@ files:
 - spec/fixtures/file with space.mp4
 - spec/fixtures/file.with :=+%sp3c!l-ch4cts().mp4
 - spec/fixtures/file.with...periods.mp4
+- spec/fixtures/file_without_metadata.low.webm
+- spec/fixtures/file_without_metadata.mp4
+- spec/fixtures/file_without_metadata.webm
 - spec/fixtures/fireworks.low.mp4
 - spec/fixtures/fireworks.mp4
 - spec/fixtures/matterhorn/cancelled_response.xml
@@ -421,6 +425,7 @@ files:
 - spec/units/core_spec.rb
 - spec/units/engine_adapter_spec.rb
 - spec/units/file_locator_spec.rb
+- spec/units/filename_sanitizer_spec.rb
 - spec/units/global_id_spec.rb
 - spec/units/input_spec.rb
 - spec/units/output_spec.rb
@@ -433,7 +438,7 @@ licenses:
 - Apache-2.0
 metadata:
   rubygems_mfa_required: 'true'
-post_install_message:
+post_install_message: 
 rdoc_options: []
 require_paths:
 - lib
@@ -448,8 +453,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.2.32
-signing_key:
+rubygems_version: 3.1.6
+signing_key: 
 specification_version: 4
 summary: Declare encode job classes that can be run by a variety of encoding services
 test_files:
@@ -520,6 +525,9 @@ test_files:
 - spec/fixtures/file with space.mp4
 - spec/fixtures/file.with :=+%sp3c!l-ch4cts().mp4
 - spec/fixtures/file.with...periods.mp4
+- spec/fixtures/file_without_metadata.low.webm
+- spec/fixtures/file_without_metadata.mp4
+- spec/fixtures/file_without_metadata.webm
 - spec/fixtures/fireworks.low.mp4
 - spec/fixtures/fireworks.mp4
 - spec/fixtures/matterhorn/cancelled_response.xml
@@ -577,6 +585,7 @@ test_files:
 - spec/units/core_spec.rb
 - spec/units/engine_adapter_spec.rb
 - spec/units/file_locator_spec.rb
+- spec/units/filename_sanitizer_spec.rb
 - spec/units/global_id_spec.rb
 - spec/units/input_spec.rb
 - spec/units/output_spec.rb