filey-diff 1.1.1 → 1.1.2

data/changelog.md

@@ -2,6 +2,10 @@
 
 This project is [Semantically Versioned](http://semver.org).
 
+## 1.1.2
+
+* Co-operate with Ruby 2.0.0 automatic gzip decoding
+
 ## 1.1.1
 
 * Fix dotfiles on Ruby 2.0.0

data/filey-diff.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name              = 'filey-diff'
-  s.version           = '1.1.1'
+  s.version           = '1.1.2'
 
   s.summary     = "Compare two data sources that contain file-like objects"
   s.description =

data/lib/filey-diff/data-sources/aws_sdk_s3.rb

@@ -18,32 +18,56 @@ module Filey
           end
 
           if (s3_object.head[:content_encoding] == "gzip")
-            last_modified, md5 =
-              last_modified_and_md5_from_gzipped(s3_object, path)
+            last_modified, md5 = last_modified_and_md5_from_gzipped(
+              s3_object, path
+            )
           else
-            last_modified = s3_object.last_modified
-            md5 = s3_object.etag.gsub(/"/, '').split('-',2).first
+            last_modified, md5 = last_modified_and_md5(s3_object)
           end
 
           normalised_path = "./#{path}"
-          Filey.new(normalised_path,
-                    name,
-                    last_modified,
-                    md5)
+          Filey.new(
+            normalised_path,
+            name,
+            last_modified,
+            md5
+          )
         }
       end
 
-      def last_modified_and_md5_from_gzipped(s3_object, path)
-        tempfile = Tempfile.new(File.basename(path))
-        tempfile.write s3_object.read
-        tempfile.close
-
-        gz = Zlib::GzipReader.open(tempfile.path)
-        last_modified = gz.mtime
-        md5 = Digest::MD5.hexdigest(gz.read)
-        gz.close
+      def last_modified_and_md5(s3_object)
+        last_modified = s3_object.last_modified
+        md5 = s3_object.etag.gsub(/"/, '').split('-',2).first
         [last_modified, md5]
       end
+
+      def last_modified_and_md5_from_gzipped(s3_object, path)
+        s3_object_contents = s3_object.read
+        if is_already_decoded s3_object_contents
+          md5 = Digest::MD5.hexdigest(s3_object_contents)
+          [s3_object.last_modified, md5]
+        else
+          tempfile = Tempfile.new(File.basename(path))
+          tempfile.write s3_object_contents
+          tempfile.close
+
+          gz = Zlib::GzipReader.open(tempfile.path)
+          last_modified = gz.mtime
+          md5 = Digest::MD5.hexdigest(gz.read)
+          gz.close
+          [last_modified, md5]
+        end
+      end
+
+      # Check if the two first bytes are the magic numbers of the gzip format.
+      # We double-check here because Ruby 2.0.0 decodes gzip'ed HTTP responses
+      # automatically. As a result, we get decoded gzip data from the
+      # s3_object#read method when we are using Ruby 2.0.0, and encoded data
+      # when we are using previous versions of Ruby.
+      def is_already_decoded(gzipped_on_server)
+        is_gzipped = gzipped_on_server.bytes.to_a[0] == 0x1f && gzipped_on_server.bytes.to_a[1] == 0x8b
+        is_gzipped == false
+      end
     end
   end
 end

data/spec/data_sources_spec.rb

@@ -61,40 +61,68 @@ describe Filey::DataSources::AwsSdkS3 do
 
   it_should_behave_like "a data source", s3_bucket
 
-  it 'provides the original md5/mtime of a gzipped file' do
-    original_object = objects.first
-    tempfile = gzip_into_tmp_file(original_object)
-    data_source = s3_data_source_from_file(tempfile, original_object[:path])
-    filey = data_source.get_fileys[0]
-
-    filey.md5.should eq(Digest::MD5.hexdigest(original_object[:content]))
-    # GzipWriter seems to cut off fractions of a second,
-    # to_i adjusts the original file to match
-    filey.last_modified.to_i.should eq(original_object[:mtime].to_i)
-  end
+  context 'gzip' do
+    let(:gzip_tempfile_and_path) {
+      original_object = objects.first
+      [
+        gzip_into_tmp_file(original_object[:content], original_object[:mtime]),
+        original_object[:path]
+      ]
+    }
 
-  def s3_data_source_from_file(file, path)
-    file.open
-    data_source = Filey::DataSources::AwsSdkS3.new(S3Bucket.new([S3Object.new(
-      path,
-      file.mtime,
-      file.read,
-      { :content_encoding => 'gzip' }
-    )]))
-  end
+    let(:data_source_with_one_gzipped_object) {
+      file, path = gzip_tempfile_and_path
+      file.open
+      data_source = Filey::DataSources::AwsSdkS3.new(S3Bucket.new([S3Object.new(
+        path,
+        file.mtime,
+        file.read,
+        { :content_encoding => 'gzip' }
+      )]))
+    }
 
-  def gzip_into_tmp_file(s3_mock_object)
-    tempfile = Tempfile.new("temp")
+    it 'provides the original md5/mtime of a gzipped file' do
+      filey = data_source_with_one_gzipped_object.get_fileys[0]
+      filey.md5.should eq(Digest::MD5.hexdigest(objects.first[:content]))
+      # GzipWriter seems to cut off fractions of a second,
+      # to_i adjusts the original file to match
+      filey.last_modified.to_i.should eq(objects.first[:mtime].to_i)
+    end
 
-    gz = Zlib::GzipWriter.open(tempfile.path, Zlib::BEST_COMPRESSION, Zlib::DEFAULT_STRATEGY)
-    gz.mtime = s3_mock_object[:mtime]
-    gz.write s3_mock_object[:content]
+    context 'working with Ruby 2.0.0 automatic decoding of gzipped HTTP responses' do
+      let(:data_source_with_decoded_object_and_gzip_header) {
+        data_source = Filey::DataSources::AwsSdkS3.new(S3Bucket.new([S3Object.new(
+          objects.first[:path],
+          objects.first[:mtime],
+          objects.first[:content],
+          { :content_encoding => 'gzip' }
+        )]))
+      }
+
+      it 'detects the case where the gzipped data has already been decoded' do
+        filey = data_source_with_decoded_object_and_gzip_header.get_fileys.first
+        filey.last_modified.should eq(objects.first[:mtime])
+      end
+
+      it 'returns the md5 of the gzip-decoded content' do
+        filey = data_source_with_decoded_object_and_gzip_header.get_fileys.first
+        filey.md5.should eq(Digest::MD5.hexdigest(objects.first[:content]))
+      end
+    end
+
+    def gzip_into_tmp_file(content, mtime)
+      tempfile = Tempfile.new("temp")
 
-    gz.flush
-    tempfile.flush
+      gz = Zlib::GzipWriter.open(tempfile.path, Zlib::BEST_COMPRESSION, Zlib::DEFAULT_STRATEGY)
+      gz.mtime = mtime
+      gz.write content
 
-    gz.close
-    tempfile
+      gz.flush
+      tempfile.flush
+
+      gz.close
+      tempfile
+    end
   end
 end
 

metadata

@@ -1,7 +1,8 @@
 --- !ruby/object:Gem::Specification
 name: filey-diff
 version: !ruby/object:Gem::Version
-  version: 1.1.1
+  version: 1.1.2
+  prerelease: 
 platform: ruby
 authors:
 - Lauri Lehmijoki
@@ -13,6 +14,7 @@ dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
+    none: false
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
@@ -20,6 +22,7 @@ dependencies:
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
+    none: false
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
@@ -27,6 +30,7 @@ dependencies:
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
+    none: false
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
@@ -34,11 +38,12 @@ dependencies:
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
+    none: false
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
         version: '2.11'
-description: "\n    Find missing or outdated files.\n    For example, compare your
+description: ! "\n    Find missing or outdated files.\n    For example, compare your
   local file system to an AWS S3 bucket.\n    "
 email: lauri.lehmijoki@iki.fi
 executables: []
@@ -64,26 +69,33 @@ files:
 - spec/spec_helper.rb
 homepage: http://github.com/laurilehmijoki/filey-diff
 licenses: []
-metadata: {}
 post_install_message: 
 rdoc_options: []
 require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
+  none: false
   requirements:
-  - - '>='
+  - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
+      segments:
+      - 0
+      hash: 4447757573038540080
 required_rubygems_version: !ruby/object:Gem::Requirement
+  none: false
   requirements:
-  - - '>='
+  - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
+      segments:
+      - 0
+      hash: 4447757573038540080
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.0.3
+rubygems_version: 1.8.25
 signing_key: 
-specification_version: 4
+specification_version: 3
 summary: Compare two data sources that contain file-like objects
 test_files:
 - spec/comparison_spec.rb

checksums.yaml

@@ -1,7 +0,0 @@
----
-SHA1:
-  metadata.gz: 226c6e3d55d656ea58bb65e10021dcb561c9ad29
-  data.tar.gz: 5c0653b7916ecb2e1f27909512506f5b36d19d99
-SHA512:
-  metadata.gz: 97e2fc29de0145379df08595789fa77c25d7f60daf12f60a17700a8bbff4638a50d2d3325e231e9461efaedc93c2f4d5459c4334011effa4de8aa39f88372d0c
-  data.tar.gz: 2ef1d709e749ee2c7be3b9d6a8c1c8f462b35b8f7982d44905a5e160199d6d76c361f5b34c30e07df148c258da86d7f8518af03b24840761ccd7aa98ff29a5c6