rika 1.1.1-java → 1.2.0-java

data/lib/rika.rb

@@ -16,6 +16,7 @@ module Rika
   import org.apache.tika.metadata.Metadata
   import org.apache.tika.Tika
   import org.apache.tika.language.LanguageIdentifier
+  import org.apache.tika.detect.DefaultDetector
   import java.io.FileInputStream
   import java.net.URL
 
@@ -34,12 +35,11 @@ module Rika
     parser.metadata
   end
 
-
   class Parser
     
-    def initialize(file_location, max_content_length = -1)
+    def initialize(file_location, max_content_length = -1, detector = DefaultDetector.new)
       @uri = file_location
-      @tika = Tika.new
+      @tika = Tika.new(detector)
       @tika.set_max_string_length(max_content_length)
       @metadata_java = Metadata.new
       @metadata_ruby = nil
@@ -64,7 +64,11 @@ module Rika
     end
 
     def media_type
-      @media_type ||= @tika.detect(input_stream) 
+      if file?
+        @media_type ||= @tika.detect(java.io.File.new(@uri))
+      else
+        @media_type ||= @tika.detect(input_stream)
+      end
     end
 
     def available_metadata

data/lib/rika/version.rb

@@ -1,3 +1,3 @@
 module Rika
-  VERSION = "1.1.1"
+  VERSION = "1.2.0"
 end

data/spec/rika_spec.rb

@@ -9,6 +9,7 @@ describe Rika::Parser do
   before(:all) do
     @txt_parser = Rika::Parser.new(file_path("text_file.txt"))
     @docx_parser = Rika::Parser.new(file_path("document.docx"))
+    @doc_parser = Rika::Parser.new(file_path("document.doc"))
     @pdf_parser = Rika::Parser.new(file_path("document.pdf"))
     @image_parser = Rika::Parser.new(file_path("image.jpg"))
     @unknown_parser = Rika::Parser.new(file_path("unknown.bin"))
@@ -151,6 +152,14 @@ describe Rika::Parser do
     it "should return application/octet-stream for unknown file" do
       @unknown_parser.media_type.should == "application/octet-stream"
     end
+
+    it "should return msword for a doc file" do
+      @doc_parser.media_type.should == "application/msword"
+    end
+
+    it "should return wordprocessingml for a docx file" do
+      @docx_parser.media_type.should == "application/vnd.openxmlformats-officedocument.wordprocessingml.document"
+    end
   end
 
   describe '#language' do

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: rika
 version: !ruby/object:Gem::Version
-  version: 1.1.1
+  version: 1.2.0
   prerelease:
 platform: java
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2013-02-26 00:00:00.000000000 Z
+date: 2013-04-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rspec