hydra-file_characterization 1.0.0 → 1.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: fc80d9f0fda674639f974b581b57939e1b7cd9b32a992b7119f90ea017c883ad
-  data.tar.gz: a4e3af95b921b20d395b3405f29b3e2bfcc3a15882265f8989338f1fcf9594ae
+  metadata.gz: 97c20a68f011d01afddf0a42c3fa2802ab17b972c1ca0044910dfce652750f75
+  data.tar.gz: 147fd5b69e6af4de7abf5f768996c2f9442624a78a620b6d9a110b3d0b68b2d1
 SHA512:
-  metadata.gz: a6fd8b4b84d9a5fa5e1a9d28c6aba85611690656a7aaee7205ee0cdf30e3f797ebdfaa8952d4a226efdd15ff608f2685bf12e6133d05b3a7f6ed491f1cedad5e
-  data.tar.gz: 243e3a2d57e41f3e2cb9a71004c4a2ccc8f136115d2008390d3a66570a4c1f0e1dec4ebfb8c490f702cb99ee3b8d25d0dd7ec8febad60463d57b5783152a0f51
+  metadata.gz: 5ff697b557ec3abdbfef62f3f45fb646da5c2b17b9aa32f26be6584798d1bf02e8f8401602a5c00a312b9a4da40fc6bffc674379eada7d82284d838963a2d815
+  data.tar.gz: 5396ae39437c6fc528cccbcc3dacff23c5759fee17740b9c3382ca16d769ecd0ef1c4bf8b1ee1b45e28d49ccac93c8ea06c77ebfc11d32390d9a4a45524b009d

data/lib/hydra/file_characterization/characterizers.rb

@@ -29,3 +29,4 @@ end
 
 require 'hydra/file_characterization/characterizers/fits'
 require 'hydra/file_characterization/characterizers/ffprobe'
+require 'hydra/file_characterization/characterizers/fits_servlet'

data/lib/hydra/file_characterization/characterizers/fits_servlet.rb

@@ -0,0 +1,23 @@
+require 'hydra/file_characterization/exceptions'
+require 'hydra/file_characterization/characterizer'
+require 'logger'
+module Hydra::FileCharacterization::Characterizers
+  class FitsServlet < Hydra::FileCharacterization::Characterizer
+
+    protected
+
+      def command
+        "curl -k -F datafile=@#{filename} #{ENV['FITS_SERVLET_URL']}/examine"
+      end
+
+      # Remove any non-XML output that precedes the <?xml> tag
+      # See: https://github.com/harvard-lts/fits/issues/20
+      #      https://github.com/harvard-lts/fits/issues/40
+      #      https://github.com/harvard-lts/fits/issues/46
+      def post_process(raw_output)
+        md = /\A(.*)(<\?xml.*)\Z/m.match(raw_output)
+        logger.warn "FITS produced non-xml output: \"#{md[1].chomp}\"" unless md[1].empty?
+        md[2]
+      end
+  end
+end

data/lib/hydra/file_characterization/version.rb

@@ -1,5 +1,5 @@
 module Hydra
   module FileCharacterization
-    VERSION = "1.0.0"
+    VERSION = "1.1.0"
   end
 end

data/spec/lib/hydra/file_characterization/characterizers/fit_servlet_spec.rb

@@ -0,0 +1,68 @@
+require 'spec_helper'
+require 'hydra/file_characterization/characterizers/fits_servlet'
+
+module Hydra::FileCharacterization::Characterizers
+  describe FitsServlet do
+    let(:fits) { Fits.new(filename) }
+
+    describe "#call", unless: ENV['TRAVIS'] do
+      subject { fits.call }
+
+      context 'validfile' do
+        let(:filename) { fixture_file('brendan_behan.jpeg') }
+        it { is_expected.to include(%(<identity format="JPEG File Interchange Format" mimetype="image/jpeg")) }
+      end
+
+      context 'invalidFile' do
+        let(:filename) { fixture_file('nofile.pdf') }
+        it "raises an error" do
+          expect { subject }.to raise_error(Hydra::FileCharacterization::FileNotFoundError)
+        end
+      end
+
+      context 'corruptFile' do
+        let(:filename) { fixture_file('brendan_broken.dxxd') }
+        it { is_expected.to include(%(<identity format="Unknown Binary" mimetype="application/octet-stream")) }
+      end
+
+      context 'zip file should be characterized not its contents' do
+        let(:filename) { fixture_file('archive.zip') }
+        it { is_expected.to include(%(<identity format="ZIP Format" mimetype="application/zip"))}
+      end
+    end
+
+    context 'when JHOVE adds non-xml' do
+      # https://github.com/harvard-lts/fits/issues/20
+      subject { fits.call }
+
+      before do
+        expect(fits.logger).to receive(:warn)
+        allow(fits).to receive(:internal_call).and_return(
+          'READBOX seen=true
+<?xml version="1.0" encoding="UTF-8"?>
+<fits xmlns="http://hul.harvard.edu/ois/xml/ns/fits/fits_output" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://hul.harvard.edu/ois/xml/ns/fits/fits_output http://hul.harvard.edu/ois/xml/xsd/fits/fits_output.xsd" version="0.8.2" timestamp="15/09/14 10:00 AM">
+<identification/></fits>')
+      end
+
+      let(:filename) { fixture_file('brendan_behan.jpeg') }
+      it { is_expected.not_to include('READBOX') }
+    end
+
+    context "when FITS itself adds non-xml" do
+      # https://github.com/harvard-lts/fits/issues/46
+      subject { fits.call }
+
+      before do
+        expect(fits.logger).to receive(:warn)
+        allow(fits).to receive(:internal_call).and_return(
+          '2015-10-15 17:14:25,761 ERROR [main] ToolBelt:79 - Thread 1 error initializing edu.harvard.hul.ois.fits.tools.droid.Droid: edu.harvard.hul.ois.fits.exceptions.FitsToolException  Message: DROID cannot run under Java 8
+<?xml version="1.0" encoding="UTF-8"?>
+<fits xmlns="http://hul.harvard.edu/ois/xml/ns/fits/fits_output" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://hul.harvard.edu/ois/xml/ns/fits/fits_output http://hul.harvard.edu/ois/xml/xsd/fits/fits_output.xsd" version="0.8.2" timestamp="15/09/14 10:00 AM">
+<identification/></fits>')
+      end
+
+      let(:filename) { fixture_file('brendan_behan.jpeg') }
+      it { is_expected.not_to include('FitsToolException') }
+    end
+  end
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: hydra-file_characterization
 version: !ruby/object:Gem::Version
-  version: 1.0.0
+  version: 1.1.0
 platform: ruby
 authors:
 - James Treacy
@@ -11,7 +11,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-08-28 00:00:00.000000000 Z
+date: 2019-05-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activesupport
@@ -125,6 +125,7 @@ files:
 - lib/hydra/file_characterization/characterizers.rb
 - lib/hydra/file_characterization/characterizers/ffprobe.rb
 - lib/hydra/file_characterization/characterizers/fits.rb
+- lib/hydra/file_characterization/characterizers/fits_servlet.rb
 - lib/hydra/file_characterization/exceptions.rb
 - lib/hydra/file_characterization/to_temp_file.rb
 - lib/hydra/file_characterization/version.rb
@@ -136,6 +137,7 @@ files:
 - spec/fixtures/brendan_broken.dxxd.fits.xml
 - spec/lib/hydra/file_characterization/characterizer_spec.rb
 - spec/lib/hydra/file_characterization/characterizers/ffprobe_spec.rb
+- spec/lib/hydra/file_characterization/characterizers/fit_servlet_spec.rb
 - spec/lib/hydra/file_characterization/characterizers/fits_spec.rb
 - spec/lib/hydra/file_characterization/characterizers_spec.rb
 - spec/lib/hydra/file_characterization/to_temp_file_spec.rb
@@ -161,7 +163,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.7.7
+rubygems_version: 2.7.9
 signing_key: 
 specification_version: 4
 summary: To provide a wrapper for file characterization
@@ -174,6 +176,7 @@ test_files:
 - spec/fixtures/brendan_broken.dxxd.fits.xml
 - spec/lib/hydra/file_characterization/characterizer_spec.rb
 - spec/lib/hydra/file_characterization/characterizers/ffprobe_spec.rb
+- spec/lib/hydra/file_characterization/characterizers/fit_servlet_spec.rb
 - spec/lib/hydra/file_characterization/characterizers/fits_spec.rb
 - spec/lib/hydra/file_characterization/characterizers_spec.rb
 - spec/lib/hydra/file_characterization/to_temp_file_spec.rb