pdf_info 0.5.0 → 0.5.1

checksums.yaml

@@ -1,15 +1,7 @@
 ---
-!binary "U0hBMQ==":
-  metadata.gz: !binary |-
-    OGYzNzE5ZTY5NDdmOWNkZmRkZTE5MTgyODk1YjM4ODFjZjY5M2E2Nw==
-  data.tar.gz: !binary |-
-    NTFiYjlhNGUzYjM0OGM3YTNhZTdmNTg0N2I4N2VlYTMwMDNiZjM2Mg==
-!binary "U0hBNTEy":
-  metadata.gz: !binary |-
-    YmEyMjM1NzFjYTRjNjQxNmUzYjk3OTQyYjZjMTc5MTI0ZDU0MmY1M2M1Yjhh
-    YzU3OWRiYzQ3MDIyMzdmMTI1YmNmNjE1MmFjNGMwMGU2NGYyZDFlNmVlN2Rk
-    Y2FiMGVjNWU3MTExYjczNmIyZTViMWUzNDUyYjc3ZmJkYmExMzk=
-  data.tar.gz: !binary |-
-    NGI0ZGViZGFjNmM5NmEyMjYzZWQ4ZmVlNGQ2NjA4MjdkNzRlYzcwODVhMjFm
-    MTk0ZjQzNzAxMmE0N2U2NjQxZjU1MjFlNjE5M2QxNDY1ZDE4NWQ4MDkzYWVm
-    MDIyYjRjMjAwOTUyZjY4NjI2NzI4MGNkYzlkNGMzMGEwNjEzZGE=
+SHA1:
+  metadata.gz: d12043f74284429c74aafe8e9784c34ed72123ba
+  data.tar.gz: 337f439eedb9250fbe5d3185b9091dbf5e811c0a
+SHA512:
+  metadata.gz: 22815c5f77af4887d1184d03904e6fce45ab572721cea97839a9fe629693837447094ad57e9f26dd74a41f02c83ca43bb67ca42691e8a8580b1e5852da56f340
+  data.tar.gz: 6dc175083822fd7f85a7bc63ece632200f33ba8496dc121f9057bc501982bae10cb5a89e7e0096d14abe020ceb9640aec96815f7b8c6a385ab2c6ac45cc0d495

data/.gitignore

@@ -4,3 +4,4 @@ doc
 .bundle
 pkg
 Gemfile.lock
+*.gem

data/lib/pdf/info.rb

@@ -18,10 +18,16 @@ module PDF
     end
 
     def command
-      output = `#{self.class.command_path} "#{@pdf_path}" -f 1 -l -1`
+      output = `#{self.class.command_path} -enc UTF-8 "#{@pdf_path}" -f 1 -l -1`
       exit_code = $? 
       case exit_code
       when 0 || nil
+        if !output.valid_encoding?
+          # It's already UTF-8, so we need to convert to UTF-16 and back to
+          # force the bad characters to be replaced.
+          output.encode!('UTF-16', :undef => :replace, :invalid => :replace, :replace => "")
+          output.encode!('UTF-8')
+        end
         return output
       else
         exit_error = PDF::Info::UnexpectedExitError.new

data/lib/pdf/info/version.rb

@@ -1,6 +1,5 @@
 module PDF
   class Info
-    VERSION = "0.5.0"
+    VERSION = "0.5.1"
   end
 end
-    

data/spec/pdf_info_spec.rb

@@ -1,3 +1,4 @@
+# encoding: utf-8
 require 'spec_helper'
 
 describe PDF::Info do
@@ -160,4 +161,18 @@ describe PDF::Info do
     its([:file_size]) { should eq "218882 bytes" }
   end
 
+  describe "running on a file with invalid utf-8 metadata" do
+    subject do
+      PDF::Info.command_path = "pdfinfo"
+      PDF::Info.new(File.join(File.dirname(__FILE__), 'assets', 'invalid-utf8.pdf')).metadata
+    end
+
+    its([:page_count]) { should == 12 }
+    its([:title]) { should eq "图形1.CDR" }
+    # the two date fields are badly encoded in this file, so the DateTime parse
+    # will fail, and we'll have a nil in these fields
+    its([:creation_date]) { should be_nil }
+    its([:modification_date]) { should be_nil }
+  end
+
 end

metadata

@@ -1,41 +1,41 @@
 --- !ruby/object:Gem::Specification
 name: pdf_info
 version: !ruby/object:Gem::Version
-  version: 0.5.0
+  version: 0.5.1
 platform: ruby
 authors:
 - Tom Taylor
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-05-16 00:00:00.000000000 Z
+date: 2013-07-22 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
         version: '0'
 description: 
@@ -57,6 +57,7 @@ files:
 - lib/pdf/info/exceptions.rb
 - lib/pdf/info/version.rb
 - pdf_info.gemspec
+- spec/assets/invalid-utf8.pdf
 - spec/assets/sample.pdf
 - spec/output/successful.txt
 - spec/pdf_info_spec.rb
@@ -72,12 +73,12 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ! '>='
+  - - '>='
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ! '>='
+  - - '>='
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
@@ -87,6 +88,7 @@ signing_key:
 specification_version: 4
 summary: Wraps the pdfinfo command line tool to provide a hash of metadata
 test_files:
+- spec/assets/invalid-utf8.pdf
 - spec/assets/sample.pdf
 - spec/output/successful.txt
 - spec/pdf_info_spec.rb