docsplit 0.7.4 → 0.7.5

checksums.yaml

@@ -1,7 +1,7 @@
---- 
-SHA512: 
-  metadata.gz: b275fe3c2adc1c24cdd0f82d443e4d435ca94b872763649519b2d30bd66a3af5b6e184d8dda521f7ae2bdb5f5e4054a16d157f97498dfbb4decbd195497153fc
-  data.tar.gz: 8630bc27e04716940919f00014e6d4cd2bfd9830a8d132c6ce6ad8b01f9068dd63579041b8d281e752e83a9b77641a0e7f53cd639d63db0699205cee77d4ae76
-SHA1: 
-  metadata.gz: a31b827c0439da61d1d38584a0c482866dca0cb0
-  data.tar.gz: 25afa3c6aabc5190c6850f4da2016494df7ef4ed
+---
+SHA1:
+  metadata.gz: ebdecba4d9b5b3a19244e08a2f3bcbaff8d8fab1
+  data.tar.gz: adb719d204a184313c1d282c837a7d1c929977ff
+SHA512:
+  metadata.gz: 8e58b660472bbff77ce500e50007c1dcdeec9adce59d527d9e331b42bde76d4ff9f2df9aece56e3793eef9d49795aafeddc8a7dfc9bd56a4fa07ca69fa080ca2
+  data.tar.gz: 1ca0706ec0b4eca050c9a1d0a45858365f5bde18b639c3ff641278dc777a1edf17128585f264b2a128e36a5388d4b2c7b9386e8f0b1993e850bc995d625c731b

data/docsplit.gemspec

@@ -1,7 +1,7 @@
 Gem::Specification.new do |s|
   s.name      = 'docsplit'
-  s.version   = '0.7.4'         # Keep version in sync with docsplit.rb
-  s.date      = '2014-02-16'
+  s.version   = '0.7.5'         # Keep version in sync with docsplit.rb
+  s.date      = '2014-05-28'
 
   s.homepage    = "http://documentcloud.github.com/docsplit/"
   s.summary     = "Break Apart Documents into Images, Text, Pages and PDFs"

data/lib/docsplit.rb

@@ -5,7 +5,7 @@ require 'shellwords'
 # The Docsplit module delegates to the Java PDF extractors.
 module Docsplit
 
-  VERSION       = '0.7.4' # Keep in sync with gemspec.
+  VERSION       = '0.7.5' # Keep in sync with gemspec.
 
   ESCAPE        = lambda {|x| Shellwords.shellescape(x) }
 

data/lib/docsplit/pdf_extractor.rb

@@ -52,6 +52,7 @@ module Docsplit
           /Applications/OpenOffice.org.app/Contents
         )
       else # probably linux/unix
+        # heroku libreoffice buildpack: https://github.com/rishihahs/heroku-buildpack-libreoffice
         search_paths = %w(
           /usr/lib/libreoffice
           /usr/lib64/libreoffice
@@ -59,6 +60,7 @@ module Docsplit
           /usr/lib/openoffice
           /usr/lib64/openoffice
           /opt/openoffice.org3
+          /app/vendor/libreoffice
         )
       end
       search_paths

data/lib/docsplit/transparent_pdfs.rb

@@ -8,19 +8,22 @@ module Docsplit
     # through further extraction.
     def ensure_pdfs(docs)
       [docs].flatten.map do |doc|
-        ext = File.extname(doc)
-        if ext.downcase == '.pdf'
+        if is_pdf?(doc)
           doc
         else
           tempdir = File.join(Dir.tmpdir, 'docsplit')
           extract_pdf([doc], {:output => tempdir})
-          File.join(tempdir, File.basename(doc, ext) + '.pdf')
+          File.join(tempdir, File.basename(doc, File.extname(doc)) + '.pdf')
         end
       end
     end
 
+    def is_pdf?(doc)
+      File.extname(doc).downcase == '.pdf' || File.open(doc, 'rb', &:readline) =~ /\A\%PDF-\d+(\.\d+)?/
+    end
+
   end
 
   extend TransparentPDFs
 
-end
+end

metadata

@@ -1,28 +1,33 @@
---- !ruby/object:Gem::Specification 
+--- !ruby/object:Gem::Specification
 name: docsplit
-version: !ruby/object:Gem::Version 
-  version: 0.7.4
+version: !ruby/object:Gem::Version
+  version: 0.7.5
 platform: ruby
-authors: 
+authors:
 - Jeremy Ashkenas
 - Samuel Clay
 - Ted Han
 autorequire: 
 bindir: bin
 cert_chain: []
-
-date: 2014-02-16 00:00:00 Z
+date: 2014-05-28 00:00:00.000000000 Z
 dependencies: []
-
-description: "    Docsplit is a command-line utility and Ruby library for splitting apart\n    documents into their component parts: searchable UTF-8 plain text, page\n    images or thumbnails in any format, PDFs, single pages, and document\n    metadata (title, author, number of pages...)\n"
+description: |2
+      Docsplit is a command-line utility and Ruby library for splitting apart
+      documents into their component parts: searchable UTF-8 plain text, page
+      images or thumbnails in any format, PDFs, single pages, and document
+      metadata (title, author, number of pages...)
 email: opensource@documentcloud.org
-executables: 
+executables:
 - docsplit
 extensions: []
-
 extra_rdoc_files: []
-
-files: 
+files:
+- LICENSE
+- README
+- bin/docsplit
+- docsplit.gemspec
+- lib/docsplit.rb
 - lib/docsplit/command_line.rb
 - lib/docsplit/image_extractor.rb
 - lib/docsplit/info_extractor.rb
@@ -31,8 +36,6 @@ files:
 - lib/docsplit/text_cleaner.rb
 - lib/docsplit/text_extractor.rb
 - lib/docsplit/transparent_pdfs.rb
-- lib/docsplit.rb
-- bin/docsplit
 - vendor/conf/document-formats.js
 - vendor/jodconverter/commons-cli-1.1.jar
 - vendor/jodconverter/commons-io-1.4.jar
@@ -43,34 +46,28 @@ files:
 - vendor/jodconverter/ridl-3.2.1.jar
 - vendor/jodconverter/unoil-3.2.1.jar
 - vendor/logging.properties
-- docsplit.gemspec
-- LICENSE
-- README
 homepage: http://documentcloud.github.com/docsplit/
-licenses: 
+licenses:
 - MIT
 metadata: {}
-
 post_install_message: 
 rdoc_options: []
-
-require_paths: 
+require_paths:
 - lib
-required_ruby_version: !ruby/object:Gem::Requirement 
-  requirements: 
-  - &id001 
-    - ">="
-    - !ruby/object:Gem::Version 
-      version: "0"
-required_rubygems_version: !ruby/object:Gem::Requirement 
-  requirements: 
-  - *id001
+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: []
-
 rubyforge_project: docsplit
-rubygems_version: 2.0.13
+rubygems_version: 2.2.2
 signing_key: 
 specification_version: 4
 summary: Break Apart Documents into Images, Text, Pages and PDFs
 test_files: []
-