chupa-text-decomposer-pdf 1.0.2 → 1.0.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 71901aa054dc878f955620473635ee11df436a2b
-  data.tar.gz: 68bbc514ce94a38865e48a27e1849c1baec7df9e
+  metadata.gz: '083dd5ec66cd3d8859092f81abcc0d3e5c6f3929'
+  data.tar.gz: 8ec37a765b3402bf97f3f93a5ef5748aa8257499
 SHA512:
-  metadata.gz: 66d265273482895c235c053c5cd69ae92a9510288abcc2a72a8a4f686d9942c15fe6ae4d442d13cf95bb21f5696b81aac68e354c7a023b25da1b292aaece0ade
-  data.tar.gz: 649fa65ae2b74fc051896afc161eb91c8db9ad04626af94d287bf32bf8961a0ce7feb521f635ccf043297172d550ce2bb1c59ea826e211c43eb6a8fa84d1a943
+  metadata.gz: 3114a7e40b3c25cca8294337108276b1f204224bf8e815201145ab97959eac8387aacbb2c786860732624e2ec282be98793c00b4a9b8e6546eb60cc122cbc0d1
+  data.tar.gz: c68f85e85556afa87c7926c862ab3bb429439048220738b8aeb33798bcb125b9729a175c98ac4f8bf3cb17837734776f6e9a18fb4b86460ff049875183c2e967

data/Rakefile

@@ -1,6 +1,6 @@
-# -*- mode: ruby; coding: utf-8 -*-
+# -*- ruby -*-
 #
-# Copyright (C) 2013  Kouhei Sutou <kou@clear-code.com>
+# Copyright (C) 2013-2017  Kouhei Sutou <kou@clear-code.com>
 #
 # This library is free software; you can redistribute it and/or
 # modify it under the terms of the GNU Lesser General Public
@@ -34,7 +34,9 @@ end
 helper.install
 spec = helper.gemspec
 
-Packnga::DocumentTask.new(spec) do
+Packnga::DocumentTask.new(spec) do |task|
+  task.original_language = "en"
+  task.translate_language = "ja"
 end
 
 Packnga::ReleaseTask.new(spec) do

data/chupa-text-decomposer-pdf.gemspec

@@ -1,6 +1,6 @@
 # -*- mode: ruby; coding: utf-8 -*-
 #
-# Copyright (C) 2013  Kouhei Sutou <kou@clear-code.com>
+# Copyright (C) 2013-2014  Kouhei Sutou <kou@clear-code.com>
 #
 # This library is free software; you can redistribute it and/or
 # modify it under the terms of the GNU Lesser General Public
@@ -22,7 +22,7 @@ end
 
 Gem::Specification.new do |spec|
   spec.name = "chupa-text-decomposer-pdf"
-  spec.version = "1.0.2"
+  spec.version = "1.0.3"
   spec.homepage = "https://github.com/ranguba/chupa-text-decomposer-pdf"
   spec.authors = ["Kouhei Sutou"]
   spec.email = ["kou@clear-code.com"]
@@ -31,7 +31,7 @@ Gem::Specification.new do |spec|
   description = clean_white_space.call(entries[entries.index("Description") + 1])
   spec.summary = description.split(/\n\n+/, 2).first
   spec.description = description
-  spec.license = "LGPLv2.1 or later"
+  spec.license = "LGPL-2.1+"
   spec.files = ["#{spec.name}.gemspec"]
   spec.files += ["README.md", "LICENSE.txt", "Rakefile", "Gemfile"]
   spec.files += [".yardopts"]

data/doc/text/news.md

@@ -1,5 +1,13 @@
 # News
 
+## 1.0.3: 2017-05-02
+
+### Improvements
+
+  * Improved PDF data detection.
+
+  * Supported source data attributes.
+
 ## 1.0.2: 2014-02-18
 
 ### Improvements

data/lib/chupa-text/decomposers/pdf.rb

@@ -24,7 +24,7 @@ module ChupaText
       registry.register("pdf", self)
 
       def target?(data)
-        data.extension == "pdf" or
+        (data.extension == "pdf" and data.body.start_with?("%PDF-1")) or
           data.mime_type == "application/pdf"
       end
 
@@ -37,8 +37,7 @@ module ChupaText
           text << page_text
           text << "\n" unless page_text.end_with?("\n")
         end
-        text_data = TextData.new(text)
-        text_data.uri = data.uri
+        text_data = TextData.new(text, :source_data => data)
         add_attribute(text_data, document, :title)
         add_attribute(text_data, document, :author)
         add_attribute(text_data, document, :subject)

data/test/test-pdf.rb

@@ -1,4 +1,4 @@
-# Copyright (C) 2013-2014  Kouhei Sutou <kou@clear-code.com>
+# Copyright (C) 2013-2017  Kouhei Sutou <kou@clear-code.com>
 #
 # This library is free software; you can redistribute it and/or
 # modify it under the terms of the GNU Lesser General Public
@@ -37,15 +37,20 @@ class TestPDF < Test::Unit::TestCase
         data = ChupaText::Data.new
         data.body = ""
         data.uri = uri
+        data.mime_type = "application/octet-stream"
         data
       end
 
       def test_pdf
-        assert_true(decomposer.target?(create_data("index.pdf")))
+        assert do
+          decomposer.target?(create_data("index.pdf"))
+        end
       end
 
       def test_html
-        assert_false(decomposer.target?(create_data("index.html")))
+        assert do
+          not decomposer.target?(create_data("index.html"))
+        end
       end
     end
 
@@ -105,7 +110,7 @@ class TestPDF < Test::Unit::TestCase
       end
 
       def test_created_time
-        assert_equal([Time.parse("2014-01-06T00:52:45+09:00")],
+        assert_equal([Time.parse("2014-01-05T06:52:45Z")],
                      decompose("created_time"))
       end
 

metadata

@@ -1,111 +1,111 @@
 --- !ruby/object:Gem::Specification
 name: chupa-text-decomposer-pdf
 version: !ruby/object:Gem::Version
-  version: 1.0.2
+  version: 1.0.3
 platform: ruby
 authors:
 - Kouhei Sutou
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-02-17 00:00:00.000000000 Z
+date: 2017-05-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: chupa-text
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: '0'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
   name: poppler
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: '0'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
   name: bundler
   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: 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: test-unit
   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: packnga
   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: redcarpet
   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: |
@@ -119,27 +119,27 @@ executables: []
 extensions: []
 extra_rdoc_files: []
 files:
-- chupa-text-decomposer-pdf.gemspec
-- README.md
+- ".yardopts"
+- Gemfile
 - LICENSE.txt
+- README.md
 - Rakefile
-- Gemfile
-- .yardopts
-- lib/chupa-text/decomposers/pdf.rb
+- chupa-text-decomposer-pdf.gemspec
 - doc/text/news.md
-- test/test-pdf.rb
-- test/fixture/one-page.pdf
-- test/fixture/multi-pages.pdf
-- test/fixture/encrypted.pdf
+- lib/chupa-text/decomposers/pdf.rb
+- test/fixture/attributes.odt
 - test/fixture/attributes.pdf
 - test/fixture/encrypted.odt
-- test/fixture/one-page.odt
+- test/fixture/encrypted.pdf
 - test/fixture/multi-pages.odt
-- test/fixture/attributes.odt
+- test/fixture/multi-pages.pdf
+- test/fixture/one-page.odt
+- test/fixture/one-page.pdf
 - test/run-test.rb
+- test/test-pdf.rb
 homepage: https://github.com/ranguba/chupa-text-decomposer-pdf
 licenses:
-- LGPLv2.1 or later
+- LGPL-2.1+
 metadata: {}
 post_install_message: 
 rdoc_options: []
@@ -147,20 +147,19 @@ 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: []
 rubyforge_project: 
-rubygems_version: 2.0.14
+rubygems_version: 2.5.2
 signing_key: 
 specification_version: 4
 summary: This is a ChupaText decomposer plugin for to extract text and meta-data from
   PDF.
 test_files: []
-has_rdoc: