chupa-text-decomposer-abiword 1.0.1 → 1.0.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 859ada0af7c396a080ec92336d58203ddcc87bbf1f82c3beb4f47db31eebed5b
-  data.tar.gz: c3929b1cc964a3b97a8c7ce690ead7482c342a0296a210a013262aaf93955509
+  metadata.gz: 73c492c3a3f54ee4fe396212e0e485ba5a52e78aafdd83803613e39e13f248b7
+  data.tar.gz: a12b3cc6676ef4c9698d0c76d417c71264aac5ce9110986fccbaa7cf7e3c57a6
 SHA512:
-  metadata.gz: f45d5a84b38180298864f8ec4829b2b21688205d8dfcf151390e426c9625bc0e453c182fac1b45553a798eec9c6bfb6a0a784c6b9fbcf18f558bee958396d8de
-  data.tar.gz: a131e344ec4d9e2f79ef08a00673fd564ddbe986e39611beedb14e0e666cb401102338d0c12264302138bea6e29a760a674398e53977969c0f094dcac268d815
+  metadata.gz: b052e072e41e4d4b747e89530de8e96238a535b6f0876b1487b9ff663e1593dccad07e6a568f8b6be7ad6d87a38c44671e81bc08bcbb3631a1ea89394620e617
+  data.tar.gz: 1064b7375d5e05a9e10ca7ecb60b2fd39feb8dcde6d408c157f33c2216b93b3abd8f7da6910a694249bcdce62f1f793173b443959ee9f9af0d8706d8937868f4

data/chupa-text-decomposer-abiword.gemspec

@@ -22,7 +22,7 @@ end
 
 Gem::Specification.new do |spec|
   spec.name = "chupa-text-decomposer-abiword"
-  spec.version = "1.0.1"
+  spec.version = "1.0.2"
   spec.homepage = "https://github.com/ranguba/chupa-text-decomposer-abiword"
   spec.authors = ["Sutou Kouhei"]
   spec.email = ["kou@clear-code.com"]
@@ -40,7 +40,6 @@ Gem::Specification.new do |spec|
   spec.files += Dir.glob("test/**/*")
 
   spec.add_runtime_dependency("chupa-text")
-  spec.add_runtime_dependency("chupa-text-decomposer-pdf")
 
   spec.add_development_dependency("bundler")
   spec.add_development_dependency("rake")

data/doc/text/news.md

@@ -1,10 +1,16 @@
 # News
 
+## 1.0.2: 2019-06-18
+
+### Improvements
+
+  * Converted to text directly.
+
 ## 1.0.1: 2019-06-13
 
 ### Improvements
 
-* Increases score.
+  * Increases score.
 
 ## 1.0.0: 2019-06-13
 

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

@@ -66,30 +66,9 @@ module ChupaText
       end
 
       def decompose(data)
-        pdf_data = convert_to_pdf(data)
-        return if pdf_data.nil?
-        yield(pdf_data)
-      end
-
-      private
-      def find_command
-        candidates = [
-          @options[:abiword],
-          ENV["ABIWORD"],
-          "abiword",
-        ]
-        candidates.each do |candidate|
-          next if candidate.nil?
-          command = ExternalCommand.new(candidate)
-          return command if command.exist?
-        end
-        nil
-      end
-
-      def convert_to_pdf(data)
-        create_tempfiles(data) do |pdf, stdout, stderr|
-          succeeded = @command.run("--to", "pdf",
-                                   "--to-name", pdf.path,
+        create_tempfiles(data) do |text, stdout, stderr|
+          succeeded = @command.run("--to", "text",
+                                   "--to-name", text.path,
                                    data.path.to_s,
                                    {
                                      data: data,
@@ -107,26 +86,38 @@ module ChupaText
                 "error: <#{stderr.read}>",
               ].join("\n")
             end
-            return nil
+            return
           end
-          normalized_pdf_uri = data.uri.to_s.gsub(/\.[^.]+\z/, ".pdf")
-          File.open(pdf.path, "rb") do |pdf_input|
-            VirtualFileData.new(normalized_pdf_uri,
-                                pdf_input,
-                                source_data: data)
+          File.open(text.path) do |text_input|
+            yield(TextData.new(text_input.read, source_data: data))
           end
         end
       end
 
+      private
+      def find_command
+        candidates = [
+          @options[:abiword],
+          ENV["ABIWORD"],
+          "abiword",
+        ]
+        candidates.each do |candidate|
+          next if candidate.nil?
+          command = ExternalCommand.new(candidate)
+          return command if command.exist?
+        end
+        nil
+      end
+
       def create_tempfiles(data)
         basename = File.basename(data.path)
-        pdf = Tempfile.new([basename, ".pdf"])
+        text = Tempfile.new([basename, ".txt"])
         stdout = Tempfile.new([basename, ".stdout.log"])
         stderr = Tempfile.new([basename, ".stderr.log"])
         begin
-          yield(pdf, stdout, stderr)
+          yield(text, stdout, stderr)
         ensure
-          pdf.close!
+          text.close!
           stdout.close!
           stderr.close!
         end

data/test/test-abw.rb

@@ -58,7 +58,7 @@ class TestAbw < Test::Unit::TestCase
 
     sub_test_case("one page") do
       def test_body
-        assert_equal(["Page1\n"], decompose.collect(&:body))
+        assert_equal(["Page1"], decompose.collect(&:body))
       end
 
       private
@@ -69,8 +69,9 @@ class TestAbw < Test::Unit::TestCase
 
     sub_test_case("multi pages") do
       def test_body
-        assert_equal([<<-BODY], decompose.collect(&:body))
+        assert_equal([<<-BODY.chomp], decompose.collect(&:body))
 Page1
+\f
 Page2
         BODY
       end

data/test/test-doc.rb

@@ -71,6 +71,7 @@ class TestDoc < Test::Unit::TestCase
       def test_body
         assert_equal([<<-BODY], decompose.collect(&:body))
 Page1
+\f
 Page2
         BODY
       end

data/test/test-docx.rb

@@ -58,7 +58,7 @@ class TestDocx < Test::Unit::TestCase
 
     sub_test_case("one page") do
       def test_body
-        assert_equal(["Page1\n"], decompose.collect(&:body))
+        assert_equal(["Page1"], decompose.collect(&:body))
       end
 
       private
@@ -69,9 +69,9 @@ class TestDocx < Test::Unit::TestCase
 
     sub_test_case("multi pages") do
       def test_body
-        assert_equal([<<-BODY], decompose.collect(&:body))
+        assert_equal([<<-BODY.chomp], decompose.collect(&:body))
 Page1
-Page2
+\fPage2
         BODY
       end
 

data/test/test-odt.rb

@@ -58,7 +58,7 @@ class TestOdt < Test::Unit::TestCase
 
     sub_test_case("one page") do
       def test_body
-        assert_equal(["Page1\n"], decompose.collect(&:body))
+        assert_equal(["Page1"], decompose.collect(&:body))
       end
 
       private
@@ -69,8 +69,9 @@ class TestOdt < Test::Unit::TestCase
 
     sub_test_case("multi pages") do
       def test_body
-        assert_equal([<<-BODY], decompose.collect(&:body))
+        assert_equal([<<-BODY.chomp], decompose.collect(&:body))
 Page1
+\f
 Page2
         BODY
       end

data/test/test-rtf.rb

@@ -58,7 +58,7 @@ class TestRtf < Test::Unit::TestCase
 
     sub_test_case("one page") do
       def test_body
-        assert_equal(["Page1\n"], decompose.collect(&:body))
+        assert_equal(["Page1"], decompose.collect(&:body))
       end
 
       private
@@ -69,7 +69,7 @@ class TestRtf < Test::Unit::TestCase
 
     sub_test_case("multi pages") do
       def test_body
-        assert_equal([<<-BODY], decompose.collect(&:body))
+        assert_equal([<<-BODY.chomp], decompose.collect(&:body))
 Page1
 Page2
         BODY

data/test/test-zabw.rb

@@ -45,7 +45,7 @@ class TestZabw < Test::Unit::TestCase
 
     sub_test_case("one page") do
       def test_body
-        assert_equal(["Page1\n"], decompose.collect(&:body))
+        assert_equal(["Page1"], decompose.collect(&:body))
       end
 
       private
@@ -56,8 +56,9 @@ class TestZabw < Test::Unit::TestCase
 
     sub_test_case("multi pages") do
       def test_body
-        assert_equal([<<-BODY], decompose.collect(&:body))
+        assert_equal([<<-BODY.chomp], decompose.collect(&:body))
 Page1
+\f
 Page2
         BODY
       end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: chupa-text-decomposer-abiword
 version: !ruby/object:Gem::Version
-  version: 1.0.1
+  version: 1.0.2
 platform: ruby
 authors:
 - Sutou Kouhei
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2019-06-13 00:00:00.000000000 Z
+date: 2019-06-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: chupa-text
@@ -24,20 +24,6 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
-- !ruby/object:Gem::Dependency
-  name: chupa-text-decomposer-pdf
-  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