doc2text 0.4.0 → 0.4.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 7b9b5aeaa63d276696f0f4f716242b181d8d3aef2e47861053c03f9623cdf498
-  data.tar.gz: a9ac2a3e0314334dda782f8ce8ef0d5a0691015ae70da0cc7a7fd79b2d6d7cd2
+  metadata.gz: dcded1a71b2126a042cde415956262082bf8bb256102aaf8fcfe731ed125161b
+  data.tar.gz: af197aac8bd0bb9b40a9b9f78358e54c788f3a582acb8cb77c0e227200cfe5f4
 SHA512:
-  metadata.gz: 88fcdc3ade712a061c62641dd7713274c676f41d37c7020348ad401e0e7af3a86b07a3726a057870396ee68a290867fccf635d3191c8376b45850507e2f566e9
-  data.tar.gz: a96c1f4cbfbb42079f5e5d6eea757531d7d7e852e01724c40cc58f94ee5ebd27e700bdffc221cc7c7202a3686a6f3c40d8a5e153f07dcdca7aa4fd542b13eac9
+  metadata.gz: 4c0aed71728b8be859273e6d8b502f9007e80ab6de17666d78095f9877d4035241656c127ef705e3c7878b33cf6d26fdd00f615f8c95e4bd2998f816f584c162
+  data.tar.gz: 5ad185c29658a5e98537221415f77dd5efc1f9b43d28144b57be9cedd998bd5c7cf808438ae4b23f9d712f0e2bb4af6d1546897e1d5951ed9ab4e54ff4263630

data/lib/doc2text/odt/markdown_odt_parser.rb

@@ -21,7 +21,7 @@ module Doc2Text
 
       def end_element_namespace(name, prefix = nil, uri = nil)
         if @current_node.parent and @current_node.parent.office_text?
-          @output << @current_node.expand
+          @output.write @current_node.expand
           @current_node.delete
         end
         @current_node = @current_node.parent

data/lib/doc2text/pptx/markdown_pptx_parser.rb

@@ -0,0 +1,55 @@
+module Doc2Text
+  module Pptx
+    module XmlNodes
+      class Node < XmlBasedDocument::XmlNodes::Node
+        def self.create_node(prefix, name, parent = nil, attrs = [], markdown_odt_parser = nil)
+          begin
+            clazz = XmlNodes.const_get "#{prefix.capitalize}::W#{name}"
+          rescue NameError => e
+            # markdown_odt_parser.logger.warn "No such <#{prefix}:#{name}> found"
+            Generic.new(parent, attrs, prefix, name, markdown_odt_parser)
+          else
+            clazz.new(parent, attrs, prefix, name, markdown_odt_parser)
+          end
+        end
+
+        def body?
+          false
+        end
+      end
+
+      class PlainText < XmlBasedDocument::XmlNodes::PlainText
+        def body?
+          false
+        end
+      end
+
+      class Generic < Node
+      end
+
+      module W
+        class Wbody < Node
+          def body?
+            true
+          end
+        end
+
+        class Wbr < Node
+          def open
+            '<br/>'
+          end
+        end
+
+        class Wp < Node
+          def open
+            "\n"
+          end
+
+          def close
+            "\n"
+          end
+        end
+      end
+    end
+  end
+end

data/lib/doc2text/pptx/pptx.rb

@@ -0,0 +1,30 @@
+module Doc2Text
+  module Pptx
+    class Document < XmlBasedDocument::DocumentFile
+
+      def self.parse_and_save(input, output_filename)
+        docx = new input
+        begin
+          docx.unpack
+          output = File.open output_filename, 'w'
+          markdown = Markdown::DocxParser.new output, nil
+          begin
+            docx.parse markdown
+          ensure
+            markdown.close
+          end
+        ensure
+          docx.clean
+        end
+      end
+
+      def contains_extracted_files?
+        File.exist? File.join(extract_path, '[Content_Types].xml')
+      end
+
+      def extract_extension
+        'unpacked_pptx'
+      end
+    end
+  end
+end

data/lib/doc2text/pptx/pptx_xml_namespaces.rb

@@ -0,0 +1,55 @@
+module Doc2Text
+  module Pptx
+    module XmlNodes
+      class Node < XmlBasedDocument::XmlNodes::Node
+        def self.create_node(prefix, name, parent = nil, attrs = [], markdown_odt_parser = nil)
+          begin
+            clazz = XmlNodes.const_get "#{prefix.capitalize}::W#{name}"
+          rescue NameError => e
+            # markdown_odt_parser.logger.warn "No such <#{prefix}:#{name}> found"
+            Generic.new(parent, attrs, prefix, name, markdown_odt_parser)
+          else
+            clazz.new(parent, attrs, prefix, name, markdown_odt_parser)
+          end
+        end
+
+        def body?
+          false
+        end
+      end
+
+      class PlainText < XmlBasedDocument::XmlNodes::PlainText
+        def body?
+          false
+        end
+      end
+
+      class Generic < Node
+      end
+
+      module W
+        class Wbody < Node
+          def body?
+            true
+          end
+        end
+
+        class Wbr < Node
+          def open
+            '<br/>'
+          end
+        end
+
+        class Wp < Node
+          def open
+            "\n"
+          end
+
+          def close
+            "\n"
+          end
+        end
+      end
+    end
+  end
+end

data/lib/doc2text/resolution.rb

@@ -4,6 +4,8 @@ module Doc2Text
       case File.extname source
         when '.docx'
           Doc2Text::Docx::Document.parse_and_save source, output
+        when '.pptx'
+          Doc2Text::Pptx::Document.parse_and_save source, output
         else
           Doc2Text::Odt::Document.parse_and_save source, output
       end

data/lib/doc2text.rb

@@ -15,4 +15,8 @@ require 'doc2text/docx/docx'
 require 'doc2text/docx/markdown_docx_parser'
 require 'doc2text/docx/docx_xml_namespaces'
 
+require 'doc2text/pptx/pptx'
+require 'doc2text/pptx/markdown_pptx_parser'
+require 'doc2text/pptx/pptx_xml_namespaces'
+
 require 'doc2text/styles_parser'

metadata

@@ -1,58 +1,53 @@
 --- !ruby/object:Gem::Specification
 name: doc2text
 version: !ruby/object:Gem::Version
-  version: 0.4.0
+  version: 0.4.4
 platform: ruby
 authors:
-- Valentin Aitken
+- Valentin A.
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2019-01-14 00:00:00.000000000 Z
+date: 2021-11-30 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '1.8'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 1.8.2
+        version: 1.12.5
+    - - "<"
+      - !ruby/object:Gem::Version
+        version: 1.13.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '1.8'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 1.8.2
+        version: 1.12.5
+    - - "<"
+      - !ruby/object:Gem::Version
+        version: 1.13.0
 - !ruby/object:Gem::Dependency
   name: rubyzip
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.2'
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: 1.2.2
+        version: 2.3.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.2'
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: 1.2.2
+        version: 2.3.0
 description: Parses odt to markdown
 email: valentin@nalisbg.com
-executables: []
+executables:
+- doc2text
 extensions: []
 extra_rdoc_files: []
 files:
@@ -66,6 +61,9 @@ files:
 - lib/doc2text/odt/markdown_odt_parser.rb
 - lib/doc2text/odt/odt.rb
 - lib/doc2text/odt/odt_xml_namespaces.rb
+- lib/doc2text/pptx/markdown_pptx_parser.rb
+- lib/doc2text/pptx/pptx.rb
+- lib/doc2text/pptx/pptx_xml_namespaces.rb
 - lib/doc2text/resolution.rb
 - lib/doc2text/styles_parser.rb
 - lib/doc2text/xml_based_document_file.rb
@@ -88,8 +86,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubyforge_project: 
-rubygems_version: 2.7.8
+rubygems_version: 3.1.2
 signing_key: 
 specification_version: 4
 summary: Translates odt to markdown