legal_summariser 0.1.0

data/lib/legal_summariser/text_extractor.rb

@@ -0,0 +1,79 @@
+# frozen_string_literal: true
+
+require 'pdf-reader'
+require 'docx'
+
+module LegalSummariser
+  class TextExtractor
+    # Extract text from various document formats
+    # @param file_path [String] Path to the document
+    # @return [String] Extracted text
+    def self.extract(file_path)
+      case File.extname(file_path).downcase
+      when '.pdf'
+        extract_from_pdf(file_path)
+      when '.docx'
+        extract_from_docx(file_path)
+      when '.txt'
+        File.read(file_path, encoding: 'UTF-8')
+      else
+        raise UnsupportedFormatError, "Unsupported file format: #{File.extname(file_path)}"
+      end
+    end
+
+    private
+
+    # Extract text from PDF files
+    # @param file_path [String] Path to PDF file
+    # @return [String] Extracted text
+    def self.extract_from_pdf(file_path)
+      reader = PDF::Reader.new(file_path)
+      text = ""
+      
+      reader.pages.each do |page|
+        text += page.text + "\n"
+      end
+      
+      # Clean up common PDF artifacts
+      clean_text(text)
+    rescue => e
+      raise Error, "Failed to extract text from PDF: #{e.message}"
+    end
+
+    # Extract text from DOCX files
+    # @param file_path [String] Path to DOCX file
+    # @return [String] Extracted text
+    def self.extract_from_docx(file_path)
+      doc = Docx::Document.open(file_path)
+      text = ""
+      
+      doc.paragraphs.each do |paragraph|
+        text += paragraph.text + "\n"
+      end
+      
+      clean_text(text)
+    rescue => e
+      raise Error, "Failed to extract text from DOCX: #{e.message}"
+    end
+
+    # Clean extracted text
+    # @param text [String] Raw extracted text
+    # @return [String] Cleaned text
+    def self.clean_text(text)
+      # Normalize line breaks first
+      text = text.gsub(/\r\n?/, "\n")
+      
+      # Remove common PDF artifacts
+      text = text.gsub(/\f/, '') # Form feed characters
+      text = text.gsub(/[\x00-\x08\x0B\x0C\x0E-\x1F\x7F]/, '') # Control characters
+      
+      # Remove excessive whitespace but preserve line breaks
+      text = text.gsub(/[ \t]+/, ' ')
+      
+      # Remove excessive newlines
+      text = text.gsub(/\n{3,}/, "\n\n")
+      
+      text.strip
+    end
+  end
+end

data/lib/legal_summariser/version.rb

@@ -0,0 +1,5 @@
+# frozen_string_literal: true
+
+module LegalSummariser
+  VERSION = "0.1.0"
+end

data/lib/legal_summariser.rb

@@ -0,0 +1,71 @@
+# frozen_string_literal: true
+
+require_relative "legal_summariser/version"
+require_relative "legal_summariser/document_parser"
+require_relative "legal_summariser/text_extractor"
+require_relative "legal_summariser/summariser"
+require_relative "legal_summariser/clause_detector"
+require_relative "legal_summariser/risk_analyzer"
+require_relative "legal_summariser/formatter"
+
+module LegalSummariser
+  class Error < StandardError; end
+  class DocumentNotFoundError < Error; end
+  class UnsupportedFormatError < Error; end
+
+  # Main entry point for the legal summariser
+  # @param file_path [String] Path to the legal document
+  # @param options [Hash] Configuration options
+  # @return [Hash] Summary results
+  def self.summarise(file_path, options = {})
+    raise DocumentNotFoundError, "File not found: #{file_path}" unless File.exist?(file_path)
+
+    # Extract text from document
+    text = TextExtractor.extract(file_path)
+    
+    # Perform analysis
+    summary = Summariser.new(text, options).generate
+    clauses = ClauseDetector.new(text).detect
+    risks = RiskAnalyzer.new(text).analyze
+    
+    # Format results
+    result = {
+      plain_text: summary[:plain_text],
+      key_points: summary[:key_points],
+      clauses: clauses,
+      risks: risks,
+      metadata: {
+        document_type: detect_document_type(text),
+        word_count: text.split.length,
+        processed_at: Time.now.strftime("%Y-%m-%dT%H:%M:%S%z")
+      }
+    }
+
+    # Apply formatting if requested
+    if options[:format]
+      Formatter.format(result, options[:format])
+    else
+      result
+    end
+  end
+
+  # Detect the type of legal document
+  # @param text [String] Document text
+  # @return [String] Document type
+  def self.detect_document_type(text)
+    case text.downcase
+    when /non.?disclosure|nda|confidentiality/
+      "nda"
+    when /service agreement|terms of service|tos/
+      "service_agreement"
+    when /employment|job|position/
+      "employment_contract"
+    when /privacy policy|data protection|gdpr|kvkk/
+      "privacy_policy"
+    when /license|licensing/
+      "license_agreement"
+    else
+      "general_contract"
+    end
+  end
+end

metadata

@@ -0,0 +1,204 @@
+--- !ruby/object:Gem::Specification
+name: legal_summariser
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Legal Summariser Team
+autorequire: 
+bindir: exe
+cert_chain: []
+date: 2025-09-09 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: pdf-reader
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.11'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.11'
+- !ruby/object:Gem::Dependency
+  name: docx
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.8'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.8'
+- !ruby/object:Gem::Dependency
+  name: nokogiri
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.13'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.13'
+- !ruby/object:Gem::Dependency
+  name: thor
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.2'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.2'
+- !ruby/object:Gem::Dependency
+  name: json
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.6'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.6'
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '1.17'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '1.17'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '13.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '13.0'
+- !ruby/object:Gem::Dependency
+  name: rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.12'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.12'
+- !ruby/object:Gem::Dependency
+  name: rubocop
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.50'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.50'
+- !ruby/object:Gem::Dependency
+  name: yard
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.9'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.9'
+description: A Ruby gem that summarises legal documents, extracts key clauses, flags
+  risks, and converts legal jargon into plain English. Supports PDF/Word documents
+  with offline processing capabilities.
+email:
+- info@legal-summariser.com
+executables:
+- legal_summariser
+extensions: []
+extra_rdoc_files: []
+files:
+- ".rspec"
+- CHANGELOG.md
+- Gemfile
+- README.md
+- Rakefile
+- exe/legal_summariser
+- lib/legal_summariser.rb
+- lib/legal_summariser/clause_detector.rb
+- lib/legal_summariser/document_parser.rb
+- lib/legal_summariser/formatter.rb
+- lib/legal_summariser/risk_analyzer.rb
+- lib/legal_summariser/summariser.rb
+- lib/legal_summariser/text_extractor.rb
+- lib/legal_summariser/version.rb
+homepage: https://github.com/legal-summariser/legal_summariser
+licenses:
+- MIT
+metadata:
+  allowed_push_host: https://rubygems.org
+  homepage_uri: https://github.com/legal-summariser/legal_summariser
+  source_code_uri: https://github.com/legal-summariser/legal_summariser
+  changelog_uri: https://github.com/legal-summariser/legal_summariser/blob/main/CHANGELOG.md
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: 2.6.0
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.0.3.1
+signing_key: 
+specification_version: 4
+summary: AI-powered legal document summarisation and analysis toolkit
+test_files: []