feat: implementar extraccion optativa de urls de pdf

lib/plaintext.rb

@@ -1,6 +1,7 @@
 # frozen_string_literal: true
 
 require 'active_support/core_ext/string'
+require 'active_support/core_ext/object/blank'
 
 require 'plaintext/version'
 require 'plaintext/configuration'
@@ -12,6 +13,7 @@
 require 'plaintext/file_handler/external_command_handler/doc_handler'
 require 'plaintext/file_handler/external_command_handler/image_handler'
 require 'plaintext/file_handler/external_command_handler/pdf_handler'
+require 'plaintext/pdf_url_extractor'
 require 'plaintext/file_handler/external_command_handler/ppt_handler'
 require 'plaintext/file_handler/external_command_handler/rtf_handler'
 require 'plaintext/file_handler/external_command_handler/xls_handler'

lib/plaintext/file_handler/external_command_handler/pdf_handler.rb

@@ -11,6 +11,19 @@ def initialize
       @command = Plaintext::Configuration['pdftotext'] || DEFAULT
     end
 
+    def text(file, options = {})
+      base_text = super(file, options)
+
+      if options[:extract_urls]
+        urls = Plaintext::PdfUrlExtractor.new(file).extract
+        if urls.any?
+          base_text += "\n\n" + urls.join("\n") + "\n"
+        end
+      end
+
+      base_text
+    end
+
     protected
 
     def utf8_stream?

lib/plaintext/pdf_url_extractor.rb

@@ -0,0 +1,42 @@
+# frozen_string_literal: true
+
+require 'pathname'
+
+module Plaintext
+  class PdfUrlExtractor
+    def initialize(file_path, command = nil)
+      @file_path = Pathname(file_path).to_s
+      @command = command || default_command
+    end
+
+    def extract
+      return [] unless available?
+
+      xml_output = `#{@command} -xml -stdout "#{@file_path}" 2>/dev/null`
+      urls = xml_output.scan(/<a\s+href="([^"]+)">/).flatten
+
+      urls.compact.uniq
+    rescue StandardError
+      []
+    end
+
+    def available?
+      @command && File.executable?(@command)
+    end
+
+    private
+
+    def default_command
+      pdftotext_cmd = Plaintext::Configuration['pdftotext']&.first || '/usr/bin/pdftotext'
+
+      # Determine pdftohtml path based on pdftotext path
+      pdftohtml_cmd = pdftotext_cmd.gsub('pdftotext', 'pdftohtml')
+
+      return pdftohtml_cmd if File.executable?(pdftohtml_cmd)
+
+      # Fallback to system PATH
+      pdftohtml_in_path = ENV['PATH'].split(File::PATH_SEPARATOR).map { |p| File.join(p, 'pdftohtml') }.find { |f| File.executable?(f) && !File.directory?(f) }
+      pdftohtml_in_path || 'pdftohtml'
+    end
+  end
+end

spec/lib/file_handler/external_command_handler/pdf_handler_spec.rb

@@ -20,6 +20,21 @@
       expect(subject.text(file)).to match /In der Küche hat es eine Kaffeemaschine/
       expect(Plaintext::Resolver.new(file, 'application/pdf').text).to match /In der Küche hat es eine Kaffeemaschine/
     end
+
+    context 'when extracting URLs is requested' do
+      let(:file) { File.new('spec/fixtures/files/text-with-links.pdf', 'r') }
+
+      it 'should NOT append URLs by default' do
+        extracted_text = subject.text(file)
+        expect(extracted_text).not_to match /http/
+      end
+
+      it 'should append URLs if extract_urls option is true' do
+        extracted_text = subject.text(file, extract_urls: true)
+        expect(extracted_text).to match /http/
+        expect(extracted_text).to match /mailto:/
+      end
+    end
   else
     warn "#{described_class.name} could not be tested as external program is not available."
   end

spec/lib/pdf_url_extractor_spec.rb

@@ -0,0 +1,20 @@
+# frozen_string_literal: true
+
+require 'spec_helper'
+
+describe Plaintext::PdfUrlExtractor do
+  let(:file) { File.new('spec/fixtures/files/text-with-links.pdf', 'r') }
+  subject { described_class.new(file) }
+
+  if described_class.new('dummy').available?
+    it 'should extract URLs from a PDF file' do
+      urls = subject.extract
+      expect(urls).to be_an(Array)
+      expect(urls.size).to be > 0
+      expect(urls).to include(match(/http/))
+      expect(urls).to include(match(/mailto:/))
+    end
+  else
+    warn "#{described_class.name} could not be tested as external program pdftohtml is not available."
+  end
+end

spec/spec_helper.rb

@@ -1,5 +1,12 @@
 # frozen_string_literal: true
 
 $LOAD_PATH.unshift File.expand_path('../../lib', __FILE__)
+require 'active_support/all'
 require 'plaintext'
 require 'byebug'
+
+pdftotext_path = ENV['PATH'].split(File::PATH_SEPARATOR).map { |p| File.join(p, 'pdftotext') }.find { |f| File.executable?(f) && !File.directory?(f) }
+if pdftotext_path
+  Plaintext::Configuration.config ||= {}
+  Plaintext::Configuration.config['pdftotext'] = [pdftotext_path, '-enc', 'UTF-8', '__FILE__', '-']
+end