add retry on failure for pdf extraction

enterprise/app/jobs/captain/documents/pdf_extraction_job.rb

@@ -1,4 +1,7 @@
 class Captain::Documents::PdfExtractionJob < ApplicationJob
+  retry_on ActiveStorage::FileNotFoundError, attempts: 3, wait: 2.seconds, on: :perform
+  retry_on Faraday::BadRequestError, attempts: 3, wait: 2.seconds, on: :perform
+
   queue_as :low
 
   def perform(document)
@@ -7,8 +10,6 @@ class Captain::Documents::PdfExtractionJob < ApplicationJob
     initialize_document_processing(document)
     result = extract_pdf_content(document)
     process_extraction_result(document, result)
-  rescue Captain::Tools::PdfExtractionService::ExtractionError => e
-    handle_pdf_extraction_error(document, e)
   end
 
   private

enterprise/app/services/captain/tools/pdf_extraction_service.rb

@@ -7,7 +7,6 @@ class Captain::Tools::PdfExtractionService
   include Captain::Tools::PdfExtractionService::SourceHandler
   include Captain::Tools::PdfExtractionService::BlobHandler
   include Captain::Tools::PdfExtractionService::TextProcessor
-  include Captain::Tools::PdfExtractionService::ErrorHandler
 
   class ExtractionError < StandardError; end
 
@@ -29,7 +28,7 @@ class Captain::Tools::PdfExtractionService
     validation_result = validate_pdf_source
     return validation_result unless validation_result[:success]
 
-    extract_pdf_with_error_handling
+    process_pdf_extraction
   end
 
   private

enterprise/app/services/captain/tools/pdf_extraction_service/error_handler.rb

@@ -1,44 +0,0 @@
-module Captain::Tools::PdfExtractionService::ErrorHandler
-  private
-
-  def extract_pdf_with_error_handling
-    process_pdf_extraction
-  rescue PDF::Reader::MalformedPDFError => e
-    handle_malformed_pdf_error(e)
-  rescue PDF::Reader::UnsupportedFeatureError => e
-    handle_unsupported_feature_error(e)
-  rescue Down::TimeoutError => e
-    handle_timeout_error(e)
-  rescue StandardError => e
-    handle_general_error(e)
-  end
-
-  def handle_malformed_pdf_error(error)
-    Rails.logger.error "Malformed PDF (#{pdf_source_type}): #{error.message}"
-    failure_response(['Invalid or corrupted PDF format'])
-  end
-
-  def handle_unsupported_feature_error(error)
-    Rails.logger.error "Unsupported PDF feature (#{pdf_source_type}): #{error.message}"
-    failure_response(['PDF contains unsupported features'])
-  end
-
-  def handle_timeout_error(error)
-    Rails.logger.error "PDF download timeout (#{pdf_source_type}): #{error.message}"
-    failure_response(['PDF download timed out'])
-  end
-
-  def handle_general_error(error)
-    Rails.logger.error "PDF extraction error (#{pdf_source_type}): #{error.message}"
-    Rails.logger.error error.backtrace.join("\n")
-    failure_response(['Failed to process PDF'])
-  end
-
-  def success_response(content)
-    { success: true, content: content }
-  end
-
-  def failure_response(errors)
-    { success: false, errors: errors }
-  end
-end

enterprise/app/services/captain/tools/pdf_validation_concern.rb

@@ -20,27 +20,7 @@ module Captain::Tools::PdfValidationConcern
   end
 
   def validate_url_format
-    uri = parse_and_validate_uri
-    validate_url_scheme(uri)
-    validate_url_length
-  end
-
-  def parse_and_validate_uri
     URI.parse(pdf_source)
-  rescue URI::InvalidURIError
-    raise StandardError, 'Malformed URL'
-  end
-
-  def validate_url_scheme(uri)
-    return if %w[http https].include?(uri.scheme)
-
-    raise StandardError, 'Invalid URL scheme'
-  end
-
-  def validate_url_length
-    return if pdf_source.length <= 2000
-
-    raise StandardError, 'URL too long'
   end
 
   def validate_file_type_and_size