remove downgraded columns

db/schema.rb

@@ -289,15 +289,10 @@ ActiveRecord::Schema[7.1].define(version: 2025_07_02_075600) do
     t.datetime "created_at", null: false
     t.datetime "updated_at", null: false
     t.integer "status", default: 0, null: false
-    t.integer "document_type", default: 0, null: false
     t.string "source_type", default: "url"
-    t.string "content_type"
-    t.integer "file_size"
     t.index ["account_id"], name: "index_captain_documents_on_account_id"
     t.index ["assistant_id", "external_link"], name: "index_captain_documents_on_assistant_id_and_external_link", unique: true
     t.index ["assistant_id"], name: "index_captain_documents_on_assistant_id"
-    t.index ["content_type"], name: "index_captain_documents_on_content_type"
-    t.index ["document_type"], name: "index_captain_documents_on_document_type"
     t.index ["source_type"], name: "index_captain_documents_on_source_type"
     t.index ["status"], name: "index_captain_documents_on_status"
   end

enterprise/app/jobs/captain/tools/pdf_extraction_parser_job.rb

@@ -3,11 +3,9 @@ class Captain::Tools::PdfExtractionParserJob < ApplicationJob
 
   def perform(assistant_id:, pdf_content:, document_id: nil)
     assistant = Captain::Assistant.find(assistant_id)
-    return unless should_process_content?(pdf_content[:content], assistant.account)
+    main_document = find_main_document(document_id)
 
-    # Find the main document instead of creating a new one
-    main_document = Captain::Document.find(document_id) if document_id
-    return unless main_document
+    return unless can_process_content?(pdf_content[:content], assistant.account, main_document)
 
     enqueue_response_builder_job_for_chunk(main_document, pdf_content)
   rescue ActiveRecord::RecordNotFound => e
@@ -18,6 +16,16 @@ class Captain::Tools::PdfExtractionParserJob < ApplicationJob
 
   private
 
+  def find_main_document(document_id)
+    return unless document_id
+
+    Captain::Document.find(document_id)
+  end
+
+  def can_process_content?(content, account, main_document)
+    main_document && should_process_content?(content, account)
+  end
+
   def enqueue_response_builder_job_for_chunk(main_document, pdf_content)
     # Create a context string for better AI processing that includes page info
     page_info = pdf_content[:page_number] ? " (Page #{pdf_content[:page_number]})" : ''
@@ -29,7 +37,7 @@ class Captain::Tools::PdfExtractionParserJob < ApplicationJob
     # Use the ResponseBuilderJob with full_content parameter to generate FAQs
     # This will link all FAQs to the main document while using the chunk content for AI processing
     # Skip reset since it's already done in the main PDF extraction job
-    Captain::Documents::ResponseBuilderJob.perform_later(main_document, context_content, skip_reset: true)
+    Captain::Documents::ResponseBuilderJob.perform_later(main_document, context_content)
   end
 
   def should_process_content?(content, account)

enterprise/app/models/captain/document.rb

@@ -4,10 +4,7 @@
 #
 #  id            :bigint           not null, primary key
 #  content       :text
-#  content_type  :string
-#  document_type :integer          default(0), not null
 #  external_link :string           not null
-#  file_size     :integer
 #  name          :string
 #  source_type   :string           default("url")
 #  status        :integer          default("in_progress"), not null
@@ -21,8 +18,6 @@
 #  index_captain_documents_on_account_id                      (account_id)
 #  index_captain_documents_on_assistant_id                    (assistant_id)
 #  index_captain_documents_on_assistant_id_and_external_link  (assistant_id,external_link) UNIQUE
-#  index_captain_documents_on_content_type                    (content_type)
-#  index_captain_documents_on_document_type                   (document_type)
 #  index_captain_documents_on_source_type                     (source_type)
 #  index_captain_documents_on_status                          (status)
 #
@@ -58,7 +53,7 @@ class Captain::Document < ApplicationRecord
   scope :for_assistant, ->(assistant_id) { where(assistant_id: assistant_id) }
 
   def pdf_document?
-    source_type == 'pdf_upload' || file.attached? || pdf_url_format?
+    source_type == 'pdf_upload' || file.attached?
   end
 
   private
@@ -93,17 +88,10 @@ class Captain::Document < ApplicationRecord
   def set_default_source_type
     return if source_type.present?
 
-    self.source_type = if file.attached? || pdf_url_format?
+    self.source_type = if file.attached?
                          'pdf_upload'
                        else
                          'url'
                        end
   end
-
-  def pdf_url_format?
-    return false if external_link.blank?
-
-    url = external_link.downcase
-    url.end_with?('.pdf') || url.include?('/rails/active_storage/blobs/')
-  end
 end

spec/enterprise/jobs/captain/tools/pdf_extraction_parser_job_spec.rb

@@ -42,7 +42,7 @@ RSpec.describe Captain::Tools::PdfExtractionParserJob, type: :job do
 
         expect(Captain::Documents::ResponseBuilderJob)
           .to receive(:perform_later)
-          .with(main_document, expected_content, skip_reset: true)
+          .with(main_document, expected_content)
 
         described_class.new.perform(
           assistant_id: assistant.id,
@@ -58,7 +58,7 @@ RSpec.describe Captain::Tools::PdfExtractionParserJob, type: :job do
 
         expect(Captain::Documents::ResponseBuilderJob)
           .to receive(:perform_later)
-          .with(main_document, expected_content, skip_reset: true)
+          .with(main_document, expected_content)
 
         described_class.new.perform(
           assistant_id: assistant.id,