Class: Ragdoll::DocumentsController

Inherits:
ApplicationController show all
Defined in:
app/controllers/ragdoll/documents_controller.rb

Instance Method Summary collapse

Instance Method Details

#bulk_delete ⇒ Object 



273
274
275
276
277
278
279
280
281
282
283
284
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 273

def bulk_delete
  if params[:document_ids].present?
    documents = ::Ragdoll::Document.where(id: params[:document_ids])
    count = documents.count
    documents.destroy_all
    flash[:notice] = "Successfully deleted #{count} documents."
  else
    flash[:alert] = "No documents selected for deletion."
  end
  
  redirect_to ragdoll.documents_path
end

#bulk_reprocess ⇒ Object 



286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 286

def bulk_reprocess
  if params[:document_ids].present?
    documents = ::Ragdoll::Document.where(id: params[:document_ids])
    documents.each do |document|
      document.all_embeddings.destroy_all
      document.update(status: 'pending')
      ::Ragdoll::GenerateEmbeddingsJob.perform_later(document.id)
    end
    flash[:notice] = "Reprocessing initiated for #{documents.count} documents."
  else
    flash[:alert] = "No documents selected for reprocessing."
  end
  
  redirect_to ragdoll.documents_path
end

#bulk_upload ⇒ Object 



166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 166

def bulk_upload
  logger.info "🔍 BULK UPLOAD METHOD CALLED"
  logger.info "📊 Request params: #{params.inspect}"
  logger.info "📊 Session ID: #{session.id}"
  
  if params[:directory_files].present?
    # Use temp_session_id from frontend, session ID, or request ID as fallback
    session_id = if params[:temp_session_id].present?
                   params[:temp_session_id]
                 elsif session.id.present?
                   session.id.to_s
                 else
                   request.request_id
                 end
    
    logger.info "📊 Using session_id: #{session_id}"
    
    uploaded_files = params[:directory_files]
    force_duplicate = params[:force_duplicate] == '1'
    
    # Ensure uploaded_files is always an array
    uploaded_files = [uploaded_files] unless uploaded_files.is_a?(Array)
    
    logger.info "📁 Processing #{uploaded_files.size} files"
    
    # Prepare file data for background job
    file_paths_data = []
    uploaded_files.each_with_index do |file, index|
      next unless file.respond_to?(:original_filename)
      
      begin
        # Generate unique temp filename
        file_id = "#{session_id}_#{index}_#{Time.current.to_i}"
        temp_path = ::Rails.root.join('tmp', 'uploads', "#{file_id}_#{file.original_filename}")
        FileUtils.mkdir_p(File.dirname(temp_path))
        File.binwrite(temp_path, file.read)
        
        file_paths_data << {
          temp_path: temp_path.to_s,
          original_filename: file.original_filename
        }
        
        logger.info "📄 Queued file: #{file.original_filename}"
      rescue => e
        logger.error "❌ Failed to prepare file #{file.original_filename}: #{e.message}"
      end
    end
    
    if file_paths_data.any?
      # Queue background job for processing
      ::Ragdoll::BulkDocumentProcessingJob.perform_later(session_id, file_paths_data, force_duplicate)
      
      logger.info "🚀 Queued bulk processing job for #{file_paths_data.size} files"
      
      # Check if this is an AJAX request
      if request.xhr? || request.format.json?
        render json: {
          success: true,
          session_id: session_id,
          file_count: file_paths_data.size,
          message: "Upload started! Processing #{file_paths_data.size} files in the background."
        }
      else
        flash[:notice] = "Upload started! Processing #{file_paths_data.size} files in the background. You can monitor progress below."
        redirect_to ragdoll.documents_path
      end
    else
      logger.warn "⚠️ No valid files found for processing"
      
      if request.xhr? || request.format.json?
        render json: {
          success: false,
          error: "No valid files found for processing."
        }, status: :unprocessable_entity
      else
        flash[:alert] = "No valid files found for processing."
        redirect_to ragdoll.documents_path
      end
    end
  else
    logger.warn "⚠️ No files provided in bulk upload"
    
    if request.xhr? || request.format.json?
      render json: {
        success: false,
        error: "Please select files to upload."
      }, status: :bad_request
    else
      flash[:alert] = "Please select files to upload."
      redirect_to ragdoll.documents_path
    end
  end
rescue => e
  logger.error "💥 Fatal error in bulk_upload: #{e.message}"
  logger.error e.backtrace.join("\n")
  
  if request.xhr? || request.format.json?
    render json: {
      success: false,
      error: "Upload failed: #{e.message}"
    }, status: :internal_server_error
  else
    flash[:alert] = "Upload failed: #{e.message}"
    redirect_to ragdoll.documents_path
  end
end

#create ⇒ Object 



29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 29

def create
  if params[:ragdoll_document] && params[:ragdoll_document][:files].present?
    uploaded_files = params[:ragdoll_document][:files]
    force_duplicate = params[:ragdoll_document][:force_duplicate] == '1'
    @results = []
    
    # Ensure uploaded_files is always an array
    uploaded_files = [uploaded_files] unless uploaded_files.is_a?(Array)
    
    uploaded_files.each do |file|
      begin
        # Skip if file is not a valid upload object
        next unless file.respond_to?(:original_filename)
        
        # Save uploaded file temporarily
        temp_path = ::Rails.root.join('tmp', 'uploads', file.original_filename)
        FileUtils.mkdir_p(File.dirname(temp_path))
        File.binwrite(temp_path, file.read)
        
        # Use Ragdoll to add document with force option
        result = ::Ragdoll.add_document(path: temp_path.to_s, force: force_duplicate)
        
        # Get the actual document object if successful
        if result[:success] && result[:document_id]
          document = ::Ragdoll::Document.find(result[:document_id])
          duplicate_detected = result[:duplicate] || (result[:message] && result[:message].include?('already exists'))
          @results << { 
            file: file.original_filename, 
            success: true, 
            document: document, 
            message: result[:message],
            duplicate: duplicate_detected,
            forced: force_duplicate
          }
        else
          @results << { file: file.original_filename, success: false, error: result[:error] || "Unknown error" }
        end
        
        # Clean up temp file
        File.delete(temp_path) if File.exist?(temp_path)
      rescue => e
        filename = file.respond_to?(:original_filename) ? file.original_filename : file.to_s
        @results << { file: filename, success: false, error: e.message }
      end
    end
    
    render :upload_results
  elsif params[:ragdoll_document] && params[:ragdoll_document][:text_content].present?
    begin
      force_duplicate = params[:ragdoll_document][:force_duplicate] == '1'
      
      # For text content, we need to save it as a file first since Ragdoll.add_document expects a file
      temp_path = ::Rails.root.join('tmp', 'uploads', "#{SecureRandom.hex(8)}.txt")
      FileUtils.mkdir_p(File.dirname(temp_path))
      File.write(temp_path, params[:ragdoll_document][:text_content])
      
      result = ::Ragdoll.add_document(path: temp_path.to_s, force: force_duplicate)
      
      # Clean up temp file
      File.delete(temp_path) if File.exist?(temp_path)
      
      if result[:success] && result[:document_id]
        document = ::Ragdoll::Document.find(result[:document_id])
        duplicate_detected = result[:duplicate] || (result[:message] && result[:message].include?('already exists'))
        
        if duplicate_detected && !force_duplicate
          redirect_to ragdoll.document_path(document), notice: 'Document already exists - returned existing document.'
        elsif duplicate_detected && force_duplicate
          redirect_to ragdoll.document_path(document), notice: 'Document was successfully created (duplicate forced).'
        else
          redirect_to ragdoll.document_path(document), notice: 'Document was successfully created.'
        end
      else
        @document = ::Ragdoll::Document.new
        @document.errors.add(:base, result[:error] || "Unknown error occurred")
        render :new
      end
    rescue => e
      @document = ::Ragdoll::Document.new
      @document.errors.add(:base, e.message)
      render :new
    end
  else
    @document = ::Ragdoll::Document.new
    @document.errors.add(:base, "Please provide either files or text content")
    render :new
  end
end

#destroy ⇒ Object 



129
130
131
132
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 129

def destroy
  @document.destroy
  redirect_to ragdoll.documents_url, notice: 'Document was successfully deleted.'
end

#download ⇒ Object 



158
159
160
161
162
163
164
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 158

def download
  if @document.location.present? && File.exist?(@document.location)
    send_file @document.location, filename: @document.title
  else
    redirect_to ragdoll.document_path(@document), alert: 'File not found.'
  end
end

#edit ⇒ Object 



118
119
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 118

def edit
end

#index ⇒ Object 



8
9
10
11
12
13
14
15
16
17
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 8

def index
  @documents = ::Ragdoll::Document.all
  @documents = @documents.where(status: params[:status]) if params[:status].present?
  @documents = @documents.where(document_type: params[:document_type]) if params[:document_type].present?
  @documents = @documents.where('title ILIKE ?', "%#{params[:search]}%") if params[:search].present?
  @documents = @documents.order(created_at: :desc)
  
  @document_types = ::Ragdoll::Document.distinct.pluck(:document_type).compact
  @statuses = ::Ragdoll::Document.distinct.pluck(:status).compact
end

#new ⇒ Object 



25
26
27
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 25

def new
  @document = ::Ragdoll::Document.new
end

#preview ⇒ Object 



134
135
136
137
138
139
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 134

def preview
  respond_to do |format|
    format.html { render layout: false }
    format.json { render json: { content: @document.content, metadata: @document.metadata } }
  end
end

#reprocess ⇒ Object 



141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 141

def reprocess
  begin
    # Delete existing embeddings
    @document.all_embeddings.destroy_all
    
    # Reprocess document
    @document.update(status: 'pending')
    
    # Process embeddings in background
    ::Ragdoll::GenerateEmbeddingsJob.perform_later(@document.id)
    
    redirect_to ragdoll.document_path(@document), notice: 'Document reprocessing initiated.'
  rescue => e
    redirect_to ragdoll.document_path(@document), alert: "Error reprocessing document: #{e.message}"
  end
end

#show ⇒ Object 



19
20
21
22
23
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 19

def show
  @embeddings = @document.all_embeddings
  # Load recent searches for sidebar
  @recent_searches = ::Ragdoll::Search.order(created_at: :desc).limit(10)
end

#status ⇒ Object 



434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 434

def status
  @processing_stats = {
    pending: ::Ragdoll::Document.where(status: 'pending').count,
    processing: ::Ragdoll::Document.where(status: 'processing').count,
    processed: ::Ragdoll::Document.where(status: 'processed').count,
    failed: ::Ragdoll::Document.where(status: 'failed').count
  }
  
  @recent_activity = ::Ragdoll::Document.order(updated_at: :desc).limit(20)
  
  respond_to do |format|
    format.html
    format.json { render json: @processing_stats }
  end
end

#update ⇒ Object 



121
122
123
124
125
126
127
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 121

def update
  if @document.update(document_params)
    redirect_to ragdoll.document_path(@document), notice: 'Document was successfully updated.'
  else
    render :edit
  end
end

#upload_async ⇒ Object 



302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
 
# File 'app/controllers/ragdoll/documents_controller.rb', line 302

def upload_async
  logger.info "upload_async called with params: #{params.inspect}"
  logger.info "Session ID: #{session.id}"
  logger.info "Request ID: #{request.request_id}"
  logger.info "Temp Session ID: #{params[:temp_session_id]}"
  
  if params[:ragdoll_document] && params[:ragdoll_document][:files].present?
    # Priority: temp_session_id from frontend, then session ID, then request ID as fallback
    session_id = if params[:temp_session_id].present?
                   params[:temp_session_id]
                 elsif session.id.present?
                   session.id.to_s
                 else
                   request.request_id
                 end
    logger.info "Using session_id: #{session_id} (source: #{params[:temp_session_id].present? ? 'temp_session_id' : session.id.present? ? 'session' : 'request'})"
    uploaded_files = params[:ragdoll_document][:files]
    
    logger.info "Files received: #{uploaded_files.inspect}"
    
    # Ensure uploaded_files is always an array
    uploaded_files = [uploaded_files] unless uploaded_files.is_a?(Array)
    
    # Log file analysis for debugging count discrepancies
    total_files = uploaded_files.count
    valid_files = uploaded_files.select { |f| f && f.respond_to?(:original_filename) && f.original_filename.present? }
    filtered_count = total_files - valid_files.count
    
    logger.info "📊 File upload analysis:"
    logger.info "   Total files in upload array: #{total_files}"
    logger.info "   Valid files with original_filename: #{valid_files.count}"
    logger.info "   Files filtered out: #{filtered_count}"
    
    if filtered_count > 0
      uploaded_files.each_with_index do |file, index|
        unless file && file.respond_to?(:original_filename) && file.original_filename.present?
          logger.warn "   Filtered file #{index + 1}: #{file.class} - #{file.inspect}"
        end
      end
    end
    
    processed_count = 0
    results = []
    
    # Process files in batches to avoid "too many open files" error
    batch_size = 50  # Process 50 files at a time
    uploaded_files.each_slice(batch_size).with_index do |file_batch, batch_index|
      file_batch.each_with_index do |file, batch_file_index|
        index = batch_index * batch_size + batch_file_index
      next unless file.respond_to?(:original_filename)
      
      logger.info "Processing file #{index + 1}: #{file.original_filename}"
      
      begin
        # Generate unique file ID
        file_id = "#{session_id}_#{index}_#{Time.current.to_i}"
        
        # Save uploaded file temporarily
        temp_path = ::Rails.root.join('tmp', 'uploads', "#{file_id}_#{file.original_filename}")
        FileUtils.mkdir_p(File.dirname(temp_path))
        File.binwrite(temp_path, file.read)
        
        logger.info "File saved to: #{temp_path}"
        
        # Try to queue background job first, fallback to direct processing
        begin
          if defined?(::Ragdoll::ProcessFileJob)
            ::Ragdoll::ProcessFileJob.perform_later(file_id, session_id, file.original_filename, temp_path.to_s)
            logger.info "Job queued for file: #{file_id}"
            results << { file: file.original_filename, status: 'queued' }
          else
            raise "ProcessFileJob not available"
          end
        rescue => job_error
          logger.warn "Background job failed, processing directly: #{job_error.message}"
          
          # Process directly if job system is not available  
          force_duplicate = params[:ragdoll_document][:force_duplicate] == '1'
          result = ::Ragdoll.add_document(path: temp_path.to_s, force: force_duplicate)
          
          if result[:success] && result[:document_id]
            document = ::Ragdoll::Document.find(result[:document_id])
            results << { 
              file: file.original_filename, 
              status: 'completed_sync',
              document_id: document.id
            }
            logger.info "File processed synchronously: #{file.original_filename}"
          else
            results << { 
              file: file.original_filename, 
              status: 'failed',
              error: result[:error] || 'Unknown error'
            }
          end
          
          # Clean up temp file for sync processing
          File.delete(temp_path) if File.exist?(temp_path)
        end
        
        processed_count += 1
      rescue => file_error
        logger.error "Error processing file #{file.original_filename}: #{file_error.message}"
        results << { 
          file: file.original_filename, 
          status: 'failed',
          error: file_error.message
        }
      end
      end
      
      # Force garbage collection after each batch to free file descriptors
      GC.start
    end
    
    logger.info "Returning success response for #{processed_count} files"
    render json: { 
      success: true, 
      session_id: session_id,
      results: results,
      message: "#{processed_count} file(s) processed" 
    }
  else
    logger.error "No files provided in upload_async"
    render json: { success: false, error: "No files provided" }, status: :bad_request
  end
rescue => e
  logger.error "Error in upload_async: #{e.message}"
  logger.error e.backtrace.join("\n")
  render json: { success: false, error: e.message }, status: :internal_server_error
end