Detect identical file uploads via MD5 hashing

- Add file_hash and is_identical_file columns to proof_versions table
- Compute MD5 hash on file upload and compare with previous version
- Display warning banner when uploading identical file as revision
- Return is_identical_file in WebSocket response and API endpoints

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

backend/alembic/versions/005_add_file_hash.py

@@ -0,0 +1,35 @@
+"""Add file_hash and is_identical_file columns to proof_versions
+
+Revision ID: 005_add_file_hash
+Revises: 004_cleanup_duplicate_dropdown_options
+Create Date: 2025-01-25
+
+"""
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision: str = '005_add_file_hash'
+down_revision: Union[str, None] = '004_cleanup_duplicate_dropdown_options'
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    # Add file_hash column (MD5 hex digest is always 32 characters)
+    op.add_column(
+        'proof_versions',
+        sa.Column('file_hash', sa.String(32), nullable=True)
+    )
+    # Add is_identical_file column to track if this version is identical to previous
+    op.add_column(
+        'proof_versions',
+        sa.Column('is_identical_file', sa.Boolean(), nullable=True, server_default='false')
+    )
+
+
+def downgrade() -> None:
+    op.drop_column('proof_versions', 'is_identical_file')
+    op.drop_column('proof_versions', 'file_hash')

backend/app/api/routes.py

@@ -245,6 +245,7 @@ async def list_proofs(
                     agent_review=v.agent_review,
                     overall_status=v.overall_status,
                     workfront_id=v.workfront_id,
+                    is_identical_file=v.is_identical_file,
                     created_at=v.created_at,
                 )
                 for v in proof.versions
@@ -283,6 +284,7 @@ async def get_proof(
                 agent_review=v.agent_review,
                 overall_status=v.overall_status,
                 workfront_id=v.workfront_id,
+                is_identical_file=v.is_identical_file,
                 created_at=v.created_at,
             )
             for v in proof.versions

backend/app/api/schemas.py

@@ -57,6 +57,7 @@ class ProofVersionResponse(BaseModel):
     agent_review: Optional[dict]
     overall_status: Optional[str]
     workfront_id: Optional[str]
+    is_identical_file: Optional[bool] = False
     created_at: datetime
 
     class Config:

backend/app/models/models.py

@@ -2,7 +2,7 @@ import uuid
 from datetime import datetime
 from typing import Optional
 
-from sqlalchemy import DateTime, ForeignKey, Integer, String, Text, UniqueConstraint, func
+from sqlalchemy import Boolean, DateTime, ForeignKey, Integer, String, Text, UniqueConstraint, func
 from sqlalchemy.dialects.postgresql import JSONB, UUID
 from sqlalchemy.orm import Mapped, mapped_column, relationship
 
@@ -100,6 +100,8 @@ class ProofVersion(Base):
     agent_review: Mapped[Optional[dict]] = mapped_column(JSONB, nullable=True)
     overall_status: Mapped[Optional[str]] = mapped_column(String(50), nullable=True)
     workfront_id: Mapped[Optional[str]] = mapped_column(String(100), nullable=True)
+    file_hash: Mapped[Optional[str]] = mapped_column(String(32), nullable=True)
+    is_identical_file: Mapped[Optional[bool]] = mapped_column(Boolean, nullable=True, default=False)
     created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), server_default=func.now())
 
     # Relationships

backend/app/repositories/proof_repository.py

@@ -95,6 +95,8 @@ class ProofRepository:
         agent_review: Optional[dict] = None,
         overall_status: Optional[str] = None,
         workfront_id: Optional[str] = None,
+        file_hash: Optional[str] = None,
+        is_identical_file: Optional[bool] = None,
     ) -> ProofVersion:
         """Create a new version of a proof."""
         proof_version = ProofVersion(
@@ -105,6 +107,8 @@ class ProofRepository:
             agent_review=agent_review,
             overall_status=overall_status,
             workfront_id=workfront_id,
+            file_hash=file_hash,
+            is_identical_file=is_identical_file,
         )
         self.session.add(proof_version)
         await self.session.flush()
@@ -148,6 +152,16 @@ class ProofRepository:
         version = result.scalar_one_or_none()
         return version if version else 0
 
+    async def get_latest_version_hash(self, proof_id: uuid.UUID) -> Optional[str]:
+        """Get the file_hash from the latest version of a proof."""
+        result = await self.session.execute(
+            select(ProofVersion.file_hash)
+            .where(ProofVersion.proof_id == proof_id)
+            .order_by(ProofVersion.version.desc())
+            .limit(1)
+        )
+        return result.scalar_one_or_none()
+
     async def get_previous_version_review(
         self,
         proof_id: uuid.UUID,
@@ -243,6 +257,8 @@ class ProofRepository:
         agent_review: dict,
         overall_status: str,
         created_by: Optional[uuid.UUID] = None,
+        file_hash: Optional[str] = None,
+        is_identical_file: Optional[bool] = None,
     ) -> tuple[Proof, ProofVersion]:
         """Create or get proof and add a new version with review results."""
         proof, is_new = await self.get_or_create_proof(
@@ -281,6 +297,8 @@ class ProofRepository:
             agent_review=agent_review,
             overall_status=overall_status,
             workfront_id=version_workfront_id,
+            file_hash=file_hash,
+            is_identical_file=is_identical_file,
         )
 
         return proof, version

backend/app/websocket/handlers.py

@@ -56,6 +56,10 @@ async def handle_analyze_message(
             })
             return
 
+        # Compute file hash for duplicate detection
+        file_hash = storage_service.get_checksum(file_data)
+        logger.info(f"[WEBSOCKET] Computed file hash: {file_hash}")
+
         # Create callback for real-time updates
         async def on_agent_update(agent_name: str, review: SubReview | None) -> None:
             if not manager.is_connected(client_id):
@@ -93,6 +97,8 @@ async def handle_analyze_message(
 
         # Fetch previous analysis if this is a revision
         previous_analysis = None
+        previous_file_hash = None
+        is_identical_file = False
         campaign_id = data.get("campaign_id")
         proof_name = data.get("proof_name")
 
@@ -106,10 +112,15 @@ async def handle_analyze_message(
                     )
                     if existing_proof:
                         previous_analysis = await proof_repo.get_latest_version_review(existing_proof.id)
+                        previous_file_hash = await proof_repo.get_latest_version_hash(existing_proof.id)
                         if previous_analysis:
                             logger.info(f"[WEBSOCKET] Found previous analysis version {previous_analysis.get('version')}")
                         else:
                             logger.info("[WEBSOCKET] No previous analysis found (new proof)")
+                        # Check if file is identical to previous version
+                        if previous_file_hash and previous_file_hash == file_hash:
+                            is_identical_file = True
+                            logger.info(f"[WEBSOCKET] Identical file detected - hash matches previous version: {file_hash}")
                     else:
                         logger.info("[WEBSOCKET] No existing proof found (new proof)")
             except Exception as e:
@@ -208,6 +219,8 @@ async def handle_analyze_message(
                         thumbnail_url=thumbnail_url,
                         agent_review=result_dict,
                         overall_status=result.overallStatus,
+                        file_hash=file_hash,
+                        is_identical_file=is_identical_file,
                     )
 
                     await session.commit()
@@ -225,6 +238,7 @@ async def handle_analyze_message(
             response = {
                 "type": "complete",
                 "result": result_dict,
+                "is_identical_file": is_identical_file,
             }
             # Include proof/version IDs if persisted
             if proof_id:

frontend/components/Campaigns.tsx

@@ -6,7 +6,7 @@ import { ArrowLeftIcon } from './icons/ArrowLeftIcon';
 import type { AgentReview, FlaggedItem, ResolvedItem, OverallStatus } from '../types';
 import { FeedbackReport } from './FeedbackReport';
 import { CreateCampaignModal } from './CreateCampaignModal';
-import { CheckCircleIcon, ArrowPathIcon } from './icons/StatusIcons';
+import { CheckCircleIcon, ArrowPathIcon, ExclamationTriangleIcon } from './icons/StatusIcons';
 import { ProofPreview } from './ProofPreview';
 import { HistoryIcon } from './icons/HistoryIcon';
 import { DropdownOptions } from '../App';
@@ -1579,9 +1579,22 @@ const ProofDetailView: React.FC<{
                     </div>
                 </div>
                 <div className="mt-12 lg:mt-0 lg:col-span-2">
-                    <FeedbackReport 
-                        feedback={selectedVersion.feedback} 
-                        onFlagSubmit={handleFlagSubmitWrapper} 
+                    {selectedVersion.isIdenticalFile && (
+                        <div className="mb-6 bg-amber-50 border border-amber-200 rounded-lg p-4 flex items-start gap-3">
+                            <ExclamationTriangleIcon className="w-5 h-5 text-amber-600 flex-shrink-0 mt-0.5" />
+                            <div>
+                                <p className="font-semibold text-amber-800">Identical File Detected</p>
+                                <p className="text-sm text-amber-700 mt-1">
+                                    This file is exactly the same as the previous version.
+                                    The analysis results shown are from the new analysis,
+                                    but no changes were made to the creative.
+                                </p>
+                            </div>
+                        </div>
+                    )}
+                    <FeedbackReport
+                        feedback={selectedVersion.feedback}
+                        onFlagSubmit={handleFlagSubmitWrapper}
                         onResolveSubmit={handleResolveSubmitWrapper}
                     />
                 </div>

frontend/services/apiService.ts

@@ -27,6 +27,7 @@ export interface ProofVersionResponse {
     agent_review: AgentReview | null;
     overall_status: string | null;
     workfront_id: string | null;
+    is_identical_file: boolean | null;
     created_at: string;
 }
 
@@ -290,6 +291,7 @@ class ApiService {
                 feedback: v.agent_review || {} as AgentReview,
                 overallStatus: v.overall_status as any,
                 fileStorageKey: v.file_storage_key || '',
+                isIdenticalFile: v.is_identical_file || false,
             })),
             _id: proof.id,
             fileStorageKey: latestVersion?.file_storage_key || '',

frontend/services/geminiService.ts

@@ -27,6 +27,7 @@ export interface AnalyzeProofResult {
     proofId?: string;
     versionId?: string;
     pdfPages?: PDFPage[];
+    isIdenticalFile?: boolean;
 }
 
 /**
@@ -121,6 +122,7 @@ export const analyzeProof = async (
                             proofId: message.proof_id,
                             versionId: message.version_id,
                             pdfPages: message.pdf_pages as PDFPage[] | undefined,
+                            isIdenticalFile: message.is_identical_file as boolean | undefined,
                         });
                         break;
 

frontend/types.ts

@@ -73,3 +73,13 @@ export interface PDFPage {
   width: number;
   height: number;
 }
+
+export interface ProofVersion {
+  version: number;
+  timestamp: string;
+  workfrontId: string;
+  proofPreviewUrl?: string;
+  fileStorageKey?: string;
+  feedback: AgentReview;
+  isIdenticalFile?: boolean;
+}