Cognee main

.gitignore

@@ -33,3 +33,5 @@ docker-compose.override.yml
 # Supabase
 .env.supabase.local
 kong.yml
+
+.DS_Store

archive/backend/.dockerignore

@@ -0,0 +1,8 @@
+__pycache__
+*.pyc
+.env
+.pytest_cache
+.mypy_cache
+.coverage
+venv/
+.venv/

archive/backend/app/api.py

@@ -0,0 +1,28 @@
+from fastapi import APIRouter, Depends
+from supabase._async.client import AsyncClient
+
+from app.core.supabase import get_async_supabase
+from app.routes.classification_routes import router as classification_router
+from app.routes.migration_routes import router as migration_router
+from app.routes.pattern_recognition_routes import router as pattern_recognition_router
+from app.routes.preprocess_routes import router as preprocess_router
+from app.routes.search_routes import router as search_router
+
+api_router = APIRouter(prefix="/api")
+
+
+@api_router.get("/health")
+async def health_check(supabase: AsyncClient = Depends(get_async_supabase)):
+    try:
+        # Check raw_files table instead of tenants
+        await supabase.table("raw_files").select("count", count="exact").execute()
+        return {"status": "healthy", "database": "connected"}
+    except Exception as e:
+        return {"status": "unhealthy", "database": "disconnected", "error": str(e)}
+
+
+api_router.include_router(preprocess_router)
+api_router.include_router(search_router)
+api_router.include_router(classification_router)
+api_router.include_router(migration_router)
+api_router.include_router(pattern_recognition_router)

archive/backend/app/core/neo4j.py

@@ -0,0 +1,30 @@
+# app/core/neo4j.py
+import os
+from app.repositories.graph_repository import GraphRepository
+
+_graph_repo: GraphRepository 
+
+
+def get_graph_repo() -> GraphRepository | None:
+    """
+    Returns a shared GraphRepository singleton.
+    Connection is created once on first call, reused after that.
+    Returns None if Neo4j is not configured or unreachable.
+    """
+    global _graph_repo
+
+    if _graph_repo is not None:
+        return _graph_repo
+
+    try:
+        uri = os.getenv("NEO4J_URI", "bolt://localhost:7687")
+        user = os.getenv("NEO4J_USER", "neo4j")
+        password = os.getenv("NEO4J_PASSWORD", "password")
+        database = os.getenv("NEO4J_DATABASE", None)
+
+        _graph_repo = GraphRepository(uri, user, password, database)
+        print("Neo4j connection established", flush=True)
+        return _graph_repo
+    except Exception as e:
+        print(f"Neo4j not available: {e}", flush=True)
+        return None

archive/backend/app/main.py

@@ -0,0 +1,62 @@
+import os
+from contextlib import asynccontextmanager
+
+from dotenv import load_dotenv
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+
+# Load env vars from .env file (looks in current or parent directories)
+load_dotenv()  # noqa: E402
+
+# Fix for local dev: Map VITE_SUPABASE_URL to SUPABASE_URL if not set
+if not os.getenv("SUPABASE_URL") and os.getenv("VITE_SUPABASE_URL"):
+    os.environ["SUPABASE_URL"] = os.getenv("VITE_SUPABASE_URL")
+
+# Fix for local dev: Map SUPABASE_SERVICE_ROLE_KEY if differently named
+if not os.getenv("SUPABASE_SERVICE_ROLE_KEY") and os.getenv(
+    "BACKEND_SUPABASE_SERVICE_ROLE_KEY"
+):
+    os.environ["SUPABASE_SERVICE_ROLE_KEY"] = os.getenv(
+        "BACKEND_SUPABASE_SERVICE_ROLE_KEY"
+    )
+
+
+from app.api import api_router  # noqa: E402
+from app.core.supabase import get_async_supabase  # noqa: E402
+from app.core.webhooks import configure_webhooks  # noqa: E402
+from app.services.extraction.preprocessing_queue import init_queue  # noqa: E402
+from app.services.supabase_check import wait_for_supabase  # noqa: E402
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    # Startup
+    print("LIFESPAN STARTING", flush=True)
+    supabase = await get_async_supabase()
+
+    await wait_for_supabase(supabase)
+
+    await configure_webhooks(supabase)
+
+    await init_queue(supabase)
+
+    yield
+    # Shutdown (if needed)
+
+
+app = FastAPI(title="Cortex ETL API", lifespan=lifespan)
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=False,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+app.include_router(api_router)
+
+
+@app.get("/")
+def read_root():
+    return {"message": "Cortex ETL Backend"}

archive/backend/example.env.local

@@ -0,0 +1,4 @@
+DATABASE_URL=postgresql://postgres:password@localhost:5432/cortex
+SECRET_KEY=your-secret-key-change-this-in-production
+ALGORITHM=HS256
+ACCESS_TOKEN_EXPIRE_MINUTES=30

archive/backend/pyproject.toml

@@ -0,0 +1,28 @@
+[tool.ruff]
+line-length = 88
+target-version = "py312"
+
+[tool.ruff.lint]
+select = [
+    "E",
+    "W", 
+    "F",
+    "I",
+    "B",
+    "C4",
+    "UP",
+]
+ignore = [
+    "E501",
+    "B008",
+    "UP007"
+]
+
+[tool.ruff.format]
+quote-style = "double"
+indent-style = "space"
+skip-magic-trailing-comma = false
+line-ending = "auto"
+
+[tool.pytest.ini_options]
+pythonpath = ["."]

archive/backend/requirements.txt

@@ -0,0 +1,41 @@
+# Web Framework & Server
+fastapi==0.119.0
+uvicorn[standard]==0.37.0
+
+# Supabase Client
+supabase>=2.7.0
+gotrue>=2.5.0
+
+# Configuration
+python-dotenv==1.0.1
+
+# Data Validation
+pydantic[email]>=2.0.0
+
+# Code Quality
+ruff==0.8.4
+
+# LLM Integration  
+litellm>=1.52.0
+openai>=1.0.0
+
+# Data Analysis
+umap-learn==0.5.7
+scikit-learn==1.5.2
+numpy==2.1.3
+sentence-transformers==3.3.1
+
+# Classification
+hdbscan>=0.8.33
+
+# docling==2.55.1
+# docling-core==2.48.4
+# docling-ibm-models==3.9.1
+# docling-parse==4.5.0
+# pdfplumber==0.11.7
+# google-ai-generativelanguage==0.6.15
+# google-api-core==2.25.2
+# google-api-python-client==2.184.0
+# google-auth==2.41.1
+# google-auth-httplib2==0.2.0
+# google-generativeai==0.8.5

archive/backend/setup.cfg

@@ -0,0 +1,8 @@
+[flake8]
+max-line-length = 88
+extend-ignore = E203, W503
+exclude = .git,__pycache__,alembic
+
+[mypy]
+python_version = 3.11
+ignore_missing_imports = True

package.json

@@ -1,12 +1,12 @@
 {
     "scripts": {
         "init-dev": "npm install && node init-dev.js",
-        "build": "docker-compose build --no-cache",
-        "up": "npm run init-dev && docker-compose up -d",
-        "down": "npx supabase stop && docker-compose down",
-        "rebuild": "docker-compose down -v && npm run build && npm run up",
-        "reset": "npx supabase db reset && npm run types:generate && docker-compose restart backend",
-        "hard-clean": "npx supabase stop --no-backup && docker-compose down -v && docker container prune -f && docker network prune -f",
+        "build": "docker compose build --no-cache",
+        "up": "npm run init-dev && docker compose up -d",
+        "down": "npx supabase stop && docker compose down",
+        "rebuild": "docker compose down -v && npm run build && npm run up",
+        "reset": "npx supabase db reset && npm run types:generate && docker compose restart backend",
+        "hard-clean": "npx supabase stop --no-backup && docker compose down -v && docker container prune -f && docker network prune -f",
         "fresh": "npm run hard-clean && npm run build && npm run up && npm run types:generate",
         "types:generate": "npm run types:frontend",
         "types:frontend": "npx supabase gen types typescript --local > frontend/src/types/database.types.ts"