Upload 45 files

Dockerfile

@@ -0,0 +1,24 @@
+FROM python:3.11-slim
+
+WORKDIR /app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+    gcc \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements first for better caching
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy application code
+COPY . .
+
+# Create data directory for SQLite
+RUN mkdir -p /app/data
+
+# Expose port (HF Spaces uses 7860)
+EXPOSE 7860
+
+# Run the application
+CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -1,11 +1,39 @@
----
-title: Numidium
-emoji: 🐢
-colorFrom: pink
-colorTo: green
-sdk: docker
-pinned: false
-license: apache-2.0
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Numidium Backend
+
+## Plataforma de Inteligência
+
+Backend do sistema Numidium/VANTAGE - uma plataforma de inteligência e análise de dados.
+
+## Stack
+
+- **FastAPI** - API REST
+- **SQLite** - Banco de dados
+- **BeautifulSoup** - Web scraping
+- **Requests** - HTTP client
+
+## Deploy
+
+Hospedado no Hugging Face Spaces com Docker.
+
+## Rodando Localmente
+
+```bash
+pip install -r requirements.txt
+uvicorn app.main:app --reload
+```
+
+## Estrutura
+
+```
+backend/
+├── app/
+│   ├── main.py          # FastAPI app
+│   ├── config.py        # Configurações
+│   ├── api/routes/      # Endpoints
+│   ├── core/            # Database, utils
+│   ├── models/          # SQLAlchemy models
+│   ├── schemas/         # Pydantic schemas
+│   └── services/        # Business logic
+├── Dockerfile
+└── requirements.txt
+```

app/__init__.py

@@ -0,0 +1 @@
+# Numidium Backend App

app/api/__init__.py

@@ -0,0 +1 @@
+# API module

app/api/routes/__init__.py

@@ -0,0 +1,2 @@
+# API Routes module
+from app.api.routes import entities, relationships, events, search, ingest

app/api/routes/entities.py

@@ -0,0 +1,163 @@
+"""
+Entity CRUD Routes
+"""
+from fastapi import APIRouter, Depends, HTTPException, Query
+from sqlalchemy.orm import Session
+from sqlalchemy import or_
+from typing import List, Optional
+
+from app.core.database import get_db
+from app.models import Entity, Relationship
+from app.schemas import EntityCreate, EntityUpdate, EntityResponse, GraphData, GraphNode, GraphEdge
+
+router = APIRouter(prefix="/entities", tags=["Entities"])
+
+
+@router.get("/", response_model=List[EntityResponse])
+def list_entities(
+    type: Optional[str] = None,
+    search: Optional[str] = None,
+    limit: int = Query(default=50, le=200),
+    offset: int = 0,
+    db: Session = Depends(get_db)
+):
+    """Lista todas as entidades com filtros opcionais"""
+    query = db.query(Entity)
+    
+    if type:
+        query = query.filter(Entity.type == type)
+    
+    if search:
+        query = query.filter(
+            or_(
+                Entity.name.ilike(f"%{search}%"),
+                Entity.description.ilike(f"%{search}%")
+            )
+        )
+    
+    query = query.order_by(Entity.created_at.desc())
+    return query.offset(offset).limit(limit).all()
+
+
+@router.get("/types")
+def get_entity_types(db: Session = Depends(get_db)):
+    """Retorna todos os tipos de entidade únicos"""
+    types = db.query(Entity.type).distinct().all()
+    return [t[0] for t in types]
+
+
+@router.get("/{entity_id}", response_model=EntityResponse)
+def get_entity(entity_id: str, db: Session = Depends(get_db)):
+    """Busca uma entidade por ID"""
+    entity = db.query(Entity).filter(Entity.id == entity_id).first()
+    if not entity:
+        raise HTTPException(status_code=404, detail="Entity not found")
+    return entity
+
+
+@router.post("/", response_model=EntityResponse, status_code=201)
+def create_entity(entity: EntityCreate, db: Session = Depends(get_db)):
+    """Cria uma nova entidade"""
+    db_entity = Entity(**entity.model_dump())
+    db.add(db_entity)
+    db.commit()
+    db.refresh(db_entity)
+    return db_entity
+
+
+@router.put("/{entity_id}", response_model=EntityResponse)
+def update_entity(entity_id: str, entity: EntityUpdate, db: Session = Depends(get_db)):
+    """Atualiza uma entidade existente"""
+    db_entity = db.query(Entity).filter(Entity.id == entity_id).first()
+    if not db_entity:
+        raise HTTPException(status_code=404, detail="Entity not found")
+    
+    update_data = entity.model_dump(exclude_unset=True)
+    for field, value in update_data.items():
+        setattr(db_entity, field, value)
+    
+    db.commit()
+    db.refresh(db_entity)
+    return db_entity
+
+
+@router.delete("/{entity_id}")
+def delete_entity(entity_id: str, db: Session = Depends(get_db)):
+    """Deleta uma entidade"""
+    db_entity = db.query(Entity).filter(Entity.id == entity_id).first()
+    if not db_entity:
+        raise HTTPException(status_code=404, detail="Entity not found")
+    
+    # Delete related relationships
+    db.query(Relationship).filter(
+        or_(
+            Relationship.source_id == entity_id,
+            Relationship.target_id == entity_id
+        )
+    ).delete()
+    
+    db.delete(db_entity)
+    db.commit()
+    return {"message": "Entity deleted"}
+
+
+@router.get("/{entity_id}/connections", response_model=GraphData)
+def get_entity_connections(
+    entity_id: str,
+    depth: int = Query(default=1, le=3),
+    db: Session = Depends(get_db)
+):
+    """
+    Retorna o grafo de conexões de uma entidade
+    Usado para visualização de rede no frontend
+    """
+    entity = db.query(Entity).filter(Entity.id == entity_id).first()
+    if not entity:
+        raise HTTPException(status_code=404, detail="Entity not found")
+    
+    nodes = {}
+    edges = []
+    visited = set()
+    
+    def explore(eid: str, current_depth: int):
+        if current_depth > depth or eid in visited:
+            return
+        visited.add(eid)
+        
+        e = db.query(Entity).filter(Entity.id == eid).first()
+        if not e:
+            return
+            
+        nodes[e.id] = GraphNode(
+            id=e.id,
+            type=e.type,
+            name=e.name,
+            properties=e.properties or {}
+        )
+        
+        # Outgoing relationships
+        for rel in db.query(Relationship).filter(Relationship.source_id == eid).all():
+            edges.append(GraphEdge(
+                source=rel.source_id,
+                target=rel.target_id,
+                type=rel.type,
+                confidence=rel.confidence
+            ))
+            explore(rel.target_id, current_depth + 1)
+        
+        # Incoming relationships
+        for rel in db.query(Relationship).filter(Relationship.target_id == eid).all():
+            edges.append(GraphEdge(
+                source=rel.source_id,
+                target=rel.target_id,
+                type=rel.type,
+                confidence=rel.confidence
+            ))
+            explore(rel.source_id, current_depth + 1)
+    
+    explore(entity_id, 0)
+    
+    return GraphData(
+        nodes=list(nodes.values()),
+        edges=edges
+    )

app/api/routes/events.py

@@ -0,0 +1,116 @@
+"""
+Events CRUD Routes
+"""
+from fastapi import APIRouter, Depends, HTTPException, Query
+from sqlalchemy.orm import Session
+from sqlalchemy import or_
+from typing import List, Optional
+from datetime import datetime
+
+from app.core.database import get_db
+from app.models import Event
+from app.schemas import EventCreate, EventResponse
+
+router = APIRouter(prefix="/events", tags=["Events"])
+
+
+@router.get("/", response_model=List[EventResponse])
+def list_events(
+    type: Optional[str] = None,
+    search: Optional[str] = None,
+    start_date: Optional[datetime] = None,
+    end_date: Optional[datetime] = None,
+    limit: int = Query(default=50, le=200),
+    offset: int = 0,
+    db: Session = Depends(get_db)
+):
+    """Lista eventos com filtros opcionais"""
+    query = db.query(Event)
+    
+    if type:
+        query = query.filter(Event.type == type)
+    
+    if search:
+        query = query.filter(
+            or_(
+                Event.title.ilike(f"%{search}%"),
+                Event.description.ilike(f"%{search}%")
+            )
+        )
+    
+    if start_date:
+        query = query.filter(Event.event_date >= start_date)
+    if end_date:
+        query = query.filter(Event.event_date <= end_date)
+    
+    query = query.order_by(Event.event_date.desc().nullslast())
+    return query.offset(offset).limit(limit).all()
+
+
+@router.get("/types")
+def get_event_types(db: Session = Depends(get_db)):
+    """Retorna todos os tipos de evento únicos"""
+    types = db.query(Event.type).distinct().all()
+    return [t[0] for t in types]
+
+
+@router.get("/timeline")
+def get_timeline(
+    entity_id: Optional[str] = None,
+    limit: int = Query(default=50, le=200),
+    db: Session = Depends(get_db)
+):
+    """
+    Retorna eventos em formato timeline
+    Útil para visualização no frontend
+    """
+    query = db.query(Event).filter(Event.event_date.isnot(None))
+    
+    if entity_id:
+        # Filter events that include this entity
+        # Note: This is a simple implementation; for production, use a proper JSON query
+        query = query.filter(Event.entity_ids.contains([entity_id]))
+    
+    events = query.order_by(Event.event_date.asc()).limit(limit).all()
+    
+    return [
+        {
+            "id": e.id,
+            "title": e.title,
+            "date": e.event_date.isoformat() if e.event_date else None,
+            "type": e.type,
+            "location": e.location_name
+        }
+        for e in events
+    ]
+
+
+@router.get("/{event_id}", response_model=EventResponse)
+def get_event(event_id: str, db: Session = Depends(get_db)):
+    """Busca um evento por ID"""
+    event = db.query(Event).filter(Event.id == event_id).first()
+    if not event:
+        raise HTTPException(status_code=404, detail="Event not found")
+    return event
+
+
+@router.post("/", response_model=EventResponse, status_code=201)
+def create_event(event: EventCreate, db: Session = Depends(get_db)):
+    """Cria um novo evento"""
+    db_event = Event(**event.model_dump())
+    db.add(db_event)
+    db.commit()
+    db.refresh(db_event)
+    return db_event
+
+
+@router.delete("/{event_id}")
+def delete_event(event_id: str, db: Session = Depends(get_db)):
+    """Deleta um evento"""
+    db_event = db.query(Event).filter(Event.id == event_id).first()
+    if not db_event:
+        raise HTTPException(status_code=404, detail="Event not found")
+    
+    db.delete(db_event)
+    db.commit()
+    return {"message": "Event deleted"}

app/api/routes/ingest.py

@@ -0,0 +1,163 @@
+"""
+Data Ingestion Routes
+Endpoints para importar dados de fontes externas
+"""
+from fastapi import APIRouter, Depends, HTTPException, BackgroundTasks
+from sqlalchemy.orm import Session
+from typing import Optional, List
+
+from app.core.database import get_db
+from app.models import Entity, Document
+from app.schemas import EntityResponse, DocumentResponse
+from app.services.ingestion import wikipedia_scraper, news_service
+
+router = APIRouter(prefix="/ingest", tags=["Data Ingestion"])
+
+
+# ========== Wikipedia ==========
+
+@router.get("/wikipedia/search")
+def search_wikipedia(q: str, limit: int = 10):
+    """Busca artigos na Wikipedia"""
+    results = wikipedia_scraper.search(q, limit)
+    return results
+
+
+@router.post("/wikipedia/entity", response_model=EntityResponse)
+def import_from_wikipedia(
+    title: str,
+    entity_type: str = "person",
+    db: Session = Depends(get_db)
+):
+    """
+    Importa uma entidade da Wikipedia
+    entity_type: person, organization, location
+    """
+    # Check if entity already exists
+    existing = db.query(Entity).filter(
+        Entity.name == title,
+        Entity.source == "wikipedia"
+    ).first()
+    
+    if existing:
+        return existing
+    
+    # Scrape based on type
+    if entity_type == "person":
+        data = wikipedia_scraper.scrape_person(title)
+    elif entity_type == "organization":
+        data = wikipedia_scraper.scrape_organization(title)
+    elif entity_type == "location":
+        data = wikipedia_scraper.scrape_location(title)
+    else:
+        data = wikipedia_scraper.scrape_person(title)  # default
+    
+    if not data:
+        raise HTTPException(status_code=404, detail="Article not found on Wikipedia")
+    
+    # Create entity
+    entity = Entity(**data)
+    db.add(entity)
+    db.commit()
+    db.refresh(entity)
+    
+    return entity
+
+
+# ========== News ==========
+
+@router.get("/news/feeds")
+def list_available_feeds():
+    """Lista os feeds de notícias disponíveis"""
+    return list(news_service.RSS_FEEDS.keys())
+
+
+@router.get("/news/fetch")
+def fetch_news(feed: Optional[str] = None):
+    """
+    Busca notícias dos feeds RSS
+    Se feed não for especificado, busca de todos
+    """
+    if feed:
+        if feed not in news_service.RSS_FEEDS:
+            raise HTTPException(status_code=404, detail="Feed not found")
+        url = news_service.RSS_FEEDS[feed]
+        articles = news_service.fetch_feed(url)
+    else:
+        articles = news_service.fetch_all_feeds()
+    
+    return articles
+
+
+@router.get("/news/search")
+def search_news(q: str):
+    """Busca notícias por palavra-chave via Google News"""
+    return news_service.search_news(q)
+
+
+@router.post("/news/import")
+def import_news(
+    query: Optional[str] = None,
+    feed: Optional[str] = None,
+    db: Session = Depends(get_db)
+):
+    """
+    Importa notícias como documentos no sistema
+    """
+    if query:
+        articles = news_service.search_news(query)
+    elif feed:
+        if feed not in news_service.RSS_FEEDS:
+            raise HTTPException(status_code=404, detail="Feed not found")
+        articles = news_service.fetch_feed(news_service.RSS_FEEDS[feed])
+    else:
+        articles = news_service.fetch_all_feeds()
+    
+    imported = 0
+    for article in articles:
+        # Check if document already exists (by URL)
+        if article.get("url"):
+            existing = db.query(Document).filter(
+                Document.source_url == article["url"]
+            ).first()
+            if existing:
+                continue
+        
+        doc_data = news_service.to_document(article)
+        doc = Document(**doc_data)
+        db.add(doc)
+        imported += 1
+    
+    db.commit()
+    
+    return {"message": f"Imported {imported} articles", "total_found": len(articles)}
+
+
+# ========== Manual Import ==========
+
+@router.post("/bulk/entities")
+def bulk_import_entities(
+    entities: List[dict],
+    db: Session = Depends(get_db)
+):
+    """
+    Importa múltiplas entidades de uma vez
+    Útil para importar de CSV/JSON
+    """
+    imported = 0
+    for entity_data in entities:
+        entity = Entity(
+            type=entity_data.get("type", "unknown"),
+            name=entity_data.get("name", "Unnamed"),
+            description=entity_data.get("description"),
+            properties=entity_data.get("properties", {}),
+            latitude=entity_data.get("latitude"),
+            longitude=entity_data.get("longitude"),
+            source=entity_data.get("source", "manual")
+        )
+        db.add(entity)
+        imported += 1
+    
+    db.commit()
+    
+    return {"message": f"Imported {imported} entities"}

app/api/routes/relationships.py

@@ -0,0 +1,71 @@
+"""
+Relationship CRUD Routes
+"""
+from fastapi import APIRouter, Depends, HTTPException, Query
+from sqlalchemy.orm import Session
+from typing import List, Optional
+
+from app.core.database import get_db
+from app.models import Relationship, Entity
+from app.schemas import RelationshipCreate, RelationshipResponse
+
+router = APIRouter(prefix="/relationships", tags=["Relationships"])
+
+
+@router.get("/", response_model=List[RelationshipResponse])
+def list_relationships(
+    type: Optional[str] = None,
+    source_id: Optional[str] = None,
+    target_id: Optional[str] = None,
+    limit: int = Query(default=50, le=200),
+    db: Session = Depends(get_db)
+):
+    """Lista relacionamentos com filtros opcionais"""
+    query = db.query(Relationship)
+    
+    if type:
+        query = query.filter(Relationship.type == type)
+    if source_id:
+        query = query.filter(Relationship.source_id == source_id)
+    if target_id:
+        query = query.filter(Relationship.target_id == target_id)
+    
+    return query.limit(limit).all()
+
+
+@router.get("/types")
+def get_relationship_types(db: Session = Depends(get_db)):
+    """Retorna todos os tipos de relacionamento únicos"""
+    types = db.query(Relationship.type).distinct().all()
+    return [t[0] for t in types]
+
+
+@router.post("/", response_model=RelationshipResponse, status_code=201)
+def create_relationship(rel: RelationshipCreate, db: Session = Depends(get_db)):
+    """Cria um novo relacionamento entre entidades"""
+    # Verify both entities exist
+    source = db.query(Entity).filter(Entity.id == rel.source_id).first()
+    target = db.query(Entity).filter(Entity.id == rel.target_id).first()
+    
+    if not source:
+        raise HTTPException(status_code=404, detail="Source entity not found")
+    if not target:
+        raise HTTPException(status_code=404, detail="Target entity not found")
+    
+    db_rel = Relationship(**rel.model_dump())
+    db.add(db_rel)
+    db.commit()
+    db.refresh(db_rel)
+    return db_rel
+
+
+@router.delete("/{relationship_id}")
+def delete_relationship(relationship_id: str, db: Session = Depends(get_db)):
+    """Deleta um relacionamento"""
+    db_rel = db.query(Relationship).filter(Relationship.id == relationship_id).first()
+    if not db_rel:
+        raise HTTPException(status_code=404, detail="Relationship not found")
+    
+    db.delete(db_rel)
+    db.commit()
+    return {"message": "Relationship deleted"}

app/api/routes/search.py

@@ -0,0 +1,133 @@
+"""
+Search and Analytics Routes
+"""
+from fastapi import APIRouter, Depends, Query
+from sqlalchemy.orm import Session
+from sqlalchemy import or_, func
+from typing import Optional, List
+
+from app.core.database import get_db
+from app.models import Entity, Relationship, Event, Document
+from app.schemas import SearchResult, SystemStats, EntityResponse, EventResponse, DocumentResponse
+
+router = APIRouter(prefix="/search", tags=["Search"])
+
+
+@router.get("/", response_model=SearchResult)
+def global_search(
+    q: str = Query(..., min_length=2, description="Search query"),
+    types: Optional[str] = Query(None, description="Entity types (comma-separated)"),
+    limit: int = Query(default=20, le=100),
+    db: Session = Depends(get_db)
+):
+    """
+    Busca global em todas as entidades, eventos e documentos
+    """
+    search_term = f"%{q}%"
+    type_filter = types.split(",") if types else None
+    
+    # Search entities
+    entity_query = db.query(Entity).filter(
+        or_(
+            Entity.name.ilike(search_term),
+            Entity.description.ilike(search_term)
+        )
+    )
+    if type_filter:
+        entity_query = entity_query.filter(Entity.type.in_(type_filter))
+    entities = entity_query.limit(limit).all()
+    
+    # Search events
+    events = db.query(Event).filter(
+        or_(
+            Event.title.ilike(search_term),
+            Event.description.ilike(search_term)
+        )
+    ).limit(limit).all()
+    
+    # Search documents
+    documents = db.query(Document).filter(
+        or_(
+            Document.title.ilike(search_term),
+            Document.content.ilike(search_term)
+        )
+    ).limit(limit).all()
+    
+    return SearchResult(
+        entities=entities,
+        events=events,
+        documents=documents
+    )
+
+
+@router.get("/stats", response_model=SystemStats)
+def get_system_stats(db: Session = Depends(get_db)):
+    """
+    Retorna estatísticas gerais do sistema
+    Usado no dashboard do VANTAGE
+    """
+    total_entities = db.query(Entity).count()
+    total_relationships = db.query(Relationship).count()
+    total_events = db.query(Event).count()
+    total_documents = db.query(Document).count()
+    
+    # Count by entity type
+    type_counts = db.query(
+        Entity.type,
+        func.count(Entity.id)
+    ).group_by(Entity.type).all()
+    
+    entities_by_type = {t: c for t, c in type_counts}
+    
+    # Recent activity (last 10 entities created)
+    recent = db.query(Entity).order_by(Entity.created_at.desc()).limit(10).all()
+    recent_activity = [
+        {
+            "id": e.id,
+            "type": e.type,
+            "name": e.name,
+            "created_at": e.created_at.isoformat()
+        }
+        for e in recent
+    ]
+    
+    return SystemStats(
+        total_entities=total_entities,
+        total_relationships=total_relationships,
+        total_events=total_events,
+        total_documents=total_documents,
+        entities_by_type=entities_by_type,
+        recent_activity=recent_activity
+    )
+
+
+@router.get("/geo")
+def get_geo_data(
+    entity_type: Optional[str] = None,
+    db: Session = Depends(get_db)
+):
+    """
+    Retorna entidades com geolocalização
+    Usado para visualização em mapa
+    """
+    query = db.query(Entity).filter(
+        Entity.latitude.isnot(None),
+        Entity.longitude.isnot(None)
+    )
+    
+    if entity_type:
+        query = query.filter(Entity.type == entity_type)
+    
+    entities = query.all()
+    
+    return [
+        {
+            "id": e.id,
+            "type": e.type,
+            "name": e.name,
+            "lat": e.latitude,
+            "lng": e.longitude,
+            "properties": e.properties
+        }
+        for e in entities
+    ]

app/config.py

@@ -0,0 +1,37 @@
+"""
+Numidium Backend Configuration
+"""
+from pydantic_settings import BaseSettings
+from functools import lru_cache
+import os
+
+
+class Settings(BaseSettings):
+    """Application settings"""
+    
+    # App Info
+    app_name: str = "Numidium"
+    app_version: str = "0.1.0"
+    debug: bool = False
+    
+    # Database
+    database_url: str = "sqlite:///./data/numidium.db"
+    
+    # APIs (opcional - pode configurar depois)
+    newsapi_key: str = ""
+    
+    # CORS
+    cors_origins: list[str] = ["*"]
+    
+    class Config:
+        env_file = ".env"
+        env_file_encoding = "utf-8"
+
+
+@lru_cache()
+def get_settings() -> Settings:
+    """Get cached settings"""
+    return Settings()
+
+
+settings = get_settings()

app/core/__init__.py

@@ -0,0 +1,2 @@
+# Core module
+from app.core.database import get_db, init_db, Base

app/core/database.py

@@ -0,0 +1,38 @@
+"""
+Database configuration and session management
+"""
+from sqlalchemy import create_engine
+from sqlalchemy.ext.declarative import declarative_base
+from sqlalchemy.orm import sessionmaker
+import os
+
+from app.config import settings
+
+# Ensure data directory exists
+os.makedirs("data", exist_ok=True)
+
+# Create engine
+engine = create_engine(
+    settings.database_url,
+    connect_args={"check_same_thread": False}  # Needed for SQLite
+)
+
+# Session factory
+SessionLocal = sessionmaker(autocommit=False, autoflush=False, bind=engine)
+
+# Base class for models
+Base = declarative_base()
+
+
+def get_db():
+    """Dependency to get database session"""
+    db = SessionLocal()
+    try:
+        yield db
+    finally:
+        db.close()
+
+
+def init_db():
+    """Initialize database tables"""
+    Base.metadata.create_all(bind=engine)

app/main.py

@@ -0,0 +1,90 @@
+"""
+Numidium Backend - Main Application
+Plataforma de Inteligência e Análise de Dados
+"""
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+from contextlib import asynccontextmanager
+
+from app.config import settings
+from app.core.database import init_db
+from app.api.routes import entities, relationships, events, search, ingest
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    """Startup and shutdown events"""
+    # Startup: Initialize database
+    init_db()
+    print("🚀 Numidium Backend started!")
+    print(f"📊 Database: {settings.database_url}")
+    yield
+    # Shutdown
+    print("👋 Numidium Backend shutting down...")
+
+
+# Create FastAPI app
+app = FastAPI(
+    title="Numidium API",
+    description="""
+    ## 🔮 Sistema de Inteligência e Análise de Dados
+    
+    Backend do VANTAGE - Uma plataforma para:
+    - 📥 Ingestão de dados de múltiplas fontes (Wikipedia, News, Manual)
+    - 🔗 Mapeamento de conexões entre entidades
+    - 🗺️ Visualização geográfica
+    - 📊 Análise de grafos e relacionamentos
+    - 🔍 Busca global
+    """,
+    version=settings.app_version,
+    lifespan=lifespan
+)
+
+# CORS middleware
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=settings.cors_origins,
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+# Include routers
+app.include_router(entities.router, prefix="/api/v1")
+app.include_router(relationships.router, prefix="/api/v1")
+app.include_router(events.router, prefix="/api/v1")
+app.include_router(search.router, prefix="/api/v1")
+app.include_router(ingest.router, prefix="/api/v1")
+
+
+@app.get("/")
+def root():
+    """Root endpoint - API info"""
+    return {
+        "name": "Numidium",
+        "version": settings.app_version,
+        "status": "online",
+        "docs": "/docs",
+        "description": "Sistema de Inteligência e Análise de Dados"
+    }
+
+
+@app.get("/health")
+def health_check():
+    """Health check endpoint for HF Spaces"""
+    return {"status": "healthy"}
+
+
+@app.get("/api/v1")
+def api_info():
+    """API v1 info"""
+    return {
+        "version": "1.0.0",
+        "endpoints": {
+            "entities": "/api/v1/entities",
+            "relationships": "/api/v1/relationships",
+            "events": "/api/v1/events",
+            "search": "/api/v1/search",
+            "ingest": "/api/v1/ingest"
+        }
+    }

app/models/__init__.py

@@ -0,0 +1,2 @@
+# Models module
+from app.models.entity import Entity, Relationship, Event, Document

app/models/entity.py

@@ -0,0 +1,135 @@
+"""
+SQLAlchemy Models for Numidium
+"""
+from sqlalchemy import Column, String, Text, DateTime, Float, JSON, ForeignKey, Table
+from sqlalchemy.orm import relationship
+from datetime import datetime
+import uuid
+
+from app.core.database import Base
+
+
+def generate_uuid():
+    return str(uuid.uuid4())
+
+
+class Entity(Base):
+    """
+    Entidade - qualquer coisa rastreável no sistema
+    Pode ser: pessoa, organização, local, veículo, evento, documento, etc.
+    """
+    __tablename__ = "entities"
+    
+    id = Column(String(36), primary_key=True, default=generate_uuid)
+    type = Column(String(50), nullable=False, index=True)  # person, organization, location, etc
+    name = Column(String(255), nullable=False, index=True)
+    description = Column(Text, nullable=True)
+    properties = Column(JSON, default=dict)  # Dados flexíveis
+    
+    # Geolocalização (opcional)
+    latitude = Column(Float, nullable=True)
+    longitude = Column(Float, nullable=True)
+    
+    # Fonte do dado
+    source = Column(String(100), nullable=True)  # wikipedia, newsapi, manual, etc
+    source_url = Column(Text, nullable=True)
+    
+    # Timestamps
+    created_at = Column(DateTime, default=datetime.utcnow)
+    updated_at = Column(DateTime, default=datetime.utcnow, onupdate=datetime.utcnow)
+    
+    # Relacionamentos
+    outgoing_relationships = relationship(
+        "Relationship",
+        foreign_keys="Relationship.source_id",
+        back_populates="source_entity"
+    )
+    incoming_relationships = relationship(
+        "Relationship",
+        foreign_keys="Relationship.target_id",
+        back_populates="target_entity"
+    )
+
+
+class Relationship(Base):
+    """
+    Relacionamento entre duas entidades
+    Exemplos: works_for, knows, owns, located_at, participated_in
+    """
+    __tablename__ = "relationships"
+    
+    id = Column(String(36), primary_key=True, default=generate_uuid)
+    source_id = Column(String(36), ForeignKey("entities.id"), nullable=False)
+    target_id = Column(String(36), ForeignKey("entities.id"), nullable=False)
+    type = Column(String(50), nullable=False, index=True)  # works_for, knows, owns, etc
+    properties = Column(JSON, default=dict)
+    confidence = Column(Float, default=1.0)  # 0-1, quão certo estamos dessa conexão
+    
+    # Fonte
+    source = Column(String(100), nullable=True)
+    
+    # Timestamps
+    created_at = Column(DateTime, default=datetime.utcnow)
+    
+    # Relacionamentos
+    source_entity = relationship("Entity", foreign_keys=[source_id], back_populates="outgoing_relationships")
+    target_entity = relationship("Entity", foreign_keys=[target_id], back_populates="incoming_relationships")
+
+
+class Event(Base):
+    """
+    Evento - algo que aconteceu envolvendo entidades
+    """
+    __tablename__ = "events"
+    
+    id = Column(String(36), primary_key=True, default=generate_uuid)
+    type = Column(String(50), nullable=False, index=True)
+    title = Column(String(255), nullable=False)
+    description = Column(Text, nullable=True)
+    
+    # Quando aconteceu
+    event_date = Column(DateTime, nullable=True)
+    
+    # Onde aconteceu
+    location_name = Column(String(255), nullable=True)
+    latitude = Column(Float, nullable=True)
+    longitude = Column(Float, nullable=True)
+    
+    # Entidades envolvidas (armazenado como JSON array de IDs)
+    entity_ids = Column(JSON, default=list)
+    
+    # Fonte
+    source = Column(String(100), nullable=True)
+    source_url = Column(Text, nullable=True)
+    
+    # Metadados
+    properties = Column(JSON, default=dict)
+    
+    # Timestamps
+    created_at = Column(DateTime, default=datetime.utcnow)
+
+
+class Document(Base):
+    """
+    Documento - texto/arquivo para análise
+    """
+    __tablename__ = "documents"
+    
+    id = Column(String(36), primary_key=True, default=generate_uuid)
+    title = Column(String(255), nullable=False)
+    content = Column(Text, nullable=True)
+    summary = Column(Text, nullable=True)  # Resumo gerado por IA
+    
+    # Tipo de documento
+    doc_type = Column(String(50), default="text")  # text, news, report, etc
+    
+    # Entidades mencionadas (extraídas por NLP)
+    mentioned_entities = Column(JSON, default=list)
+    
+    # Fonte
+    source = Column(String(100), nullable=True)
+    source_url = Column(Text, nullable=True)
+    
+    # Timestamps
+    published_at = Column(DateTime, nullable=True)
+    created_at = Column(DateTime, default=datetime.utcnow)

app/schemas/__init__.py

@@ -0,0 +1,10 @@
+# Schemas module
+from app.schemas.schemas import (
+    EntityCreate, EntityUpdate, EntityResponse,
+    RelationshipCreate, RelationshipResponse,
+    EventCreate, EventResponse,
+    DocumentCreate, DocumentResponse,
+    GraphData, GraphNode, GraphEdge,
+    SearchQuery, SearchResult,
+    SystemStats
+)

app/schemas/schemas.py

@@ -0,0 +1,163 @@
+"""
+Pydantic Schemas for API validation
+"""
+from pydantic import BaseModel, Field
+from typing import Optional, List, Any
+from datetime import datetime
+
+
+# ========== Entity Schemas ==========
+
+class EntityBase(BaseModel):
+    type: str = Field(..., description="Tipo da entidade: person, organization, location, etc")
+    name: str = Field(..., description="Nome da entidade")
+    description: Optional[str] = None
+    properties: dict = Field(default_factory=dict)
+    latitude: Optional[float] = None
+    longitude: Optional[float] = None
+    source: Optional[str] = None
+    source_url: Optional[str] = None
+
+
+class EntityCreate(EntityBase):
+    pass
+
+
+class EntityUpdate(BaseModel):
+    type: Optional[str] = None
+    name: Optional[str] = None
+    description: Optional[str] = None
+    properties: Optional[dict] = None
+    latitude: Optional[float] = None
+    longitude: Optional[float] = None
+
+
+class EntityResponse(EntityBase):
+    id: str
+    created_at: datetime
+    updated_at: datetime
+    
+    class Config:
+        from_attributes = True
+
+
+# ========== Relationship Schemas ==========
+
+class RelationshipBase(BaseModel):
+    source_id: str
+    target_id: str
+    type: str = Field(..., description="Tipo: works_for, knows, owns, located_at, etc")
+    properties: dict = Field(default_factory=dict)
+    confidence: float = Field(default=1.0, ge=0, le=1)
+    source: Optional[str] = None
+
+
+class RelationshipCreate(RelationshipBase):
+    pass
+
+
+class RelationshipResponse(RelationshipBase):
+    id: str
+    created_at: datetime
+    
+    class Config:
+        from_attributes = True
+
+
+# ========== Event Schemas ==========
+
+class EventBase(BaseModel):
+    type: str
+    title: str
+    description: Optional[str] = None
+    event_date: Optional[datetime] = None
+    location_name: Optional[str] = None
+    latitude: Optional[float] = None
+    longitude: Optional[float] = None
+    entity_ids: List[str] = Field(default_factory=list)
+    source: Optional[str] = None
+    source_url: Optional[str] = None
+    properties: dict = Field(default_factory=dict)
+
+
+class EventCreate(EventBase):
+    pass
+
+
+class EventResponse(EventBase):
+    id: str
+    created_at: datetime
+    
+    class Config:
+        from_attributes = True
+
+
+# ========== Document Schemas ==========
+
+class DocumentBase(BaseModel):
+    title: str
+    content: Optional[str] = None
+    doc_type: str = "text"
+    source: Optional[str] = None
+    source_url: Optional[str] = None
+    published_at: Optional[datetime] = None
+
+
+class DocumentCreate(DocumentBase):
+    pass
+
+
+class DocumentResponse(DocumentBase):
+    id: str
+    summary: Optional[str] = None
+    mentioned_entities: List[str] = []
+    created_at: datetime
+    
+    class Config:
+        from_attributes = True
+
+
+# ========== Graph Schemas ==========
+
+class GraphNode(BaseModel):
+    id: str
+    type: str
+    name: str
+    properties: dict = {}
+
+
+class GraphEdge(BaseModel):
+    source: str
+    target: str
+    type: str
+    confidence: float = 1.0
+
+
+class GraphData(BaseModel):
+    nodes: List[GraphNode]
+    edges: List[GraphEdge]
+
+
+# ========== Search Schemas ==========
+
+class SearchQuery(BaseModel):
+    query: str
+    entity_types: Optional[List[str]] = None
+    limit: int = Field(default=20, le=100)
+
+
+class SearchResult(BaseModel):
+    entities: List[EntityResponse]
+    events: List[EventResponse]
+    documents: List[DocumentResponse]
+
+
+# ========== Stats Schemas ==========
+
+class SystemStats(BaseModel):
+    total_entities: int
+    total_relationships: int
+    total_events: int
+    total_documents: int
+    entities_by_type: dict
+    recent_activity: List[dict]

app/services/__init__.py

@@ -0,0 +1 @@
+# Services module

app/services/analysis/__init__.py

@@ -0,0 +1 @@
+# Analysis services

app/services/ingestion/__init__.py

@@ -0,0 +1,3 @@
+# Ingestion services
+from app.services.ingestion.wikipedia import wikipedia_scraper
+from app.services.ingestion.news import news_service

app/services/ingestion/news.py

@@ -0,0 +1,86 @@
+"""
+News API Client Service
+Usa RSS feeds públicos para não precisar de API key
+"""
+import feedparser
+import requests
+from typing import List, Dict
+from datetime import datetime
+import re
+
+
+class NewsService:
+    """Serviço para buscar notícias de fontes públicas via RSS"""
+    
+    # RSS feeds públicos brasileiros e internacionais
+    RSS_FEEDS = {
+        "g1": "https://g1.globo.com/rss/g1/",
+        "folha": "https://feeds.folha.uol.com.br/folha/rss/rss091.xml",
+        "bbc_brasil": "https://www.bbc.com/portuguese/articles/rss.xml",
+        "reuters": "https://www.reutersagency.com/feed/",
+        "google_news_br": "https://news.google.com/rss?hl=pt-BR&gl=BR&ceid=BR:pt-419"
+    }
+    
+    def fetch_feed(self, feed_url: str) -> List[Dict]:
+        """Busca artigos de um feed RSS"""
+        try:
+            feed = feedparser.parse(feed_url)
+            articles = []
+            
+            for entry in feed.entries[:20]:  # Limitar a 20 artigos
+                published = None
+                if hasattr(entry, 'published_parsed') and entry.published_parsed:
+                    published = datetime(*entry.published_parsed[:6])
+                
+                articles.append({
+                    "title": entry.get("title", ""),
+                    "description": self._clean_html(entry.get("summary", "")),
+                    "url": entry.get("link", ""),
+                    "published_at": published,
+                    "source": feed.feed.get("title", "Unknown")
+                })
+            
+            return articles
+        except Exception as e:
+            print(f"Error fetching feed {feed_url}: {e}")
+            return []
+    
+    def fetch_all_feeds(self) -> List[Dict]:
+        """Busca artigos de todos os feeds configurados"""
+        all_articles = []
+        for name, url in self.RSS_FEEDS.items():
+            articles = self.fetch_feed(url)
+            for article in articles:
+                article["feed_name"] = name
+            all_articles.extend(articles)
+        return all_articles
+    
+    def search_news(self, query: str) -> List[Dict]:
+        """
+        Busca notícias pelo Google News RSS
+        """
+        # Google News RSS search
+        search_url = f"https://news.google.com/rss/search?q={query}&hl=pt-BR&gl=BR&ceid=BR:pt-419"
+        return self.fetch_feed(search_url)
+    
+    def _clean_html(self, text: str) -> str:
+        """Remove HTML tags do texto"""
+        clean = re.compile('<.*?>')
+        return re.sub(clean, '', text)
+    
+    def to_document(self, article: Dict) -> Dict:
+        """
+        Converte um artigo de notícia para o formato Document
+        """
+        return {
+            "title": article["title"],
+            "content": article.get("description", ""),
+            "doc_type": "news",
+            "source": article.get("source", "news"),
+            "source_url": article.get("url"),
+            "published_at": article.get("published_at")
+        }
+
+
+# Singleton instance
+news_service = NewsService()

app/services/ingestion/wikipedia.py

@@ -0,0 +1,186 @@
+"""
+Wikipedia Scraper Service
+"""
+import requests
+from bs4 import BeautifulSoup
+from typing import Optional, Dict, List
+import re
+
+
+class WikipediaScraper:
+    """Scraper para extrair dados da Wikipedia"""
+    
+    BASE_URL = "https://pt.wikipedia.org"
+    API_URL = "https://pt.wikipedia.org/w/api.php"
+    
+    def search(self, query: str, limit: int = 10) -> List[Dict]:
+        """
+        Busca artigos na Wikipedia
+        """
+        params = {
+            "action": "query",
+            "list": "search",
+            "srsearch": query,
+            "srlimit": limit,
+            "format": "json"
+        }
+        
+        response = requests.get(self.API_URL, params=params)
+        data = response.json()
+        
+        results = []
+        for item in data.get("query", {}).get("search", []):
+            results.append({
+                "title": item["title"],
+                "snippet": BeautifulSoup(item["snippet"], "html.parser").get_text(),
+                "pageid": item["pageid"]
+            })
+        
+        return results
+    
+    def get_article(self, title: str) -> Optional[Dict]:
+        """
+        Busca informações completas de um artigo
+        """
+        params = {
+            "action": "query",
+            "titles": title,
+            "prop": "extracts|pageimages|coordinates|categories",
+            "exintro": True,
+            "explaintext": True,
+            "pithumbsize": 300,
+            "format": "json"
+        }
+        
+        response = requests.get(self.API_URL, params=params)
+        data = response.json()
+        
+        pages = data.get("query", {}).get("pages", {})
+        for page_id, page in pages.items():
+            if page_id == "-1":
+                return None
+            
+            result = {
+                "title": page.get("title"),
+                "extract": page.get("extract"),
+                "pageid": page.get("pageid"),
+                "url": f"{self.BASE_URL}/wiki/{page.get('title', '').replace(' ', '_')}",
+                "thumbnail": page.get("thumbnail", {}).get("source"),
+                "categories": [c["title"].replace("Categoria:", "") 
+                              for c in page.get("categories", [])]
+            }
+            
+            # Coordenadas se disponíveis
+            if "coordinates" in page:
+                coords = page["coordinates"][0]
+                result["latitude"] = coords.get("lat")
+                result["longitude"] = coords.get("lon")
+            
+            return result
+        
+        return None
+    
+    def get_infobox(self, title: str) -> Dict:
+        """
+        Tenta extrair dados estruturados do infobox de um artigo
+        """
+        url = f"{self.BASE_URL}/wiki/{title.replace(' ', '_')}"
+        response = requests.get(url)
+        soup = BeautifulSoup(response.text, "lxml")
+        
+        infobox = soup.find("table", class_="infobox")
+        if not infobox:
+            return {}
+        
+        data = {}
+        for row in infobox.find_all("tr"):
+            header = row.find("th")
+            cell = row.find("td")
+            if header and cell:
+                key = header.get_text(strip=True)
+                value = cell.get_text(strip=True)
+                # Clean up the value
+                value = re.sub(r'\[\d+\]', '', value)  # Remove references
+                data[key] = value
+        
+        return data
+    
+    def scrape_person(self, name: str) -> Optional[Dict]:
+        """
+        Scrape dados de uma pessoa da Wikipedia
+        Retorna dados formatados para criar uma Entity
+        """
+        article = self.get_article(name)
+        if not article:
+            return None
+        
+        infobox = self.get_infobox(name)
+        
+        return {
+            "type": "person",
+            "name": article["title"],
+            "description": article.get("extract"),
+            "source": "wikipedia",
+            "source_url": article["url"],
+            "properties": {
+                "thumbnail": article.get("thumbnail"),
+                "categories": article.get("categories", []),
+                **infobox
+            },
+            "latitude": article.get("latitude"),
+            "longitude": article.get("longitude")
+        }
+    
+    def scrape_organization(self, name: str) -> Optional[Dict]:
+        """
+        Scrape dados de uma organização da Wikipedia
+        """
+        article = self.get_article(name)
+        if not article:
+            return None
+        
+        infobox = self.get_infobox(name)
+        
+        return {
+            "type": "organization",
+            "name": article["title"],
+            "description": article.get("extract"),
+            "source": "wikipedia",
+            "source_url": article["url"],
+            "properties": {
+                "thumbnail": article.get("thumbnail"),
+                "categories": article.get("categories", []),
+                **infobox
+            },
+            "latitude": article.get("latitude"),
+            "longitude": article.get("longitude")
+        }
+    
+    def scrape_location(self, name: str) -> Optional[Dict]:
+        """
+        Scrape dados de um local da Wikipedia
+        """
+        article = self.get_article(name)
+        if not article:
+            return None
+        
+        infobox = self.get_infobox(name)
+        
+        return {
+            "type": "location",
+            "name": article["title"],
+            "description": article.get("extract"),
+            "source": "wikipedia",
+            "source_url": article["url"],
+            "properties": {
+                "thumbnail": article.get("thumbnail"),
+                "categories": article.get("categories", []),
+                **infobox
+            },
+            "latitude": article.get("latitude"),
+            "longitude": article.get("longitude")
+        }
+
+
+# Singleton instance
+wikipedia_scraper = WikipediaScraper()

requirements.txt

@@ -0,0 +1,12 @@
+fastapi==0.104.1
+uvicorn[standard]==0.24.0
+sqlalchemy==2.0.23
+pydantic==2.5.2
+pydantic-settings==2.1.0
+requests==2.31.0
+beautifulsoup4==4.12.2
+lxml==4.9.3
+httpx==0.25.2
+python-multipart==0.0.6
+aiohttp==3.9.1
+feedparser==6.0.10