Delete app
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- app/__init__.py +0 -1
- app/__pycache__/__init__.cpython-311.pyc +0 -0
- app/__pycache__/config.cpython-311.pyc +0 -0
- app/api/__init__.py +0 -1
- app/api/__pycache__/__init__.cpython-311.pyc +0 -0
- app/api/__pycache__/deps.cpython-311.pyc +0 -0
- app/api/deps.py +0 -35
- app/api/routes/__init__.py +0 -2
- app/api/routes/__pycache__/__init__.cpython-311.pyc +0 -0
- app/api/routes/__pycache__/entities.cpython-311.pyc +0 -0
- app/api/routes/__pycache__/events.cpython-311.pyc +0 -0
- app/api/routes/__pycache__/ingest.cpython-311.pyc +0 -0
- app/api/routes/__pycache__/investigate.cpython-311.pyc +0 -0
- app/api/routes/__pycache__/relationships.cpython-311.pyc +0 -0
- app/api/routes/__pycache__/search.cpython-311.pyc +0 -0
- app/api/routes/aethermap.py +0 -307
- app/api/routes/analyze.py +0 -309
- app/api/routes/chat.py +0 -63
- app/api/routes/dados_publicos.py +0 -155
- app/api/routes/entities.py +0 -353
- app/api/routes/events.py +0 -113
- app/api/routes/graph.py +0 -173
- app/api/routes/ingest.py +0 -341
- app/api/routes/investigate.py +0 -207
- app/api/routes/projects.py +0 -135
- app/api/routes/relationships.py +0 -76
- app/api/routes/research.py +0 -158
- app/api/routes/search.py +0 -126
- app/api/routes/session.py +0 -44
- app/api/routes/timeline.py +0 -165
- app/config.py +0 -47
- app/core/__init__.py +0 -2
- app/core/__pycache__/__init__.cpython-311.pyc +0 -0
- app/core/__pycache__/database.cpython-311.pyc +0 -0
- app/core/database.py +0 -115
- app/main.py +0 -99
- app/models/__init__.py +0 -3
- app/models/__pycache__/__init__.cpython-311.pyc +0 -0
- app/models/__pycache__/entity.cpython-311.pyc +0 -0
- app/models/__pycache__/project.cpython-311.pyc +0 -0
- app/models/entity.py +0 -143
- app/models/project.py +0 -29
- app/schemas/__init__.py +0 -10
- app/schemas/__pycache__/__init__.cpython-311.pyc +0 -0
- app/schemas/__pycache__/schemas.cpython-311.pyc +0 -0
- app/schemas/schemas.py +0 -163
- app/services/__init__.py +0 -1
- app/services/__pycache__/__init__.cpython-311.pyc +0 -0
- app/services/__pycache__/brazil_apis.cpython-311.pyc +0 -0
- app/services/__pycache__/geocoding.cpython-311.pyc +0 -0
app/__init__.py
DELETED
|
@@ -1 +0,0 @@
|
|
| 1 |
-
# Numidium Backend App
|
|
|
|
|
|
app/__pycache__/__init__.cpython-311.pyc
DELETED
|
Binary file (156 Bytes)
|
|
|
app/__pycache__/config.cpython-311.pyc
DELETED
|
Binary file (1.76 kB)
|
|
|
app/api/__init__.py
DELETED
|
@@ -1 +0,0 @@
|
|
| 1 |
-
# API module
|
|
|
|
|
|
app/api/__pycache__/__init__.cpython-311.pyc
DELETED
|
Binary file (162 Bytes)
|
|
|
app/api/__pycache__/deps.cpython-311.pyc
DELETED
|
Binary file (1.64 kB)
|
|
|
app/api/deps.py
DELETED
|
@@ -1,35 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
API dependencies.
|
| 3 |
-
"""
|
| 4 |
-
from typing import Generator, Optional
|
| 5 |
-
|
| 6 |
-
from fastapi import Cookie, Header
|
| 7 |
-
from sqlalchemy.orm import Session
|
| 8 |
-
|
| 9 |
-
from app.core.database import get_db_for_session, get_default_session
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
def get_session_id(
|
| 13 |
-
x_session_id: Optional[str] = Header(None),
|
| 14 |
-
numidium_session: Optional[str] = Cookie(None)
|
| 15 |
-
) -> Optional[str]:
|
| 16 |
-
"""Return the session id from header or cookie."""
|
| 17 |
-
return x_session_id or numidium_session
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
def get_scoped_db(
|
| 21 |
-
x_session_id: Optional[str] = Header(None),
|
| 22 |
-
numidium_session: Optional[str] = Cookie(None)
|
| 23 |
-
) -> Generator[Session, None, None]:
|
| 24 |
-
"""
|
| 25 |
-
Provide a session-scoped DB if available, otherwise the default DB.
|
| 26 |
-
"""
|
| 27 |
-
session_id = x_session_id or numidium_session
|
| 28 |
-
if session_id:
|
| 29 |
-
db = get_db_for_session(session_id)
|
| 30 |
-
else:
|
| 31 |
-
db = get_default_session()
|
| 32 |
-
try:
|
| 33 |
-
yield db
|
| 34 |
-
finally:
|
| 35 |
-
db.close()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/__init__.py
DELETED
|
@@ -1,2 +0,0 @@
|
|
| 1 |
-
# API Routes module
|
| 2 |
-
from app.api.routes import entities, relationships, events, search, ingest
|
|
|
|
|
|
|
|
|
app/api/routes/__pycache__/__init__.cpython-311.pyc
DELETED
|
Binary file (329 Bytes)
|
|
|
app/api/routes/__pycache__/entities.cpython-311.pyc
DELETED
|
Binary file (18.5 kB)
|
|
|
app/api/routes/__pycache__/events.cpython-311.pyc
DELETED
|
Binary file (7.14 kB)
|
|
|
app/api/routes/__pycache__/ingest.cpython-311.pyc
DELETED
|
Binary file (16 kB)
|
|
|
app/api/routes/__pycache__/investigate.cpython-311.pyc
DELETED
|
Binary file (10.1 kB)
|
|
|
app/api/routes/__pycache__/relationships.cpython-311.pyc
DELETED
|
Binary file (5.04 kB)
|
|
|
app/api/routes/__pycache__/search.cpython-311.pyc
DELETED
|
Binary file (7 kB)
|
|
|
app/api/routes/aethermap.py
DELETED
|
@@ -1,307 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
AetherMap Routes - Document Mapping & Semantic Search
|
| 3 |
-
Integrates with AetherMap API for document clustering, NER, and semantic search.
|
| 4 |
-
"""
|
| 5 |
-
from fastapi import APIRouter, HTTPException, UploadFile, File, Form, Depends
|
| 6 |
-
from pydantic import BaseModel, Field
|
| 7 |
-
from typing import Optional, List, Dict, Any
|
| 8 |
-
from sqlalchemy.orm import Session
|
| 9 |
-
import io
|
| 10 |
-
|
| 11 |
-
from app.api.deps import get_scoped_db
|
| 12 |
-
from app.services.aethermap_client import aethermap, ProcessResult, SearchResult, EntityGraphResult
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
router = APIRouter()
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
# ============================================================================
|
| 19 |
-
# Request/Response Models
|
| 20 |
-
# ============================================================================
|
| 21 |
-
|
| 22 |
-
class IndexDocumentsRequest(BaseModel):
|
| 23 |
-
"""Request to index documents from text list"""
|
| 24 |
-
documents: List[str] = Field(..., description="Lista de textos para indexar")
|
| 25 |
-
fast_mode: bool = Field(True, description="Modo rápido (PCA) ou preciso (UMAP)")
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
class IndexEntitiesRequest(BaseModel):
|
| 29 |
-
"""Request to index entities from NUMIDIUM database"""
|
| 30 |
-
entity_types: Optional[List[str]] = Field(None, description="Filtrar por tipos de entidade")
|
| 31 |
-
limit: int = Field(500, description="Limite de entidades")
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
class SemanticSearchRequest(BaseModel):
|
| 35 |
-
"""Request for semantic search"""
|
| 36 |
-
query: str = Field(..., description="Termo de busca")
|
| 37 |
-
turbo_mode: bool = Field(True, description="Modo turbo (mais rápido)")
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
class IndexResponse(BaseModel):
|
| 41 |
-
"""Response from indexing"""
|
| 42 |
-
job_id: str
|
| 43 |
-
num_documents: int
|
| 44 |
-
num_clusters: int
|
| 45 |
-
num_noise: int
|
| 46 |
-
metrics: Dict[str, Any] = {}
|
| 47 |
-
cluster_analysis: Dict[str, Any] = {}
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
class SearchResponse(BaseModel):
|
| 51 |
-
"""Response from search"""
|
| 52 |
-
summary: str
|
| 53 |
-
results: List[Dict[str, Any]] = []
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
class EntityGraphResponse(BaseModel):
|
| 57 |
-
"""Response from NER extraction"""
|
| 58 |
-
hubs: List[Dict[str, Any]] = []
|
| 59 |
-
insights: Dict[str, Any] = {}
|
| 60 |
-
node_count: int = 0
|
| 61 |
-
edge_count: int = 0
|
| 62 |
-
|
| 63 |
-
|
| 64 |
-
class StatusResponse(BaseModel):
|
| 65 |
-
"""AetherMap status"""
|
| 66 |
-
connected: bool
|
| 67 |
-
job_id: Optional[str] = None
|
| 68 |
-
documents_indexed: int = 0
|
| 69 |
-
|
| 70 |
-
|
| 71 |
-
# ============================================================================
|
| 72 |
-
# Endpoints
|
| 73 |
-
# ============================================================================
|
| 74 |
-
|
| 75 |
-
@router.get("/status", response_model=StatusResponse)
|
| 76 |
-
async def get_status():
|
| 77 |
-
"""
|
| 78 |
-
Get AetherMap connection status.
|
| 79 |
-
"""
|
| 80 |
-
return StatusResponse(
|
| 81 |
-
connected=True,
|
| 82 |
-
job_id=aethermap.current_job_id,
|
| 83 |
-
documents_indexed=0 # TODO: track this
|
| 84 |
-
)
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
@router.post("/index", response_model=IndexResponse)
|
| 88 |
-
async def index_documents(request: IndexDocumentsRequest):
|
| 89 |
-
"""
|
| 90 |
-
Index a list of documents for semantic search.
|
| 91 |
-
|
| 92 |
-
The documents will be:
|
| 93 |
-
- Embedded using sentence transformers
|
| 94 |
-
- Clustered using HDBSCAN
|
| 95 |
-
- Indexed in FAISS + BM25 for hybrid search
|
| 96 |
-
"""
|
| 97 |
-
try:
|
| 98 |
-
if not request.documents:
|
| 99 |
-
raise HTTPException(status_code=400, detail="Nenhum documento fornecido")
|
| 100 |
-
|
| 101 |
-
result = await aethermap.process_documents(
|
| 102 |
-
texts=request.documents,
|
| 103 |
-
fast_mode=request.fast_mode
|
| 104 |
-
)
|
| 105 |
-
|
| 106 |
-
return IndexResponse(
|
| 107 |
-
job_id=result.job_id,
|
| 108 |
-
num_documents=result.num_documents,
|
| 109 |
-
num_clusters=result.num_clusters,
|
| 110 |
-
num_noise=result.num_noise,
|
| 111 |
-
metrics=result.metrics,
|
| 112 |
-
cluster_analysis=result.cluster_analysis
|
| 113 |
-
)
|
| 114 |
-
|
| 115 |
-
except Exception as e:
|
| 116 |
-
raise HTTPException(status_code=500, detail=str(e))
|
| 117 |
-
|
| 118 |
-
|
| 119 |
-
@router.post("/index-entities", response_model=IndexResponse)
|
| 120 |
-
async def index_entities(
|
| 121 |
-
request: IndexEntitiesRequest,
|
| 122 |
-
db: Session = Depends(get_scoped_db)
|
| 123 |
-
):
|
| 124 |
-
"""
|
| 125 |
-
Index entities from NUMIDIUM database.
|
| 126 |
-
|
| 127 |
-
Collects entity names and descriptions, sends to AetherMap for processing.
|
| 128 |
-
"""
|
| 129 |
-
from app.models.entity import Entity
|
| 130 |
-
|
| 131 |
-
try:
|
| 132 |
-
query = db.query(Entity)
|
| 133 |
-
|
| 134 |
-
if request.entity_types:
|
| 135 |
-
query = query.filter(Entity.type.in_(request.entity_types))
|
| 136 |
-
|
| 137 |
-
entities = query.limit(request.limit).all()
|
| 138 |
-
|
| 139 |
-
if not entities:
|
| 140 |
-
raise HTTPException(status_code=404, detail="Nenhuma entidade encontrada")
|
| 141 |
-
|
| 142 |
-
# Build text representations
|
| 143 |
-
documents = []
|
| 144 |
-
for e in entities:
|
| 145 |
-
text = f"{e.name} ({e.type})"
|
| 146 |
-
if e.description:
|
| 147 |
-
text += f": {e.description[:1000]}"
|
| 148 |
-
documents.append(text)
|
| 149 |
-
|
| 150 |
-
result = await aethermap.process_documents(
|
| 151 |
-
texts=documents,
|
| 152 |
-
fast_mode=request.fast_mode if hasattr(request, 'fast_mode') else True
|
| 153 |
-
)
|
| 154 |
-
|
| 155 |
-
return IndexResponse(
|
| 156 |
-
job_id=result.job_id,
|
| 157 |
-
num_documents=result.num_documents,
|
| 158 |
-
num_clusters=result.num_clusters,
|
| 159 |
-
num_noise=result.num_noise,
|
| 160 |
-
metrics=result.metrics,
|
| 161 |
-
cluster_analysis=result.cluster_analysis
|
| 162 |
-
)
|
| 163 |
-
|
| 164 |
-
except HTTPException:
|
| 165 |
-
raise
|
| 166 |
-
except Exception as e:
|
| 167 |
-
raise HTTPException(status_code=500, detail=str(e))
|
| 168 |
-
|
| 169 |
-
|
| 170 |
-
@router.post("/upload", response_model=IndexResponse)
|
| 171 |
-
async def upload_documents(
|
| 172 |
-
file: UploadFile = File(...),
|
| 173 |
-
fast_mode: bool = Form(True)
|
| 174 |
-
):
|
| 175 |
-
"""
|
| 176 |
-
Upload a file (TXT or CSV) for indexing.
|
| 177 |
-
|
| 178 |
-
- TXT: One document per line
|
| 179 |
-
- CSV: Will use first text column found
|
| 180 |
-
"""
|
| 181 |
-
try:
|
| 182 |
-
content = await file.read()
|
| 183 |
-
text = content.decode('utf-8', errors='ignore')
|
| 184 |
-
|
| 185 |
-
# Split by lines for TXT
|
| 186 |
-
documents = [line.strip() for line in text.splitlines() if line.strip()]
|
| 187 |
-
|
| 188 |
-
if not documents:
|
| 189 |
-
raise HTTPException(status_code=400, detail="Arquivo vazio ou sem texto válido")
|
| 190 |
-
|
| 191 |
-
result = await aethermap.process_documents(
|
| 192 |
-
texts=documents,
|
| 193 |
-
fast_mode=fast_mode
|
| 194 |
-
)
|
| 195 |
-
|
| 196 |
-
return IndexResponse(
|
| 197 |
-
job_id=result.job_id,
|
| 198 |
-
num_documents=result.num_documents,
|
| 199 |
-
num_clusters=result.num_clusters,
|
| 200 |
-
num_noise=result.num_noise,
|
| 201 |
-
metrics=result.metrics,
|
| 202 |
-
cluster_analysis=result.cluster_analysis
|
| 203 |
-
)
|
| 204 |
-
|
| 205 |
-
except HTTPException:
|
| 206 |
-
raise
|
| 207 |
-
except Exception as e:
|
| 208 |
-
raise HTTPException(status_code=500, detail=str(e))
|
| 209 |
-
|
| 210 |
-
|
| 211 |
-
@router.post("/search", response_model=SearchResponse)
|
| 212 |
-
async def semantic_search(request: SemanticSearchRequest):
|
| 213 |
-
"""
|
| 214 |
-
Semantic search in indexed documents.
|
| 215 |
-
|
| 216 |
-
Uses hybrid RAG (FAISS + BM25 + reranking + LLM).
|
| 217 |
-
Returns a summary answering the query with citations.
|
| 218 |
-
"""
|
| 219 |
-
try:
|
| 220 |
-
if not aethermap.current_job_id:
|
| 221 |
-
raise HTTPException(status_code=400, detail="Nenhum documento indexado. Use /index primeiro.")
|
| 222 |
-
|
| 223 |
-
result = await aethermap.semantic_search(
|
| 224 |
-
query=request.query,
|
| 225 |
-
turbo_mode=request.turbo_mode
|
| 226 |
-
)
|
| 227 |
-
|
| 228 |
-
return SearchResponse(
|
| 229 |
-
summary=result.summary,
|
| 230 |
-
results=result.results
|
| 231 |
-
)
|
| 232 |
-
|
| 233 |
-
except HTTPException:
|
| 234 |
-
raise
|
| 235 |
-
except Exception as e:
|
| 236 |
-
raise HTTPException(status_code=500, detail=str(e))
|
| 237 |
-
|
| 238 |
-
|
| 239 |
-
@router.post("/entities", response_model=EntityGraphResponse)
|
| 240 |
-
async def extract_entities():
|
| 241 |
-
"""
|
| 242 |
-
Extract named entities (NER) from indexed documents.
|
| 243 |
-
|
| 244 |
-
Returns:
|
| 245 |
-
- Hub entities (most connected)
|
| 246 |
-
- Relationship insights
|
| 247 |
-
- Graph metrics
|
| 248 |
-
"""
|
| 249 |
-
try:
|
| 250 |
-
if not aethermap.current_job_id:
|
| 251 |
-
raise HTTPException(status_code=400, detail="Nenhum documento indexado. Use /index primeiro.")
|
| 252 |
-
|
| 253 |
-
result = await aethermap.extract_entities()
|
| 254 |
-
|
| 255 |
-
return EntityGraphResponse(
|
| 256 |
-
hubs=result.hubs,
|
| 257 |
-
insights=result.insights,
|
| 258 |
-
node_count=len(result.nodes),
|
| 259 |
-
edge_count=len(result.edges)
|
| 260 |
-
)
|
| 261 |
-
|
| 262 |
-
except HTTPException:
|
| 263 |
-
raise
|
| 264 |
-
except Exception as e:
|
| 265 |
-
raise HTTPException(status_code=500, detail=str(e))
|
| 266 |
-
|
| 267 |
-
|
| 268 |
-
@router.post("/analyze")
|
| 269 |
-
async def analyze_graph():
|
| 270 |
-
"""
|
| 271 |
-
Analyze entity graph using LLM.
|
| 272 |
-
|
| 273 |
-
Returns semantic insights about relationships and patterns.
|
| 274 |
-
"""
|
| 275 |
-
try:
|
| 276 |
-
if not aethermap.current_job_id:
|
| 277 |
-
raise HTTPException(status_code=400, detail="Nenhum documento indexado. Use /index primeiro.")
|
| 278 |
-
|
| 279 |
-
result = await aethermap.analyze_graph()
|
| 280 |
-
|
| 281 |
-
return {
|
| 282 |
-
"analysis": result.analysis,
|
| 283 |
-
"key_entities": result.key_entities,
|
| 284 |
-
"relationships": result.relationships
|
| 285 |
-
}
|
| 286 |
-
|
| 287 |
-
except HTTPException:
|
| 288 |
-
raise
|
| 289 |
-
except Exception as e:
|
| 290 |
-
raise HTTPException(status_code=500, detail=str(e))
|
| 291 |
-
|
| 292 |
-
|
| 293 |
-
@router.post("/describe-clusters")
|
| 294 |
-
async def describe_clusters():
|
| 295 |
-
"""
|
| 296 |
-
Get LLM descriptions for each cluster found.
|
| 297 |
-
"""
|
| 298 |
-
try:
|
| 299 |
-
if not aethermap.current_job_id:
|
| 300 |
-
raise HTTPException(status_code=400, detail="Nenhum documento indexado. Use /index primeiro.")
|
| 301 |
-
|
| 302 |
-
result = await aethermap.describe_clusters()
|
| 303 |
-
|
| 304 |
-
return result
|
| 305 |
-
|
| 306 |
-
except Exception as e:
|
| 307 |
-
raise HTTPException(status_code=500, detail=str(e))
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/analyze.py
DELETED
|
@@ -1,309 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Analyze API Routes - LLM-based text analysis
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, Depends, HTTPException
|
| 5 |
-
from pydantic import BaseModel, Field
|
| 6 |
-
from typing import Optional, List
|
| 7 |
-
from sqlalchemy.orm import Session
|
| 8 |
-
import traceback
|
| 9 |
-
|
| 10 |
-
from app.api.deps import get_scoped_db
|
| 11 |
-
from app.services.nlp import entity_extractor
|
| 12 |
-
from app.services.geocoding import geocode
|
| 13 |
-
from app.models.entity import Entity, Relationship, Event
|
| 14 |
-
from app.config import settings
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
router = APIRouter(prefix="/analyze", tags=["Analysis"])
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
class AnalyzeRequest(BaseModel):
|
| 21 |
-
"""Request model for text analysis"""
|
| 22 |
-
text: str = Field(..., min_length=10, description="Text to analyze")
|
| 23 |
-
auto_create: bool = Field(default=False, description="Auto-create extracted entities in database")
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
class ExtractedEntityResponse(BaseModel):
|
| 27 |
-
"""Response model for an extracted entity"""
|
| 28 |
-
name: str
|
| 29 |
-
type: str
|
| 30 |
-
role: Optional[str] = None
|
| 31 |
-
aliases: Optional[List[str]] = None
|
| 32 |
-
description: Optional[str] = None
|
| 33 |
-
created: bool = False # Whether it was created in DB
|
| 34 |
-
entity_id: Optional[str] = None # DB ID if created
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
class ExtractedRelationshipResponse(BaseModel):
|
| 38 |
-
"""Response model for an extracted relationship"""
|
| 39 |
-
source: str
|
| 40 |
-
target: str
|
| 41 |
-
relationship_type: str
|
| 42 |
-
context: Optional[str] = None
|
| 43 |
-
created: bool = False
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
class ExtractedEventResponse(BaseModel):
|
| 47 |
-
"""Response model for an extracted event"""
|
| 48 |
-
description: str
|
| 49 |
-
event_type: Optional[str] = None
|
| 50 |
-
date: Optional[str] = None
|
| 51 |
-
location: Optional[str] = None
|
| 52 |
-
participants: Optional[List[str]] = None
|
| 53 |
-
created: bool = False
|
| 54 |
-
event_id: Optional[str] = None
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
class AnalyzeResponse(BaseModel):
|
| 58 |
-
"""Response model for analysis"""
|
| 59 |
-
entities: List[ExtractedEntityResponse]
|
| 60 |
-
relationships: List[ExtractedRelationshipResponse]
|
| 61 |
-
events: List[ExtractedEventResponse]
|
| 62 |
-
stats: dict
|
| 63 |
-
|
| 64 |
-
|
| 65 |
-
@router.post("", response_model=AnalyzeResponse)
|
| 66 |
-
async def analyze_text(request: AnalyzeRequest, db: Session = Depends(get_scoped_db)):
|
| 67 |
-
"""
|
| 68 |
-
Analyze text using LLM to extract entities, relationships, and events.
|
| 69 |
-
|
| 70 |
-
Uses Cerebras API with Qwen 3 235B for intelligent extraction.
|
| 71 |
-
|
| 72 |
-
Args:
|
| 73 |
-
text: Text to analyze (min 10 characters)
|
| 74 |
-
auto_create: If true, automatically creates entities in the database
|
| 75 |
-
|
| 76 |
-
Returns:
|
| 77 |
-
Extracted entities, relationships, events, and statistics
|
| 78 |
-
"""
|
| 79 |
-
try:
|
| 80 |
-
# Extract using LLM
|
| 81 |
-
result = await entity_extractor.extract(request.text)
|
| 82 |
-
|
| 83 |
-
# Prepare response
|
| 84 |
-
entities_response = []
|
| 85 |
-
relationships_response = []
|
| 86 |
-
events_response = []
|
| 87 |
-
|
| 88 |
-
created_entities = 0
|
| 89 |
-
created_relationships = 0
|
| 90 |
-
created_events = 0
|
| 91 |
-
|
| 92 |
-
# Helper function to parse date strings
|
| 93 |
-
def parse_date(date_str):
|
| 94 |
-
if not date_str:
|
| 95 |
-
return None
|
| 96 |
-
from datetime import datetime
|
| 97 |
-
try:
|
| 98 |
-
# Try YYYY-MM-DD format
|
| 99 |
-
return datetime.strptime(date_str[:10], "%Y-%m-%d")
|
| 100 |
-
except:
|
| 101 |
-
try:
|
| 102 |
-
# Try YYYY format
|
| 103 |
-
return datetime.strptime(date_str[:4], "%Y")
|
| 104 |
-
except:
|
| 105 |
-
return None
|
| 106 |
-
|
| 107 |
-
# Process entities
|
| 108 |
-
for entity in result.entities:
|
| 109 |
-
entity_data = ExtractedEntityResponse(
|
| 110 |
-
name=entity.name,
|
| 111 |
-
type=entity.type,
|
| 112 |
-
role=entity.role,
|
| 113 |
-
aliases=entity.aliases,
|
| 114 |
-
description=entity.description,
|
| 115 |
-
created=False
|
| 116 |
-
)
|
| 117 |
-
|
| 118 |
-
if request.auto_create and entity.name:
|
| 119 |
-
# Check if entity already exists
|
| 120 |
-
existing = db.query(Entity).filter(
|
| 121 |
-
Entity.name.ilike(f"%{entity.name}%")
|
| 122 |
-
).first()
|
| 123 |
-
|
| 124 |
-
if not existing:
|
| 125 |
-
# Get coordinates for location entities
|
| 126 |
-
lat, lng = None, None
|
| 127 |
-
if entity.type == "location":
|
| 128 |
-
coords = await geocode(entity.name)
|
| 129 |
-
if coords:
|
| 130 |
-
lat, lng = coords
|
| 131 |
-
|
| 132 |
-
# Parse event_date if available
|
| 133 |
-
event_date = parse_date(getattr(entity, 'event_date', None))
|
| 134 |
-
|
| 135 |
-
# Create new entity
|
| 136 |
-
new_entity = Entity(
|
| 137 |
-
name=entity.name,
|
| 138 |
-
type=entity.type if entity.type in ["person", "organization", "location", "event"] else "person",
|
| 139 |
-
description=entity.description or entity.role or "",
|
| 140 |
-
source="llm_extraction",
|
| 141 |
-
latitude=lat,
|
| 142 |
-
longitude=lng,
|
| 143 |
-
event_date=event_date,
|
| 144 |
-
properties={"role": entity.role, "aliases": entity.aliases}
|
| 145 |
-
)
|
| 146 |
-
db.add(new_entity)
|
| 147 |
-
db.commit()
|
| 148 |
-
db.refresh(new_entity)
|
| 149 |
-
|
| 150 |
-
entity_data.created = True
|
| 151 |
-
entity_data.entity_id = new_entity.id
|
| 152 |
-
created_entities += 1
|
| 153 |
-
else:
|
| 154 |
-
entity_data.entity_id = existing.id
|
| 155 |
-
|
| 156 |
-
entities_response.append(entity_data)
|
| 157 |
-
|
| 158 |
-
# Process relationships
|
| 159 |
-
for rel in result.relationships:
|
| 160 |
-
rel_data = ExtractedRelationshipResponse(
|
| 161 |
-
source=rel.source,
|
| 162 |
-
target=rel.target,
|
| 163 |
-
relationship_type=rel.relationship_type,
|
| 164 |
-
context=rel.context,
|
| 165 |
-
created=False
|
| 166 |
-
)
|
| 167 |
-
|
| 168 |
-
if request.auto_create:
|
| 169 |
-
# Find source and target entities
|
| 170 |
-
source_entity = db.query(Entity).filter(
|
| 171 |
-
Entity.name.ilike(f"%{rel.source}%")
|
| 172 |
-
).first()
|
| 173 |
-
target_entity = db.query(Entity).filter(
|
| 174 |
-
Entity.name.ilike(f"%{rel.target}%")
|
| 175 |
-
).first()
|
| 176 |
-
|
| 177 |
-
if source_entity and target_entity:
|
| 178 |
-
# Check if relationship exists
|
| 179 |
-
existing_rel = db.query(Relationship).filter(
|
| 180 |
-
Relationship.source_id == source_entity.id,
|
| 181 |
-
Relationship.target_id == target_entity.id,
|
| 182 |
-
Relationship.type == rel.relationship_type
|
| 183 |
-
).first()
|
| 184 |
-
|
| 185 |
-
if not existing_rel:
|
| 186 |
-
# Parse event_date if available
|
| 187 |
-
rel_event_date = parse_date(getattr(rel, 'event_date', None))
|
| 188 |
-
|
| 189 |
-
new_rel = Relationship(
|
| 190 |
-
source_id=source_entity.id,
|
| 191 |
-
target_id=target_entity.id,
|
| 192 |
-
type=rel.relationship_type,
|
| 193 |
-
event_date=rel_event_date,
|
| 194 |
-
properties={"context": rel.context}
|
| 195 |
-
)
|
| 196 |
-
db.add(new_rel)
|
| 197 |
-
db.commit()
|
| 198 |
-
rel_data.created = True
|
| 199 |
-
created_relationships += 1
|
| 200 |
-
|
| 201 |
-
relationships_response.append(rel_data)
|
| 202 |
-
|
| 203 |
-
# Process events
|
| 204 |
-
for event in result.events:
|
| 205 |
-
event_data = ExtractedEventResponse(
|
| 206 |
-
description=event.description,
|
| 207 |
-
event_type=event.event_type,
|
| 208 |
-
date=event.date,
|
| 209 |
-
location=event.location,
|
| 210 |
-
participants=event.participants,
|
| 211 |
-
created=False
|
| 212 |
-
)
|
| 213 |
-
|
| 214 |
-
if request.auto_create and event.description:
|
| 215 |
-
# Create event
|
| 216 |
-
new_event = Event(
|
| 217 |
-
title=event.description[:100] if len(event.description) > 100 else event.description,
|
| 218 |
-
description=event.description,
|
| 219 |
-
type=event.event_type or "general",
|
| 220 |
-
source="llm_extraction"
|
| 221 |
-
)
|
| 222 |
-
db.add(new_event)
|
| 223 |
-
db.commit()
|
| 224 |
-
db.refresh(new_event)
|
| 225 |
-
|
| 226 |
-
event_data.created = True
|
| 227 |
-
event_data.event_id = new_event.id
|
| 228 |
-
created_events += 1
|
| 229 |
-
|
| 230 |
-
events_response.append(event_data)
|
| 231 |
-
|
| 232 |
-
return AnalyzeResponse(
|
| 233 |
-
entities=entities_response,
|
| 234 |
-
relationships=relationships_response,
|
| 235 |
-
events=events_response,
|
| 236 |
-
stats={
|
| 237 |
-
"total_entities": len(entities_response),
|
| 238 |
-
"total_relationships": len(relationships_response),
|
| 239 |
-
"total_events": len(events_response),
|
| 240 |
-
"created_entities": created_entities,
|
| 241 |
-
"created_relationships": created_relationships,
|
| 242 |
-
"created_events": created_events
|
| 243 |
-
}
|
| 244 |
-
)
|
| 245 |
-
|
| 246 |
-
except Exception as e:
|
| 247 |
-
# Log the full error with traceback
|
| 248 |
-
print(f"=== ANALYZE ERROR ===")
|
| 249 |
-
print(f"Error type: {type(e).__name__}")
|
| 250 |
-
print(f"Error message: {str(e)}")
|
| 251 |
-
print(f"Traceback:")
|
| 252 |
-
traceback.print_exc()
|
| 253 |
-
print(f"=== END ERROR ===")
|
| 254 |
-
raise HTTPException(status_code=500, detail=f"Analysis failed: {str(e)}")
|
| 255 |
-
|
| 256 |
-
|
| 257 |
-
@router.get("/debug")
|
| 258 |
-
async def debug_config():
|
| 259 |
-
"""
|
| 260 |
-
Debug endpoint to check if API is configured correctly.
|
| 261 |
-
"""
|
| 262 |
-
api_key = settings.cerebras_api_key
|
| 263 |
-
return {
|
| 264 |
-
"cerebras_api_key_configured": bool(api_key),
|
| 265 |
-
"cerebras_api_key_length": len(api_key) if api_key else 0,
|
| 266 |
-
"cerebras_api_key_preview": f"{api_key[:8]}...{api_key[-4:]}" if api_key and len(api_key) > 12 else "NOT SET"
|
| 267 |
-
}
|
| 268 |
-
|
| 269 |
-
|
| 270 |
-
@router.post("/quick")
|
| 271 |
-
async def quick_analyze(request: AnalyzeRequest):
|
| 272 |
-
"""
|
| 273 |
-
Quick analysis without database operations.
|
| 274 |
-
Returns only extracted data without creating anything.
|
| 275 |
-
"""
|
| 276 |
-
try:
|
| 277 |
-
result = await entity_extractor.extract(request.text)
|
| 278 |
-
|
| 279 |
-
return {
|
| 280 |
-
"entities": [
|
| 281 |
-
{
|
| 282 |
-
"name": e.name,
|
| 283 |
-
"type": e.type,
|
| 284 |
-
"role": e.role,
|
| 285 |
-
"aliases": e.aliases
|
| 286 |
-
}
|
| 287 |
-
for e in result.entities
|
| 288 |
-
],
|
| 289 |
-
"relationships": [
|
| 290 |
-
{
|
| 291 |
-
"source": r.source,
|
| 292 |
-
"target": r.target,
|
| 293 |
-
"type": r.relationship_type,
|
| 294 |
-
"context": r.context
|
| 295 |
-
}
|
| 296 |
-
for r in result.relationships
|
| 297 |
-
],
|
| 298 |
-
"events": [
|
| 299 |
-
{
|
| 300 |
-
"description": ev.description,
|
| 301 |
-
"type": ev.event_type,
|
| 302 |
-
"date": ev.date,
|
| 303 |
-
"participants": ev.participants
|
| 304 |
-
}
|
| 305 |
-
for ev in result.events
|
| 306 |
-
]
|
| 307 |
-
}
|
| 308 |
-
except Exception as e:
|
| 309 |
-
raise HTTPException(status_code=500, detail=f"Analysis failed: {str(e)}")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/chat.py
DELETED
|
@@ -1,63 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Chat API Routes - Intelligent chat with RAG
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, Depends, HTTPException
|
| 5 |
-
from pydantic import BaseModel, Field
|
| 6 |
-
from typing import Optional
|
| 7 |
-
from sqlalchemy.orm import Session
|
| 8 |
-
|
| 9 |
-
from app.api.deps import get_scoped_db, get_session_id
|
| 10 |
-
from app.services.chat import chat_service
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
router = APIRouter(prefix="/chat", tags=["Chat"])
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
class ChatRequest(BaseModel):
|
| 17 |
-
"""Chat request model"""
|
| 18 |
-
message: str = Field(..., min_length=1, description="User message")
|
| 19 |
-
use_web: bool = Field(default=True, description="Include web search")
|
| 20 |
-
use_history: bool = Field(default=True, description="Use conversation history")
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
class ChatResponse(BaseModel):
|
| 24 |
-
"""Chat response model"""
|
| 25 |
-
answer: str
|
| 26 |
-
local_context_used: bool
|
| 27 |
-
web_context_used: bool
|
| 28 |
-
entities_found: int
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
@router.post("", response_model=ChatResponse)
|
| 32 |
-
async def chat(
|
| 33 |
-
request: ChatRequest,
|
| 34 |
-
db: Session = Depends(get_scoped_db),
|
| 35 |
-
session_id: Optional[str] = Depends(get_session_id)
|
| 36 |
-
):
|
| 37 |
-
"""
|
| 38 |
-
Send a message and get an intelligent response.
|
| 39 |
-
|
| 40 |
-
Uses:
|
| 41 |
-
- Local NUMIDIUM knowledge (entities/relationships)
|
| 42 |
-
- Lancer web search (if enabled)
|
| 43 |
-
- Cerebras LLM for synthesis
|
| 44 |
-
"""
|
| 45 |
-
try:
|
| 46 |
-
result = await chat_service.chat(
|
| 47 |
-
message=request.message,
|
| 48 |
-
db=db,
|
| 49 |
-
use_web=request.use_web,
|
| 50 |
-
use_history=request.use_history,
|
| 51 |
-
session_id=session_id
|
| 52 |
-
)
|
| 53 |
-
return ChatResponse(**result)
|
| 54 |
-
|
| 55 |
-
except Exception as e:
|
| 56 |
-
raise HTTPException(status_code=500, detail=str(e))
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
@router.post("/clear")
|
| 60 |
-
async def clear_history(session_id: Optional[str] = Depends(get_session_id)):
|
| 61 |
-
"""Clear conversation history"""
|
| 62 |
-
chat_service.clear_history(session_id=session_id)
|
| 63 |
-
return {"message": "Historico limpo"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/dados_publicos.py
DELETED
|
@@ -1,155 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Public Data API Routes - IBGE and TSE data access
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, HTTPException, Query
|
| 5 |
-
from pydantic import BaseModel, Field
|
| 6 |
-
from typing import Optional, List, Dict, Any
|
| 7 |
-
|
| 8 |
-
from app.services.ibge_api import (
|
| 9 |
-
listar_estados,
|
| 10 |
-
listar_municipios,
|
| 11 |
-
buscar_municipio,
|
| 12 |
-
enriquecer_localizacao
|
| 13 |
-
)
|
| 14 |
-
from app.services.tse_api import (
|
| 15 |
-
listar_eleicoes,
|
| 16 |
-
buscar_candidatos,
|
| 17 |
-
obter_candidato_detalhes,
|
| 18 |
-
buscar_politico
|
| 19 |
-
)
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
router = APIRouter(prefix="/dados", tags=["Public Data"])
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
# ========== IBGE Endpoints ==========
|
| 26 |
-
|
| 27 |
-
class EstadoResponse(BaseModel):
|
| 28 |
-
id: int
|
| 29 |
-
sigla: str
|
| 30 |
-
nome: str
|
| 31 |
-
regiao: str
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
class MunicipioResponse(BaseModel):
|
| 35 |
-
id: int
|
| 36 |
-
nome: str
|
| 37 |
-
estado_sigla: str
|
| 38 |
-
estado_nome: str
|
| 39 |
-
regiao: str
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
@router.get("/ibge/estados", response_model=List[EstadoResponse])
|
| 43 |
-
async def get_estados():
|
| 44 |
-
"""List all Brazilian states"""
|
| 45 |
-
estados = await listar_estados()
|
| 46 |
-
return [EstadoResponse(**e.__dict__) for e in estados]
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
@router.get("/ibge/municipios/{uf}", response_model=List[MunicipioResponse])
|
| 50 |
-
async def get_municipios(uf: str):
|
| 51 |
-
"""List municipalities in a state"""
|
| 52 |
-
municipios = await listar_municipios(uf)
|
| 53 |
-
return [MunicipioResponse(**m.__dict__) for m in municipios]
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
@router.get("/ibge/buscar")
|
| 57 |
-
async def buscar_cidade(
|
| 58 |
-
nome: str = Query(..., min_length=2),
|
| 59 |
-
uf: Optional[str] = None
|
| 60 |
-
):
|
| 61 |
-
"""Search for a municipality by name"""
|
| 62 |
-
municipios = await buscar_municipio(nome, uf)
|
| 63 |
-
return [MunicipioResponse(**m.__dict__) for m in municipios]
|
| 64 |
-
|
| 65 |
-
|
| 66 |
-
@router.get("/ibge/enriquecer")
|
| 67 |
-
async def enriquecer_cidade(
|
| 68 |
-
cidade: str = Query(..., min_length=2),
|
| 69 |
-
uf: Optional[str] = None
|
| 70 |
-
):
|
| 71 |
-
"""Enrich a location name with IBGE data"""
|
| 72 |
-
return await enriquecer_localizacao(cidade, uf)
|
| 73 |
-
|
| 74 |
-
|
| 75 |
-
# ========== TSE Endpoints ==========
|
| 76 |
-
|
| 77 |
-
class EleicaoResponse(BaseModel):
|
| 78 |
-
id: int
|
| 79 |
-
ano: int
|
| 80 |
-
descricao: str
|
| 81 |
-
turno: int
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
class CandidatoResponse(BaseModel):
|
| 85 |
-
id: int
|
| 86 |
-
nome: str
|
| 87 |
-
nome_urna: str
|
| 88 |
-
numero: str
|
| 89 |
-
cargo: str
|
| 90 |
-
partido_sigla: str
|
| 91 |
-
uf: str
|
| 92 |
-
municipio: str
|
| 93 |
-
situacao: str
|
| 94 |
-
total_bens: float
|
| 95 |
-
|
| 96 |
-
|
| 97 |
-
class CandidatoDetalhadoResponse(BaseModel):
|
| 98 |
-
id: int
|
| 99 |
-
nome: str
|
| 100 |
-
nome_urna: str
|
| 101 |
-
numero: str
|
| 102 |
-
cargo: str
|
| 103 |
-
partido_sigla: str
|
| 104 |
-
partido_nome: str
|
| 105 |
-
uf: str
|
| 106 |
-
municipio: str
|
| 107 |
-
situacao: str
|
| 108 |
-
data_nascimento: str
|
| 109 |
-
genero: str
|
| 110 |
-
grau_instrucao: str
|
| 111 |
-
ocupacao: str
|
| 112 |
-
total_bens: float
|
| 113 |
-
bens: List[Dict[str, Any]]
|
| 114 |
-
|
| 115 |
-
|
| 116 |
-
@router.get("/tse/eleicoes", response_model=List[EleicaoResponse])
|
| 117 |
-
async def get_eleicoes():
|
| 118 |
-
"""List available elections"""
|
| 119 |
-
eleicoes = await listar_eleicoes()
|
| 120 |
-
return [EleicaoResponse(**e.__dict__) for e in eleicoes]
|
| 121 |
-
|
| 122 |
-
|
| 123 |
-
@router.get("/tse/candidatos")
|
| 124 |
-
async def get_candidatos(
|
| 125 |
-
nome: str = Query(..., min_length=3),
|
| 126 |
-
ano: int = Query(default=2024),
|
| 127 |
-
uf: Optional[str] = None,
|
| 128 |
-
cargo: Optional[str] = None
|
| 129 |
-
):
|
| 130 |
-
"""Search for candidates by name"""
|
| 131 |
-
candidatos = await buscar_candidatos(nome, ano=ano, uf=uf, cargo=cargo)
|
| 132 |
-
return [CandidatoResponse(**c.__dict__) for c in candidatos]
|
| 133 |
-
|
| 134 |
-
|
| 135 |
-
@router.get("/tse/candidato/{id_candidato}")
|
| 136 |
-
async def get_candidato_detalhes(
|
| 137 |
-
id_candidato: int,
|
| 138 |
-
ano: int = Query(default=2024)
|
| 139 |
-
):
|
| 140 |
-
"""Get detailed candidate information including assets"""
|
| 141 |
-
candidato = await obter_candidato_detalhes(id_candidato, ano=ano)
|
| 142 |
-
|
| 143 |
-
if not candidato:
|
| 144 |
-
raise HTTPException(status_code=404, detail="Candidato não encontrado")
|
| 145 |
-
|
| 146 |
-
return CandidatoDetalhadoResponse(**candidato.__dict__)
|
| 147 |
-
|
| 148 |
-
|
| 149 |
-
@router.get("/tse/politico")
|
| 150 |
-
async def pesquisar_politico(nome: str = Query(..., min_length=3)):
|
| 151 |
-
"""
|
| 152 |
-
Search for a politician across multiple elections.
|
| 153 |
-
Returns consolidated career information.
|
| 154 |
-
"""
|
| 155 |
-
return await buscar_politico(nome)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/entities.py
DELETED
|
@@ -1,353 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Entity CRUD Routes
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, Depends, HTTPException, Query
|
| 5 |
-
from sqlalchemy.orm import Session
|
| 6 |
-
from sqlalchemy import or_
|
| 7 |
-
from typing import List, Optional
|
| 8 |
-
|
| 9 |
-
from app.api.deps import get_scoped_db
|
| 10 |
-
from app.models import Entity, Relationship
|
| 11 |
-
from app.schemas import EntityCreate, EntityUpdate, EntityResponse, GraphData, GraphNode, GraphEdge
|
| 12 |
-
|
| 13 |
-
router = APIRouter(prefix="/entities", tags=["Entities"])
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
@router.get("", response_model=List[EntityResponse])
|
| 17 |
-
def list_entities(
|
| 18 |
-
type: Optional[str] = None,
|
| 19 |
-
search: Optional[str] = None,
|
| 20 |
-
project_id: Optional[str] = None,
|
| 21 |
-
limit: int = Query(default=50, le=200),
|
| 22 |
-
offset: int = 0,
|
| 23 |
-
db: Session = Depends(get_scoped_db)
|
| 24 |
-
):
|
| 25 |
-
"""Lista todas as entidades com filtros opcionais"""
|
| 26 |
-
query = db.query(Entity)
|
| 27 |
-
|
| 28 |
-
if project_id:
|
| 29 |
-
query = query.filter(Entity.project_id == project_id)
|
| 30 |
-
|
| 31 |
-
if type:
|
| 32 |
-
query = query.filter(Entity.type == type)
|
| 33 |
-
|
| 34 |
-
if search:
|
| 35 |
-
query = query.filter(
|
| 36 |
-
or_(
|
| 37 |
-
Entity.name.ilike(f"%{search}%"),
|
| 38 |
-
Entity.description.ilike(f"%{search}%")
|
| 39 |
-
)
|
| 40 |
-
)
|
| 41 |
-
|
| 42 |
-
query = query.order_by(Entity.created_at.desc())
|
| 43 |
-
return query.offset(offset).limit(limit).all()
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
@router.get("/types")
|
| 47 |
-
def get_entity_types(db: Session = Depends(get_scoped_db)):
|
| 48 |
-
"""Retorna todos os tipos de entidade únicos"""
|
| 49 |
-
types = db.query(Entity.type).distinct().all()
|
| 50 |
-
return [t[0] for t in types]
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
@router.get("/suggest-merge")
|
| 54 |
-
async def suggest_merge_candidates(
|
| 55 |
-
limit: int = Query(default=10, le=50),
|
| 56 |
-
db: Session = Depends(get_scoped_db)
|
| 57 |
-
):
|
| 58 |
-
"""
|
| 59 |
-
Use LLM to find potential duplicate entities that could be merged.
|
| 60 |
-
Returns pairs of entities that might be the same.
|
| 61 |
-
"""
|
| 62 |
-
import httpx
|
| 63 |
-
import json
|
| 64 |
-
import re
|
| 65 |
-
from app.config import settings
|
| 66 |
-
|
| 67 |
-
# Get all entities
|
| 68 |
-
entities = db.query(Entity).order_by(Entity.name).limit(200).all()
|
| 69 |
-
|
| 70 |
-
if len(entities) < 2:
|
| 71 |
-
return {"candidates": [], "message": "Not enough entities to compare"}
|
| 72 |
-
|
| 73 |
-
# Build entity list for LLM
|
| 74 |
-
entity_list = []
|
| 75 |
-
for e in entities:
|
| 76 |
-
aliases = (e.properties or {}).get("aliases", [])
|
| 77 |
-
entity_list.append({
|
| 78 |
-
"id": e.id,
|
| 79 |
-
"name": e.name,
|
| 80 |
-
"type": e.type,
|
| 81 |
-
"aliases": aliases[:5] if aliases else []
|
| 82 |
-
})
|
| 83 |
-
|
| 84 |
-
# Ask LLM to find duplicates
|
| 85 |
-
prompt = f"""Analise esta lista de entidades e encontre possíveis DUPLICATAS (mesma pessoa/organização/local com nomes diferentes).
|
| 86 |
-
|
| 87 |
-
Entidades:
|
| 88 |
-
{entity_list[:100]}
|
| 89 |
-
|
| 90 |
-
Retorne APENAS um JSON válido com pares de IDs que são provavelmente a mesma entidade:
|
| 91 |
-
```json
|
| 92 |
-
{{
|
| 93 |
-
"duplicates": [
|
| 94 |
-
{{
|
| 95 |
-
"id1": "uuid1",
|
| 96 |
-
"id2": "uuid2",
|
| 97 |
-
"confidence": 0.95,
|
| 98 |
-
"reason": "Mesmo nome com variação"
|
| 99 |
-
}}
|
| 100 |
-
]
|
| 101 |
-
}}
|
| 102 |
-
```
|
| 103 |
-
|
| 104 |
-
Se não houver duplicatas, retorne: {{"duplicates": []}}
|
| 105 |
-
"""
|
| 106 |
-
|
| 107 |
-
try:
|
| 108 |
-
async with httpx.AsyncClient(timeout=30.0) as client:
|
| 109 |
-
response = await client.post(
|
| 110 |
-
"https://api.cerebras.ai/v1/chat/completions",
|
| 111 |
-
headers={
|
| 112 |
-
"Authorization": f"Bearer {settings.cerebras_api_key}",
|
| 113 |
-
"Content-Type": "application/json"
|
| 114 |
-
},
|
| 115 |
-
json={
|
| 116 |
-
"model": "zai-glm-4.7",
|
| 117 |
-
"messages": [
|
| 118 |
-
{"role": "system", "content": "Você é um especialista em detecção de entidades duplicadas. Responda apenas em JSON válido."},
|
| 119 |
-
{"role": "user", "content": prompt}
|
| 120 |
-
],
|
| 121 |
-
"temperature": 0.1,
|
| 122 |
-
"max_tokens": 1024
|
| 123 |
-
}
|
| 124 |
-
)
|
| 125 |
-
|
| 126 |
-
if response.status_code != 200:
|
| 127 |
-
return {"candidates": [], "error": "LLM API error"}
|
| 128 |
-
|
| 129 |
-
data = response.json()
|
| 130 |
-
content = data["choices"][0]["message"]["content"]
|
| 131 |
-
|
| 132 |
-
# Parse JSON from response
|
| 133 |
-
json_match = re.search(r'\{.*\}', content, re.DOTALL)
|
| 134 |
-
if json_match:
|
| 135 |
-
result = json.loads(json_match.group(0))
|
| 136 |
-
|
| 137 |
-
# Enrich with entity names
|
| 138 |
-
candidates = []
|
| 139 |
-
for dup in result.get("duplicates", [])[:limit]:
|
| 140 |
-
e1 = next((e for e in entities if e.id == dup.get("id1")), None)
|
| 141 |
-
e2 = next((e for e in entities if e.id == dup.get("id2")), None)
|
| 142 |
-
if e1 and e2:
|
| 143 |
-
candidates.append({
|
| 144 |
-
"entity1": {"id": e1.id, "name": e1.name, "type": e1.type},
|
| 145 |
-
"entity2": {"id": e2.id, "name": e2.name, "type": e2.type},
|
| 146 |
-
"confidence": dup.get("confidence", 0.5),
|
| 147 |
-
"reason": dup.get("reason", "Possível duplicata")
|
| 148 |
-
})
|
| 149 |
-
|
| 150 |
-
return {"candidates": candidates}
|
| 151 |
-
|
| 152 |
-
return {"candidates": [], "message": "No duplicates found"}
|
| 153 |
-
|
| 154 |
-
except Exception as e:
|
| 155 |
-
return {"candidates": [], "error": str(e)}
|
| 156 |
-
|
| 157 |
-
|
| 158 |
-
@router.get("/{entity_id}", response_model=EntityResponse)
|
| 159 |
-
def get_entity(entity_id: str, db: Session = Depends(get_scoped_db)):
|
| 160 |
-
"""Busca uma entidade por ID"""
|
| 161 |
-
entity = db.query(Entity).filter(Entity.id == entity_id).first()
|
| 162 |
-
if not entity:
|
| 163 |
-
raise HTTPException(status_code=404, detail="Entity not found")
|
| 164 |
-
return entity
|
| 165 |
-
|
| 166 |
-
|
| 167 |
-
@router.post("", response_model=EntityResponse, status_code=201)
|
| 168 |
-
def create_entity(entity: EntityCreate, db: Session = Depends(get_scoped_db)):
|
| 169 |
-
"""Cria uma nova entidade"""
|
| 170 |
-
db_entity = Entity(**entity.model_dump())
|
| 171 |
-
db.add(db_entity)
|
| 172 |
-
db.commit()
|
| 173 |
-
db.refresh(db_entity)
|
| 174 |
-
return db_entity
|
| 175 |
-
|
| 176 |
-
|
| 177 |
-
@router.put("/{entity_id}", response_model=EntityResponse)
|
| 178 |
-
def update_entity(entity_id: str, entity: EntityUpdate, db: Session = Depends(get_scoped_db)):
|
| 179 |
-
"""Atualiza uma entidade existente"""
|
| 180 |
-
db_entity = db.query(Entity).filter(Entity.id == entity_id).first()
|
| 181 |
-
if not db_entity:
|
| 182 |
-
raise HTTPException(status_code=404, detail="Entity not found")
|
| 183 |
-
|
| 184 |
-
update_data = entity.model_dump(exclude_unset=True)
|
| 185 |
-
for field, value in update_data.items():
|
| 186 |
-
setattr(db_entity, field, value)
|
| 187 |
-
|
| 188 |
-
db.commit()
|
| 189 |
-
db.refresh(db_entity)
|
| 190 |
-
return db_entity
|
| 191 |
-
|
| 192 |
-
|
| 193 |
-
@router.delete("/{entity_id}")
|
| 194 |
-
def delete_entity(entity_id: str, db: Session = Depends(get_scoped_db)):
|
| 195 |
-
"""Deleta uma entidade"""
|
| 196 |
-
db_entity = db.query(Entity).filter(Entity.id == entity_id).first()
|
| 197 |
-
if not db_entity:
|
| 198 |
-
raise HTTPException(status_code=404, detail="Entity not found")
|
| 199 |
-
|
| 200 |
-
# Delete related relationships
|
| 201 |
-
db.query(Relationship).filter(
|
| 202 |
-
or_(
|
| 203 |
-
Relationship.source_id == entity_id,
|
| 204 |
-
Relationship.target_id == entity_id
|
| 205 |
-
)
|
| 206 |
-
).delete()
|
| 207 |
-
|
| 208 |
-
db.delete(db_entity)
|
| 209 |
-
db.commit()
|
| 210 |
-
return {"message": "Entity deleted"}
|
| 211 |
-
|
| 212 |
-
|
| 213 |
-
@router.get("/{entity_id}/connections", response_model=GraphData)
|
| 214 |
-
def get_entity_connections(
|
| 215 |
-
entity_id: str,
|
| 216 |
-
depth: int = Query(default=1, le=3),
|
| 217 |
-
db: Session = Depends(get_scoped_db)
|
| 218 |
-
):
|
| 219 |
-
"""
|
| 220 |
-
Retorna o grafo de conexões de uma entidade
|
| 221 |
-
Usado para visualização de rede no frontend
|
| 222 |
-
"""
|
| 223 |
-
entity = db.query(Entity).filter(Entity.id == entity_id).first()
|
| 224 |
-
if not entity:
|
| 225 |
-
raise HTTPException(status_code=404, detail="Entity not found")
|
| 226 |
-
|
| 227 |
-
nodes = {}
|
| 228 |
-
edges = []
|
| 229 |
-
visited = set()
|
| 230 |
-
|
| 231 |
-
def explore(eid: str, current_depth: int):
|
| 232 |
-
if current_depth > depth or eid in visited:
|
| 233 |
-
return
|
| 234 |
-
visited.add(eid)
|
| 235 |
-
|
| 236 |
-
e = db.query(Entity).filter(Entity.id == eid).first()
|
| 237 |
-
if not e:
|
| 238 |
-
return
|
| 239 |
-
|
| 240 |
-
nodes[e.id] = GraphNode(
|
| 241 |
-
id=e.id,
|
| 242 |
-
type=e.type,
|
| 243 |
-
name=e.name,
|
| 244 |
-
properties=e.properties or {}
|
| 245 |
-
)
|
| 246 |
-
|
| 247 |
-
# Outgoing relationships
|
| 248 |
-
for rel in db.query(Relationship).filter(Relationship.source_id == eid).all():
|
| 249 |
-
edges.append(GraphEdge(
|
| 250 |
-
source=rel.source_id,
|
| 251 |
-
target=rel.target_id,
|
| 252 |
-
type=rel.type,
|
| 253 |
-
confidence=rel.confidence
|
| 254 |
-
))
|
| 255 |
-
explore(rel.target_id, current_depth + 1)
|
| 256 |
-
|
| 257 |
-
# Incoming relationships
|
| 258 |
-
for rel in db.query(Relationship).filter(Relationship.target_id == eid).all():
|
| 259 |
-
edges.append(GraphEdge(
|
| 260 |
-
source=rel.source_id,
|
| 261 |
-
target=rel.target_id,
|
| 262 |
-
type=rel.type,
|
| 263 |
-
confidence=rel.confidence
|
| 264 |
-
))
|
| 265 |
-
explore(rel.source_id, current_depth + 1)
|
| 266 |
-
|
| 267 |
-
explore(entity_id, 0)
|
| 268 |
-
|
| 269 |
-
return GraphData(
|
| 270 |
-
nodes=list(nodes.values()),
|
| 271 |
-
edges=edges
|
| 272 |
-
)
|
| 273 |
-
|
| 274 |
-
|
| 275 |
-
@router.post("/merge")
|
| 276 |
-
def merge_entities(
|
| 277 |
-
primary_id: str,
|
| 278 |
-
secondary_id: str,
|
| 279 |
-
db: Session = Depends(get_scoped_db)
|
| 280 |
-
):
|
| 281 |
-
"""
|
| 282 |
-
Merge two entities into one.
|
| 283 |
-
The primary entity is kept, the secondary is deleted.
|
| 284 |
-
All relationships from secondary are transferred to primary.
|
| 285 |
-
"""
|
| 286 |
-
if primary_id == secondary_id:
|
| 287 |
-
raise HTTPException(status_code=400, detail="Cannot merge entity with itself")
|
| 288 |
-
|
| 289 |
-
primary = db.query(Entity).filter(Entity.id == primary_id).first()
|
| 290 |
-
secondary = db.query(Entity).filter(Entity.id == secondary_id).first()
|
| 291 |
-
|
| 292 |
-
if not primary:
|
| 293 |
-
raise HTTPException(status_code=404, detail="Primary entity not found")
|
| 294 |
-
if not secondary:
|
| 295 |
-
raise HTTPException(status_code=404, detail="Secondary entity not found")
|
| 296 |
-
|
| 297 |
-
# Merge properties
|
| 298 |
-
primary_props = primary.properties or {}
|
| 299 |
-
secondary_props = secondary.properties or {}
|
| 300 |
-
|
| 301 |
-
# Add aliases from secondary
|
| 302 |
-
aliases = primary_props.get("aliases", []) or []
|
| 303 |
-
if secondary.name not in aliases:
|
| 304 |
-
aliases.append(secondary.name)
|
| 305 |
-
secondary_aliases = secondary_props.get("aliases", []) or []
|
| 306 |
-
for alias in secondary_aliases:
|
| 307 |
-
if alias not in aliases:
|
| 308 |
-
aliases.append(alias)
|
| 309 |
-
primary_props["aliases"] = aliases
|
| 310 |
-
|
| 311 |
-
# Add merge history
|
| 312 |
-
merge_history = primary_props.get("merged_from", []) or []
|
| 313 |
-
merge_history.append({
|
| 314 |
-
"id": secondary.id,
|
| 315 |
-
"name": secondary.name,
|
| 316 |
-
"source": secondary.source
|
| 317 |
-
})
|
| 318 |
-
primary_props["merged_from"] = merge_history
|
| 319 |
-
|
| 320 |
-
# Combine descriptions if primary has none
|
| 321 |
-
if not primary.description and secondary.description:
|
| 322 |
-
primary.description = secondary.description
|
| 323 |
-
|
| 324 |
-
primary.properties = primary_props
|
| 325 |
-
|
| 326 |
-
# Transfer relationships from secondary to primary
|
| 327 |
-
# Update source_id
|
| 328 |
-
db.query(Relationship).filter(
|
| 329 |
-
Relationship.source_id == secondary_id
|
| 330 |
-
).update({"source_id": primary_id})
|
| 331 |
-
|
| 332 |
-
# Update target_id
|
| 333 |
-
db.query(Relationship).filter(
|
| 334 |
-
Relationship.target_id == secondary_id
|
| 335 |
-
).update({"target_id": primary_id})
|
| 336 |
-
|
| 337 |
-
# Delete duplicate relationships (same source, target, type)
|
| 338 |
-
# This is a simple approach - in production you'd want more sophisticated deduplication
|
| 339 |
-
|
| 340 |
-
# Delete the secondary entity
|
| 341 |
-
db.delete(secondary)
|
| 342 |
-
db.commit()
|
| 343 |
-
db.refresh(primary)
|
| 344 |
-
|
| 345 |
-
return {
|
| 346 |
-
"message": f"Merged '{secondary.name}' into '{primary.name}'",
|
| 347 |
-
"primary": {
|
| 348 |
-
"id": primary.id,
|
| 349 |
-
"name": primary.name,
|
| 350 |
-
"aliases": aliases
|
| 351 |
-
}
|
| 352 |
-
}
|
| 353 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/events.py
DELETED
|
@@ -1,113 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Events CRUD Routes
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, Depends, HTTPException, Query
|
| 5 |
-
from sqlalchemy.orm import Session
|
| 6 |
-
from sqlalchemy import or_
|
| 7 |
-
from typing import List, Optional
|
| 8 |
-
from datetime import datetime
|
| 9 |
-
|
| 10 |
-
from app.api.deps import get_scoped_db
|
| 11 |
-
from app.models import Event
|
| 12 |
-
from app.schemas import EventCreate, EventResponse
|
| 13 |
-
|
| 14 |
-
router = APIRouter(prefix="/events", tags=["Events"])
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
@router.get("/", response_model=List[EventResponse])
|
| 18 |
-
def list_events(
|
| 19 |
-
type: Optional[str] = None,
|
| 20 |
-
search: Optional[str] = None,
|
| 21 |
-
start_date: Optional[datetime] = None,
|
| 22 |
-
end_date: Optional[datetime] = None,
|
| 23 |
-
limit: int = Query(default=50, le=200),
|
| 24 |
-
offset: int = 0,
|
| 25 |
-
db: Session = Depends(get_scoped_db)
|
| 26 |
-
):
|
| 27 |
-
"""Lista eventos com filtros opcionais"""
|
| 28 |
-
query = db.query(Event)
|
| 29 |
-
|
| 30 |
-
if type:
|
| 31 |
-
query = query.filter(Event.type == type)
|
| 32 |
-
|
| 33 |
-
if search:
|
| 34 |
-
query = query.filter(
|
| 35 |
-
or_(
|
| 36 |
-
Event.title.ilike(f"%{search}%"),
|
| 37 |
-
Event.description.ilike(f"%{search}%")
|
| 38 |
-
)
|
| 39 |
-
)
|
| 40 |
-
|
| 41 |
-
if start_date:
|
| 42 |
-
query = query.filter(Event.event_date >= start_date)
|
| 43 |
-
if end_date:
|
| 44 |
-
query = query.filter(Event.event_date <= end_date)
|
| 45 |
-
|
| 46 |
-
query = query.order_by(Event.event_date.desc().nullslast())
|
| 47 |
-
return query.offset(offset).limit(limit).all()
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
@router.get("/types")
|
| 51 |
-
def get_event_types(db: Session = Depends(get_scoped_db)):
|
| 52 |
-
"""Retorna todos os tipos de evento unicos"""
|
| 53 |
-
types = db.query(Event.type).distinct().all()
|
| 54 |
-
return [t[0] for t in types]
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
@router.get("/timeline")
|
| 58 |
-
def get_timeline(
|
| 59 |
-
entity_id: Optional[str] = None,
|
| 60 |
-
limit: int = Query(default=50, le=200),
|
| 61 |
-
db: Session = Depends(get_scoped_db)
|
| 62 |
-
):
|
| 63 |
-
"""
|
| 64 |
-
Retorna eventos em formato timeline.
|
| 65 |
-
"""
|
| 66 |
-
query = db.query(Event).filter(Event.event_date.isnot(None))
|
| 67 |
-
|
| 68 |
-
if entity_id:
|
| 69 |
-
query = query.filter(Event.entity_ids.contains([entity_id]))
|
| 70 |
-
|
| 71 |
-
events = query.order_by(Event.event_date.asc()).limit(limit).all()
|
| 72 |
-
|
| 73 |
-
return [
|
| 74 |
-
{
|
| 75 |
-
"id": e.id,
|
| 76 |
-
"title": e.title,
|
| 77 |
-
"date": e.event_date.isoformat() if e.event_date else None,
|
| 78 |
-
"type": e.type,
|
| 79 |
-
"location": e.location_name
|
| 80 |
-
}
|
| 81 |
-
for e in events
|
| 82 |
-
]
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
@router.get("/{event_id}", response_model=EventResponse)
|
| 86 |
-
def get_event(event_id: str, db: Session = Depends(get_scoped_db)):
|
| 87 |
-
"""Busca um evento por ID"""
|
| 88 |
-
event = db.query(Event).filter(Event.id == event_id).first()
|
| 89 |
-
if not event:
|
| 90 |
-
raise HTTPException(status_code=404, detail="Event not found")
|
| 91 |
-
return event
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
@router.post("/", response_model=EventResponse, status_code=201)
|
| 95 |
-
def create_event(event: EventCreate, db: Session = Depends(get_scoped_db)):
|
| 96 |
-
"""Cria um novo evento"""
|
| 97 |
-
db_event = Event(**event.model_dump())
|
| 98 |
-
db.add(db_event)
|
| 99 |
-
db.commit()
|
| 100 |
-
db.refresh(db_event)
|
| 101 |
-
return db_event
|
| 102 |
-
|
| 103 |
-
|
| 104 |
-
@router.delete("/{event_id}")
|
| 105 |
-
def delete_event(event_id: str, db: Session = Depends(get_scoped_db)):
|
| 106 |
-
"""Deleta um evento"""
|
| 107 |
-
db_event = db.query(Event).filter(Event.id == event_id).first()
|
| 108 |
-
if not db_event:
|
| 109 |
-
raise HTTPException(status_code=404, detail="Event not found")
|
| 110 |
-
|
| 111 |
-
db.delete(db_event)
|
| 112 |
-
db.commit()
|
| 113 |
-
return {"message": "Event deleted"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/graph.py
DELETED
|
@@ -1,173 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Graph API Routes - Network visualization endpoints
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, Depends, HTTPException, Query
|
| 5 |
-
from typing import Optional, List
|
| 6 |
-
from sqlalchemy.orm import Session
|
| 7 |
-
from sqlalchemy import or_
|
| 8 |
-
|
| 9 |
-
from app.api.deps import get_scoped_db
|
| 10 |
-
from app.models.entity import Entity, Relationship
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
router = APIRouter(prefix="/graph", tags=["Graph"])
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
@router.get("")
|
| 17 |
-
async def get_graph(
|
| 18 |
-
entity_type: Optional[str] = Query(None, description="Filter by entity type"),
|
| 19 |
-
limit: int = Query(100, le=500, description="Maximum number of entities"),
|
| 20 |
-
db: Session = Depends(get_scoped_db)
|
| 21 |
-
):
|
| 22 |
-
"""
|
| 23 |
-
Get graph data for visualization.
|
| 24 |
-
Returns nodes (entities) and edges (relationships).
|
| 25 |
-
"""
|
| 26 |
-
try:
|
| 27 |
-
# Get entities
|
| 28 |
-
query = db.query(Entity)
|
| 29 |
-
if entity_type:
|
| 30 |
-
query = query.filter(Entity.type == entity_type)
|
| 31 |
-
|
| 32 |
-
entities = query.limit(limit).all()
|
| 33 |
-
entity_ids = [e.id for e in entities]
|
| 34 |
-
|
| 35 |
-
# Get relationships between these entities
|
| 36 |
-
relationships = db.query(Relationship).filter(
|
| 37 |
-
or_(
|
| 38 |
-
Relationship.source_id.in_(entity_ids),
|
| 39 |
-
Relationship.target_id.in_(entity_ids)
|
| 40 |
-
)
|
| 41 |
-
).all()
|
| 42 |
-
|
| 43 |
-
# Format for Cytoscape.js
|
| 44 |
-
nodes = []
|
| 45 |
-
for e in entities:
|
| 46 |
-
nodes.append({
|
| 47 |
-
"data": {
|
| 48 |
-
"id": e.id,
|
| 49 |
-
"label": e.name[:30] + "..." if len(e.name) > 30 else e.name,
|
| 50 |
-
"fullName": e.name,
|
| 51 |
-
"type": e.type,
|
| 52 |
-
"description": e.description[:100] if e.description else "",
|
| 53 |
-
"source": e.source or "unknown"
|
| 54 |
-
}
|
| 55 |
-
})
|
| 56 |
-
|
| 57 |
-
edges = []
|
| 58 |
-
for r in relationships:
|
| 59 |
-
if r.source_id in entity_ids and r.target_id in entity_ids:
|
| 60 |
-
edges.append({
|
| 61 |
-
"data": {
|
| 62 |
-
"id": r.id,
|
| 63 |
-
"source": r.source_id,
|
| 64 |
-
"target": r.target_id,
|
| 65 |
-
"label": r.type,
|
| 66 |
-
"type": r.type
|
| 67 |
-
}
|
| 68 |
-
})
|
| 69 |
-
|
| 70 |
-
return {
|
| 71 |
-
"nodes": nodes,
|
| 72 |
-
"edges": edges,
|
| 73 |
-
"stats": {
|
| 74 |
-
"total_nodes": len(nodes),
|
| 75 |
-
"total_edges": len(edges)
|
| 76 |
-
}
|
| 77 |
-
}
|
| 78 |
-
|
| 79 |
-
except Exception as e:
|
| 80 |
-
raise HTTPException(status_code=500, detail=f"Failed to get graph: {str(e)}")
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
@router.get("/entity/{entity_id}")
|
| 84 |
-
async def get_entity_graph(
|
| 85 |
-
entity_id: str,
|
| 86 |
-
depth: int = Query(1, ge=1, le=3, description="How many levels of connections to include"),
|
| 87 |
-
db: Session = Depends(get_scoped_db)
|
| 88 |
-
):
|
| 89 |
-
"""
|
| 90 |
-
Get graph centered on a specific entity.
|
| 91 |
-
"""
|
| 92 |
-
try:
|
| 93 |
-
# Get the central entity
|
| 94 |
-
central = db.query(Entity).filter(Entity.id == entity_id).first()
|
| 95 |
-
if not central:
|
| 96 |
-
raise HTTPException(status_code=404, detail="Entity not found")
|
| 97 |
-
|
| 98 |
-
# Collect entity IDs at each depth level
|
| 99 |
-
collected_ids = {entity_id}
|
| 100 |
-
current_level = {entity_id}
|
| 101 |
-
|
| 102 |
-
for _ in range(depth):
|
| 103 |
-
rels = db.query(Relationship).filter(
|
| 104 |
-
or_(
|
| 105 |
-
Relationship.source_id.in_(current_level),
|
| 106 |
-
Relationship.target_id.in_(current_level)
|
| 107 |
-
)
|
| 108 |
-
).all()
|
| 109 |
-
|
| 110 |
-
next_level = set()
|
| 111 |
-
for r in rels:
|
| 112 |
-
next_level.add(r.source_id)
|
| 113 |
-
next_level.add(r.target_id)
|
| 114 |
-
|
| 115 |
-
current_level = next_level - collected_ids
|
| 116 |
-
collected_ids.update(next_level)
|
| 117 |
-
|
| 118 |
-
# Get all entities
|
| 119 |
-
entities = db.query(Entity).filter(Entity.id.in_(collected_ids)).all()
|
| 120 |
-
|
| 121 |
-
# Get all relationships between collected entities
|
| 122 |
-
relationships = db.query(Relationship).filter(
|
| 123 |
-
Relationship.source_id.in_(collected_ids),
|
| 124 |
-
Relationship.target_id.in_(collected_ids)
|
| 125 |
-
).all()
|
| 126 |
-
|
| 127 |
-
# Format for Cytoscape
|
| 128 |
-
nodes = []
|
| 129 |
-
for e in entities:
|
| 130 |
-
nodes.append({
|
| 131 |
-
"data": {
|
| 132 |
-
"id": e.id,
|
| 133 |
-
"label": e.name[:30] + "..." if len(e.name) > 30 else e.name,
|
| 134 |
-
"fullName": e.name,
|
| 135 |
-
"type": e.type,
|
| 136 |
-
"description": e.description[:100] if e.description else "",
|
| 137 |
-
"source": e.source or "unknown",
|
| 138 |
-
"isCentral": e.id == entity_id
|
| 139 |
-
}
|
| 140 |
-
})
|
| 141 |
-
|
| 142 |
-
edges = []
|
| 143 |
-
for r in relationships:
|
| 144 |
-
edges.append({
|
| 145 |
-
"data": {
|
| 146 |
-
"id": r.id,
|
| 147 |
-
"source": r.source_id,
|
| 148 |
-
"target": r.target_id,
|
| 149 |
-
"label": r.type,
|
| 150 |
-
"type": r.type
|
| 151 |
-
}
|
| 152 |
-
})
|
| 153 |
-
|
| 154 |
-
return {
|
| 155 |
-
"central": {
|
| 156 |
-
"id": central.id,
|
| 157 |
-
"name": central.name,
|
| 158 |
-
"type": central.type
|
| 159 |
-
},
|
| 160 |
-
"nodes": nodes,
|
| 161 |
-
"edges": edges,
|
| 162 |
-
"stats": {
|
| 163 |
-
"total_nodes": len(nodes),
|
| 164 |
-
"total_edges": len(edges),
|
| 165 |
-
"depth": depth
|
| 166 |
-
}
|
| 167 |
-
}
|
| 168 |
-
|
| 169 |
-
except HTTPException:
|
| 170 |
-
raise
|
| 171 |
-
except Exception as e:
|
| 172 |
-
raise HTTPException(status_code=500, detail=f"Failed to get entity graph: {str(e)}")
|
| 173 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/ingest.py
DELETED
|
@@ -1,341 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Data Ingestion Routes
|
| 3 |
-
Endpoints para importar dados de fontes externas
|
| 4 |
-
"""
|
| 5 |
-
from fastapi import APIRouter, Depends, HTTPException, BackgroundTasks
|
| 6 |
-
from sqlalchemy.orm import Session
|
| 7 |
-
from typing import Optional, List
|
| 8 |
-
from datetime import datetime
|
| 9 |
-
import asyncio
|
| 10 |
-
|
| 11 |
-
from app.api.deps import get_scoped_db
|
| 12 |
-
from app.models import Entity, Document, Relationship
|
| 13 |
-
from app.schemas import EntityResponse, DocumentResponse
|
| 14 |
-
from app.services.ingestion import wikipedia_scraper, news_service
|
| 15 |
-
from app.services.nlp import entity_extractor
|
| 16 |
-
from app.services.geocoding import geocode
|
| 17 |
-
|
| 18 |
-
router = APIRouter(prefix="/ingest", tags=["Data Ingestion"])
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
def parse_event_date(date_str):
|
| 22 |
-
"""Parse date string to datetime object"""
|
| 23 |
-
if not date_str:
|
| 24 |
-
return None
|
| 25 |
-
try:
|
| 26 |
-
# Try YYYY-MM-DD format
|
| 27 |
-
return datetime.strptime(date_str[:10], "%Y-%m-%d")
|
| 28 |
-
except:
|
| 29 |
-
try:
|
| 30 |
-
# Try YYYY format
|
| 31 |
-
return datetime.strptime(date_str[:4], "%Y")
|
| 32 |
-
except:
|
| 33 |
-
return None
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
# ========== Wikipedia ==========
|
| 37 |
-
|
| 38 |
-
@router.get("/wikipedia/search")
|
| 39 |
-
def search_wikipedia(q: str, limit: int = 10):
|
| 40 |
-
"""Busca artigos na Wikipedia"""
|
| 41 |
-
results = wikipedia_scraper.search(q, limit)
|
| 42 |
-
return results
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
@router.post("/wikipedia/entity", response_model=EntityResponse)
|
| 46 |
-
async def import_from_wikipedia(
|
| 47 |
-
title: str,
|
| 48 |
-
entity_type: str = "person",
|
| 49 |
-
project_id: Optional[str] = None,
|
| 50 |
-
auto_extract: bool = True,
|
| 51 |
-
db: Session = Depends(get_scoped_db)
|
| 52 |
-
):
|
| 53 |
-
"""
|
| 54 |
-
Importa uma entidade da Wikipedia
|
| 55 |
-
entity_type: person, organization, location
|
| 56 |
-
project_id: ID do projeto para associar a entidade
|
| 57 |
-
auto_extract: Se True, usa LLM para extrair entidades relacionadas
|
| 58 |
-
"""
|
| 59 |
-
# Check if entity already exists
|
| 60 |
-
existing = db.query(Entity).filter(
|
| 61 |
-
Entity.name == title,
|
| 62 |
-
Entity.source == "wikipedia"
|
| 63 |
-
).first()
|
| 64 |
-
|
| 65 |
-
if existing:
|
| 66 |
-
return existing
|
| 67 |
-
|
| 68 |
-
# Scrape based on type
|
| 69 |
-
if entity_type == "person":
|
| 70 |
-
data = wikipedia_scraper.scrape_person(title)
|
| 71 |
-
elif entity_type == "organization":
|
| 72 |
-
data = wikipedia_scraper.scrape_organization(title)
|
| 73 |
-
elif entity_type == "location":
|
| 74 |
-
data = wikipedia_scraper.scrape_location(title)
|
| 75 |
-
else:
|
| 76 |
-
data = wikipedia_scraper.scrape_person(title) # default
|
| 77 |
-
|
| 78 |
-
if not data:
|
| 79 |
-
raise HTTPException(status_code=404, detail="Article not found on Wikipedia")
|
| 80 |
-
|
| 81 |
-
# Create main entity with project_id
|
| 82 |
-
entity = Entity(**data)
|
| 83 |
-
entity.project_id = project_id
|
| 84 |
-
db.add(entity)
|
| 85 |
-
db.commit()
|
| 86 |
-
db.refresh(entity)
|
| 87 |
-
|
| 88 |
-
# Auto-extract entities and relationships using LLM
|
| 89 |
-
if auto_extract and data.get("description"):
|
| 90 |
-
try:
|
| 91 |
-
# Limit text to avoid token limits
|
| 92 |
-
text_to_analyze = data["description"][:3000]
|
| 93 |
-
result = await entity_extractor.extract(text_to_analyze)
|
| 94 |
-
|
| 95 |
-
# Create extracted entities
|
| 96 |
-
created_entities = {}
|
| 97 |
-
for ext_entity in result.entities:
|
| 98 |
-
# Skip if same as main entity
|
| 99 |
-
if ext_entity.name.lower() == title.lower():
|
| 100 |
-
created_entities[ext_entity.name] = entity
|
| 101 |
-
continue
|
| 102 |
-
|
| 103 |
-
# Check if entity exists (by similar name)
|
| 104 |
-
existing_ent = db.query(Entity).filter(
|
| 105 |
-
Entity.name.ilike(f"%{ext_entity.name}%")
|
| 106 |
-
).first()
|
| 107 |
-
|
| 108 |
-
if existing_ent:
|
| 109 |
-
created_entities[ext_entity.name] = existing_ent
|
| 110 |
-
else:
|
| 111 |
-
# Get coordinates for location entities
|
| 112 |
-
lat, lng = None, None
|
| 113 |
-
if ext_entity.type == "location":
|
| 114 |
-
coords = await geocode(ext_entity.name)
|
| 115 |
-
if coords:
|
| 116 |
-
lat, lng = coords
|
| 117 |
-
|
| 118 |
-
# Parse event_date
|
| 119 |
-
event_date = parse_event_date(getattr(ext_entity, 'event_date', None))
|
| 120 |
-
|
| 121 |
-
new_ent = Entity(
|
| 122 |
-
name=ext_entity.name,
|
| 123 |
-
type=ext_entity.type if ext_entity.type in ["person", "organization", "location", "event"] else "person",
|
| 124 |
-
description=ext_entity.description or ext_entity.role,
|
| 125 |
-
source="wikipedia_extraction",
|
| 126 |
-
latitude=lat,
|
| 127 |
-
longitude=lng,
|
| 128 |
-
event_date=event_date,
|
| 129 |
-
project_id=project_id,
|
| 130 |
-
properties={"role": ext_entity.role, "aliases": ext_entity.aliases, "extracted_from": title}
|
| 131 |
-
)
|
| 132 |
-
db.add(new_ent)
|
| 133 |
-
db.commit()
|
| 134 |
-
db.refresh(new_ent)
|
| 135 |
-
created_entities[ext_entity.name] = new_ent
|
| 136 |
-
|
| 137 |
-
# Create relationships
|
| 138 |
-
for rel in result.relationships:
|
| 139 |
-
source_ent = created_entities.get(rel.source) or db.query(Entity).filter(Entity.name.ilike(f"%{rel.source}%")).first()
|
| 140 |
-
target_ent = created_entities.get(rel.target) or db.query(Entity).filter(Entity.name.ilike(f"%{rel.target}%")).first()
|
| 141 |
-
|
| 142 |
-
if source_ent and target_ent and source_ent.id != target_ent.id:
|
| 143 |
-
# Check if relationship exists
|
| 144 |
-
existing_rel = db.query(Relationship).filter(
|
| 145 |
-
Relationship.source_id == source_ent.id,
|
| 146 |
-
Relationship.target_id == target_ent.id,
|
| 147 |
-
Relationship.type == rel.relationship_type
|
| 148 |
-
).first()
|
| 149 |
-
|
| 150 |
-
if not existing_rel:
|
| 151 |
-
# Parse relationship event_date
|
| 152 |
-
rel_event_date = parse_event_date(getattr(rel, 'event_date', None))
|
| 153 |
-
|
| 154 |
-
new_rel = Relationship(
|
| 155 |
-
source_id=source_ent.id,
|
| 156 |
-
target_id=target_ent.id,
|
| 157 |
-
type=rel.relationship_type,
|
| 158 |
-
event_date=rel_event_date,
|
| 159 |
-
properties={"context": rel.context, "extracted_from": title}
|
| 160 |
-
)
|
| 161 |
-
db.add(new_rel)
|
| 162 |
-
|
| 163 |
-
db.commit()
|
| 164 |
-
|
| 165 |
-
except Exception as e:
|
| 166 |
-
print(f"NER extraction error: {e}")
|
| 167 |
-
# Continue without extraction if it fails
|
| 168 |
-
|
| 169 |
-
return entity
|
| 170 |
-
|
| 171 |
-
|
| 172 |
-
# ========== News ==========
|
| 173 |
-
|
| 174 |
-
@router.get("/news/feeds")
|
| 175 |
-
def list_available_feeds():
|
| 176 |
-
"""Lista os feeds de notícias disponíveis"""
|
| 177 |
-
return list(news_service.RSS_FEEDS.keys())
|
| 178 |
-
|
| 179 |
-
|
| 180 |
-
@router.get("/news/fetch")
|
| 181 |
-
def fetch_news(feed: Optional[str] = None):
|
| 182 |
-
"""
|
| 183 |
-
Busca notícias dos feeds RSS
|
| 184 |
-
Se feed não for especificado, busca de todos
|
| 185 |
-
"""
|
| 186 |
-
if feed:
|
| 187 |
-
if feed not in news_service.RSS_FEEDS:
|
| 188 |
-
raise HTTPException(status_code=404, detail="Feed not found")
|
| 189 |
-
url = news_service.RSS_FEEDS[feed]
|
| 190 |
-
articles = news_service.fetch_feed(url)
|
| 191 |
-
else:
|
| 192 |
-
articles = news_service.fetch_all_feeds()
|
| 193 |
-
|
| 194 |
-
return articles
|
| 195 |
-
|
| 196 |
-
|
| 197 |
-
@router.get("/news/search")
|
| 198 |
-
def search_news(q: str):
|
| 199 |
-
"""Busca notícias por palavra-chave via Google News"""
|
| 200 |
-
return news_service.search_news(q)
|
| 201 |
-
|
| 202 |
-
|
| 203 |
-
@router.post("/news/import")
|
| 204 |
-
async def import_news(
|
| 205 |
-
query: Optional[str] = None,
|
| 206 |
-
feed: Optional[str] = None,
|
| 207 |
-
auto_extract: bool = True,
|
| 208 |
-
db: Session = Depends(get_scoped_db)
|
| 209 |
-
):
|
| 210 |
-
"""
|
| 211 |
-
Importa notícias como documentos no sistema
|
| 212 |
-
auto_extract: Se True, usa LLM para extrair entidades de cada notícia
|
| 213 |
-
"""
|
| 214 |
-
if query:
|
| 215 |
-
articles = news_service.search_news(query)
|
| 216 |
-
elif feed:
|
| 217 |
-
if feed not in news_service.RSS_FEEDS:
|
| 218 |
-
raise HTTPException(status_code=404, detail="Feed not found")
|
| 219 |
-
articles = news_service.fetch_feed(news_service.RSS_FEEDS[feed])
|
| 220 |
-
else:
|
| 221 |
-
articles = news_service.fetch_all_feeds()
|
| 222 |
-
|
| 223 |
-
imported = 0
|
| 224 |
-
extracted_entities = 0
|
| 225 |
-
|
| 226 |
-
for article in articles:
|
| 227 |
-
# Check if document already exists (by URL)
|
| 228 |
-
if article.get("url"):
|
| 229 |
-
existing = db.query(Document).filter(
|
| 230 |
-
Document.source_url == article["url"]
|
| 231 |
-
).first()
|
| 232 |
-
if existing:
|
| 233 |
-
continue
|
| 234 |
-
|
| 235 |
-
doc_data = news_service.to_document(article)
|
| 236 |
-
doc = Document(**doc_data)
|
| 237 |
-
db.add(doc)
|
| 238 |
-
db.commit()
|
| 239 |
-
imported += 1
|
| 240 |
-
|
| 241 |
-
# Extract entities from article content
|
| 242 |
-
if auto_extract:
|
| 243 |
-
try:
|
| 244 |
-
text_to_analyze = f"{article.get('title', '')} {article.get('description', '')}".strip()
|
| 245 |
-
if len(text_to_analyze) >= 20:
|
| 246 |
-
result = await entity_extractor.extract(text_to_analyze[:2000])
|
| 247 |
-
|
| 248 |
-
created_entities = {}
|
| 249 |
-
for ext_entity in result.entities:
|
| 250 |
-
# Check if entity exists
|
| 251 |
-
existing_ent = db.query(Entity).filter(
|
| 252 |
-
Entity.name.ilike(f"%{ext_entity.name}%")
|
| 253 |
-
).first()
|
| 254 |
-
|
| 255 |
-
if existing_ent:
|
| 256 |
-
created_entities[ext_entity.name] = existing_ent
|
| 257 |
-
else:
|
| 258 |
-
# Get coordinates for location entities
|
| 259 |
-
lat, lng = None, None
|
| 260 |
-
if ext_entity.type == "location":
|
| 261 |
-
coords = await geocode(ext_entity.name)
|
| 262 |
-
if coords:
|
| 263 |
-
lat, lng = coords
|
| 264 |
-
|
| 265 |
-
new_ent = Entity(
|
| 266 |
-
name=ext_entity.name,
|
| 267 |
-
type=ext_entity.type if ext_entity.type in ["person", "organization", "location", "event"] else "person",
|
| 268 |
-
description=ext_entity.description or ext_entity.role,
|
| 269 |
-
source="news_extraction",
|
| 270 |
-
latitude=lat,
|
| 271 |
-
longitude=lng,
|
| 272 |
-
properties={"role": ext_entity.role, "aliases": ext_entity.aliases, "from_article": article.get('title', '')}
|
| 273 |
-
)
|
| 274 |
-
db.add(new_ent)
|
| 275 |
-
db.commit()
|
| 276 |
-
db.refresh(new_ent)
|
| 277 |
-
created_entities[ext_entity.name] = new_ent
|
| 278 |
-
extracted_entities += 1
|
| 279 |
-
|
| 280 |
-
# Create relationships
|
| 281 |
-
for rel in result.relationships:
|
| 282 |
-
source_ent = created_entities.get(rel.source) or db.query(Entity).filter(Entity.name.ilike(f"%{rel.source}%")).first()
|
| 283 |
-
target_ent = created_entities.get(rel.target) or db.query(Entity).filter(Entity.name.ilike(f"%{rel.target}%")).first()
|
| 284 |
-
|
| 285 |
-
if source_ent and target_ent and source_ent.id != target_ent.id:
|
| 286 |
-
existing_rel = db.query(Relationship).filter(
|
| 287 |
-
Relationship.source_id == source_ent.id,
|
| 288 |
-
Relationship.target_id == target_ent.id,
|
| 289 |
-
Relationship.type == rel.relationship_type
|
| 290 |
-
).first()
|
| 291 |
-
|
| 292 |
-
if not existing_rel:
|
| 293 |
-
new_rel = Relationship(
|
| 294 |
-
source_id=source_ent.id,
|
| 295 |
-
target_id=target_ent.id,
|
| 296 |
-
type=rel.relationship_type,
|
| 297 |
-
properties={"context": rel.context}
|
| 298 |
-
)
|
| 299 |
-
db.add(new_rel)
|
| 300 |
-
|
| 301 |
-
db.commit()
|
| 302 |
-
|
| 303 |
-
except Exception as e:
|
| 304 |
-
print(f"NER extraction error for article: {e}")
|
| 305 |
-
# Continue without extraction
|
| 306 |
-
|
| 307 |
-
return {
|
| 308 |
-
"message": f"Imported {imported} articles",
|
| 309 |
-
"total_found": len(articles),
|
| 310 |
-
"extracted_entities": extracted_entities
|
| 311 |
-
}
|
| 312 |
-
|
| 313 |
-
|
| 314 |
-
# ========== Manual Import ==========
|
| 315 |
-
|
| 316 |
-
@router.post("/bulk/entities")
|
| 317 |
-
def bulk_import_entities(
|
| 318 |
-
entities: List[dict],
|
| 319 |
-
db: Session = Depends(get_scoped_db)
|
| 320 |
-
):
|
| 321 |
-
"""
|
| 322 |
-
Importa múltiplas entidades de uma vez
|
| 323 |
-
Útil para importar de CSV/JSON
|
| 324 |
-
"""
|
| 325 |
-
imported = 0
|
| 326 |
-
for entity_data in entities:
|
| 327 |
-
entity = Entity(
|
| 328 |
-
type=entity_data.get("type", "unknown"),
|
| 329 |
-
name=entity_data.get("name", "Unnamed"),
|
| 330 |
-
description=entity_data.get("description"),
|
| 331 |
-
properties=entity_data.get("properties", {}),
|
| 332 |
-
latitude=entity_data.get("latitude"),
|
| 333 |
-
longitude=entity_data.get("longitude"),
|
| 334 |
-
source=entity_data.get("source", "manual")
|
| 335 |
-
)
|
| 336 |
-
db.add(entity)
|
| 337 |
-
imported += 1
|
| 338 |
-
|
| 339 |
-
db.commit()
|
| 340 |
-
|
| 341 |
-
return {"message": f"Imported {imported} entities"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/investigate.py
DELETED
|
@@ -1,207 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Investigation API Routes - Build dossiers on companies and people
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, HTTPException, Depends
|
| 5 |
-
from pydantic import BaseModel, Field
|
| 6 |
-
from typing import Optional, List, Dict, Any
|
| 7 |
-
from sqlalchemy.orm import Session
|
| 8 |
-
|
| 9 |
-
from app.services.investigation import (
|
| 10 |
-
investigar_empresa,
|
| 11 |
-
investigar_pessoa,
|
| 12 |
-
dossier_to_dict
|
| 13 |
-
)
|
| 14 |
-
from app.services.brazil_apis import consultar_cnpj
|
| 15 |
-
from app.services.investigator_agent import investigator_agent
|
| 16 |
-
from app.api.deps import get_scoped_db
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
router = APIRouter(prefix="/investigate", tags=["Investigation"])
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
class InvestigateCompanyRequest(BaseModel):
|
| 23 |
-
"""Request to investigate a company"""
|
| 24 |
-
cnpj: str = Field(..., min_length=11, description="CNPJ da empresa")
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
class InvestigatePersonRequest(BaseModel):
|
| 28 |
-
"""Request to investigate a person"""
|
| 29 |
-
nome: str = Field(..., min_length=2, description="Nome da pessoa")
|
| 30 |
-
cpf: Optional[str] = Field(None, description="CPF (opcional)")
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
class DossierResponse(BaseModel):
|
| 34 |
-
"""Dossier response"""
|
| 35 |
-
tipo: str
|
| 36 |
-
alvo: str
|
| 37 |
-
cnpj_cpf: Optional[str]
|
| 38 |
-
red_flags: List[str]
|
| 39 |
-
score_risco: int
|
| 40 |
-
data_geracao: str
|
| 41 |
-
fonte_dados: List[str]
|
| 42 |
-
secoes: Dict[str, Any]
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
class CNPJResponse(BaseModel):
|
| 46 |
-
"""Quick CNPJ lookup response"""
|
| 47 |
-
cnpj: str
|
| 48 |
-
razao_social: str
|
| 49 |
-
nome_fantasia: str
|
| 50 |
-
situacao: str
|
| 51 |
-
data_abertura: str
|
| 52 |
-
capital_social: float
|
| 53 |
-
endereco: str
|
| 54 |
-
telefone: str
|
| 55 |
-
email: str
|
| 56 |
-
atividade: str
|
| 57 |
-
socios: List[Dict[str, Any]]
|
| 58 |
-
|
| 59 |
-
|
| 60 |
-
@router.post("/company", response_model=DossierResponse)
|
| 61 |
-
async def investigate_company(request: InvestigateCompanyRequest):
|
| 62 |
-
"""
|
| 63 |
-
Build a comprehensive dossier on a company.
|
| 64 |
-
|
| 65 |
-
Collects:
|
| 66 |
-
- Cadastral data from CNPJ
|
| 67 |
-
- Partners/owners
|
| 68 |
-
- Sanctions (CEIS, CNEP, CEPIM)
|
| 69 |
-
- News and media mentions
|
| 70 |
-
- Related entities
|
| 71 |
-
|
| 72 |
-
Returns risk score and red flags.
|
| 73 |
-
"""
|
| 74 |
-
try:
|
| 75 |
-
dossier = await investigar_empresa(request.cnpj)
|
| 76 |
-
return DossierResponse(**dossier_to_dict(dossier))
|
| 77 |
-
|
| 78 |
-
except Exception as e:
|
| 79 |
-
raise HTTPException(status_code=500, detail=str(e))
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
@router.post("/person", response_model=DossierResponse)
|
| 83 |
-
async def investigate_person(request: InvestigatePersonRequest):
|
| 84 |
-
"""
|
| 85 |
-
Build a dossier on a person.
|
| 86 |
-
|
| 87 |
-
Note: Due to LGPD, personal data is limited.
|
| 88 |
-
Mainly uses web search for public information.
|
| 89 |
-
"""
|
| 90 |
-
try:
|
| 91 |
-
dossier = await investigar_pessoa(request.nome, request.cpf)
|
| 92 |
-
return DossierResponse(**dossier_to_dict(dossier))
|
| 93 |
-
|
| 94 |
-
except Exception as e:
|
| 95 |
-
raise HTTPException(status_code=500, detail=str(e))
|
| 96 |
-
|
| 97 |
-
|
| 98 |
-
@router.get("/cnpj/{cnpj}", response_model=CNPJResponse)
|
| 99 |
-
async def lookup_cnpj(cnpj: str):
|
| 100 |
-
"""
|
| 101 |
-
Quick CNPJ lookup - returns basic company data.
|
| 102 |
-
"""
|
| 103 |
-
try:
|
| 104 |
-
data = await consultar_cnpj(cnpj)
|
| 105 |
-
|
| 106 |
-
if not data:
|
| 107 |
-
raise HTTPException(status_code=404, detail="CNPJ não encontrado")
|
| 108 |
-
|
| 109 |
-
return CNPJResponse(
|
| 110 |
-
cnpj=data.cnpj,
|
| 111 |
-
razao_social=data.razao_social,
|
| 112 |
-
nome_fantasia=data.nome_fantasia,
|
| 113 |
-
situacao=data.situacao,
|
| 114 |
-
data_abertura=data.data_abertura,
|
| 115 |
-
capital_social=data.capital_social,
|
| 116 |
-
endereco=f"{data.logradouro}, {data.numero} - {data.bairro}, {data.cidade}/{data.uf}",
|
| 117 |
-
telefone=data.telefone,
|
| 118 |
-
email=data.email,
|
| 119 |
-
atividade=f"{data.cnae_principal} - {data.cnae_descricao}",
|
| 120 |
-
socios=data.socios
|
| 121 |
-
)
|
| 122 |
-
|
| 123 |
-
except HTTPException:
|
| 124 |
-
raise
|
| 125 |
-
except Exception as e:
|
| 126 |
-
raise HTTPException(status_code=500, detail=str(e))
|
| 127 |
-
|
| 128 |
-
|
| 129 |
-
# ===========================================
|
| 130 |
-
# Autonomous Investigation Agent
|
| 131 |
-
# ===========================================
|
| 132 |
-
|
| 133 |
-
class AgentInvestigateRequest(BaseModel):
|
| 134 |
-
"""Request for autonomous investigation"""
|
| 135 |
-
mission: str = Field(..., min_length=5, description="Missão de investigação em linguagem natural")
|
| 136 |
-
max_iterations: int = Field(10, ge=1, le=20, description="Máximo de iterações do agente")
|
| 137 |
-
|
| 138 |
-
|
| 139 |
-
class FindingResponse(BaseModel):
|
| 140 |
-
"""A finding from investigation"""
|
| 141 |
-
title: str
|
| 142 |
-
content: str
|
| 143 |
-
source: str
|
| 144 |
-
timestamp: str
|
| 145 |
-
|
| 146 |
-
|
| 147 |
-
class AgentInvestigateResponse(BaseModel):
|
| 148 |
-
"""Response from autonomous investigation"""
|
| 149 |
-
mission: str
|
| 150 |
-
status: str
|
| 151 |
-
report: str
|
| 152 |
-
findings: List[FindingResponse]
|
| 153 |
-
entities_discovered: int
|
| 154 |
-
connections_mapped: int
|
| 155 |
-
iterations: int
|
| 156 |
-
tools_used: List[str]
|
| 157 |
-
|
| 158 |
-
|
| 159 |
-
@router.post("/agent", response_model=AgentInvestigateResponse)
|
| 160 |
-
async def investigate_with_agent(
|
| 161 |
-
request: AgentInvestigateRequest,
|
| 162 |
-
db: Session = Depends(get_scoped_db)
|
| 163 |
-
):
|
| 164 |
-
"""
|
| 165 |
-
Autonomous investigation with AI agent.
|
| 166 |
-
|
| 167 |
-
The agent will:
|
| 168 |
-
1. Search NUMIDIUM for existing entities
|
| 169 |
-
2. Query CNPJ data for Brazilian companies
|
| 170 |
-
3. Search the web for news and public info
|
| 171 |
-
4. Follow leads and connections
|
| 172 |
-
5. Generate a comprehensive report
|
| 173 |
-
|
| 174 |
-
Example missions:
|
| 175 |
-
- "Investigue a rede de empresas de João Silva"
|
| 176 |
-
- "Descubra os sócios da empresa CNPJ 11.222.333/0001-44"
|
| 177 |
-
- "Pesquise sobre a empresa XYZ e suas conexões"
|
| 178 |
-
"""
|
| 179 |
-
try:
|
| 180 |
-
result = await investigator_agent.investigate(
|
| 181 |
-
mission=request.mission,
|
| 182 |
-
db=db,
|
| 183 |
-
max_iterations=request.max_iterations
|
| 184 |
-
)
|
| 185 |
-
|
| 186 |
-
return AgentInvestigateResponse(
|
| 187 |
-
mission=result.mission,
|
| 188 |
-
status=result.status,
|
| 189 |
-
report=result.report,
|
| 190 |
-
findings=[
|
| 191 |
-
FindingResponse(
|
| 192 |
-
title=f.title,
|
| 193 |
-
content=f.content,
|
| 194 |
-
source=f.source,
|
| 195 |
-
timestamp=f.timestamp
|
| 196 |
-
)
|
| 197 |
-
for f in result.findings
|
| 198 |
-
],
|
| 199 |
-
entities_discovered=len(result.entities_discovered),
|
| 200 |
-
connections_mapped=len(result.connections_mapped),
|
| 201 |
-
iterations=result.iterations,
|
| 202 |
-
tools_used=result.tools_used
|
| 203 |
-
)
|
| 204 |
-
|
| 205 |
-
except Exception as e:
|
| 206 |
-
raise HTTPException(status_code=500, detail=str(e))
|
| 207 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/projects.py
DELETED
|
@@ -1,135 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Projects API Routes - Workspace management
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, Depends, HTTPException
|
| 5 |
-
from pydantic import BaseModel
|
| 6 |
-
from typing import Optional, List
|
| 7 |
-
from datetime import datetime
|
| 8 |
-
from sqlalchemy.orm import Session
|
| 9 |
-
|
| 10 |
-
from app.api.deps import get_scoped_db
|
| 11 |
-
from app.models import Project, Entity, Relationship
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
router = APIRouter(prefix="/projects", tags=["Projects"])
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
class ProjectCreate(BaseModel):
|
| 18 |
-
name: str
|
| 19 |
-
description: Optional[str] = None
|
| 20 |
-
color: str = "#00d4ff"
|
| 21 |
-
icon: str = "folder"
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
class ProjectResponse(BaseModel):
|
| 25 |
-
id: str
|
| 26 |
-
name: str
|
| 27 |
-
description: Optional[str]
|
| 28 |
-
color: str
|
| 29 |
-
icon: str
|
| 30 |
-
entity_count: int = 0
|
| 31 |
-
created_at: datetime
|
| 32 |
-
|
| 33 |
-
class Config:
|
| 34 |
-
from_attributes = True
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
@router.get("", response_model=List[ProjectResponse])
|
| 38 |
-
def list_projects(db: Session = Depends(get_scoped_db)):
|
| 39 |
-
"""List all projects"""
|
| 40 |
-
projects = db.query(Project).order_by(Project.created_at.desc()).all()
|
| 41 |
-
|
| 42 |
-
result = []
|
| 43 |
-
for p in projects:
|
| 44 |
-
entity_count = db.query(Entity).filter(Entity.project_id == p.id).count()
|
| 45 |
-
result.append(ProjectResponse(
|
| 46 |
-
id=p.id,
|
| 47 |
-
name=p.name,
|
| 48 |
-
description=p.description,
|
| 49 |
-
color=p.color,
|
| 50 |
-
icon=p.icon,
|
| 51 |
-
entity_count=entity_count,
|
| 52 |
-
created_at=p.created_at
|
| 53 |
-
))
|
| 54 |
-
|
| 55 |
-
return result
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
@router.post("", response_model=ProjectResponse)
|
| 59 |
-
def create_project(project: ProjectCreate, db: Session = Depends(get_scoped_db)):
|
| 60 |
-
"""Create a new project"""
|
| 61 |
-
new_project = Project(
|
| 62 |
-
name=project.name,
|
| 63 |
-
description=project.description,
|
| 64 |
-
color=project.color,
|
| 65 |
-
icon=project.icon
|
| 66 |
-
)
|
| 67 |
-
db.add(new_project)
|
| 68 |
-
db.commit()
|
| 69 |
-
db.refresh(new_project)
|
| 70 |
-
|
| 71 |
-
return ProjectResponse(
|
| 72 |
-
id=new_project.id,
|
| 73 |
-
name=new_project.name,
|
| 74 |
-
description=new_project.description,
|
| 75 |
-
color=new_project.color,
|
| 76 |
-
icon=new_project.icon,
|
| 77 |
-
entity_count=0,
|
| 78 |
-
created_at=new_project.created_at
|
| 79 |
-
)
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
@router.get("/{project_id}", response_model=ProjectResponse)
|
| 83 |
-
def get_project(project_id: str, db: Session = Depends(get_scoped_db)):
|
| 84 |
-
"""Get project by ID"""
|
| 85 |
-
project = db.query(Project).filter(Project.id == project_id).first()
|
| 86 |
-
|
| 87 |
-
if not project:
|
| 88 |
-
raise HTTPException(status_code=404, detail="Project not found")
|
| 89 |
-
|
| 90 |
-
entity_count = db.query(Entity).filter(Entity.project_id == project_id).count()
|
| 91 |
-
|
| 92 |
-
return ProjectResponse(
|
| 93 |
-
id=project.id,
|
| 94 |
-
name=project.name,
|
| 95 |
-
description=project.description,
|
| 96 |
-
color=project.color,
|
| 97 |
-
icon=project.icon,
|
| 98 |
-
entity_count=entity_count,
|
| 99 |
-
created_at=project.created_at
|
| 100 |
-
)
|
| 101 |
-
|
| 102 |
-
|
| 103 |
-
@router.delete("/{project_id}")
|
| 104 |
-
def delete_project(project_id: str, db: Session = Depends(get_scoped_db)):
|
| 105 |
-
"""Delete project and optionally its entities"""
|
| 106 |
-
project = db.query(Project).filter(Project.id == project_id).first()
|
| 107 |
-
|
| 108 |
-
if not project:
|
| 109 |
-
raise HTTPException(status_code=404, detail="Project not found")
|
| 110 |
-
|
| 111 |
-
# Set entities and relationships to no project (null)
|
| 112 |
-
db.query(Entity).filter(Entity.project_id == project_id).update({"project_id": None})
|
| 113 |
-
db.query(Relationship).filter(Relationship.project_id == project_id).update({"project_id": None})
|
| 114 |
-
|
| 115 |
-
db.delete(project)
|
| 116 |
-
db.commit()
|
| 117 |
-
|
| 118 |
-
return {"message": f"Project '{project.name}' deleted"}
|
| 119 |
-
|
| 120 |
-
|
| 121 |
-
@router.put("/{project_id}")
|
| 122 |
-
def update_project(project_id: str, project: ProjectCreate, db: Session = Depends(get_scoped_db)):
|
| 123 |
-
"""Update project"""
|
| 124 |
-
existing = db.query(Project).filter(Project.id == project_id).first()
|
| 125 |
-
|
| 126 |
-
if not existing:
|
| 127 |
-
raise HTTPException(status_code=404, detail="Project not found")
|
| 128 |
-
|
| 129 |
-
existing.name = project.name
|
| 130 |
-
existing.description = project.description
|
| 131 |
-
existing.color = project.color
|
| 132 |
-
existing.icon = project.icon
|
| 133 |
-
db.commit()
|
| 134 |
-
|
| 135 |
-
return {"message": "Project updated"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/relationships.py
DELETED
|
@@ -1,76 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Relationship CRUD Routes
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, Depends, HTTPException, Query
|
| 5 |
-
from sqlalchemy.orm import Session
|
| 6 |
-
from typing import List, Optional
|
| 7 |
-
|
| 8 |
-
from app.api.deps import get_scoped_db
|
| 9 |
-
from app.models import Relationship, Entity
|
| 10 |
-
from app.schemas import RelationshipCreate, RelationshipResponse
|
| 11 |
-
|
| 12 |
-
router = APIRouter(prefix="/relationships", tags=["Relationships"])
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
@router.get("/", response_model=List[RelationshipResponse])
|
| 16 |
-
def list_relationships(
|
| 17 |
-
type: Optional[str] = None,
|
| 18 |
-
source_id: Optional[str] = None,
|
| 19 |
-
target_id: Optional[str] = None,
|
| 20 |
-
limit: int = Query(default=50, le=200),
|
| 21 |
-
db: Session = Depends(get_scoped_db)
|
| 22 |
-
):
|
| 23 |
-
"""Lista relacionamentos com filtros opcionais"""
|
| 24 |
-
query = db.query(Relationship)
|
| 25 |
-
|
| 26 |
-
if type:
|
| 27 |
-
query = query.filter(Relationship.type == type)
|
| 28 |
-
if source_id:
|
| 29 |
-
query = query.filter(Relationship.source_id == source_id)
|
| 30 |
-
if target_id:
|
| 31 |
-
query = query.filter(Relationship.target_id == target_id)
|
| 32 |
-
|
| 33 |
-
return query.limit(limit).all()
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
@router.get("/types")
|
| 37 |
-
def get_relationship_types(db: Session = Depends(get_scoped_db)):
|
| 38 |
-
"""Retorna todos os tipos de relacionamento unicos"""
|
| 39 |
-
types = db.query(Relationship.type).distinct().all()
|
| 40 |
-
return [t[0] for t in types]
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
@router.post("/", response_model=RelationshipResponse, status_code=201)
|
| 44 |
-
def create_relationship(
|
| 45 |
-
rel: RelationshipCreate,
|
| 46 |
-
db: Session = Depends(get_scoped_db)
|
| 47 |
-
):
|
| 48 |
-
"""Cria um novo relacionamento entre entidades"""
|
| 49 |
-
source = db.query(Entity).filter(Entity.id == rel.source_id).first()
|
| 50 |
-
target = db.query(Entity).filter(Entity.id == rel.target_id).first()
|
| 51 |
-
|
| 52 |
-
if not source:
|
| 53 |
-
raise HTTPException(status_code=404, detail="Source entity not found")
|
| 54 |
-
if not target:
|
| 55 |
-
raise HTTPException(status_code=404, detail="Target entity not found")
|
| 56 |
-
|
| 57 |
-
db_rel = Relationship(**rel.model_dump())
|
| 58 |
-
db.add(db_rel)
|
| 59 |
-
db.commit()
|
| 60 |
-
db.refresh(db_rel)
|
| 61 |
-
return db_rel
|
| 62 |
-
|
| 63 |
-
|
| 64 |
-
@router.delete("/{relationship_id}")
|
| 65 |
-
def delete_relationship(
|
| 66 |
-
relationship_id: str,
|
| 67 |
-
db: Session = Depends(get_scoped_db)
|
| 68 |
-
):
|
| 69 |
-
"""Deleta um relacionamento"""
|
| 70 |
-
db_rel = db.query(Relationship).filter(Relationship.id == relationship_id).first()
|
| 71 |
-
if not db_rel:
|
| 72 |
-
raise HTTPException(status_code=404, detail="Relationship not found")
|
| 73 |
-
|
| 74 |
-
db.delete(db_rel)
|
| 75 |
-
db.commit()
|
| 76 |
-
return {"message": "Relationship deleted"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/research.py
DELETED
|
@@ -1,158 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Research API Routes - Deep research with automatic entity extraction
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, Depends, HTTPException
|
| 5 |
-
from pydantic import BaseModel, Field
|
| 6 |
-
from typing import Optional, List
|
| 7 |
-
import traceback
|
| 8 |
-
from sqlalchemy.orm import Session
|
| 9 |
-
|
| 10 |
-
from app.api.deps import get_scoped_db
|
| 11 |
-
from app.services import lancer
|
| 12 |
-
from app.services.nlp import entity_extractor
|
| 13 |
-
from app.services.geocoding import geocode
|
| 14 |
-
from app.models.entity import Entity, Relationship
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
router = APIRouter(prefix="/research", tags=["Research"])
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
class ResearchRequest(BaseModel):
|
| 21 |
-
"""Request model for research"""
|
| 22 |
-
query: str = Field(..., min_length=3, description="Research query")
|
| 23 |
-
mode: str = Field(default="search", description="Research mode: search, deep, heavy")
|
| 24 |
-
max_results: int = Field(default=10, le=20)
|
| 25 |
-
auto_extract: bool = Field(default=True, description="Auto-extract entities using NER")
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
class ResearchResponse(BaseModel):
|
| 29 |
-
"""Response model for research"""
|
| 30 |
-
query: str
|
| 31 |
-
answer: Optional[str]
|
| 32 |
-
sources: List[dict]
|
| 33 |
-
citations: List[dict]
|
| 34 |
-
extracted_entities: int
|
| 35 |
-
extracted_relationships: int
|
| 36 |
-
processing_time_ms: float
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
@router.post("", response_model=ResearchResponse)
|
| 40 |
-
async def research(request: ResearchRequest, db: Session = Depends(get_scoped_db)):
|
| 41 |
-
"""
|
| 42 |
-
Perform AI-powered research using Lancer API and optionally extract entities.
|
| 43 |
-
|
| 44 |
-
Modes:
|
| 45 |
-
- search: Fast search with AI synthesis
|
| 46 |
-
- deep: Multi-dimensional deep research (slower, more comprehensive)
|
| 47 |
-
- heavy: Search with full content scraping
|
| 48 |
-
"""
|
| 49 |
-
try:
|
| 50 |
-
# Call Lancer API based on mode
|
| 51 |
-
if request.mode == "deep":
|
| 52 |
-
result = await lancer.deep_research(request.query)
|
| 53 |
-
elif request.mode == "heavy":
|
| 54 |
-
result = await lancer.heavy_search(request.query, request.max_results)
|
| 55 |
-
else:
|
| 56 |
-
result = await lancer.search(request.query, request.max_results)
|
| 57 |
-
|
| 58 |
-
extracted_entities = 0
|
| 59 |
-
extracted_relationships = 0
|
| 60 |
-
|
| 61 |
-
# Extract entities if enabled
|
| 62 |
-
if request.auto_extract and result.raw_text:
|
| 63 |
-
try:
|
| 64 |
-
# Limit text to avoid token limits
|
| 65 |
-
text_to_analyze = result.raw_text[:5000]
|
| 66 |
-
ner_result = await entity_extractor.extract(text_to_analyze)
|
| 67 |
-
|
| 68 |
-
created_entities = {}
|
| 69 |
-
|
| 70 |
-
# Create entities
|
| 71 |
-
for entity in ner_result.entities:
|
| 72 |
-
# Check if exists
|
| 73 |
-
existing = db.query(Entity).filter(
|
| 74 |
-
Entity.name.ilike(f"%{entity.name}%")
|
| 75 |
-
).first()
|
| 76 |
-
|
| 77 |
-
if existing:
|
| 78 |
-
created_entities[entity.name] = existing
|
| 79 |
-
else:
|
| 80 |
-
# Geocode if location
|
| 81 |
-
lat, lng = None, None
|
| 82 |
-
if entity.type == "location":
|
| 83 |
-
coords = await geocode(entity.name)
|
| 84 |
-
if coords:
|
| 85 |
-
lat, lng = coords
|
| 86 |
-
|
| 87 |
-
new_entity = Entity(
|
| 88 |
-
name=entity.name,
|
| 89 |
-
type=entity.type if entity.type in ["person", "organization", "location", "event"] else "person",
|
| 90 |
-
description=entity.description or entity.role or "",
|
| 91 |
-
source="lancer_research",
|
| 92 |
-
latitude=lat,
|
| 93 |
-
longitude=lng,
|
| 94 |
-
properties={
|
| 95 |
-
"role": entity.role,
|
| 96 |
-
"aliases": entity.aliases,
|
| 97 |
-
"research_query": request.query
|
| 98 |
-
}
|
| 99 |
-
)
|
| 100 |
-
db.add(new_entity)
|
| 101 |
-
db.commit()
|
| 102 |
-
db.refresh(new_entity)
|
| 103 |
-
created_entities[entity.name] = new_entity
|
| 104 |
-
extracted_entities += 1
|
| 105 |
-
|
| 106 |
-
# Create relationships
|
| 107 |
-
for rel in ner_result.relationships:
|
| 108 |
-
source_ent = created_entities.get(rel.source) or db.query(Entity).filter(Entity.name.ilike(f"%{rel.source}%")).first()
|
| 109 |
-
target_ent = created_entities.get(rel.target) or db.query(Entity).filter(Entity.name.ilike(f"%{rel.target}%")).first()
|
| 110 |
-
|
| 111 |
-
if source_ent and target_ent and source_ent.id != target_ent.id:
|
| 112 |
-
existing_rel = db.query(Relationship).filter(
|
| 113 |
-
Relationship.source_id == source_ent.id,
|
| 114 |
-
Relationship.target_id == target_ent.id,
|
| 115 |
-
Relationship.type == rel.relationship_type
|
| 116 |
-
).first()
|
| 117 |
-
|
| 118 |
-
if not existing_rel:
|
| 119 |
-
new_rel = Relationship(
|
| 120 |
-
source_id=source_ent.id,
|
| 121 |
-
target_id=target_ent.id,
|
| 122 |
-
type=rel.relationship_type,
|
| 123 |
-
properties={"context": rel.context, "research_query": request.query}
|
| 124 |
-
)
|
| 125 |
-
db.add(new_rel)
|
| 126 |
-
extracted_relationships += 1
|
| 127 |
-
|
| 128 |
-
db.commit()
|
| 129 |
-
|
| 130 |
-
except Exception as e:
|
| 131 |
-
print(f"NER extraction error: {e}")
|
| 132 |
-
traceback.print_exc()
|
| 133 |
-
|
| 134 |
-
# Prepare sources for response
|
| 135 |
-
sources = [
|
| 136 |
-
{
|
| 137 |
-
"title": r.title,
|
| 138 |
-
"url": r.url,
|
| 139 |
-
"content": r.content[:300] if r.content else "",
|
| 140 |
-
"score": r.score
|
| 141 |
-
}
|
| 142 |
-
for r in result.results[:10]
|
| 143 |
-
]
|
| 144 |
-
|
| 145 |
-
return ResearchResponse(
|
| 146 |
-
query=result.query,
|
| 147 |
-
answer=result.answer,
|
| 148 |
-
sources=sources,
|
| 149 |
-
citations=result.citations,
|
| 150 |
-
extracted_entities=extracted_entities,
|
| 151 |
-
extracted_relationships=extracted_relationships,
|
| 152 |
-
processing_time_ms=result.processing_time_ms
|
| 153 |
-
)
|
| 154 |
-
|
| 155 |
-
except Exception as e:
|
| 156 |
-
print(f"Research error: {e}")
|
| 157 |
-
traceback.print_exc()
|
| 158 |
-
raise HTTPException(status_code=500, detail=str(e))
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/search.py
DELETED
|
@@ -1,126 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Search and Analytics Routes
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, Depends, Query
|
| 5 |
-
from sqlalchemy.orm import Session
|
| 6 |
-
from sqlalchemy import or_, func
|
| 7 |
-
from typing import Optional
|
| 8 |
-
|
| 9 |
-
from app.api.deps import get_scoped_db
|
| 10 |
-
from app.models import Entity, Relationship, Event, Document
|
| 11 |
-
from app.schemas import SearchResult, SystemStats
|
| 12 |
-
|
| 13 |
-
router = APIRouter(prefix="/search", tags=["Search"])
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
@router.get("", response_model=SearchResult)
|
| 17 |
-
def global_search(
|
| 18 |
-
q: str = Query(..., min_length=2, description="Search query"),
|
| 19 |
-
types: Optional[str] = Query(None, description="Entity types (comma-separated)"),
|
| 20 |
-
limit: int = Query(default=20, le=100),
|
| 21 |
-
db: Session = Depends(get_scoped_db)
|
| 22 |
-
):
|
| 23 |
-
"""
|
| 24 |
-
Busca global em todas as entidades, eventos e documentos.
|
| 25 |
-
"""
|
| 26 |
-
search_term = f"%{q}%"
|
| 27 |
-
type_filter = types.split(",") if types else None
|
| 28 |
-
|
| 29 |
-
entity_query = db.query(Entity).filter(
|
| 30 |
-
or_(
|
| 31 |
-
Entity.name.ilike(search_term),
|
| 32 |
-
Entity.description.ilike(search_term)
|
| 33 |
-
)
|
| 34 |
-
)
|
| 35 |
-
if type_filter:
|
| 36 |
-
entity_query = entity_query.filter(Entity.type.in_(type_filter))
|
| 37 |
-
entities = entity_query.limit(limit).all()
|
| 38 |
-
|
| 39 |
-
events = db.query(Event).filter(
|
| 40 |
-
or_(
|
| 41 |
-
Event.title.ilike(search_term),
|
| 42 |
-
Event.description.ilike(search_term)
|
| 43 |
-
)
|
| 44 |
-
).limit(limit).all()
|
| 45 |
-
|
| 46 |
-
documents = db.query(Document).filter(
|
| 47 |
-
or_(
|
| 48 |
-
Document.title.ilike(search_term),
|
| 49 |
-
Document.content.ilike(search_term)
|
| 50 |
-
)
|
| 51 |
-
).limit(limit).all()
|
| 52 |
-
|
| 53 |
-
return SearchResult(
|
| 54 |
-
entities=entities,
|
| 55 |
-
events=events,
|
| 56 |
-
documents=documents
|
| 57 |
-
)
|
| 58 |
-
|
| 59 |
-
|
| 60 |
-
@router.get("/stats", response_model=SystemStats)
|
| 61 |
-
def get_system_stats(db: Session = Depends(get_scoped_db)):
|
| 62 |
-
"""
|
| 63 |
-
Retorna estatisticas gerais do sistema.
|
| 64 |
-
"""
|
| 65 |
-
total_entities = db.query(Entity).count()
|
| 66 |
-
total_relationships = db.query(Relationship).count()
|
| 67 |
-
total_events = db.query(Event).count()
|
| 68 |
-
total_documents = db.query(Document).count()
|
| 69 |
-
|
| 70 |
-
type_counts = db.query(
|
| 71 |
-
Entity.type,
|
| 72 |
-
func.count(Entity.id)
|
| 73 |
-
).group_by(Entity.type).all()
|
| 74 |
-
|
| 75 |
-
entities_by_type = {t: c for t, c in type_counts}
|
| 76 |
-
|
| 77 |
-
recent = db.query(Entity).order_by(Entity.created_at.desc()).limit(10).all()
|
| 78 |
-
recent_activity = [
|
| 79 |
-
{
|
| 80 |
-
"id": e.id,
|
| 81 |
-
"type": e.type,
|
| 82 |
-
"name": e.name,
|
| 83 |
-
"created_at": e.created_at.isoformat()
|
| 84 |
-
}
|
| 85 |
-
for e in recent
|
| 86 |
-
]
|
| 87 |
-
|
| 88 |
-
return SystemStats(
|
| 89 |
-
total_entities=total_entities,
|
| 90 |
-
total_relationships=total_relationships,
|
| 91 |
-
total_events=total_events,
|
| 92 |
-
total_documents=total_documents,
|
| 93 |
-
entities_by_type=entities_by_type,
|
| 94 |
-
recent_activity=recent_activity
|
| 95 |
-
)
|
| 96 |
-
|
| 97 |
-
|
| 98 |
-
@router.get("/geo")
|
| 99 |
-
def get_geo_data(
|
| 100 |
-
entity_type: Optional[str] = None,
|
| 101 |
-
db: Session = Depends(get_scoped_db)
|
| 102 |
-
):
|
| 103 |
-
"""
|
| 104 |
-
Retorna entidades com geolocalizacao.
|
| 105 |
-
"""
|
| 106 |
-
query = db.query(Entity).filter(
|
| 107 |
-
Entity.latitude.isnot(None),
|
| 108 |
-
Entity.longitude.isnot(None)
|
| 109 |
-
)
|
| 110 |
-
|
| 111 |
-
if entity_type:
|
| 112 |
-
query = query.filter(Entity.type == entity_type)
|
| 113 |
-
|
| 114 |
-
entities = query.all()
|
| 115 |
-
|
| 116 |
-
return [
|
| 117 |
-
{
|
| 118 |
-
"id": e.id,
|
| 119 |
-
"type": e.type,
|
| 120 |
-
"name": e.name,
|
| 121 |
-
"lat": e.latitude,
|
| 122 |
-
"lng": e.longitude,
|
| 123 |
-
"properties": e.properties
|
| 124 |
-
}
|
| 125 |
-
for e in entities
|
| 126 |
-
]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/session.py
DELETED
|
@@ -1,44 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Session management routes
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, Header, Cookie, Response, Request
|
| 5 |
-
from typing import Optional
|
| 6 |
-
import uuid
|
| 7 |
-
|
| 8 |
-
from app.core.database import create_new_session_id
|
| 9 |
-
from app.config import settings
|
| 10 |
-
|
| 11 |
-
router = APIRouter(prefix="/session", tags=["Session"])
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
@router.post("/create")
|
| 15 |
-
def create_session(response: Response, request: Request):
|
| 16 |
-
"""Create a new session and return session_id"""
|
| 17 |
-
session_id = create_new_session_id()
|
| 18 |
-
secure = settings.cookie_secure
|
| 19 |
-
samesite = settings.cookie_samesite
|
| 20 |
-
proto = request.headers.get("x-forwarded-proto", request.url.scheme)
|
| 21 |
-
if proto != "https" and secure:
|
| 22 |
-
secure = False
|
| 23 |
-
samesite = "lax"
|
| 24 |
-
response.set_cookie(
|
| 25 |
-
key="numidium_session",
|
| 26 |
-
value=session_id,
|
| 27 |
-
max_age=60*60*24*365, # 1 year
|
| 28 |
-
httponly=True,
|
| 29 |
-
samesite=samesite,
|
| 30 |
-
secure=secure
|
| 31 |
-
)
|
| 32 |
-
return {"session_id": session_id}
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
@router.get("/current")
|
| 36 |
-
def get_current_session(
|
| 37 |
-
numidium_session: Optional[str] = Cookie(None),
|
| 38 |
-
x_session_id: Optional[str] = Header(None)
|
| 39 |
-
):
|
| 40 |
-
"""Get current session ID"""
|
| 41 |
-
session_id = x_session_id or numidium_session
|
| 42 |
-
if not session_id:
|
| 43 |
-
return {"session_id": None, "message": "No session. Call POST /session/create"}
|
| 44 |
-
return {"session_id": session_id}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/api/routes/timeline.py
DELETED
|
@@ -1,165 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Timeline API Routes - Temporal view of entities and relationships
|
| 3 |
-
"""
|
| 4 |
-
from fastapi import APIRouter, Depends, Query
|
| 5 |
-
from pydantic import BaseModel
|
| 6 |
-
from typing import Optional, List, Dict, Any
|
| 7 |
-
from datetime import datetime, timedelta
|
| 8 |
-
from collections import defaultdict
|
| 9 |
-
from sqlalchemy.orm import Session
|
| 10 |
-
|
| 11 |
-
from app.api.deps import get_scoped_db
|
| 12 |
-
from app.models.entity import Entity, Relationship
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
router = APIRouter(prefix="/timeline", tags=["Timeline"])
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
class TimelineEvent(BaseModel):
|
| 19 |
-
id: str
|
| 20 |
-
type: str # "entity" or "relationship"
|
| 21 |
-
entity_type: Optional[str] = None
|
| 22 |
-
name: str
|
| 23 |
-
description: Optional[str] = None
|
| 24 |
-
date: str
|
| 25 |
-
icon: str
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
class TimelineGroup(BaseModel):
|
| 29 |
-
date: str
|
| 30 |
-
label: str
|
| 31 |
-
events: List[TimelineEvent]
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
class TimelineResponse(BaseModel):
|
| 35 |
-
groups: List[TimelineGroup]
|
| 36 |
-
total_events: int
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
@router.get("", response_model=TimelineResponse)
|
| 40 |
-
async def get_timeline(
|
| 41 |
-
days: int = Query(default=30, ge=1, le=365),
|
| 42 |
-
entity_type: Optional[str] = None,
|
| 43 |
-
limit: int = Query(default=100, ge=1, le=500),
|
| 44 |
-
db: Session = Depends(get_scoped_db)
|
| 45 |
-
):
|
| 46 |
-
"""
|
| 47 |
-
Get timeline of recent entities and relationships.
|
| 48 |
-
Groups events by date.
|
| 49 |
-
"""
|
| 50 |
-
# Calculate date range
|
| 51 |
-
end_date = datetime.now()
|
| 52 |
-
start_date = end_date - timedelta(days=days)
|
| 53 |
-
|
| 54 |
-
events = []
|
| 55 |
-
|
| 56 |
-
# Get entities
|
| 57 |
-
query = db.query(Entity).filter(
|
| 58 |
-
Entity.created_at >= start_date
|
| 59 |
-
)
|
| 60 |
-
|
| 61 |
-
if entity_type:
|
| 62 |
-
query = query.filter(Entity.type == entity_type)
|
| 63 |
-
|
| 64 |
-
entities = query.order_by(Entity.created_at.desc()).limit(limit).all()
|
| 65 |
-
|
| 66 |
-
icon_map = {
|
| 67 |
-
"person": "👤",
|
| 68 |
-
"organization": "🏢",
|
| 69 |
-
"location": "📍",
|
| 70 |
-
"event": "📅",
|
| 71 |
-
"concept": "💡",
|
| 72 |
-
"product": "📦"
|
| 73 |
-
}
|
| 74 |
-
|
| 75 |
-
for e in entities:
|
| 76 |
-
# Prefer event_date over created_at
|
| 77 |
-
date = e.event_date if e.event_date else e.created_at
|
| 78 |
-
events.append(TimelineEvent(
|
| 79 |
-
id=e.id,
|
| 80 |
-
type="entity",
|
| 81 |
-
entity_type=e.type,
|
| 82 |
-
name=e.name,
|
| 83 |
-
description=e.description[:100] if e.description else None,
|
| 84 |
-
date=date.isoformat() if date else datetime.now().isoformat(),
|
| 85 |
-
icon=icon_map.get(e.type, "📄")
|
| 86 |
-
))
|
| 87 |
-
|
| 88 |
-
# Get relationships
|
| 89 |
-
relationships = db.query(Relationship).filter(
|
| 90 |
-
Relationship.created_at >= start_date
|
| 91 |
-
).order_by(Relationship.created_at.desc()).limit(limit // 2).all()
|
| 92 |
-
|
| 93 |
-
for r in relationships:
|
| 94 |
-
source = db.query(Entity).filter(Entity.id == r.source_id).first()
|
| 95 |
-
target = db.query(Entity).filter(Entity.id == r.target_id).first()
|
| 96 |
-
|
| 97 |
-
if source and target:
|
| 98 |
-
# Prefer event_date over created_at
|
| 99 |
-
date = r.event_date if r.event_date else r.created_at
|
| 100 |
-
events.append(TimelineEvent(
|
| 101 |
-
id=r.id,
|
| 102 |
-
type="relationship",
|
| 103 |
-
name=f"{source.name} → {target.name}",
|
| 104 |
-
description=r.type,
|
| 105 |
-
date=date.isoformat() if date else datetime.now().isoformat(),
|
| 106 |
-
icon="🔗"
|
| 107 |
-
))
|
| 108 |
-
|
| 109 |
-
# Sort by date
|
| 110 |
-
events.sort(key=lambda x: x.date, reverse=True)
|
| 111 |
-
|
| 112 |
-
# Group by date
|
| 113 |
-
groups_dict = defaultdict(list)
|
| 114 |
-
for event in events:
|
| 115 |
-
date_key = event.date[:10] # YYYY-MM-DD
|
| 116 |
-
groups_dict[date_key].append(event)
|
| 117 |
-
|
| 118 |
-
# Format groups
|
| 119 |
-
groups = []
|
| 120 |
-
for date_key in sorted(groups_dict.keys(), reverse=True):
|
| 121 |
-
try:
|
| 122 |
-
dt = datetime.fromisoformat(date_key)
|
| 123 |
-
label = dt.strftime("%d %b %Y")
|
| 124 |
-
except:
|
| 125 |
-
label = date_key
|
| 126 |
-
|
| 127 |
-
groups.append(TimelineGroup(
|
| 128 |
-
date=date_key,
|
| 129 |
-
label=label,
|
| 130 |
-
events=groups_dict[date_key]
|
| 131 |
-
))
|
| 132 |
-
|
| 133 |
-
return TimelineResponse(
|
| 134 |
-
groups=groups,
|
| 135 |
-
total_events=len(events)
|
| 136 |
-
)
|
| 137 |
-
|
| 138 |
-
|
| 139 |
-
@router.get("/stats")
|
| 140 |
-
async def get_timeline_stats(db: Session = Depends(get_scoped_db)):
|
| 141 |
-
"""Get statistics for timeline visualization"""
|
| 142 |
-
|
| 143 |
-
# Count entities by type
|
| 144 |
-
entity_counts = {}
|
| 145 |
-
for entity_type in ["person", "organization", "location", "event", "concept"]:
|
| 146 |
-
count = db.query(Entity).filter(Entity.type == entity_type).count()
|
| 147 |
-
entity_counts[entity_type] = count
|
| 148 |
-
|
| 149 |
-
# Count relationships
|
| 150 |
-
relationship_count = db.query(Relationship).count()
|
| 151 |
-
|
| 152 |
-
# Recent activity (last 7 days)
|
| 153 |
-
week_ago = datetime.now() - timedelta(days=7)
|
| 154 |
-
recent_entities = db.query(Entity).filter(Entity.created_at >= week_ago).count()
|
| 155 |
-
recent_relationships = db.query(Relationship).filter(Relationship.created_at >= week_ago).count()
|
| 156 |
-
|
| 157 |
-
return {
|
| 158 |
-
"entity_counts": entity_counts,
|
| 159 |
-
"relationship_count": relationship_count,
|
| 160 |
-
"recent_activity": {
|
| 161 |
-
"entities": recent_entities,
|
| 162 |
-
"relationships": recent_relationships,
|
| 163 |
-
"total": recent_entities + recent_relationships
|
| 164 |
-
}
|
| 165 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/config.py
DELETED
|
@@ -1,47 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Numidium Backend Configuration
|
| 3 |
-
"""
|
| 4 |
-
from pydantic_settings import BaseSettings
|
| 5 |
-
from functools import lru_cache
|
| 6 |
-
import os
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
class Settings(BaseSettings):
|
| 10 |
-
"""Application settings"""
|
| 11 |
-
|
| 12 |
-
# App Info
|
| 13 |
-
app_name: str = "Numidium"
|
| 14 |
-
app_version: str = "0.1.0"
|
| 15 |
-
debug: bool = False
|
| 16 |
-
|
| 17 |
-
# Database
|
| 18 |
-
database_url: str = "sqlite:///./data/numidium.db"
|
| 19 |
-
|
| 20 |
-
# APIs (opcional - pode configurar depois)
|
| 21 |
-
newsapi_key: str = ""
|
| 22 |
-
|
| 23 |
-
# Cerebras API for LLM-based entity extraction
|
| 24 |
-
cerebras_api_key: str = ""
|
| 25 |
-
|
| 26 |
-
# AetherMap API for semantic search and NER
|
| 27 |
-
aethermap_url: str = "https://madras1-aethermap.hf.space"
|
| 28 |
-
|
| 29 |
-
# CORS
|
| 30 |
-
cors_origins: list[str] = ["*"]
|
| 31 |
-
|
| 32 |
-
# Session cookie
|
| 33 |
-
cookie_secure: bool = True
|
| 34 |
-
cookie_samesite: str = "none"
|
| 35 |
-
|
| 36 |
-
class Config:
|
| 37 |
-
env_file = ".env"
|
| 38 |
-
env_file_encoding = "utf-8"
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
@lru_cache()
|
| 42 |
-
def get_settings() -> Settings:
|
| 43 |
-
"""Get cached settings"""
|
| 44 |
-
return Settings()
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
settings = get_settings()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/core/__init__.py
DELETED
|
@@ -1,2 +0,0 @@
|
|
| 1 |
-
# Core module
|
| 2 |
-
from app.core.database import get_db, init_db, Base
|
|
|
|
|
|
|
|
|
app/core/__pycache__/__init__.cpython-311.pyc
DELETED
|
Binary file (270 Bytes)
|
|
|
app/core/__pycache__/database.cpython-311.pyc
DELETED
|
Binary file (5.58 kB)
|
|
|
app/core/database.py
DELETED
|
@@ -1,115 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Database configuration and session management
|
| 3 |
-
Per-session databases - each user session gets its own SQLite file
|
| 4 |
-
"""
|
| 5 |
-
from sqlalchemy import create_engine, text
|
| 6 |
-
from sqlalchemy.ext.declarative import declarative_base
|
| 7 |
-
from sqlalchemy.orm import sessionmaker, Session
|
| 8 |
-
from typing import Optional
|
| 9 |
-
import os
|
| 10 |
-
import uuid
|
| 11 |
-
|
| 12 |
-
# Ensure data directory exists
|
| 13 |
-
os.makedirs("data/sessions", exist_ok=True)
|
| 14 |
-
|
| 15 |
-
# Base class for models
|
| 16 |
-
Base = declarative_base()
|
| 17 |
-
|
| 18 |
-
# Cache for session engines
|
| 19 |
-
_session_engines = {}
|
| 20 |
-
_session_makers = {}
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
def get_session_engine(session_id: str):
|
| 24 |
-
"""Get or create engine for a specific session"""
|
| 25 |
-
if session_id not in _session_engines:
|
| 26 |
-
db_path = f"data/sessions/{session_id}.db"
|
| 27 |
-
engine = create_engine(
|
| 28 |
-
f"sqlite:///./{db_path}",
|
| 29 |
-
connect_args={"check_same_thread": False}
|
| 30 |
-
)
|
| 31 |
-
_session_engines[session_id] = engine
|
| 32 |
-
_session_makers[session_id] = sessionmaker(autocommit=False, autoflush=False, bind=engine)
|
| 33 |
-
|
| 34 |
-
# Initialize tables for this session
|
| 35 |
-
Base.metadata.create_all(bind=engine)
|
| 36 |
-
_run_migrations(engine)
|
| 37 |
-
|
| 38 |
-
return _session_engines[session_id]
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
def get_session_db(session_id: str):
|
| 42 |
-
"""Get database session for a specific user session"""
|
| 43 |
-
get_session_engine(session_id) # Ensure engine exists
|
| 44 |
-
SessionLocal = _session_makers[session_id]
|
| 45 |
-
db = SessionLocal()
|
| 46 |
-
try:
|
| 47 |
-
yield db
|
| 48 |
-
finally:
|
| 49 |
-
db.close()
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
def get_db_for_session(session_id: str) -> Session:
|
| 53 |
-
"""Direct session getter (non-generator) for routes"""
|
| 54 |
-
get_session_engine(session_id)
|
| 55 |
-
SessionLocal = _session_makers[session_id]
|
| 56 |
-
return SessionLocal()
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
# Legacy - default database for backwards compatibility
|
| 60 |
-
from app.config import settings
|
| 61 |
-
engine = create_engine(
|
| 62 |
-
settings.database_url,
|
| 63 |
-
connect_args={"check_same_thread": False}
|
| 64 |
-
)
|
| 65 |
-
SessionLocal = sessionmaker(autocommit=False, autoflush=False, bind=engine)
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
def get_default_session() -> Session:
|
| 69 |
-
"""Create a new session for the default database."""
|
| 70 |
-
return SessionLocal()
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
def get_db():
|
| 74 |
-
"""Legacy: Default database session"""
|
| 75 |
-
db = get_default_session()
|
| 76 |
-
try:
|
| 77 |
-
yield db
|
| 78 |
-
finally:
|
| 79 |
-
db.close()
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
def _run_migrations(eng):
|
| 83 |
-
"""Run migrations on an engine"""
|
| 84 |
-
with eng.connect() as conn:
|
| 85 |
-
try:
|
| 86 |
-
conn.execute(text("ALTER TABLE entities ADD COLUMN event_date DATETIME"))
|
| 87 |
-
conn.commit()
|
| 88 |
-
except Exception:
|
| 89 |
-
pass
|
| 90 |
-
try:
|
| 91 |
-
conn.execute(text("ALTER TABLE relationships ADD COLUMN event_date DATETIME"))
|
| 92 |
-
conn.commit()
|
| 93 |
-
except Exception:
|
| 94 |
-
pass
|
| 95 |
-
try:
|
| 96 |
-
conn.execute(text("ALTER TABLE entities ADD COLUMN project_id VARCHAR(36)"))
|
| 97 |
-
conn.commit()
|
| 98 |
-
except Exception:
|
| 99 |
-
pass
|
| 100 |
-
try:
|
| 101 |
-
conn.execute(text("ALTER TABLE relationships ADD COLUMN project_id VARCHAR(36)"))
|
| 102 |
-
conn.commit()
|
| 103 |
-
except Exception:
|
| 104 |
-
pass
|
| 105 |
-
|
| 106 |
-
|
| 107 |
-
def init_db():
|
| 108 |
-
"""Initialize default database tables"""
|
| 109 |
-
Base.metadata.create_all(bind=engine)
|
| 110 |
-
_run_migrations(engine)
|
| 111 |
-
|
| 112 |
-
|
| 113 |
-
def create_new_session_id() -> str:
|
| 114 |
-
"""Generate a new session ID"""
|
| 115 |
-
return str(uuid.uuid4())
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/main.py
DELETED
|
@@ -1,99 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Numidium Backend - Main Application
|
| 3 |
-
Plataforma de Inteligência e Análise de Dados
|
| 4 |
-
"""
|
| 5 |
-
from fastapi import FastAPI
|
| 6 |
-
from fastapi.middleware.cors import CORSMiddleware
|
| 7 |
-
from contextlib import asynccontextmanager
|
| 8 |
-
|
| 9 |
-
from app.config import settings
|
| 10 |
-
from app.core.database import init_db
|
| 11 |
-
from app.api.routes import entities, relationships, events, search, ingest, analyze, graph, research, chat, investigate, dados_publicos, timeline, session, aethermap
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
@asynccontextmanager
|
| 15 |
-
async def lifespan(app: FastAPI):
|
| 16 |
-
"""Startup and shutdown events"""
|
| 17 |
-
# Startup: Initialize database
|
| 18 |
-
init_db()
|
| 19 |
-
print("🚀 Numidium Backend started!")
|
| 20 |
-
print(f"📊 Database: {settings.database_url}")
|
| 21 |
-
yield
|
| 22 |
-
# Shutdown
|
| 23 |
-
print("👋 Numidium Backend shutting down...")
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
# Create FastAPI app
|
| 27 |
-
app = FastAPI(
|
| 28 |
-
title="Numidium API",
|
| 29 |
-
description="""
|
| 30 |
-
## 🔮 Sistema de Inteligência e Análise de Dados
|
| 31 |
-
|
| 32 |
-
Backend do VANTAGE - Uma plataforma para:
|
| 33 |
-
- 📥 Ingestão de dados de múltiplas fontes (Wikipedia, News, Manual)
|
| 34 |
-
- 🔗 Mapeamento de conexões entre entidades
|
| 35 |
-
- 🗺️ Visualização geográfica
|
| 36 |
-
- 📊 Análise de grafos e relacionamentos
|
| 37 |
-
- 🔍 Busca global
|
| 38 |
-
""",
|
| 39 |
-
version=settings.app_version,
|
| 40 |
-
lifespan=lifespan
|
| 41 |
-
)
|
| 42 |
-
|
| 43 |
-
# CORS middleware
|
| 44 |
-
app.add_middleware(
|
| 45 |
-
CORSMiddleware,
|
| 46 |
-
allow_origins=settings.cors_origins,
|
| 47 |
-
allow_credentials=True,
|
| 48 |
-
allow_methods=["*"],
|
| 49 |
-
allow_headers=["*"],
|
| 50 |
-
)
|
| 51 |
-
|
| 52 |
-
# Include routers
|
| 53 |
-
app.include_router(entities.router, prefix="/api/v1")
|
| 54 |
-
app.include_router(relationships.router, prefix="/api/v1")
|
| 55 |
-
app.include_router(events.router, prefix="/api/v1")
|
| 56 |
-
app.include_router(search.router, prefix="/api/v1")
|
| 57 |
-
app.include_router(ingest.router, prefix="/api/v1")
|
| 58 |
-
app.include_router(analyze.router, prefix="/api/v1")
|
| 59 |
-
app.include_router(graph.router, prefix="/api/v1")
|
| 60 |
-
app.include_router(research.router, prefix="/api/v1")
|
| 61 |
-
app.include_router(chat.router, prefix="/api/v1")
|
| 62 |
-
app.include_router(investigate.router, prefix="/api/v1")
|
| 63 |
-
app.include_router(dados_publicos.router, prefix="/api/v1")
|
| 64 |
-
app.include_router(timeline.router, prefix="/api/v1")
|
| 65 |
-
app.include_router(session.router, prefix="/api/v1")
|
| 66 |
-
app.include_router(aethermap.router, prefix="/api/v1/aethermap", tags=["aethermap"])
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
@app.get("/")
|
| 70 |
-
def root():
|
| 71 |
-
"""Root endpoint - API info"""
|
| 72 |
-
return {
|
| 73 |
-
"name": "Numidium",
|
| 74 |
-
"version": settings.app_version,
|
| 75 |
-
"status": "online",
|
| 76 |
-
"docs": "/docs",
|
| 77 |
-
"description": "Sistema de Inteligência e Análise de Dados"
|
| 78 |
-
}
|
| 79 |
-
|
| 80 |
-
|
| 81 |
-
@app.get("/health")
|
| 82 |
-
def health_check():
|
| 83 |
-
"""Health check endpoint for HF Spaces"""
|
| 84 |
-
return {"status": "healthy"}
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
@app.get("/api/v1")
|
| 88 |
-
def api_info():
|
| 89 |
-
"""API v1 info"""
|
| 90 |
-
return {
|
| 91 |
-
"version": "1.0.0",
|
| 92 |
-
"endpoints": {
|
| 93 |
-
"entities": "/api/v1/entities",
|
| 94 |
-
"relationships": "/api/v1/relationships",
|
| 95 |
-
"events": "/api/v1/events",
|
| 96 |
-
"search": "/api/v1/search",
|
| 97 |
-
"ingest": "/api/v1/ingest"
|
| 98 |
-
}
|
| 99 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/models/__init__.py
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
# Models module
|
| 2 |
-
from app.models.entity import Entity, Relationship, Event, Document
|
| 3 |
-
from app.models.project import Project
|
|
|
|
|
|
|
|
|
|
|
|
app/models/__pycache__/__init__.cpython-311.pyc
DELETED
|
Binary file (367 Bytes)
|
|
|
app/models/__pycache__/entity.cpython-311.pyc
DELETED
|
Binary file (6.45 kB)
|
|
|
app/models/__pycache__/project.cpython-311.pyc
DELETED
|
Binary file (1.76 kB)
|
|
|
app/models/entity.py
DELETED
|
@@ -1,143 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
SQLAlchemy Models for Numidium
|
| 3 |
-
"""
|
| 4 |
-
from sqlalchemy import Column, String, Text, DateTime, Float, JSON, ForeignKey, Table
|
| 5 |
-
from sqlalchemy.orm import relationship
|
| 6 |
-
from datetime import datetime
|
| 7 |
-
import uuid
|
| 8 |
-
|
| 9 |
-
from app.core.database import Base
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
def generate_uuid():
|
| 13 |
-
return str(uuid.uuid4())
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
class Entity(Base):
|
| 17 |
-
"""
|
| 18 |
-
Entidade - qualquer coisa rastreável no sistema
|
| 19 |
-
Pode ser: pessoa, organização, local, veículo, evento, documento, etc.
|
| 20 |
-
"""
|
| 21 |
-
__tablename__ = "entities"
|
| 22 |
-
|
| 23 |
-
id = Column(String(36), primary_key=True, default=generate_uuid)
|
| 24 |
-
project_id = Column(String(36), ForeignKey("projects.id"), nullable=True, index=True)
|
| 25 |
-
type = Column(String(50), nullable=False, index=True) # person, organization, location, etc
|
| 26 |
-
name = Column(String(255), nullable=False, index=True)
|
| 27 |
-
description = Column(Text, nullable=True)
|
| 28 |
-
properties = Column(JSON, default=dict) # Dados flexíveis
|
| 29 |
-
|
| 30 |
-
# Geolocalização (opcional)
|
| 31 |
-
latitude = Column(Float, nullable=True)
|
| 32 |
-
longitude = Column(Float, nullable=True)
|
| 33 |
-
|
| 34 |
-
# Data histórica do evento/entidade (quando aconteceu, não quando foi adicionado)
|
| 35 |
-
event_date = Column(DateTime, nullable=True)
|
| 36 |
-
|
| 37 |
-
# Fonte do dado
|
| 38 |
-
source = Column(String(100), nullable=True) # wikipedia, newsapi, manual, etc
|
| 39 |
-
source_url = Column(Text, nullable=True)
|
| 40 |
-
|
| 41 |
-
# Timestamps
|
| 42 |
-
created_at = Column(DateTime, default=datetime.utcnow)
|
| 43 |
-
updated_at = Column(DateTime, default=datetime.utcnow, onupdate=datetime.utcnow)
|
| 44 |
-
|
| 45 |
-
# Relacionamentos
|
| 46 |
-
outgoing_relationships = relationship(
|
| 47 |
-
"Relationship",
|
| 48 |
-
foreign_keys="Relationship.source_id",
|
| 49 |
-
back_populates="source_entity"
|
| 50 |
-
)
|
| 51 |
-
incoming_relationships = relationship(
|
| 52 |
-
"Relationship",
|
| 53 |
-
foreign_keys="Relationship.target_id",
|
| 54 |
-
back_populates="target_entity"
|
| 55 |
-
)
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
class Relationship(Base):
|
| 59 |
-
"""
|
| 60 |
-
Relacionamento entre duas entidades
|
| 61 |
-
Exemplos: works_for, knows, owns, located_at, participated_in
|
| 62 |
-
"""
|
| 63 |
-
__tablename__ = "relationships"
|
| 64 |
-
|
| 65 |
-
id = Column(String(36), primary_key=True, default=generate_uuid)
|
| 66 |
-
project_id = Column(String(36), ForeignKey("projects.id"), nullable=True, index=True)
|
| 67 |
-
source_id = Column(String(36), ForeignKey("entities.id"), nullable=False)
|
| 68 |
-
target_id = Column(String(36), ForeignKey("entities.id"), nullable=False)
|
| 69 |
-
type = Column(String(50), nullable=False, index=True) # works_for, knows, owns, etc
|
| 70 |
-
properties = Column(JSON, default=dict)
|
| 71 |
-
confidence = Column(Float, default=1.0) # 0-1, quão certo estamos dessa conexão
|
| 72 |
-
|
| 73 |
-
# Data histórica do relacionamento (quando aconteceu)
|
| 74 |
-
event_date = Column(DateTime, nullable=True)
|
| 75 |
-
|
| 76 |
-
# Fonte
|
| 77 |
-
source = Column(String(100), nullable=True)
|
| 78 |
-
|
| 79 |
-
# Timestamps
|
| 80 |
-
created_at = Column(DateTime, default=datetime.utcnow)
|
| 81 |
-
|
| 82 |
-
# Relacionamentos
|
| 83 |
-
source_entity = relationship("Entity", foreign_keys=[source_id], back_populates="outgoing_relationships")
|
| 84 |
-
target_entity = relationship("Entity", foreign_keys=[target_id], back_populates="incoming_relationships")
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
class Event(Base):
|
| 88 |
-
"""
|
| 89 |
-
Evento - algo que aconteceu envolvendo entidades
|
| 90 |
-
"""
|
| 91 |
-
__tablename__ = "events"
|
| 92 |
-
|
| 93 |
-
id = Column(String(36), primary_key=True, default=generate_uuid)
|
| 94 |
-
type = Column(String(50), nullable=False, index=True)
|
| 95 |
-
title = Column(String(255), nullable=False)
|
| 96 |
-
description = Column(Text, nullable=True)
|
| 97 |
-
|
| 98 |
-
# Quando aconteceu
|
| 99 |
-
event_date = Column(DateTime, nullable=True)
|
| 100 |
-
|
| 101 |
-
# Onde aconteceu
|
| 102 |
-
location_name = Column(String(255), nullable=True)
|
| 103 |
-
latitude = Column(Float, nullable=True)
|
| 104 |
-
longitude = Column(Float, nullable=True)
|
| 105 |
-
|
| 106 |
-
# Entidades envolvidas (armazenado como JSON array de IDs)
|
| 107 |
-
entity_ids = Column(JSON, default=list)
|
| 108 |
-
|
| 109 |
-
# Fonte
|
| 110 |
-
source = Column(String(100), nullable=True)
|
| 111 |
-
source_url = Column(Text, nullable=True)
|
| 112 |
-
|
| 113 |
-
# Metadados
|
| 114 |
-
properties = Column(JSON, default=dict)
|
| 115 |
-
|
| 116 |
-
# Timestamps
|
| 117 |
-
created_at = Column(DateTime, default=datetime.utcnow)
|
| 118 |
-
|
| 119 |
-
|
| 120 |
-
class Document(Base):
|
| 121 |
-
"""
|
| 122 |
-
Documento - texto/arquivo para análise
|
| 123 |
-
"""
|
| 124 |
-
__tablename__ = "documents"
|
| 125 |
-
|
| 126 |
-
id = Column(String(36), primary_key=True, default=generate_uuid)
|
| 127 |
-
title = Column(String(255), nullable=False)
|
| 128 |
-
content = Column(Text, nullable=True)
|
| 129 |
-
summary = Column(Text, nullable=True) # Resumo gerado por IA
|
| 130 |
-
|
| 131 |
-
# Tipo de documento
|
| 132 |
-
doc_type = Column(String(50), default="text") # text, news, report, etc
|
| 133 |
-
|
| 134 |
-
# Entidades mencionadas (extraídas por NLP)
|
| 135 |
-
mentioned_entities = Column(JSON, default=list)
|
| 136 |
-
|
| 137 |
-
# Fonte
|
| 138 |
-
source = Column(String(100), nullable=True)
|
| 139 |
-
source_url = Column(Text, nullable=True)
|
| 140 |
-
|
| 141 |
-
# Timestamps
|
| 142 |
-
published_at = Column(DateTime, nullable=True)
|
| 143 |
-
created_at = Column(DateTime, default=datetime.utcnow)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/models/project.py
DELETED
|
@@ -1,29 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Project Model - Workspaces for organizing investigations
|
| 3 |
-
"""
|
| 4 |
-
from sqlalchemy import Column, String, Text, DateTime
|
| 5 |
-
from datetime import datetime
|
| 6 |
-
import uuid
|
| 7 |
-
|
| 8 |
-
from app.core.database import Base
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
def generate_uuid():
|
| 12 |
-
return str(uuid.uuid4())
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
class Project(Base):
|
| 16 |
-
"""
|
| 17 |
-
Projeto/Workspace - agrupa entidades e relacionamentos por investigação
|
| 18 |
-
"""
|
| 19 |
-
__tablename__ = "projects"
|
| 20 |
-
|
| 21 |
-
id = Column(String(36), primary_key=True, default=generate_uuid)
|
| 22 |
-
name = Column(String(255), nullable=False)
|
| 23 |
-
description = Column(Text, nullable=True)
|
| 24 |
-
color = Column(String(7), default="#00d4ff") # Hex color for UI
|
| 25 |
-
icon = Column(String(50), default="folder") # Icon name
|
| 26 |
-
|
| 27 |
-
# Timestamps
|
| 28 |
-
created_at = Column(DateTime, default=datetime.utcnow)
|
| 29 |
-
updated_at = Column(DateTime, default=datetime.utcnow, onupdate=datetime.utcnow)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/schemas/__init__.py
DELETED
|
@@ -1,10 +0,0 @@
|
|
| 1 |
-
# Schemas module
|
| 2 |
-
from app.schemas.schemas import (
|
| 3 |
-
EntityCreate, EntityUpdate, EntityResponse,
|
| 4 |
-
RelationshipCreate, RelationshipResponse,
|
| 5 |
-
EventCreate, EventResponse,
|
| 6 |
-
DocumentCreate, DocumentResponse,
|
| 7 |
-
GraphData, GraphNode, GraphEdge,
|
| 8 |
-
SearchQuery, SearchResult,
|
| 9 |
-
SystemStats
|
| 10 |
-
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/schemas/__pycache__/__init__.cpython-311.pyc
DELETED
|
Binary file (725 Bytes)
|
|
|
app/schemas/__pycache__/schemas.cpython-311.pyc
DELETED
|
Binary file (9.17 kB)
|
|
|
app/schemas/schemas.py
DELETED
|
@@ -1,163 +0,0 @@
|
|
| 1 |
-
"""
|
| 2 |
-
Pydantic Schemas for API validation
|
| 3 |
-
"""
|
| 4 |
-
from pydantic import BaseModel, Field
|
| 5 |
-
from typing import Optional, List, Any
|
| 6 |
-
from datetime import datetime
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
# ========== Entity Schemas ==========
|
| 10 |
-
|
| 11 |
-
class EntityBase(BaseModel):
|
| 12 |
-
type: str = Field(..., description="Tipo da entidade: person, organization, location, etc")
|
| 13 |
-
name: str = Field(..., description="Nome da entidade")
|
| 14 |
-
description: Optional[str] = None
|
| 15 |
-
properties: dict = Field(default_factory=dict)
|
| 16 |
-
latitude: Optional[float] = None
|
| 17 |
-
longitude: Optional[float] = None
|
| 18 |
-
source: Optional[str] = None
|
| 19 |
-
source_url: Optional[str] = None
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
class EntityCreate(EntityBase):
|
| 23 |
-
pass
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
class EntityUpdate(BaseModel):
|
| 27 |
-
type: Optional[str] = None
|
| 28 |
-
name: Optional[str] = None
|
| 29 |
-
description: Optional[str] = None
|
| 30 |
-
properties: Optional[dict] = None
|
| 31 |
-
latitude: Optional[float] = None
|
| 32 |
-
longitude: Optional[float] = None
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
class EntityResponse(EntityBase):
|
| 36 |
-
id: str
|
| 37 |
-
created_at: datetime
|
| 38 |
-
updated_at: datetime
|
| 39 |
-
|
| 40 |
-
class Config:
|
| 41 |
-
from_attributes = True
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
# ========== Relationship Schemas ==========
|
| 45 |
-
|
| 46 |
-
class RelationshipBase(BaseModel):
|
| 47 |
-
source_id: str
|
| 48 |
-
target_id: str
|
| 49 |
-
type: str = Field(..., description="Tipo: works_for, knows, owns, located_at, etc")
|
| 50 |
-
properties: dict = Field(default_factory=dict)
|
| 51 |
-
confidence: float = Field(default=1.0, ge=0, le=1)
|
| 52 |
-
source: Optional[str] = None
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
class RelationshipCreate(RelationshipBase):
|
| 56 |
-
pass
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
class RelationshipResponse(RelationshipBase):
|
| 60 |
-
id: str
|
| 61 |
-
created_at: datetime
|
| 62 |
-
|
| 63 |
-
class Config:
|
| 64 |
-
from_attributes = True
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
# ========== Event Schemas ==========
|
| 68 |
-
|
| 69 |
-
class EventBase(BaseModel):
|
| 70 |
-
type: str
|
| 71 |
-
title: str
|
| 72 |
-
description: Optional[str] = None
|
| 73 |
-
event_date: Optional[datetime] = None
|
| 74 |
-
location_name: Optional[str] = None
|
| 75 |
-
latitude: Optional[float] = None
|
| 76 |
-
longitude: Optional[float] = None
|
| 77 |
-
entity_ids: List[str] = Field(default_factory=list)
|
| 78 |
-
source: Optional[str] = None
|
| 79 |
-
source_url: Optional[str] = None
|
| 80 |
-
properties: dict = Field(default_factory=dict)
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
class EventCreate(EventBase):
|
| 84 |
-
pass
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
class EventResponse(EventBase):
|
| 88 |
-
id: str
|
| 89 |
-
created_at: datetime
|
| 90 |
-
|
| 91 |
-
class Config:
|
| 92 |
-
from_attributes = True
|
| 93 |
-
|
| 94 |
-
|
| 95 |
-
# ========== Document Schemas ==========
|
| 96 |
-
|
| 97 |
-
class DocumentBase(BaseModel):
|
| 98 |
-
title: str
|
| 99 |
-
content: Optional[str] = None
|
| 100 |
-
doc_type: str = "text"
|
| 101 |
-
source: Optional[str] = None
|
| 102 |
-
source_url: Optional[str] = None
|
| 103 |
-
published_at: Optional[datetime] = None
|
| 104 |
-
|
| 105 |
-
|
| 106 |
-
class DocumentCreate(DocumentBase):
|
| 107 |
-
pass
|
| 108 |
-
|
| 109 |
-
|
| 110 |
-
class DocumentResponse(DocumentBase):
|
| 111 |
-
id: str
|
| 112 |
-
summary: Optional[str] = None
|
| 113 |
-
mentioned_entities: List[str] = []
|
| 114 |
-
created_at: datetime
|
| 115 |
-
|
| 116 |
-
class Config:
|
| 117 |
-
from_attributes = True
|
| 118 |
-
|
| 119 |
-
|
| 120 |
-
# ========== Graph Schemas ==========
|
| 121 |
-
|
| 122 |
-
class GraphNode(BaseModel):
|
| 123 |
-
id: str
|
| 124 |
-
type: str
|
| 125 |
-
name: str
|
| 126 |
-
properties: dict = {}
|
| 127 |
-
|
| 128 |
-
|
| 129 |
-
class GraphEdge(BaseModel):
|
| 130 |
-
source: str
|
| 131 |
-
target: str
|
| 132 |
-
type: str
|
| 133 |
-
confidence: float = 1.0
|
| 134 |
-
|
| 135 |
-
|
| 136 |
-
class GraphData(BaseModel):
|
| 137 |
-
nodes: List[GraphNode]
|
| 138 |
-
edges: List[GraphEdge]
|
| 139 |
-
|
| 140 |
-
|
| 141 |
-
# ========== Search Schemas ==========
|
| 142 |
-
|
| 143 |
-
class SearchQuery(BaseModel):
|
| 144 |
-
query: str
|
| 145 |
-
entity_types: Optional[List[str]] = None
|
| 146 |
-
limit: int = Field(default=20, le=100)
|
| 147 |
-
|
| 148 |
-
|
| 149 |
-
class SearchResult(BaseModel):
|
| 150 |
-
entities: List[EntityResponse]
|
| 151 |
-
events: List[EventResponse]
|
| 152 |
-
documents: List[DocumentResponse]
|
| 153 |
-
|
| 154 |
-
|
| 155 |
-
# ========== Stats Schemas ==========
|
| 156 |
-
|
| 157 |
-
class SystemStats(BaseModel):
|
| 158 |
-
total_entities: int
|
| 159 |
-
total_relationships: int
|
| 160 |
-
total_events: int
|
| 161 |
-
total_documents: int
|
| 162 |
-
entities_by_type: dict
|
| 163 |
-
recent_activity: List[dict]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
app/services/__init__.py
DELETED
|
@@ -1 +0,0 @@
|
|
| 1 |
-
# Services module
|
|
|
|
|
|
app/services/__pycache__/__init__.cpython-311.pyc
DELETED
|
Binary file (167 Bytes)
|
|
|
app/services/__pycache__/brazil_apis.cpython-311.pyc
DELETED
|
Binary file (12.3 kB)
|
|
|
app/services/__pycache__/geocoding.cpython-311.pyc
DELETED
|
Binary file (3.14 kB)
|
|
|