Spaces:
Sleeping
Sleeping
File size: 19,210 Bytes
7dfe46c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 |
import logging
import time
from pathlib import Path
from typing import Dict, List, Any, Optional, Tuple
from dataclasses import dataclass
from concurrent.futures import ThreadPoolExecutor, as_completed
import hashlib
import os
import sys
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
from src.document_processor import (
DocumentProcessor, DocumentProcessorFactory, ProcessedDocument,
DocumentChunk, ProcessingStatus, DocumentType
)
from src.embedding_system import EmbeddingSystem
from src.vector_store import QdrantVectorStore
from src.metadata_manager import MetadataManager, DocumentMetadata
from src.image_processor import ImageProcessor
try:
from logger.custom_logger import CustomLoggerTracker
custom_log = CustomLoggerTracker()
logger = custom_log.get_logger("ingestion_pipeline")
except ImportError:
# Fallback to standard logging if custom logger not available
logger = logging.getLogger("ingestion_pipeline")
@dataclass
class IngestionResult:
"""Result of document ingestion."""
document_id: str
filename: str
success: bool
processing_time: float
chunks_created: int
chunks_indexed: int
error_message: Optional[str] = None
warnings: List[str] = None
def __post_init__(self):
if self.warnings is None:
self.warnings = []
@dataclass
class IngestionStats:
"""Statistics for batch ingestion."""
total_documents: int
successful_documents: int
failed_documents: int
total_chunks: int
total_processing_time: float
average_processing_time: float
documents_by_type: Dict[str, int]
errors: List[str]
def jina_embeddings(text: str) -> List[float]:
JINA_API_KEY= "jina_a75b55a8a9524bb697ea016b164211ebF5IduSgA0Ku8lmI0pS9fnXoZ83Su"
import requests
headers = {
'Content-Type': 'application/json',
'Authorization': 'Bearer jina_a75b55a8a9524bb697ea016b164211ebF5IduSgA0Ku8lmI0pS9fnXoZ83Su'}
data = {
"model": "jina-embeddings-v3",
"task": "retrieval.passage",
"input": text}
response = requests.post('https://api.jina.ai/v1/embeddings', headers=headers, json=data)
return response.json()['data'][0]['embedding']
class DocumentIngestionPipeline:
def __init__(self, config: Dict[str, Any]):
self.config = config
# Initialize components
self.embedding_system = EmbeddingSystem(config)
self.vector_store = QdrantVectorStore(config)
self.metadata_manager = MetadataManager(config)
# Initialize components with correct vector dimensions
self.vector_size = config.get('vector_size', 1024) # Match Jina's dimension
self.config['vector_size'] = self.vector_size # Update config for other components
# Initialize image processor for OCR
self.image_processor = ImageProcessor(config)
# Pipeline settings
self.chunk_size = config.get('chunk_size', 512)
self.chunk_overlap = config.get('chunk_overlap', 50)
self.batch_size = config.get('embedding_batch_size', 32)
self.max_workers = config.get('max_workers', 4)
self.enable_ocr = config.get('image_processing', True)
logger.info(f"Document ingestion pipeline initialized")
def ingest_document(self, file_path: str, document_id: Optional[str] = None) -> IngestionResult:
"""
Ingest a single document through the complete pipeline.
Args:
file_path: Path to the document file
document_id: Optional custom document ID
Returns:
IngestionResult with processing details
"""
start_time = time.time()
file_path_obj = Path(file_path)
filename = file_path_obj.name
try:
logger.info(f"Starting ingestion of document: {filename}")
# Generate document ID if not provided
if not document_id:
document_id = self._generate_document_id(file_path)
# Check if document already exists
existing_metadata = self.metadata_manager.get_document_metadata(document_id)
if existing_metadata and existing_metadata.processing_status == ProcessingStatus.COMPLETED:
logger.info(f"Document {filename} already processed, skipping")
return IngestionResult(
document_id=document_id,
filename=filename,
success=True,
processing_time=0.0,
chunks_created=existing_metadata.total_chunks,
chunks_indexed=existing_metadata.total_chunks,
warnings=["Document already processed"]
)
# Step 1: Process document
processed_doc = self._process_document(file_path)
if processed_doc.processing_status == ProcessingStatus.FAILED:
return IngestionResult(
document_id=document_id,
filename=filename,
success=False,
processing_time=time.time() - start_time,
chunks_created=0,
chunks_indexed=0,
error_message=processed_doc.error_message
)
# Step 2: Process images with OCR if enabled
if self.enable_ocr and processed_doc.images:
processed_doc.images = self.image_processor.batch_process_images(processed_doc.images)
# Step 3: Create document chunks
processor = DocumentProcessorFactory.create_processor(file_path, self.config)
chunks = processor.extract_chunks(processed_doc, self.chunk_size, self.chunk_overlap)
if not chunks:
logger.warning(f"No chunks created for document: {filename}")
return IngestionResult(
document_id=document_id,
filename=filename,
success=False,
processing_time=time.time() - start_time,
chunks_created=0,
chunks_indexed=0,
error_message="No content chunks could be created"
)
# Step 4: Generate embeddings
chunk_texts = [chunk.content for chunk in chunks]
logger.info(chunk_texts[:2])
# embeddings = self.embedding_system.generate_embeddings(chunk_texts)
embeddings = [jina_embeddings(text) for text in chunk_texts]
if not embeddings or len(embeddings) != len(chunks):
logger.error(f"Embedding generation failed for document: {filename}")
return IngestionResult(
document_id=document_id,
filename=filename,
success=False,
processing_time=time.time() - start_time,
chunks_created=len(chunks),
chunks_indexed=0,
error_message="Failed to generate embeddings"
)
# Attach embeddings to chunks
for chunk, embedding in zip(chunks, embeddings):
chunk.embedding = embedding
# Step 5: Store in vector database
vector_success = self.vector_store.add_documents(chunks)
if not vector_success:
logger.error(f"Failed to store vectors for document: {filename}")
return IngestionResult(
document_id=document_id,
filename=filename,
success=False,
processing_time=time.time() - start_time,
chunks_created=len(chunks),
chunks_indexed=0,
error_message="Failed to store document vectors"
)
# Step 6: Store metadata
processing_time = time.time() - start_time
metadata = DocumentMetadata(
document_id=document_id,
filename=filename,
file_path=file_path,
file_type=processed_doc.document_type.value,
upload_timestamp=processed_doc.processing_timestamp,
processing_status=ProcessingStatus.COMPLETED,
total_chunks=len(chunks),
file_size=processed_doc.file_size,
checksum=processed_doc.checksum,
processing_time=processing_time,
metadata_json=self._serialize_metadata(processed_doc.metadata)
)
metadata_success = self.metadata_manager.store_document_metadata(document_id, metadata)
if not metadata_success:
logger.warning(f"Failed to store metadata for document: {filename}")
logger.info(f"Successfully ingested document {filename}: {len(chunks)} chunks in {processing_time:.2f}s")
return IngestionResult(
document_id=document_id,
filename=filename,
success=True,
processing_time=processing_time,
chunks_created=len(chunks),
chunks_indexed=len(chunks)
)
except Exception as e:
error_msg = f"Ingestion failed for {filename}: {str(e)}"
logger.error(error_msg)
# Update metadata with error status
if document_id:
self.metadata_manager.update_document_status(
document_id,
ProcessingStatus.FAILED,
error_msg,
time.time() - start_time
)
return IngestionResult(
document_id=document_id or "unknown",
filename=filename,
success=False,
processing_time=time.time() - start_time,
chunks_created=0,
chunks_indexed=0,
error_message=error_msg
)
def ingest_batch(self, file_paths: List[str], max_workers: Optional[int] = None) -> IngestionStats:
"""
Ingest multiple documents in parallel.
Args:
file_paths: List of file paths to process
max_workers: Maximum number of worker threads
Returns:
IngestionStats with batch processing results
"""
start_time = time.time()
max_workers = max_workers or self.max_workers
logger.info(f"Starting batch ingestion of {len(file_paths)} documents with {max_workers} workers")
results = []
errors = []
documents_by_type = {}
# Process documents in parallel
with ThreadPoolExecutor(max_workers=max_workers) as executor:
# Submit all tasks
future_to_path = {
executor.submit(self.ingest_document, file_path): file_path
for file_path in file_paths
}
# Collect results
for future in as_completed(future_to_path):
file_path = future_to_path[future]
try:
result = future.result()
results.append(result)
# Track document types
file_ext = Path(file_path).suffix.lower()
documents_by_type[file_ext] = documents_by_type.get(file_ext, 0) + 1
if not result.success:
errors.append(f"{result.filename}: {result.error_message}")
except Exception as e:
error_msg = f"Failed to process {file_path}: {str(e)}"
errors.append(error_msg)
logger.error(error_msg)
# Calculate statistics
successful_results = [r for r in results if r.success]
failed_results = [r for r in results if not r.success]
total_processing_time = time.time() - start_time
total_chunks = sum(r.chunks_indexed for r in successful_results)
avg_processing_time = (
sum(r.processing_time for r in results) / len(results)
if results else 0.0
)
stats = IngestionStats(
total_documents=len(file_paths),
successful_documents=len(successful_results),
failed_documents=len(failed_results),
total_chunks=total_chunks,
total_processing_time=total_processing_time,
average_processing_time=avg_processing_time,
documents_by_type=documents_by_type,
errors=errors
)
logger.info(f"Batch ingestion completed: {stats.successful_documents}/{stats.total_documents} "
f"documents processed successfully in {total_processing_time:.2f}s")
return stats
def reprocess_document(self, document_id: str) -> IngestionResult:
"""
Reprocess an existing document.
Args:
document_id: ID of the document to reprocess
Returns:
IngestionResult with reprocessing details
"""
# Get existing metadata
metadata = self.metadata_manager.get_document_metadata(document_id)
if not metadata:
return IngestionResult(
document_id=document_id,
filename="unknown",
success=False,
processing_time=0.0,
chunks_created=0,
chunks_indexed=0,
error_message="Document not found in metadata"
)
# Delete existing vectors
self.vector_store.delete_document(document_id)
# Reprocess the document
return self.ingest_document(metadata.file_path, document_id)
def delete_document(self, document_id: str) -> bool:
"""
Delete a document and all associated data.
Args:
document_id: ID of the document to delete
Returns:
True if successful, False otherwise
"""
try:
# Delete from vector store
vector_success = self.vector_store.delete_document(document_id)
# Delete from metadata
metadata_success = self.metadata_manager.delete_document(document_id)
success = vector_success and metadata_success
if success:
logger.info(f"Successfully deleted document: {document_id}")
else:
logger.warning(f"Partial deletion of document: {document_id}")
return success
except Exception as e:
logger.error(f"Failed to delete document {document_id}: {e}")
return False
def _process_document(self, file_path: str) -> ProcessedDocument:
try:
processor = DocumentProcessorFactory.create_processor(file_path, self.config)
return processor.process_document(file_path)
except Exception as e:
logger.error(f"Document processing failed for {file_path}: {e}")
# Return failed document
document_id = self._generate_document_id(file_path)
return ProcessedDocument(
document_id=document_id,
filename=Path(file_path).name,
file_path=file_path,
document_type=DocumentType.UNKNOWN,
content="",
metadata={},
processing_status=ProcessingStatus.FAILED,
error_message=str(e)
)
def _generate_document_id(self, file_path: str) -> str:
# Use file path and modification time for uniqueness
file_path_obj = Path(file_path)
if file_path_obj.exists():
mtime = file_path_obj.stat().st_mtime
content = f"{file_path}_{mtime}"
else:
content = f"{file_path}_{time.time()}"
return hashlib.md5(content.encode()).hexdigest()
def _serialize_metadata(self, metadata: Dict[str, Any]) -> str:
try:
import json
return json.dumps(metadata, default=str, ensure_ascii=False)
except Exception as e:
logger.warning(f"Failed to serialize metadata: {e}")
return "{}"
def get_pipeline_stats(self) -> Dict[str, Any]:
"""
Get statistics about the ingestion pipeline.
Returns:
Dictionary with pipeline statistics
"""
try:
# Get component statistics
vector_stats = self.vector_store.get_collection_info()
metadata_stats = self.metadata_manager.get_statistics()
embedding_stats = self.embedding_system.get_cache_stats()
return {
"vector_store": vector_stats.__dict__ if vector_stats else {},
"metadata_manager": metadata_stats,
"embedding_system": embedding_stats,
"pipeline_config": {
"chunk_size": self.chunk_size,
"chunk_overlap": self.chunk_overlap,
"batch_size": self.batch_size,
"max_workers": self.max_workers,
"enable_ocr": self.enable_ocr
}
}
except Exception as e:
logger.error(f"Failed to get pipeline stats: {e}")
return {"error": str(e)}
def health_check(self) -> Dict[str, bool]:
"""
Check health of all pipeline components.
Returns:
Dictionary with health status of each component
"""
return {
"vector_store": self.vector_store.health_check(),
"metadata_manager": True, # SQLite is always available if file system works
"embedding_system": True # Will be checked during actual usage
}
if __name__=="__main__":
logger.info(f"Ingestion Pipe init ..")
## Example usage
import yaml
with open("src/config.yaml", 'r') as f:
config = yaml.safe_load(f)
pipeline = DocumentIngestionPipeline(config)
stats = pipeline.get_pipeline_stats()
logger.info(f"Pipeline stats: {stats}")
# Example single document ingestion
result = pipeline.ingest_document("data/documents/3.수불확인등록.xlsx")
logger.info(f"Ingestion result: {result}")
# Example batch ingestion
# batch_result = pipeline.ingest_batch(["sample_data/sample.pdf", "sample_data/sample.docx"])
# logger.info(f"Batch ingestion stats: {batch_result}")
|