idchlife/rag-solution
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Checking properly source of the file for metadata, with instanceof

Browse Source
This commit is contained in:
idchlife
2026-02-11 16:23:27 +03:00
parent f5659675ec
commit 93d538ecc6
1 changed files with 8 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
services/rag/langchain/enrichment.py
View File

@@ -118,7 +118,7 @@ def try_guess_file_type(extension: str) -> str:
def identify_adaptive_file_source(adaptive_file: _AdaptiveFile) -> str: def identify_adaptive_file_source(adaptive_file: _AdaptiveFile) -> str:
if adaptive_file is YandexDiskAdaptiveFile: if isinstance(adaptive_file, YandexDiskAdaptiveFile):
return "Яндекс Диск" return "Яндекс Диск"
else: else:
return "Локальный Файл" return "Локальный Файл"
@@ -279,9 +279,11 @@ class DocumentEnricher:
file_hash = self._get_file_hash(local_file_path) file_hash = self._get_file_hash(local_file_path)
if self._is_document_hash_processed(file_hash): if self._is_document_hash_processed(file_hash):
logger.info( logger.info(
f"Skipping already processed document hash for: {source_identifier}" f"SKIPPING already processed document hash for: {source_identifier}"
) )
return return
else:
logger.info("Document is not processed! Doing it")
loader = self._get_loader_for_extension(local_file_path) loader = self._get_loader_for_extension(local_file_path)
if loader is None: if loader is None:
@@ -326,6 +328,7 @@ class DocumentEnricher:
self.ADAPTIVE_FILES_QUEUE.put(adaptive_file) self.ADAPTIVE_FILES_QUEUE.put(adaptive_file)
logger.debug("ADAPTIVE COLLECTION DEPLETED!")
self.collection_finished.set() self.collection_finished.set()
# Phase 13 API: reads adaptive files and writes processed docs into PROCESSED_DOCUMENTS_QUEUE # Phase 13 API: reads adaptive files and writes processed docs into PROCESSED_DOCUMENTS_QUEUE
@@ -368,8 +371,9 @@ class DocumentEnricher:
processed_record[0], processed_record[1] processed_record[0], processed_record[1]
) )
except Exception as error: except Exception as error:
logger.error(f"Error uploading processed documents: {error}") logger.error(
raise f"Error uploading processed documents: {error}. But swallowing error. NOT raising."
)
finally: finally:
self.PROCESSED_DOCUMENTS_QUEUE.task_done() self.PROCESSED_DOCUMENTS_QUEUE.task_done()