mirror of
https://github.com/tiennm99/DocsGPT.git
synced 2026-10-03 20:12:55 +00:00
Alex
574f96341e
The backend import package is now docsgpt, the name it will carry on PyPI; application was far too generic to install into anyone's site-packages. git mv plus a mechanical rewrite of every import, dotted string and path reference: 734 Python files, the compose files, Dockerfile, workflows, docs, setup scripts, devcontainer, k8s manifests, vscode config, pytest and coverage config, .gitignore. Behaviour is unchanged. Kept for one release: - A top-level application package whose meta-path finder resolves application.x.y to the already-imported docsgpt.x.y object, so old imports and entry points (celery -A application.app.celery, uvicorn application.asgi:asgi_app) keep working with a FutureWarning. - Celery registers every application.* task name as an alias of its docsgpt.* task on start-up, so messages queued by the previous release still run. The redbeat key prefix moves to redbeat:docsgpt:v2: so schedule entries the previous release wrote are left unread instead of firing twice. The backend image builds from the repository root (docker build -f docsgpt/Dockerfile .) so it can ship the alias package; a root .dockerignore allow-lists docsgpt/ and application/ and keeps caches, local data, .env files, the sample index files and the Dockerfile out. Compose and the image workflows point at the new context.
105 lines
3.7 KiB
Python
105 lines
3.7 KiB
Python
import pymongo
|
|
import os
|
|
import shutil
|
|
import logging
|
|
from tqdm import tqdm
|
|
|
|
# Configure logging
|
|
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
|
|
logger = logging.getLogger()
|
|
|
|
# Configuration
|
|
MONGO_URI = "mongodb://localhost:27017/"
|
|
MONGO_ATLAS_URI = "mongodb+srv://<username>:<password>@<cluster>/<dbname>?retryWrites=true&w=majority"
|
|
DB_NAME = "docsgpt"
|
|
|
|
def backup_collection(collection, backup_collection_name):
|
|
logger.info(f"Backing up collection {collection.name} to {backup_collection_name}")
|
|
collection.aggregate([{"$out": backup_collection_name}])
|
|
logger.info("Backup completed")
|
|
|
|
def migrate_to_v1_vectorstore_mongo():
|
|
client = pymongo.MongoClient(MONGO_URI)
|
|
db = client[DB_NAME]
|
|
vectors_collection = db["vectors"]
|
|
sources_collection = db["sources"]
|
|
|
|
# Backup collections before migration
|
|
backup_collection(vectors_collection, "vectors_backup")
|
|
backup_collection(sources_collection, "sources_backup")
|
|
|
|
vectors = list(vectors_collection.find())
|
|
for vector in tqdm(vectors, desc="Updating vectors"):
|
|
if "location" in vector:
|
|
del vector["location"]
|
|
if "retriever" not in vector:
|
|
vector["retriever"] = "classic"
|
|
vector["remote_data"] = None
|
|
vectors_collection.update_one({"_id": vector["_id"]}, {"$set": vector})
|
|
|
|
# Move data from vectors_collection to sources_collection
|
|
for vector in tqdm(vectors, desc="Moving to sources"):
|
|
sources_collection.insert_one(vector)
|
|
|
|
vectors_collection.drop()
|
|
client.close()
|
|
logger.info("Migration completed")
|
|
|
|
def migrate_faiss_to_v1_vectorstore():
|
|
client = pymongo.MongoClient(MONGO_URI)
|
|
db = client[DB_NAME]
|
|
vectors_collection = db["vectors"]
|
|
|
|
vectors = list(vectors_collection.find())
|
|
for vector in tqdm(vectors, desc="Migrating FAISS vectors"):
|
|
old_path = f"./docsgpt/indexes/{vector['user']}/{vector['name']}"
|
|
new_path = f"./docsgpt/indexes/{vector['_id']}"
|
|
try:
|
|
os.makedirs(os.path.dirname(new_path), exist_ok=True)
|
|
shutil.move(old_path, new_path)
|
|
except OSError as e:
|
|
logger.error(f"Error moving {old_path} to {new_path}: {e}")
|
|
|
|
client.close()
|
|
logger.info("FAISS migration completed")
|
|
|
|
def migrate_mongo_atlas_vector_to_v1_vectorstore():
|
|
client = pymongo.MongoClient(MONGO_ATLAS_URI)
|
|
db = client[DB_NAME]
|
|
vectors_collection = db["vectors"]
|
|
documents_collection = db["documents"]
|
|
|
|
# Backup collections before migration
|
|
backup_collection(vectors_collection, "vectors_backup")
|
|
backup_collection(documents_collection, "documents_backup")
|
|
|
|
vectors = list(vectors_collection.find())
|
|
for vector in tqdm(vectors, desc="Updating Mongo Atlas vectors"):
|
|
documents_collection.update_many(
|
|
{"store": vector["user"] + "/" + vector["name"]},
|
|
{"$set": {"source_id": str(vector["_id"])}}
|
|
)
|
|
|
|
client.close()
|
|
logger.info("Mongo Atlas migration completed")
|
|
|
|
if __name__ == "__main__":
|
|
try:
|
|
logger.info("Starting FAISS migration...")
|
|
migrate_faiss_to_v1_vectorstore()
|
|
logger.info("FAISS migration completed successfully ")
|
|
|
|
logger.info("Starting local Mongo migration...")
|
|
migrate_to_v1_vectorstore_mongo()
|
|
logger.info("Local Mongo migration completed successfully ")
|
|
|
|
logger.info("Starting Mongo Atlas migration...")
|
|
migrate_mongo_atlas_vector_to_v1_vectorstore()
|
|
logger.info("Mongo Atlas migration completed successfully ")
|
|
|
|
logger.info(" All migrations completed successfully!")
|
|
|
|
except Exception as e:
|
|
logger.error(f" Migration failed due to error: {e}")
|
|
logger.warning(" Please verify database state or restore from backups if necessary.")
|