pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2025-12-08 20:42:30 +08:00
Code Issues Packages Projects Releases Wiki Activity

Move some vars to globals (#11017)

Browse Source 
### What problem does this PR solve?

As title.

### Type of change

- [x] Refactoring

---------

Signed-off-by: Jin Hai <haijin.chn@gmail.com>
This commit is contained in:
Jin Hai
2025-11-05 14:14:38 +08:00
committed by GitHub
parent cf9611c96f
commit 1a9215bc6f
35 changed files with 185 additions and 164 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
api/apps/sdk/dataset.py
View File

@ -20,7 +20,6 @@ import os
import json
from flask import request
from peewee import OperationalError
from api import settings
from api.db.db_models import File
from api.db.services.document_service import DocumentService
from api.db.services.file2document_service import File2DocumentService
@ -49,6 +48,7 @@ from api.utils.validation_utils import (
)
from rag.nlp import search
from rag.settings import PAGERANK_FLD
from common import globals
@manager.route("/datasets", methods=["POST"]) # noqa: F821
@ -360,11 +360,11 @@ def update(tenant_id, dataset_id):
return get_error_argument_result(message="'pagerank' can only be set when doc_engine is elasticsearch")
if req["pagerank"] > 0:
settings.docStoreConn.update({"kb_id": kb.id}, {PAGERANK_FLD: req["pagerank"]},
globals.docStoreConn.update({"kb_id": kb.id}, {PAGERANK_FLD: req["pagerank"]},
search.index_name(kb.tenant_id), kb.id)
else:
# Elasticsearch requires PAGERANK_FLD be non-zero!
settings.docStoreConn.update({"exists": PAGERANK_FLD}, {"remove": PAGERANK_FLD},
globals.docStoreConn.update({"exists": PAGERANK_FLD}, {"remove": PAGERANK_FLD},
search.index_name(kb.tenant_id), kb.id)
if not KnowledgebaseService.update_by_id(kb.id, req):
@ -493,9 +493,9 @@ def knowledge_graph(tenant_id, dataset_id):
}
obj = {"graph": {}, "mind_map": {}}
if not settings.docStoreConn.indexExist(search.index_name(kb.tenant_id), dataset_id):
if not globals.docStoreConn.indexExist(search.index_name(kb.tenant_id), dataset_id):
return get_result(data=obj)
sres = settings.retriever.search(req, search.index_name(kb.tenant_id), [dataset_id])
sres = globals.retriever.search(req, search.index_name(kb.tenant_id), [dataset_id])
if not len(sres.ids):
return get_result(data=obj)
@ -528,7 +528,7 @@ def delete_knowledge_graph(tenant_id, dataset_id):
code=RetCode.AUTHENTICATION_ERROR
)
_, kb = KnowledgebaseService.get_by_id(dataset_id)
settings.docStoreConn.delete({"knowledge_graph_kwd": ["graph", "subgraph", "entity", "relation"]},
globals.docStoreConn.delete({"knowledge_graph_kwd": ["graph", "subgraph", "entity", "relation"]},
search.index_name(kb.tenant_id), dataset_id)
return get_result(data=True)

3
api/apps/sdk/dify_retrieval.py
View File

@ -25,6 +25,7 @@ from api.utils.api_utils import validate_request, build_error_result, apikey_req
from rag.app.tag import label_question
from api.db.services.dialog_service import meta_filter, convert_conditions
from common.constants import RetCode, LLMType
from common import globals
@manager.route('/dify/retrieval', methods=['POST']) # noqa: F821
@apikey_required
@ -137,7 +138,7 @@ def retrieval(tenant_id):
# print("doc_ids", doc_ids)
if not doc_ids and metadata_condition is not None:
doc_ids = ['-999']
ranks = settings.retriever.retrieval(
ranks = globals.retriever.retrieval(
question,
embd_mdl,
kb.tenant_id,

25
api/apps/sdk/doc.py
View File

@ -44,6 +44,7 @@ from rag.prompts.generator import cross_languages, keyword_extraction
from rag.utils.storage_factory import STORAGE_IMPL
from common.string_utils import remove_redundant_spaces
from common.constants import RetCode, LLMType, ParserType, TaskStatus, FileSource
from common import globals
MAXIMUM_OF_UPLOADING_FILES = 256
@ -307,7 +308,7 @@ def update_doc(tenant_id, dataset_id, document_id):
)
if not e:
return get_error_data_result(message="Document not found!")
settings.docStoreConn.delete({"doc_id": doc.id}, search.index_name(tenant_id), dataset_id)
globals.docStoreConn.delete({"doc_id": doc.id}, search.index_name(tenant_id), dataset_id)
if "enabled" in req:
status = int(req["enabled"])
@ -316,7 +317,7 @@ def update_doc(tenant_id, dataset_id, document_id):
if not DocumentService.update_by_id(doc.id, {"status": str(status)}):
return get_error_data_result(message="Database error (Document update)!")
settings.docStoreConn.update({"doc_id": doc.id}, {"available_int": status}, search.index_name(kb.tenant_id), doc.kb_id)
globals.docStoreConn.update({"doc_id": doc.id}, {"available_int": status}, search.index_name(kb.tenant_id), doc.kb_id)
return get_result(data=True)
except Exception as e:
return server_error_response(e)
@ -755,7 +756,7 @@ def parse(tenant_id, dataset_id):
return get_error_data_result("Can't parse document that is currently being processed")
info = {"run": "1", "progress": 0, "progress_msg": "", "chunk_num": 0, "token_num": 0}
DocumentService.update_by_id(id, info)
settings.docStoreConn.delete({"doc_id": id}, search.index_name(tenant_id), dataset_id)
globals.docStoreConn.delete({"doc_id": id}, search.index_name(tenant_id), dataset_id)
TaskService.filter_delete([Task.doc_id == id])
e, doc = DocumentService.get_by_id(id)
doc = doc.to_dict()
@ -835,7 +836,7 @@ def stop_parsing(tenant_id, dataset_id):
return get_error_data_result("Can't stop parsing document with progress at 0 or 1")
info = {"run": "2", "progress": 0, "chunk_num": 0}
DocumentService.update_by_id(id, info)
settings.docStoreConn.delete({"doc_id": doc[0].id}, search.index_name(tenant_id), dataset_id)
globals.docStoreConn.delete({"doc_id": doc[0].id}, search.index_name(tenant_id), dataset_id)
success_count += 1
if duplicate_messages:
if success_count > 0:
@ -968,7 +969,7 @@ def list_chunks(tenant_id, dataset_id, document_id):
res = {"total": 0, "chunks": [], "doc": renamed_doc}
if req.get("id"):
chunk = settings.docStoreConn.get(req.get("id"), search.index_name(tenant_id), [dataset_id])
chunk = globals.docStoreConn.get(req.get("id"), search.index_name(tenant_id), [dataset_id])
if not chunk:
return get_result(message=f"Chunk not found: {dataset_id}/{req.get('id')}", code=RetCode.NOT_FOUND)
k = []
@ -995,8 +996,8 @@ def list_chunks(tenant_id, dataset_id, document_id):
res["chunks"].append(final_chunk)
_ = Chunk(**final_chunk)
elif settings.docStoreConn.indexExist(search.index_name(tenant_id), dataset_id):
sres = settings.retriever.search(query, search.index_name(tenant_id), [dataset_id], emb_mdl=None, highlight=True)
elif globals.docStoreConn.indexExist(search.index_name(tenant_id), dataset_id):
sres = globals.retriever.search(query, search.index_name(tenant_id), [dataset_id], emb_mdl=None, highlight=True)
res["total"] = sres.total
for id in sres.ids:
d = {
@ -1120,7 +1121,7 @@ def add_chunk(tenant_id, dataset_id, document_id):
v, c = embd_mdl.encode([doc.name, req["content"] if not d["question_kwd"] else "\n".join(d["question_kwd"])])
v = 0.1 * v[0] + 0.9 * v[1]
d["q_%d_vec" % len(v)] = v.tolist()
settings.docStoreConn.insert([d], search.index_name(tenant_id), dataset_id)
globals.docStoreConn.insert([d], search.index_name(tenant_id), dataset_id)
DocumentService.increment_chunk_num(doc.id, doc.kb_id, c, 1, 0)
# rename keys
@ -1201,7 +1202,7 @@ def rm_chunk(tenant_id, dataset_id, document_id):
if "chunk_ids" in req:
unique_chunk_ids, duplicate_messages = check_duplicate_ids(req["chunk_ids"], "chunk")
condition["id"] = unique_chunk_ids
chunk_number = settings.docStoreConn.delete(condition, search.index_name(tenant_id), dataset_id)
chunk_number = globals.docStoreConn.delete(condition, search.index_name(tenant_id), dataset_id)
if chunk_number != 0:
DocumentService.decrement_chunk_num(document_id, dataset_id, 1, chunk_number, 0)
if "chunk_ids" in req and chunk_number != len(unique_chunk_ids):
@ -1273,7 +1274,7 @@ def update_chunk(tenant_id, dataset_id, document_id, chunk_id):
schema:
type: object
"""
chunk = settings.docStoreConn.get(chunk_id, search.index_name(tenant_id), [dataset_id])
chunk = globals.docStoreConn.get(chunk_id, search.index_name(tenant_id), [dataset_id])
if chunk is None:
return get_error_data_result(f"Can't find this chunk {chunk_id}")
if not KnowledgebaseService.accessible(kb_id=dataset_id, user_id=tenant_id):
@ -1318,7 +1319,7 @@ def update_chunk(tenant_id, dataset_id, document_id, chunk_id):
v, c = embd_mdl.encode([doc.name, d["content_with_weight"] if not d.get("question_kwd") else "\n".join(d["question_kwd"])])
v = 0.1 * v[0] + 0.9 * v[1] if doc.parser_id != ParserType.QA else v[1]
d["q_%d_vec" % len(v)] = v.tolist()
settings.docStoreConn.update({"id": chunk_id}, d, search.index_name(tenant_id), dataset_id)
globals.docStoreConn.update({"id": chunk_id}, d, search.index_name(tenant_id), dataset_id)
return get_result()
@ -1464,7 +1465,7 @@ def retrieval_test(tenant_id):
chat_mdl = LLMBundle(kb.tenant_id, LLMType.CHAT)
question += keyword_extraction(chat_mdl, question)
ranks = settings.retriever.retrieval(
ranks = globals.retriever.retrieval(
question,
embd_mdl,
tenant_ids,

3
api/apps/sdk/session.py
View File

@ -41,6 +41,7 @@ from rag.app.tag import label_question
from rag.prompts.template import load_prompt
from rag.prompts.generator import cross_languages, gen_meta_filter, keyword_extraction, chunks_format
from common.constants import RetCode, LLMType, StatusEnum
from common import globals
@manager.route("/chats/<chat_id>/sessions", methods=["POST"]) # noqa: F821
@token_required
@ -1015,7 +1016,7 @@ def retrieval_test_embedded():
question += keyword_extraction(chat_mdl, question)
labels = label_question(question, [kb])
ranks = settings.retriever.retrieval(
ranks = globals.retriever.retrieval(
question, embd_mdl, tenant_ids, kb_ids, page, size, similarity_threshold, vector_similarity_weight, top,
doc_ids, rerank_mdl=rerank_mdl, highlight=req.get("highlight"), rank_feature=labels
)