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

3
api/db/init_data.py
View File

@ -32,6 +32,7 @@ from api.db.services.user_service import TenantService, UserTenantService
from api import settings
from common.constants import LLMType
from common.file_utils import get_project_base_directory
from common import globals
from api.common.base64 import encode_to_base64
@ -49,7 +50,7 @@ def init_superuser():
"id": user_info["id"],
"name": user_info["nickname"] + "‘s Kingdom",
"llm_id": settings.CHAT_MDL,
"embd_id": settings.EMBEDDING_MDL,
"embd_id": globals.EMBEDDING_MDL,
"asr_id": settings.ASR_MDL,
"parser_ids": settings.PARSERS,
"img2txt_id": settings.IMAGE2TEXT_MDL

7
api/db/joint_services/user_account_service.py
View File

@ -38,6 +38,7 @@ from api.db.services.user_service import TenantService, UserService, UserTenantS
from rag.utils.storage_factory import STORAGE_IMPL
from rag.nlp import search
from common.constants import ActiveEnum
from common import globals
def create_new_user(user_info: dict) -> dict:
"""
@ -63,7 +64,7 @@ def create_new_user(user_info: dict) -> dict:
"id": user_id,
"name": user_info["nickname"] + "‘s Kingdom",
"llm_id": settings.CHAT_MDL,
"embd_id": settings.EMBEDDING_MDL,
"embd_id": globals.EMBEDDING_MDL,
"asr_id": settings.ASR_MDL,
"parser_ids": settings.PARSERS,
"img2txt_id": settings.IMAGE2TEXT_MDL,
@ -179,7 +180,7 @@ def delete_user_data(user_id: str) -> dict:
)
done_msg += f"- Deleted {file2doc_delete_res} document-file relation records.\n"
# step1.1.3 delete chunk in es
r = settings.docStoreConn.delete({"kb_id": kb_ids},
r = globals.docStoreConn.delete({"kb_id": kb_ids},
search.index_name(tenant_id), kb_ids)
done_msg += f"- Deleted {r} chunk records.\n"
kb_delete_res = KnowledgebaseService.delete_by_ids(kb_ids)
@ -237,7 +238,7 @@ def delete_user_data(user_id: str) -> dict:
kb_doc_info = {}
for _tenant_id, kb_doc in kb_grouped_doc.items():
for _kb_id, docs in kb_doc.items():
chunk_delete_res += settings.docStoreConn.delete(
chunk_delete_res += globals.docStoreConn.delete(
{"doc_id": [d["id"] for d in docs]},
search.index_name(_tenant_id), _kb_id
)

9
api/db/services/dialog_service.py
View File

@ -44,6 +44,7 @@ from rag.prompts.generator import chunks_format, citation_prompt, cross_language
from common.token_utils import num_tokens_from_string
from rag.utils.tavily_conn import Tavily
from common.string_utils import remove_redundant_spaces
from common import globals
class DialogService(CommonService):
@ -371,7 +372,7 @@ def chat(dialog, messages, stream=True, **kwargs):
chat_mdl.bind_tools(toolcall_session, tools)
bind_models_ts = timer()
retriever = settings.retriever
retriever = globals.retriever
questions = [m["content"] for m in messages if m["role"] == "user"][-3:]
attachments = kwargs["doc_ids"].split(",") if "doc_ids" in kwargs else []
if "doc_ids" in messages[-1]:
@ -663,7 +664,7 @@ Please write the SQL, only SQL, without any other explanations or text.
logging.debug(f"{question} get SQL(refined): {sql}")
tried_times += 1
return settings.retriever.sql_retrieval(sql, format="json"), sql
return globals.retriever.sql_retrieval(sql, format="json"), sql
tbl, sql = get_table()
if tbl is None:
@ -757,7 +758,7 @@ def ask(question, kb_ids, tenant_id, chat_llm_name=None, search_config={}):
embedding_list = list(set([kb.embd_id for kb in kbs]))
is_knowledge_graph = all([kb.parser_id == ParserType.KG for kb in kbs])
retriever = settings.retriever if not is_knowledge_graph else settings.kg_retriever
retriever = globals.retriever if not is_knowledge_graph else settings.kg_retriever
embd_mdl = LLMBundle(tenant_id, LLMType.EMBEDDING, embedding_list[0])
chat_mdl = LLMBundle(tenant_id, LLMType.CHAT, chat_llm_name)
@ -853,7 +854,7 @@ def gen_mindmap(question, kb_ids, tenant_id, search_config={}):
if not doc_ids:
doc_ids = None
ranks = settings.retriever.retrieval(
ranks = globals.retriever.retrieval(
question=question,
embd_mdl=embd_mdl,
tenant_ids=tenant_ids,

25
api/db/services/document_service.py
View File

@ -26,7 +26,6 @@ import trio
import xxhash
from peewee import fn, Case, JOIN
from api import settings
from api.constants import IMG_BASE64_PREFIX, FILE_NAME_LEN_LIMIT
from api.db import FileType, UserTenantRole, CanvasCategory
from api.db.db_models import DB, Document, Knowledgebase, Task, Tenant, UserTenant, File2Document, File, UserCanvas, \
@ -42,7 +41,7 @@ from rag.settings import get_svr_queue_name, SVR_CONSUMER_GROUP_NAME
from rag.utils.redis_conn import REDIS_CONN
from rag.utils.storage_factory import STORAGE_IMPL
from rag.utils.doc_store_conn import OrderByExpr
from common import globals
class DocumentService(CommonService):
model = Document
@ -309,10 +308,10 @@ class DocumentService(CommonService):
page_size = 1000
all_chunk_ids = []
while True:
chunks = settings.docStoreConn.search(["img_id"], [], {"doc_id": doc.id}, [], OrderByExpr(),
chunks = globals.docStoreConn.search(["img_id"], [], {"doc_id": doc.id}, [], OrderByExpr(),
page * page_size, page_size, search.index_name(tenant_id),
[doc.kb_id])
chunk_ids = settings.docStoreConn.getChunkIds(chunks)
chunk_ids = globals.docStoreConn.getChunkIds(chunks)
if not chunk_ids:
break
all_chunk_ids.extend(chunk_ids)
@ -323,19 +322,19 @@ class DocumentService(CommonService):
if doc.thumbnail and not doc.thumbnail.startswith(IMG_BASE64_PREFIX):
if STORAGE_IMPL.obj_exist(doc.kb_id, doc.thumbnail):
STORAGE_IMPL.rm(doc.kb_id, doc.thumbnail)
settings.docStoreConn.delete({"doc_id": doc.id}, search.index_name(tenant_id), doc.kb_id)
globals.docStoreConn.delete({"doc_id": doc.id}, search.index_name(tenant_id), doc.kb_id)
graph_source = settings.docStoreConn.getFields(
settings.docStoreConn.search(["source_id"], [], {"kb_id": doc.kb_id, "knowledge_graph_kwd": ["graph"]}, [], OrderByExpr(), 0, 1, search.index_name(tenant_id), [doc.kb_id]), ["source_id"]
graph_source = globals.docStoreConn.getFields(
globals.docStoreConn.search(["source_id"], [], {"kb_id": doc.kb_id, "knowledge_graph_kwd": ["graph"]}, [], OrderByExpr(), 0, 1, search.index_name(tenant_id), [doc.kb_id]), ["source_id"]
)
if len(graph_source) > 0 and doc.id in list(graph_source.values())[0]["source_id"]:
settings.docStoreConn.update({"kb_id": doc.kb_id, "knowledge_graph_kwd": ["entity", "relation", "graph", "subgraph", "community_report"], "source_id": doc.id},
globals.docStoreConn.update({"kb_id": doc.kb_id, "knowledge_graph_kwd": ["entity", "relation", "graph", "subgraph", "community_report"], "source_id": doc.id},
{"remove": {"source_id": doc.id}},
search.index_name(tenant_id), doc.kb_id)
settings.docStoreConn.update({"kb_id": doc.kb_id, "knowledge_graph_kwd": ["graph"]},
globals.docStoreConn.update({"kb_id": doc.kb_id, "knowledge_graph_kwd": ["graph"]},
{"removed_kwd": "Y"},
search.index_name(tenant_id), doc.kb_id)
settings.docStoreConn.delete({"kb_id": doc.kb_id, "knowledge_graph_kwd": ["entity", "relation", "graph", "subgraph", "community_report"], "must_not": {"exists": "source_id"}},
globals.docStoreConn.delete({"kb_id": doc.kb_id, "knowledge_graph_kwd": ["entity", "relation", "graph", "subgraph", "community_report"], "must_not": {"exists": "source_id"}},
search.index_name(tenant_id), doc.kb_id)
except Exception:
pass
@ -996,10 +995,10 @@ def doc_upload_and_parse(conversation_id, file_objs, user_id):
d["q_%d_vec" % len(v)] = v
for b in range(0, len(cks), es_bulk_size):
if try_create_idx:
if not settings.docStoreConn.indexExist(idxnm, kb_id):
settings.docStoreConn.createIdx(idxnm, kb_id, len(vects[0]))
if not globals.docStoreConn.indexExist(idxnm, kb_id):
globals.docStoreConn.createIdx(idxnm, kb_id, len(vects[0]))
try_create_idx = False
settings.docStoreConn.insert(cks[b:b + es_bulk_size], idxnm, kb_id)
globals.docStoreConn.insert(cks[b:b + es_bulk_size], idxnm, kb_id)
DocumentService.increment_chunk_num(
doc_id, kb.id, token_counts[doc_id], chunk_counts[doc_id], 0)

4
api/db/services/task_service.py
View File

@ -34,7 +34,7 @@ from deepdoc.parser.excel_parser import RAGFlowExcelParser
from rag.settings import get_svr_queue_name
from rag.utils.storage_factory import STORAGE_IMPL
from rag.utils.redis_conn import REDIS_CONN
from api import settings
from common import globals
from rag.nlp import search
CANVAS_DEBUG_DOC_ID = "dataflow_x"
@ -418,7 +418,7 @@ def queue_tasks(doc: dict, bucket: str, name: str, priority: int):
if pre_task["chunk_ids"]:
pre_chunk_ids.extend(pre_task["chunk_ids"].split())
if pre_chunk_ids:
settings.docStoreConn.delete({"id": pre_chunk_ids}, search.index_name(chunking_config["tenant_id"]),
globals.docStoreConn.delete({"id": pre_chunk_ids}, search.index_name(chunking_config["tenant_id"]),
chunking_config["kb_id"])
DocumentService.update_by_id(doc["id"], {"chunk_num": ck_num})