Refa: Clean the folders. (#12890)

### Type of change

- [x] Refactoring

CLAUDE.md

@@ -27,7 +27,7 @@ RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine based on d
 - **Document Processing**: `deepdoc/` - PDF parsing, OCR, layout analysis
 - **LLM Integration**: `rag/llm/` - Model abstractions for chat, embedding, reranking
 - **RAG Pipeline**: `rag/flow/` - Chunking, parsing, tokenization
-- **Graph RAG**: `graphrag/` - Knowledge graph construction and querying
+- **Graph RAG**: `rag/graphrag/` - Knowledge graph construction and querying
 
 ### Agent System (`/agent/`)
 - **Components**: Modular workflow components (LLM, retrieval, categorize, etc.)
@@ -113,4 +113,4 @@ RAGFlow supports switching between Elasticsearch (default) and Infinity:
 - Node.js >=18.20.4
 - Docker & Docker Compose
 - uv package manager
-- 16GB+ RAM, 50GB+ disk space
+- 16GB+ RAM, 50GB+ disk space

Dockerfile

@@ -192,10 +192,8 @@ COPY conf conf
 COPY deepdoc deepdoc
 COPY rag rag
 COPY agent agent
-COPY graphrag graphrag
 COPY pyproject.toml uv.lock ./
 COPY mcp mcp
-COPY plugin plugin
 COPY common common
 COPY memory memory
 

agent/plugin/README.md

@@ -23,7 +23,7 @@ All the execution logic of this tool should go into this method.
 When you start RAGFlow, you can see your plugin was loaded in the log:
 
 ```
-2025-05-15 19:29:08,959 INFO     34670 Recursively importing plugins from path `/some-path/ragflow/plugin/embedded_plugins`
+2025-05-15 19:29:08,959 INFO     34670 Recursively importing plugins from path `/some-path/ragflow/agent/plugin/embedded_plugins`
 2025-05-15 19:29:08,960 INFO     34670 Loaded llm_tools plugin BadCalculatorPlugin version 1.0.0
 ```
 

agent/plugin/README_zh.md

@@ -23,7 +23,7 @@ RAGFlow将会从`embedded_plugins`子文件夹中递归加载所有的插件。
 当你启动RAGFlow时，你会在日志中看见你的插件被加载了：
 
 ```
-2025-05-15 19:29:08,959 INFO     34670 Recursively importing plugins from path `/some-path/ragflow/plugin/embedded_plugins`
+2025-05-15 19:29:08,959 INFO     34670 Recursively importing plugins from path `/some-path/ragflow/agent/plugin/embedded_plugins`
 2025-05-15 19:29:08,960 INFO     34670 Loaded llm_tools plugin BadCalculatorPlugin version 1.0.0
 ```
 

agent/plugin/embedded_plugins/llm_tools/bad_calculator.py

@@ -1,5 +1,5 @@
 import logging
-from plugin.llm_tool_plugin import LLMToolMetadata, LLMToolPlugin
+from agent.plugin.llm_tool_plugin import LLMToolMetadata, LLMToolPlugin
 
 
 class BadCalculatorPlugin(LLMToolPlugin):

api/apps/plugin_app.py

@@ -18,7 +18,7 @@
 from quart import Response
 from api.apps import login_required
 from api.utils.api_utils import get_json_result
-from plugin import GlobalPluginManager
+from agent.plugin import GlobalPluginManager
 
 
 @manager.route('/llm_tools', methods=['GET'])  # noqa: F821

api/db/services/dialog_service.py

@@ -35,7 +35,7 @@ from api.db.services.llm_service import LLMBundle
 from common.metadata_utils import apply_meta_data_filter
 from api.db.services.tenant_llm_service import TenantLLMService
 from common.time_utils import current_timestamp, datetime_format
-from graphrag.general.mind_map_extractor import MindMapExtractor
+from rag.graphrag.general.mind_map_extractor import MindMapExtractor
 from rag.advanced_rag import DeepResearcher
 from rag.app.tag import label_question
 from rag.nlp.search import index_name

api/db/services/document_service.py

@@ -1077,7 +1077,7 @@ def doc_upload_and_parse(conversation_id, file_objs, user_id):
         cks = [c for c in docs if c["doc_id"] == doc_id]
 
         if parser_ids[doc_id] != ParserType.PICTURE.value:
-            from graphrag.general.mind_map_extractor import MindMapExtractor
+            from rag.graphrag.general.mind_map_extractor import MindMapExtractor
             mindmap = MindMapExtractor(llm_bdl)
             try:
                 mind_map = asyncio.run(mindmap([c["content_with_weight"] for c in docs if c["doc_id"] == doc_id]))

api/ragflow_server.py

@@ -41,7 +41,7 @@ from common.versions import get_ragflow_version
 from common.config_utils import show_configs
 from common.mcp_tool_call_conn import shutdown_all_mcp_sessions
 from common.log_utils import init_root_logger
-from plugin import GlobalPluginManager
+from agent.plugin import GlobalPluginManager
 from rag.utils.redis_conn import RedisDistributedLock
 
 stop_event = threading.Event()

common/settings.py

@@ -317,7 +317,7 @@ def init_settings():
 
     global retriever, kg_retriever
     retriever = search.Dealer(docStoreConn)
-    from graphrag import search as kg_search
+    from rag.graphrag import search as kg_search
 
     kg_retriever = kg_search.KGSearch(docStoreConn)
 

rag/app/tag.py

@@ -124,7 +124,7 @@ def chunk(filename, binary=None, lang="Chinese", callback=None, **kwargs):
 
 def label_question(question, kbs):
     from api.db.services.knowledgebase_service import KnowledgebaseService
-    from graphrag.utils import get_tags_from_cache, set_tags_to_cache
+    from rag.graphrag.utils import get_tags_from_cache, set_tags_to_cache
     tags = None
     tag_kb_ids = []
     for kb in kbs:

rag/graphrag/entity_resolution.py

@@ -23,12 +23,12 @@ from typing import Any, Callable
 
 import networkx as nx
 
-from graphrag.general.extractor import Extractor
+from rag.graphrag.general.extractor import Extractor
 from rag.nlp import is_english
 import editdistance
-from graphrag.entity_resolution_prompt import ENTITY_RESOLUTION_PROMPT
+from rag.graphrag.entity_resolution_prompt import ENTITY_RESOLUTION_PROMPT
 from rag.llm.chat_model import Base as CompletionLLM
-from graphrag.utils import perform_variable_replacements, chat_limiter, GraphChange
+from rag.graphrag.utils import perform_variable_replacements, chat_limiter, GraphChange
 from api.db.services.task_service import has_canceled
 from common.exceptions import TaskCanceledException
 

rag/graphrag/general/community_reports_extractor.py

@@ -21,12 +21,12 @@ import pandas as pd
 from api.db.services.task_service import has_canceled
 from common.exceptions import TaskCanceledException
 from common.connection_utils import timeout
-from graphrag.general import leiden
+from rag.graphrag.general import leiden
-from graphrag.general.community_report_prompt import COMMUNITY_REPORT_PROMPT
+from rag.graphrag.general.community_report_prompt import COMMUNITY_REPORT_PROMPT
-from graphrag.general.extractor import Extractor
+from rag.graphrag.general.extractor import Extractor
-from graphrag.general.leiden import add_community_info2graph
+from rag.graphrag.general.leiden import add_community_info2graph
 from rag.llm.chat_model import Base as CompletionLLM
-from graphrag.utils import perform_variable_replacements, dict_has_keys_with_types, chat_limiter
+from rag.graphrag.utils import perform_variable_replacements, dict_has_keys_with_types, chat_limiter
 from common.token_utils import num_tokens_from_string
 
 @dataclass

rag/graphrag/general/entity_embedding.py

@@ -9,7 +9,7 @@ from typing import Any
 import numpy as np
 import networkx as nx
 from dataclasses import dataclass
-from graphrag.general.leiden import stable_largest_connected_component
+from rag.graphrag.general.leiden import stable_largest_connected_component
 import graspologic as gc
 
 
@@ -63,4 +63,4 @@ def run(graph: nx.Graph, args: dict[str, Any]) -> dict:
     pairs = zip(embeddings.nodes, embeddings.embeddings.tolist(), strict=True)
     sorted_pairs = sorted(pairs, key=lambda x: x[0])
 
-    return dict(sorted_pairs)
+    return dict(sorted_pairs)

rag/graphrag/general/extractor.py

@@ -26,8 +26,8 @@ import networkx as nx
 from api.db.services.task_service import has_canceled
 from common.connection_utils import timeout
 from common.token_utils import truncate
-from graphrag.general.graph_prompt import SUMMARIZE_DESCRIPTIONS_PROMPT
+from rag.graphrag.general.graph_prompt import SUMMARIZE_DESCRIPTIONS_PROMPT
-from graphrag.utils import (
+from rag.graphrag.utils import (
     GraphChange,
     chat_limiter,
     flat_uniq_list,

rag/graphrag/general/graph_extractor.py

@@ -13,9 +13,9 @@ from typing import Any
 from dataclasses import dataclass
 import tiktoken
 
-from graphrag.general.extractor import Extractor, ENTITY_EXTRACTION_MAX_GLEANINGS
+from rag.graphrag.general.extractor import Extractor, ENTITY_EXTRACTION_MAX_GLEANINGS
-from graphrag.general.graph_prompt import GRAPH_EXTRACTION_PROMPT, CONTINUE_PROMPT, LOOP_PROMPT
+from rag.graphrag.general.graph_prompt import GRAPH_EXTRACTION_PROMPT, CONTINUE_PROMPT, LOOP_PROMPT
-from graphrag.utils import ErrorHandlerFn, perform_variable_replacements, chat_limiter, split_string_by_multi_markers
+from rag.graphrag.utils import ErrorHandlerFn, perform_variable_replacements, chat_limiter, split_string_by_multi_markers
 from rag.llm.chat_model import Base as CompletionLLM
 import networkx as nx
 from common.token_utils import num_tokens_from_string

rag/graphrag/general/index.py

@@ -25,12 +25,12 @@ from api.db.services.task_service import has_canceled
 from common.exceptions import TaskCanceledException
 from common.misc_utils import get_uuid
 from common.connection_utils import timeout
-from graphrag.entity_resolution import EntityResolution
+from rag.graphrag.entity_resolution import EntityResolution
-from graphrag.general.community_reports_extractor import CommunityReportsExtractor
+from rag.graphrag.general.community_reports_extractor import CommunityReportsExtractor
-from graphrag.general.extractor import Extractor
+from rag.graphrag.general.extractor import Extractor
-from graphrag.general.graph_extractor import GraphExtractor as GeneralKGExt
+from rag.graphrag.general.graph_extractor import GraphExtractor as GeneralKGExt
-from graphrag.light.graph_extractor import GraphExtractor as LightKGExt
+from rag.graphrag.light.graph_extractor import GraphExtractor as LightKGExt
-from graphrag.utils import (
+from rag.graphrag.utils import (
     GraphChange,
     chunk_id,
     does_graph_contains,

rag/graphrag/general/mind_map_extractor.py

@@ -21,9 +21,9 @@ import re
 from typing import Any
 from dataclasses import dataclass
 
-from graphrag.general.extractor import Extractor
+from rag.graphrag.general.extractor import Extractor
-from graphrag.general.mind_map_prompt import MIND_MAP_EXTRACTION_PROMPT
+from rag.graphrag.general.mind_map_prompt import MIND_MAP_EXTRACTION_PROMPT
-from graphrag.utils import ErrorHandlerFn, perform_variable_replacements, chat_limiter
+from rag.graphrag.utils import ErrorHandlerFn, perform_variable_replacements, chat_limiter
 from rag.llm.chat_model import Base as CompletionLLM
 import markdown_to_json
 from functools import reduce

rag/graphrag/general/smoke.py

@@ -25,8 +25,8 @@ from api.db.services.document_service import DocumentService
 from api.db.services.knowledgebase_service import KnowledgebaseService
 from api.db.services.llm_service import LLMBundle
 from api.db.services.user_service import TenantService
-from graphrag.general.graph_extractor import GraphExtractor
+from rag.graphrag.general.graph_extractor import GraphExtractor
-from graphrag.general.index import update_graph, with_resolution, with_community
+from rag.graphrag.general.index import update_graph, with_resolution, with_community
 from common import settings
 
 settings.init_settings()

rag/graphrag/light/graph_extractor.py

@@ -15,9 +15,9 @@ from typing import Any
 
 import networkx as nx
 
-from graphrag.general.extractor import ENTITY_EXTRACTION_MAX_GLEANINGS, Extractor
+from rag.graphrag.general.extractor import ENTITY_EXTRACTION_MAX_GLEANINGS, Extractor
-from graphrag.light.graph_prompt import PROMPTS
+from rag.graphrag.light.graph_prompt import PROMPTS
-from graphrag.utils import chat_limiter, pack_user_ass_to_openai_messages, split_string_by_multi_markers
+from rag.graphrag.utils import chat_limiter, pack_user_ass_to_openai_messages, split_string_by_multi_markers
 from rag.llm.chat_model import Base as CompletionLLM
 from common.token_utils import num_tokens_from_string
 

rag/graphrag/light/smoke.py

@@ -25,8 +25,8 @@ from api.db.services.document_service import DocumentService
 from api.db.services.knowledgebase_service import KnowledgebaseService
 from api.db.services.llm_service import LLMBundle
 from api.db.services.user_service import TenantService
-from graphrag.general.index import update_graph
+from rag.graphrag.general.index import update_graph
-from graphrag.light.graph_extractor import GraphExtractor
+from rag.graphrag.light.graph_extractor import GraphExtractor
 from common import settings
 
 settings.init_settings()

rag/graphrag/search.py

@@ -22,8 +22,8 @@ import json_repair
 import pandas as pd
 
 from common.misc_utils import get_uuid
-from graphrag.query_analyze_prompt import PROMPTS
+from rag.graphrag.query_analyze_prompt import PROMPTS
-from graphrag.utils import get_entity_type2samples, get_llm_cache, set_llm_cache, get_relation
+from rag.graphrag.utils import get_entity_type2samples, get_llm_cache, set_llm_cache, get_relation
 from common.token_utils import num_tokens_from_string
 
 from rag.nlp.search import Dealer, index_name

rag/prompts/generator.py

@@ -492,7 +492,7 @@ async def gen_meta_filter(chat_mdl, meta_data: dict, query: str) -> dict:
 
 
 async def gen_json(system_prompt: str, user_prompt: str, chat_mdl, gen_conf={}, max_retry=2):
-    from graphrag.utils import get_llm_cache, set_llm_cache
+    from rag.graphrag.utils import get_llm_cache, set_llm_cache
     cached = get_llm_cache(chat_mdl.llm_name, system_prompt, user_prompt, gen_conf)
     if cached:
         return json_repair.loads(cached)
@@ -928,4 +928,4 @@ async def multi_queries_gen(chat_mdl, question: str, query:str, missing_infos:li
         )
     except Exception as e:
         logging.exception(e)
-    return {}
+    return {}

rag/raptor.py

@@ -25,7 +25,7 @@ from api.db.services.task_service import has_canceled
 from common.connection_utils import timeout
 from common.exceptions import TaskCanceledException
 from common.token_utils import truncate
-from graphrag.utils import (
+from rag.graphrag.utils import (
     chat_limiter,
     get_embed_cache,
     get_llm_cache,

rag/svr/task_executor.py

@@ -40,8 +40,8 @@ from rag.utils.base64_image import image2id
 from rag.utils.raptor_utils import should_skip_raptor, get_skip_reason
 from common.log_utils import init_root_logger
 from common.config_utils import show_configs
-from graphrag.general.index import run_graphrag_for_kb
+from rag.graphrag.general.index import run_graphrag_for_kb
-from graphrag.utils import get_llm_cache, set_llm_cache, get_tags_from_cache, set_tags_to_cache
+from rag.graphrag.utils import get_llm_cache, set_llm_cache, get_tags_from_cache, set_tags_to_cache
 from rag.prompts.generator import keyword_extraction, question_proposal, content_tagging, run_toc_from_text, \
     gen_metadata
 import logging
@@ -73,7 +73,7 @@ from rag.nlp import search, rag_tokenizer, add_positions
 from rag.raptor import RecursiveAbstractiveProcessing4TreeOrganizedRetrieval as Raptor
 from common.token_utils import num_tokens_from_string, truncate
 from rag.utils.redis_conn import REDIS_CONN, RedisDistributedLock
-from graphrag.utils import chat_limiter
+from rag.graphrag.utils import chat_limiter
 from common.signal_utils import start_tracemalloc_and_snapshot, stop_tracemalloc
 from common.exceptions import TaskCanceledException
 from common import settings