Skip to content

RAGFlow is an agent-native database: a Retrieval-Augmented Generation (RAG) engine that grew into context infrastructure.

Fix: get rid of graphrag raptor in parser_config in REST API (#19161)

Wang Qi6ec9c6a7376d
diff --git a/api/apps/restful_apis/document_api.py b/api/apps/restful_apis/document_api.py
index 78cb021670..4f5b1b2c5b 100644
--- a/api/apps/restful_apis/document_api.py
+++ b/api/apps/restful_apis/document_api.py
@@ -60,6 +60,7 @@ from api.utils.api_utils import (
get_request_json,
get_error_argument_result,
check_duplicate_ids,
+ strip_graphrag_raptor_config,
)
from api.utils.pagination_utils import DEFAULT_PAGE, DEFAULT_PAGE_SIZE, validate_rest_api_ids, validate_rest_api_page, validate_rest_api_page_size
from api.utils.validation_utils import (
@@ -713,7 +714,7 @@ async def _upload_local_documents(kb, tenant_id):
return_raw_files = request.args.get("return_raw_files", "false").lower() == "true"
if return_raw_files:
- doc_data = files
+ doc_data = [strip_graphrag_raptor_config(doc) for doc in files]
else:
doc_data = [map_doc_keys_with_run_status(doc, run_status="0") for doc in files]
@@ -1296,7 +1297,7 @@ async def update_metadata_config(tenant_id, dataset_id, document_id):
except Exception as e:
return get_json_result(code=RetCode.EXCEPTION_ERROR, message=repr(e))
- return get_result(data=doc.to_dict())
+ return get_result(data=strip_graphrag_raptor_config(doc.to_dict()))
@manager.route("/thumbnails", methods=["GET"]) # noqa: F821
diff --git a/api/apps/services/document_api_service.py b/api/apps/services/document_api_service.py
index aa4663661a..707894abad 100644
--- a/api/apps/services/document_api_service.py
+++ b/api/apps/services/document_api_service.py
@@ -22,7 +22,7 @@ from api.db.services.file_service import FileService
from api.utils import validation_utils
from common import settings
from common.constants import TaskStatus
-from api.utils.api_utils import get_error_data_result, server_error_response, get_parser_config
+from api.utils.api_utils import get_error_data_result, server_error_response, get_parser_config, strip_graphrag_raptor_config
from api.utils.validation_utils import UpdateDocumentReq
from rag.nlp import rag_tokenizer, search
@@ -282,7 +282,7 @@ def _process_key_mappings(doc):
for key, value in items:
new_key = key_mapping.get(key, key)
renamed_doc[new_key] = value
- return renamed_doc
+ return strip_graphrag_raptor_config(renamed_doc)
def _process_run_mapping(doc, run_status):
diff --git a/api/utils/api_utils.py b/api/utils/api_utils.py
index d2a2ef4ea1..a112f09e61 100644
--- a/api/utils/api_utils.py
+++ b/api/utils/api_utils.py
@@ -371,57 +371,25 @@ def get_parser_config(chunk_method, parser_config):
"auto_questions": 0,
"html4excel": False,
"topn_tags": 3,
- "raptor": {
- "use_raptor": True,
- "prompt": "Summarize the paragraphs below without inventing facts or changing numbers.\nOutput exactly two parts in the same language as the source:\n1. First line: a concise title only.\n2. Following lines: a concise summary of the content.\nDo not output labels, Markdown headings, bullet points, or any other commentary.\n\nParagraphs:\n{cluster_content}",
- "max_token": 512,
- "clustering_threshold": 0.3,
- "clustering_ratio": 0.5,
- "max_cluster": 64,
- "random_seed": 0,
- },
- "graphrag": {
- "use_graphrag": True,
- "entity_types": [
- "organization",
- "person",
- "geo",
- "event",
- "category",
- ],
- "method": "light",
- "batch_chunk_token_size": 4096,
- "retry_attempts": 2,
- "retry_backoff_seconds": 2.0,
- "retry_backoff_max_seconds": 60.0,
- "build_subgraph_timeout_per_chunk_seconds": 300,
- "build_subgraph_min_timeout_seconds": 600,
- "merge_timeout_seconds": 180,
- "resolution_timeout_seconds": 1800,
- "community_timeout_seconds": 1800,
- "lock_acquire_timeout_seconds": 600,
- },
"parent_child": {
"use_parent_child": False,
"children_delimiter": "\n",
},
},
- "qa": {"raptor": {"use_raptor": False}, "graphrag": {"use_graphrag": False}},
+ "qa": {},
"tag": None,
"resume": None,
- "manual": {"raptor": {"use_raptor": False}, "graphrag": {"use_graphrag": False}},
+ "manual": {},
"table": None,
- "paper": {"raptor": {"use_raptor": False}, "graphrag": {"use_graphrag": False}},
- "book": {"raptor": {"use_raptor": False}, "graphrag": {"use_graphrag": False}},
- "laws": {"raptor": {"use_raptor": False}, "graphrag": {"use_graphrag": False}},
- "presentation": {"raptor": {"use_raptor": False}, "graphrag": {"use_graphrag": False}},
+ "paper": {},
+ "book": {},
+ "laws": {},
+ "presentation": {},
"one": None,
"knowledge_graph": {
"chunk_token_num": 8192,
"delimiter": "\n",
"entity_types": ["organization", "person", "location", "event", "time"],
- "raptor": {"use_raptor": False},
- "graphrag": {"use_graphrag": False},
},
"email": None,
"picture": None,
@@ -644,6 +612,14 @@ def deep_merge(default: dict, custom: dict) -> dict:
return merged
+def strip_graphrag_raptor_config(source_data: dict) -> dict:
+ result = dict(source_data)
+ parser_config = source_data.get("parser_config")
+ if isinstance(parser_config, dict):
+ result["parser_config"] = {key: value for key, value in parser_config.items() if key not in ("graphrag", "raptor")}
+ return result
+
+
def remap_dictionary_keys(source_data: dict, key_aliases: dict = None) -> dict:
"""
Transform dictionary keys using a configurable mapping schema.
@@ -676,7 +652,7 @@ def remap_dictionary_keys(source_data: dict, key_aliases: dict = None) -> dict:
mapped_key = mapping.get(original_key, original_key)
transformed_data[mapped_key] = value
- return transformed_data
+ return strip_graphrag_raptor_config(transformed_data)
def group_by(list_of_dict, key):
diff --git a/test/testcases/configs.py b/test/testcases/configs.py
index 608b25d59e..280b659ec4 100644
--- a/test/testcases/configs.py
+++ b/test/testcases/configs.py
@@ -86,36 +86,6 @@ DEFAULT_PARSER_CONFIG = {
"table_context_size": 0,
"topn_tags": 3,
"llm_id": "glm-4-flash@CI@ZHIPU-AI",
- "raptor": {
- "use_raptor": True,
- "prompt": "Summarize the paragraphs below without inventing facts or changing numbers.\nOutput exactly two parts in the same language as the source:\n1. First line: a concise title only.\n2. Following lines: a concise summary of the content.\nDo not output labels, Markdown headings, bullet points, or any other commentary.\n\nParagraphs:\n{cluster_content}",
- "max_token": 512,
- "clustering_threshold": 0.3,
- "clustering_ratio": 0.5,
- "max_cluster": 64,
- "random_seed": 0,
- },
- "graphrag": {
- "use_graphrag": True,
- "entity_types": [
- "organization",
- "person",
- "geo",
- "event",
- "category",
- ],
- "method": "light",
- "batch_chunk_token_size": 4096,
- "retry_attempts": 2,
- "retry_backoff_seconds": 2.0,
- "retry_backoff_max_seconds": 60.0,
- "build_subgraph_timeout_per_chunk_seconds": 300,
- "build_subgraph_min_timeout_seconds": 600,
- "merge_timeout_seconds": 180,
- "resolution_timeout_seconds": 1800,
- "community_timeout_seconds": 1800,
- "lock_acquire_timeout_seconds": 600,
- },
"parent_child": {
"use_parent_child": False,
"children_delimiter": "\n",
diff --git a/test/testcases/restful_api/test_datasets.py b/test/testcases/restful_api/test_datasets.py
index 366eb87ba9..77233bec89 100644
--- a/test/testcases/restful_api/test_datasets.py
+++ b/test/testcases/restful_api/test_datasets.py
@@ -362,6 +362,9 @@ def test_dataset_update_parser_config_valid_matrix_contract(rest_client, clear_d
assert list_payload["code"] == 0, list_payload
actual_parser_config = list_payload["data"][0]["parser_config"]
for key, expected_value in parser_config.items():
+ if key in {"graphrag", "raptor"}:
+ assert key not in actual_parser_config, list_payload
+ continue
if isinstance(expected_value, dict):
for nested_key, nested_expected in expected_value.items():
assert actual_parser_config[key][nested_key] == nested_expected, list_payload
@@ -1296,10 +1299,8 @@ def test_dataset_create_parser_config_missing_raptor_and_graphrag(rest_client, c
body = res.json()
assert body["code"] == 0, body
parser_config = body["data"]["parser_config"]
- assert "raptor" in parser_config, body
- assert "graphrag" in parser_config, body
- assert parser_config["raptor"]["use_raptor"] is False, body
- assert parser_config["graphrag"]["use_graphrag"] is False, body
+ assert "raptor" not in parser_config, body
+ assert "graphrag" not in parser_config, body
assert parser_config["chunk_token_num"] == 1024, body
@@ -1448,6 +1449,9 @@ def test_dataset_create_parser_config_valid_matrix_contract(rest_client, clear_d
assert body["code"] == 0, body
actual_parser_config = body["data"]["parser_config"]
for key, expected_value in parser_config.items():
+ if key in {"graphrag", "raptor"}:
+ assert key not in actual_parser_config, body
+ continue
if isinstance(expected_value, dict):
for nested_key, nested_expected in expected_value.items():
assert actual_parser_config[key][nested_key] == nested_expected, body
@@ -1457,20 +1461,18 @@ def test_dataset_create_parser_config_valid_matrix_contract(rest_client, clear_d
@pytest.mark.p1
@pytest.mark.parametrize(
- "name, parser_config, expected_raptor, expected_graphrag",
+ "name, parser_config",
[
- ("test_parser_config_only_raptor", {"chunk_token_num": 1024, "raptor": {"use_raptor": True}}, True, False),
- ("test_parser_config_only_graphrag", {"chunk_token_num": 1024, "graphrag": {"use_graphrag": True}}, False, True),
+ ("test_parser_config_only_raptor", {"chunk_token_num": 1024, "raptor": {"use_raptor": True}}),
+ ("test_parser_config_only_graphrag", {"chunk_token_num": 1024, "graphrag": {"use_graphrag": True}}),
(
"test_parser_config_both_fields",
{"chunk_token_num": 1024, "raptor": {"use_raptor": True}, "graphrag": {"use_graphrag": True}},
- True,
- True,
),
],
ids=["only_raptor", "only_graphrag", "both_fields"],
)
-def test_dataset_create_parser_config_bugfix_contract(rest_client, clear_datasets, name, parser_config, expected_raptor, expected_graphrag):
+def test_dataset_create_parser_config_bugfix_contract(rest_client, clear_datasets, name, parser_config):
if IS_GO_PROXY:
pytest.skip("Go CreateDataset does not accept parser_config")
res = rest_client.post("/datasets", json={"name": name, "parser_config": parser_config})
@@ -1478,10 +1480,8 @@ def test_dataset_create_parser_config_bugfix_contract(rest_client, clear_dataset
body = res.json()
assert body["code"] == 0, body
actual_parser_config = body["data"]["parser_config"]
- assert "raptor" in actual_parser_config, body
- assert "graphrag" in actual_parser_config, body
- assert actual_parser_config["raptor"]["use_raptor"] is expected_raptor, body
- assert actual_parser_config["graphrag"]["use_graphrag"] is expected_graphrag, body
+ assert "raptor" not in actual_parser_config, body
+ assert "graphrag" not in actual_parser_config, body
assert actual_parser_config["chunk_token_num"] == 1024, body
@@ -1505,10 +1505,8 @@ def test_dataset_create_parser_config_different_chunk_methods_contract(rest_clie
assert body["code"] == 0, body
parser_config = body["data"]["parser_config"]
assert parser_config["chunk_token_num"] == 512, body
- assert "raptor" in parser_config, body
- assert "graphrag" in parser_config, body
- assert parser_config["raptor"]["use_raptor"] is False, body
- assert parser_config["graphrag"]["use_graphrag"] is False, body
+ assert "raptor" not in parser_config, body
+ assert "graphrag" not in parser_config, body
def test_dataset_create_name_invalid_and_duplicate_contract(rest_client, clear_datasets):
diff --git a/test/testcases/restful_api/test_documents.py b/test/testcases/restful_api/test_documents.py
index c10e12f417..c7981da8f3 100644
--- a/test/testcases/restful_api/test_documents.py
+++ b/test/testcases/restful_api/test_documents.py
@@ -824,6 +824,9 @@ def test_documents_update_parser_config_contract(rest_client, create_dataset, tm
assert doc_parser_config == DEFAULT_PARSER_CONFIG, (parser_config, list_body)
else:
for key, value in parser_config.items():
+ if key in {"graphrag", "raptor"}:
+ assert key not in doc_parser_config, (parser_config, list_body)
+ continue
if isinstance(value, dict):
for sub_key, sub_value in value.items():
assert doc_parser_config[key][sub_key] == sub_value, (parser_config, list_body)
diff --git a/test/testcases/test_sdk_api/test_dataset_mangement/test_create_dataset.py b/test/testcases/test_sdk_api/test_dataset_mangement/test_create_dataset.py
index 8a60697e14..39de107137 100644
--- a/test/testcases/test_sdk_api/test_dataset_mangement/test_create_dataset.py
+++ b/test/testcases/test_sdk_api/test_dataset_mangement/test_create_dataset.py
@@ -463,6 +463,9 @@ class TestDatasetCreate:
payload = {"name": name, "parser_config": parser_config_o}
dataset = client.create_dataset(**payload)
for k, v in parser_config.items():
+ if k in {"graphrag", "raptor"}:
+ assert not hasattr(dataset.parser_config, k), str(dataset)
+ continue
if isinstance(v, dict):
for kk, vv in v.items():
assert attrgetter(f"{k}.{kk}")(dataset.parser_config) == vv, str(dataset)
@@ -653,10 +656,8 @@ class TestParserConfigBugFix:
dataset = client.create_dataset(**payload)
config = dataset.parser_config
- assert hasattr(config, "raptor"), "raptor field should be present"
- assert hasattr(config, "graphrag"), "graphrag field should be present"
- assert config.raptor.use_raptor is False, "raptor.use_raptor should default to False"
- assert config.graphrag.use_graphrag is False, "graphrag.use_graphrag should default to False"
+ assert not hasattr(config, "raptor"), "raptor field should not be exposed"
+ assert not hasattr(config, "graphrag"), "graphrag field should not be exposed"
assert config.chunk_token_num == 1024, "User-provided chunk_token_num should be preserved"
@pytest.mark.p1
@@ -666,9 +667,8 @@ class TestParserConfigBugFix:
dataset = client.create_dataset(**payload)
config = dataset.parser_config
- assert config.raptor.use_raptor is True, "User-provided raptor.use_raptor should be preserved"
- assert hasattr(config, "graphrag"), "graphrag field should be present"
- assert config.graphrag.use_graphrag is False, "graphrag.use_graphrag should default to False"
+ assert not hasattr(config, "raptor"), "raptor field should not be exposed"
+ assert not hasattr(config, "graphrag"), "graphrag field should not be exposed"
@pytest.mark.p1
def test_parser_config_with_only_graphrag(self, client):
@@ -677,9 +677,8 @@ class TestParserConfigBugFix:
dataset = client.create_dataset(**payload)
config = dataset.parser_config
- assert hasattr(config, "raptor"), "raptor field should be present"
- assert config.raptor.use_raptor is False, "raptor.use_raptor should default to False"
- assert config.graphrag.use_graphrag is True, "User-provided graphrag.use_graphrag should be preserved"
+ assert not hasattr(config, "raptor"), "raptor field should not be exposed"
+ assert not hasattr(config, "graphrag"), "graphrag field should not be exposed"
@pytest.mark.p1
def test_parser_config_with_both_fields(self, client):
@@ -688,8 +687,8 @@ class TestParserConfigBugFix:
dataset = client.create_dataset(**payload)
config = dataset.parser_config
- assert config.raptor.use_raptor is True, "User-provided raptor.use_raptor should be preserved"
- assert config.graphrag.use_graphrag is True, "User-provided graphrag.use_graphrag should be preserved"
+ assert not hasattr(config, "raptor"), "raptor field should not be exposed"
+ assert not hasattr(config, "graphrag"), "graphrag field should not be exposed"
@pytest.mark.p2
@pytest.mark.parametrize("chunk_method", ["qa", "manual", "paper", "book", "laws", "presentation"])
@@ -699,7 +698,5 @@ class TestParserConfigBugFix:
dataset = client.create_dataset(**payload)
config = dataset.parser_config
- assert hasattr(config, "raptor"), f"raptor field should be present for {chunk_method}"
- assert hasattr(config, "graphrag"), f"graphrag field should be present for {chunk_method}"
- assert config.raptor.use_raptor is False, f"raptor.use_raptor should default to False for {chunk_method}"
- assert config.graphrag.use_graphrag is False, f"graphrag.use_graphrag should default to False for {chunk_method}"
+ assert not hasattr(config, "raptor"), f"raptor field should not be exposed for {chunk_method}"
+ assert not hasattr(config, "graphrag"), f"graphrag field should not be exposed for {chunk_method}"
diff --git a/test/testcases/test_sdk_api/test_dataset_mangement/test_update_dataset.py b/test/testcases/test_sdk_api/test_dataset_mangement/test_update_dataset.py
index 4047ec9b26..543245e0dd 100644
--- a/test/testcases/test_sdk_api/test_dataset_mangement/test_update_dataset.py
+++ b/test/testcases/test_sdk_api/test_dataset_mangement/test_update_dataset.py
@@ -524,6 +524,9 @@ class TestDatasetUpdate:
dataset = add_dataset_func
dataset.update({"parser_config": parser_config})
for k, v in parser_config.items():
+ if k in {"graphrag", "raptor"}:
+ assert not hasattr(dataset.parser_config, k), str(dataset)
+ continue
if isinstance(v, dict):
for kk, vv in v.items():
assert attrgetter(f"{k}.{kk}")(dataset.parser_config) == vv, str(dataset)
@@ -532,6 +535,9 @@ class TestDatasetUpdate:
retrieved_dataset = client.get_dataset(name=dataset.name)
for k, v in parser_config.items():
+ if k in {"graphrag", "raptor"}:
+ assert not hasattr(retrieved_dataset.parser_config, k), str(retrieved_dataset)
+ continue
if isinstance(v, dict):
for kk, vv in v.items():
assert attrgetter(f"{k}.{kk}")(retrieved_dataset.parser_config) == vv, str(retrieved_dataset)
@@ -688,8 +694,6 @@ class TestDatasetUpdate:
expected_config = DataSet.ParserConfig(
client,
{
- "raptor": {"use_raptor": False},
- "graphrag": {"use_graphrag": False},
"image_context_size": 0,
"table_context_size": 0,
},
@@ -706,8 +710,6 @@ class TestDatasetUpdate:
expected_config = DataSet.ParserConfig(
client,
{
- "raptor": {"use_raptor": False},
- "graphrag": {"use_graphrag": False},
"image_context_size": 0,
"table_context_size": 0,
},
@@ -724,8 +726,6 @@ class TestDatasetUpdate:
expected_config = DataSet.ParserConfig(
client,
{
- "raptor": {"use_raptor": False},
- "graphrag": {"use_graphrag": False},
"image_context_size": 0,
"table_context_size": 0,
},
diff --git a/test/testcases/test_sdk_api/test_file_management_within_dataset/test_update_document.py b/test/testcases/test_sdk_api/test_file_management_within_dataset/test_update_document.py
index 9ad523b498..81abe6363b 100644
--- a/test/testcases/test_sdk_api/test_file_management_within_dataset/test_update_document.py
+++ b/test/testcases/test_sdk_api/test_file_management_within_dataset/test_update_document.py
@@ -498,6 +498,9 @@ class TestUpdateDocumentParserConfig:
updated_doc = [doc for doc in docs if doc.id == document.id][0]
if parser_config:
for k, v in parser_config.items():
+ if k in {"graphrag", "raptor"}:
+ assert not hasattr(updated_doc.parser_config, k), str(updated_doc)
+ continue
if isinstance(v, dict):
for kk, vv in v.items():
assert attrgetter(f"{k}.{kk}")(updated_doc.parser_config) == vv, str(updated_doc)
diff --git a/test/unit_test/api/apps/services/test_update_document_name_only.py b/test/unit_test/api/apps/services/test_update_document_name_only.py
index 2b1ecac7ed..72b8fa8225 100644
--- a/test/unit_test/api/apps/services/test_update_document_name_only.py
+++ b/test/unit_test/api/apps/services/test_update_document_name_only.py
@@ -45,6 +45,7 @@ def _load_update_document_name_only_module(monkeypatch, *, file_lookup):
doc = SimpleNamespace(id="doc-1", kb_id="kb-1")
f2d_row = SimpleNamespace(file_id="missing-file-id")
+ _stub(monkeypatch, "api.db.services.document_counter_service", release_reparse_counters=lambda doc_id: None)
_stub(
monkeypatch,
"api.db.services.document_service",
@@ -75,6 +76,7 @@ def _load_update_document_name_only_module(monkeypatch, *, file_lookup):
get_error_data_result=lambda **kwargs: kwargs,
server_error_response=lambda e: {"error": str(e)},
get_parser_config=lambda doc: {},
+ strip_graphrag_raptor_config=lambda data: data,
)
_stub(monkeypatch, "api.utils.validation_utils", UpdateDocumentReq=object)
_stub(monkeypatch, "api.utils", validation_utils=sys.modules["api.utils.validation_utils"])
diff --git a/web/src/hooks/parser-config-utils.ts b/web/src/hooks/parser-config-utils.ts
index 08ab8850cf..d0ab2462d2 100644
--- a/web/src/hooks/parser-config-utils.ts
+++ b/web/src/hooks/parser-config-utils.ts
@@ -15,9 +15,9 @@
*/
/**
- * Utility functions for extracting parser and raptor config extensions.
- * These functions extract known fields from parser/raptor config objects
- * and merge unknown fields into the `ext` field for flexible configuration.
+ * Utility functions for extracting parser config extensions.
+ * These functions extract known fields from parser config objects and merge
+ * unknown fields into the `ext` field for flexible configuration.
*/
/**
@@ -34,52 +34,6 @@ export const isPipelineParserConfig = (
return Object.keys(parserConfig).some((key) => key.includes(':'));
};
-/**
- * Extracts Raptor configuration with extra fields merged into ext.
- * @param raptorConfig - The raptor configuration object
- * @returns Processed raptor config with extra fields in ext
- */
-export const extractRaptorConfigExt = (
- raptorConfig: Record<string, any> | undefined,
-) => {
- if (!raptorConfig) return raptorConfig;
- const {
- use_raptor,
- prompt,
- max_token,
- threshold,
- max_cluster,
- random_seed,
- scope,
- clustering_method,
- tree_builder,
- auto_disable_for_structured_data,
- ext,
- ...raptorExt
- } = raptorConfig;
- const extClusteringMethod = ext?.clustering_method;
- const normalizedClusteringMethod =
- clustering_method ?? extClusteringMethod ?? 'gmm';
- const normalizedTreeBuilder = tree_builder ?? ext?.tree_builder ?? 'raptor';
-
- return {
- use_raptor,
- prompt,
- max_token,
- threshold,
- max_cluster,
- random_seed,
- scope,
- auto_disable_for_structured_data,
- ext: {
- ...ext,
- ...raptorExt,
- clustering_method: normalizedClusteringMethod,
- tree_builder: normalizedTreeBuilder,
- },
- };
-};
-
/**
* Extracts Parser configuration with extra fields merged into ext.
* @param parserConfig - The parser configuration object
@@ -94,10 +48,8 @@ export const extractParserConfigExt = (
auto_questions,
chunk_token_num,
delimiter,
- graphrag,
html4excel,
layout_recognize,
- raptor,
tag_kb_ids,
topn_tags,
filename_embd_weight,
@@ -109,15 +61,15 @@ export const extractParserConfigExt = (
ext,
...parserExt
} = parserConfig;
+ delete parserExt.graphrag;
+ delete parserExt.raptor;
return {
auto_keywords,
auto_questions,
chunk_token_num,
delimiter,
- graphrag,
html4excel,
layout_recognize,
- raptor: extractRaptorConfigExt(raptor),
tag_kb_ids,
topn_tags,
filename_embd_weight,
diff --git a/web/src/hooks/tests/parser-config-utils.test.ts b/web/src/hooks/tests/parser-config-utils.test.ts
deleted file mode 100644
index 6bbfcf0cb6..0000000000
--- a/web/src/hooks/tests/parser-config-utils.test.ts
+++ /dev/null
@@ -1,45 +0,0 @@
-import { extractParserConfigExt } from '../parser-config-utils';
-
-describe('extractParserConfigExt', () => {
- it('serializes RAPTOR clustering fields through ext for API compatibility', () => {
- const result = extractParserConfigExt({
- raptor: {
- use_raptor: true,
- prompt: 'Summarize {cluster_content}',
- max_token: 256,
- threshold: 0.1,
- max_cluster: 317,
- random_seed: 0,
- scope: 'file',
- clustering_method: 'ahc',
- tree_builder: 'raptor',
- },
- });
-
- expect(result?.raptor).not.toHaveProperty('clustering_method');
- expect(result?.raptor).not.toHaveProperty('tree_builder');
- expect(result?.raptor?.ext).toMatchObject({
- clustering_method: 'ahc',
- tree_builder: 'raptor',
- });
- });
-
- it('preserves existing RAPTOR ext clustering values when the top-level field is absent', () => {
- const result = extractParserConfigExt({
- raptor: {
- max_cluster: 512,
- ext: {
- clustering_method: 'ahc',
- tree_builder: 'raptor',
- psi_bucket_size: 1024,
- },
- },
- });
-
- expect(result?.raptor?.ext).toMatchObject({
- clustering_method: 'ahc',
- tree_builder: 'raptor',
- psi_bucket_size: 1024,
- });
- });
-});
diff --git a/web/src/pages/dataset/setting/python/form-schema.ts b/web/src/pages/dataset/setting/python/form-schema.ts
index 4227e5bae3..6f92ba9dfc 100644
--- a/web/src/pages/dataset/setting/python/form-schema.ts
+++ b/web/src/pages/dataset/setting/python/form-schema.ts
@@ -45,60 +45,6 @@ export const formSchema = z
mineru_formula_enable: z.boolean().optional(),
mineru_table_enable: z.boolean().optional(),
mineru_lang: z.string().optional(),
- raptor: z
- .object({
- use_raptor: z.boolean().optional(),
- prompt: z.string().optional(),
- max_token: z.coerce.number().optional(),
- threshold: z.coerce.number().optional(),
- max_cluster: z.coerce.number().optional(),
- random_seed: z.coerce.number().optional(),
- scope: z.string().optional(),
- clustering_method: z.enum(['gmm', 'ahc']).optional(),
- tree_builder: z.enum(['raptor', 'psi']).optional(),
- ext: z.record(z.string(), z.any()).optional(),
- })
- .refine(
- (data) => {
- if (data.use_raptor && !data.prompt) {
- return false;
- }
- return true;
- },
- {
- message: 'Prompt is required',
- path: ['prompt'],
- },
- ),
- graphrag: z
- .object({
- use_graphrag: z.boolean().optional(),
- entity_types: z.array(z.string()).optional(),
- method: z.string().optional(),
- resolution: z.boolean().optional(),
- community: z.boolean().optional(),
- batch_chunk_token_size: z
- .number()
- .int()
- .min(512)
- .max(8196)
- .optional(),
- })
- .refine(
- (data) => {
- if (
- data.use_graphrag &&
- (!data.entity_types || data.entity_types.length === 0)
- ) {
- return false;
- }
- return true;
- },
- {
- message: 'Please enter Entity types',
- path: ['entity_types'],
- },
- ),
metadata: z.any().optional(),
built_in_metadata: z
.array(
diff --git a/web/src/pages/dataset/setting/python/hooks.test.tsx b/web/src/pages/dataset/setting/python/hooks.test.tsx
deleted file mode 100644
index 0df505209c..0000000000
--- a/web/src/pages/dataset/setting/python/hooks.test.tsx
+++ /dev/null
@@ -1,78 +0,0 @@
-import { useFetchKnowledgeBaseConfiguration } from '@/hooks/use-knowledge-request';
-import { renderHook } from '@testing-library/react';
-import { useFetchKnowledgeConfigurationOnMount } from './hooks';
-
-jest.mock('@/hooks/use-knowledge-request', () => ({
- useFetchKnowledgeBaseConfiguration: jest.fn(),
-}));
-jest.mock('@/hooks/common-hooks', () => ({
- useSetModalState: jest.fn(),
-}));
-jest.mock('@/hooks/use-user-setting-request', () => ({
- useSelectParserList: jest.fn(() => []),
-}));
-jest.mock('@/services/knowledge-service', () => ({
- checkEmbedding: jest.fn(),
-}));
-jest.mock('react-router', () => ({
- useParams: jest.fn(() => ({})),
- useSearchParams: jest.fn(() => [new URLSearchParams()]),
-}));
-
-const mockUseFetchKnowledgeBaseConfiguration = jest.mocked(
- useFetchKnowledgeBaseConfiguration,
-);
-
-function renderConfigurationHook(parserConfig: Record<string, unknown>) {
- mockUseFetchKnowledgeBaseConfiguration.mockReturnValue({
- data: {
- name: 'Dataset',
- parser_config: parserConfig,
- embedding_model: 'BAAI/bge-large-zh-v1.5',
- chunk_method: 'naive',
- },
- loading: false,
- } as never);
-
- const reset = jest.fn();
- const form = {
- formState: {
- defaultValues: {
- parser_config: {
- raptor: { use_raptor: true },
- graphrag: { use_graphrag: true },
- },
- },
- },
- reset,
- } as never;
-
- renderHook(() => useFetchKnowledgeConfigurationOnMount(form));
-
- return reset.mock.calls[0][0].parser_config;
-}
-
-describe('useFetchKnowledgeConfigurationOnMount', () => {
- it('preserves explicitly disabled RAPTOR and GraphRAG settings', () => {
- const parserConfig = renderConfigurationHook({
- raptor: {
- use_raptor: false,
- ext: { clustering_method: 'gmm' },
- },
- graphrag: { use_graphrag: false },
- });
-
- expect(parserConfig.raptor.use_raptor).toBe(false);
- expect(parserConfig.graphrag.use_graphrag).toBe(false);
- });
-
- it('keeps form defaults when saved settings omit the enable flags', () => {
- const parserConfig = renderConfigurationHook({
- raptor: { ext: { clustering_method: 'gmm' } },
- graphrag: {},
- });
-
- expect(parserConfig.raptor.use_raptor).toBe(true);
- expect(parserConfig.graphrag.use_graphrag).toBe(true);
- });
-});
diff --git a/web/src/pages/dataset/setting/python/hooks.ts b/web/src/pages/dataset/setting/python/hooks.ts
index eed3e89014..42aebe4985 100644
--- a/web/src/pages/dataset/setting/python/hooks.ts
+++ b/web/src/pages/dataset/setting/python/hooks.ts
@@ -37,16 +37,6 @@ export const useFetchKnowledgeConfigurationOnMount = (
const parser_config = {
...form.formState?.defaultValues?.parser_config,
...knowledgeDetails.parser_config,
- raptor: {
- ...form.formState?.defaultValues?.parser_config?.raptor,
- ...knowledgeDetails.parser_config?.raptor,
- clustering_method:
- knowledgeDetails.parser_config?.raptor?.ext?.clustering_method,
- },
- graphrag: {
- ...form.formState?.defaultValues?.parser_config?.graphrag,
- ...knowledgeDetails.parser_config?.graphrag,
- },
};
const formValues = {
...pick({ ...knowledgeDetails, parser_config: parser_config }, [
diff --git a/web/src/pages/dataset/setting/python/index.tsx b/web/src/pages/dataset/setting/python/index.tsx
index 0e004c3b12..c7fd5ccbe9 100644
--- a/web/src/pages/dataset/setting/python/index.tsx
+++ b/web/src/pages/dataset/setting/python/index.tsx
@@ -41,20 +41,6 @@ export const DataSetContext = createContext<{
knowledgeDetails: IDataset;
}>({ loading: false, knowledgeDetails: {} as IDataset });
-const initialEntityTypes = [
- 'organization',
- 'person',
- 'geo',
- 'event',
- 'category',
-];
-
-const enum MethodValue {
- General = 'general',
- Light = 'light',
- NER = 'ner',
-}
-
export default function DatasetSettings() {
const { t } = useTranslation();
@@ -82,23 +68,6 @@ export default function DatasetSettings() {
mineru_formula_enable: true,
mineru_table_enable: true,
mineru_lang: 'English',
- raptor: {
- use_raptor: true,
- max_token: 256,
- threshold: 0.1,
- max_cluster: 64,
- random_seed: 0,
- scope: 'file',
- clustering_method: 'gmm',
- tree_builder: 'raptor',
- prompt: t('knowledgeConfiguration.promptText'),
- },
- graphrag: {
- use_graphrag: true,
- entity_types: initialEntityTypes,
- method: MethodValue.Light,
- batch_chunk_token_size: 4096,
- },
metadata: {
type: 'object',
properties: {},