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

Fix: add the validation for parser_config. (#5755)

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

#5719

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
This commit is contained in:
Kevin Hu
2025-03-07 15:34:34 +08:00
committed by GitHub
parent b1bbb9e210
commit da3f279495
3 changed files with 43 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
sdk/python/test/test_http_api/test_dataset_mangement/test_create_dataset.py
View File

@ -178,7 +178,7 @@ class TestAdvancedConfigurations:
@pytest.mark.parametrize("name, chunk_method, parser_config, expected_code", [
("naive_default", "naive",
{"chunk_token_count": 128,
{"chunk_token_num": 128,
"layout_recognize": "DeepDOC",
"html4excel": False,
"delimiter": "\n!?。；！？",
@ -187,20 +187,20 @@ class TestAdvancedConfigurations:
},
0),
("naive_empty", "naive", {}, 0),
pytest.param("naive_chunk_token_count_negative", "naive",
{"chunk_token_count": -1},
pytest.param("naive_chunk_token_num_negative", "naive",
{"chunk_token_num": -1},
102, marks=pytest.mark.xfail(reason='issue#5719')),
pytest.param("naive_chunk_token_count_zero", "naive",
{"chunk_token_count": 0},
pytest.param("naive_chunk_token_num_zero", "naive",
{"chunk_token_num": 0},
102, marks=pytest.mark.xfail(reason='issue#5719')),
pytest.param("naive_chunk_token_count_float", "naive",
{"chunk_token_count": 3.14},
pytest.param("naive_chunk_token_num_float", "naive",
{"chunk_token_num": 3.14},
102, marks=pytest.mark.xfail(reason='issue#5719')),
pytest.param("naive_chunk_token_count_max", "naive",
{"chunk_token_count": 1024*1024*1024},
pytest.param("naive_chunk_token_num_max", "naive",
{"chunk_token_num": 1024*1024*1024},
102, marks=pytest.mark.xfail(reason='issue#5719')),
pytest.param("naive_chunk_token_count_str", "naive",
{"chunk_token_count": '1024'},
pytest.param("naive_chunk_token_num_str", "naive",
{"chunk_token_num": '1024'},
102, marks=pytest.mark.xfail(reason='issue#5719')),
("naive_layout_recognize_DeepDOC", "naive",
{"layout_recognize": "DeepDOC"}, 0),
@ -231,10 +231,6 @@ class TestAdvancedConfigurations:
102, marks=pytest.mark.xfail(reason='issue#5719')),
("naive_raptor_true", "naive", {"raptor": {"use_raptor": True}}, 0),
("naive_raptor_false", "naive", {"raptor": {"use_raptor": False}}, 0),
("knowledge_graph_entity_types_default", "knowledge_graph", {
"entity_types": ["organization", "person", "location", "event", "time"]}, 0),
pytest.param("knowledge_graph_entity_types_not_list", "knowledge_graph", {
"entity_types": "organization,person,location,event,time"}, 102, marks=pytest.mark.xfail(reason='issue#5719'))
])
def test_parser_configs(self, get_http_api_auth, name, chunk_method, parser_config, expected_code):
payload = {