opensource/dify
2
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-09-08 11:04:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix(api): default parent_mode to paragraph for hierarchical chunking via API (#34635)

Browse Source 
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
This commit is contained in:
plind 2026-04-10 19:55:40 +09:00 committed by GitHub
parent 2eb43b1e1f
commit 2dc015b360
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 31 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
api/services/dataset_service.py
View File

@ -2822,6 +2822,10 @@ class DocumentService:
knowledge_config.process_rule.rules.pre_processing_rules = list(unique_pre_processing_rule_dicts.values()) knowledge_config.process_rule.rules.pre_processing_rules = list(unique_pre_processing_rule_dicts.values())
if knowledge_config.process_rule.mode == ProcessRuleMode.HIERARCHICAL:
if not knowledge_config.process_rule.rules.parent_mode:
knowledge_config.process_rule.rules.parent_mode = "paragraph"
if not knowledge_config.process_rule.rules.segmentation: if not knowledge_config.process_rule.rules.segmentation:
raise ValueError("Process rule segmentation is required") raise ValueError("Process rule segmentation is required")

27
api/tests/unit_tests/services/test_dataset_service_document.py
View File

@ -1069,6 +1069,33 @@ class TestDocumentServiceCreateValidation:
assert len(knowledge_config.process_rule.rules.pre_processing_rules) == 1 assert len(knowledge_config.process_rule.rules.pre_processing_rules) == 1
assert knowledge_config.process_rule.rules.pre_processing_rules[0].enabled is False assert knowledge_config.process_rule.rules.pre_processing_rules[0].enabled is False
def test_process_rule_args_validate_hierarchical_defaults_parent_mode_to_paragraph(self):
knowledge_config = KnowledgeConfig(
indexing_technique="economy",
data_source=DataSource(
info_list=InfoList(
data_source_type="upload_file",
file_info_list=FileInfo(file_ids=["file-1"]),
)
),
process_rule=ProcessRule(
mode="hierarchical",
rules=Rule(
pre_processing_rules=[
PreProcessingRule(id="remove_extra_spaces", enabled=True),
],
segmentation=Segmentation(separator="\n", max_tokens=1024),
subchunk_segmentation=Segmentation(separator="\n", max_tokens=512),
),
),
)
DocumentService.process_rule_args_validate(knowledge_config)
assert knowledge_config.process_rule is not None
assert knowledge_config.process_rule.rules is not None
assert knowledge_config.process_rule.rules.parent_mode == "paragraph"
class TestDocumentServiceSaveDocumentWithDatasetId: class TestDocumentServiceSaveDocumentWithDatasetId:
"""Unit tests for non-SQL validation branches in save_document_with_dataset_id.""" """Unit tests for non-SQL validation branches in save_document_with_dataset_id."""