Feat/support parent child chunk (#12092)

This commit is contained in:
Jyong
2024-12-25 19:49:07 +08:00
committed by GitHub
parent 017d7538ae
commit 9231fdbf4c
54 changed files with 2578 additions and 808 deletions

View File

@@ -73,6 +73,7 @@ dataset_detail_fields = {
"embedding_available": fields.Boolean,
"retrieval_model_dict": fields.Nested(dataset_retrieval_model_fields),
"tags": fields.List(fields.Nested(tag_fields)),
"doc_form": fields.String,
"external_knowledge_info": fields.Nested(external_knowledge_info_fields),
"external_retrieval_model": fields.Nested(external_retrieval_model_fields, allow_null=True),
}

View File

@@ -34,6 +34,7 @@ document_with_segments_fields = {
"data_source_info": fields.Raw(attribute="data_source_info_dict"),
"data_source_detail_dict": fields.Raw(attribute="data_source_detail_dict"),
"dataset_process_rule_id": fields.String,
"process_rule_dict": fields.Raw(attribute="process_rule_dict"),
"name": fields.String,
"created_from": fields.String,
"created_by": fields.String,

View File

@@ -34,8 +34,16 @@ segment_fields = {
"document": fields.Nested(document_fields),
}
child_chunk_fields = {
"id": fields.String,
"content": fields.String,
"position": fields.Integer,
"score": fields.Float,
}
hit_testing_record_fields = {
"segment": fields.Nested(segment_fields),
"child_chunks": fields.List(fields.Nested(child_chunk_fields)),
"score": fields.Float,
"tsne_position": fields.Raw,
}

View File

@@ -2,6 +2,17 @@ from flask_restful import fields # type: ignore
from libs.helper import TimestampField
child_chunk_fields = {
"id": fields.String,
"segment_id": fields.String,
"content": fields.String,
"position": fields.Integer,
"word_count": fields.Integer,
"type": fields.String,
"created_at": TimestampField,
"updated_at": TimestampField,
}
segment_fields = {
"id": fields.String,
"position": fields.Integer,
@@ -20,10 +31,13 @@ segment_fields = {
"status": fields.String,
"created_by": fields.String,
"created_at": TimestampField,
"updated_at": TimestampField,
"updated_by": fields.String,
"indexing_at": TimestampField,
"completed_at": TimestampField,
"error": fields.String,
"stopped_at": TimestampField,
"child_chunks": fields.List(fields.Nested(child_chunk_fields)),
}
segment_list_response = {