mirror of
https://github.com/csunny/DB-GPT.git
synced 2025-07-29 06:47:30 +00:00
FIX:1598 Use PageTextSplitter for DatasourceKnowledge (#1599)
Co-authored-by: shenk-b <shenk-b@glodon.com> Co-authored-by: aries_ckt <916701291@qq.com>
This commit is contained in:
parent
59b7aa790b
commit
20e7ccc831
@ -52,6 +52,7 @@ class DatasourceKnowledge(Knowledge):
|
|||||||
return [
|
return [
|
||||||
ChunkStrategy.CHUNK_BY_SIZE,
|
ChunkStrategy.CHUNK_BY_SIZE,
|
||||||
ChunkStrategy.CHUNK_BY_SEPARATOR,
|
ChunkStrategy.CHUNK_BY_SEPARATOR,
|
||||||
|
ChunkStrategy.CHUNK_BY_PAGE,
|
||||||
]
|
]
|
||||||
|
|
||||||
@classmethod
|
@classmethod
|
||||||
@ -63,3 +64,12 @@ class DatasourceKnowledge(Knowledge):
|
|||||||
def document_type(cls) -> DocumentType:
|
def document_type(cls) -> DocumentType:
|
||||||
"""Return document type."""
|
"""Return document type."""
|
||||||
return DocumentType.DATASOURCE
|
return DocumentType.DATASOURCE
|
||||||
|
|
||||||
|
@classmethod
|
||||||
|
def default_chunk_strategy(cls) -> ChunkStrategy:
|
||||||
|
"""Return default chunk strategy.
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
ChunkStrategy: default chunk strategy
|
||||||
|
"""
|
||||||
|
return ChunkStrategy.CHUNK_BY_PAGE
|
||||||
|
@ -101,8 +101,10 @@ class DBSummaryClient:
|
|||||||
from dbgpt.rag.assembler.db_schema import DBSchemaAssembler
|
from dbgpt.rag.assembler.db_schema import DBSchemaAssembler
|
||||||
|
|
||||||
db_assembler = DBSchemaAssembler.load_from_connection(
|
db_assembler = DBSchemaAssembler.load_from_connection(
|
||||||
connector=db_summary_client.db, vector_store_connector=vector_connector
|
connector=db_summary_client.db,
|
||||||
|
vector_store_connector=vector_connector,
|
||||||
)
|
)
|
||||||
|
|
||||||
if len(db_assembler.get_chunks()) > 0:
|
if len(db_assembler.get_chunks()) > 0:
|
||||||
db_assembler.persist()
|
db_assembler.persist()
|
||||||
else:
|
else:
|
||||||
|
Loading…
Reference in New Issue
Block a user