phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2026-01-02 20:20:01 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Added output type to PreprocessorResponse.

Browse source
This commit is contained in:
ilya-kolchinsky 2025-03-06 14:05:05 +01:00
parent b981181b25
commit 4c81a72214
7 changed files with 33 additions and 25 deletions
Download patch file Download diff file Expand all files Collapse all files

5
llama_stack/providers/inline/preprocessing/basic/basic.py
View file

@ -88,13 +88,14 @@ class InclineBasicPreprocessorImpl(Preprocessing, PreprocessorsProtocolPrivate):
results.append(document)
return PreprocessorResponse(status=True, results=results)
return PreprocessorResponse(
success=True, preprocessor_output_type=PreprocessingDataType.raw_text_document, results=results
)
async def chain_preprocess(
self,
preprocessors: PreprocessorChain,
preprocessor_inputs: List[PreprocessorInput],
is_rag_chain: Optional[bool] = False,
) -> PreprocessorResponse:
return await self.preprocess(preprocessor_id="", preprocessor_inputs=preprocessor_inputs)

6
llama_stack/providers/inline/preprocessing/docling/docling.py
View file

@ -75,12 +75,14 @@ class InclineDoclingPreprocessorImpl(Preprocessing, PreprocessorsProtocolPrivate
result = converted_document.export_to_markdown()
results.append(result)
return PreprocessorResponse(status=True, results=results)
preprocessor_output_type = (
PreprocessingDataType.chunks if self.config.chunk else PreprocessingDataType.raw_text_document
)
return PreprocessorResponse(success=True, preprocessor_output_type=preprocessor_output_type, results=results)
async def chain_preprocess(
self,
preprocessors: PreprocessorChain,
preprocessor_inputs: List[PreprocessorInput],
is_rag_chain: Optional[bool] = False,
) -> PreprocessorResponse:
return await self.preprocess(preprocessor_id="", preprocessor_inputs=preprocessor_inputs)

3
llama_stack/providers/inline/preprocessing/simple_chunking/simple_chunking.py
View file

@ -62,13 +62,12 @@ class InclineSimpleChunkingImpl(Preprocessing, PreprocessorsProtocolPrivate):
)
chunks.extend(new_chunks)
return PreprocessorResponse(status=True, results=chunks)
return PreprocessorResponse(success=True, preprocessor_output_type=PreprocessingDataType.chunks, results=chunks)
async def chain_preprocess(
self,
preprocessors: PreprocessorChain,
preprocessor_inputs: List[PreprocessorInput],
is_rag_chain: Optional[bool] = False,
) -> PreprocessorResponse:
return await self.preprocess(preprocessor_id="", preprocessor_inputs=preprocessor_inputs)

13
llama_stack/providers/inline/tool_runtime/rag/memory.py
View file

@ -21,6 +21,7 @@ from llama_stack.apis.inference import Inference
from llama_stack.apis.preprocessing import (
Preprocessing,
PreprocessingDataFormat,
PreprocessingDataType,
PreprocessorChainElement,
PreprocessorInput,
)
@ -81,9 +82,19 @@ class MemoryToolRuntimeImpl(ToolsProtocolPrivate, ToolRuntime, RAGToolRuntime):
preprocessors=preprocessor_chain, preprocessor_inputs=preprocessor_inputs
)
chunks = preprocessor_response.results
if not preprocessor_response.success:
log.error("Preprocessor chain returned an error")
return
if preprocessor_response.preprocessor_output_type != PreprocessingDataType.chunks:
log.error(
f"Preprocessor chain returned {preprocessor_response.preprocessor_output_type} instead of {PreprocessingDataType.chunks}"
)
return
chunks = preprocessor_response.results
if not chunks:
log.error("No chunks returned by the preprocessor chain")
return
await self.vector_io_api.insert_chunks(