[fel] update llama splitter tool meta

Author: Mark-Tanbowu

framework/fel/python/plugins/fel_llama_splitter_tools/callable_registers.py

@@ -6,7 +6,7 @@
 import traceback
 from typing import Tuple, List, Any, Callable
 
-from fitframework import fit_logger
+from fitframework import fit_logger, fitable
 from llama_index.core.node_parser import (
     SentenceSplitter,
     TokenTextSplitter,
@@ -17,11 +17,11 @@
 from llama_index.core.schema import Document as LDocument
 from llama_index.embeddings.openai import OpenAIEmbedding
 
-from .callable_registers import register_callable_tool
 from .node_utils import to_llama_index_document
+from .types.semantic_splitter_options import SemanticSplitterOptions
 
-
-def sentence_splitter(text: str, separator: str, chunk_size: int, chunk_overlap: int, **kwargs) -> List[str]:
+@fitable("llama.tools.sentence_splitter", "default")
+def sentence_splitter(text: str, separator: str, chunk_size: int, chunk_overlap: int) -> List[str]:
     """Parse text with a preference for complete sentences."""
     if len(text) == 0:
         return []
@@ -38,7 +38,8 @@ def sentence_splitter(text: str, separator: str, chunk_size: int, chunk_overlap:
         return []
 
 
-def token_text_splitter(text: str, separator: str, chunk_size: int, chunk_overlap: int, **kwargs) -> List[str]:
+@fitable("llama.tools.token_text_splitter", "default")
+def token_text_splitter(text: str, separator: str, chunk_size: int, chunk_overlap: int) -> List[str]:
     """Splitting text that looks at word tokens."""
     if len(text) == 0:
         return []
@@ -55,14 +56,15 @@ def token_text_splitter(text: str, separator: str, chunk_size: int, chunk_overla
         return []
 
 
-def semantic_splitter(buffer_size: int, breakpoint_percentile_threshold: int, docs: List[LDocument], **kwargs) \
+# @fitable("llama.tools.semantic_splitter", "default")
+def semantic_splitter(buffer_size: int, breakpoint_percentile_threshold: int, docs: List[LDocument], options: SemanticSplitterOptions) \
         -> List[BaseNode]:
     """Splitting text that looks at word tokens."""
     if len(docs) == 0:
         return []
-    api_key = kwargs.get("api_key")
-    model_name = kwargs.get("model_name")
-    api_base = kwargs.get("api_base")
+    api_key = options.api_key
+    model_name = options.model_name
+    api_base = options.api_base
 
     embed_model = OpenAIEmbedding(model_name=model_name, api_base=api_base, api_key=api_key, max_tokens=4096)
 
@@ -80,8 +82,9 @@ def semantic_splitter(buffer_size: int, breakpoint_percentile_threshold: int, do
         return []
 
 
+# @fitable("llama.tools.sentence_window_node_parser", "default")
 def sentence_window_node_parser(window_size: int, window_metadata_key: str, original_text_metadata_key: str,
-                                docs: List[LDocument], **kwargs) -> List[BaseNode]:
+                                docs: List[LDocument]) -> List[BaseNode]:
     """Splitting text that looks at word tokens."""
     if len(docs) == 0:
         return []
@@ -96,26 +99,4 @@ def sentence_window_node_parser(window_size: int, window_metadata_key: str, orig
     except BaseException:
         fit_logger.error("Invoke semantic splitter failed.")
         traceback.print_exc()
-        return []
-
-
-# Tuple 结构： (tool_func, config_args, return_description)
-splitter_basic_toolkit: List[Tuple[Callable[..., Any], List[str], str]] = [
-    (sentence_splitter, ["text", "separator", "chunk_size", "chunk_overlap"], "Split sentences by sentence."),
-    (token_text_splitter, ["text", "separator", "chunk_size", "chunk_overlap"], "Split sentences by token."),
-    (semantic_splitter,
-     ["docs", "buffer_size", "breakpoint_percentile_threshold", "chunk_overlap", "model_name", "api_key", "api_base"],
-     "Split sentences by semantic."),
-    (sentence_window_node_parser, ["docs", "window_size", "window_metadata_key", "original_text_metadata_key"],
-     "Splits all documents into individual sentences")
-]
-
-for tool in splitter_basic_toolkit:
-    register_callable_tool(tool, sentence_splitter.__module__, "llama_index.rag.toolkit")
-
-if __name__ == '__main__':
-    import time
-    from .llama_schema_helper import dump_llama_schema
-
-    current_timestamp = time.strftime('%Y%m%d%H%M%S')
-    dump_llama_schema(splitter_basic_toolkit, f"./llama_tool_schema-{str(current_timestamp)}.json")
+        return []