Refactor DocumentLoaderVertex to use record instead of doc and update display message format

This commit is contained in:
Gabriel Luiz Freitas Almeida 2024-03-04 15:18:04 -03:00
commit 16c8d5be0c

View file

@ -1,6 +1,7 @@
import ast import ast
import json import json
from typing import AsyncIterator, Callable, Dict, Iterator, List, Optional, Union from typing import (AsyncIterator, Callable, Dict, Iterator, List, Optional,
Union)
import yaml import yaml
from langchain_core.messages import AIMessage from langchain_core.messages import AIMessage
@ -124,13 +125,13 @@ class DocumentLoaderVertex(Vertex):
if not isinstance(self._built_object, UnbuiltObject): if not isinstance(self._built_object, UnbuiltObject):
avg_length = sum( avg_length = sum(
len(doc.page_content) len(record.text)
for doc in self._built_object for record in self._built_object
if hasattr(doc, "page_content") if hasattr(record, "text")
) / len(self._built_object) ) / len(self._built_object)
return f"""{self.display_name}({len(self._built_object)} documents) return f"""{self.display_name}({len(self._built_object)} records)
\nAvg. Document Length (characters): {int(avg_length)} \nAvg. Record Length (characters): {int(avg_length)}
Documents: {self._built_object[:3]}...""" Records: {self._built_object[:3]}..."""
return f"{self.vertex_type}()" return f"{self.vertex_type}()"