Source code for langchain_community.document_loaders.xml
"""加载Microsoft Excel文件。"""
from pathlib import Path
from typing import Any, List, Union
from langchain_community.document_loaders.unstructured import (
UnstructuredFileLoader,
validate_unstructured_version,
)
[docs]class UnstructuredXMLLoader(UnstructuredFileLoader):
"""使用`Unstructured`加载`XML`文件。
可以在两种模式中的一种中运行加载程序:"single"和"elements"。
如果使用"single"模式,文档将作为单个`langchain Document`对象返回。
如果使用"elements"模式,`unstructured`库将文档拆分为诸如Title和NarrativeText之类的元素。
您可以在模式之后传递额外的`unstructured kwargs`以应用不同的`unstructured settings`。
示例
--------
```python
from langchain_community.document_loaders import UnstructuredXMLLoader
loader = UnstructuredXMLLoader(
"example.xml", mode="elements", strategy="fast",
)
docs = loader.load()
```
参考资料
----------
https://unstructured-io.github.io/unstructured/bricks.html#partition-xml"""
[docs] def __init__(
self,
file_path: Union[str, Path],
mode: str = "single",
**unstructured_kwargs: Any,
):
file_path = str(file_path)
validate_unstructured_version(min_unstructured_version="0.6.7")
super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
def _get_elements(self) -> List:
from unstructured.partition.xml import partition_xml
return partition_xml(filename=self.file_path, **self.unstructured_kwargs)