Files
langchain-python/langchain/document_loaders/s3_directory.py
T
vinoyang 8073bc849f Minor: Remove duplicated word in error message (#2706)
Removed the duplicated word "it" from the error message.
From:
`Please it install it with xxx`
To:
`Please install it with xxx`.
2023-04-11 13:10:33 -07:00

33 lines
1.1 KiB
Python

"""Loading logic for loading documents from an s3 directory."""
from typing import List
from langchain.docstore.document import Document
from langchain.document_loaders.base import BaseLoader
from langchain.document_loaders.s3_file import S3FileLoader
class S3DirectoryLoader(BaseLoader):
"""Loading logic for loading documents from s3."""
def __init__(self, bucket: str, prefix: str = ""):
"""Initialize with bucket and key name."""
self.bucket = bucket
self.prefix = prefix
def load(self) -> List[Document]:
"""Load documents."""
try:
import boto3
except ImportError:
raise ValueError(
"Could not import boto3 python package. "
"Please install it with `pip install boto3`."
)
s3 = boto3.resource("s3")
bucket = s3.Bucket(self.bucket)
docs = []
for obj in bucket.objects.filter(Prefix=self.prefix):
loader = S3FileLoader(self.bucket, obj.key)
docs.extend(loader.load())
return docs