熊猫和蔚蓝：DataFrame to_csv blob

Question

我想将python pandas DataFrame存储到Azure，但是Jay Gong（Python 2.7）的解决方案给了我错误：

from azure.storage.blob import BlockBlobService
service = BlockBlobService(
    account_name=STORAGE_ACCOUNT_NAME,
    account_key=STORAGE_ACCOUNT_KEY
)

with io.StringIO() as stream:
    df.to_csv(stream, encoding="utf-8")
    service.create_blob_from_text('containert', 'example.txt', stream)

返回：

  

AttributeError：'_io.StringIO'对象没有属性'encode'

• python 3.7
• 熊猫0.23.4
• azure-blob存储1.3.1

Answer 1

这有效，但并不十分优雅：

with io.StringIO() as streamio:
    df.to_csv(streamio, encoding = "utf-8", index=False)
    streamio.seek(0)
    service.create_blob_from_text('containert', 'example.txt', "".join(streamio.readlines()))