A Python library to read files in Azure Blob Storage as Pandas DataFrames

The latest version is available on PyPI:

pip install bluepandas

The library reads an environment variable containing the connection string to your blob storage account. The variable must be named AZ_<STORAGE-ACCOUNT-NAME> and set to your connection string, which is obtained from the Azure Portal by navigating to you storage account under settings/access keys as shown below

import bluepandas

df = bluepandas.read_csv("wasbs://<container-name>@<storage-account-name>.blob.core.windows.net/path/to/your.csv")

This returns a BluePandas data frame, which subclasses the pandas data frame, but allows you to write to Blob Storage using the write_csv() method.

import bluepandas

# Import the iris dataset
from sklearn import datasets

values = datasets.load_iris()['data']
columns = datasets.load_iris()['feature_names']

df = bluepandas.DataFrame(values, columns=columns)

df.to_csv("wasbs://<container-name>@<storage-account-name>.blob.core.windows.net/path/to/iris.csv", index=False)

import pandas as pd
import bluepandas

df = pd.DataFrame([(1,2,3),(4,5,6)], columns=['A','B','C'])

# BluePandas data frame
bpd_df = bluepandas.DataFrame(df.values, columns=df.columns)