Azure 功能 - Python（Blob 触发器和绑定）

Question

I have reviewed the documentation provided by Microsoft on Triggers. [https://docs.microsoft.com/en-us/azure/azure-functions/functions-bindings-storage-blob-trigger?tabs=python][1]

Indeed, using the func.InputStream parameter in the Azure Function allows us to retrieve the blob and some properties ( name, uri, length ) we can also read the bytes using the read() function, but how do we transform the bytes into an object we can manipulate such as a Pandas dataframe (or any other types of object for other types of files ie jpg)?

My host.json file can be found below:

{
  "scriptFile": "__init__.py",
  "bindings": [
    {
      "name": "myblob",
      "type": "blobTrigger",
      "direction": "in",
      "path": "statscan/raw/ncdb/{name}",
      "connection": ""
    },
    {
      "type": "blob",
      "direction": "out",
      "name": "outputBlob",
      "path": "statscan/enriched/func/{name}.csv",
      "connection": ""
    }
  ]
}

The Blob Trigger function can be found here below:

import pandas as pd
import logging
import azure.functions as func

def main(myblob: func.InputStream, outputBlob: func.Out[str]):

    logging.info(f"Blob trigger executed!")
    logging.info(f"Blob Name: {myblob.name} ({myblob.length}) bytes")
    logging.info(f"Full Blob URI: {myblob.uri}")

    ### Manipulate with Pandas ###

    ### Output ###
    output = ''
    outputBlob.set(output)

Answer 1

We have multiple ways to check the file content and read it accordingly, in your case lets consider as csv format as blob.

• To achieve this we can download the blob and then read the data to dataframe, below is the way I tried this from MS Docs( https://docs.microsoft.com/en-us/azure/architecture/data-science-process/explore-data-blob ):

   from azure.storage.blob import BlobServiceClient import pandas as pd STORAGEACCOUNTURL= <storage_account_url> STORAGEACCOUNTKEY= <storage_account_key> LOCALFILENAME= <local_file_name> CONTAINERNAME= <container_name> BLOBNAME= <blob_name> #download from blob t1=time.time() blob_service_client_instance = BlobServiceClient(account_url=STORAGEACCOUNTURL, credential=STORAGEACCOUNTKEY) blob_client_instance = blob_service_client_instance.get_blob_client(CONTAINERNAME, BLOBNAME, snapshot=None) with open(LOCALFILENAME, "wb") as my_blob: blob_data = blob_client_instance.download_blob() blob_data.readinto(my_blob) t2=time.time() print(("It takes %s seconds to download "+BLOBNAME) % (t2 - t1))

• Else we can convert is as below directly with blob sas url:

   from io import StringIO blobstring = blob_service.get_blob_to_text(CONTAINERNAME,BLOBNAME).content df = pd.read_csv(StringIO(blobstring))

• The other way is with blob sas url, we can get it after right clicking on the blob and opting for "Generate SAS":

   import pandas as pd data = pd.read_csv('blob_sas_url')