Python Azure Blob触发函数文件名打印异常问题排查

问题场景 #

通过Blob触发事件运行Python Azure函数，容器samples-workitems中已存在base.csv文件。当容器接收新文件new.csv时，函数通过InputStream读取同一容器内的base.csv和new.csv。

函数代码 #

import logging
import pandas as pd
from io import BytesIO
import azure.functions as func

def main(myblob: func.InputStream, base: func.InputStream):
    logging.info(f"Python blob trigger function processed blob \n"
                 f"Name: {myblob.name}\n"
                 f"Blob Size: {myblob.length} bytes")
    logging.info(f"Base file info \n"
                 f"Name: {base.name}\n"
                 f"Blob Size: {base.length} bytes")
                 
    df_base = pd.read_csv(BytesIO(base.read()))
    df_new = pd.read_csv(BytesIO(myblob.read()))
    print(df_new.head())
    print("printing base dataframe")
    print(df_base.head())

运行输出 #

Python blob trigger function processed blob 
Name: samples-workitems/new.csv
Blob Size: None bytes
Base file info 
Name: samples-workitems/new.csv
Blob Size: None bytes
first 5 rows of df_new (cannot show data here)
printing base dataframe
first 5 rows of df_base (cannot show data here)

配置文件（function.json） #

{
  "scriptFile": "__init__.py",
  "bindings": [
    {
      "name": "myblob",
      "type": "blobTrigger",
      "direction": "in",
      "path": "samples-workitems/{name}",
      "connection": "my_storage"
    },
    {
      "type": "blob",
      "name": "base",
      "path": "samples-workitems/base.csv",
      "connection": "my_storage",
      "direction": "in"
    }
  ]
}

异常现象 #

两个文件的内容能正常打印，但myblob.name和base.name均显示为samples-workitems/new.csv，不符合预期——base.name应该显示samples-workitems/base.csv。

问题原因 #

这是Azure Functions Python运行时的已知行为：当Blob触发器绑定使用了占位符（如{name}），同一函数内其他Blob输入绑定的name属性会被触发器的占位符值覆盖，导致元数据显示错误，但实际读取的Blob内容是正确的。

解决方法 #

方法1：手动指定固定文件名

由于base.csv的路径是固定的，可直接在代码中使用硬编码的文件名，无需依赖base.name：

# 修改日志输出部分
logging.info(f"Base file info \n"
             f"Name: samples-workitems/base.csv\n"
             f"Blob Size: {base.length} bytes")

方法2：从Blob URI解析正确名称

通过base.uri属性解析出真实的Blob路径：

import logging
import pandas as pd
from io import BytesIO
import azure.functions as func
from urllib.parse import urlparse

def main(myblob: func.InputStream, base: func.InputStream):
    logging.info(f"Python blob trigger function processed blob \n"
                 f"Name: {myblob.name}\n"
                 f"Blob Size: {myblob.length} bytes")
    
    # 解析base文件的正确名称
    base_uri = urlparse(base.uri)
    base_file_name = base_uri.path.lstrip('/')
    logging.info(f"Base file info \n"
                 f"Name: {base_file_name}\n"
                 f"Blob Size: {base.length} bytes")
                 
    df_base = pd.read_csv(BytesIO(base.read()))
    df_new = pd.read_csv(BytesIO(myblob.read()))
    print(df_new.head())
    print("printing base dataframe")
    print(df_base.head())

方法3：修改触发器占位符名称

将触发器绑定中的占位符{name}改为其他名称（如{blobname}），避免名称冲突：

{
  "scriptFile": "__init__.py",
  "bindings": [
    {
      "name": "myblob",
      "type": "blobTrigger",
      "direction": "in",
      "path": "samples-workitems/{blobname}",
      "connection": "my_storage"
    },
    {
      "type": "blob",
      "name": "base",
      "path": "samples-workitems/base.csv",
      "connection": "my_storage",
      "direction": "in"
    }
  ]
}

内容的提问来源于stack exchange，提问作者Shaida Muhammad