Python2转Python3：已弃用的AppUrlOpener与FancyURLopener替代方案咨询

Hey there! Let's get that User-Agent setup sorted for your XML download script—Python 3 did away with urllib.request.FancyURLopener, so we need a modern alternative that plays nice with the current urllib APIs.

Option 1: Set a Global Custom Opener (Match Your Original Workflow) #

This replicates what your old AppUrlOpener was doing—setting a default User-Agent for all urllib requests in your script. Here's how to do it:

First, keep your exact User-Agent string:

USER_AGENT = 'Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3112.113 Safari/537.36'

Then create and install a custom opener that uses this header:

import urllib.request

# Build a custom opener instance
opener = urllib.request.build_opener()
# Attach your User-Agent to the opener's default headers
opener.addheaders = [('User-agent', USER_AGENT)]
# Install this opener as the default for all urllib.request calls
urllib.request.install_opener(opener)

Now whenever you call urllib.request.urlopen() later in your code, it will automatically send the correct User-Agent header—no need to mess with the internal _urlopener variable (which was never meant to be modified directly anyway!).

Option 2: Per-Request Headers (More Flexible) #

If you only need the custom User-Agent for specific requests (not all of them), you can wrap each URL in a Request object with the headers included:

import urllib.request

USER_AGENT = 'Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3112.113 Safari/537.36'

# For each target URL, create a Request with your custom headers
url = "https://your-target-xml-url.com"
req = urllib.request.Request(url, headers={'User-Agent': USER_AGENT})

# Open the request and process the XML data
with urllib.request.urlopen(req) as response:
    xml_content = response.read()
    # Proceed with your XML conversion logic here

This approach is great if you might need different headers for different requests later on.

Why Your Original Attempt Didn't Work #

urlopen() is a function that makes a single request—it can't be assigned to _urlopener like your old AppUrlOpener instance. The modern Python 3 way is either using install_opener for global defaults or Request objects for per-call customization.

内容的提问来源于stack exchange，提问作者Ian Waller