How do you handle redirects in Mechanize?

When using Mechanize, a library in Python for programmatic web browsing, handling redirects is typically straightforward because Mechanize automatically handles HTTP redirects (like 301 and 302 responses) by default. However, there are times when you might want to customize or monitor how redirects are being handled.

Here's how you can work with redirects in Mechanize:

Detecting Redirects

You can detect when a redirect happens by checking the response code or the URL before and after a request. Here's a basic example:

import mechanize

# Create a Browser instance
br = mechanize.Browser()

# Open a URL that you expect to redirect
response = br.open('http://example.com/some-redirect-url')

# Check if the final response is a result of a redirect
if response.code in (301, 302) or response.geturl() != 'http://example.com/some-redirect-url':
    print('We were redirected!')
    print('Final URL after redirects:', response.geturl())

Customizing Redirect Behavior

If you need to customize the behavior, such as to limit the number of redirects, you can subclass HTTPRedirectHandler and then add your custom handler to your Browser instance.

Here's an example of how you could limit the number of redirects:

import mechanize
from mechanize._http import HTTPRedirectHandler

class LimitRedirectHandler(HTTPRedirectHandler):
    def http_error_302(self, req, fp, code, msg, hdrs):
        if hasattr(self, 'redirect_count'):
            self.redirect_count += 1
        else:
            self.redirect_count = 1

        # Set a redirect limit (e.g., 3)
        if self.redirect_count > 3:
            raise mechanize.HTTPError(req.get_full_url(), code, 
                                      "Redirect limit reached", hdrs, fp)

        # Call the parent class method to actually perform the redirect
        return HTTPRedirectHandler.http_error_302(self, req, fp, code, msg, hdrs)

    http_error_301 = http_error_303 = http_error_307 = http_error_302

# Create a Browser instance
br = mechanize.Browser()

# Add our custom redirect handler
br.add_handler(LimitRedirectHandler())

# Now when you open a URL, it will raise an HTTPError if it redirects more than 3 times
try:
    response = br.open('http://example.com/some-redirecting-url')
except mechanize.HTTPError as e:
    print('Redirect limit reached:', e)

In this example, we're overriding the http_error_302 method (as well as other redirect-related methods) to count the number of redirects and raise an HTTPError if a certain limit is exceeded.

Disabling Redirects

If for some reason you want to disable following redirects altogether, you can do so by removing the redirect handlers from your Browser instance:

import mechanize

# Create a Browser instance
br = mechanize.Browser()

# Remove the HTTPRedirectHandler to disable following redirects
br.set_handle_redirect(False)

# Now when you open a URL that redirects, it will not follow the redirect
response = br.open('http://example.com/some-redirect-url')
print('Response code:', response.code)  # Expected to be a 3xx code
print('Content at the redirect URL:', response.read())  # Will be the redirect response, not the final destination

By setting set_handle_redirect to False, you disable Mechanize's automatic redirection following. This means that if a URL returns a redirect response, Mechanize will not attempt to follow the redirect; instead, it will return the redirect response to you directly.

These examples should help you handle redirects using Mechanize in a way that fits your specific needs. Mechanize offers a lot of flexibility for managing HTTP interactions, including redirects.