Falling Back to Cache if 404

Is there a recognised pattern or method for using the cache if the live page 404s or there is no network connection?

For example, one way might be to try to get the page without the cache, and if that fails, then try to use the cache:

try:
    with requests_cache.disabled():
        r=requests.get('http://httpbin.org/ip')
        if r.status_code!= requests.codes.ok:
            raise ValueError()
        #If this is good, add it to the cache somehow?
        #...
except:
    r=requests.get('http://httpbin.org/ip')

print(getattr(r, 'from_cache', False),r.text)

[JWCook]

Fyi, changes for the first option I mentioned (manually saving a response) are now in the master branch and in pre-release build 0.7.0.dev255. You can use CachedSession.cache.save_response(response) to manually save a response object.

The second option (adding a raise_for_status option) will be done in issue #210. I’ll close this issue for now, then, unless there’s a different way you would prefer to handle 404s.

[yakutovicha]

In case someone else has similar issue as mine, here is the solution I came up with:

import requests
import requests_cache
from IPython.lib import backgroundjobs as bg

def update_cache():
    """Run this process asynchronously"""
    requests_cache.install_cache(cache_name='apps_meta', backend='sqlite', expire_after=1, old_data_on_error=True) # Make cache expired
    requests.get(registry_url) # Update cache
    requests_cache.install_cache(cache_name='apps_meta', backend='sqlite') # Return to initial setup


requests_cache.install_cache(cache_name='apps_meta', backend='sqlite') # at start getting data from cache
update_cache_background = bg.BackgroundJobFunc(update_cache) # create cache update job
update_cache_background.start() # start cache update in the background