This script can be used to periodically pull events from an iCal feed and insert them into a selected Google Calendar using the API for that service.

Why do this instead of importing the iCal URL straight into GCal? The rate at which GCal refreshes iCal feeds is glacially slow, typically somewhere between 1-2 days. This has been the case for the best part of a decade now and Google show absolutely no interest in providing a sensible way to even trigger a manual refresh (e.g. https://productforums.google.com/forum/#!msg/calendar/iXp8fZfgU2E/wK9Qf6nfI48J). This script is a simple way to work around that limitation - it's not much use to me if I add an event to my todo list and then check my calendar the next day and forget about it because it hasn't been synced from the associated iCal feed yet.

I've been running this script on an RPi as a cronjob and it's working well for me. I'm putting the code here in case it's useful to anyone similarly frustrated with Google Calendar and its handling of iCal feeds. Note that it's not particularly polished or well-packaged, and importantly doesn't try to handle all possible types of iCal events. It only does the minimum I needed for my own workflow.

Using the script

NOTE: requires Python 3.7+

1. Copy config.py.example to a new file config.py or a custom file (see Multiple Configurations below)
2. Modify the value of ICAL_FEEDS to configure your calendars. It should contain a list with one or more entries where each entry is a dict with the following structure:

ICAL_FEEDS = [
    {
        # source of calendar events. normally this is an iCal feed URL, but you can also use a local path
        # containing .ics files as a data source instead (in that case set 'files' to True)
        'source': '<ICAL URL OR DIRECTORY PATH>',
        # the ID of the Google calendar to insert events into. this can be "primary" if you want to use the
        # default main calendar, or a '[email protected]' string for secondary calendars. You
        # can find the long calendar ID on its settings page.
        'destination': '<GOOGLE CAL ID>',
        # set to False if source is a URL, True if it's a local path
        'files': False,
    },
]

3. If your iCal feed is password protected you should also set the variables ICAL_FEED_USER and ICAL_FEED_PASS appropriately.
4. Create and activate a virtualenv and then run pip install -r requirements.txt
5. Go through the process of registering an app in the Google Calendar API dashboard in order to obtain an OAuth client ID. This process is described at https://developers.google.com/google-apps/calendar/quickstart/python. It's important to select "Desktop app" for the OAuth "Application Type" field. Once the credentials are created, download the JSON file, rename it to ical_to_gcal_sync_client_secret.json and place it in the same location as the script.
6. Until recently you could leave your Google Cloud project in "testing" mode and the OAuth flow would work indefinitely. However as described here any tokens for apps in this mode will now expire after 7 days, including refresh tokens. To avoid having to manually re-auth every time this happens, go to your OAuth consent page configuration and set the "Publishing status" to "Production". This will display a warning that you need to do a lot of verification steps, but things still seem to work if you ignore the warnings.
7. Run the script. This should trigger the OAuth2 authentication process and prompt you to allow the app you created in step 5 to access your calendars. If successful it should store the credentials in ical_to_gcal_sync_credentials.json.
8. Subsequent runs of the script should not require any further interaction unless the credentials are invalidated/changed.

OAuth workarounds

If you're running the script on a headless device you may have some issues with step 7 above. It normally attempts to auto-open a browser to allow you to authorise the request, and will print a URL to visit instead if it can't find a browser/display. You can visit this URL from another device, but the final step in the auth flow is an HTTP request sent back to the server started by auth.py. This will fail when you're using another device because the URL it redirects your browser to will be http://localhost:port/.... To workaround this there are a few options:

• clone the repo to another machine, run the auth flow there, and then copy the saved credentials file to the headless device
• copy the http://localhost:port/... URL and then in an SSH session on the headless device run curl <URL>
• alternatively if your headless device has open ports, you can modify the run_local_server() line in auth.py to have the Google redirect point to a hostname other than localhost. See the host and port parameters in the documentation

Multiple Configurations / Alternate Config Location

If you want to specify an alternate location for the config.py file, use the environment variable CONFIG_PATH:

CONFIG_PATH='/path/to/my-custom-config.py' python ical_to_gcal_sync.py

Rewriting Events / Skipping Events

If you specify a function in the config file called EVENT_PREPROESSOR, you can use that function to rewrite or even skip events from being synced to the Google Calendar.

Some example rewrite rules:

import icalevents
def EVENT_PREPROCESSOR(ev: icalevents.icalparser.Event) -> bool:
    from datetime import timedelta

    # Skip Bob's out of office messages
    if ev.summary == "Bob OOO":
        return False

    # Skip gaming events when we're playing Monopoly
    if ev.summary == "Gaming" and "Monopoly" in ev.description:
        return False

    # convert fire drill events to all-day events
    if ev.summary == "Fire Drill":
        ev.all_day = True
        ev.start = ev.start.replace(hour=0, minute=0, second=0)
        ev.end = ev.start + timedelta(days=1)

    # include all other entries
    return True