Hubspot Company Cache

Hubspot has a spotty (heh) API which has some missing pieces. One of the worse is the inability to search companies without download them all. This is discuess in https://integrate.hubspot.com/t/search-for-companies-by-name/3024 on their forum.

They do have an API for getting recently created and recently modified companies. These can be used for maintaining a cache of companies. I wrote some code to support such a cache but due to complexity I didn't end up using it. My primary use was to reduce the duplication of companies by name. But instead I decided to do that manually. This code is for people who still need to do searching locally.

In order to use this code first install it

yarn add hubspot-company-cache

This package assumes you have a hubspot object from https://github.com/brainflake/node-hubspot that you can inject into it. I will call this hspot throughout the documentation. Pass this into the HubspotCompanyCache object thus:

import HubspotCompanyCache from 'hubspot-company-cache';
const companyCache = new HubspotCompanyCache(hspot);

Early in startup you might want to start the cache loading.

companyCache.fill();

If it doesn't finishg filling before your first call then your first call will block until it is done filling the cache.

You can use this to find companies by name like this:

companyCache.findCompanyInCache('My Company');

The current cache policy is to fill when there is a miss. There is no attempt to update companies once they are fetched.

At the moment I've just saved code that I ripped out of another codebase. I won't even claim it really works at the moment.

Thus the API isn't stable in any sense of the word.

Work to be done before this is something I would use includes

1. Tests
2. An interface to iterate over the cache (search other then name)
3. Stable API
4. Remove the few bits of lodash use to reduce weight.
5. Remove dependence on https://github.com/brainflake/node-hubspot ( Note that I like this API, just a lot to drag in.)