Cron + Marketing Support Solutions (MSS) / SPExpress (SPE) + Stitch Labs + Google Drive
This is a piece of software I made to automate some of the more manual tasks that take place behind the scenes at the Holstee office.
At Holstee we use Stitch Labs for inventory management, we use it as our middleman between our fulfillment center(s) and Shopify. One of our fulfillment centers MSS (recently acquired by SPE) doesn't talk to Stitch, so I built this software to bridge the gap. This is a script that runs on a daily basis. The script retrieves the daily stock, quantity or inventory numbers data from the SPE FTP server and pulling data from Stitch matching variants via SKU and updates the stock or quantity in Stitch with the updated one from MSS.
There are two modules created specifically for this project. One is the for the SPExpress API and the other for StitchLabs.
npm i spexpress stitchlabs --save
- Paginate over all products in Stitch
- Connect to the SPE FTP server and pull the latest inventory document
- Map over all the products in SPE and check Stitch
- Build a data report of all the products that are filtered out
- Return all variants that need their quantity updated
- Loop over all the variants and send quantity update requests
- Push the two reports as CSV's to Google Drive
The two defaults are CRONTAB which is set to 0 1 * * *, which is everyday at 1:00 AM America/New_York, and the timezone which is set to New York. The SPExpress FTP server places a new file every day at 9:15 PM America/Phoenix.
CRONTAB=0 1 * * *
TZ=America/New_York
STITCH_URL=
STITCH_CLIENT_ID=
STITCH_CLIENT_SECRET=
STITCH_ACCESS_TOKEN=
FTP_HOST=
FTP_USERNAME=
FTP_PASSWORD=
GD_WEB_CLIENT_ID=
GD_WEB_CLIENT_SECRET=
GD_WEB_REDIRECT_URI=
GD_USER_ACCESS_TOKEN=
GD_USER_TOKEN_TYPE=
GD_USER_REFRESH_TOKEN=
GD_USER_EXPIRY_DATE=
GD_FILE_UPLOAD_PARENT_ID=
Go to Google Developer Console make a new web app, add those keys to the .env file.
GD_WEB_CLIENT_ID=
GD_WEB_CLIENT_SECRET=
GD_WEB_REDIRECT_URI=
Start up the google web server
npm run server
Open the webpage to login to google
open http://localhost:3000/google
This will create a .token file that you can copy and paste into .env and easily push to heroku with the following.
heroku config:push
Note this isn't really a webserver, it connects to the internet but it doesn't need to be accessed on any ports or anything. Heroku has two distinctions for this, web process and worker process. Soon after deploying even having the worker prefix in the Procfile my process exited. I ran heroku ps:scale web=0 worker=1 and now it looks like it's up.