Futures

It would be great to have the ability to perform automatic backups of tasks / scripts. The Google Drive integration is there, so it would be possible to build off of that.

For example, when a Merchant sends a collection to a task from an admin action link, it would sometimes be helpful to give them form options.

We ran into an API that requires multipart POST requests having a part with headers like these: Content-Disposition: form-data; name="file"; filename="file.csv" Content-Type: text/csv The "files" option in Mechanic's HTTP action uses hash keys as both name and filename for each file. There's no way direct way to use file generators and end up with distinct name vs filename values. This future is about a direct path forward there. In the meantime, something like this should work, as an indirect path: {% assign boundary = "MechanicMultipartBoundary" %} {% capture multipart_body %} --{{ boundary }} Content-Disposition: form-data; name="hello" world --{{ boundary }} Content-Disposition: form-data; name="file"; filename="robots.csv" Content-Type: text/csv Sku Code,MRP,Cost Price SKU1,0.00,0.00 SKU2,0.00,0.00 --{{ boundary }}-- {% endcapture %} {% action "http" %} { "method": "post", "url": "https://webhook.site/your-webhook-url", "headers": { "Content-Type": "multipart/form-data; boundary={{ boundary }}" }, "body": {{ multipart_body | json }} } {% endaction %}

Current Google Sheets implementation only supports 1 sheet per file. Looking to be able to create/modify a single file with multiple sheets.

Store user data in private git repo

Currently Mechanic reacts to all type of actions as long as you subscribe for an action it does react to it, sometimes though, you might not need a response for each and every one, a parameter in action tag to disable the task from reacting to that specific action would be nice.

Assuming this is possible of course. I don't believe these would be app-specific webhooks.

Shopify (currently) only allows for a single bulk op to be running per app, per store. Would be great if Mechanic could automatically handle a few retries.

Because I'm lazy, it would be awesome if I didn't have to manage GraphQL cursors and pagination when querying Shopify objects. This feels like something Mechanic could do automatically for us on the backend. Could Mechanic autodetect edge references and handle pagination for any references that are paired with pageInfo ? Or maybe auto-paginate any edge reference that is unbounded? I imagine this is something that would also find it's way into Liquid GraphQL objects if that ever becomes a thing.