Can this even be done? Can't map open ai output into airtable

The goal is to map output from openai into airtable fields. I have fields called prompt1-30 in airtable.

I am using the open ai module to create 30 prompts. I want to map each prompt to its corresponding field in airtable. This is the information received from my webhook

The iterator then successfully outputs it all into seperate bundles:

I then have this output from open ai:

The issue is that no matter what I do with text parsers, iterators, variables, nothing seems to get this data into a structure where i can simply map prompt1, prompt2 etc into corresponding airtable fields. Completely baffled :confused:


Welcome to the Make community!

How are you then mapping it into the Airtable module?

You might need an aggregator, or not.

Every result (item/record) from a search/match module will output a bundle. To “combine” them into a single structure, you’ll need to use an aggregator of some sort.

Aggregators are modules that accumulate multiple bundles into one single bundle. An example of a commonly-used aggregator module is the Array aggregator module. The next popular aggregator is the Text Aggregator which is very flexible and has applies to many use-cases.

2 Likes

Thanks for getting back to me. Really tearing my hair out with this. I just can’t seem to get the data into Bundle 1, Bundle 2, Bundle 3 etc, no matter what i do each prompt continues to be named Bundle 1, Bundle 1 meaning that I can’t map anything. I have even (successfully) tried to get each prompt into a datastore but i can’t even pull this data out! :rage:

Everything is named “bundle 1” because they are in different operations. You’ll just need to aggregate the results into a single array like I mentioned.

2 Likes

I tried that and it just gave an empty array. It was taking too much time and after days gave up. I think Make could probably up their online tutorial/docs game. In the end I passed the payload of 30 values into lambda, split them out and pushed them to different scenarios with filters on ‘promptX’ exists. Clearly a massive hack but i just needed this to work asap.