Skip to main content
Question

Integracion de automatizar

  • January 7, 2026
  • 2 replies
  • 41 views

Forum|alt.badge.img

I’m working with a company that manages products, materials, and services across multiple Airtable tables.

I want to use the Airtable API with Python to perform data integration / migration between two tables:

  • A source table containing general product records

  • A destination table where records must be created based on specific selected fields

One of the main challenges is that I cannot rely on fixed column names, and I need a way to define which fields should be considered by the script when creating new records.

I’ve read that working with JSON is the recommended approach by Airtable, but I’m unsure about the best practice to:

  • Select only specific fields from a table

  • Map them to the destination schema

  • Maintain a relationship with a linked table (“categories”)

My question is:
What is the recommended way to automate field selection and mapping when creating new records via the Airtable API, especially when schemas are not fully standardized?

Any guidance, examples, or design patterns would be appreciated.

2 replies

ibayub
Forum|alt.badge.img+10
  • Participating Frequently
  • January 8, 2026

i dont have much experience with this but to start you could use the base scheme metadata endpoint to get the schema of all the tables in the base so you an list available fields, validate or auto-gen mappings to destination table, and identify linked fields like categories. then instead of hardcoding mappings, you could use an external config file to define source fieds, how they may to destination fields (if they follow a pattern could also use regex or fuzzy matching). pyairtable is also a popular wrapper if you’re not using it already 


TheTimeSavingCo
Forum|alt.badge.img+31

Hmm, it sounds like you’re trying to:

  1. Grab a specific list of fields whose names can change
  2. For each of these fields, put its data into a specific field in a destination table that doesn’t change
  3. Both the source and destination table have a linked field to a table called ‘Categories’, which you also want to link up
  4. This is all within the same base

It seems like points 1 and 2 would be solved by grabbing the field IDs of the fields you want and using those in your script to create the records in the destination table?  If you’re saying that the actual fields that you want (and thus the field IDs) can change, i.e. someone deletes a field (Field A), creates a new one (Field B) and you now want the data from Field B to work with the script, then that won’t work though

For point 3, may I know what issues you’re facing with that?  If you just reuse the data you’re grabbing from the original record for the record creation that would link up fine

 

I feel like I’m missing something though, apologies if I’ve misunderstood!