Metadata API for schema and mutating tables

[Simon_Horup_Es1]

Is there any plan for an API to pull metadata? It’s currently very difficult to infer the schema, you have to resort to hacks like reading all records and then trying to infer types and column names from that. It’d be great if you could pull the schema information for a table more easily!

It’d also be awesome if you’d be able to add and change tables from the API, but that’s less pressing to me :slightly_smiling_face:

[Hugh_Campbell]

I feel your hurt Tuur - between

• the lack of a metadata API
• the inability to list all bases accessible with a given API key (hmm Zapier can do it, why can’t we? What endpoint have they been granted access to that we haven’t?)
• the inconsistent way that fields are returned (empty values are sometimes just missing, other times empty arrays)

developers will struggle to write useful, reusable tools. The silence on this front is frustrating. If a small % of the effort going into front end tools like Blocks was put into these small enhancements, the sky is the limit.

[Tuur]

I’m afraid the silence (and removing of old references) is telling…

They probably have other priorities.

[Mike_Preuss]

I’d love to see a Metadata API as well…

[Matt_Scheurich]

Why would it take so long? They use something automated to generate the table’s API documentation (I doubt there’s a poor intern building each table model for each table’s API docs…).

I’m sure all they’d have to do is expose it as an API endpoint that we can read.

[Zim_One]

@Simon_Horup_Eskildse any chance we can do a similar schema hack for “view” meta data besides columns?

[Martin_Edwards]

This is a blocker for me. Oh well, back to Google Sheets.

[R_D_Connect_MF]

Hello, is there any official update on this? The blocks seems to be using a private API endpoint to fetch the data I want.

It would be cool if we could get access to the same API even if it’s released in Beta. At least we can build some proof-of-concept’s until a mature API is ready

[Ryan1]

+1 for making metadata a first class citizen

[Bill_French]

+1 and then some…

Schema…

If Airtable wants to see intelligent, non-brittle integrations emerge around their product, they will expand API access as soon as is practical. Forcing the developer community to script hard-coded solutions doesn’t allow us to create sustainable outcomes. This is bad for business.

Blocks…

It’s also not wise (in my opinion) for Airtable to create classes of API access that are denied to the general user community. It tends to stifle innovation. Ideally, we should all be able to define and register custom blocks using open web standards.

[Matt_Irvin]

+1 for this feature, the ability to create tables and modify table schemas would be enormously helpful in our use case. I’d be happy to pay extra for this too.