Help

This Product Ideas board is currently undergoing updates, but please continue to submit your ideas.

Search in attached files

cancel
Showing results for 
Search instead for 
Did you mean: 
raoulwegat
6 - Interface Innovator
6 - Interface Innovator

TXT, PDF, DOC, ALS, etc. This would actually be a huge feature.

4 Comments
Bill_French
17 - Neptune
17 - Neptune

I sort’a did this -

Moe
10 - Mercury
10 - Mercury

We built this extension that allows you to convert PDF to TXT on Airtable, so that attachments can be searched or filtered.

Extract Text From PDF Attachments on Airtable

Michael_Tchong
5 - Automation Enthusiast
5 - Automation Enthusiast

Why is this still not possible? I want to create a content database in Airtable but would like to attach text files instead of laboriously pasting the text in a field, surprised Airtable can’t do that! It supports regex searches and what not, but can’t search attachments, DOH! :smirk:

Bill_French
17 - Neptune
17 - Neptune

Many reasons.

You make it sound like these are the only two pathways to successful searching in the current Airtable feature set. It can do that and hundreds of users are doing that - where “that” is using Airtable as a comprehensive text and content management backend. Have you thought about automating text updates and management?

You are conflating two very distinct aspects of findability; filtering and indexing.

While Airtable is pretty good about filtering and using a variety of formula functions to do so including the latest RegEx features, formulas work on fields, not full text or inverted indices and certainly not attachments. Since Airtable does not support full-text indexing, we are all presently confined to one dimension of a two-dimensional concept. unless – we do something more like this.

Indexing Attachments

This is yet another entirely complex layer of full-text indexing that needs to designed either by Airtable or third-party integrators. I have built search systems for clients and it requires a number of services and skills that are typically well outside the realm of no-code platforms. For example, indexing PDF and other binary file format documents require significant server time that needs to constantly craw the data set. Even text attachments must use an inverted index to remove stop-words and optimize discovery for boolean queries.