Aug 17, 2021 05:06 PM
I use an Automation to classify records that are generated from incoming emails, using string matching. The script is triggered whenever a new record enters an Airtable view.
Over time, these incoming emails change, so I frequently need to make adjustments and tweaks to the classifier script as new email types are encountered that were not successfully classified in the first pass of the script.
This creates a problem where by the time I’ve updated the script to accommodate some newly discovered content in a new email type, the system may have already received many tens of emails of this new type, and these will be unclassified, so I need to manually classify them.
I want to find the best way to re-run the classification script over these records already in the view - ie so that every time I make a change and save an updated version of the classifier script, the script would then re-run over the records in the view and the second pass would be able to successfully classify any emails of this new type that already present.
What’s the best way to achieve this?
Aug 17, 2021 05:23 PM
In short, you need to find some way to make the unclassified records leave the view, then re-enter it again. The easiest way that I know of is to delete the records, then undo to bring them back, though that might have side effects in some rare circumstances. A safer option is to temporarily add a filter condition to the view that these records will never meet, causing them to leave the view. Remove the added condition and the records will re-enter the view and trigger the automation.
Aug 18, 2021 01:20 PM
I always add a temporary update flag (checkbox field) and either change the filter for a view or the conditions for the automation.
Sep 04, 2021 12:20 AM
Thanks for suggestions @Justin_Barrett and @Tuur , I’m going to try using the new scheduled trigger instead which I think (hope!) should deliver the outcome I’m looking for …