Duplicate Output Files Where There Shouldn't Be

I have one set of 12 rules watching a subfolder of "Downloads called "FileThisFetch." The general idea behind these rules is as follows: 1. The watched folder is linked to FileItFetch, which periodically deposits into it PDF invoices and statements. Ten of the rules are set up to "read" the invoice files and determine which vendor's bill has been placed in the folder. When it detects a valid bill, the relevant rule: a. renames the file using the contents of that file (such as Vendor Name, Bill Amount, Date of Bill), and replaces any existing tags with new standard ones; b. an applescript (pulled from forums) uploads the newly named and tagged file to Evernote; and c. the original file is trashed. (Minor detail: some documents fail all tests because they are either not bills, or they have not been programmed yet. They remain in the watch folder for now.)

The 10 watching rules are ordered third through twelfth in Hazel Rule order. Rule #1 deletes empty folders (FileThis delivers multiple documents from one vendor in a folder). And Rule #2 is the so-called Subfolder Rule you posted to this forum, copied exactly as given.

Each rule was independently debugged and seemed to be working, so I activated all twelve and set them loose for a couple of weeks.

When I came back to check on the Evernote folder that contains the fixed files, I found multiple duplicates had been created for many of the files that Fetch had sent, even though the files should only have been detected and fixed just once before trashing the original.

I have the logs, but regretfully I don't know what I'm looking for or at. I know this is a breezy outline of the problem, but perhaps I'm doing something glaringly wrong that a trained eye will catch. My only debugging guess is that the rules are firing multiple times before the file trashing takes place. Any ideas appreciated. John

Email support with the logs and a copy of your rules. Click the "Contact Support" button in Hazel so you can get a Mail window.