HoldMyTwistMake Google Drive: check files a created-time watcher missed
This guide explains how to tell if Make is able to detect an older file that was moved to a Google Drive folder. This is done by checking that the created-time watcher works compared to a basic folder search. The idea is to do this without any downstream automation yet.
With By Created Time, I was not able to catch an old file that we moved to Make's Google Drive watcher folder. But a new file uploaded to the same folder was captured. (We tested it with make free using dummy text files). You can check it out in the below exercise, to compare the watcher output with files present in your folder.
Before you start
Open a browser (on a desktop) and log into your own Google Drive account (and Make Free) - or create one if you don't have one yet: create Make account (opens in a new tab)
Try with small dummy text files this time :)
Make sure scheduling is disabled.
One module for one scenario, no email or anything else downstream.
We may receive a commission through our Make links. Google Drive is a trademark of Google LLC. This guide is not affiliated with or endorsed by Google.
In this guide, we will assume that you already have Google Drive connected to Make. If you haven't connected it yet, be sure to go through Make's instructions on how to connect your Google Drive (opens in a new tab). Please take a look at the Google permissions that you are being asked for before granting access. Please note that picking a practice folder does not limit the connection to that specific folder.
1. Set aside an older file
Reader creates intake in My drive, uploads older.txt outside of intake, then adds watcher. So we have a file existing before it's added to watched folder. Reader uploads three other files to intake. two separate files called sample.txt. and one file called third.txt (if google drive calls the second upload sample (1).txt then they rename it to sample.txt). You can open all those files to see ID in the address between /d/ and /view. Google drive allows multiple files with the same name. Both sample.txt files need to have different IDs. The IDs will be later used for comparison.
2. Check the starting files
Add Google Drive -> Watch files in a folder to a new Make scenario. Use your existing connection. Watch files -> By created time. Choose a Drive -> My Drive Folder -> Intake. File types -> All. Limit -> 2. Save module. Choose where to start -> All. Save scenario. Click Run once. Look at File IDs in bundles output from run. For us it was 2 files on our 1st run. 3rd file on our 2nd run. No new data on our 3rd run. Make sure you've seen all 3 file ids at start before moving the older.txt.
3. Move the old file, then upload a new one
Take older.txt and throw it into Intake without modifying it. Then run the watcher again. We found that it didn't generate any new data, despite the file being in the folder.
Upload new.txt into intake, run watcher again. We see our new file in the output. This proved that the watcher/connection was still capable of pulling a file that was just uploaded after the moved file run that got nothing.
Watch out
A file could appear in folder at some point after it was created somewhere else. In case the reader would like to include old moved files in intake process - IDs of the files should be checked in watcher output before attaching some downstream action to send/move/delete something.
Same By Created Time / My Drive, no new data on rerun (unchanged rerun). Content change to existing file, no new data. Remove the old file from intake and put back, no new data. These are three separate instances. (I didn't try By Modified Time nor Shared Drives).
4. Search the folder and check the limit
Make a new scenario using "Google Drive -> Search for Files/Folders". Use the same connection. Use "Enter manually". Copy your Intake Folder ID from after /folders/ in your Drive address. (Shared Drive ID leave blank if My Drive) Retrieve = Files. Search = blank Limit = 2. Do search twice without changing the folder. Record all File IDs that are found. In our case, same 2 sample.txt IDs came back both times. No other files were found in the second search.
For example:
Set Limit to 10. Then try the search again. We got exactly the 5 file ids we expected (and older.txt was among them). Check to see if your expected list of file ids match the output of your search (comparing each file id from your expected list vs. output of your search). It's only helpful when you got the same file ids if the count matches.
This is just our 5 file folder that we know about. You'll have to check for completeness on larger or other folders. We don't know that all pages of search results are checked. It is possible to get the same file in a recurring search. Keep track of all the file ids that have downstream work done. (See the duplicate record guide for data store for that other problem.) The log of downstream work done has no way to find a file that was missing in the search results.
5. Choose the next step
If you have some batch sporadically then you can manually check the folder in Drive. But for repetitive automation you need to specify which files are arrivals. Pull all files that should be arrivals. And you need to remember the past work. Our test we did not get older.txt for created time watcher. Limit 2 search only got same two sample.txt (same in both searches). Did not get older.txt with those searches. Limit 10 search got all 5 files including older.txt. Overall for our test it took 6 credits for 11 manual runs in Make Free without scheduling. We have not checked cost of scheduling, crash recovery, parallel modifications, temporary arrivals, once by id.
If even the new upload is missing
If new.txt isn't showing up in the output either, check your connection, folder, file type filter, starting point. Disconnect anything further down the pipeline while troubleshooting. If you reset Choose where to start it may reintroduce older files so consider the possibility of handling the same files twice before resetting a well established workflow. Afterwards, once you're done with this exercise, move the one use scenarios/fictional files to Trash/Bin. If you granted a temporary Make connection for Google Account Linked apps, check if other Make integrations require the same connection grant before revoking it.
Sources
Make Google Drive modules (opens in a new tab); Google Drive file fields (opens in a new tab); Google Drive file moves (opens in a new tab); Related Make Community moved-file question (opens in a new tab).
Share this guide
From holdmytwist.com/guides/make-google-drive-moved-files/
Related guides
Check blank and over-limit text in Make before routing
Make two make routes for this empty/short/over-limit text. Careful of the filters and the empty string problem. Don't even think of using real submissions until you have the two.
In this guide
Test AI feedback categories in Make before acting on them
Use the Make AI Toolkit "Categorize text" module to categorize the given fictional feedback. Review the resulting categories and explanations to see how they match up to the original text. Look for additional categories, incorrect negations, and changes in the overall categories string before proceeding.
Clean a customer CSV in Make Free before importing it
Create a make free csv cleaning activity with 5 fake rows. Clean up the start and end spaces on name, email, segment values. Drop empty email rows. Preview the new csv text. (Note this does not import actual customer records).