2700 pdf's via Google Doc via Analyze image

[Donna]

Hey,

Need some brain juice

I have a folder with 2700 pdf’s. In each of them are several photos’ from an item. An AI needs to write 5 lines of text which will be used on a Shopify website.

How do I set this up?

If I try the run based on 3 pdf’s … it does not work …

@Wasay-Gumloop what do you think?

D

[Gumloop_Bot]

Hey @Donna! If you’re reporting an issue with a flow or an error in a run, please include the run link and make sure it’s shareable so we can take a look.

1. Find your run link on the history page. Format: https://www.gumloop.com/pipeline?run_id={{your_run_id}}&workbook_id={{workbook_id}}

2. Make it shareable by clicking “Share” → ‘Anyone with the link can view’ in the top-left corner of the flow screen.
   

3. Provide details about the issue—more context helps us troubleshoot faster.

You can find your run history here: https://www.gumloop.com/history

[Wasay-Gumloop]

Hey @Donna – Can you share the run link where it’s not working please so I can view the inputs/outputs. From your screenshot I can see that the Analyze Image node has an empty input field.

In general you can set it up like this: https://www.gumloop.com/pipeline?workbook_id=4LCjtFG7qbsBE1PxRSGs39

Reason why I setup a subflow: https://docs.gumloop.com/core-concepts/subflows#pro-tip%3A-build-flows-backwards-using-subflows

Let me know if this makes sense and works for you

[Donna]

Hey @Wasay-Gumloop, thanks for the tip, it works.

So I now have this folder with 2600 files.

I think I can not process them in one hit, so I included a list trimmer.

Initiatly I have put it at 50 files per run.

But when I click on RUN, it takes for ages to “read the GDrive folder”

I think it is first gong to read 2600 files, before selecting the first 50 (to process)

Is this the case?

And if so, how should I attack this problem then.

be well,

Bart

[Wasay-Gumloop]

No problem! Would it be possible to separate these files in smaller folders? I think that’d be the best solution. With the List Trimmer the Google Drive Folder Reader node would always run first and read all 2.7k files.

Another option is to export all the drive links to a spreadsheet and read 50 rows at a time from that and use the Use Link option on the Drive File Reader node.

[Donna]

The second option … can you clarify?
How would I “export all the drive links” (of the 2700 files?) to a spreasheet?
Thanks Bart

[Wasay-Gumloop]

Ah sorry, should’ve clarified — this would be outside of Gumloop, done via an app script. I’m assuming it’s a one-time large batch run, which is why I made the suggestion. If that’s not feasible, then using separate folders is probably your best bet if you’re looking for efficiency in the flow without having to read all files each time.

[Donna]

Got it. Will try the script and come back to you.

[Donna]

Hey, tried this :

But does noet work …

Can you have a look @Wasay-Gumloop

[Wasay-Gumloop]

You’re passing the spreadsheet link here instead of the drive link. You should just enable Use Link manually in the node (instead of populating it dynamically) and then feed the Drive Link from your spreadsheet in the Link input not the Use Link input. The Use Link option is a parameter on the node, the toggle that you see here:

[Donna]

@Wasay-Gumloop thanks

Looking at the forum I think it will be a good idea to write one dedicated blogpost about all possibilities and potetial issues about Google Drive Folder reader, File reader, sheets, links, file creation etc etc With use cases
Instead of you having to reply to all different flavours of questions (because it is still a bit confusing)

What do you think?

[Wasay-Gumloop]

That makes sense, I could probably create a doc page for that. Thanks for the suggestion!