Combining list to table is so challenging

[Karthik]

I have a call to BlueSky

It returns 5 fields, each as one list. I’ve tried combine list and Extract to Table. Both are not elegant in getting a tabular output. How would you approach this problem generally?

If I use combine list, I get each column stacked on top of the other and the AI does a poor job turning into into a table.

If I feed it directly into Extract Table, that also does a poor job. The easiest solution would be to join all lists into a table as opposed to a stack.

[Gumloop_Bot]

Hey @Karthik! 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 @Karthik — if I understand correctly, you’re sending this data from Bluesky to an Extract to Table node, right? I’d recommend using a Combine Text node to add labels around the data — this helps the AI better recognize each field. You can then use the Join List Items node (if needed) to smush each post into one text block instead of processing each post in a loop.

Here’s an example: https://www.gumloop.com/pipeline?workbook_id=jAzrgvQ4TufPCuqN9eDD5e

Let me know if this works better.

Also, if you’re just writing this data to a spreadsheet, you can skip the AI and directly connect the Bluesky node to a Google Sheet Writer node.

[Karthik]

Thanks Wasay!

I tried it but it doesn’t work: See here:

https://www.gumloop.com/pipeline?workbook_id=6SESJUnCct7EV36Syh7Dbc&run_id=cDfxURY5Uxrbi9SCRTYknV

And with data: https://www.gumloop.com/pipeline?workbook_id=6SESJUnCct7EV36Syh7Dbc&run_id=cDfxURY5Uxrbi9SCRTYknV

I basically want to retreive 50-500 posts, and feed them all to an AI node to them summarize interesting discussions, links, most liked posts etc. I want to feed one dataset (which I thought would be best as a table) to the AI.

In my example above, the Bluesky node works fine, But the combine text node has only one list. What is wrong here?

[Karthik]

Here is another approach. As you suggested, connecting BlueSky to Google Sheet Writer worked great. But alas, feeding that to AI failed. See here: https://www.gumloop.com/pipeline?workbook_id=imJSCC5GcGRSYwcH5Z4AqE

Adding a Google Sheet Reader after Google Sheet Writer failed because that returns one column at a time as a list, defeating the whole write process to begin with.

[Karthik]

Just figured out another bug. The last input to combine text being a number breaks the output.

[Karthik]

@Wasay-Gumloop Might be an interesting bug for your team. I made it work here with your example.

https://www.gumloop.com/pipeline?workbook_id=6SESJUnCct7EV36Syh7Dbc

But in the Combine Text node, if I add Like Counts or Repost Counts (anywhere, end, the middle etc), it breaks. I tried to put it through a text formatter to convert number to text using trim (as suggested by the gumloop AI helper) but no dice. Right now I made it work without the like count and repost count but would like it in there.

[Wasay-Gumloop]

@Karthik – Really appreciate you flagging this. Could you share the run link as well so I can view the inputs/outputs please? You can find the run link on the https://www.gumloop.com/history page or through the Previous Runs tab on the canvas.

https://www.gumloop.com/pipeline?workbook_id=6SESJUnCct7EV36Syh7Dbc

[Karthik]

Here it is: https://www.gumloop.com/pipeline?run_id=cmjSgx5MbkfxP5CHTminwX&workbook_id=6SESJUnCct7EV36Syh7Dbc

[Wasay-Gumloop]

Appreciate it! Created a ticket on our end to fix this and also sent you some credits for the inconvenience.