I’m using n8n on a Linux VM using dockers and using a windows machine browser to actually use it. I created a workflow in which I pick a pdf from google drive and extract data from it which is later to be used by AI models. My workflow’s Google Drive node successfully takes the pdf from Google Drive and sends it to the “Extract from pdf” node’s input. However, upon extraction, it says that the pdf has no binary data.
My drive node has “data” set in output binary field. Google Drive’s API is enabled and the file is properly shared with my drive’s service account. The pdf has text and no image. I even checked an online pdf binary extractor and it did extract data from it. I checked & applied the settings on relevant posts on this forum and applied the solutions as well which AI suggested but no luck.
Hope all is going well, it would appear you just need to run the node before it, it should be the node directly before the extract, once you ensure the download node runs, it should work on the extract node correctly.
update: you can share the workflow too, this will help I can test locally too.
Thats awesome and good to hear, just remember everything connects
For workflows created in version 1.0 and above: executes each branch in turn, completing one branch before starting another. n8n orders the branches based on their position on the canvas, from topmost to bottommost. If two branches are at the same height, the leftmost branch executes first.*
I’d be really greatful if you could mark as solution if all okay, feel free to reach out to me or tag me if you have any further questions.
