Pinterest to Dropbox

[mateitudor]

Hello,
I have made a workflow that works perfectly up until one point. The issue is clearly me. My talent ran out.

Basically I have made a JSON that has two properties: an url property that contains the link to a Pinterest image, and a filename property (parsed using js).

Inspired by this thread: Pinterest Node

How can I transform the JSON in a way that it downloads the images from the url property in a Dropbox folder?

I have tried the http request component but it doesn’t have a Pinterest auth type.
I have tried converting the url to a binary file but I am not smart enough.

Please help.
Thank you in advance.

[MutedJam]

Hi @mateitudor, welcome to the community

I am not sure that’s a question of talent, many websites make it intentionally hard to extract content from them (to get real visitors instead of bots simply downloading a file). Unfortunately, you didn’t provide example data (and I am not a Pinterest user) so I am not 100% sure how your URLs look like.

However, on a general level I think downloading multiple files and then passing them on to Dropbox could work like so (assuming you are running the most recent version of n8n):

[mateitudor]

Hey @MutedJam, thank you so much for being so kind!

First, I am reading the RSS of a Pinterest board:

720 Typotheque ideas in 2023 | typography design, typography, typography letters

Then I am iterating through them, replacing the low res url with the full res one:

The urls are valid, check this one for example:

After this step I tried using the http request with GET but I am met with an authentication error.

[MutedJam]

I am afraid I don’t get any error when trying to download the example URL you have provided using n8n. Any chance you can provide an example workflow using which your problem can be reproduced?

[mateitudor]

Sure, here you go:

[MutedJam]

Thank you! I could see the problem and it seems that one of the URLs you are reading is not publicly accessible. Opening https://i.pinimg.com/originals/8d/ca/63/8dca6337923e829f5ad6c22d5988de97.jpg will also result in an error outside of n8n:

You can account for this by having the HTTP Request node ignore errors and storing the response status code, then checking this in the next step.

For example like so:

This will split the downloaded files in a “good” (24 items) and “bad” (1 item) branch:

Hope this helps!

PS: I have created this workflow using a recent version of n8n. You didn’t specify which version you are running, so you might need to upgrade your n8n instance before you can import my example workflow.