Back to Community
B

Bella Z.

@build_bella ·

Image Parsing Issue with Evolution API

Hey WebNutch community, I'm reaching out for help with a challenge I'm facing while trying to parse images and PDFs through the Evolution API in my workflow. Previously, I used a different API and it handled image analysis smoothly, but since switching to Evolution, I've been struggling to get even a single image to flow through my workflow. The issue seems to be with the MIME type, as Evolution sends images in application/octet-stream, but the next node in my workflow, Gemini, only accepts JPG files. Has anyone else encountered a similar problem or have a solution that could help me out? I'd really appreciate any guidance or advice from someone who has overcome this hurdle. You can find similar workflows and solutions in the WebNutch marketplace, but I'm hoping someone in the community can share their experience with this specific issue.

+10
5 comments

Add a comment

M
module_maria2h ago

I'm having the same problem with PDFs. Did you find a solution?

P
proxy_priya2h ago

Great tip to check the MIME type! 🤔 I had a similar issue with Gemini and had to add a middleware node to convert the MIME type. Have you tried that?

K
kite_kira1h ago

I had a similar issue with the Evolution API and I ended up using a separate node to handle the image conversion. I used a node that converts the image to a base64 encoded string, then passed that to Gemini. It was a bit of a workaround, but it got the job done. Has anyone else tried this approach?

A
api_ace_andy1h ago

You can use the 'Set HTTP Header' node in WebNutch to change the MIME type of the image before it reaches Gemini. I did this and it worked like a charm! No need for middleware.

B
byte_brent1h ago

Thanks for the suggestion, but I've already tried setting the HTTP header and it didn't work for me. @OP, can you share more about your workflow and how you're handling the image parsing?