Standardize API response across 2 variations so that Zapier treats the response fields as the same during configuration

  • 18 July 2022
  • 1 reply

  • I have a call in a private app for Zapier to retrieve the Metadata of an asset.
  • The call covers both images and videos, which are 2 different endpoints. The only different in the request being whether it says images or videos in the URL:

  • The response returns mostly the same fields, but as they are wrapped by either images {} or videos {}, Zapier treats them as different fields, which makes it very cumbersome to configure metadata mapping in subsequent steps, as you basically need to test the entire 2 zaps twice, once with an image, once with a video, so that you can map the fields correctly.

  • I’m looking at a way to modify the response structure so that both videos and images fields are considered as the same by Zapier in subsequent steps.  

  • Sample image response:

{ "images": [ "allowed_use", "artist", "artist_title", "asset_family", "call_for_image", "caption", "city", "collection_code", "collection_id", "collection_name", "color_type", "copyright", "country", "credit_line", "date_created", "date_submitted", "download_sizes", "editorial_segments", "event_ids", "graphical_style", "license_model", "max_dimensions", "orientation", "product_types", "quality_rank", "referral_destinations", "state_province", "title" ] }

  • Sample video response:

{ "videos": [ "allowed_use", "artist", "asset_family", "call_for_image", "caption", "clip_length", "collection_code", "collection_id", "collection_name", "color_type", "copyright", "date_created", "display_sizes": [ { "name": "comp" }, { "name": "preview" }, { "name": "thumb" } ], "download_sizes", "era", "event_ids", "license_model", "mastered_to", "originally_shot_on", "product_types", "quality_rank", "shot_speed", "source", "title" ] }

  • Note, I use endpoints which always return a maximum of 1 result.
  • Interestingly, Zapier only pre-pends Images to the image response fields in Zapier. Not sure why.
  • Current code:

const options = {
  url: `${bundle.inputData.asset_type}s/${bundle.inputData.asset_id}`,
  method: 'GET',
  headers: {
    'Accept': 'application/json',
    'API-Key': bundle.authData.client_id,
    'Authorization': `Bearer ${bundle.authData.access_token}`
  params: {
    'fields': 'detail_set,largest_downloads,keywords'

return z.request(options)
  .then((response) => {
    const results = response.json;

    // You can do any parsing you need for results here before returning them

    return results;


  • bundle.inputData.asset_type is what determines whether the endpoint is for images or videos.

Thanks in advance for any guidance

This post has been closed for comments. Please create a new post if you need help or have a question about this topic.

1 reply

Userlevel 6
Badge +8

Hey there!

The key part of your code is that comment “// You can do any parsing you need for results here before returning them”. That’s exactly what you need to do. If you’re only ever getting one image or video returned I think the following would work:

if (results.hasOwnProperty(‘images’)) {

    return results.images[0];

} else {

    return results.videos[0];



Do you want to try that and let us know how ti goes?