Datahub Importer - HTTP - Data Selector for JSON Payload

[cancan101]

Is there any way to control the "JSONPath" of the data payload to be importer when using an HTTP Data Source? For example the endpoint returns the data like this:

{
  "data": {...},
  "pagination": {...}
}

and I want to tell Pimcore that the relevant data can be found under the data key. Other REST / Json importers generally offer a concept of an extraction "path" that specifies where the data can be found. (something like this: https://jmespath.org/).

[fashxp]

Nope, see https://pimcore.com/docs/platform/Data_Importer/Configuration/File_Formats#json

[cancan101]

The adapter expects an array of json objects and reads all first level attributes as separate fields.

Seems to be the limitation I am running into. That prevents me from using the transform pipeline to pull out the fields that I need.

Would Pimcore be open to a PR adding a path extractor for JSON? Similar fields already exist for XML (XPath) and to a lesser extent CSV / Excel ("Skip First Row").

[fashxp]

yes, definitely!

[cancan101]

PR: pimcore/data-importer#437