Samyak2 commented on issue #7715: URL: https://github.com/apache/arrow-rs/issues/7715#issuecomment-3058494490
> 1. What should the "path" argument be? A String? A JSON path? Some structured thing (Vec)`? I don't know if this is the right issue to discuss this, but I have some thoughts. I don't think it should be a JSON path representation. `variant_get` in databricks only supports a subset of JSON path. Specifically only field accesses (`$.fieldName` or `$['fieldName']`) and array indexing (`$[0]`) are supported. Here's the [reference](https://docs.databricks.com/aws/en/sql/language-manual/sql-ref-json-path-expression) (look for variant) for this. We can use an enum specifically for variant paths that supports only these two accesses. Then it would be a simple vec of paths like @adriangb mentioned above (`Vec<VariantPathSegment>`). > 2\. Should we also provide a "requested data type" field? Similar to the data bricks function IMO, yes. If we don't specify a type, how do we infer what type to extract the value as? In a variant, the type could change for every row. Databricks chooses not to do any type inference and explicitly requires the user to specify a type for any variant access. When the type is not specified, like in `variant_col:someField`, databricks returns another variant instead of trying to infer a type. This behavior makes sense to me. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: github-unsubscr...@arrow.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
