jq Helper – transform JSON with jq
Pricing
from $0.01 / successful conversion
Run jq programs over inline JSON or a linked Apify dataset. Built for chaining into data-processing and enrichment workflows.
The jq program to run (standard jq 1.8 syntax). Compiled once and applied per the apply mode below. A filter may emit zero, one, or many values per input — each becomes one output item; non-object values are wrapped under the scalar wrap key. IMPORTANT — do NOT prefix with .[] in per-item mode: each item is already a single object, so .websiteUrl is correct and .[].websiteUrl errors with 'Cannot index string with string'. Use a leading .[] only in whole-input mode, where the whole array is fed in at once.
How the filter maps over the input. Per-item (default): the array is iterated for you and the filter runs on each element as a single object — write .field, NOT .[].field. Best for mapping/filtering/enriching rows. Whole-input: the entire array is fed to the filter as one value, so use .[] to reach elements. Best for group_by, reduce, sort, unique, and other cross-record aggregations.
Raw JSON (object, array, or scalar) or JSONL (one JSON value per line) to transform. Provide either this or a Dataset ID, not both. In per-item mode, a top-level array is iterated element by element; a single object/scalar is treated as one item. Runs on inline input are FREE — prototype your jq snippets here at no cost.
ID of an Apify dataset to read items from (e.g. the default dataset of an upstream actor run). Provide either this or inline JSON, not both. This is the field to map when wiring this actor into an integration/workflow. Processing a dataset is a paid conversion ($0.01 per successful run, regardless of size).
Dataset items must be JSON objects. When the filter emits a non-object value (string, number, array, boolean, null), it is wrapped as { : value }.
