{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://api.sentinel-hub.com/schemas/ArrayParameters",
  "title": "ArrayParameters",
  "required": [
    "dtype",
    "chunks",
    "order",
    "fill_value"
  ],
  "properties": {
    "dtype": {
      "description": "Data type/encoding. Allowed values depend on the `sampleType` defined in `evalscript`:\n * `|u1`: 8-bit unsigned integer, recommended for `sampleType` UINT8 and AUTO,\n * `|i1`: 8-bit signed integer, recommended for `sampleType` INT8,\n * `<u2`,`>u2`: 16-bit unsigned integer (little and big endian, respectively), recommended for `sampleType` UINT16, allowed for UINT8 and AUTO,\n * `<i2`,`>i2`: 16-bit signed integer (little and big endian, respectively), recommended for `sampleType` INT16, allowed for UINT8, INT8 and AUTO,\n * `<f4`, `>f4`, `<f8`, `>f8`: float (little/big endian single precision, little/big endian double precision, respectively),\n   recommended for `sampleType` FLOAT32, allowed for any `sampleType`.\n\n*Recommended* values encode the chosen `sampleType` losslessly, while other allowed values encode the same values in a wider data type but do not add any more precision.\n",
      "example": "<f4",
      "type": "string",
      "externalDocs": {
        "description": "Zarr data type specification version 2, Attributes",
        "url": "https://zarr-specs.readthedocs.io/en/latest/v2/v2.0.html#data-type-encoding"
      }
    },
    "order": {
      "description": "Layout of values within each chunk of the array. Currently only \"C\" is supported, which means row-major order.",
      "type": "string",
      "default": "C"
    },
    "chunks": {
      "description": "A list of integers defining the length of each dimension of a chunk of the array, e.g. `[1, 1000, 1000]`.\n\nThe first element (time dimension chunking) must be 1.\n\nThe second and third (latidude/y and longitude/x-dimension chunking, respectively)\nmust evenly divide the batch output tile raster size.\nFor example, when using the [LAEA 100km grid](https://docs.planet.com/develop/apis/batch-processing/#1-tiling-grid)\nwith an output resolution of 50 m, each batch tile will be 2000 x 2000 pixels,\nthus valid chunking sizes are 2000, 1000, 500, 400 etc.\n",
      "example": [
        1,
        1000,
        1000
      ],
      "type": "array",
      "minItems": 3,
      "maxItems": 3,
      "items": {
        "type": "number",
        "format": "integer"
      }
    },
    "fill_value": {
      "type": "number",
      "description": "A scalar value providing the default value for portions of the array corresponding to non-existing chunks:\n* any chunks consisting solely of this value will **not** be written,\n* the value will be included in the output Zarr metadata.\n\n**Note**: `fill_value` must be representable by the array's `dtype`.\n\n**Note**: any grid tiles that are within Zarr envelope but outside of `processRequest.input.bounds.geometry` will not be processed by batch at all.\nNo chunks will thus be written for those tiles, thus `fill_value` is required to ensure a valid Zarr is created.\n"
    },
    "compressor": {
      "description": "Compressor type and parameters, which will be used to compress the data chunks.\n",
      "default": {
        "id": "blosc",
        "cname": "lz4",
        "clevel": 5,
        "shuffle": 1,
        "blocksize": 0
      },
      "type": "object"
    }
  }
}