add_date

The add_date processor adds the current time to the configured field for any DataEntity or DataWindow.

The format option controls how the date is written — an ISO 8601 string by default, or epoch seconds/milliseconds as a number, or any date-fns format string. See Use a different date format for the full list.

The date is generated per record at the time the record is processed, so records in the same slice may have slightly different values.

Usage

Add an ISO 8601 processing timestamp to every record

Example of a job using the add_date processor

{
    "name" : "testing",
    "workers" : 1,
    "slicers" : 1,
    "lifecycle" : "once",
    "assets" : [
        "standard"
    ],
    "operations" : [
        {
            "_op": "test-reader"
        },
        {
            "_op": "add_date",
            "field": "processed_at"
        }
    ]
}

Example of the data and the expected results

const data = [
    DataEntity.make({ name: 'lilly', otherField: 1 }),
    DataEntity.make({ name: 'willy', otherField: 2  }),
]

const results = await processor.run(data);

DataEntity.make({ name: 'lilly', otherField: 1, processed_at: '2026-09-11T17:04:22.314Z' }),
DataEntity.make({ name: 'willy', otherField: 2, processed_at: '2026-09-11T17:04:22.314Z' }),

Use a different date format

The format option accepts any of the following:

format	Example output	Type
iso_8601 (default)	2026-09-11T17:04:22.314Z	String
epoch_millis	1789664662314	Number
milliseconds	1789664662314	Number
epoch	1789664662	Number
seconds	1789664662	Number
any date-fns format string	2026-09-11 for yyyy-MM-dd	String

{
    "_op": "add_date",
    "field": "processed_at",
    "format": "epoch_millis"
}

date-fns formats and UTC

A date-fns format string renders in UTC by default, which matches the iso_8601 and epoch formats:

{
    "_op": "add_date",
    "field": "processed_at",
    "format": "yyyy-MM-dd HH:mm:ss"
}

For the instant 2026-09-11T02:30:00.000Z this writes "2026-09-11 02:30:00" — the UTC wall clock, regardless of the worker's timezone.

Including an x or X timezone token switches the output to the worker's local time, with the offset rendered by the token:

{
    "_op": "add_date",
    "field": "processed_at",
    "format": "yyyy-MM-dd'T'HH:mm:ss.SSSXXX"
}

On a worker in America/Denver the same instant writes "2026-09-10T20:30:00.000-06:00". That is the same point in time, just expressed in local terms.

Note: the local-time offset is read once when the worker process starts. A long-running worker in a timezone that observes daylight saving will keep using the offset that was in effect at startup, so x/X output can drift by an hour after a DST transition. Use the default UTC rendering, iso_8601, or an epoch format if that matters.

Overwrite an existing field

By default a record that already has the field is left alone. Set overwrite to true to always stamp the field.

{
    "_op": "add_date",
    "field": "processed_at",
    "overwrite": true
}

Parameters

Configuration	Description	Type	Notes
_op	Name of operation, it must reflect the exact name of the file	String	required
field	Name of the field that the date is written to	String	required, no default
format	Output format of the date, either a DateFormat or a date-fns format string	String	optional, defaults to iso_8601
overwrite	Option to set the field even if it is already on the document	Boolean	optional, defaults to false