Skip to content

Data generation

Krzysztof Słysz edited this page Nov 21, 2026 · 1 revision

Data generation

Required user role: MEMBER

The Generate feature must be enabled on the view (since 3.20.0; it is off in views created earlier).

The Generate data… item of the Import / Export menu (⇅) in the table toolbar fills the bucket with random data rows - up to 10 000 at once. You don't write the rows: for each property you say how to fill it, and Databucket makes the rows (with Datafaker).

Generate data

Properties

The dialog opens with a table of properties, one row for each property column of the view:

Column
Property name The title of the column in the view (- for a property added by hand)
Property path The JSON path of the property, e.g. $.address.city; paths with dots build nested objects
Generator How the property is filled: Text / Datafaker, Number, One of, Date, Date and time
Settings The text, range or values of the generator

Click a cell to change it. The generator of each property is proposed from its column:

  • a property with an enum - One of its values; true/false - One of true, false,
  • a number - Number in a range (an age 18-90, a price or amount with 2 decimals),
  • a date - Date in the last year (a birth date 1950-2005); a date and time - Date and time in the last year,
  • a text - a Datafaker expression guessed from the name: email → user-{n}@example.com, firstName → #{Name.firstName}, lastName → #{Name.lastName}, login → user-{n}, phone → #{numerify '###-###-###'}, city → #{Address.city}, country, street, zip, company, iban, url, description...; other texts get #{Lorem.word}.

The icons above the table: Add property (a property that isn't a column of the view) and Propose from the columns (starts again from the proposals). The trash icon removes a property from the generated rows.

Generator Settings Value
Text / Datafaker #{Name.firstName} A Datafaker expression: #{Name.fullName}, #{Address.city}, #{Internet.emailAddress}, #{Company.name}, #{Commerce.productName}, #{Finance.iban}, #{Lorem.sentence}... (all providers); #{numerify '###'} (digits), #{letterify '???'} (letters), #{regexify '[A-Z]{3}[0-9]{2}'}. Text around it stays, e.g. +48 #{numerify '### ### ###'}
Text / Datafaker user-{n} {n} is the number of the row - unique. It continues after the last row of the bucket, so generating again doesn't repeat the values
Text / Datafaker customer Any other text is the same in every row
Number 18 – 90, decimals 0 A number in the range, with the given number of decimal places
One of active, blocked One of the values, separated with commas (true/false are booleans; numbers are numbers in a numeric column)
Date 2024-01-01 – 2024-12-31 A date yyyy-MM-dd in the range
Date and time 2025-01-01 – 2025-12-31 A UTC timestamp (ISO 8601) in the range

Generating

Below the table:

  • Rows - how many rows to generate, 1 to 10 000.
  • Language - the language of the Datafaker values: names, cities, phone numbers... (en, pl, de, fr...).
  • Seed - optional: the same seed and the same properties give the same rows again, for repeatable test data.
  • Tag - optional: the tag of the new rows.
  • Not reserved / Reserved for me - whether the new rows are reserved for you (you become the owner).

Preview shows the first 10 rows as a table, without saving anything. With a seed, these are exactly the rows that Generate then saves. Generate N rows saves the rows and refreshes the table.

All rows are saved in one transaction: when saving fails, none of them is saved. A property Databucket can't generate - an expression Datafaker doesn't know, a range from 90 to 18, an empty value - is reported with its path before anything is saved.

The properties, the number of rows, the language and the seed of the last generation are remembered in your browser for each bucket.

Without the web application

The same generator is available:

Clone this wiki locally