Command line
bash
hippodatamus-cli --config customers.yamlUsing seed 20260822 (from config)
Generating 1000 rows to customers.csv
Done. 1000 rows in 0.31s (3,225 rows/s) -> customers.csvOptions
Required:
-c, --config <path> YAML schema describing the columns to generate, or a
.hippo bundle. A bundle is unpacked to a temporary
folder and run from there, so it needs nothing else
present.
Options:
-o, --output <path> Output file. Format is inferred from the extension
(.csv, .json, .xlsx, .xls). Defaults to the config's
output.path.
-r, --rows <n> Override the schema's row count. Must be > 0.
-s, --seed <n> Override the seed. Falls back to the config's seed, then
to a random seed, which is printed.
-q, --quiet Print only errors.
-v, --verbose Include stack traces on failure.
-h, --help Show this help.
--version Show the version.Exit codes
| Code | Meaning |
|---|---|
0 | Generation completed |
1 | Generation failed |
2 | Invalid arguments |
Distinguishing 1 from 2 matters in a script: a 2 means the command was wrong, and retrying will not help.
In CI
bash
hippodatamus-cli --config fixtures/orders.yaml --output "$ARTIFACTS/orders.csv" --quietBecause output is byte-identical for a fixed configuration and seed, you can check a hash of the result and detect an unintended change to your fixtures:
bash
hippodatamus-cli --config fixtures/orders.yaml --output orders.csv --quiet
sha256sum -c fixtures/orders.csv.sha256Note that generated files use the platform's line ending, so a hash taken on Windows will not match one taken on Linux. Normalise first, or take the hash on one platform.
Overriding rows and seed
bash
hippodatamus-cli --config orders.yaml --rows 50 --seed 7Useful for a quick look at a schema without editing it. Remember that a different seed means different data - if you want the dataset the file describes, do not pass --seed.