Skip to main content

# kater test readiness

Says whether a shared model is ready to hand off to the people who will use it.

{/* Written by apps/landing-page/scripts/docs/write_cli_reference.py from the kater CLI's own help and scripts/docs/cli-reference.md. Do not edit it by hand: change those, then run the script. */}

It reads the last run of each test on this machine, so run [`kater test`](https://kater.ai/docs/cli-reference/test/) first. Readiness only informs: nothing stops you from using a model that is not ready.

## Examples

```bash
# Run the tests of the shared ad_reporting model
kater test --select semantic.ad_reporting
# Check whether the model is ready to hand off
kater test readiness --model semantic.ad_reporting
```

## Options

<div class="cli-options">

| Option | Description |
| --- | --- |
| `--connection <name>` | The connection, by its folder name under `connections/`. Leave it out when only one connection has the `--model`. |
| `--model <model>` | Required. The model to check, by its name or typed address, such as `semantic.ad_reporting`. |

</div>

It also takes the [global options](https://kater.ai/docs/cli-reference/global-options/) `--json`, `--kater-root`, `--scope` and `--help`.

## What ready means

Only the tests in the shared layer's file count. A model is ready when each of these checks is met:

| Check | Met when |
| --- | --- |
| Every test passes | Every test passed on its last run, and nothing that it reads changed since. |
| No drafts | No test is still a draft. |
| Every measure in a checked test | Each measure of the model is in at least one checked test. |
| Matches native SQL | A `same_as` test against native SQL, such as a query of a dbt mart, passes. |
| Every rule pinned | Each `NOT_ADDITIVE_ACROSS` rule that the model declares has a passing refused test. |

It also lists how many of the model's business questions a checked test answers. It lists known issues too: the tests that the last `kater test generate` could not write. These two do not block hand-off.

## What it prints

One line for each part: Tests, Measures, Families, Picker path, Oracle, Refusals and Questions, then any known issues. The last line says "Ready to hand off: yes", or "not yet" with the checks still to meet. The command exits with `0` when the model is ready and `1` when it is not, so a script or a CI job can wait for it.

## See also

- [`kater test`](https://kater.ai/docs/cli-reference/test/)
- [All commands](https://kater.ai/docs/cli-reference/kater/)

On this page

Examples
Options
What ready means
What it prints
See also

Kater CLI Build and check test

kater test readiness

Says whether a shared model is ready to hand off to the people who will use it.

It reads the last run of each test on this machine, so run kater test first. Readiness only informs: nothing stops you from using a model that is not ready.

Examples

# Run the tests of the shared ad_reporting model
kater test --select semantic.ad_reporting
# Check whether the model is ready to hand off
kater test readiness --model semantic.ad_reporting

Options

Option	Description
`--connection <name>`	The connection, by its folder name under `connections/`. Leave it out when only one connection has the `--model`.
`--model <model>`	Required. The model to check, by its name or typed address, such as `semantic.ad_reporting`.

It also takes the global options --json, --kater-root, --scope and --help.

What ready means

Only the tests in the shared layer’s file count. A model is ready when each of these checks is met:

Check	Met when
Every test passes	Every test passed on its last run, and nothing that it reads changed since.
No drafts	No test is still a draft.
Every measure in a checked test	Each measure of the model is in at least one checked test.
Matches native SQL	A `same_as` test against native SQL, such as a query of a dbt mart, passes.
Every rule pinned	Each `NOT_ADDITIVE_ACROSS` rule that the model declares has a passing refused test.

It also lists how many of the model’s business questions a checked test answers. It lists known issues too: the tests that the last kater test generate could not write. These two do not block hand-off.

What it prints

One line for each part: Tests, Measures, Families, Picker path, Oracle, Refusals and Questions, then any known issues. The last line says “Ready to hand off: yes”, or “not yet” with the checks still to meet. The command exits with 0 when the model is ready and 1 when it is not, so a script or a CI job can wait for it.

See also