Skip to main content

Expectation Suites

The Expectations page at /data-quality/expectations integrates Great Expectations (GX) suites with Clone-Xs's catalog and rule engine. Use it when DQX's lightweight rules aren't enough — for example when you need multi-step validations, custom Python expectations, or full Data Docs.

DQX vs. Expectations​

DQXExpectations (GX)
Rule complexitySingle predicateMulti-step, programmable
RuntimeSQL onlySQL + Python
LatencySub-secondSeconds-to-minutes
Inline cloneYesNo (post-clone job)
Data DocsNoYes

Default to DQX for fast inline checks. Reach for Expectations when you need Python custom expectations or auto-generated documentation.

Suite structure​

A GX suite groups multiple expectations against one batch of data. Clone-Xs stores suites under clxs.yaml's dq.expectations.path (default _clxs_artifacts/expectations/).

Authoring​

The page lets you:

  • Create suite — pick a table, choose expectations from the GX catalog, parametrise
  • Edit YAML — direct edit of the suite YAML
  • Validate — run the suite against the table, view results
  • Generate Data Docs — produce the GX HTML site

Scheduling​

Suites run via DQ Automation. Common patterns:

  • Post-clone — validate cloned data before promoting it to the destination
  • Hourly — recurring quality check on production tables
  • Pre-publish — gate a Data Product release

Custom expectations​

Python custom expectations live under _clxs_artifacts/expectations/custom/. They're auto-loaded at runtime — no rebuild needed. See the GX docs for the expectation interface.

API​

GET    /data-quality/expectations/suites
POST   /data-quality/expectations/suites
POST   /data-quality/expectations/suites/{name}/validate
POST   /data-quality/expectations/suites/{name}/docs    # generate HTML