cuallee

What it is and what it does

Cuallee is a data quality validation framework designed to work across multiple dataframe libraries—PySpark, Snowpark, DuckDB, Polars, Daft, and BigQuery—without requiring a separate implementation for each. It was built as a pure-Python alternative to pydeequ, addressing usability and performance concerns with that framework. The library lets you define checks for data integrity (completeness, uniqueness, patterns, ranges, statistical anomalies) and run them against your dataframe, returning pass/fail results with detailed metrics.

The package depends on toolz and requests, keeping its footprint minimal. It supports checks across multiple data types and includes specialized validators for dates, membership tests, regular expressions, and workflow sequences (process mining). You can compose checks into a fluent API, run them individually or as controls across entire dataframes, and get results as either detailed validation reports or simple boolean assertions.

Use it for:

• Validate completeness and uniqueness on PySpark DataFrames before loading data into a warehouse.
• Run date range and continuity checks (e.g., is_daily) on time-series data in Snowpark or DuckDB.
• Test that categorical columns conform to an allowed set of values using is_contained_in across multiple dataframe backends.
• Detect statistical anomalies (e.g., interquartile range outliers) in numeric columns without writing custom aggregation logic.
• Verify business process workflows (e.g., Order-to-Cash sequences) using has_workflow to ensure event ordering and state transitions.
• Apply completeness checks across all columns in a dataframe using Control.completeness for quick data profiling.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Cuallee provides a dataframe-agnostic API to define and run data quality checks across PySpark, pandas, Snowpark, DuckDB, Polars, Daft, and BigQuery, replacing pydeequ with a pure-Python implementation.

Yes, with conditions. Cuallee is worth installing if you work with multiple dataframe libraries and need a unified data quality API. The low install friction (pure Python, two dependencies), permissive Apache 2.0 license, and broad dataframe support make it accessible. However, the aging maintenance status (311 days since last release, last commit 2026-02-05) means you should verify that the specific dataframe versions you use are still supported in 0.15.4 before committing to production use. No known security vulnerabilities.

Install

cuallee on PyPI

pip install cuallee

uv add cuallee

poetry add cuallee

Installing cuallee

Before you install

Low friction: pure Python wheel with only two runtime dependencies (toolz, requests). Maintenance status is aging—last release was 311 days ago and the last commit on 2026-02-05, though the repo remains active with 248 stars and is not archived.

License in practice

Apache License 2.0 is permissive; you may use, modify, and distribute cuallee freely in commercial and private projects, provided you include a copy of the license and note any changes you make.

Quickstart

pip install cuallee

from cuallee import Check, CheckLevel

check = Check(CheckLevel.WARNING, "Completeness")
check.is_complete("id").is_unique("id").validate(df)

Requires Python 3.10 or later. The package is dataframe-agnostic but each target dataframe library (PySpark, Snowpark, etc.) must be installed separately.

Verify before relying

• Whether all supported dataframe providers (PySpark 4.0.1, Snowpark 1.11.1, pandas 2.0.2, DuckDB 1.4.0, Polars 1.34.0, Daft 0.2.24) are tested and maintained in the current 0.15.4 release.
• Current performance characteristics relative to pydeequ—the description cites a benchmark but does not specify the test dataset size or hardware used.
• Status of the 'new version of validate output' mentioned as under construction in the documentation.

Package facts

Evidence: cuallee-0.15.4-py3-none-any.whl

Tags

More Quality Assurance packages