truero 0.2.1__tar.gz

truero-0.2.1/LICENSE

@@ -0,0 +1,16 @@
+Copyright (c) 2026 Truero.ai Limited. All rights reserved.
+
+This software and associated documentation files (the "Software") are the
+proprietary property of Truero.ai Limited.
+
+Subject to the Truero Terms & Conditions (https://truero.ai/terms), you are
+granted a limited, non-exclusive, non-transferable licence to install and use
+the Software for your internal business or personal purposes.
+
+You may not sublicence, sell, resell, reverse-engineer, or otherwise exploit
+the Software commercially without prior written consent from Truero.ai Limited.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND. See the Terms
+& Conditions and Privacy Policy (https://truero.ai/privacy) for full details.
+
+Contact: legal@truero.ai

truero-0.2.1/PKG-INFO

@@ -0,0 +1,84 @@
+Metadata-Version: 2.4
+Name: truero
+Version: 0.2.1
+Summary: Understand and clean your data before it causes problems
+Home-page: https://github.com/Thisizzellie/Truero
+Author: Truero
+Author-email: hello@truero.ai
+License: Proprietary
+Project-URL: Homepage, https://www.truero.ai
+Project-URL: Dashboard, https://dashboard.truero.ai
+Project-URL: Source, https://github.com/Thisizzellie/Truero
+Keywords: ml data-quality validation bias audit cli
+Classifier: Development Status :: 4 - Beta
+Classifier: Environment :: Console
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: Other/Proprietary License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: click>=8.0
+Requires-Dist: requests>=2.28
+Requires-Dist: rich>=13.0
+Requires-Dist: pandas>=1.5
+Requires-Dist: numpy>=1.23
+Requires-Dist: flask>=3.0
+Requires-Dist: openpyxl>=3.0
+Requires-Dist: pyarrow>=10.0
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: keywords
+Dynamic: license
+Dynamic: license-file
+Dynamic: project-url
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
+
+# Truero CLI
+
+Understand and clean your data before it causes problems.
+
+## Install
+
+```bash
+pip install truero
+```
+
+## Usage
+
+```bash
+truero scan data.csv
+truero web
+```
+
+## Options
+
+| Flag / env | Purpose |
+|------------|---------|
+| `--api-key` / `TRUERO_API_KEY` | API key sent to the Truero backend |
+| `--user-id` / `TRUERO_USER_ID` | Supabase auth UUID for dashboard upload |
+| `--no-upload` | Skip dashboard upload prompt |
+| `TRUERO_API_URL` | Backend base URL (default: hosted Railway API for pip installs; `http://localhost:8001` in monorepo) |
+| `TRUERO_DASHBOARD_URL` | Dashboard base URL (default: `https://dashboard.truero.ai`) |
+
+You can also set API/dashboard URLs in the web wizard Settings, or in `~/.truero/config.json` (`web_api_url`, `web_dashboard_url`).
+
+## Notes
+
+- Cleaning checks run on your machine.
+- ML checks (Option A) call the Truero backend API.
+- Upload sends the report to the backend so it appears in the dashboard.
+- Supported file types: CSV, JSON, Excel (`.xlsx`/`.xls`), Parquet.
+- First run prompts you to accept the Terms & Conditions (or pass `--accept-terms`).

truero-0.2.1/README.md

@@ -0,0 +1,36 @@
+# Truero CLI
+
+Understand and clean your data before it causes problems.
+
+## Install
+
+```bash
+pip install truero
+```
+
+## Usage
+
+```bash
+truero scan data.csv
+truero web
+```
+
+## Options
+
+| Flag / env | Purpose |
+|------------|---------|
+| `--api-key` / `TRUERO_API_KEY` | API key sent to the Truero backend |
+| `--user-id` / `TRUERO_USER_ID` | Supabase auth UUID for dashboard upload |
+| `--no-upload` | Skip dashboard upload prompt |
+| `TRUERO_API_URL` | Backend base URL (default: hosted Railway API for pip installs; `http://localhost:8001` in monorepo) |
+| `TRUERO_DASHBOARD_URL` | Dashboard base URL (default: `https://dashboard.truero.ai`) |
+
+You can also set API/dashboard URLs in the web wizard Settings, or in `~/.truero/config.json` (`web_api_url`, `web_dashboard_url`).
+
+## Notes
+
+- Cleaning checks run on your machine.
+- ML checks (Option A) call the Truero backend API.
+- Upload sends the report to the backend so it appears in the dashboard.
+- Supported file types: CSV, JSON, Excel (`.xlsx`/`.xls`), Parquet.
+- First run prompts you to accept the Terms & Conditions (or pass `--accept-terms`).

truero-0.2.1/pyproject.toml

@@ -0,0 +1,3 @@
+[build-system]
+requires = ["setuptools>=61", "wheel"]
+build-backend = "setuptools.build_meta"

truero-0.2.1/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

truero-0.2.1/setup.py

@@ -0,0 +1,60 @@
+from pathlib import Path
+
+from setuptools import setup, find_packages
+
+_README = Path(__file__).with_name("README.md")
+_long_description = _README.read_text(encoding="utf-8") if _README.exists() else ""
+
+setup(
+    name="truero",
+    version="0.2.1",
+    description="Understand and clean your data before it causes problems",
+    long_description=_long_description,
+    long_description_content_type="text/markdown",
+    author="Truero",
+    author_email="hello@truero.ai",
+    url="https://github.com/Thisizzellie/Truero",
+    project_urls={
+        "Homepage": "https://www.truero.ai",
+        "Dashboard": "https://dashboard.truero.ai",
+        "Source": "https://github.com/Thisizzellie/Truero",
+    },
+    license="Proprietary",
+    license_files=("LICENSE",),
+    packages=find_packages(exclude=["tests", "tests.*"]),
+    package_data={
+        "truero": ["legal/*.json"],
+        "truero.web": ["static/*.html", "static/*.js", "static/*.css"],
+    },
+    include_package_data=True,
+    install_requires=[
+        "click>=8.0",
+        "requests>=2.28",
+        "rich>=13.0",
+        "pandas>=1.5",
+        "numpy>=1.23",
+        "flask>=3.0",
+        "openpyxl>=3.0",
+        "pyarrow>=10.0",
+    ],
+    entry_points={
+        "console_scripts": [
+            "truero=truero.main:cli",
+        ],
+    },
+    python_requires=">=3.9",
+    classifiers=[
+        "Development Status :: 4 - Beta",
+        "Environment :: Console",
+        "Intended Audience :: Developers",
+        "Intended Audience :: Science/Research",
+        "License :: Other/Proprietary License",
+        "Programming Language :: Python :: 3",
+        "Programming Language :: Python :: 3.9",
+        "Programming Language :: Python :: 3.10",
+        "Programming Language :: Python :: 3.11",
+        "Programming Language :: Python :: 3.12",
+        "Topic :: Scientific/Engineering :: Artificial Intelligence",
+    ],
+    keywords="ml data-quality validation bias audit cli",
+)

truero-0.2.1/tests/test_cleaning_checks.py

@@ -0,0 +1,390 @@
+"""Unit tests for CLI cleaning checks — all 8 checks in cleaning_engine.py.
+
+Run with:
+  cd cli && source .venv/bin/activate && pip install -e . && pytest tests/ -v
+"""
+
+import pytest
+import pandas as pd
+import numpy as np
+from truero.cleaning_engine import (
+    check_duplicate_rows,
+    check_missing_values,
+    check_outliers,
+    check_data_types,
+    check_cardinality,
+    check_correlation,
+    check_format_consistency,
+    check_text_consistency,
+    run_cleaning_checks,
+)
+
+
+# ── Duplicate Rows ───────────────────────────────────────────────────
+
+def test_duplicate_rows_pass():
+    df = pd.DataFrame({"a": [1, 2, 3], "b": ["x", "y", "z"]})
+    result = check_duplicate_rows(df)
+    assert result["status"] == "PASS"
+    assert result["detail"]["duplicate_count"] == 0
+    assert result["detail"]["duplicate_rate"] == 0.0
+
+
+def test_duplicate_rows_warning():
+    # 2 dupes out of 102 rows ≈ 2.0% (between 1% and 5%)
+    uniques = pd.DataFrame({"a": range(100), "b": [f"x{i}" for i in range(100)]})
+    dupes = pd.DataFrame({"a": [0, 1], "b": ["x0", "x1"]})
+    df = pd.concat([uniques, dupes], ignore_index=True)
+    result = check_duplicate_rows(df)
+    assert result["status"] == "WARNING"
+    assert result["detail"]["duplicate_count"] == 2
+    assert "df.drop_duplicates()" in result["suggestion"]
+
+
+def test_duplicate_rows_critical():
+    # 10 dupes out of 110 rows ≈ 9.1% (above 5%)
+    uniques = pd.DataFrame({"a": range(100), "b": [f"x{i}" for i in range(100)]})
+    dupes = pd.DataFrame({"a": range(10), "b": [f"x{i}" for i in range(10)]})
+    df = pd.concat([uniques, dupes], ignore_index=True)
+    result = check_duplicate_rows(df)
+    assert result["status"] == "CRITICAL"
+    assert result["detail"]["duplicate_count"] == 10
+    assert "df.drop_duplicates()" in result["suggestion"]
+
+
+def test_duplicate_rows_empty():
+    df = pd.DataFrame()
+    result = check_duplicate_rows(df)
+    assert result["status"] == "PASS"
+    assert result["detail"]["duplicate_count"] == 0
+
+
+def test_duplicate_rows_suggestion_includes_count():
+    uniques = pd.DataFrame({"a": range(100)})
+    dupes = pd.DataFrame({"a": range(10)})
+    df = pd.concat([uniques, dupes], ignore_index=True)
+    result = check_duplicate_rows(df)
+    assert "10 duplicate rows" in result["suggestion"]
+
+
+# ── Missing Values ────────────────────────────────────────────────────
+
+def test_missing_values_pass():
+    df = pd.DataFrame({"a": [1, 2, 3], "b": ["x", "y", "z"]})
+    result = check_missing_values(df, critical_threshold=0.20)
+    assert result["status"] == "PASS"
+    assert result["detail"]["total_missing"] == 0
+
+
+def test_missing_values_critical():
+    # 50% missing should exceed even the lowest critical threshold
+    df = pd.DataFrame({"a": [1, None], "b": [None, "x"]})
+    result = check_missing_values(df, critical_threshold=0.10)
+    assert result["status"] == "CRITICAL"
+    assert result["detail"]["total_missing"] > 0
+
+
+def test_missing_values_warning():
+    # 10% missing (below 20% critical, above 5% warning)
+    df = pd.DataFrame({"a": [1, 2, 3, 4, 5, 6, 7, 8, 9, None]})
+    result = check_missing_values(df, critical_threshold=0.20)
+    assert result["status"] in ("WARNING", "CRITICAL")
+
+
+def test_missing_values_suggestion_includes_fillna():
+    df = pd.DataFrame({"age": [25, None, 35, None, 45]})
+    result = check_missing_values(df, critical_threshold=0.20)
+    assert "fillna" in result["suggestion"]
+    assert "age" in result["suggestion"]
+
+
+def test_missing_values_empty():
+    df = pd.DataFrame()
+    result = check_missing_values(df)
+    assert result["status"] == "PASS"
+
+
+# ── Outlier Detection ─────────────────────────────────────────────────
+
+def test_outliers_pass():
+    np.random.seed(42)
+    df = pd.DataFrame({"score": np.random.randn(100) * 10 + 50})
+    result = check_outliers(df, threshold=0.05)
+    assert result["status"] == "PASS"
+    assert result["detail"]["columns_with_outliers"] == 0
+
+
+def test_outliers_warning():
+    np.random.seed(42)
+    df = pd.DataFrame(
+        {"score": list(np.random.randn(100) * 10 + 50) + [500, 600, -100, -200]}
+    )
+    result = check_outliers(df, threshold=0.02)
+    assert result["status"] == "WARNING"
+    assert result["detail"]["columns_with_outliers"] > 0
+
+
+def test_outliers_suggestion_includes_clip():
+    np.random.seed(42)
+    df = pd.DataFrame(
+        {"score": list(np.random.randn(50) * 10 + 50) + [500, 600]}
+    )
+    result = check_outliers(df, threshold=0.02)
+    if result["status"] != "PASS":
+        assert "clip" in result["suggestion"]
+        assert "score" in result["suggestion"]
+
+
+# ── Data Type Consistency ─────────────────────────────────────────────
+
+def test_data_types_pass():
+    df = pd.DataFrame({"age": [25, 30, 35], "name": ["Alice", "Bob", "Carol"]})
+    result = check_data_types(df)
+    assert result["status"] == "PASS"
+    assert result["detail"]["issues_found"] == 0
+
+
+def test_data_types_date_keyword():
+    df = pd.DataFrame({"join_date": ["2024-01-01", "2024-02-01", "2024-03-01"]})
+    result = check_data_types(df)
+    assert any(i["column"] == "join_date" for i in result["detail"]["issues"])
+
+
+def test_data_types_numeric_keyword():
+    df = pd.DataFrame({"age": ["25", "30", "35"]})
+    result = check_data_types(df)
+    assert any(i["column"] == "age" for i in result["detail"]["issues"])
+
+
+def test_data_types_value_sampling_catches_numeric():
+    """Column with non-obvious name but 100% numeric values should be caught."""
+    df = pd.DataFrame({"info": ["100", "200", "300"]})
+    result = check_data_types(df)
+    assert any(
+        i["column"] == "info" and "numeric" in i["issue"]
+        for i in result["detail"]["issues"]
+    )
+
+
+def test_data_types_value_sampling_catches_dates():
+    """Column with non-obvious name but 100% date values should be caught."""
+    df = pd.DataFrame({"field_x": ["2024-01-01", "2024-02-01", "2024-03-01"]})
+    result = check_data_types(df)
+    assert any(
+        i["column"] == "field_x" and "date" in i["issue"]
+        for i in result["detail"]["issues"]
+    )
+
+
+# ── Cardinality ────────────────────────────────────────────────────────
+
+def test_cardinality_pass():
+    df = pd.DataFrame({"category": ["A", "B", "A", "C", "B", "D"]})
+    result = check_cardinality(df)
+    assert result["status"] == "PASS"
+    assert result["detail"]["issues_found"] == 0
+
+
+def test_cardinality_constant_column():
+    df = pd.DataFrame({"always_x": ["x", "x", "x"], "varies": [1, 2, 3]})
+    result = check_cardinality(df)
+    assert any(
+        i["column"] == "always_x" and i["type"] == "constant"
+        for i in result["detail"]["all_issues"]
+    )
+
+
+def test_cardinality_likely_id():
+    df = pd.DataFrame({"id": [f"ID_{i:05d}" for i in range(200)]})
+    result = check_cardinality(df)
+    assert any(
+        i["column"] == "id" and i["type"] == "likely_id"
+        for i in result["detail"]["all_issues"]
+    )
+
+
+# ── Correlation ────────────────────────────────────────────────────────
+
+def test_correlation_pass():
+    # Uncorrelated data should pass
+    np.random.seed(42)
+    df = pd.DataFrame({"a": np.random.randn(100), "b": np.random.randn(100)})
+    result = check_correlation(df)
+    assert result["status"] == "PASS"
+
+
+def test_correlation_warning():
+    df = pd.DataFrame({"a": range(100), "b": [x + np.random.randn() * 0.01 for x in range(100)]})
+    result = check_correlation(df)
+    assert result["status"] == "WARNING"
+    assert result["detail"]["pairs_found"] > 0
+
+
+def test_correlation_not_enough_numeric():
+    df = pd.DataFrame({"a": ["x", "y", "z"], "b": ["p", "q", "r"]})
+    result = check_correlation(df)
+    assert result["status"] == "PASS"
+
+
+# ── Format Consistency ─────────────────────────────────────────────────
+
+def test_format_consistency_mixed_dates():
+    df = pd.DataFrame({
+        "dt": ["2024-01-01", "01/15/2024", "2024-03-20", "12/31/2024"]
+    })
+    result = check_format_consistency(df)
+    assert result["status"] != "PASS"
+    assert any(
+        "mixed date formats" in i["issue"]
+        for i in result["detail"]["issues"]
+    )
+
+
+def test_format_consistency_number_formatting():
+    df = pd.DataFrame({
+        "revenue": ["$50,000", "$75,200", "$120,000.50", "$90K", "$1.2M"]
+    })
+    result = check_format_consistency(df)
+    assert any(
+        "number formatting" in i["issue"]
+        for i in result["detail"]["issues"]
+    )
+
+
+def test_format_consistency_whitespace():
+    df = pd.DataFrame({
+        "name": ["Alice ", " Bob", "Carol"]
+    })
+    result = check_format_consistency(df)
+    assert any(
+        "whitespace" in i["issue"]
+        for i in result["detail"]["issues"]
+    )
+
+
+def test_format_consistency_mixed_case():
+    df = pd.DataFrame({
+        "dept": ["ENGINEERING", "Sales", "HR", "engineering", "marketing"]
+    })
+    result = check_format_consistency(df)
+    assert any(
+        "mixed text case" in i["issue"]
+        for i in result["detail"]["issues"]
+    )
+
+
+def test_format_consistency_unit_inconsistency():
+    df = pd.DataFrame({
+        "weight": ["kg", "KG", "Kilogram", "kilograms", "kg", "kgs"]
+    })
+    result = check_format_consistency(df)
+    assert any(
+        "inconsistent units" in i["issue"]
+        for i in result["detail"]["issues"]
+    )
+
+
+def test_format_consistency_pass():
+    df = pd.DataFrame({
+        "date": ["2024-01-01", "2024-02-01"],
+        "amount": ["50000", "75000"],
+        "name": ["Alice", "Bob"],
+    })
+    result = check_format_consistency(df)
+    assert result["status"] == "PASS"
+
+
+# ── Text Consistency ───────────────────────────────────────────────────
+
+def test_text_consistency_case_variants():
+    df = pd.DataFrame({
+        "dept": ["Engineering", "engineering", "ENGINEERING"]
+    })
+    result = check_text_consistency(df)
+    assert result["status"] != "PASS"
+    assert any(
+        "case/space variants" in i["issue"]
+        for i in result["detail"]["issues"]
+    )
+
+
+def test_text_consistency_near_duplicates():
+    df = pd.DataFrame({
+        "product": ["iPhone", "I Phone", "iPhone ", "iphone"]
+    })
+    result = check_text_consistency(df)
+    # Should catch at least case variants
+    assert result["detail"]["issues_found"] > 0
+
+
+def test_text_consistency_skips_numeric_columns():
+    df = pd.DataFrame({
+        "revenue": ["$50,000", "$75,200", "$120,000"]
+    })
+    result = check_text_consistency(df)
+    # Revenue values are numbers with formatting — should be skipped
+    issues = [
+        i for i in result["detail"]["issues"] if i["column"] == "revenue"
+    ]
+    assert len(issues) == 0
+
+
+def test_text_consistency_skips_date_columns():
+    df = pd.DataFrame({
+        "dt": ["2024-01-01", "2024-02-01", "2024-03-01"]
+    })
+    result = check_text_consistency(df)
+    issues = [i for i in result["detail"]["issues"] if i["column"] == "dt"]
+    assert len(issues) == 0
+
+
+def test_text_consistency_pass():
+    df = pd.DataFrame({"name": ["Alice", "Bob", "Carol"]})
+    result = check_text_consistency(df)
+    assert result["status"] == "PASS"
+
+
+def test_text_consistency_rows_affected_is_int():
+    df = pd.DataFrame({
+        "dept": ["Sales", "SALES", "sales"]
+    })
+    result = check_text_consistency(df)
+    for i in result["detail"]["issues"]:
+        assert isinstance(i["rows_affected"], (int, np.integer))
+
+
+# ── Run All Checks ─────────────────────────────────────────────────────
+
+def test_run_cleaning_checks_returns_all_8():
+    df = pd.DataFrame({
+        "age": [25, 30, 35, None, 45, 50],
+        "name": ["Alice", "Bob", "Carol", "Dave", "Eve", "Frank"],
+        "salary": [50000, 60000, 70000, 80000, 90000, 100000],
+    })
+    context = {"missing_critical": 20, "outlier_threshold": 0.05}
+    results = run_cleaning_checks(df, context)
+    assert len(results) == 8
+    expected = {
+        "duplicate_rows", "missing_values", "outlier_detection",
+        "data_type_consistency", "cardinality", "correlation",
+        "format_consistency", "text_consistency",
+    }
+    actual = {r["check_name"] for r in results}
+    assert actual == expected
+
+
+def test_run_cleaning_checks_all_have_categories():
+    df = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
+    context = {"missing_critical": 20, "outlier_threshold": 0.05}
+    results = run_cleaning_checks(df, context)
+    for r in results:
+        assert "category" in r, f"Check {r['check_name']} missing category"
+        assert r["category"] in ("integrity", "reliability", "ethics")
+
+
+def test_run_cleaning_checks_all_return_status():
+    df = pd.DataFrame({"a": [1, 2, 3]})
+    context = {"missing_critical": 20, "outlier_threshold": 0.05}
+    results = run_cleaning_checks(df, context)
+    for r in results:
+        assert r["status"] in ("PASS", "WARNING", "CRITICAL")