truthound 1.0.8__py3-none-any.whl

truthound/cli_modules/core/check.py

@@ -0,0 +1,123 @@
+"""Check command - Validate data quality.
+
+This module implements the `truthound check` command for validating
+data quality in files.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import Annotated, Optional
+
+import typer
+
+from truthound.cli_modules.common.errors import error_boundary, require_file
+from truthound.cli_modules.common.options import parse_list_callback
+
+
+@error_boundary
+def check_cmd(
+    file: Annotated[
+        Path,
+        typer.Argument(help="Path to the data file"),
+    ],
+    validators: Annotated[
+        Optional[list[str]],
+        typer.Option("--validators", "-v", help="Comma-separated list of validators"),
+    ] = None,
+    min_severity: Annotated[
+        Optional[str],
+        typer.Option(
+            "--min-severity",
+            "-s",
+            help="Minimum severity level (low, medium, high, critical)",
+        ),
+    ] = None,
+    schema_file: Annotated[
+        Optional[Path],
+        typer.Option("--schema", help="Schema file for validation"),
+    ] = None,
+    auto_schema: Annotated[
+        bool,
+        typer.Option("--auto-schema", help="Auto-learn and cache schema (zero-config mode)"),
+    ] = False,
+    format: Annotated[
+        str,
+        typer.Option("--format", "-f", help="Output format (console, json, html)"),
+    ] = "console",
+    output: Annotated[
+        Optional[Path],
+        typer.Option("--output", "-o", help="Output file path"),
+    ] = None,
+    strict: Annotated[
+        bool,
+        typer.Option("--strict", help="Exit with code 1 if issues are found"),
+    ] = False,
+) -> None:
+    """Validate data quality in a file.
+
+    This command runs data quality validators on the specified file and
+    reports any issues found.
+
+    Examples:
+        truthound check data.csv
+        truthound check data.parquet --validators null,duplicate,range
+        truthound check data.csv --min-severity high --strict
+        truthound check data.csv --auto-schema
+        truthound check data.csv --format json -o report.json
+    """
+    from truthound.api import check
+
+    # Validate files exist
+    require_file(file)
+    if schema_file:
+        require_file(schema_file, "Schema file")
+
+    # Parse validators if provided
+    validator_list = parse_list_callback(validators) if validators else None
+
+    try:
+        report = check(
+            str(file),
+            validators=validator_list,
+            min_severity=min_severity,
+            schema=schema_file,
+            auto_schema=auto_schema,
+        )
+    except Exception as e:
+        typer.echo(f"Error: {e}", err=True)
+        raise typer.Exit(1)
+
+    # Output the report
+    if format == "json":
+        result = report.to_json()
+        if output:
+            output.write_text(result)
+            typer.echo(f"Report written to {output}")
+        else:
+            typer.echo(result)
+
+    elif format == "html":
+        if not output:
+            typer.echo("Error: --output is required for HTML format", err=True)
+            raise typer.Exit(1)
+        try:
+            from truthound.html_report import generate_html_report
+
+            html = generate_html_report(report)
+            output.write_text(html)
+            typer.echo(f"HTML report written to {output}")
+        except ImportError:
+            typer.echo(
+                "Error: HTML reports require jinja2. "
+                "Install with: pip install truthound[reports]",
+                err=True,
+            )
+            raise typer.Exit(1)
+
+    else:
+        report.print()
+
+    # Exit with error if strict mode and issues found
+    if strict and report.has_issues:
+        raise typer.Exit(1)

truthound/cli_modules/core/compare.py

@@ -0,0 +1,104 @@
+"""Compare command - Compare datasets for drift.
+
+This module implements the `truthound compare` command for detecting
+data drift between two datasets.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import Annotated, Optional
+
+import typer
+
+from truthound.cli_modules.common.errors import error_boundary, require_file
+from truthound.cli_modules.common.options import parse_list_callback
+
+
+@error_boundary
+def compare_cmd(
+    baseline: Annotated[
+        Path,
+        typer.Argument(help="Baseline (reference) data file"),
+    ],
+    current: Annotated[
+        Path,
+        typer.Argument(help="Current data file to compare"),
+    ],
+    columns: Annotated[
+        Optional[list[str]],
+        typer.Option("--columns", "-c", help="Columns to compare (comma-separated)"),
+    ] = None,
+    method: Annotated[
+        str,
+        typer.Option("--method", "-m", help="Detection method (auto, ks, psi, chi2, js)"),
+    ] = "auto",
+    threshold: Annotated[
+        Optional[float],
+        typer.Option("--threshold", "-t", help="Custom drift threshold"),
+    ] = None,
+    format: Annotated[
+        str,
+        typer.Option("--format", "-f", help="Output format (console, json)"),
+    ] = "console",
+    output: Annotated[
+        Optional[Path],
+        typer.Option("--output", "-o", help="Output file path"),
+    ] = None,
+    strict: Annotated[
+        bool,
+        typer.Option("--strict", help="Exit with code 1 if drift is detected"),
+    ] = False,
+) -> None:
+    """Compare two datasets and detect data drift.
+
+    This command compares a baseline dataset with a current dataset and
+    detects statistical drift in column distributions.
+
+    Detection Methods:
+        - auto: Automatically select best method per column
+        - ks: Kolmogorov-Smirnov test (numeric)
+        - psi: Population Stability Index
+        - chi2: Chi-squared test (categorical)
+        - js: Jensen-Shannon divergence
+
+    Examples:
+        truthound compare baseline.csv current.csv
+        truthound compare ref.parquet new.parquet --method psi
+        truthound compare old.csv new.csv --threshold 0.2 --strict
+        truthound compare old.csv new.csv --columns price,quantity
+    """
+    from truthound.drift import compare
+
+    # Validate files exist
+    require_file(baseline, "Baseline file")
+    require_file(current, "Current file")
+
+    # Parse columns if provided
+    column_list = parse_list_callback(columns) if columns else None
+
+    try:
+        drift_report = compare(
+            str(baseline),
+            str(current),
+            columns=column_list,
+            method=method,
+            threshold=threshold,
+        )
+    except Exception as e:
+        typer.echo(f"Error: {e}", err=True)
+        raise typer.Exit(1)
+
+    if format == "json":
+        result = drift_report.to_json()
+        if output:
+            output.write_text(result)
+            typer.echo(f"Drift report written to {output}")
+        else:
+            typer.echo(result)
+    else:
+        drift_report.print()
+
+    # Exit with error if strict mode and drift found
+    if strict and drift_report.has_drift:
+        raise typer.Exit(1)

truthound/cli_modules/core/learn.py

@@ -0,0 +1,57 @@
+"""Learn command - Learn schema from data files.
+
+This module implements the `truthound learn` command for inferring
+schema from data files.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import Annotated
+
+import typer
+
+from truthound.cli_modules.common.errors import error_boundary, require_file
+
+
+@error_boundary
+def learn_cmd(
+    file: Annotated[
+        Path,
+        typer.Argument(help="Path to the data file to learn from"),
+    ],
+    output: Annotated[
+        Path,
+        typer.Option("--output", "-o", help="Output schema file path"),
+    ] = Path("schema.yaml"),
+    no_constraints: Annotated[
+        bool,
+        typer.Option("--no-constraints", help="Don't infer constraints from data"),
+    ] = False,
+) -> None:
+    """Learn schema from a data file.
+
+    This command analyzes the data file and generates a schema definition
+    that captures column types, constraints, and patterns.
+
+    Examples:
+        truthound learn data.csv
+        truthound learn data.parquet -o my_schema.yaml
+        truthound learn data.csv --no-constraints
+    """
+    from truthound.schema import learn
+
+    # Validate file exists
+    require_file(file)
+
+    try:
+        schema = learn(str(file), infer_constraints=not no_constraints)
+        schema.save(output)
+
+        typer.echo(f"Schema saved to {output}")
+        typer.echo(f"  Columns: {len(schema.columns)}")
+        typer.echo(f"  Rows: {schema.row_count:,}")
+
+    except Exception as e:
+        typer.echo(f"Error: {e}", err=True)
+        raise typer.Exit(1)

truthound/cli_modules/core/mask.py

@@ -0,0 +1,77 @@
+"""Mask command - Mask sensitive data.
+
+This module implements the `truthound mask` command for masking
+sensitive data in files.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import Annotated, Optional
+
+import typer
+
+from truthound.cli_modules.common.errors import error_boundary, require_file
+from truthound.cli_modules.common.options import parse_list_callback
+
+
+@error_boundary
+def mask_cmd(
+    file: Annotated[
+        Path,
+        typer.Argument(help="Path to the data file"),
+    ],
+    output: Annotated[
+        Path,
+        typer.Option("--output", "-o", help="Output file path"),
+    ],
+    columns: Annotated[
+        Optional[list[str]],
+        typer.Option("--columns", "-c", help="Columns to mask (comma-separated)"),
+    ] = None,
+    strategy: Annotated[
+        str,
+        typer.Option("--strategy", "-s", help="Masking strategy (redact, hash, fake)"),
+    ] = "redact",
+) -> None:
+    """Mask sensitive data in a file.
+
+    This command creates a copy of the data file with sensitive columns
+    masked using the specified strategy.
+
+    Strategies:
+        - redact: Replace values with asterisks
+        - hash: Replace values with hashed versions
+        - fake: Replace values with realistic fake data
+
+    Examples:
+        truthound mask data.csv -o masked.csv
+        truthound mask data.csv -o masked.csv --columns email,phone
+        truthound mask data.csv -o masked.csv --strategy hash
+    """
+    from truthound.api import mask
+
+    # Validate file exists
+    require_file(file)
+
+    # Parse columns if provided
+    column_list = parse_list_callback(columns) if columns else None
+
+    try:
+        masked_df = mask(str(file), columns=column_list, strategy=strategy)
+    except Exception as e:
+        typer.echo(f"Error: {e}", err=True)
+        raise typer.Exit(1)
+
+    # Write output based on file extension
+    suffix = output.suffix.lower()
+    if suffix == ".csv":
+        masked_df.write_csv(output)
+    elif suffix == ".parquet":
+        masked_df.write_parquet(output)
+    elif suffix == ".json":
+        masked_df.write_json(output)
+    else:
+        masked_df.write_csv(output)
+
+    typer.echo(f"Masked data written to {output}")

truthound/cli_modules/core/profile.py

@@ -0,0 +1,65 @@
+"""Profile command - Generate data profiles.
+
+This module implements the `truthound profile` command for generating
+statistical profiles of data files.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import Annotated, Optional
+
+import typer
+
+from truthound.cli_modules.common.errors import error_boundary, require_file
+
+
+@error_boundary
+def profile_cmd(
+    file: Annotated[
+        Path,
+        typer.Argument(help="Path to the data file"),
+    ],
+    format: Annotated[
+        str,
+        typer.Option("--format", "-f", help="Output format (console, json)"),
+    ] = "console",
+    output: Annotated[
+        Optional[Path],
+        typer.Option("--output", "-o", help="Output file path"),
+    ] = None,
+) -> None:
+    """Generate a statistical profile of the data.
+
+    This command analyzes the data file and generates statistics including:
+    - Row and column counts
+    - Null ratios per column
+    - Unique value counts
+    - Data type distribution
+    - Value ranges for numeric columns
+
+    Examples:
+        truthound profile data.csv
+        truthound profile data.parquet --format json
+        truthound profile data.csv -o profile.json
+    """
+    from truthound.api import profile
+
+    # Validate file exists
+    require_file(file)
+
+    try:
+        profile_report = profile(str(file))
+    except Exception as e:
+        typer.echo(f"Error: {e}", err=True)
+        raise typer.Exit(1)
+
+    if format == "json":
+        result = profile_report.to_json()
+        if output:
+            output.write_text(result)
+            typer.echo(f"Profile written to {output}")
+        else:
+            typer.echo(result)
+    else:
+        profile_report.print()

truthound/cli_modules/core/scan.py

@@ -0,0 +1,61 @@
+"""Scan command - Scan for PII.
+
+This module implements the `truthound scan` command for detecting
+personally identifiable information in data files.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import Annotated, Optional
+
+import typer
+
+from truthound.cli_modules.common.errors import error_boundary, require_file
+
+
+@error_boundary
+def scan_cmd(
+    file: Annotated[
+        Path,
+        typer.Argument(help="Path to the data file"),
+    ],
+    format: Annotated[
+        str,
+        typer.Option("--format", "-f", help="Output format (console, json)"),
+    ] = "console",
+    output: Annotated[
+        Optional[Path],
+        typer.Option("--output", "-o", help="Output file path"),
+    ] = None,
+) -> None:
+    """Scan for personally identifiable information (PII).
+
+    This command analyzes data files to detect columns that may contain
+    PII such as names, emails, phone numbers, SSNs, etc.
+
+    Examples:
+        truthound scan data.csv
+        truthound scan data.parquet --format json
+        truthound scan data.csv -o pii_report.json
+    """
+    from truthound.api import scan
+
+    # Validate file exists
+    require_file(file)
+
+    try:
+        pii_report = scan(str(file))
+    except Exception as e:
+        typer.echo(f"Error: {e}", err=True)
+        raise typer.Exit(1)
+
+    if format == "json":
+        result = pii_report.to_json()
+        if output:
+            output.write_text(result)
+            typer.echo(f"Report written to {output}")
+        else:
+            typer.echo(result)
+    else:
+        pii_report.print()

truthound/cli_modules/profiler/__init__.py

@@ -0,0 +1,51 @@
+"""Profiler CLI commands for Truthound.
+
+This package contains auto-profiling and rule generation commands:
+    - auto_profile: Profile data with auto-detection
+    - generate_suite: Generate validation rules from profile
+    - quick_suite: Profile and generate rules in one step
+    - list_formats: List available output formats
+    - list_presets: List available configuration presets
+    - list_categories: List available rule categories
+"""
+
+import typer
+
+from truthound.cli_modules.profiler.auto_profile import auto_profile_cmd
+from truthound.cli_modules.profiler.suite import (
+    generate_suite_cmd,
+    quick_suite_cmd,
+)
+from truthound.cli_modules.profiler.metadata import (
+    list_formats_cmd,
+    list_presets_cmd,
+    list_categories_cmd,
+)
+
+# Profiler app isn't a subcommand - these are top-level commands
+# We just organize them here for modularity
+
+
+def register_commands(parent_app: typer.Typer) -> None:
+    """Register profiler commands with the parent app.
+
+    Args:
+        parent_app: Parent Typer app to register commands to
+    """
+    parent_app.command(name="auto-profile")(auto_profile_cmd)
+    parent_app.command(name="generate-suite")(generate_suite_cmd)
+    parent_app.command(name="quick-suite")(quick_suite_cmd)
+    parent_app.command(name="list-formats")(list_formats_cmd)
+    parent_app.command(name="list-presets")(list_presets_cmd)
+    parent_app.command(name="list-categories")(list_categories_cmd)
+
+
+__all__ = [
+    "register_commands",
+    "auto_profile_cmd",
+    "generate_suite_cmd",
+    "quick_suite_cmd",
+    "list_formats_cmd",
+    "list_presets_cmd",
+    "list_categories_cmd",
+]