truthound-dashboard 1.3.1__py3-none-any.whl → 1.4.1__py3-none-any.whl

truthound_dashboard/core/plugins/validator_executor.py

@@ -0,0 +1,401 @@
+"""Custom Validator Executor.
+
+This module provides execution of custom validators in a safe,
+sandboxed environment.
+"""
+
+from __future__ import annotations
+
+import logging
+import uuid
+from dataclasses import dataclass, field
+from datetime import datetime
+from typing import Any
+
+from sqlalchemy.ext.asyncio import AsyncSession
+
+from truthound_dashboard.db.models import CustomValidator, PluginExecutionLog
+
+from .sandbox import create_sandbox, SandboxConfig, SandboxResult
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class ValidatorResult:
+    """Result of custom validator execution.
+
+    Attributes:
+        passed: Whether validation passed.
+        issues: List of validation issues found.
+        message: Summary message.
+        details: Additional details (JSON-serializable).
+        execution_time_ms: Execution time in milliseconds.
+    """
+
+    passed: bool
+    issues: list[dict[str, Any]] = field(default_factory=list)
+    message: str = ""
+    details: dict[str, Any] = field(default_factory=dict)
+    execution_time_ms: float = 0
+
+
+@dataclass
+class ValidatorContext:
+    """Context provided to custom validators.
+
+    Attributes:
+        column_name: Name of the column being validated.
+        column_values: List of values in the column.
+        parameters: Validator parameters.
+        schema: Column schema information.
+        row_count: Total number of rows.
+    """
+
+    column_name: str
+    column_values: list[Any]
+    parameters: dict[str, Any] = field(default_factory=dict)
+    schema: dict[str, Any] = field(default_factory=dict)
+    row_count: int = 0
+
+    def to_dict(self) -> dict[str, Any]:
+        """Convert to dictionary for sandbox execution."""
+        return {
+            "column_name": self.column_name,
+            "values": self.column_values,
+            "params": self.parameters,
+            "schema": self.schema,
+            "row_count": self.row_count,
+        }
+
+
+# Template code that wraps custom validator code
+VALIDATOR_WRAPPER_CODE = '''
+# Custom validator wrapper
+import re
+import math
+import statistics
+from datetime import datetime, date
+from collections import Counter
+
+# User-provided validator code
+{user_code}
+
+# Entry point
+def _execute_validator(column_name, values, params, schema, row_count):
+    """Execute the custom validator.
+
+    Args:
+        column_name: Name of the column.
+        values: List of column values.
+        params: Validator parameters.
+        schema: Column schema.
+        row_count: Total row count.
+
+    Returns:
+        Dictionary with 'passed', 'issues', 'message', 'details'.
+    """
+    # Call the user-defined validate function
+    result = validate(column_name, values, params, schema, row_count)
+
+    # Ensure result is properly formatted
+    if isinstance(result, bool):
+        return {{
+            "passed": result,
+            "issues": [],
+            "message": "Validation passed" if result else "Validation failed",
+            "details": {{}}
+        }}
+    elif isinstance(result, dict):
+        return {{
+            "passed": result.get("passed", False),
+            "issues": result.get("issues", []),
+            "message": result.get("message", ""),
+            "details": result.get("details", {{}})
+        }}
+    else:
+        return {{
+            "passed": False,
+            "issues": [],
+            "message": f"Invalid result type: {{type(result).__name__}}",
+            "details": {{}}
+        }}
+'''
+
+
+class CustomValidatorExecutor:
+    """Executor for custom validators.
+
+    This class handles the execution of custom validators
+    in a sandboxed environment with logging and monitoring.
+
+    Attributes:
+        sandbox: Plugin sandbox for secure execution.
+        log_executions: Whether to log executions to database.
+    """
+
+    def __init__(
+        self,
+        sandbox_config: SandboxConfig | None = None,
+        log_executions: bool = True,
+    ) -> None:
+        """Initialize the executor.
+
+        Args:
+            sandbox_config: Sandbox configuration.
+            log_executions: Whether to log executions.
+        """
+        self.sandbox = create_sandbox(sandbox_config or SandboxConfig())
+        self.log_executions = log_executions
+
+    def validate_validator_code(self, code: str) -> tuple[bool, list[str]]:
+        """Validate custom validator code before execution.
+
+        Args:
+            code: Python code implementing the validator.
+
+        Returns:
+            Tuple of (is_valid, list of issues).
+        """
+        issues = []
+
+        # Check for required function
+        if "def validate(" not in code:
+            issues.append("Missing required 'validate' function")
+
+        # Check for dangerous patterns
+        dangerous = [
+            "os.system",
+            "subprocess",
+            "exec(",
+            "eval(",
+            "__import__",
+            "open(",
+            "file(",
+        ]
+        for pattern in dangerous:
+            if pattern in code:
+                issues.append(f"Dangerous pattern detected: {pattern}")
+
+        # Analyze with sandbox
+        sandbox_issues, _ = self.sandbox.analyze_code(code)
+        issues.extend(sandbox_issues)
+
+        return len(issues) == 0, issues
+
+    async def execute(
+        self,
+        validator: CustomValidator,
+        context: ValidatorContext,
+        session: AsyncSession | None = None,
+        source_id: str | None = None,
+    ) -> ValidatorResult:
+        """Execute a custom validator.
+
+        Args:
+            validator: CustomValidator model.
+            context: Validation context.
+            session: Optional database session for logging.
+            source_id: Optional source ID for logging.
+
+        Returns:
+            ValidatorResult with execution results.
+        """
+        execution_id = str(uuid.uuid4())
+        start_time = datetime.utcnow()
+
+        logger.debug(
+            f"Executing custom validator: {validator.name} "
+            f"on column: {context.column_name}"
+        )
+
+        # Create execution log if logging enabled
+        log_entry = None
+        if self.log_executions and session and validator.plugin_id:
+            log_entry = PluginExecutionLog(
+                plugin_id=validator.plugin_id,
+                validator_id=str(validator.id),
+                execution_id=execution_id,
+                source_id=source_id,
+                status="running",
+            )
+            session.add(log_entry)
+            await session.flush()
+
+        try:
+            # Prepare the code
+            wrapped_code = VALIDATOR_WRAPPER_CODE.format(user_code=validator.code)
+
+            # Execute in sandbox
+            sandbox_result = self.sandbox.execute(
+                code=wrapped_code,
+                entry_point="_execute_validator",
+                entry_args=context.to_dict(),
+            )
+
+            if sandbox_result.success and sandbox_result.result:
+                result = ValidatorResult(
+                    passed=sandbox_result.result.get("passed", False),
+                    issues=sandbox_result.result.get("issues", []),
+                    message=sandbox_result.result.get("message", ""),
+                    details=sandbox_result.result.get("details", {}),
+                    execution_time_ms=sandbox_result.execution_time_ms,
+                )
+            else:
+                result = ValidatorResult(
+                    passed=False,
+                    message=sandbox_result.error or "Execution failed",
+                    details={
+                        "stdout": sandbox_result.stdout,
+                        "stderr": sandbox_result.stderr,
+                    },
+                    execution_time_ms=sandbox_result.execution_time_ms,
+                )
+
+            # Update validator usage stats
+            validator.increment_usage()
+
+            # Update execution log
+            if log_entry:
+                if result.passed:
+                    log_entry.mark_completed(
+                        result={"passed": result.passed, "issues_count": len(result.issues)},
+                        memory_used_mb=sandbox_result.memory_used_mb,
+                    )
+                else:
+                    log_entry.mark_failed(result.message)
+
+            if session:
+                await session.flush()
+
+            return result
+
+        except Exception as e:
+            logger.error(f"Custom validator execution failed: {e}")
+
+            if log_entry:
+                log_entry.mark_failed(str(e))
+                if session:
+                    await session.flush()
+
+            return ValidatorResult(
+                passed=False,
+                message=f"Execution error: {e}",
+            )
+
+    async def test_validator(
+        self,
+        code: str,
+        parameters: list[dict[str, Any]],
+        test_data: dict[str, Any],
+        param_values: dict[str, Any] | None = None,
+    ) -> dict[str, Any]:
+        """Test a custom validator without saving.
+
+        Args:
+            code: Validator code.
+            parameters: Parameter definitions.
+            test_data: Test data (column_name and values).
+            param_values: Parameter values to use.
+
+        Returns:
+            Dictionary with test results.
+        """
+        # Validate code first
+        is_valid, issues = self.validate_validator_code(code)
+        if not is_valid:
+            return {
+                "success": False,
+                "passed": None,
+                "error": f"Code validation failed: {'; '.join(issues)}",
+                "execution_time_ms": 0,
+                "warnings": issues,
+            }
+
+        # Create context
+        context = ValidatorContext(
+            column_name=test_data.get("column_name", "test_column"),
+            column_values=test_data.get("values", []),
+            parameters=param_values or {},
+            schema=test_data.get("schema", {}),
+            row_count=len(test_data.get("values", [])),
+        )
+
+        # Execute
+        wrapped_code = VALIDATOR_WRAPPER_CODE.format(user_code=code)
+        sandbox_result = self.sandbox.execute(
+            code=wrapped_code,
+            entry_point="_execute_validator",
+            entry_args=context.to_dict(),
+        )
+
+        if sandbox_result.success and sandbox_result.result:
+            return {
+                "success": True,
+                "passed": sandbox_result.result.get("passed", False),
+                "result": sandbox_result.result,
+                "execution_time_ms": sandbox_result.execution_time_ms,
+                "warnings": sandbox_result.warnings,
+            }
+        else:
+            return {
+                "success": False,
+                "passed": None,
+                "error": sandbox_result.error,
+                "execution_time_ms": sandbox_result.execution_time_ms,
+                "warnings": sandbox_result.warnings,
+            }
+
+    def get_validator_template(self) -> str:
+        """Get a template for creating custom validators.
+
+        Returns:
+            Template code string.
+        """
+        return '''
+def validate(column_name, values, params, schema, row_count):
+    """Custom validator function.
+
+    Args:
+        column_name: Name of the column being validated.
+        values: List of values in the column.
+        params: Dictionary of parameter values.
+        schema: Column schema information.
+        row_count: Total number of rows.
+
+    Returns:
+        Dictionary with:
+            - passed: bool - Whether validation passed
+            - issues: list - List of issue dictionaries
+            - message: str - Summary message
+            - details: dict - Additional details
+    """
+    issues = []
+
+    # Example: Check for null values
+    null_count = sum(1 for v in values if v is None)
+    if null_count > 0:
+        issues.append({
+            "row": None,
+            "message": f"Found {null_count} null values",
+            "severity": "warning"
+        })
+
+    # Example: Custom validation logic
+    # threshold = params.get("threshold", 0.1)
+    # ...
+
+    return {
+        "passed": len(issues) == 0,
+        "issues": issues,
+        "message": f"Validation completed with {len(issues)} issues",
+        "details": {
+            "null_count": null_count,
+            "total_values": len(values)
+        }
+    }
+'''
+
+
+# Default executor instance
+validator_executor = CustomValidatorExecutor()

truthound_dashboard/core/plugins/versioning/__init__.py

@@ -0,0 +1,51 @@
+"""Plugin Versioning and Dependency Management Module.
+
+This module provides:
+- Semantic versioning (semver) parsing and comparison
+- Version constraint parsing (^, ~, >=, <, etc.)
+- Dependency graph with cycle detection
+- Topological sorting for load order
+"""
+
+from __future__ import annotations
+
+from .semver import (
+    Version,
+    parse_version,
+    compare_versions,
+)
+from .constraints import (
+    VersionConstraint,
+    VersionRange,
+    parse_constraint,
+    satisfies,
+    find_best_version,
+)
+from .dependencies import (
+    DependencyType,
+    Dependency,
+    DependencyGraph,
+    DependencyResolver,
+    DependencyResolutionError,
+    CyclicDependencyError,
+)
+
+__all__ = [
+    # Semver
+    "Version",
+    "parse_version",
+    "compare_versions",
+    # Constraints
+    "VersionConstraint",
+    "VersionRange",
+    "parse_constraint",
+    "satisfies",
+    "find_best_version",
+    # Dependencies
+    "DependencyType",
+    "Dependency",
+    "DependencyGraph",
+    "DependencyResolver",
+    "DependencyResolutionError",
+    "CyclicDependencyError",
+]