crackerjack 0.32.0__py3-none-any.whl → 0.33.0__py3-none-any.whl

crackerjack/services/parallel_executor.py

@@ -0,0 +1,474 @@
+"""Parallel execution coordinator for hooks and other independent operations.
+
+This module provides safe parallel execution of hooks while respecting
+security levels, dependencies, and resource constraints.
+"""
+
+import asyncio
+import time
+import typing as t
+from concurrent.futures import ThreadPoolExecutor
+from dataclasses import dataclass, field
+from enum import Enum
+from pathlib import Path
+
+from crackerjack.config.hooks import HookDefinition, SecurityLevel
+from crackerjack.services.logging import get_logger
+from crackerjack.services.performance_cache import get_performance_cache
+
+
+class ExecutionStrategy(Enum):
+    """Execution strategy for parallel operations."""
+
+    SEQUENTIAL = "sequential"
+    PARALLEL_SAFE = (
+        "parallel_safe"  # Only parallel execution of safe, independent hooks
+    )
+    PARALLEL_AGGRESSIVE = "parallel_aggressive"  # More aggressive parallelization
+
+
+@dataclass
+class ExecutionGroup:
+    """Group of operations that can be executed together."""
+
+    name: str
+    operations: list[t.Any]
+    max_workers: int = 3
+    timeout_seconds: int = 300
+    dependencies: set[str] = field(default_factory=set)
+    security_level: SecurityLevel = SecurityLevel.MEDIUM
+
+
+@dataclass
+class ExecutionResult:
+    """Result of executing an operation."""
+
+    operation_id: str
+    success: bool
+    duration_seconds: float
+    output: str = ""
+    error: str = ""
+    exit_code: int = 0
+    metadata: dict[str, t.Any] = field(default_factory=dict)
+
+
+@dataclass
+class ParallelExecutionResult:
+    """Result of parallel execution batch."""
+
+    group_name: str
+    total_operations: int
+    successful_operations: int
+    failed_operations: int
+    total_duration_seconds: float
+    results: list[ExecutionResult]
+
+    @property
+    def success_rate(self) -> float:
+        """Calculate success rate."""
+        return (
+            self.successful_operations / self.total_operations
+            if self.total_operations > 0
+            else 0.0
+        )
+
+    @property
+    def overall_success(self) -> bool:
+        """Check if all operations succeeded."""
+        return self.failed_operations == 0
+
+
+class ParallelHookExecutor:
+    """Parallel executor for hook operations with safety constraints."""
+
+    def __init__(
+        self,
+        max_workers: int = 3,
+        timeout_seconds: int = 300,
+        strategy: ExecutionStrategy = ExecutionStrategy.PARALLEL_SAFE,
+    ):
+        self.max_workers = max_workers
+        self.timeout_seconds = timeout_seconds
+        self.strategy = strategy
+        self._logger = get_logger("crackerjack.parallel_executor")
+        self._cache = get_performance_cache()
+
+    def analyze_hook_dependencies(
+        self,
+        hooks: list[HookDefinition],
+    ) -> dict[str, list[HookDefinition]]:
+        """Analyze hook dependencies and group hooks for parallel execution."""
+        groups: dict[str, list[HookDefinition]] = {
+            "formatting": [],
+            "validation": [],
+            "security": [],
+            "comprehensive": [],
+        }
+
+        for hook in hooks:
+            if hook.is_formatting or hook.security_level == SecurityLevel.LOW:
+                groups["formatting"].append(hook)
+            elif hook.security_level == SecurityLevel.CRITICAL:
+                groups["security"].append(hook)
+            elif hook.name in {"check-yaml", "check-json", "check-toml"}:
+                groups["validation"].append(hook)
+            else:
+                groups["comprehensive"].append(hook)
+
+        # Remove empty groups
+        return {k: v for k, v in groups.items() if v}
+
+    def can_execute_in_parallel(
+        self,
+        hook1: HookDefinition,
+        hook2: HookDefinition,
+    ) -> bool:
+        """Check if two hooks can be safely executed in parallel."""
+        # Never parallelize different security levels for safety
+        if hook1.security_level != hook2.security_level:
+            return False
+
+        # Don't parallelize hooks that modify files with hooks that read files
+        if hook1.is_formatting and not hook2.is_formatting:
+            return False
+
+        # Safe combinations for parallel execution
+        safe_parallel_combinations = [
+            # All formatting hooks can run in parallel
+            (lambda h: h.is_formatting, lambda h: h.is_formatting),
+            # Validation hooks can run in parallel
+            (
+                lambda h: h.name in {"check-yaml", "check-json", "check-toml"},
+                lambda h: h.name in {"check-yaml", "check-json", "check-toml"},
+            ),
+            # Same security level non-formatting hooks
+            (
+                lambda h: not h.is_formatting
+                and h.security_level == SecurityLevel.MEDIUM,
+                lambda h: not h.is_formatting
+                and h.security_level == SecurityLevel.MEDIUM,
+            ),
+        ]
+
+        for check1, check2 in safe_parallel_combinations:
+            if check1(hook1) and check2(hook2):
+                return True
+
+        return False
+
+    async def execute_hooks_parallel(
+        self,
+        hooks: list[HookDefinition],
+        hook_runner: t.Callable[[HookDefinition], t.Awaitable[ExecutionResult]],
+    ) -> ParallelExecutionResult:
+        """Execute hooks in parallel where safe, sequential otherwise."""
+        start_time = time.time()
+
+        if self.strategy == ExecutionStrategy.SEQUENTIAL:
+            return await self._execute_sequential(hooks, hook_runner, start_time)
+
+        # Group hooks for parallel execution
+        groups = self.analyze_hook_dependencies(hooks)
+        all_results: list[ExecutionResult] = []
+
+        self._logger.info(
+            f"Executing {len(hooks)} hooks in {len(groups)} parallel groups"
+        )
+
+        for group_name, group_hooks in groups.items():
+            if len(group_hooks) == 1 or not self._can_parallelize_group(group_hooks):
+                # Execute single hook or non-parallelizable group sequentially
+                for hook in group_hooks:
+                    result = await hook_runner(hook)
+                    all_results.append(result)
+            else:
+                # Execute group in parallel
+                group_results = await self._execute_group_parallel(
+                    group_hooks,
+                    hook_runner,
+                    group_name,
+                )
+                all_results.extend(group_results)
+
+        total_duration = time.time() - start_time
+        successful = sum(1 for r in all_results if r.success)
+        failed = len(all_results) - successful
+
+        return ParallelExecutionResult(
+            group_name="all_hooks",
+            total_operations=len(hooks),
+            successful_operations=successful,
+            failed_operations=failed,
+            total_duration_seconds=total_duration,
+            results=all_results,
+        )
+
+    async def _execute_sequential(
+        self,
+        hooks: list[HookDefinition],
+        hook_runner: t.Callable[[HookDefinition], t.Awaitable[ExecutionResult]],
+        start_time: float,
+    ) -> ParallelExecutionResult:
+        """Execute hooks sequentially (fallback)."""
+        results: list[ExecutionResult] = []
+
+        for hook in hooks:
+            result = await hook_runner(hook)
+            results.append(result)
+
+        total_duration = time.time() - start_time
+        successful = sum(1 for r in results if r.success)
+        failed = len(results) - successful
+
+        return ParallelExecutionResult(
+            group_name="sequential",
+            total_operations=len(hooks),
+            successful_operations=successful,
+            failed_operations=failed,
+            total_duration_seconds=total_duration,
+            results=results,
+        )
+
+    def _can_parallelize_group(self, hooks: list[HookDefinition]) -> bool:
+        """Check if a group of hooks can be parallelized."""
+        if len(hooks) < 2:
+            return False
+
+        # Check pairwise compatibility
+        for i, hook1 in enumerate(hooks):
+            for hook2 in hooks[i + 1 :]:
+                if not self.can_execute_in_parallel(hook1, hook2):
+                    return False
+
+        return True
+
+    async def _execute_group_parallel(
+        self,
+        hooks: list[HookDefinition],
+        hook_runner: t.Callable[[HookDefinition], t.Awaitable[ExecutionResult]],
+        group_name: str,
+    ) -> list[ExecutionResult]:
+        """Execute a group of hooks in parallel."""
+        self._logger.debug(f"Executing {len(hooks)} {group_name} hooks in parallel")
+
+        # Limit parallelism for safety
+        max_workers = min(self.max_workers, len(hooks))
+
+        # Create semaphore to limit concurrent executions
+        semaphore = asyncio.Semaphore(max_workers)
+
+        async def run_with_semaphore(hook: HookDefinition) -> ExecutionResult:
+            async with semaphore:
+                return await hook_runner(hook)
+
+        # Execute all hooks concurrently with limited parallelism
+        tasks = [run_with_semaphore(hook) for hook in hooks]
+        results = await asyncio.gather(*tasks, return_exceptions=True)
+
+        # Process results and handle exceptions
+        processed_results: list[ExecutionResult] = []
+        for i, result in enumerate(results):
+            if isinstance(result, Exception):
+                # Create error result for exception
+                error_result = ExecutionResult(
+                    operation_id=hooks[i].name,
+                    success=False,
+                    duration_seconds=0.0,
+                    error=str(result),
+                )
+                processed_results.append(error_result)
+                self._logger.error(
+                    f"Hook {hooks[i].name} failed with exception: {result}"
+                )
+            else:
+                # result must be ExecutionResult here due to type narrowing
+                processed_results.append(result)  # type: ignore[arg-type]
+
+        successful = sum(1 for r in processed_results if r.success)
+        self._logger.info(
+            f"Parallel {group_name} execution: {successful}/{len(hooks)} succeeded"
+        )
+
+        return processed_results
+
+
+class AsyncCommandExecutor:
+    """Asynchronous command executor with optimization and caching."""
+
+    def __init__(
+        self,
+        max_workers: int = 4,
+        cache_results: bool = True,
+    ):
+        self.max_workers = max_workers
+        self.cache_results = cache_results
+        self._logger = get_logger("crackerjack.async_executor")
+        self._cache = get_performance_cache()
+        self._thread_pool = ThreadPoolExecutor(max_workers=max_workers)
+
+    async def execute_command(
+        self,
+        command: list[str],
+        cwd: Path | None = None,
+        timeout: int = 60,
+        cache_ttl: int = 120,
+    ) -> ExecutionResult:
+        """Execute a command asynchronously with caching."""
+        # Check cache first
+        if self.cache_results:
+            cached_result = await self._get_cached_result(command, cwd)
+            if cached_result:
+                self._logger.debug(
+                    f"Using cached result for command: {' '.join(command)}"
+                )
+                return cached_result
+
+        # Execute command
+        start_time = time.time()
+        result = await self._run_command_async(command, cwd, timeout)
+        result.duration_seconds = time.time() - start_time
+
+        # Cache successful results
+        if self.cache_results and result.success:
+            await self._cache_result(command, result, cache_ttl, cwd)
+
+        return result
+
+    async def execute_commands_batch(
+        self,
+        commands: list[tuple[list[str], Path | None]],
+        timeout: int = 60,
+    ) -> list[ExecutionResult]:
+        """Execute multiple commands concurrently."""
+        self._logger.info(f"Executing {len(commands)} commands in batch")
+
+        tasks = [self.execute_command(cmd, cwd, timeout) for cmd, cwd in commands]
+
+        results = await asyncio.gather(*tasks, return_exceptions=True)
+
+        # Process results and handle exceptions
+        processed_results: list[ExecutionResult] = []
+        for i, result in enumerate(results):
+            if isinstance(result, Exception):
+                error_result = ExecutionResult(
+                    operation_id=f"command_{i}",
+                    success=False,
+                    duration_seconds=0.0,
+                    error=str(result),
+                )
+                processed_results.append(error_result)
+            else:
+                # result must be ExecutionResult here due to type narrowing
+                processed_results.append(result)  # type: ignore[arg-type]
+
+        successful = sum(1 for r in processed_results if r.success)
+        self._logger.info(
+            f"Batch execution: {successful}/{len(commands)} commands succeeded"
+        )
+
+        return processed_results
+
+    async def _run_command_async(
+        self,
+        command: list[str],
+        cwd: Path | None = None,
+        timeout: int = 60,
+    ) -> ExecutionResult:
+        """Run command asynchronously in thread pool."""
+        loop = asyncio.get_event_loop()
+
+        def run_sync_command() -> ExecutionResult:
+            import subprocess
+
+            try:
+                result = subprocess.run(
+                    command,
+                    cwd=cwd,
+                    capture_output=True,
+                    text=True,
+                    timeout=timeout,
+                    check=False,
+                )
+
+                return ExecutionResult(
+                    operation_id=" ".join(command),
+                    success=result.returncode == 0,
+                    duration_seconds=0.0,  # Set by caller
+                    output=result.stdout,
+                    error=result.stderr,
+                    exit_code=result.returncode,
+                )
+
+            except subprocess.TimeoutExpired:
+                return ExecutionResult(
+                    operation_id=" ".join(command),
+                    success=False,
+                    duration_seconds=timeout,
+                    error=f"Command timeout after {timeout}s",
+                    exit_code=-1,
+                )
+            except Exception as e:
+                return ExecutionResult(
+                    operation_id=" ".join(command),
+                    success=False,
+                    duration_seconds=0.0,
+                    error=str(e),
+                    exit_code=-1,
+                )
+
+        return await loop.run_in_executor(self._thread_pool, run_sync_command)
+
+    async def _get_cached_result(
+        self,
+        command: list[str],
+        cwd: Path | None = None,
+    ) -> ExecutionResult | None:
+        """Get cached command result."""
+        from crackerjack.services.performance_cache import get_command_cache
+
+        return get_command_cache().get_command_result(command, cwd)
+
+    async def _cache_result(
+        self,
+        command: list[str],
+        result: ExecutionResult,
+        ttl_seconds: int,
+        cwd: Path | None = None,
+    ) -> None:
+        """Cache command result."""
+        from crackerjack.services.performance_cache import get_command_cache
+
+        command_cache = get_command_cache()
+        command_cache.set_command_result(command, result, cwd, ttl_seconds)
+
+    def __del__(self):
+        """Clean up thread pool."""
+        if hasattr(self, "_thread_pool"):
+            self._thread_pool.shutdown(wait=False)
+
+
+# Global executor instances
+_parallel_executor: ParallelHookExecutor | None = None
+_async_executor: AsyncCommandExecutor | None = None
+
+
+def get_parallel_executor(
+    max_workers: int = 3,
+    strategy: ExecutionStrategy = ExecutionStrategy.PARALLEL_SAFE,
+) -> ParallelHookExecutor:
+    """Get global parallel hook executor instance."""
+    global _parallel_executor
+    if _parallel_executor is None:
+        _parallel_executor = ParallelHookExecutor(
+            max_workers=max_workers,
+            strategy=strategy,
+        )
+    return _parallel_executor
+
+
+def get_async_executor(max_workers: int = 4) -> AsyncCommandExecutor:
+    """Get global async command executor instance."""
+    global _async_executor
+    if _async_executor is None:
+        _async_executor = AsyncCommandExecutor(max_workers=max_workers)
+    return _async_executor