crackerjack 0.18.2__py3-none-any.whl → 0.45.2__py3-none-any.whl

crackerjack/mcp/state.py

@@ -0,0 +1,407 @@
+import asyncio
+import json
+import time
+import typing as t
+import uuid
+from dataclasses import asdict, dataclass
+from enum import Enum
+from pathlib import Path
+
+
+class StageStatus(str, Enum):
+    PENDING = "pending"
+    RUNNING = "running"
+    COMPLETED = "completed"
+    FAILED = "failed"
+    ERROR = "error"
+
+
+class Priority(str, Enum):
+    CRITICAL = "critical"
+    HIGH = "high"
+    MEDIUM = "medium"
+    LOW = "low"
+
+
+@dataclass
+class Issue:
+    id: str
+    type: str
+    message: str
+    file_path: str
+    line_number: int | None = None
+    priority: Priority = Priority.MEDIUM
+    stage: str = ""
+    suggested_fix: str | None = None
+    auto_fixable: bool = False
+
+    def to_dict(self) -> dict[str, t.Any]:
+        return asdict(self)
+
+
+@dataclass
+class StageResult:
+    stage: str
+    status: StageStatus
+    start_time: float
+    end_time: float | None = None
+    duration: float | None = None
+    issues_found: list[Issue] | None = None
+    fixes_applied: list[str] | None = None
+    error_message: str | None = None
+
+    def __post_init__(self) -> None:
+        if self.issues_found is None:
+            self.issues_found = []
+        if self.fixes_applied is None:
+            self.fixes_applied = []
+        if self.end_time and self.start_time:
+            self.duration = self.end_time - self.start_time
+
+    def to_dict(self) -> dict[str, t.Any]:
+        data = asdict(self)
+        data["issues_found"] = [issue.to_dict() for issue in self.issues_found or []]
+        return data
+
+
+@dataclass
+class SessionState:
+    session_id: str
+    start_time: float
+    current_stage: str | None = None
+    stages: dict[str, StageResult] | None = None
+    global_issues: list[Issue] | None = None
+    fixes_applied: list[str] | None = None
+    metadata: dict[str, t.Any] | None = None
+
+    def __post_init__(self) -> None:
+        if self.stages is None:
+            self.stages = {}
+        if self.global_issues is None:
+            self.global_issues = []
+        if self.fixes_applied is None:
+            self.fixes_applied = []
+        if self.metadata is None:
+            self.metadata = {}
+
+    def to_dict(self) -> dict[str, t.Any]:
+        data = asdict(self)
+        data["stages"] = {k: v.to_dict() for k, v in (self.stages or {}).items()}
+        data["global_issues"] = [issue.to_dict() for issue in self.global_issues or []]
+        return data
+
+
+class StateManager:
+    def __init__(
+        self, state_dir: Path | None = None, batched_saver: t.Any | None = None
+    ) -> None:
+        self._lock = asyncio.Lock()
+        self.state_dir = state_dir or Path.home() / ".cache" / "crackerjack-mcp"
+        self.state_dir.mkdir(exist_ok=True)
+        self.session_state = SessionState(
+            session_id=self._generate_session_id(),
+            start_time=time.time(),
+        )
+        self.checkpoints_dir = self.state_dir / "checkpoints"
+        self.checkpoints_dir.mkdir(exist_ok=True)
+        self._batched_saver = batched_saver
+
+    def _generate_session_id(self) -> str:
+        return str(uuid.uuid4())[:8]
+
+    async def start_stage(self, stage: str) -> None:
+        async with self._lock:
+            if not self.session_state.stages:
+                self.session_state.stages = {}
+            self.session_state.current_stage = stage
+            self.session_state.stages[stage] = StageResult(
+                stage=stage,
+                status=StageStatus.RUNNING,
+                start_time=time.time(),
+            )
+            self._save_state()
+
+    async def complete_stage(
+        self,
+        stage: str,
+        issues: list[Issue] | None = None,
+        fixes: list[str] | None = None,
+    ) -> None:
+        async with self._lock:
+            stage_result = self._get_stage_result(stage)
+            if not stage_result:
+                return
+
+            self._update_stage_completion(stage_result)
+            self._process_stage_issues(stage_result, issues)
+            self._process_stage_fixes(stage_result, fixes)
+            self._clear_current_stage(stage)
+            self._save_state()
+
+    def _get_stage_result(self, stage: str) -> StageResult | None:
+        if not self.session_state.stages or stage not in self.session_state.stages:
+            return None
+        return self.session_state.stages[stage]
+
+    def _update_stage_completion(self, stage_result: StageResult) -> None:
+        stage_result.status = StageStatus.COMPLETED
+        stage_result.end_time = time.time()
+        stage_result.duration = stage_result.end_time - stage_result.start_time
+
+    def _process_stage_issues(
+        self,
+        stage_result: StageResult,
+        issues: list[Issue] | None,
+    ) -> None:
+        if not issues:
+            return
+        stage_result.issues_found = issues
+        if not self.session_state.global_issues:
+            self.session_state.global_issues = []
+        self.session_state.global_issues.extend(issues)
+
+    def _process_stage_fixes(
+        self,
+        stage_result: StageResult,
+        fixes: list[str] | None,
+    ) -> None:
+        if not fixes:
+            return
+        stage_result.fixes_applied = fixes
+        if not self.session_state.fixes_applied:
+            self.session_state.fixes_applied = []
+        self.session_state.fixes_applied.extend(fixes)
+
+    def _clear_current_stage(self, stage: str) -> None:
+        if self.session_state.current_stage == stage:
+            self.session_state.current_stage = None
+
+    async def fail_stage(self, stage: str, error_message: str) -> None:
+        async with self._lock:
+            if not self.session_state.stages or stage not in self.session_state.stages:
+                return
+            stage_result = self.session_state.stages[stage]
+            stage_result.status = StageStatus.FAILED
+            stage_result.end_time = time.time()
+            stage_result.duration = stage_result.end_time - stage_result.start_time
+            stage_result.error_message = error_message
+            if self.session_state.current_stage == stage:
+                self.session_state.current_stage = None
+            self._save_state()
+
+    async def update_stage_status(self, stage: str, status: str) -> None:
+        async with self._lock:
+            if not self.session_state.stages:
+                self.session_state.stages = {}
+            if stage not in self.session_state.stages:
+                self.session_state.stages[stage] = StageResult(
+                    stage=stage,
+                    status=StageStatus(status),
+                    start_time=time.time(),
+                )
+            else:
+                self.session_state.stages[stage].status = StageStatus(status)
+                if status in ("completed", "failed", "error"):
+                    self.session_state.stages[stage].end_time = time.time()
+            self._save_state()
+
+    async def add_issue(self, issue: Issue) -> None:
+        async with self._lock:
+            if not self.session_state.global_issues:
+                self.session_state.global_issues = []
+            self.session_state.global_issues.append(issue)
+            self._save_state()
+
+    def remove_issue(self, issue_id: str) -> bool:
+        if not self.session_state.global_issues:
+            return False
+        initial_count = len(self.session_state.global_issues)
+        self.session_state.global_issues = [
+            issue for issue in self.session_state.global_issues if issue.id != issue_id
+        ]
+        removed = len(self.session_state.global_issues) < initial_count
+        if removed:
+            self._save_state()
+
+        return removed
+
+    def get_issues_by_priority(self, priority: Priority) -> list[Issue]:
+        if not self.session_state.global_issues:
+            return []
+
+        return [
+            issue
+            for issue in self.session_state.global_issues
+            if issue.priority == priority
+        ]
+
+    def get_issues_by_type(self, issue_type: str) -> list[Issue]:
+        if not self.session_state.global_issues:
+            return []
+
+        return [
+            issue
+            for issue in self.session_state.global_issues
+            if issue.type == issue_type
+        ]
+
+    def get_auto_fixable_issues(self) -> list[Issue]:
+        if not self.session_state.global_issues:
+            return []
+
+        return [
+            issue for issue in self.session_state.global_issues if issue.auto_fixable
+        ]
+
+    def get_session_summary(self) -> dict[str, t.Any]:
+        stages = self.session_state.stages or {}
+        issues = self.session_state.global_issues or []
+        fixes = self.session_state.fixes_applied or []
+        priority_counts = {}
+        for priority in Priority:
+            priority_counts[priority.value] = len(self.get_issues_by_priority(priority))
+        type_counts: dict[str, int] = {}
+        for issue in issues:
+            type_counts[issue.type] = type_counts.get(issue.type, 0) + 1
+        stage_status = {}
+        for stage_name, stage_result in stages.items():
+            stage_status[stage_name] = stage_result.status.value
+
+        return {
+            "session_id": self.session_state.session_id,
+            "duration": time.time() - self.session_state.start_time,
+            "current_stage": self.session_state.current_stage,
+            "stages": stage_status,
+            "total_issues": len(issues),
+            "issues_by_priority": priority_counts,
+            "issues_by_type": type_counts,
+            "total_fixes": len(fixes),
+            "auto_fixable_issues": len(self.get_auto_fixable_issues()),
+        }
+
+    async def save_checkpoint(self, name: str) -> None:
+        async with self._lock:
+            checkpoint_file = self.checkpoints_dir / f"{name}.json"
+            checkpoint_data = {
+                "name": name,
+                "timestamp": time.time(),
+                "session_state": self.session_state.to_dict(),
+            }
+            with checkpoint_file.open("w") as f:
+                json.dump(checkpoint_data, f, indent=2)
+
+    def load_checkpoint(self, name: str) -> bool:
+        checkpoint_file = self.checkpoints_dir / f"{name}.json"
+        if not checkpoint_file.exists():
+            return False
+        try:
+            with checkpoint_file.open() as f:
+                checkpoint_data = json.load(f)
+            session_data = checkpoint_data["session_state"]
+            self.session_state = SessionState(
+                session_id=session_data["session_id"],
+                start_time=session_data["start_time"],
+                current_stage=session_data.get("current_stage"),
+                metadata=session_data.get("metadata", {}),
+            )
+            stages = {}
+            for stage_name, stage_data in session_data.get("stages", {}).items():
+                issues = [
+                    Issue(**issue_data)
+                    for issue_data in stage_data.get("issues_found", [])
+                ]
+                stages[stage_name] = StageResult(
+                    stage=stage_data["stage"],
+                    status=StageStatus(stage_data["status"]),
+                    start_time=stage_data["start_time"],
+                    end_time=stage_data.get("end_time"),
+                    duration=stage_data.get("duration"),
+                    issues_found=issues,
+                    fixes_applied=stage_data.get("fixes_applied", []),
+                    error_message=stage_data.get("error_message"),
+                )
+            self.session_state.stages = stages
+            global_issues = [
+                Issue(**issue_data)
+                for issue_data in session_data.get("global_issues", [])
+            ]
+            self.session_state.global_issues = global_issues
+            self.session_state.fixes_applied = session_data.get("fixes_applied", [])
+            self._save_state()
+            return True
+        except Exception:
+            return False
+
+    def list_checkpoints(self) -> list[dict[str, t.Any]]:
+        checkpoints: list[dict[str, t.Any]] = []
+        for checkpoint_file in self.checkpoints_dir.glob(" * .json"):
+            try:
+                with checkpoint_file.open() as f:
+                    data = json.load(f)
+                checkpoints.append(
+                    {
+                        "name": data.get("name", checkpoint_file.stem),
+                        "timestamp": data.get("timestamp", 0),
+                        "file": str(checkpoint_file),
+                    },
+                )
+            except Exception:
+                continue
+        import operator
+
+        checkpoints.sort(key=operator.itemgetter("timestamp"), reverse=True)
+        return checkpoints
+
+    def start_session(self) -> None:
+        self._save_state()
+
+    def complete_session(self) -> None:
+        if not self.session_state.metadata:
+            self.session_state.metadata = {}
+        self.session_state.metadata["status"] = "completed"
+        self.session_state.metadata["completed_time"] = time.time()
+        self._save_state()
+
+    async def reset_session(self) -> None:
+        async with self._lock:
+            self.session_state = SessionState(
+                session_id=self._generate_session_id(),
+                start_time=time.time(),
+            )
+            self._save_state()
+
+    def _save_state(self) -> None:
+        if self._batched_saver:
+            save_func = self._save_state_sync
+
+            save_func()
+        else:
+            self._save_state_sync()
+
+    def _save_state_sync(self) -> None:
+        state_file = self.state_dir / "current_session.json"
+        try:
+            with state_file.open("w") as f:
+                json.dump(self.session_state.to_dict(), f, indent=2)
+        except (OSError, json.JSONEncodeError):
+            pass
+        except Exception:
+            pass
+
+    def _load_state(self) -> bool:
+        state_file = self.state_dir / "current_session.json"
+        if not state_file.exists():
+            return False
+        try:
+            with state_file.open() as f:
+                session_data = json.load(f)
+            checkpoint_data = {"session_state": session_data}
+            temp_checkpoint = self.checkpoints_dir / "_temp.json"
+            with temp_checkpoint.open("w") as f:
+                json.dump(checkpoint_data, f)
+            result = self.load_checkpoint("_temp")
+            temp_checkpoint.unlink(missing_ok=True)
+
+            return result
+        except Exception:
+            return False

crackerjack/mcp/task_manager.py

@@ -0,0 +1,259 @@
+import asyncio
+import logging
+import time
+import typing as t
+from contextlib import asynccontextmanager, suppress
+from dataclasses import dataclass
+
+from acb import console
+
+# console imported from acb
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class TaskInfo:
+    task_id: str
+    task: asyncio.Task[t.Any]
+    created_at: float
+    description: str = ""
+    timeout_seconds: float | None = None
+
+
+class AsyncTaskManager:
+    def __init__(self, max_concurrent_tasks: int = 10) -> None:
+        self.max_concurrent_tasks = max_concurrent_tasks
+        self._tasks: dict[str, TaskInfo] = {}
+        self._task_semaphore = asyncio.Semaphore(max_concurrent_tasks)
+        self._cleanup_task: asyncio.Task[t.Any] | None = None
+        self._running = False
+        self._lock = asyncio.Lock()
+
+    async def start(self) -> None:
+        self._running = True
+        self._cleanup_task = asyncio.create_task(self._cleanup_loop())
+        try:
+            console.print(
+                f"[green]🎯 Task Manager started (max {self.max_concurrent_tasks} concurrent)[ / green]",
+            )
+        except (ValueError, OSError):
+            import logging
+
+            logging.info(
+                f"Task Manager started (max {self.max_concurrent_tasks} concurrent) ",
+            )
+
+    async def stop(self) -> None:
+        self._running = False
+
+        if self._cleanup_task:
+            self._cleanup_task.cancel()
+            with suppress(asyncio.CancelledError):
+                await self._cleanup_task
+
+        await self._cancel_all_tasks()
+        try:
+            console.print("[yellow]🎯 Task Manager stopped[ / yellow]")
+        except (ValueError, OSError):
+            import logging
+
+            logging.info("Task Manager stopped")
+
+    async def create_task(
+        self,
+        coro: t.Coroutine[t.Any, t.Any, t.Any],
+        task_id: str,
+        description: str = "",
+        timeout_seconds: float | None = None,
+    ) -> asyncio.Task[t.Any]:
+        async with self._lock:
+            if task_id in self._tasks:
+                msg = f"Task {task_id} already exists"
+                raise ValueError(msg)
+
+            if len(self._tasks) >= self.max_concurrent_tasks:
+                msg = f"Maximum concurrent tasks ({self.max_concurrent_tasks}) reached"
+                raise RuntimeError(
+                    msg,
+                )
+
+        if timeout_seconds:
+            coro = asyncio.wait_for(coro, timeout=timeout_seconds)
+
+        task = asyncio.create_task(self._wrap_task(coro, task_id))
+
+        task_info = TaskInfo(
+            task_id=task_id,
+            task=task,
+            created_at=time.time(),
+            description=description,
+            timeout_seconds=timeout_seconds,
+        )
+
+        async with self._lock:
+            self._tasks[task_id] = task_info
+
+        logger.info(f"Created task {task_id}: {description}")
+        from contextlib import suppress
+
+        with suppress(ValueError, OSError):
+            console.print(f"[blue]🚀 Task {task_id} created: {description}[ / blue]")
+        return task
+
+    async def _wrap_task(
+        self, coro: t.Coroutine[t.Any, t.Any, t.Any], task_id: str
+    ) -> t.Any:
+        try:
+            async with self._task_semaphore:
+                result = await coro
+                logger.info(f"Task {task_id} completed successfully")
+                return result
+        except asyncio.CancelledError:
+            logger.info(f"Task {task_id} was cancelled")
+            raise
+        except Exception as e:
+            logger.exception(f"Task {task_id} failed: {e}")
+            from contextlib import suppress
+
+            with suppress(ValueError, OSError):
+                console.print(f"[red]❌ Task {task_id} failed: {e}[ / red]")
+            raise
+        finally:
+            async with self._lock:
+                if task_id in self._tasks:
+                    del self._tasks[task_id]
+
+    async def cancel_task(self, task_id: str) -> bool:
+        async with self._lock:
+            task_info = self._tasks.get(task_id)
+            if not task_info:
+                return False
+
+            task_info.task.cancel()
+            logger.info(f"Cancelled task {task_id}")
+            from contextlib import suppress
+
+            with suppress(ValueError, OSError):
+                console.print(f"[yellow]🚫 Task {task_id} cancelled[ / yellow]")
+            return True
+
+    async def get_task_status(self, task_id: str) -> dict[str, t.Any] | None:
+        async with self._lock:
+            task_info = self._tasks.get(task_id)
+            if not task_info:
+                return None
+
+            return {
+                "task_id": task_id,
+                "description": task_info.description,
+                "created_at": task_info.created_at,
+                "running_time": time.time() - task_info.created_at,
+                "done": task_info.task.done(),
+                "cancelled": task_info.task.cancelled(),
+                "timeout_seconds": task_info.timeout_seconds,
+            }
+
+    async def list_active_tasks(self) -> list[dict[str, t.Any]]:
+        async with self._lock:
+            tasks = []
+            for task_id, task_info in self._tasks.items():
+                task_status = {
+                    "task_id": task_id,
+                    "description": task_info.description,
+                    "created_at": task_info.created_at,
+                    "running_time": time.time() - task_info.created_at,
+                    "done": task_info.task.done(),
+                    "cancelled": task_info.task.cancelled(),
+                    "timeout_seconds": task_info.timeout_seconds,
+                }
+                tasks.append(task_status)
+            return tasks
+
+    async def wait_for_task(self, task_id: str, timeout: float | None = None) -> t.Any:
+        async with self._lock:
+            task_info = self._tasks.get(task_id)
+            if not task_info:
+                msg = f"Task {task_id} not found"
+                raise ValueError(msg)
+
+        try:
+            if timeout:
+                return await asyncio.wait_for(task_info.task, timeout=timeout)
+            return await task_info.task
+        except TimeoutError:
+            logger.warning(f"Timeout waiting for task {task_id}")
+            raise
+
+    @asynccontextmanager
+    async def managed_task(
+        self,
+        coro: t.Coroutine[t.Any, t.Any, t.Any],
+        task_id: str,
+        description: str = "",
+        timeout_seconds: float | None = None,
+    ) -> t.AsyncGenerator[asyncio.Task[t.Any]]:
+        task = await self.create_task(coro, task_id, description, timeout_seconds)
+        try:
+            yield task
+        finally:
+            if not task.done():
+                task.cancel()
+                with suppress(asyncio.CancelledError):
+                    await task
+
+    async def _cancel_all_tasks(self) -> None:
+        async with self._lock:
+            tasks_to_cancel = list[t.Any](self._tasks.values())
+
+        if not tasks_to_cancel:
+            return
+
+        try:
+            console.print(
+                f"[yellow]🧹 Cancelling {len(tasks_to_cancel)} running tasks[ / yellow]",
+            )
+        except (ValueError, OSError):
+            import logging
+
+            logging.info(f"Cancelling {len(tasks_to_cancel)} running tasks")
+
+        for task_info in tasks_to_cancel:
+            task_info.task.cancel()
+
+        await asyncio.gather(
+            *[task_info.task for task_info in tasks_to_cancel],
+            return_exceptions=True,
+        )
+
+        async with self._lock:
+            self._tasks.clear()
+
+    async def _cleanup_loop(self) -> None:
+        while self._running:
+            try:
+                await self._cleanup_completed_tasks()
+                await asyncio.sleep(30)
+            except asyncio.CancelledError:
+                break
+            except Exception as e:
+                logger.exception(f"Error in task cleanup loop: {e}")
+                await asyncio.sleep(5)
+
+    async def _cleanup_completed_tasks(self) -> None:
+        async with self._lock:
+            completed_tasks = []
+            for task_id, task_info in list[t.Any](self._tasks.items()):
+                if task_info.task.done():
+                    completed_tasks.append(task_id)
+                    del self._tasks[task_id]
+
+        if completed_tasks:
+            logger.info(f"Cleaned up {len(completed_tasks)} completed tasks")
+
+    def get_stats(self) -> dict[str, t.Any]:
+        return {
+            "running": self._running,
+            "active_tasks": len(self._tasks),
+            "max_concurrent_tasks": self.max_concurrent_tasks,
+            "available_slots": self._task_semaphore._value,
+        }

crackerjack/mcp/tools/README.md

@@ -0,0 +1,27 @@
+# MCP Tools
+
+MCP tool definitions and adapters for the Crackerjack MCP server.
+
+## Tool Categories
+
+- **`execution_tools.py`** - Execute crackerjack workflows, get job progress
+- **`monitoring_tools.py`** - System health, status reporting, workflow monitoring
+- **`progress_tools.py`** - Real-time progress tracking for test execution
+- **`semantic_tools.py`** - Semantic search, file indexing, embeddings
+- **`intelligence_tools.py`** - AI agent selection, recommendations, performance analysis
+- **`error_analyzer.py`** - Smart error analysis and pattern detection
+- **`proactive_tools.py`** - Proactive monitoring and preventive quality checks
+- **`utility_tools.py`** - Helper utilities and common operations
+- **`workflow_executor.py`** - Workflow orchestration and execution management
+
+## Core Tools
+
+The MCP server provides these primary capabilities:
+
+1. **Workflow Execution** - Run crackerjack quality checks via MCP
+1. **Job Tracking** - Monitor async job progress and status
+1. **Error Analysis** - Intelligent error detection and root cause analysis
+1. **Semantic Search** - Code search with vector embeddings
+1. **Agent Intelligence** - Smart agent selection based on task context
+
+See parent `mcp/README.md` for server architecture details.