OCR/backend/app/routers/tasks.py

"""
Tool_OCR - Task Management Router
Handles OCR task operations with user isolation
"""

import logging
from typing import Optional
from pathlib import Path
import shutil
import hashlib

from fastapi import APIRouter, Depends, HTTPException, status, Query, UploadFile, File, BackgroundTasks
from fastapi.responses import FileResponse
from sqlalchemy.orm import Session
import json
from datetime import datetime

from app.core.deps import get_db, get_current_user
from app.core.config import settings
from app.models.user import User
from app.models.task import TaskStatus, TaskFile
from app.schemas.task import (
    TaskCreate,
    TaskUpdate,
    TaskResponse,
    TaskDetailResponse,
    TaskListResponse,
    TaskStatsResponse,
    TaskStatusEnum,
    UploadResponse,
    ProcessingTrackEnum,
    ProcessingOptions,
    AnalyzeRequest,
    DocumentAnalysisResponse,
    ProcessingMetadata,
    TaskResponseWithMetadata,
    ExportOptions,
    PreprocessingModeEnum,
    PreprocessingConfig,
    PreprocessingPreviewRequest,
    PreprocessingPreviewResponse,
    ImageQualityMetrics,
    TableDetectionConfig,
)
from app.services.task_service import task_service
from app.services.file_access_service import file_access_service
from app.services.ocr_service import OCRService
from app.services.service_pool import get_service_pool, PoolConfig

# Import dual-track components
try:
    from app.services.document_type_detector import DocumentTypeDetector
    DUAL_TRACK_AVAILABLE = True
except ImportError:
    DUAL_TRACK_AVAILABLE = False

# Service pool availability
SERVICE_POOL_AVAILABLE = True
try:
    from app.services.memory_manager import get_model_manager
except ImportError:
    SERVICE_POOL_AVAILABLE = False

logger = logging.getLogger(__name__)

router = APIRouter(prefix="/api/v2/tasks", tags=["Tasks"])


def process_task_ocr(
    task_id: str,
    task_db_id: int,
    file_path: str,
    filename: str,
    use_dual_track: bool = True,
    force_track: Optional[str] = None,
    language: str = 'ch',
    layout_model: Optional[str] = "chinese",
    preprocessing_mode: Optional[str] = "auto",
    preprocessing_config: Optional[dict] = None,
    table_detection_config: Optional[dict] = None
):
    """
    Background task to process OCR for a task with dual-track support.

    Uses OCRServicePool to acquire a shared service instance instead of
    creating a new one, preventing GPU memory proliferation.

    Args:
        task_id: Task UUID string
        task_db_id: Task database ID
        file_path: Path to uploaded file
        filename: Original filename
        use_dual_track: Enable dual-track processing
        force_track: Force specific track ('ocr' or 'direct')
        language: OCR language code
        layout_model: Layout detection model ('chinese', 'default', 'cdla')
        preprocessing_mode: Preprocessing mode ('auto', 'manual', 'disabled')
        preprocessing_config: Manual preprocessing config dict (contrast, sharpen, binarize)
        table_detection_config: Table detection config dict (enable_wired_table, enable_wireless_table, enable_region_detection)
    """
    from app.core.database import SessionLocal
    from app.models.task import Task

    db = SessionLocal()
    start_time = datetime.now()
    pooled_service = None

    try:
        logger.info(f"Starting OCR processing for task {task_id}, file: {filename}")
        logger.info(f"Processing options: dual_track={use_dual_track}, force_track={force_track}, lang={language}")
        logger.info(f"Preprocessing options: mode={preprocessing_mode}, config={preprocessing_config}")
        logger.info(f"Table detection options: {table_detection_config}")

        # Convert preprocessing parameters to proper types
        preprocess_mode_enum = None
        preprocess_config_obj = None
        if preprocessing_mode:
            preprocess_mode_enum = PreprocessingModeEnum(preprocessing_mode)
        if preprocessing_config:
            from app.schemas.task import PreprocessingContrastEnum
            preprocess_config_obj = PreprocessingConfig(
                contrast=PreprocessingContrastEnum(preprocessing_config.get("contrast", "clahe")),
                contrast_strength=preprocessing_config.get("contrast_strength", 1.0),
                sharpen=preprocessing_config.get("sharpen", True),
                sharpen_strength=preprocessing_config.get("sharpen_strength", 1.0),
                binarize=preprocessing_config.get("binarize", False)
            )

        # Convert table detection config to object
        table_det_config_obj = None
        if table_detection_config:
            table_det_config_obj = TableDetectionConfig(
                enable_wired_table=table_detection_config.get("enable_wired_table", True),
                enable_wireless_table=table_detection_config.get("enable_wireless_table", True),
                enable_region_detection=table_detection_config.get("enable_region_detection", True)
            )

        # Get task directly by database ID (bypass user isolation for background task)
        task = db.query(Task).filter(Task.id == task_db_id).first()
        if not task:
            logger.error(f"Task {task_id} not found in database")
            return

        # Acquire OCR service from pool (or create new if pool disabled)
        ocr_service = None
        if settings.enable_service_pool and SERVICE_POOL_AVAILABLE:
            try:
                service_pool = get_service_pool()
                pooled_service = service_pool.acquire(
                    device="GPU:0",
                    timeout=settings.service_acquire_timeout_seconds,
                    task_id=task_id
                )
                if pooled_service:
                    ocr_service = pooled_service.service
                    logger.info(f"Acquired OCR service from pool for task {task_id}")
                else:
                    logger.warning(f"Timeout acquiring service from pool, creating new instance")
            except Exception as e:
                logger.warning(f"Failed to acquire service from pool: {e}, creating new instance")

        # Fallback: create new instance if pool acquisition failed
        if ocr_service is None:
            logger.info("Creating new OCRService instance (pool disabled or unavailable)")
            ocr_service = OCRService()

        # Create result directory before OCR processing (needed for saving extracted images)
        result_dir = Path(settings.result_dir) / task_id
        result_dir.mkdir(parents=True, exist_ok=True)

        # Process the file with OCR
        # Use dual-track processing if:
        # 1. use_dual_track is True (auto-detection)
        # 2. OR force_track is specified (explicit track selection)
        if (use_dual_track or force_track) and hasattr(ocr_service, 'process'):
            # Use new dual-track processing (or forced track)
            ocr_result = ocr_service.process(
                file_path=Path(file_path),
                lang=language,
                detect_layout=True,
                output_dir=result_dir,
                use_dual_track=use_dual_track,
                force_track=force_track,
                layout_model=layout_model,
                preprocessing_mode=preprocess_mode_enum,
                preprocessing_config=preprocess_config_obj,
                table_detection_config=table_det_config_obj
            )
        else:
            # Fall back to traditional processing (no force_track support)
            ocr_result = ocr_service.process_image(
                image_path=Path(file_path),
                lang=language,
                detect_layout=True,
                output_dir=result_dir,
                layout_model=layout_model,
                preprocessing_mode=preprocess_mode_enum,
                preprocessing_config=preprocess_config_obj,
                table_detection_config=table_det_config_obj
            )

        # Calculate processing time
        processing_time_ms = int((datetime.now() - start_time).total_seconds() * 1000)

        # Save results using OCR service (includes JSON, Markdown, and PDF generation)
        json_path, markdown_path, pdf_path = ocr_service.save_results(
            result=ocr_result,
            output_dir=result_dir,
            file_id=Path(filename).stem,
            source_file_path=Path(file_path)
        )

        # Release service back to pool (success case)
        if pooled_service:
            try:
                service_pool = get_service_pool()
                service_pool.release(pooled_service, error=None)
                logger.info(f"Released OCR service back to pool for task {task_id}")
                pooled_service = None  # Prevent double release in finally
            except Exception as e:
                logger.warning(f"Failed to release service to pool: {e}")

        # Close old session and create fresh one to avoid MySQL timeout
        # (long OCR processing may cause connection to become stale)
        db.close()
        db = SessionLocal()

        # Re-fetch task with fresh connection
        task = db.query(Task).filter(Task.id == task_db_id).first()
        if not task:
            logger.error(f"Task {task_id} not found after OCR processing")
            return

        # Update task with results (direct database update)
        task.result_json_path = str(json_path) if json_path else None
        task.result_markdown_path = str(markdown_path) if markdown_path else None
        task.result_pdf_path = str(pdf_path) if pdf_path else None
        task.processing_time_ms = processing_time_ms
        task.status = TaskStatus.COMPLETED
        task.completed_at = datetime.utcnow()
        task.updated_at = datetime.utcnow()

        db.commit()

        logger.info(f"OCR processing completed for task {task_id} in {processing_time_ms}ms")

    except Exception as e:
        logger.exception(f"OCR processing failed for task {task_id}")

        # Release service back to pool with error
        if pooled_service:
            try:
                service_pool = get_service_pool()
                service_pool.release(pooled_service, error=e)
                pooled_service = None
            except Exception as release_error:
                logger.warning(f"Failed to release service to pool: {release_error}")

        # Update task status to failed (direct database update)
        try:
            task = db.query(Task).filter(Task.id == task_db_id).first()
            if task:
                task.status = TaskStatus.FAILED
                task.error_message = str(e)
                task.updated_at = datetime.utcnow()
                db.commit()
        except Exception as update_error:
            logger.error(f"Failed to update task status: {update_error}")

    finally:
        # Ensure service is released in case of any missed release
        if pooled_service:
            try:
                service_pool = get_service_pool()
                service_pool.release(pooled_service, error=None)
            except Exception:
                pass
        db.close()


@router.post("/", response_model=TaskResponse, status_code=status.HTTP_201_CREATED)
async def create_task(
    task_data: TaskCreate,
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Create a new OCR task

    - **filename**: Original filename (optional)
    - **file_type**: File MIME type (optional)
    """
    try:
        task = task_service.create_task(
            db=db,
            user_id=current_user.id,
            filename=task_data.filename,
            file_type=task_data.file_type
        )

        logger.info(f"Created task {task.task_id} for user {current_user.email}")
        return task

    except Exception as e:
        logger.exception(f"Failed to create task for user {current_user.id}")
        raise HTTPException(
            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
            detail=f"Failed to create task: {str(e)}"
        )


@router.get("/", response_model=TaskListResponse)
async def list_tasks(
    status_filter: Optional[TaskStatusEnum] = Query(None, alias="status"),
    filename_search: Optional[str] = Query(None, alias="filename"),
    date_from: Optional[str] = Query(None, alias="date_from"),
    date_to: Optional[str] = Query(None, alias="date_to"),
    page: int = Query(1, ge=1),
    page_size: int = Query(50, ge=1, le=100),
    order_by: str = Query("created_at"),
    order_desc: bool = Query(True),
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    List user's tasks with pagination and filtering

    - **status**: Filter by task status (optional)
    - **filename**: Search by filename (partial match, optional)
    - **date_from**: Filter tasks from this date (YYYY-MM-DD, optional)
    - **date_to**: Filter tasks until this date (YYYY-MM-DD, optional)
    - **page**: Page number (starts from 1)
    - **page_size**: Number of tasks per page (max 100)
    - **order_by**: Sort field (created_at, updated_at, completed_at)
    - **order_desc**: Sort descending (default: true)
    """
    try:
        # Convert enum to model enum if provided
        status_enum = TaskStatus[status_filter.value.upper()] if status_filter else None

        # Parse date strings
        from datetime import datetime
        date_from_dt = datetime.fromisoformat(date_from) if date_from else None
        date_to_dt = datetime.fromisoformat(date_to) if date_to else None

        # Calculate offset
        skip = (page - 1) * page_size

        # Get tasks
        tasks, total = task_service.get_user_tasks(
            db=db,
            user_id=current_user.id,
            status=status_enum,
            filename_search=filename_search,
            date_from=date_from_dt,
            date_to=date_to_dt,
            skip=skip,
            limit=page_size,
            order_by=order_by,
            order_desc=order_desc
        )

        # Calculate pagination
        has_more = (skip + len(tasks)) < total

        return {
            "tasks": tasks,
            "total": total,
            "page": page,
            "page_size": page_size,
            "has_more": has_more
        }

    except Exception as e:
        logger.exception(f"Failed to list tasks for user {current_user.id}")
        raise HTTPException(
            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
            detail=f"Failed to list tasks: {str(e)}"
        )


@router.get("/stats", response_model=TaskStatsResponse)
async def get_task_stats(
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Get task statistics for current user

    Returns counts by status
    """
    try:
        stats = task_service.get_user_stats(db=db, user_id=current_user.id)
        return stats

    except Exception as e:
        logger.exception(f"Failed to get stats for user {current_user.id}")
        raise HTTPException(
            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
            detail=f"Failed to get statistics: {str(e)}"
        )


@router.get("/{task_id}", response_model=TaskDetailResponse)
async def get_task(
    task_id: str,
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Get task details by ID

    - **task_id**: Task UUID
    """
    task = task_service.get_task_by_id(
        db=db,
        task_id=task_id,
        user_id=current_user.id
    )

    if not task:
        raise HTTPException(
            status_code=status.HTTP_404_NOT_FOUND,
            detail="Task not found"
        )

    # Extract processing_track from result JSON metadata if available
    processing_track = None
    if task.result_json_path:
        try:
            import json
            from pathlib import Path
            result_path = Path(task.result_json_path)
            if result_path.exists():
                with open(result_path) as f:
                    result_data = json.load(f)
                    metadata = result_data.get("metadata", {})
                    track_str = metadata.get("processing_track")
                    # Convert string to enum to avoid Pydantic serialization warning
                    if track_str:
                        try:
                            processing_track = ProcessingTrackEnum(track_str)
                        except ValueError:
                            processing_track = None
        except Exception:
            pass  # Silently ignore errors reading the result file

    # Create response with processing_track
    response = TaskDetailResponse.model_validate(task)
    response.processing_track = processing_track
    return response


@router.patch("/{task_id}", response_model=TaskResponse)
async def update_task(
    task_id: str,
    task_update: TaskUpdate,
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Update task status and results

    - **task_id**: Task UUID
    - **status**: New task status (optional)
    - **error_message**: Error message if failed (optional)
    - **processing_time_ms**: Processing time in milliseconds (optional)
    - **result_json_path**: Path to JSON result (optional)
    - **result_markdown_path**: Path to Markdown result (optional)
    - **result_pdf_path**: Path to searchable PDF (optional)
    """
    try:
        # Update status if provided
        if task_update.status:
            status_enum = TaskStatus[task_update.status.value.upper()]
            task = task_service.update_task_status(
                db=db,
                task_id=task_id,
                user_id=current_user.id,
                status=status_enum,
                error_message=task_update.error_message,
                processing_time_ms=task_update.processing_time_ms
            )

            if not task:
                raise HTTPException(
                    status_code=status.HTTP_404_NOT_FOUND,
                    detail="Task not found"
                )

        # Update result paths if provided
        if any([
            task_update.result_json_path,
            task_update.result_markdown_path,
            task_update.result_pdf_path
        ]):
            task = task_service.update_task_results(
                db=db,
                task_id=task_id,
                user_id=current_user.id,
                result_json_path=task_update.result_json_path,
                result_markdown_path=task_update.result_markdown_path,
                result_pdf_path=task_update.result_pdf_path
            )

            if not task:
                raise HTTPException(
                    status_code=status.HTTP_404_NOT_FOUND,
                    detail="Task not found"
                )

        return task

    except HTTPException:
        raise
    except Exception as e:
        logger.exception(f"Failed to update task {task_id}")
        raise HTTPException(
            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
            detail=f"Failed to update task: {str(e)}"
        )


@router.delete("/{task_id}", status_code=status.HTTP_204_NO_CONTENT)
async def delete_task(
    task_id: str,
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Delete a task

    - **task_id**: Task UUID
    """
    success = task_service.delete_task(
        db=db,
        task_id=task_id,
        user_id=current_user.id
    )

    if not success:
        raise HTTPException(
            status_code=status.HTTP_404_NOT_FOUND,
            detail="Task not found"
        )

    logger.info(f"Deleted task {task_id} for user {current_user.email}")
    return None


@router.get("/{task_id}/download/json", summary="Download JSON result")
async def download_json(
    task_id: str,
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Download task result as JSON file

    - **task_id**: Task UUID
    """
    # Get task
    task = task_service.get_task_by_id(
        db=db,
        task_id=task_id,
        user_id=current_user.id
    )

    if not task:
        raise HTTPException(
            status_code=status.HTTP_404_NOT_FOUND,
            detail="Task not found"
        )

    # Validate file access
    is_valid, error_msg = file_access_service.validate_file_access(
        db=db,
        user_id=current_user.id,
        task_id=task_id,
        file_path=task.result_json_path
    )

    if not is_valid:
        raise HTTPException(
            status_code=status.HTTP_403_FORBIDDEN,
            detail=error_msg
        )

    # Return file
    filename = f"{task.filename or task_id}_result.json"
    return FileResponse(
        path=task.result_json_path,
        filename=filename,
        media_type="application/json"
    )


@router.get("/{task_id}/download/markdown", summary="Download Markdown result")
async def download_markdown(
    task_id: str,
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Download task result as Markdown file

    - **task_id**: Task UUID
    """
    # Get task
    task = task_service.get_task_by_id(
        db=db,
        task_id=task_id,
        user_id=current_user.id
    )

    if not task:
        raise HTTPException(
            status_code=status.HTTP_404_NOT_FOUND,
            detail="Task not found"
        )

    # Validate file access
    is_valid, error_msg = file_access_service.validate_file_access(
        db=db,
        user_id=current_user.id,
        task_id=task_id,
        file_path=task.result_markdown_path
    )

    if not is_valid:
        raise HTTPException(
            status_code=status.HTTP_403_FORBIDDEN,
            detail=error_msg
        )

    # Return file
    filename = f"{task.filename or task_id}_result.md"
    return FileResponse(
        path=task.result_markdown_path,
        filename=filename,
        media_type="text/markdown"
    )


@router.get("/{task_id}/download/pdf", summary="Download PDF result")
async def download_pdf(
    task_id: str,
    format: Optional[str] = Query(
        None,
        description="PDF format: 'layout' (default) preserves original coordinates, 'reflow' provides flowing text with consistent font sizes"
    ),
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Download task result as PDF file

    - **task_id**: Task UUID
    - **format**: Optional format parameter
      - `layout` (default): Preserves original document layout and coordinates
      - `reflow`: Flowing text with consistent font sizes for better readability

    Returns a PDF generated from OCR JSON data. The PDF is cached for subsequent requests.
    """
    from pathlib import Path
    from app.services.pdf_generator_service import pdf_generator_service

    # Get task
    task = task_service.get_task_by_id(
        db=db,
        task_id=task_id,
        user_id=current_user.id
    )

    if not task:
        raise HTTPException(
            status_code=status.HTTP_404_NOT_FOUND,
            detail="Task not found"
        )

    # Check if task is completed
    if task.status.value != "completed":
        raise HTTPException(
            status_code=status.HTTP_400_BAD_REQUEST,
            detail="Task is not completed yet. Please wait for OCR processing to finish."
        )

    # Determine format (default to layout)
    use_reflow = format and format.lower() == "reflow"

    # Check if PDF path is stored in database (only for layout format, as reflow is always generated)
    if not use_reflow and task.result_pdf_path and Path(task.result_pdf_path).exists():
        pdf_path = Path(task.result_pdf_path)
        logger.info(f"Using pre-generated PDF from database: {pdf_path.name}")
    else:
        # Generate PDF on-demand
        result_dir = Path(settings.result_dir) / task_id

        # Use stored JSON path or construct it
        if task.result_json_path and Path(task.result_json_path).exists():
            json_path = Path(task.result_json_path)
        else:
            # Try to find JSON file in result directory
            json_files = list(result_dir.glob("*_result.json"))
            if not json_files:
                raise HTTPException(
                    status_code=status.HTTP_404_NOT_FOUND,
                    detail="OCR result JSON not found"
                )
            json_path = json_files[0]

        # Construct PDF path based on JSON filename and format
        format_suffix = "_reflow" if use_reflow else "_layout"
        pdf_filename = json_path.stem.replace("_result", format_suffix) + ".pdf"
        pdf_path = result_dir / pdf_filename

        # Generate PDF if it doesn't exist
        if not pdf_path.exists():
            logger.info(f"Generating {'reflow' if use_reflow else 'layout-preserving'} PDF for task {task_id}")

            # Get source file path if available
            source_file = None
            task_file = db.query(TaskFile).filter(TaskFile.task_id == task.id).first()
            if task_file and task_file.stored_path and Path(task_file.stored_path).exists():
                source_file = Path(task_file.stored_path)

            # Generate PDF based on format
            if use_reflow:
                # For reflow, pass result_dir as source_file_path (contains extracted images)
                success = pdf_generator_service.generate_reflow_pdf(
                    json_path=json_path,
                    output_path=pdf_path,
                    source_file_path=result_dir
                )
            else:
                success = pdf_generator_service.generate_layout_pdf(
                    json_path=json_path,
                    output_path=pdf_path,
                    source_file_path=source_file
                )

            if not success:
                raise HTTPException(
                    status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
                    detail="Failed to generate PDF. Please check server logs."
                )

            logger.info(f"PDF generated successfully: {pdf_path.name}")

    # Validate file access
    is_valid, error_msg = file_access_service.validate_file_access(
        db=db,
        user_id=current_user.id,
        task_id=task_id,
        file_path=str(pdf_path)
    )

    if not is_valid:
        raise HTTPException(
            status_code=status.HTTP_403_FORBIDDEN,
            detail=error_msg
        )

    # Return file with format indication in filename
    base_name = task.filename or task_id
    format_suffix = "_reflow" if use_reflow else "_layout"
    filename = f"{base_name}{format_suffix}.pdf"
    return FileResponse(
        path=str(pdf_path),
        filename=filename,
        media_type="application/pdf"
    )


@router.post("/{task_id}/start", response_model=TaskResponse, summary="Start task processing")
async def start_task(
    task_id: str,
    background_tasks: BackgroundTasks,
    options: Optional[ProcessingOptions] = None,
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Start processing a pending task with dual-track support and layout model selection

    - **task_id**: Task UUID
    - **options**: Processing options (in request body):
        - **use_dual_track**: Enable intelligent track selection (default: true)
        - **force_track**: Force specific processing track ('ocr' or 'direct')
        - **language**: OCR language code (default: 'ch')
        - **layout_model**: Layout detection model ('chinese', 'default', 'cdla')
        - **table_detection**: Table detection config (enable_wired_table, enable_wireless_table, enable_region_detection)
    """
    try:
        # Parse processing options with defaults
        if options is None:
            options = ProcessingOptions()

        use_dual_track = options.use_dual_track
        force_track = options.force_track.value if options.force_track else None
        language = options.language

        # Extract layout model (default to 'chinese' for best Chinese document support)
        layout_model = options.layout_model.value if options.layout_model else "chinese"
        logger.info(f"Using layout model: {layout_model}")

        # Extract preprocessing options
        preprocessing_mode = options.preprocessing_mode.value if options.preprocessing_mode else "auto"
        preprocessing_config = None
        if options.preprocessing_config:
            preprocessing_config = {
                "contrast": options.preprocessing_config.contrast.value,
                "contrast_strength": options.preprocessing_config.contrast_strength,
                "sharpen": options.preprocessing_config.sharpen,
                "sharpen_strength": options.preprocessing_config.sharpen_strength,
                "binarize": options.preprocessing_config.binarize
            }
        logger.info(f"Preprocessing: mode={preprocessing_mode}, config={preprocessing_config}")

        # Extract table detection options
        table_detection_config = None
        if options.table_detection:
            table_detection_config = {
                "enable_wired_table": options.table_detection.enable_wired_table,
                "enable_wireless_table": options.table_detection.enable_wireless_table,
                "enable_region_detection": options.table_detection.enable_region_detection
            }
        logger.info(f"Table detection: {table_detection_config}")

        # Get task details
        task = task_service.get_task_by_id(
            db=db,
            task_id=task_id,
            user_id=current_user.id
        )

        if not task:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Task not found"
            )

        # Check if task is in pending status
        if task.status != TaskStatus.PENDING:
            raise HTTPException(
                status_code=status.HTTP_400_BAD_REQUEST,
                detail=f"Cannot start task in '{task.status.value}' status"
            )

        # Get task file
        task_file = db.query(TaskFile).filter(TaskFile.task_id == task.id).first()
        if not task_file:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Task file not found"
            )

        # Update task status to processing
        task = task_service.update_task_status(
            db=db,
            task_id=task_id,
            user_id=current_user.id,
            status=TaskStatus.PROCESSING
        )

        # Start OCR processing in background with dual-track parameters and layout model
        background_tasks.add_task(
            process_task_ocr,
            task_id=task_id,
            task_db_id=task.id,
            file_path=task_file.stored_path,
            filename=task_file.original_name,
            use_dual_track=use_dual_track,
            force_track=force_track,
            language=language,
            layout_model=layout_model,
            preprocessing_mode=preprocessing_mode,
            preprocessing_config=preprocessing_config,
            table_detection_config=table_detection_config
        )

        logger.info(f"Started OCR processing task {task_id} for user {current_user.email}")
        logger.info(f"Options: dual_track={use_dual_track}, force_track={force_track}, lang={language}, layout_model={layout_model}, preprocessing={preprocessing_mode}, table_detection={table_detection_config}")
        return task

    except HTTPException:
        raise
    except Exception as e:
        logger.exception(f"Failed to start task {task_id}")
        raise HTTPException(
            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
            detail=f"Failed to start task: {str(e)}"
        )


@router.post("/{task_id}/cancel", response_model=TaskResponse, summary="Cancel task")
async def cancel_task(
    task_id: str,
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Cancel a pending or processing task

    - **task_id**: Task UUID
    """
    try:
        # Get current task
        task = task_service.get_task_by_id(
            db=db,
            task_id=task_id,
            user_id=current_user.id
        )

        if not task:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Task not found"
            )

        # Only allow canceling pending or processing tasks
        if task.status not in [TaskStatus.PENDING, TaskStatus.PROCESSING]:
            raise HTTPException(
                status_code=status.HTTP_400_BAD_REQUEST,
                detail=f"Cannot cancel task in '{task.status.value}' status"
            )

        # Update to failed status with cancellation message
        task = task_service.update_task_status(
            db=db,
            task_id=task_id,
            user_id=current_user.id,
            status=TaskStatus.FAILED,
            error_message="Task cancelled by user"
        )

        logger.info(f"Cancelled task {task_id} for user {current_user.email}")
        return task

    except HTTPException:
        raise
    except Exception as e:
        logger.exception(f"Failed to cancel task {task_id}")
        raise HTTPException(
            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
            detail=f"Failed to cancel task: {str(e)}"
        )


@router.post("/{task_id}/retry", response_model=TaskResponse, summary="Retry failed task")
async def retry_task(
    task_id: str,
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Retry a failed task

    - **task_id**: Task UUID
    """
    try:
        # Get current task
        task = task_service.get_task_by_id(
            db=db,
            task_id=task_id,
            user_id=current_user.id
        )

        if not task:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Task not found"
            )

        # Only allow retrying failed tasks
        if task.status != TaskStatus.FAILED:
            raise HTTPException(
                status_code=status.HTTP_400_BAD_REQUEST,
                detail=f"Cannot retry task in '{task.status.value}' status"
            )

        # Reset task to pending status
        task = task_service.update_task_status(
            db=db,
            task_id=task_id,
            user_id=current_user.id,
            status=TaskStatus.PENDING,
            error_message=None
        )

        logger.info(f"Retrying task {task_id} for user {current_user.email}")
        return task

    except HTTPException:
        raise
    except Exception as e:
        logger.exception(f"Failed to retry task {task_id}")
        raise HTTPException(
            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
            detail=f"Failed to retry task: {str(e)}"
        )


# ===== Document Analysis Endpoints =====

@router.post("/{task_id}/analyze", response_model=DocumentAnalysisResponse, summary="Analyze document type")
async def analyze_document(
    task_id: str,
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Analyze document to determine recommended processing track

    Returns document type analysis with recommended processing track
    (OCR for scanned documents, DIRECT for editable PDFs)

    - **task_id**: Task UUID
    """
    try:
        if not DUAL_TRACK_AVAILABLE:
            raise HTTPException(
                status_code=status.HTTP_501_NOT_IMPLEMENTED,
                detail="Dual-track processing not available"
            )

        # Get task details
        task = task_service.get_task_by_id(
            db=db,
            task_id=task_id,
            user_id=current_user.id
        )

        if not task:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Task not found"
            )

        # Get task file
        task_file = db.query(TaskFile).filter(TaskFile.task_id == task.id).first()
        if not task_file:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Task file not found"
            )

        # Analyze document (using detect method)
        detector = DocumentTypeDetector()
        recommendation = detector.detect(Path(task_file.stored_path))

        # Build response
        response = DocumentAnalysisResponse(
            task_id=task_id,
            filename=task_file.original_name or "",
            recommended_track=ProcessingTrackEnum(recommendation.track),
            confidence=recommendation.confidence,
            reason=recommendation.reason,
            document_info=recommendation.metadata or {},
            is_editable=recommendation.track == "direct",
            text_coverage=recommendation.metadata.get("text_coverage") if recommendation.metadata else None,
            page_count=recommendation.metadata.get("total_pages") if recommendation.metadata else None
        )

        logger.info(f"Document analysis for task {task_id}: {recommendation.track} (confidence: {recommendation.confidence})")
        return response

    except HTTPException:
        raise
    except Exception as e:
        logger.exception(f"Failed to analyze document for task {task_id}")
        raise HTTPException(
            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
            detail=f"Failed to analyze document: {str(e)}"
        )


@router.get("/{task_id}/metadata", response_model=ProcessingMetadata, summary="Get processing metadata")
async def get_processing_metadata(
    task_id: str,
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Get processing metadata for a completed task

    Returns detailed processing information including track used,
    element counts, and statistics.

    - **task_id**: Task UUID
    """
    try:
        # Get task details
        task = task_service.get_task_by_id(
            db=db,
            task_id=task_id,
            user_id=current_user.id
        )

        if not task:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Task not found"
            )

        if task.status != TaskStatus.COMPLETED:
            raise HTTPException(
                status_code=status.HTTP_400_BAD_REQUEST,
                detail="Task not completed"
            )

        # Load JSON result to get metadata
        if not task.result_json_path:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Result JSON not found"
            )

        json_path = Path(task.result_json_path)
        if not json_path.exists():
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Result file not found"
            )

        with open(json_path, 'r', encoding='utf-8') as f:
            result_data = json.load(f)

        # Extract metadata
        metadata = result_data.get('metadata', {})
        statistics = result_data.get('statistics', {})

        response = ProcessingMetadata(
            processing_track=ProcessingTrackEnum(metadata.get('processing_track', 'ocr')),
            processing_time_seconds=metadata.get('processing_time', 0),
            language=metadata.get('language', 'ch'),
            page_count=statistics.get('page_count', 1),
            total_elements=statistics.get('total_elements', 0),
            total_text_regions=len(result_data.get('text_regions', [])) if 'text_regions' in result_data else statistics.get('total_elements', 0),
            total_tables=statistics.get('total_tables', 0),
            total_images=statistics.get('total_images', 0),
            average_confidence=result_data.get('average_confidence'),
            unified_format=metadata.get('processing_info', {}).get('export_format') == 'unified_document_v1'
        )

        return response

    except HTTPException:
        raise
    except Exception as e:
        logger.exception(f"Failed to get metadata for task {task_id}")
        raise HTTPException(
            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
            detail=f"Failed to get metadata: {str(e)}"
        )


@router.get("/{task_id}/download/unified", summary="Download unified format")
async def download_unified(
    task_id: str,
    include_metadata: bool = Query(True, description="Include processing metadata"),
    include_statistics: bool = Query(True, description="Include document statistics"),
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Download results in unified document format

    Returns JSON with full UnifiedDocument structure including
    all elements, coordinates, and metadata.

    - **task_id**: Task UUID
    - **include_metadata**: Include processing metadata
    - **include_statistics**: Include document statistics
    """
    try:
        # Get task details
        task = task_service.get_task_by_id(
            db=db,
            task_id=task_id,
            user_id=current_user.id
        )

        if not task:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Task not found"
            )

        if task.status != TaskStatus.COMPLETED:
            raise HTTPException(
                status_code=status.HTTP_400_BAD_REQUEST,
                detail="Task not completed"
            )

        # Get JSON result path
        if not task.result_json_path:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Result JSON not found"
            )

        json_path = Path(task.result_json_path)
        if not json_path.exists():
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Result file not found"
            )

        # Return the unified format JSON
        return FileResponse(
            path=str(json_path),
            filename=f"{task_id}_unified.json",
            media_type="application/json"
        )

    except HTTPException:
        raise
    except Exception as e:
        logger.exception(f"Failed to download unified format for task {task_id}")
        raise HTTPException(
            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
            detail=f"Failed to download: {str(e)}"
        )


# ===== Preprocessing Preview Endpoints =====

@router.post("/{task_id}/preview/preprocessing", response_model=PreprocessingPreviewResponse, summary="Preview preprocessing effect")
async def preview_preprocessing(
    task_id: str,
    request: PreprocessingPreviewRequest,
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Preview the effect of image preprocessing before OCR processing.

    Shows side-by-side comparison of original and preprocessed images,
    along with image quality metrics and auto-detected configuration.

    - **task_id**: Task UUID
    - **page**: Page number to preview (1-based)
    - **mode**: Preprocessing mode ('auto', 'manual', 'disabled')
    - **config**: Manual preprocessing config (only used when mode='manual')
    """
    from pdf2image import convert_from_path
    import base64
    import io
    from PIL import Image
    from app.services.layout_preprocessing_service import get_layout_preprocessing_service

    try:
        # Get task details
        task = task_service.get_task_by_id(
            db=db,
            task_id=task_id,
            user_id=current_user.id
        )

        if not task:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Task not found"
            )

        # Get task file
        task_file = db.query(TaskFile).filter(TaskFile.task_id == task.id).first()
        if not task_file:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Task file not found"
            )

        file_path = Path(task_file.stored_path)
        if not file_path.exists():
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Source file not found"
            )

        # Get the page image
        page_num = request.page
        if file_path.suffix.lower() == '.pdf':
            # Convert specific page from PDF
            images = convert_from_path(
                str(file_path),
                first_page=page_num,
                last_page=page_num,
                dpi=150
            )
            if not images:
                raise HTTPException(
                    status_code=status.HTTP_400_BAD_REQUEST,
                    detail=f"Page {page_num} not found in PDF"
                )
            original_image = images[0]
        else:
            # Direct image file
            if page_num != 1:
                raise HTTPException(
                    status_code=status.HTTP_400_BAD_REQUEST,
                    detail="Single image file only has page 1"
                )
            original_image = Image.open(file_path)

        # Get preprocessing service
        preprocessing_service = get_layout_preprocessing_service()

        # Apply preprocessing
        preprocessed_image, preprocess_result = preprocessing_service.preprocess_to_pil(
            original_image,
            mode=request.mode,
            config=request.config
        )

        # Create result directory for preview images
        preview_dir = Path(settings.result_dir) / task_id / "preview"
        preview_dir.mkdir(parents=True, exist_ok=True)

        # Save preview images
        original_filename = f"page_{page_num}_original.png"
        preprocessed_filename = f"page_{page_num}_preprocessed.png"

        original_path = preview_dir / original_filename
        preprocessed_path = preview_dir / preprocessed_filename

        original_image.save(str(original_path), "PNG")
        preprocessed_image.save(str(preprocessed_path), "PNG")

        # Build URLs (relative paths that can be served)
        base_url = f"/api/v2/tasks/{task_id}/preview/image"
        original_url = f"{base_url}?type=original&page={page_num}"
        preprocessed_url = f"{base_url}?type=preprocessed&page={page_num}"

        return PreprocessingPreviewResponse(
            original_url=original_url,
            preprocessed_url=preprocessed_url,
            quality_metrics=preprocess_result.quality_metrics,
            auto_config=preprocess_result.config_used,
            mode_used=request.mode
        )

    except HTTPException:
        raise
    except Exception as e:
        logger.exception(f"Failed to preview preprocessing for task {task_id}")
        raise HTTPException(
            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
            detail=f"Failed to preview preprocessing: {str(e)}"
        )


@router.get("/{task_id}/preview/image", summary="Get preview image")
async def get_preview_image(
    task_id: str,
    type: str = Query(..., description="Image type: 'original' or 'preprocessed'"),
    page: int = Query(1, ge=1, description="Page number"),
    db: Session = Depends(get_db),
    current_user: User = Depends(get_current_user)
):
    """
    Get a preview image (original or preprocessed).

    - **task_id**: Task UUID
    - **type**: Image type ('original' or 'preprocessed')
    - **page**: Page number
    """
    try:
        # Verify task ownership
        task = task_service.get_task_by_id(
            db=db,
            task_id=task_id,
            user_id=current_user.id
        )

        if not task:
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail="Task not found"
            )

        # Validate type parameter
        if type not in ['original', 'preprocessed']:
            raise HTTPException(
                status_code=status.HTTP_400_BAD_REQUEST,
                detail="Invalid type. Must be 'original' or 'preprocessed'"
            )

        # Build image path
        preview_dir = Path(settings.result_dir) / task_id / "preview"
        image_filename = f"page_{page}_{type}.png"
        image_path = preview_dir / image_filename

        if not image_path.exists():
            raise HTTPException(
                status_code=status.HTTP_404_NOT_FOUND,
                detail=f"Preview image not found. Please call preview/preprocessing first."
            )

        return FileResponse(
            path=str(image_path),
            media_type="image/png",
            filename=image_filename
        )

    except HTTPException:
        raise
    except Exception as e:
        logger.exception(f"Failed to get preview image for task {task_id}")
        raise HTTPException(
            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
            detail=f"Failed to get preview image: {str(e)}"
        )