dss/dss-mvp1/dss/api/export_import_routes.py

"""
FastAPI routes for DSS Export/Import system

Provides REST API endpoints for project export, import, merge, and analysis.
All operations support both synchronous and asynchronous (background job) modes.
"""

from fastapi import APIRouter, File, UploadFile, HTTPException, BackgroundTasks, Query
from fastapi.responses import FileResponse
from pydantic import BaseModel
from pathlib import Path
from datetime import datetime
from typing import Optional, List, Dict, Any
import logging

from dss.export_import.service import DSSProjectService, ExportSummary, ImportSummary, MergeSummary
from dss.models.project import Project

logger = logging.getLogger(__name__)
router = APIRouter(prefix="/api/projects", tags=["export_import"])

# Initialize service layer
service = DSSProjectService(busy_timeout_ms=5000)

# In-memory job tracking (replace with Redis/database in production)
_jobs: Dict[str, Dict[str, Any]] = {}


# ============================================================================
# Pydantic Models for API Responses
# ============================================================================

class ExportResponse(BaseModel):
    """Response from export endpoint"""
    success: bool
    file_size_bytes: Optional[int] = None
    token_count: Optional[int] = None
    component_count: Optional[int] = None
    error: Optional[str] = None
    duration_seconds: Optional[float] = None


class ImportResponse(BaseModel):
    """Response from import endpoint"""
    success: bool
    project_name: Optional[str] = None
    token_count: Optional[int] = None
    component_count: Optional[int] = None
    migration_performed: Optional[bool] = None
    warnings: Optional[List[str]] = None
    error: Optional[str] = None
    duration_seconds: Optional[float] = None
    job_id: Optional[str] = None


class MergeResponse(BaseModel):
    """Response from merge endpoint"""
    success: bool
    new_items: Optional[int] = None
    updated_items: Optional[int] = None
    conflicts: Optional[int] = None
    resolution_strategy: Optional[str] = None
    error: Optional[str] = None
    duration_seconds: Optional[float] = None
    job_id: Optional[str] = None


class AnalysisResponse(BaseModel):
    """Response from analysis endpoint"""
    is_valid: bool
    project_name: Optional[str] = None
    schema_version: Optional[str] = None
    token_count: Optional[int] = None
    component_count: Optional[int] = None
    migration_needed: Optional[bool] = None
    errors: Optional[List[str]] = None
    warnings: Optional[List[str]] = None


class JobStatus(BaseModel):
    """Status of a background job"""
    job_id: str
    status: str  # pending, running, completed, failed
    result: Optional[Dict[str, Any]] = None
    error: Optional[str] = None
    created_at: str
    completed_at: Optional[str] = None


# ============================================================================
# Export Endpoints
# ============================================================================

@router.post("/{project_id}/export", response_class=FileResponse)
async def export_project(
    project_id: str,
    background_tasks: Optional[BackgroundTasks] = None,
    background: bool = Query(False, description="Run as background job")
) -> FileResponse:
    """
    Export a project to a .dss archive file

    Args:
        project_id: ID of project to export
        background: If true, schedule as background job (for large projects)

    Returns:
        .dss archive file download

    Examples:
        ```bash
        # Export synchronously
        curl -X POST http://localhost:8000/api/projects/my-project/export \
          -o my-project.dss

        # Export as background job
        curl -X POST "http://localhost:8000/api/projects/my-project/export?background=true"
        ```
    """
    try:
        # Load project (adapt to your data source)
        project = _load_project(project_id)
        if not project:
            raise HTTPException(status_code=404, detail=f"Project not found: {project_id}")

        # Export
        output_path = Path("/tmp") / f"{project_id}_export.dss"
        result: ExportSummary = service.export_project(project, output_path)

        if not result.success:
            raise HTTPException(status_code=400, detail=result.error)

        # Return file for download
        return FileResponse(
            result.archive_path,
            media_type="application/zip",
            filename=f"{project.name}.dss"
        )

    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"Export failed for {project_id}: {e}")
        raise HTTPException(status_code=500, detail=str(e))


# ============================================================================
# Import Endpoints
# ============================================================================

@router.post("/import", response_model=ImportResponse)
async def import_project(
    file: UploadFile = File(...),
    strategy: str = Query("replace", description="Import strategy: replace or merge"),
    background: bool = Query(False, description="Run as background job")
) -> ImportResponse:
    """
    Import a project from a .dss archive file

    Args:
        file: .dss archive file to import
        strategy: Import strategy (replace=full restoration, merge=smart update)
        background: If true, schedule as background job (for large archives)

    Returns:
        Import result summary

    Examples:
        ```bash
        # Import synchronously
        curl -X POST http://localhost:8000/api/projects/import \
          -F "file=@my-project.dss"

        # Import with merge strategy
        curl -X POST "http://localhost:8000/api/projects/import?strategy=merge" \
          -F "file=@updates.dss"

        # Import as background job
        curl -X POST "http://localhost:8000/api/projects/import?background=true" \
          -F "file=@large-project.dss"
        ```
    """
    archive_path = None
    job_id = None

    try:
        # Save uploaded file
        archive_path = Path("/tmp") / f"import_{datetime.now().timestamp()}.dss"
        contents = await file.read()
        archive_path.write_bytes(contents)

        # Check if should run as background job
        if service._should_schedule_background(archive_path):
            # Schedule background job
            job_id = _create_job_id()
            _jobs[job_id] = {
                "status": "pending",
                "created_at": datetime.now().isoformat(),
                "type": "import",
                "archive_path": str(archive_path),
                "strategy": strategy
            }

            # In production: queue with Celery, RQ, or similar
            # For now: return job ID for polling
            return ImportResponse(
                success=True,
                job_id=job_id,
                duration_seconds=0
            )

        # Run synchronously
        result: ImportSummary = service.import_project(archive_path, strategy)

        if not result.success:
            raise HTTPException(status_code=400, detail=result.error)

        return ImportResponse(
            success=True,
            project_name=result.project_name,
            token_count=result.item_counts.get("tokens") if result.item_counts else None,
            component_count=result.item_counts.get("components") if result.item_counts else None,
            migration_performed=result.migration_performed,
            warnings=result.warnings or [],
            duration_seconds=result.duration_seconds
        )

    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"Import failed: {e}")
        raise HTTPException(status_code=500, detail=str(e))
    finally:
        # Cleanup uploaded file after async processing
        if archive_path and archive_path.exists() and job_id is None:
            try:
                archive_path.unlink()
            except Exception:
                pass


# ============================================================================
# Merge Endpoints
# ============================================================================

@router.post("/{project_id}/merge", response_model=MergeResponse)
async def merge_project(
    project_id: str,
    file: UploadFile = File(...),
    strategy: str = Query("keep_local", description="Conflict resolution: overwrite, keep_local, or fork"),
    background: bool = Query(False, description="Run as background job")
) -> MergeResponse:
    """
    Merge updates from a .dss archive into a project

    Args:
        project_id: ID of project to merge into
        file: .dss archive with updates
        strategy: Conflict resolution strategy
        background: If true, schedule as background job

    Returns:
        Merge result summary

    Examples:
        ```bash
        # Merge with keep_local strategy (preserve local changes)
        curl -X POST "http://localhost:8000/api/projects/my-project/merge?strategy=keep_local" \
          -F "file=@updates.dss"

        # Merge with overwrite strategy (accept remote changes)
        curl -X POST "http://localhost:8000/api/projects/my-project/merge?strategy=overwrite" \
          -F "file=@updates.dss"

        # Merge as background job (for large archives)
        curl -X POST "http://localhost:8000/api/projects/my-project/merge?background=true" \
          -F "file=@large-update.dss"
        ```
    """
    archive_path = None
    job_id = None

    try:
        # Load project
        project = _load_project(project_id)
        if not project:
            raise HTTPException(status_code=404, detail=f"Project not found: {project_id}")

        # Save uploaded file
        archive_path = Path("/tmp") / f"merge_{datetime.now().timestamp()}.dss"
        contents = await file.read()
        archive_path.write_bytes(contents)

        # Check if should run as background job
        if service._should_schedule_background(archive_path):
            job_id = _create_job_id()
            _jobs[job_id] = {
                "status": "pending",
                "created_at": datetime.now().isoformat(),
                "type": "merge",
                "project_id": project_id,
                "archive_path": str(archive_path),
                "strategy": strategy
            }
            return MergeResponse(
                success=True,
                job_id=job_id,
                duration_seconds=0
            )

        # Run synchronously
        result: MergeSummary = service.merge_project(project, archive_path, strategy)

        if not result.success:
            raise HTTPException(status_code=400, detail=result.error)

        return MergeResponse(
            success=True,
            new_items=result.new_items_count,
            updated_items=result.updated_items_count,
            conflicts=result.conflicts_count,
            resolution_strategy=result.resolution_strategy,
            duration_seconds=result.duration_seconds
        )

    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"Merge failed for {project_id}: {e}")
        raise HTTPException(status_code=500, detail=str(e))
    finally:
        if archive_path and archive_path.exists() and job_id is None:
            try:
                archive_path.unlink()
            except Exception:
                pass


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

@router.post("/{project_id}/analyze-merge")
async def analyze_merge(
    project_id: str,
    file: UploadFile = File(...)
) -> AnalysisResponse:
    """
    Analyze merge without applying it (safe preview)

    Args:
        project_id: ID of project to analyze merge into
        file: .dss archive to analyze

    Returns:
        Merge analysis (what changes would happen)

    Examples:
        ```bash
        curl -X POST http://localhost:8000/api/projects/my-project/analyze-merge \
          -F "file=@updates.dss"
        ```
    """
    archive_path = None

    try:
        # Load project
        project = _load_project(project_id)
        if not project:
            raise HTTPException(status_code=404, detail=f"Project not found: {project_id}")

        # Save uploaded file
        archive_path = Path("/tmp") / f"analyze_{datetime.now().timestamp()}.dss"
        contents = await file.read()
        archive_path.write_bytes(contents)

        # Analyze
        analysis = service.analyze_merge(project, archive_path)

        return AnalysisResponse(
            is_valid=analysis.is_valid,
            new_items=len(analysis.new_items.get("tokens", [])),
            conflicts=len(analysis.conflicted_items)
        )

    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"Merge analysis failed for {project_id}: {e}")
        raise HTTPException(status_code=500, detail=str(e))
    finally:
        if archive_path and archive_path.exists():
            try:
                archive_path.unlink()
            except Exception:
                pass


@router.post("/analyze-archive")
async def analyze_archive(
    file: UploadFile = File(...)
) -> AnalysisResponse:
    """
    Analyze a .dss archive without importing it (safe preview)

    Args:
        file: .dss archive to analyze

    Returns:
        Archive analysis details

    Examples:
        ```bash
        curl -X POST http://localhost:8000/api/projects/analyze-archive \
          -F "file=@project.dss"
        ```
    """
    archive_path = None

    try:
        # Save uploaded file
        archive_path = Path("/tmp") / f"analyze_archive_{datetime.now().timestamp()}.dss"
        contents = await file.read()
        archive_path.write_bytes(contents)

        # Analyze
        analysis = service.analyze_import(archive_path)

        return AnalysisResponse(
            is_valid=analysis.is_valid,
            project_name=analysis.project_name,
            schema_version=analysis.schema_version,
            token_count=analysis.content_summary.get("tokens", {}).get("count"),
            component_count=analysis.content_summary.get("components", {}).get("count"),
            migration_needed=analysis.migration_needed,
            errors=[e.message for e in analysis.errors],
            warnings=analysis.warnings
        )

    except Exception as e:
        logger.error(f"Archive analysis failed: {e}")
        raise HTTPException(status_code=500, detail=str(e))
    finally:
        if archive_path and archive_path.exists():
            try:
                archive_path.unlink()
            except Exception:
                pass


# ============================================================================
# Job Status Endpoint
# ============================================================================

@router.get("/jobs/{job_id}", response_model=JobStatus)
async def get_job_status(job_id: str) -> JobStatus:
    """
    Get status of a background job

    Args:
        job_id: ID of the job (returned from async endpoint)

    Returns:
        Current job status and result (if completed)

    Examples:
        ```bash
        curl http://localhost:8000/api/projects/jobs/job-123
        ```
    """
    job = _jobs.get(job_id)
    if not job:
        raise HTTPException(status_code=404, detail=f"Job not found: {job_id}")

    return JobStatus(
        job_id=job_id,
        status=job.get("status", "unknown"),
        result=job.get("result"),
        error=job.get("error"),
        created_at=job.get("created_at", ""),
        completed_at=job.get("completed_at")
    )


# ============================================================================
# Helper Functions
# ============================================================================

def _load_project(project_id: str) -> Optional[Project]:
    """
    Load a project by ID

    ADAPT THIS to your actual data source (database, API, etc.)
    """
    try:
        # Example: Load from database
        # return db.query(Project).filter(Project.id == project_id).first()

        # For now: return a dummy project
        # In production: implement actual loading
        logger.warning(f"Using dummy project for {project_id} - implement _load_project()")
        return Project(
            name=project_id,
            description="Auto-loaded project",
            author="system"
        )
    except Exception as e:
        logger.error(f"Failed to load project {project_id}: {e}")
        return None


def _create_job_id() -> str:
    """Generate unique job ID"""
    import uuid
    return str(uuid.uuid4())[:8]


# ============================================================================
# Export router for inclusion in FastAPI app
# ============================================================================

__all__ = ["router"]