backend/app/api/datasets.py

import json
from pathlib import Path

import pandas as pd
from fastapi import APIRouter, Depends, File, HTTPException, UploadFile
from sqlalchemy.orm import Session

from backend.app.db import get_db
from backend.app.repositories.dataset_repo import create_dataset, get_dataset, list_datasets
from backend.app.services.profiling_service import profile_dataframe
from backend.app.utils.ids import make_dataset_id

router = APIRouter(tags=["datasets"])

UPLOAD_DIR = Path("/data/uploads")
UPLOAD_DIR.mkdir(parents=True, exist_ok=True)


@router.get("/datasets")
def datasets_list(db: Session = Depends(get_db)):
    datasets = list_datasets(db)
    return {
        "datasets": [
            {
                "id": d.id,
                "name": d.name,
                "row_count": d.row_count,
                "column_count": d.column_count,
            }
            for d in datasets
        ]
    }


@router.post("/datasets/upload")
async def upload_dataset(file: UploadFile = File(...), db: Session = Depends(get_db)):
    if not file.filename:
        raise HTTPException(status_code=400, detail="Missing file name")

    suffix = Path(file.filename).suffix.lower()
    if suffix not in {".csv", ".xlsx", ".xls"}:
        raise HTTPException(status_code=400, detail="Only CSV and Excel files are supported")

    dataset_id = make_dataset_id()
    path = UPLOAD_DIR / f"{dataset_id}{suffix}"

    content = await file.read()
    path.write_bytes(content)

    if suffix == ".csv":
        df = pd.read_csv(path)
    else:
        df = pd.read_excel(path)

    profile = profile_dataframe(df)

    create_dataset(
        db=db,
        id=dataset_id,
        name=file.filename,
        file_path=str(path),
        row_count=int(len(df)),
        column_count=int(len(df.columns)),
        schema_json=json.dumps({"columns": list(df.columns)}),
        profile_json=json.dumps(profile),
    )

    return {
        "dataset_id": dataset_id,
        "name": file.filename,
        "row_count": int(len(df)),
        "column_count": int(len(df.columns)),
    }


@router.get("/datasets/{dataset_id}/profile")
def dataset_profile(dataset_id: str, db: Session = Depends(get_db)):
    dataset = get_dataset(db, dataset_id)
    if not dataset:
        raise HTTPException(status_code=404, detail="Dataset not found")

    return {
        "dataset_id": dataset.id,
        "name": dataset.name,
        "row_count": dataset.row_count,
        "column_count": dataset.column_count,
        "schema": json.loads(dataset.schema_json),
        "profile": json.loads(dataset.profile_json),
    }