infoscreen/server/routes/holidays.py

from flask import Blueprint, request, jsonify
from server.permissions import admin_or_higher
from server.database import Session
from models.models import SchoolHoliday
from datetime import datetime
import csv
import io

holidays_bp = Blueprint("holidays", __name__, url_prefix="/api/holidays")


@holidays_bp.route("", methods=["GET"])
def list_holidays():
    session = Session()
    region = request.args.get("region")
    q = session.query(SchoolHoliday)
    if region:
        q = q.filter(SchoolHoliday.region == region)
    rows = q.order_by(SchoolHoliday.start_date.asc()).all()
    data = [r.to_dict() for r in rows]
    session.close()
    return jsonify({"holidays": data})


@holidays_bp.route("/upload", methods=["POST"])
@admin_or_higher
def upload_holidays():
    """
    Accepts a CSV/TXT file upload (multipart/form-data).

    Supported formats:
    1) Headered CSV with columns (case-insensitive): name, start_date, end_date[, region]
       - Dates: YYYY-MM-DD, DD.MM.YYYY, YYYY/MM/DD, or YYYYMMDD
    2) Headerless CSV/TXT lines with columns:
       [internal, name, start_yyyymmdd, end_yyyymmdd, optional_internal]
       - Only columns 2-4 are used; 1 and 5 are ignored.
    """
    if "file" not in request.files:
        return jsonify({"error": "No file part"}), 400
    file = request.files["file"]
    if file.filename == "":
        return jsonify({"error": "No selected file"}), 400

    try:
        raw = file.read()
        # Try UTF-8 first (strict), then cp1252, then latin-1 as last resort
        try:
            content = raw.decode("utf-8")
        except UnicodeDecodeError:
            try:
                content = raw.decode("cp1252")
            except UnicodeDecodeError:
                content = raw.decode("latin-1", errors="replace")

        sniffer = csv.Sniffer()
        dialect = None
        try:
            sample = content[:2048]
            # Some files may contain a lot of quotes; allow Sniffer to guess delimiter
            dialect = sniffer.sniff(sample)
        except Exception:
            pass

        def parse_date(s: str):
            s = (s or "").strip()
            if not s:
                return None
            # Numeric YYYYMMDD
            if s.isdigit() and len(s) == 8:
                try:
                    return datetime.strptime(s, "%Y%m%d").date()
                except ValueError:
                    pass
            # Common formats
            for fmt in ("%Y-%m-%d", "%d.%m.%Y", "%Y/%m/%d"):
                try:
                    return datetime.strptime(s, fmt).date()
                except ValueError:
                    continue
            raise ValueError(f"Unsupported date format: {s}")

        session = Session()
        inserted = 0
        updated = 0

        # First, try headered CSV via DictReader
        dict_reader = csv.DictReader(io.StringIO(
            content), dialect=dialect) if dialect else csv.DictReader(io.StringIO(content))
        fieldnames_lower = [h.lower() for h in (dict_reader.fieldnames or [])]
        has_required_headers = {"name", "start_date",
                                "end_date"}.issubset(set(fieldnames_lower))

        def upsert(name: str, start_date, end_date, region=None):
            nonlocal inserted, updated
            if not name or not start_date or not end_date:
                return
            existing = (
                session.query(SchoolHoliday)
                .filter(
                    SchoolHoliday.name == name,
                    SchoolHoliday.start_date == start_date,
                    SchoolHoliday.end_date == end_date,
                    SchoolHoliday.region.is_(
                        region) if region is None else SchoolHoliday.region == region,
                )
                .first()
            )
            if existing:
                existing.region = region
                existing.source_file_name = file.filename
                updated += 1
            else:
                session.add(SchoolHoliday(
                    name=name,
                    start_date=start_date,
                    end_date=end_date,
                    region=region,
                    source_file_name=file.filename,
                ))
                inserted += 1

        if has_required_headers:
            for row in dict_reader:
                norm = {k.lower(): (v or "").strip() for k, v in row.items()}
                name = norm.get("name")
                try:
                    start_date = parse_date(norm.get("start_date"))
                    end_date = parse_date(norm.get("end_date"))
                except ValueError:
                    # Skip rows with unparseable dates
                    continue
                region = (norm.get("region")
                          or None) if "region" in norm else None
                upsert(name, start_date, end_date, region)
        else:
            # Fallback: headerless rows -> use columns [1]=name, [2]=start, [3]=end
            reader = csv.reader(io.StringIO(
                content), dialect=dialect) if dialect else csv.reader(io.StringIO(content))
            for row in reader:
                if not row:
                    continue
                # tolerate varying column counts (4 or 5); ignore first and optional last
                cols = [c.strip() for c in row]
                if len(cols) < 4:
                    # Not enough data
                    continue
                name = cols[1].strip().strip('"')
                start_raw = cols[2]
                end_raw = cols[3]
                try:
                    start_date = parse_date(start_raw)
                    end_date = parse_date(end_raw)
                except ValueError:
                    continue
                upsert(name, start_date, end_date, None)

        session.commit()
        session.close()
        return jsonify({"success": True, "inserted": inserted, "updated": updated})
    except Exception as e:
        return jsonify({"error": str(e)}), 400