first ruggedy draft

apps/local-agent/mousehold_agent/csv_importer.py

@@ -0,0 +1,86 @@
+from __future__ import annotations
+
+import csv
+from datetime import date
+from decimal import Decimal
+from pathlib import Path
+from typing import Any
+
+
+def read_transaction_csv(
+    path: Path,
+    *,
+    household_id: str,
+    source_type: str,
+    source_account_id: str | None,
+    owner_user_id: str | None,
+) -> list[dict[str, Any]]:
+    rows: list[dict[str, Any]] = []
+    with path.open(newline="", encoding="utf-8-sig") as handle:
+        reader = csv.DictReader(handle)
+        for row in reader:
+            event_date = _pick(
+                row, "event_date", "date", "booking_date", "Buchungstag", "Valutadatum"
+            )
+            amount = _pick(row, "amount", "Betrag", "value")
+            if not event_date or not amount:
+                continue
+            merchant = _pick(
+                row,
+                "merchant",
+                "counterparty",
+                "Auftraggeber/Empfänger",
+                "Name Zahlungsbeteiligter",
+            )
+            description = _pick(row, "description", "purpose", "Verwendungszweck", "Buchungstext")
+            currency = _pick(row, "currency", "Waehrung", "Währung") or "EUR"
+            provider_transaction_id = _pick(
+                row, "provider_transaction_id", "transaction_id", "Umsatz ID"
+            )
+            normalized = {
+                "household_id": household_id,
+                "source_type": source_type,
+                "source_account_id": source_account_id,
+                "owner_user_id": owner_user_id,
+                "source_reference": provider_transaction_id,
+                "provider_transaction_id": provider_transaction_id,
+                "raw_payload_json": row,
+                "amount": str(_parse_amount(amount)),
+                "currency": currency,
+                "event_date": _parse_date(event_date).isoformat(),
+                "merchant": merchant,
+                "description": description,
+                "event_type": "expense" if _parse_amount(amount) < 0 else "income",
+                "status": "confirmed",
+            }
+            normalized["amount"] = str(abs(_parse_amount(amount)))
+            rows.append(normalized)
+    return rows
+
+
+def _pick(row: dict[str, str], *keys: str) -> str | None:
+    lowered = {key.lower(): value for key, value in row.items()}
+    for key in keys:
+        value = row.get(key) or lowered.get(key.lower())
+        if value:
+            return value.strip()
+    return None
+
+
+def _parse_amount(value: str) -> Decimal:
+    normalized = value.strip().replace(".", "").replace(",", ".")
+    return Decimal(normalized)
+
+
+def _parse_date(value: str) -> date:
+    stripped = value.strip()
+    for fmt in ("%Y-%m-%d", "%d.%m.%Y", "%d/%m/%Y"):
+        try:
+            return (
+                date.fromisoformat(stripped)
+                if fmt == "%Y-%m-%d"
+                else __import__("datetime").datetime.strptime(stripped, fmt).date()
+            )
+        except ValueError:
+            continue
+    raise ValueError(f"Unsupported date format: {value}")