Compare commits

...
Author SHA1 Message Date
mathias2784 5cea645062 Some more refactoring
Tests / test (push) Successful in 10s
Tests / lint (push) Successful in 11s
2026-10-16 16:27:50 +02:00
mathias2784 ef1a65adeb some refactoring 2026-10-16 15:27:41 +02:00
mathias2784 696823406e Extract Category model into src/models/category.py
Tests / test (push) Successful in 10s
Tests / lint (push) Successful in 11s
2026-10-16 12:29:11 +02:00
mathias2784 c7e4b68dfb Fix comdirect CSV reader for neu group header rows 2026-10-16 12:28:57 +02:00
mathias2784 435f7b92bf Add category_budget satellite table for time-bounded per-category budgets
Tests / test (push) Successful in 10s
Tests / lint (push) Successful in 12s
2026-10-16 11:45:13 +02:00
19 changed files with 364 additions and 121 deletions
+1 -1
View File
@@ -71,7 +71,7 @@ Two sources with shared patterns:
- **comdirect**: `ComdirectGiro`, `ComdirectCreditCard`, `ComdirectDepot`, import to `comdirect_giro` table
- **unified**: `Transaction`, `TransactionDb` – both ingest pipelines project their staging rows (`import_b4w`, `comdirect_giro`) into one `transactions` table via `src/ingest/transactions.py` (`map_*` mappers + `consolidate_transactions`). Dedup uses a business-key `hash_id` (source + IBAN + value date + amount + text + counterparty) with a UNIQUE constraint and SQLite `INSERT OR IGNORE`, so re-imports are idempotent. Amounts are canonical signed (negative = money out). Depot/credit-card/tagesgeld are not projected yet.
All models inherit from common base in `src/models/_common.py`:
All models inherit from common base in `src/models/common.py`:
- `_hash()`: SHA256-based deduplication key
- `_parse_date_german()`: Handles "07.09.2026" format and values like "offen", "--"
- `_parse_decimal_german()`: Converts "1,23" to Decimal
+1 -1
View File
@@ -6,7 +6,7 @@ from sqlalchemy import engine_from_config, pool
from alembic import context
from src.db import get_set_engine
from src.models._common import BaseModelDb
from src.models.common import BaseModelDb
load_dotenv() # reads .env into os.environ
@@ -0,0 +1,52 @@
"""create category_budget table
Revision ID: fe23064de1c5
Revises: 78ec55b9e1d1
Create Date: 2026-10-16 10:43:24.952699
"""
from collections.abc import Sequence
import sqlalchemy as sa
from alembic import op
# revision identifiers, used by Alembic.
revision: str = "fe23064de1c5"
down_revision: str | Sequence[str] | None = "78ec55b9e1d1"
branch_labels: str | Sequence[str] | None = None
depends_on: str | Sequence[str] | None = None
def upgrade() -> None:
"""Upgrade schema."""
# ### commands auto generated by Alembic - please adjust! ###
op.create_table(
"category_budget",
sa.Column("id", sa.Integer(), autoincrement=True, nullable=False),
sa.Column("category_id", sa.Integer(), nullable=False),
sa.Column("amount_eur", sa.Numeric(precision=12, scale=2), nullable=False),
sa.Column("valid_from", sa.Date(), nullable=True),
sa.Column("valid_to", sa.Date(), nullable=True),
sa.ForeignKeyConstraint(
["category_id"],
["category.id"],
),
sa.PrimaryKeyConstraint("id"),
)
op.create_index(
"ix_category_budget_category_id",
"category_budget",
["category_id"],
unique=False,
)
# ### end Alembic commands ###
def downgrade() -> None:
"""Downgrade schema."""
# ### commands auto generated by Alembic - please adjust! ###
op.drop_index("ix_category_budget_category_id", table_name="category_budget")
op.drop_table("category_budget")
# ### end Alembic commands ###
+5 -15
View File
@@ -1,20 +1,16 @@
import csv
from pathlib import Path
from typing import Any
from sqlalchemy.orm import Session
from src.models._common import BaseModelDb
from sqlalchemy.orm import DeclarativeBase, Session
def _query_existing(
session: Session, DbModel: type[BaseModelDb], query_parameters: dict[str, Any]
) -> BaseModelDb | None:
def _query_existing[DbModelT: DeclarativeBase](
session: Session, DbModel: type[DbModelT], query_parameters: dict[str, Any]
) -> DbModelT | None:
query = session.query(DbModel).filter_by(**query_parameters)
return query.first()
def _upsert[DbModelT: BaseModelDb](
def upsert[DbModelT: DeclarativeBase](
session: Session,
DbModel: type[DbModelT],
parameters_existing: dict[str, Any],
@@ -28,7 +24,6 @@ def _upsert[DbModelT: BaseModelDb](
if do_update:
for key, value in parameters_new.items():
setattr(existing_db_row, key, value)
assert isinstance(existing_db_row, DbModel)
return existing_db_row
new_db_row = DbModel(**parameters_new)
@@ -37,8 +32,3 @@ def _upsert[DbModelT: BaseModelDb](
if do_commit:
session.commit()
return new_db_row
def _reader_raw(path: Path):
with path.open() as fid:
return list(csv.DictReader(fid))
+1
View File
@@ -0,0 +1 @@
"""Database helper for application-specific persistence operations."""
+22
View File
@@ -0,0 +1,22 @@
from sqlalchemy.orm import Session
from src.db_utils import upsert
from src.models.category import Category, CategoryDb
def upsert_category(
session: Session, category: Category | None, do_commit: bool = True
) -> CategoryDb | None:
if category is None:
return None
parent = upsert_category(session, category.parent, do_commit=False)
parameters = {"name": category.name, "parent": parent}
return upsert(
session,
CategoryDb,
parameters,
parameters,
do_update=False,
do_commit=do_commit,
)
+5 -22
View File
@@ -2,9 +2,11 @@ from pathlib import Path
from sqlalchemy.orm import Session
from src.ingest._common import _reader_raw, _upsert
from src.db_utils import upsert
from src.helper.category import upsert_category
from src.ingest.common import _reader_raw
from src.ingest.transactions import consolidate_transactions, map_b4w_to_transaction
from src.models.b4w import B4wDb, B4wExport, Category, CategoryDb
from src.models.b4w import B4wDb, B4wExport
def reader_b4w(path: Path) -> list[B4wExport]:
@@ -12,25 +14,6 @@ def reader_b4w(path: Path) -> list[B4wExport]:
return list(map(B4wExport.model_validate, raw))
def upsert_category(
session: Session, category: Category | None, do_commit=True
) -> CategoryDb | None:
if category is None:
return None
parameters = {
"name": category.name,
"parent": upsert_category(session, category.parent),
}
return _upsert(
session,
CategoryDb,
parameters,
parameters,
do_update=False,
do_commit=do_commit,
)
def upsert_b4w(
session: Session, data: B4wExport, do_update=False, do_commit=True
) -> B4wDb | None:
@@ -54,7 +37,7 @@ def upsert_b4w(
"notes": data.notes,
}
return _upsert(
return upsert(
session,
B4wDb,
parameters_existing,
+18 -9
View File
@@ -6,18 +6,18 @@ from typing import cast
from sqlalchemy.orm import Session
from settings import BANK_ACCOUNTS
from src.ingest._common import _upsert
from src.db_utils import upsert
from src.ingest.transactions import (
consolidate_transactions,
map_comdirect_giro_to_transaction,
)
from src.models._common import Csv2Dict, CsV2DictMulti
from src.models.comdirect import (
ComdirectCreditCard,
ComdirectDepot,
ComdirectGiro,
ComdirectGiroDb,
)
from src.models.common import Csv2Dict, CsV2DictMulti
from src.models.settings import Account, Source
PATTERN_ACCOUNT_NAME = r'^"Umsätze ([^"]+)";"Zeitraum: .*$'
@@ -73,6 +73,10 @@ def _reader_raw_comdirect(path: Path) -> CsV2DictMulti:
account_name: str | None = None
header: list[str] = []
result: CsV2DictMulti = {}
# Comdirect groups "new" transactions (marked with "neu") under a header
# line that carries only their Buchungstag; remember it until the next
# transaction row for the same account.
pending_book_date: dict[str, str] = {}
for line in lines:
if expect_next == EXPECT_LINE.ACCOUNT_NAME:
matches = re.findall(PATTERN_ACCOUNT_NAME, line)
@@ -91,13 +95,18 @@ def _reader_raw_comdirect(path: Path) -> CsV2DictMulti:
expect_next = EXPECT_LINE.DATA
elif expect_next == EXPECT_LINE.DATA:
data = re.findall(PATTERN_DATA, line)
if data:
assert account_name is not None
result[account_name].append(
{column: value for column, value in zip(header, data)}
)
else:
if not data:
expect_next = EXPECT_LINE.ACCOUNT_NAME
elif len(data) == 1:
# Group header line (single Buchungstag for the "neu" rows below).
assert account_name is not None
pending_book_date[account_name] = data[0]
else:
assert account_name is not None
row = {column: value for column, value in zip(header, data)}
if data[0] == "neu" and account_name in pending_book_date:
row["Buchungstag"] = pending_book_date.pop(account_name)
result[account_name].append(row)
else:
expect_next = EXPECT_LINE.ACCOUNT_NAME
return result
@@ -128,7 +137,7 @@ def upsert_comdirect_giro(
"text": data.text,
}
return _upsert(
return upsert(
session,
ComdirectGiroDb,
parameters_existing,
+7
View File
@@ -0,0 +1,7 @@
import csv
from pathlib import Path
def _reader_raw(path: Path):
with path.open() as fid:
return list(csv.DictReader(fid))
+1 -1
View File
@@ -4,9 +4,9 @@ from sqlalchemy import select
from sqlalchemy.dialects.sqlite import insert as sqlite_insert
from sqlalchemy.orm import Session
from src.models._common import BaseModelDb
from src.models.b4w import B4wDb
from src.models.comdirect import ComdirectGiroDb
from src.models.common import BaseModelDb
from src.models.transactions import Transaction, TransactionDb
+35 -1
View File
@@ -1,5 +1,39 @@
from importlib import import_module
from typing import TYPE_CHECKING, Any
_EXPORTS = {
"B4wExport": ".b4w",
"Category": ".category",
"CategoryBudgetDb": ".category_budget",
"ComdirectCreditCard": ".comdirect",
"ComdirectDepot": ".comdirect",
"ComdirectGiro": ".comdirect",
"Account": ".settings",
"Source": ".settings",
"Transaction": ".transactions",
"TransactionDb": ".transactions",
}
__all__ = list(_EXPORTS)
def __getattr__(name: str) -> Any:
module_name = _EXPORTS.get(name)
if module_name is None:
raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
value = getattr(import_module(module_name, __name__), name)
globals()[name] = value
return value
def __dir__() -> list[str]:
return sorted([*globals(), *__all__])
if TYPE_CHECKING:
from src.models.b4w import B4wExport as B4wExport
from src.models.b4w import Category as Category
from src.models.category import Category as Category
from src.models.category_budget import CategoryBudgetDb as CategoryBudgetDb
from src.models.comdirect import ComdirectCreditCard as ComdirectCreditCard
from src.models.comdirect import ComdirectDepot as ComdirectDepot
from src.models.comdirect import ComdirectGiro as ComdirectGiro
+5 -55
View File
@@ -2,17 +2,17 @@ from datetime import date
from decimal import Decimal
from pydantic import Field, computed_field, field_validator
from sqlalchemy import Date, ForeignKey, Integer, String, UniqueConstraint
from sqlalchemy import Date, ForeignKey, Integer, String
from sqlalchemy.orm import Mapped, mapped_column, relationship
from src.constants import IBAN_PATTERN, Currency
from src.models._common import (
BaseModelDb,
from src.models.category import Category
from src.models.common import (
ExportBaseModel,
IngestdbBase,
Sign,
_hash,
date2str,
hashify,
)
@@ -59,7 +59,7 @@ class B4wExport(ExportBaseModel):
@computed_field # type: ignore[prop-decorator]
@property
def hash_id(self) -> str:
return _hash(self.id)
return hashify(self.id)
@property
def amount_signed(self) -> Decimal:
@@ -87,56 +87,6 @@ class B4wExport(ExportBaseModel):
return category
class Category(ExportBaseModel):
name: str = Field(pattern=r"[^:\r\n]+")
parent: Category | None = None
@computed_field # type: ignore[prop-decorator]
@property
def hash_id(self) -> str:
return _hash(self.name, self.parent)
@classmethod
def from_str(cls, s: str) -> Category | None:
if not s:
return None
components = s.split(":")
name = components[-1]
rest = ":".join(components[:-1])
return cls(name=name, parent=cls.from_str(rest))
def to_str(self) -> str | None:
if not self.parent:
return self.name
else:
return f"{self.parent.to_str()}:{self.name}"
def __str__(self):
return self.to_str()
def __repr__(self):
return f'Category.from_str("{self.__str__()}")'
class CategoryDb(BaseModelDb):
__tablename__ = "category"
__table_args__ = (
UniqueConstraint("name", "parent_id", name="uq_category_name_parent"),
)
id: Mapped[int] = mapped_column(primary_key=True, autoincrement=True)
name: Mapped[str] = mapped_column(String, nullable=False)
parent_id: Mapped[int | None] = mapped_column(
ForeignKey("category.id"), nullable=True
)
# Relationships
parent = relationship(
"CategoryDb", remote_side="CategoryDb.id", back_populates="children"
)
children = relationship("CategoryDb", back_populates="parent")
class B4wDb(IngestdbBase):
__tablename__ = "import_b4w"
+56
View File
@@ -0,0 +1,56 @@
from pydantic import Field, computed_field
from sqlalchemy import ForeignKey, String, UniqueConstraint
from sqlalchemy.orm import Mapped, mapped_column, relationship
from src.models.common import BaseModelDb, ExportBaseModel, hashify
class Category(ExportBaseModel):
name: str = Field(pattern=r"[^:\r\n]+")
parent: Category | None = None
@computed_field # type: ignore[prop-decorator]
@property
def hash_id(self) -> str:
return hashify(self.name, self.parent)
@classmethod
def from_str(cls, s: str) -> Category | None:
if not s:
return None
components = s.split(":")
name = components[-1]
rest = ":".join(components[:-1])
return cls(name=name, parent=cls.from_str(rest))
def to_str(self) -> str | None:
if not self.parent:
return self.name
else:
return f"{self.parent.to_str()}:{self.name}"
def __str__(self):
return self.to_str()
def __repr__(self):
return f'Category.from_str("{self.__str__()}")'
class CategoryDb(BaseModelDb):
__tablename__ = "category"
__table_args__ = (
UniqueConstraint("name", "parent_id", name="uq_category_name_parent"),
)
id: Mapped[int] = mapped_column(primary_key=True, autoincrement=True)
name: Mapped[str] = mapped_column(String, nullable=False)
parent_id: Mapped[int | None] = mapped_column(
ForeignKey("category.id"), nullable=True
)
# Relationships
parent = relationship(
"CategoryDb", remote_side="CategoryDb.id", back_populates="children"
)
children = relationship("CategoryDb", back_populates="parent")
budgets = relationship("CategoryBudgetDb", back_populates="category")
+27
View File
@@ -0,0 +1,27 @@
from datetime import date
from decimal import Decimal
from sqlalchemy import Date, ForeignKey, Numeric
from sqlalchemy.orm import Mapped, mapped_column, relationship
from src.models.common import BaseModelDb
class CategoryBudgetDb(BaseModelDb):
"""Budget amount for a category, optionally bounded to a time window.
Satellite table to ``category``: one category may have several budgets
(e.g. one per month). ``valid_from``/``valid_to`` are nullable; ``None``
means unbounded on that side.
"""
__tablename__ = "category_budget"
id: Mapped[int] = mapped_column(primary_key=True, autoincrement=True)
category_id: Mapped[int] = mapped_column(ForeignKey("category.id"), nullable=False)
amount_eur: Mapped[Decimal] = mapped_column(Numeric(12, 2), nullable=False)
valid_from: Mapped[date | None] = mapped_column(Date, nullable=True, default=None)
valid_to: Mapped[date | None] = mapped_column(Date, nullable=True, default=None)
# relationship
category = relationship("CategoryDb", back_populates="budgets")
+3 -3
View File
@@ -5,14 +5,14 @@ from pydantic import ConfigDict, Field, computed_field, field_validator
from sqlalchemy import Date, String
from sqlalchemy.orm import Mapped, mapped_column
from src.models._common import (
from src.models.common import (
ExportBaseModel,
IngestdbBase,
_hash,
_parse_date_german,
_parse_decimal_german,
date2str,
decimal2str,
hashify,
)
@@ -72,7 +72,7 @@ class ComdirectGiro(ComdirectBase):
@computed_field # type: ignore[prop-decorator]
@property
def hash_id(self) -> str:
return _hash(self.date_val, self.date_book, self.text)
return hashify(self.date_val, self.date_book, self.text)
@field_validator("date_book", mode="before")
@classmethod
@@ -8,7 +8,7 @@ from sqlalchemy import Date, Numeric, String
from sqlalchemy.orm import DeclarativeBase, Mapped, mapped_column
def _hash(*args):
def hashify(*args):
raw = "".join(str(args))
return hashlib.sha256(raw.encode()).hexdigest()
+2 -2
View File
@@ -5,7 +5,7 @@ from pydantic import Field, computed_field
from sqlalchemy import Date, ForeignKey, Index, Integer, String, UniqueConstraint
from sqlalchemy.orm import Mapped, mapped_column, relationship
from src.models._common import ExportBaseModel, IngestdbBase, _hash, date2str
from src.models.common import ExportBaseModel, IngestdbBase, date2str, hashify
class Transaction(ExportBaseModel):
@@ -33,7 +33,7 @@ class Transaction(ExportBaseModel):
@computed_field # type: ignore[prop-decorator]
@property
def hash_id(self) -> str:
return _hash(
return hashify(
self.source,
self.account_iban,
self.date_valid,
+112
View File
@@ -0,0 +1,112 @@
from datetime import date
from decimal import Decimal
from sqlalchemy import inspect
from sqlalchemy.orm import Session
from src.models.category import CategoryDb
from src.models.category_budget import CategoryBudgetDb
def test_category_budget_schema(db_session: Session):
inspector = inspect(db_session.get_bind())
columns = {c["name"]: c for c in inspector.get_columns("category_budget")}
assert set(columns) == {
"id",
"category_id",
"amount_eur",
"valid_from",
"valid_to",
}
assert columns["id"]["nullable"] is False
assert columns["category_id"]["nullable"] is False
assert columns["amount_eur"]["nullable"] is False
assert columns["valid_from"]["nullable"] is True
assert columns["valid_to"]["nullable"] is True
assert str(columns["amount_eur"]["type"]) == "NUMERIC(12, 2)"
foreign_keys = inspector.get_foreign_keys("category_budget")
assert len(foreign_keys) == 1
assert foreign_keys[0]["constrained_columns"] == ["category_id"]
assert foreign_keys[0]["referred_table"] == "category"
def test_category_budget_defaults_null_dates(db_session: Session):
category = CategoryDb(name="Lebenshaltung")
db_session.add(category)
db_session.flush()
budget = CategoryBudgetDb(category_id=category.id, amount_eur=Decimal("150.00"))
db_session.add(budget)
db_session.commit()
db_session.refresh(budget)
assert budget.id is not None
assert budget.category_id == category.id
assert budget.amount_eur == Decimal("150.00")
# Null means unbounded in either direction.
assert budget.valid_from is None
assert budget.valid_to is None
def test_category_budget_date_window(db_session: Session):
category = CategoryDb(name="Freizeit")
db_session.add(category)
db_session.flush()
budget = CategoryBudgetDb(
category_id=category.id,
amount_eur=Decimal("200.50"),
valid_from=date(2026, 10, 1),
valid_to=date(2026, 10, 31),
)
db_session.add(budget)
db_session.commit()
db_session.refresh(budget)
assert budget.valid_from == date(2026, 10, 1)
assert budget.valid_to == date(2026, 10, 31)
def test_category_budget_relationship_both_ways(db_session: Session):
category = CategoryDb(name="Beruf")
db_session.add(category)
db_session.flush()
budget = CategoryBudgetDb(category_id=category.id, amount_eur=Decimal("1.00"))
db_session.add(budget)
db_session.commit()
# Force re-read from the database so the relationship is loaded, not
# just resolved from the session identity map.
db_session.expire_all()
assert [b.id for b in category.budgets] == [budget.id]
assert budget.category.id == category.id
def test_category_has_multiple_budgets(db_session: Session):
category = CategoryDb(name="Wohnen")
db_session.add(category)
db_session.flush()
budgets = [
CategoryBudgetDb(
category_id=category.id,
amount_eur=Decimal("1200.00"),
valid_from=date(2026, 10, 1),
valid_to=date(2026, 10, 31),
),
CategoryBudgetDb(
category_id=category.id,
amount_eur=Decimal("1300.00"),
valid_from=date(2026, 11, 1),
valid_to=date(2026, 11, 30),
),
]
for budget in budgets:
db_session.add(budget)
db_session.commit()
db_session.expire_all()
assert [b.id for b in category.budgets] == [b.id for b in budgets]
+2 -2
View File
@@ -1,9 +1,9 @@
from sqlalchemy import select
from sqlalchemy.orm import Session
from src.ingest.b4w import upsert_category
from src.helper.category import upsert_category
from src.models import Category
from src.models.b4w import CategoryDb
from src.models.category import CategoryDb
def test_upsert_category_returns_orm_object(db_session: Session):