aoiro 0.0.3__py3-none-any.whl → 0.1.0__py3-none-any.whl

aoiro/account.yml

@@ -0,0 +1,68 @@
+資産: #貸方
+  資産:
+    [
+      現金,
+      当座預金,
+      定期預金,
+      その他の預金,
+      受取手形,
+      売掛金,
+      有価証券,
+      棚卸資産,
+      前払金,
+      貸付金,
+      建物,
+      建物附属設備,
+      機械装置,
+      車両運搬具,
+      工具器具備品,
+      土地,
+    ]
+  事業主貸: [事業主貸]
+負債:
+  負債: [支払手形, 買掛金, 借入金, 未払金, 前受金, 預り金]
+  貸倒引当金: [貸倒引当金]
+  事業主借: [事業主借]
+純資産:
+  純資産: [元入金]
+収益: #貸方
+  売上: [売上, 雑収入]
+  売上原価: [仕入返品, 期末商品棚卸高]
+  各種引当金・準備金等: [貸倒引当金繰戻]
+  追加: [家事消費]
+費用: #借方
+  売上: [売上返品]
+  売上原価: [仕入, 期首商品棚卸高]
+  経費:
+    [
+      租税公課,
+      荷造運費,
+      水道光熱費,
+      旅費交通費,
+      通信費,
+      広告宣伝費,
+      接待交際費,
+      損害保険料,
+      修繕費,
+      消耗品費,
+      減価償却費,
+      福利厚生費,
+      給料賃金,
+      外注工賃,
+      利子割引料,
+      地代家賃,
+      貸倒金,
+      雑費,
+    ]
+  各種引当金・準備金等: [専従者給与, 貸倒引当金]
+  追加:
+    [
+      法定福利費,
+      税理士・弁護士報酬,
+      支払手数料,
+      新聞図書費,
+      車両費,
+      研修費,
+      会議費,
+      諸会費,
+    ]

aoiro/cli.py

@@ -1,12 +1,102 @@
-import typer
+from datetime import datetime
+from pathlib import Path
+
+# from rich import print
+import attrs
+import cyclopts
+import networkx as nx
+import pandas as pd
+from account_codes_jp import (
+    get_account_type_factory,
+    get_blue_return_accounts,
+    get_node_from_label,
+)
+from networkx.readwrite.text import generate_network_text
 from rich import print
 
-from .main import add
+from ._ledger import (
+    generalledger_to_multiledger,
+    multiledger_to_ledger,
+)
+from ._multidimensional import multidimensional_ledger_to_ledger
+from ._sheets import get_sheets
+from .reader._expenses import ledger_from_expenses
+from .reader._io import read_general_ledger
+from .reader._sales import ledger_from_sales
+
+app = cyclopts.App(name="aoiro")
+
+
+@app.default
+def metrics(path: Path, year: int | None = None, drop: bool = True) -> None:
+    """
+    Calculate metrics needed for tax declaration.
+
+    Parameters
+    ----------
+    path : Path
+        The path to the directory containing CSV files.
+    year : int | None, optional
+        The year to calculate, by default None.
+        If None, the previous year would be used.
+    drop : bool, optional
+        Whether to drop unused accounts, by default True.
+
+    """
+    if year is None:
+        year = datetime.now().year - 1
+
+    def patch_G(G: nx.DiGraph) -> nx.DiGraph:
+        G.add_node(-1, label="為替差益")
+        G.add_node(-2, label="為替差損")
+        G.add_edge(next(n for n, d in G.nodes(data=True) if d["label"] == "売上"), -1)
+        G.add_edge(
+            next(n for n, d in G.nodes(data=True) if d["label"] == "経費追加"), -2
+        )
+        return G
+
+    G = get_blue_return_accounts(patch_G)
+
+    gledger_vec = (
+        list(ledger_from_sales(path, G))
+        + list(ledger_from_expenses(path))
+        + list(read_general_ledger(path))
+    )
+    f = get_account_type_factory(G)
 
-app = typer.Typer()
+    def is_debit(x: str) -> bool:
+        v = getattr(f(x), "debit", None)
+        if v is None:
+            raise ValueError(f"Account {x} not recognized")
+        return v
 
+    gledger = multidimensional_ledger_to_ledger(gledger_vec, is_debit=is_debit)
+    ledger = multiledger_to_ledger(
+        generalledger_to_multiledger(gledger, is_debit=is_debit)
+    )
+    ledger_now = [line for line in ledger if line.date.year == year]
+    with pd.option_context("display.max_rows", None, "display.max_columns", None):
+        print(
+            pd.DataFrame([attrs.asdict(line) for line in ledger_now])  # type: ignore
+            .set_index("date")
+            .sort_index(axis=0)
+        )
+    gledger_now = [line for line in gledger if line.date.year == year]
+    G = get_sheets(gledger_now, G, drop=drop)
+    G_print = G.copy()
+    for n, d in G_print.nodes(data=True):
+        G_print.nodes[n]["label"] = f"{d['label']}/{d['sum_natural'].get('', 0)}"
+    for line in generate_network_text(G_print, with_labels=True):
+        print(line)
 
-@app.command()
-def main(n1: int, n2: int) -> None:
-    """Add the arguments and print the result."""
-    print(add(n1, n2))
+    # sales per month
+    print("Sales per month")
+    for month in range(1, 13):
+        G_month = get_sheets(
+            [line for line in gledger_now if line.date.month == month], G, drop=False
+        )
+        sales_deeper_node = get_node_from_label(
+            G, "売上", lambda x: not G.nodes[x]["abstract"]
+        )
+        sales_deeper = G_month.nodes[sales_deeper_node]["sum_natural"].get("", 0)
+        print(f"{month}: {sales_deeper}")

aoiro/reader/__init__.py

@@ -0,0 +1,12 @@
+from ._expenses import ledger_from_expenses
+from ._io import read_all_csvs, read_general_ledger, read_simple_csvs
+from ._sales import ledger_from_sales, withholding_tax
+
+__all__ = [
+    "ledger_from_expenses",
+    "ledger_from_sales",
+    "read_all_csvs",
+    "read_general_ledger",
+    "read_simple_csvs",
+    "withholding_tax",
+]

aoiro/reader/_expenses.py

@@ -0,0 +1,51 @@
+from collections.abc import Sequence
+from pathlib import Path
+from typing import Any
+
+from .._ledger import GeneralLedgerLineImpl, LedgerElementImpl
+from ._io import read_simple_csvs
+
+
+def ledger_from_expenses(
+    path: Path,
+) -> Sequence[GeneralLedgerLineImpl[Any, Any]]:
+    """
+    Generate ledger from expenses.
+
+    The CSV files are assumed to have columns
+    ["勘定科目"].
+    The relative path of the CSV file would be used as "取引先".
+
+    Parameters
+    ----------
+    path : Path
+        The path to the directory containing CSV files.
+
+    Returns
+    -------
+    Sequence[GeneralLedgerLineImpl[Any, Any]]
+        The ledger lines.
+
+    """
+    df = read_simple_csvs(path / "expenses")
+    if df.empty:
+        return []
+    df["取引先"] = df["path"]
+    res: list[GeneralLedgerLineImpl[Any, Any]] = []
+    for date, row in df.iterrows():
+        res.append(
+            GeneralLedgerLineImpl(
+                date=date,
+                values=[
+                    LedgerElementImpl(
+                        account="事業主借", amount=row["金額"], currency=row["通貨"]
+                    ),
+                    LedgerElementImpl(
+                        account=row["勘定科目"],
+                        amount=row["金額"],
+                        currency=row["通貨"],
+                    ),
+                ],
+            )
+        )
+    return res

aoiro/reader/_io.py

@@ -0,0 +1,177 @@
+import re
+import warnings
+from collections.abc import Iterable
+from decimal import Decimal
+from pathlib import Path
+from typing import Any
+
+import pandas as pd
+from dateparser import parse
+
+from .._ledger import GeneralLedgerLineImpl, LedgerElementImpl
+
+
+def read_all_csvs(path: Path, /, **kwargs: Any) -> pd.DataFrame:
+    """
+    Read all CSV files in the path.
+
+    Parameters
+    ----------
+    path : Path
+        The path to the directory containing CSV files.
+    **kwargs : Any
+        The keyword arguments for `pd.read_csv`.
+
+    Returns
+    -------
+    pd.DataFrame
+        The concatenated DataFrame with
+        column "path" containing the relative path of the CSV file added.
+
+    """
+    dfs = []
+    for p in path.rglob("*.csv"):
+        df = pd.read_csv(p, **kwargs)
+        df["path"] = p.relative_to(path).as_posix()
+        dfs.append(df)
+    if not dfs:
+        return pd.DataFrame(columns=["path"])
+    return pd.concat(dfs)
+
+
+def parse_date(s: str) -> pd.Timestamp:
+    """
+    Parse date.
+
+    Prefer the last day of the month if the day is not provided.
+
+    Parameters
+    ----------
+    s : str
+        The string to parse.
+
+    Returns
+    -------
+    pd.Timestamp
+        The parsed date.
+
+    """
+    return pd.Timestamp(parse(s, settings={"PREFER_DAY_OF_MONTH": "last"}))
+
+
+def parse_money(
+    s: str, currency: str | None = None
+) -> tuple[Decimal | None, str | None]:
+    """
+    Parse money.
+
+    Parameters
+    ----------
+    s : str
+        The string to parse.
+    currency : str | None, optional
+        The currency, by default None.
+        If provided, the currency
+        in the string would be ignored and replaced by this.
+
+    Returns
+    -------
+    tuple[Decimal | None, str | None]
+        The amount and the currency.
+
+    """
+    match = re.search(r"[\d.]+", s)
+    if match is None:
+        return None, None
+    amount = Decimal(match.group())
+    if currency is None:
+        currency = re.sub(r"\s+", "", s[: match.start()] + s[match.end() :])
+    return amount, currency
+
+
+def read_simple_csvs(path: Path) -> pd.DataFrame:
+    """
+    Read all CSV files in the path.
+
+    The CSV files are assumed to have columns
+    ["発生日", "金額"].
+
+    Parameters
+    ----------
+    path : Path
+        The path to the directory containing CSV files.
+
+    Returns
+    -------
+    pd.DataFrame
+        The concatenated DataFrame with columns
+        ["発生日", "金額", "通貨", "path"].
+
+    """
+    df = read_all_csvs(path, dtype=str)
+    for col in ["発生日", "金額"]:
+        if col not in df.columns:
+            df[col] = None
+
+    # parse date
+    for k in df.columns:
+        if "日" not in k:
+            continue
+        df[k] = df[k].map(parse_date)
+
+    # parse money
+    df[["金額", "通貨"]] = pd.DataFrame(
+        df["金額"].map(parse_money).tolist(), index=df.index
+    )
+
+    # set date as index
+    df.set_index("発生日", inplace=True, drop=False)
+    return df
+
+
+def read_general_ledger(path: Path) -> Iterable[GeneralLedgerLineImpl[Any, Any]]:
+    """
+    Read general ledger.
+
+    The first column is assumed to be the date.
+    For all n in N. the 2n-1-th column is assumed to be
+    the account name, and the 2n-th column
+    is assumed to be the amount.
+
+    Parameters
+    ----------
+    path : Path
+        The path to the CSV file.
+
+    Returns
+    -------
+    Iterable[GeneralLedgerLineImpl[Any, Any]]
+        The general ledger.
+
+    """
+    df = read_all_csvs(path / "general", header=None, dtype=str)
+    df.drop(columns="path", inplace=True)
+    if df.empty:
+        return
+    if len(df.columns) % 2 != 1:
+        raise ValueError("The number of columns should be odd.")
+    if len(df.columns) < 3:
+        raise ValueError("The number of columns should be at least 3.")
+    for _, row in df.iterrows():
+        values: list[LedgerElementImpl[Any, Any]] = []
+        for i in range(1, len(row), 2):
+            amount, currency = parse_money(row[i + 1])
+            if amount is None:
+                warnings.warn(f"Amount not found in {row[i + 1]}", stacklevel=2)
+                continue
+            values.append(
+                LedgerElementImpl(
+                    account=row[i],
+                    amount=amount,
+                    currency=currency,
+                )
+            )
+        yield GeneralLedgerLineImpl(
+            values=values,
+            date=parse_date(row[0]),
+        )

aoiro/reader/_sales.py

@@ -0,0 +1,158 @@
+from collections.abc import Sequence
+from decimal import ROUND_DOWN, Decimal, localcontext
+from pathlib import Path
+from typing import Any
+
+import networkx as nx
+import pandas as pd
+from account_codes_jp import get_node_from_label
+
+from .._ledger import GeneralLedgerLineImpl, LedgerElementImpl
+from ._io import read_simple_csvs
+
+
+def withholding_tax(amount: Decimal) -> Decimal:
+    """
+    Withholding tax calculation for most 源泉徴収が必要な報酬・料金等.
+
+    Parameters
+    ----------
+    amount : Decimal
+        The raw amount.
+
+    Returns
+    -------
+    Decimal
+        The withholding tax amount.
+
+    References
+    ----------
+    https://www.nta.go.jp/taxes/shiraberu/taxanswer/gensen/2792.htm
+
+    """
+    with localcontext() as ctx:
+        ctx.rounding = ROUND_DOWN
+        if amount > 1000000:
+            return round(
+                Decimal(
+                    1000000 * Decimal("0.1021") + (amount - 1000000) * Decimal("0.2042")
+                ),
+                0,
+            )
+        else:
+            return round(amount * Decimal("0.1021"), 0)
+
+
+def ledger_from_sales(
+    path: Path,
+    G: nx.DiGraph | None = None,
+) -> Sequence[GeneralLedgerLineImpl[Any, Any]]:
+    """
+    Generate ledger from sales.
+
+    The CSV files are assumed to have columns
+    ["発生日", "金額", "振込日", "源泉徴収", "手数料"].
+    If "源泉徴収" is True, the amount would be assumed by `withholding_tax()`.
+    If "源泉徴収" if False or NaN, the amount would be assumed as 0.
+    If "源泉徴収" is numeric, the amount would be assumed as 源泉徴収額.
+    The relative path of the CSV file would be used as "取引先".
+
+    Parameters
+    ----------
+    path : Path
+        The path to the directory containing CSV files.
+    G : nx.DiGraph | None
+        The graph of accounts, by default None.
+        If provided, each "取引先" would be added as a child node of "売上".
+
+    Returns
+    -------
+    Sequence[GneralLedgerLineImpl[Any, Any]]
+        The ledger lines.
+
+    Raises
+    ------
+    ValueError
+        If the transaction date is later than the transfer date.
+    ValueError
+        If withholding tax is included in transactions with different currencies.
+
+    """
+    df = read_simple_csvs(path / "sales")
+    if df.empty:
+        return []
+    df["取引先"] = df["path"].str.replace(".csv", "")
+    df["手数料"] = df["手数料"].apply(
+        lambda x: Decimal(x) if pd.notna(x) else Decimal(0)
+    )
+    df["源泉徴収"] = df["源泉徴収"].replace(
+        {"True": True, "False": False, "true": True, "false": False}
+    )
+    df.fillna({"源泉徴収": Decimal(0)}, inplace=True)
+
+    if G is not None:
+        for ca in ["売上", "仮払税金"]:
+            parent_node = get_node_from_label(
+                G, ca, lambda x: not G.nodes[x]["abstract"]
+            )
+            parent_node_attrs = G.nodes[parent_node]
+            for t in df["取引先"].unique():
+                t_attrs = {**parent_node_attrs, "label": f"{ca}({t})"}
+                t_id = f"{ca}({t})"
+                G.add_node(t_id, **t_attrs)
+                G.add_edge(parent_node, t_id)
+
+    ledger_lines: list[GeneralLedgerLineImpl[Any, Any]] = []
+    for date, row in df.iterrows():
+        ledger_lines.append(
+            GeneralLedgerLineImpl(
+                date=date,
+                values=[
+                    LedgerElementImpl(
+                        account="売掛金", amount=row["金額"], currency=row["通貨"]
+                    ),
+                    LedgerElementImpl(
+                        account="売上" if G is None else f"売上({row['取引先']})",
+                        amount=row["金額"],
+                        currency=row["通貨"],
+                    ),
+                ],
+            )
+        )
+    for (t, date, currency), df_ in df.groupby(["取引先", "振込日", "通貨"]):
+        amount = Decimal(df_["金額"].sum())
+        if currency == "":
+            withholding = withholding_tax(
+                df_.loc[df_["源泉徴収"] == True, "金額"].sum()
+            )
+            values = [
+                LedgerElementImpl(
+                    account="事業主貸", amount=amount - withholding, currency=currency
+                )
+            ]
+            if withholding > 0:
+                values.append(
+                    LedgerElementImpl(
+                        account=f"仮払税金({t})", amount=withholding, currency=currency
+                    )
+                )
+        else:
+            if (df_["源泉徴収"] == True).any():
+                raise ValueError("通貨が異なる取引に源泉徴収が含まれています。")
+            values = [
+                LedgerElementImpl(account="事業主貸", amount=amount, currency=currency)
+            ]
+        ledger_lines.append(
+            GeneralLedgerLineImpl(
+                date=date,
+                values=[
+                    *values,
+                    LedgerElementImpl(
+                        account="売掛金", amount=-amount, currency=currency
+                    ),
+                ],
+            )
+        )
+    if (df["発生日"] > df["振込日"]).any():
+        raise ValueError("発生日が振込日より後の取引があります。")
+    return ledger_lines