Merge pull request #866 from santhreal/fix/journal-to-float-strip-currency

fix(journal): strip currency symbols before float coerce

agent/src/tools/trade_journal_parsers.py

@@ -9,12 +9,20 @@ Excel (.xlsx/.xls) always opens as utf-8 internally via openpyxl/xlrd.
 
 from __future__ import annotations
 
+import re
 from dataclasses import asdict, dataclass
 from pathlib import Path
 from typing import Any
 
 import pandas as pd
 
+# Broker CSV/Excel cells often include ISO codes or currency glyphs around the
+# number (Schwab/IBKR "$1,234.56", JP/CN "¥1000"). Commas are already stripped;
+# without stripping these tokens float() fails and we silently store 0.0.
+_CURRENCY_TOKEN_RE = re.compile(
+    r"(?i)(?<![A-Za-z])(?:USDT|USDC|USD|EUR|GBP|JPY|CNY|HKD)(?![A-Za-z])|[$€£¥￥]"
+)
+
 FormatName = str  # "tonghuashun" | "eastmoney" | "futu" | "generic" | "unknown"
 
 _A_SHARE_EXCHANGE_MAP = {
@@ -202,7 +210,8 @@ def _to_float(val: Any, default: float = 0.0) -> float:
     if val is None:
         return default
     try:
-        s = str(val).replace(",", "").strip()
+        s = str(val).strip().replace("\u2212", "-")
+        s = _CURRENCY_TOKEN_RE.sub("", s).replace(",", "").strip()
         return float(s) if s else default
     except (ValueError, TypeError):
         return default

agent/tests/test_trade_journal.py

@@ -19,6 +19,7 @@ from src.tools.trade_journal_parsers import (
     _infer_market_from_symbol,
     _normalize_side,
     _qualify_a_share,
+    _to_float,
     load_dataframe,
     parse_tonghuashun,
     parse_eastmoney,
@@ -656,3 +657,35 @@ def test_parse_tonghuashun_float_code_cell() -> None:
     rec = parse_tonghuashun(df)
     assert len(rec) == 1
     assert rec[0].symbol == "600519.SH"
+
+
+@pytest.mark.parametrize(
+    "raw,expected",
+    [
+        ("$1,234.56", 1234.56),
+        ("USD 12.5", 12.5),
+        ("€99.00", 99.0),
+        ("¥1,000", 1000.0),
+        ("\u221212.5", -12.5),
+        ("CNY12.5", 12.5),
+    ],
+)
+def test_to_float_strips_currency_and_unicode_minus(raw: str, expected: float) -> None:
+    """US/EU/Asia broker cells with currency glyphs must not become 0.0."""
+    assert _to_float(raw) == expected
+
+
+def test_parse_generic_currency_price_not_zero() -> None:
+    df = pd.DataFrame([{
+        "datetime": "2024-01-02 10:00:00",
+        "symbol": "AAPL",
+        "side": "buy",
+        "quantity": "10",
+        "price": "$150.25",
+        "fee": "$1.00",
+    }])
+    rec = parse_generic(df)
+    assert len(rec) == 1
+    assert rec[0].price == 150.25
+    assert rec[0].fee == 1.0
+    assert rec[0].amount == 1502.5