Accounts tab and page: savings and mortgage balances kept separate from spending and the budget

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

statement_loader.py

@@ -67,6 +67,7 @@ class Statement:
     txns: list = field(default_factory=list)
     stated_out: float | None = None   # total "money out" the statement itself reports (for checking)
     kind: str = "pdf"
+    snaps: list = field(default_factory=list)   # savings / loan balances (kept on the separate Accounts tab, never in spending)
 
 # ---------------------------------------------------------------- PDF parsers
 def _pages(path):
@@ -125,6 +126,19 @@ def parse_ally(path, pages):
             cur.detail = (cur.detail + " | " + line.strip()).strip(" |")
         if re.match(r"^\d{6}/\d+//", line): cur = None
     st.txns = [t for t in st.txns if t.account == "Ally Checking"]   # savings has no spending
+    heads = list(re.finditer(r"Product: (.+?) Account Ownership", full))
+    for i, h in enumerate(heads):
+        prod = h.group(1).strip()
+        if prod.lower().startswith("spending"): continue
+        sec = full[h.end(): heads[i + 1].start() if i + 1 < len(heads) else h.end() + 3000]
+        g = lambda rx: (lambda m: money(m.group(1)) if m else None)(re.search(rx, sec))
+        bal = g(r"Ending Balance, as of \S+ \$([\d,]+\.\d\d)")
+        if bal is None: continue
+        apy = re.search(r"Annual Percentage Yield Earned ([\d.]+)%", sec)
+        st.snaps.append(dict(account="Ally " + re.sub(r" Account$", "", prod), group="Savings", date=end, balance=bal,
+                             interest=g(r"Interest Paid This Period \$([\d,]+\.\d\d)"), ytd=g(r"Interest Paid Year to Date \$([\d,]+\.\d\d)"),
+                             deposits=g(r"Deposits and Other Credits \$([\d,]+\.\d\d)"), withdrawals=g(r"Withdrawals and Other Debits -?\$([\d,]+\.\d\d)"),
+                             rate=float(apy.group(1)) if apy else None))
     return st
 
 CV_TX = re.compile(r"^(\d\d/\d\d/\d{4}) (.+?) \$([\d,]+\.\d\d) \$([\d,]+\.\d\d)$")
@@ -160,8 +174,38 @@ def parse_cvfcu(path, pages):
     for t in st.txns:
         if "SCHEDULED TRANSFER TO LOAN" in t.desc.upper() and pr and fin:
             t.extra = f"Principal ${pr:,.2f} / Interest ${fin:,.2f}"
+    st.snaps = cvfcu_snapshots(full, end)
     return st
 
+def cvfcu_snapshots(full, end):
+    """Savings/share and loan accounts from a CVFCU statement (checking is handled as transactions)."""
+    bal, kind, loan = {}, {}, False
+    for line in full.splitlines():
+        if line.startswith("Loan Account Account Number"): loan = True; continue
+        if line.startswith("Account Type Account Number"): loan = False; continue
+        m = re.match(r"^([A-Z0-9][A-Z0-9 &]+?) XXXXXX\d{4} \$(-?[\d,]+\.\d\d)$", line)
+        if m and "CHECKING" not in m.group(1):
+            bal[m.group(1)] = money(m.group(2)); kind[m.group(1)] = "Loan" if loan else "Savings"
+    info, blk = {n: {} for n in bal}, None
+    pats = dict(interest=r"Interest Paid This Period \$([\d,]+\.\d\d)", ytd=r"Interest Paid (?:Year-to-Date|YTD) \$([\d,]+\.\d\d)",
+                rate=r"Interest Rate ([\d.]+)%", payment=r"Payment Amount \$([\d,]+\.\d\d)", due=r"Payment Due Date (\d\d/\d\d/\d{4})",
+                principal=r"Principal Amount -?\$([\d,]+\.\d\d)")
+    for line in full.splitlines():
+        h = re.match(r"^([A-Z0-9][A-Z0-9 &/\.]+?) - XXXXXX\d{4}$", line)
+        if h: blk = h.group(1); continue
+        if blk in info:
+            for k, rx in pats.items():
+                m = re.search(rx, line)
+                if m and k not in info[blk]: info[blk][k] = pdate(m.group(1)) if k == "due" else float(m.group(1).replace(",", ""))
+    out = []
+    for n, b in bal.items():
+        i = info[n]; note = None
+        if kind[n] == "Loan" and i.get("payment") and i.get("principal") is not None:
+            note = f"Payment ${i['payment']:,.2f}: principal ${i['principal']:,.2f} + interest ${i.get('interest', 0):,.2f}"
+        out.append(dict(account="CVFCU " + n.title(), group=kind[n], date=end, balance=b, interest=i.get("interest"), ytd=i.get("ytd"),
+                        principal=i.get("principal"), rate=i.get("rate"), payment=i.get("payment"), due=i.get("due"), note=note))
+    return out
+
 # Online "Account Details" page saved as PDF (recent activity, no statement dates). Account is named by its product line.
 ACTIVITY_ACCOUNTS = {"PREFERRED CHECKING": "CVFCU Checking"}
 ACT_TX = re.compile(r"^([A-Z][a-z]{2}) (\d\d) (.+?) ([+-])\$([\d,]+\.\d\d) \$(-?[\d,]+\.\d\d)$")
@@ -218,7 +262,7 @@ def parse_capone_csv(path):
 
 _CACHE = {}
 _DISK = DATA / ".statement_cache.pkl"      # parsed statements survive restarts (re-reading ~30 PDFs takes several seconds)
-_CACHE_VERSION = 1
+_CACHE_VERSION = 2
 _loaded = False
 
 def _load_cache():
@@ -440,7 +484,52 @@ def _copy_style(src, dst):
     from copy import copy
     dst.font, dst.fill, dst.border, dst.alignment, dst.number_format = copy(src.font), copy(src.fill), copy(src.border), copy(src.alignment), src.number_format
 
-def apply(new_rows, workbook=WORKBOOK):
+ACCOUNT_COLS = ["Statement date", "Account", "Type", "Balance", "Interest this period", "Interest year to date", "Deposits", "Withdrawals",
+                "Principal this period", "Rate %", "Payment", "Next due", "Note"]
+ACCOUNT_KEYS = ["date", "account", "group", "balance", "interest", "ytd", "deposits", "withdrawals", "principal", "rate", "payment", "due", "note"]
+
+def account_rows(wb):
+    """Existing snapshots on the Accounts tab as dicts (empty if the tab doesn't exist yet)."""
+    if "Accounts" not in wb.sheetnames: return []
+    out = []
+    for r in wb["Accounts"].iter_rows(min_row=2, max_col=len(ACCOUNT_KEYS), values_only=True):
+        if r[0] is None or r[1] is None: continue
+        d = dict(zip(ACCOUNT_KEYS, r))
+        for k in ("date", "due"):
+            if isinstance(d[k], dt.datetime): d[k] = d[k].date()
+        out.append(d)
+    return out
+
+def pending_snaps(wb, statements):
+    have = {(r["account"], r["date"]) for r in account_rows(wb)}; out = []
+    for st in statements:
+        for sn in st.snaps:
+            if (sn["account"], sn["date"]) not in have: have.add((sn["account"], sn["date"])); out.append(sn)
+    return sorted(out, key=lambda x: (x["date"], x["account"]))
+
+def _write_accounts(wb, snaps):
+    from openpyxl.styles import Font, PatternFill
+    if "Accounts" not in wb.sheetnames:
+        ws = wb.create_sheet("Accounts")
+        for c, h in enumerate(ACCOUNT_COLS, 1):
+            cell = ws.cell(1, c, h); cell.font = Font(name="Arial", bold=True, color="FFFFFF"); cell.fill = PatternFill("solid", fgColor="1F3864")
+        ws["O1"] = "Savings and loan balances from statements. Kept apart from the Bill Log, spending totals and budget; see the Accounts page in the app."
+        ws["O1"].font = Font(name="Arial", italic=True)
+        for c, w in zip("ABCDEFGHIJKLM", (14, 26, 10, 14, 12, 12, 12, 12, 12, 8, 12, 12, 44)): ws.column_dimensions[c].width = w
+        ws.freeze_panes = "A2"
+    ws = wb["Accounts"]
+    rows = account_rows(wb) + list(snaps)
+    for r in range(2, ws.max_row + 1):
+        for c in range(1, len(ACCOUNT_KEYS) + 1): ws.cell(r, c).value = None
+    for i, d in enumerate(sorted(rows, key=lambda x: (x["date"], x["account"])), 2):
+        for c, k in enumerate(ACCOUNT_KEYS, 1):
+            v = d.get(k)
+            if isinstance(v, dt.date) and not isinstance(v, dt.datetime): v = dt.datetime.combine(v, dt.time())
+            cell = ws.cell(i, c, v); cell.font = Font(name="Arial", size=10)
+            if k in ("date", "due"): cell.number_format = "m/d/yyyy"
+            elif k in ("balance", "interest", "ytd", "deposits", "withdrawals", "principal", "payment"): cell.number_format = "$#,##0.00;($#,##0.00);-"
+
+def apply(new_rows, workbook=WORKBOOK, snaps=None):
     """Back up, then append new rows. Returns a summary dict."""
     from openpyxl import load_workbook
     BACKUPS.mkdir(exist_ok=True)
@@ -509,12 +598,13 @@ def apply(new_rows, workbook=WORKBOOK):
                                "Added to the Lists tab (category from the built-in merchant rules, or 'Other / Unsorted' when unknown) so they appear in every summary.",
                                "Set the right Category/Type on the Lists tab (and rename the payee if you like)."), 1):
             rv.cell(rr, c).value = v; _copy_style(rv.cell(rr - 1, c), rv.cell(rr, c))
+    if snaps: _write_accounts(wb, snaps)
     wb.calculation.fullCalcOnLoad = True
     tmp = Path(str(workbook) + ".new")
     wb.save(tmp)
     _restore_charts(workbook, tmp)
     tmp.replace(workbook)
-    return dict(backup=str(backup), log_rows=len(log_rows), excluded_rows=len(ex_rows), new_payees=added_payees)
+    return dict(backup=str(backup), log_rows=len(log_rows), excluded_rows=len(ex_rows), new_payees=added_payees, accounts=len(snaps or []))
 
 
 # ---------------------------------------------------------------- categorizing existing payees
@@ -647,7 +737,7 @@ def archived_statements():
 def archive_statements():
     """Once everything read is in the workbook, move the statement files into Statements/YYYY/MM and remember what they covered."""
     res = run_preview(DATA)
-    if res["new"]: raise ValueError("Add the new transactions to the workbook first.")
+    if res["new"] or res["snaps"]: raise ValueError("Add the new transactions to the workbook first.")
     idx, batch, moved = _index(), f"{dt.datetime.now():%Y%m%d%H%M%S}", 0
     for st in res["statements"]:
         if st.file in res["bad"]: continue
@@ -685,7 +775,7 @@ def run_preview(folder=DATA):
     good = [s for s in sts if s.file not in bad]
     wb = load_workbook(WORKBOOK)
     new, already = plan(wb, good)
-    return dict(statements=sts, skipped=skipped, dups=dups, checks=checks, bad=bad, new=new, already=already)
+    return dict(statements=sts, skipped=skipped, dups=dups, checks=checks, bad=bad, new=new, already=already, snaps=pending_snaps(wb, good))
 
 if __name__ == "__main__":
     cmd = sys.argv[1] if len(sys.argv) > 1 else "preview"