aboutsummaryrefslogtreecommitdiffstats
path: root/python/report_ops/cash.py
blob: c71fcdc071ce7d6dd50b4f3ba53515657f850af8 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
import datetime
from typing import ClassVar
from functools import partial
from dataclasses import dataclass
import pandas as pd

from serenitas.utils.env import DAILY_DIR
from serenitas.utils.db2 import dbconn
from serenitas.analytics.dates import prev_business_day
from serenitas.analytics.exceptions import MissingDataError

from .misc import get_dir, dt_from_fname, Custodian


@dataclass
class CashReport:
    custodian: ClassVar[Custodian]
    date: datetime.date
    dtkey: ClassVar
    _conn: ClassVar = dbconn("dawndb")
    _staging_queue: ClassVar[set] = set()
    _insert_sql = "INSERT INTO cash_balances VALUES (%s, %s, %s, %s, %s, %s) ON CONFLICT DO NOTHING"
    _registry = {}

    def __init_subclass__(cls, custodian, dtkey):
        cls.custodian = custodian
        cls.dtkey = dtkey
        cls._registry[custodian] = cls

    def __class_getitem__(cls, key):
        return cls._registry[key]

    @classmethod
    def get_report(cls, date, fund, prefix=None):
        report_dir = get_dir(date)
        report_dir.mkdir(exist_ok=True, parents=True)
        prefix = prefix if prefix else f"{cls.custodian}_CASH_{fund}"
        p = max(
            [f for f in get_dir(date).iterdir() if f.name.startswith(prefix)],
            key=partial(dt_from_fname, dt_format=cls.dtkey),
            default=None,
        )
        if not p:
            raise MissingDataError(
                f"No reports found for fund: {prefix.split('_')[-1]} date: {date}"
            )
        return p

    @classmethod
    def commit(cls):
        with cls._conn.cursor() as c:
            c.executemany(cls._insert_sql, cls._staging_queue)
        cls._conn.commit()

    @classmethod
    def clear(cls):
        cls._staging_queue.clear()

    @classmethod
    def stage_from_row(cls, row, date, fund):
        (account, currency), amount = row
        cls._staging_queue.add(
            (
                prev_business_day(date),
                fund,
                f"{cls.custodian} Custody Account {fund}",
                account,
                currency,
                amount,
            )
        )

    @classmethod
    def to_db(cls, date, fund):
        for row in cls.yield_rows(date, fund):
            cls.stage_from_row(row, date, fund)
        cls.commit()
        cls.clear()


class NTCashReport(CashReport, custodian="NT", dtkey="%Y%m%d%H%M"):
    @classmethod
    def yield_rows(cls, date, fund):
        p = cls.get_report(date, fund)
        df = pd.read_csv(p, on_bad_lines="warn")
        df = df[df["T-NARR-LONG"] == "CLOSING BALANCE"]
        df = df[["Consolidation", "Currency code", "A-TRAN-AMT"]]
        df.columns = df.columns.str.replace(" |-|_", "", regex=True).str.lower()
        df = df.set_index(["consolidation", "currencycode"])
        yield from df.itertuples()


class UMBCashReport(CashReport, custodian="UMB", dtkey="%Y%m%d%H%M"):
    @classmethod
    def yield_rows(cls, date, fund):
        p = cls.get_report(date, fund)
        df = pd.read_excel(p, skiprows=3)
        yield from df.groupby(["Portfolio #", "Currency"]).sum(numeric_only=True)[
            "Current Balance"
        ].items()


class BNYCashReport(CashReport, custodian="BNY", dtkey="%Y%m%d%H%M%S"):
    @classmethod
    def yield_rows(cls, date, fund):
        p = cls.get_report(date, fund)
        df = pd.read_csv(p)
        df["Beginning Balance Local"] = df["Beginning Balance Local"].apply(
            lambda s: "-" + s[1:-1] if s.startswith("(") else s
        )
        df["Beginning Balance Local"] = pd.to_numeric(
            df["Beginning Balance Local"].str.replace(",", "")
        )
        yield from df.groupby(["Account Number", "Local Currency Code"]).sum(
            numeric_only=True
        )["Beginning Balance Local"].items()


class ScotiaCashReport(CashReport, custodian="SCOTIA", dtkey="%Y%m%d%H%M%S"):
    @classmethod
    def yield_rows(cls, date, fund):
        p = cls.get_report(date, fund)
        df = pd.read_excel(p, skipfooter=1)
        if df.empty:
            # No wires, so we can't skip the footer
            df = pd.read_excel(p)
        yield (
            (int(df.loc[0]["Account"]), df.loc[0]["Curr."]),
            df.loc[0]["Closing Bal."],
        )

    @classmethod
    def get_report(cls, date, fund):
        REPORT_DIR = DAILY_DIR / "Selene" / "Scotia_reports"
        try:
            return next(
                REPORT_DIR.glob(
                    f"IsoSelene_{prev_business_day(date):%d-%b-%Y}_*_xlsx.JOAAPKO3.JOAAPKO1"
                )
            )
        except StopIteration as e:
            raise MissingDataError(f"No file available for Scotia on {date}") from e