calminer/services/export_serializers.py

from __future__ import annotations

import csv
from dataclasses import dataclass
from datetime import date, datetime, timezone
from decimal import Decimal, InvalidOperation, ROUND_HALF_UP
from enum import Enum
from io import StringIO
from typing import Any, Callable, Iterable, Iterator, Sequence

CSVValueFormatter = Callable[[Any], str]
Accessor = Callable[[Any], Any]

__all__ = [
    "CSVExportColumn",
    "CSVExporter",
    "default_project_columns",
    "default_scenario_columns",
    "stream_projects_to_csv",
    "stream_scenarios_to_csv",
    "default_formatter",
    "format_datetime_utc",
    "format_date_iso",
    "format_decimal",
]


@dataclass(slots=True)
class CSVExportColumn:
    """Declarative description of a CSV export column."""

    header: str
    accessor: Accessor | str
    formatter: CSVValueFormatter | None = None
    required: bool = False

    def resolve_accessor(self) -> Accessor:
        if isinstance(self.accessor, str):
            return _coerce_accessor(self.accessor)
        return self.accessor

    def value_for(self, entity: Any) -> Any:
        accessor = self.resolve_accessor()
        try:
            return accessor(entity)
        except Exception:  # pragma: no cover - defensive safeguard
            return None


class CSVExporter:
    """Stream Python objects as UTF-8 encoded CSV rows."""

    def __init__(
        self,
        columns: Sequence[CSVExportColumn],
        *,
        include_header: bool = True,
        line_terminator: str = "\n",
    ) -> None:
        if not columns:
            raise ValueError("At least one column is required for CSV export.")
        self._columns: tuple[CSVExportColumn, ...] = tuple(columns)
        self._include_header = include_header
        self._line_terminator = line_terminator

    @property
    def columns(self) -> tuple[CSVExportColumn, ...]:
        return self._columns

    def headers(self) -> tuple[str, ...]:
        return tuple(column.header for column in self._columns)

    def iter_bytes(self, records: Iterable[Any]) -> Iterator[bytes]:
        buffer = StringIO()
        writer = csv.writer(buffer, lineterminator=self._line_terminator)

        if self._include_header:
            writer.writerow(self.headers())
            yield _drain_buffer(buffer)

        for record in records:
            writer.writerow(self._format_row(record))
            yield _drain_buffer(buffer)

    def _format_row(self, record: Any) -> list[str]:
        formatted: list[str] = []
        for column in self._columns:
            accessor = column.resolve_accessor()
            raw_value = accessor(record)
            formatter = column.formatter or default_formatter
            formatted.append(formatter(raw_value))
        return formatted


def default_project_columns(
    *,
    include_description: bool = True,
    include_timestamps: bool = True,
) -> tuple[CSVExportColumn, ...]:
    columns: list[CSVExportColumn] = [
        CSVExportColumn("name", "name", required=True),
        CSVExportColumn("location", "location"),
        CSVExportColumn("operation_type", "operation_type"),
    ]
    if include_description:
        columns.append(CSVExportColumn("description", "description"))
    if include_timestamps:
        columns.extend(
            (
                CSVExportColumn("created_at", "created_at",
                                formatter=format_datetime_utc),
                CSVExportColumn("updated_at", "updated_at",
                                formatter=format_datetime_utc),
            )
        )
    return tuple(columns)


def default_scenario_columns(
    *,
    include_description: bool = True,
    include_timestamps: bool = True,
) -> tuple[CSVExportColumn, ...]:
    columns: list[CSVExportColumn] = [
        CSVExportColumn(
            "project_name",
            lambda scenario: getattr(
                getattr(scenario, "project", None), "name", None),
            required=True,
        ),
        CSVExportColumn("name", "name", required=True),
        CSVExportColumn("status", "status"),
        CSVExportColumn("start_date", "start_date", formatter=format_date_iso),
        CSVExportColumn("end_date", "end_date", formatter=format_date_iso),
        CSVExportColumn("discount_rate", "discount_rate",
                        formatter=format_decimal),
        CSVExportColumn("currency", "currency"),
        CSVExportColumn("primary_resource", "primary_resource"),
    ]
    if include_description:
        columns.append(CSVExportColumn("description", "description"))
    if include_timestamps:
        columns.extend(
            (
                CSVExportColumn("created_at", "created_at",
                                formatter=format_datetime_utc),
                CSVExportColumn("updated_at", "updated_at",
                                formatter=format_datetime_utc),
            )
        )
    return tuple(columns)


def stream_projects_to_csv(
    projects: Iterable[Any],
    *,
    columns: Sequence[CSVExportColumn] | None = None,
) -> Iterator[bytes]:
    resolved_columns = tuple(columns or default_project_columns())
    exporter = CSVExporter(resolved_columns)
    yield from exporter.iter_bytes(projects)


def stream_scenarios_to_csv(
    scenarios: Iterable[Any],
    *,
    columns: Sequence[CSVExportColumn] | None = None,
) -> Iterator[bytes]:
    resolved_columns = tuple(columns or default_scenario_columns())
    exporter = CSVExporter(resolved_columns)
    yield from exporter.iter_bytes(scenarios)


def default_formatter(value: Any) -> str:
    if value is None:
        return ""
    if isinstance(value, Enum):
        return str(value.value)
    if isinstance(value, Decimal):
        return format_decimal(value)
    if isinstance(value, datetime):
        return format_datetime_utc(value)
    if isinstance(value, date):
        return format_date_iso(value)
    if isinstance(value, bool):
        return "true" if value else "false"
    return str(value)


def format_datetime_utc(value: Any) -> str:
    if not isinstance(value, datetime):
        return ""
    if value.tzinfo is None:
        value = value.replace(tzinfo=timezone.utc)
    value = value.astimezone(timezone.utc)
    return value.isoformat().replace("+00:00", "Z")


def format_date_iso(value: Any) -> str:
    if not isinstance(value, date):
        return ""
    return value.isoformat()


def format_decimal(value: Any) -> str:
    if value is None:
        return ""
    if isinstance(value, Decimal):
        try:
            quantised = value.quantize(Decimal("0.01"), rounding=ROUND_HALF_UP)
        except InvalidOperation:  # pragma: no cover - unexpected precision issues
            quantised = value
        return format(quantised, "f")
    if isinstance(value, (int, float)):
        return f"{value:.2f}"
    return default_formatter(value)


def _coerce_accessor(accessor: Accessor | str) -> Accessor:
    if callable(accessor):
        return accessor

    path = [segment for segment in accessor.split(".") if segment]

    def _resolve(entity: Any) -> Any:
        current: Any = entity
        for segment in path:
            if current is None:
                return None
            current = getattr(current, segment, None)
        return current

    return _resolve


def _drain_buffer(buffer: StringIO) -> bytes:
    data = buffer.getvalue()
    buffer.seek(0)
    buffer.truncate(0)
    return data.encode("utf-8")