Eric-ontology/app/mock_data.py

from __future__ import annotations

from datetime import date, datetime, time, timedelta, timezone
from pathlib import Path
from uuid import uuid4

from pydantic import BaseModel, Field
from rdflib import Literal, Namespace, RDF, RDFS, XSD

from app.market import MarketService
from app.repository import RDFRepository


GESLI = Namespace("https://gesli.example/ontology#")
MOCK = Namespace("https://gesli.example/runtime-mock/")


class MockBatchRequest(BaseModel):
    origin: str
    destination: str
    supplier_count: int = Field(default=4, ge=1, le=8)
    customer_count: int = Field(default=2, ge=0, le=5)
    evaluation_date: date = date(2026, 7, 20)


class LocalMockDataService:
    def __init__(
        self,
        root: Path,
        repository: RDFRepository,
        market_service: MarketService,
    ):
        self.root = root
        self.repository = repository
        self.market_service = market_service

    def catalog(self) -> dict:
        candidate_rows = self.repository.query_file(
            self.root / "queries" / "market-candidates.rq"
        )["rows"]
        context_rows = self.repository.query_file(
            self.root / "queries" / "market-context.rq"
        )["rows"]
        customers = self.market_service.customers()
        evaluation_date = date(2026, 7, 20)

        routes: dict[tuple[str, str], dict] = {}
        for row in candidate_rows:
            origin = row.get("originLocation", "")
            destination = row.get("destinationLocation", "")
            if not origin or not destination:
                continue
            key = (origin, destination)
            route = routes.setdefault(key, {
                "origin": origin,
                "destination": destination,
                "products": set(),
                "suppliers": set(),
                "quotes": set(),
                "current_quotes": 0,
                "transport_modes": set(),
            })
            route["products"].add(row.get("product", ""))
            route["suppliers"].add(row.get("supplier", ""))
            route["quotes"].add(row.get("quote", ""))
            if row.get("transportMode"):
                route["transport_modes"].add(row["transportMode"])
            if _within(evaluation_date, row.get("validFrom"), row.get("validTo")):
                route["current_quotes"] += 1

        route_rows = [
            {
                "origin": route["origin"],
                "destination": route["destination"],
                "transport_modes": " / ".join(sorted(route["transport_modes"])),
                "product_count": len(route["products"]),
                "supplier_count": len(route["suppliers"]),
                "quote_count": len(route["quotes"]),
                "current_quote_count": route["current_quotes"],
            }
            for route in routes.values()
        ]
        route_rows.sort(key=lambda item: (item["origin"], item["destination"]))

        quote_rows = []
        for row in candidate_rows:
            valid_from = row.get("validFrom", "")
            valid_to = row.get("validTo", "")
            if _within(evaluation_date, valid_from, valid_to):
                status = "当前有效"
            elif valid_to and valid_to < evaluation_date.isoformat():
                status = "已过期"
            else:
                status = "尚未生效"
            quote_rows.append({
                "quote_no": row.get("quoteNo", ""),
                "route": f"{row.get('originLocation', '')} → {row.get('destinationLocation', '')}",
                "transport_mode": row.get("transportMode", ""),
                "product_name": row.get("productName", ""),
                "supplier_name": row.get("supplierName", ""),
                "supplier_cost_amount": _as_float(row.get("supplierCostAmount")),
                "currency": row.get("currency", ""),
                "valid_to": valid_to,
                "status": status,
                "dangerous_goods": row.get("supportsDangerousGoods", "false") == "true",
                "relationship_tier": row.get("relationshipTier") or "基础合作",
                "relationship_strength": _as_float(row.get("relationshipStrength"), 0.60),
                "low_price_access_probability": _as_float(
                    row.get("lowPriceAccessProbability"), 0.50
                ),
                "priority_space_probability": _as_float(
                    row.get("prioritySpaceProbability"), 0.60
                ),
                "relationship_evidence": (
                    row.get("relationshipEvidence") or "按标准合作机制执行"
                ),
                "spot_space_status": row.get("spotSpaceStatus") or "需确认",
                "spot_space_teu": _as_float(row.get("spotSpaceTeu")),
                "runtime_mock": row.get("quote", "").startswith(str(MOCK)),
            })
        quote_rows.sort(
            key=lambda item: (
                item["route"], item["status"], item["supplier_cost_amount"]
            )
        )

        strategies = [
            {
                "name": row.get("strategyName", ""),
                "objective": row.get("strategyObjective", ""),
                "route": f"{row.get('origin', '')} → {row.get('destination', '')}",
                "adjustment_rate": _as_float(row.get("strategyAdjustmentRate")),
                "valid_to": row.get("validTo", ""),
            }
            for row in context_rows
            if row.get("kind") == "strategy"
        ]

        return {
            "summary": {
                "triple_count": self.repository.triple_count,
                "in_memory_triple_count": self.repository.in_memory_triple_count,
                "customer_count": len(customers),
                "supplier_count": len({row.get("supplier") for row in candidate_rows}),
                "product_count": len({row.get("product") for row in candidate_rows}),
                "quote_count": len({row.get("quote") for row in candidate_rows}),
                "route_count": len(route_rows),
                "strategy_count": len(strategies),
            },
            "routes": route_rows,
            "customers": customers,
            "quotes": quote_rows,
            "strategies": strategies,
            "loaded_files": self.repository.loaded_file_names,
        }

    def generate(self, request: MockBatchRequest) -> dict:
        route_context = next(
            (
                route
                for route in self.market_service.form_options()["routes"]
                if route["origin"] == request.origin
                and route["destination"] == request.destination
            ),
            None,
        )
        if not route_context:
            raise ValueError("只能为已经配置市场基准和策略的路线生成Mock数据")

        token = uuid4().hex[:8]
        benchmark = float(route_context["target_amount"])
        transport_mode = self._route_transport_mode(
            request.origin, request.destination
        )
        triples = []
        generated_suppliers = []
        generated_customers = []

        for index in range(request.supplier_count):
            suffix = f"{token}-{index + 1}"
            brand = MOCK[f"brand-{suffix}"]
            supplier = MOCK[f"supplier-{suffix}"]
            product = MOCK[f"product-{suffix}"]
            offering = MOCK[f"offering-{suffix}"]
            schedule = MOCK[f"schedule-{suffix}"]
            quote = MOCK[f"quote-{suffix}"]
            performance = MOCK[f"performance-{suffix}"]
            profile = self._supplier_profile(index)
            supplier_name = (
                f"印度航线Mock{profile['supplier_label']} {token[-3:].upper()}{index + 1}"
            )
            product_name = (
                f"{request.origin}-{request.destination}{profile['product_label']}方案{index + 1}"
            )
            departure = request.evaluation_date + timedelta(days=8 + index)
            transit_days = self._transit_days(transport_mode, index)
            arrival = departure + timedelta(days=transit_days)
            cost = round(benchmark * profile["cost_ratio"], -2)
            success_rate = profile["booking_success_rate"]
            product_class = (
                GESLI.AirFreightProduct
                if transport_mode == "空运"
                else GESLI.ContainerOceanProduct
            )
            route_type = "直达" if transport_mode == "空运" and index < 2 else profile["route_type"]

            triples.extend([
                (brand, RDF.type, GESLI.ServiceBrand),
                (brand, GESLI.brandCode, Literal(f"MOCK-{token}-{index + 1}")),
                (brand, GESLI.brandName, Literal(f"Mock品牌 {index + 1}")),
                (supplier, RDF.type, profile["supplier_class"]),
                (supplier, RDFS.label, Literal(supplier_name, lang="zh")),
                (supplier, GESLI.platformSupplierType, Literal(profile["supplier_type"])),
                (supplier, GESLI.isActivePartner, Literal(True)),
                (product, RDF.type, product_class),
                (product, GESLI.productCode, Literal(f"MOCK-PRODUCT-{suffix}")),
                (product, GESLI.productName, Literal(product_name)),
                (product, GESLI.hasServiceBrand, brand),
                (product, GESLI.hasProductPerformanceProfile, performance),
                (product, GESLI.transportMode, Literal(transport_mode)),
                (product, GESLI.serviceForm, Literal(profile["service_form"])),
                (product, GESLI.routeType, Literal(route_type)),
                (product, GESLI.originLocation, Literal(request.origin)),
                (product, GESLI.destinationLocation, Literal(request.destination)),
                (product, GESLI.estimatedTransitDays, Literal(transit_days, datatype=XSD.decimal)),
                (performance, RDF.type, GESLI.ProductPerformanceProfile),
                (performance, GESLI.etdAtdOnTimeRate, Literal(success_rate, datatype=XSD.decimal)),
                (performance, GESLI.etaAtaOnTimeRate, Literal(success_rate - 0.02, datatype=XSD.decimal)),
                (offering, RDF.type, GESLI.SupplierProductOffering),
                (offering, GESLI.offeringCode, Literal(f"MOCK-OFFER-{suffix}")),
                (offering, GESLI.offeredBySupplier, supplier),
                (offering, GESLI.offeringProduct, product),
                (offering, GESLI.offeringChannel, Literal(profile["offering_channel"])),
                (offering, GESLI.agencyLevel, Literal(profile["agency_level"])),
                (offering, GESLI.capacityCommitment, Literal(profile["capacity_commitment"])),
                (offering, GESLI.capacityAvailable, Literal(True)),
                (offering, GESLI.spotSpaceStatus, Literal(profile["spot_space_status"])),
                (offering, GESLI.spotSpaceTeu, Literal(profile["spot_space_teu"], datatype=XSD.decimal)),
                (offering, GESLI.supportsDangerousGoods, Literal(index == 0)),
                (offering, GESLI.solutionAssuranceScore, Literal(profile["solution_assurance"], datatype=XSD.decimal)),
                (offering, GESLI.serviceSupportScore, Literal(profile["service_support_score"], datatype=XSD.decimal)),
                (offering, GESLI.relationshipTier, Literal(profile["relationship_tier"])),
                (offering, GESLI.relationshipStrength, Literal(profile["relationship_strength"], datatype=XSD.decimal)),
                (offering, GESLI.lowPriceAccessProbability, Literal(profile["low_price_access_probability"], datatype=XSD.decimal)),
                (offering, GESLI.prioritySpaceProbability, Literal(profile["priority_space_probability"], datatype=XSD.decimal)),
                (offering, GESLI.relationshipEvidence, Literal(profile["relationship_evidence"])),
                (offering, GESLI.bookingSuccessRate, Literal(success_rate, datatype=XSD.decimal)),
                (offering, GESLI.spaceReleaseSuccessRate, Literal(profile["space_release_success_rate"], datatype=XSD.decimal)),
                (schedule, RDF.type, GESLI.ScheduledServiceInstance),
                (schedule, GESLI.scheduleCode, Literal(f"MOCK-SCHEDULE-{suffix}")),
                (schedule, GESLI.scheduleOfProduct, product),
                (schedule, GESLI.documentCutoffTime, _date_time_literal(departure - timedelta(days=2))),
                (schedule, GESLI.estimatedDepartureTime, _date_time_literal(departure)),
                (schedule, GESLI.estimatedArrivalTime, _date_time_literal(arrival)),
                (schedule, GESLI.estimatedTransitDays, Literal(transit_days, datatype=XSD.decimal)),
                (quote, RDF.type, GESLI.SupplierQuote),
                (quote, GESLI.quoteNo, Literal(f"MOCK-QUOTE-{suffix}")),
                (quote, GESLI.quotedBySupplier, supplier),
                (quote, GESLI.quotedForOffering, offering),
                (quote, GESLI.quoteAppliesToSchedule, schedule),
                (quote, GESLI.supplierCostAmount, Literal(cost, datatype=XSD.decimal)),
                (quote, GESLI.currency, Literal("CNY")),
                (quote, GESLI.validFrom, Literal(request.evaluation_date, datatype=XSD.date)),
                (quote, GESLI.validTo, Literal(request.evaluation_date + timedelta(days=30), datatype=XSD.date)),
            ])
            generated_suppliers.append(supplier_name)

        for index in range(request.customer_count):
            suffix = f"{token}-{index + 1}"
            customer = MOCK[f"customer-{suffix}"]
            trade = MOCK[f"customer-trade-{suffix}"]
            credit = MOCK[f"customer-credit-{suffix}"]
            customer_name = f"印度目的港模拟客户 {token[-3:].upper()}{index + 1}"
            triples.extend([
                (customer, RDF.type, GESLI.Customer),
                (customer, RDFS.label, Literal(customer_name, lang="zh")),
                (customer, GESLI.hasTradeProfile, trade),
                (customer, GESLI.hasCreditProfile, credit),
                (trade, RDF.type, GESLI.CustomerTradeProfile),
                (trade, GESLI.industry, Literal("印度市场Mock行业")),
                (trade, GESLI.routeName, Literal(f"{request.origin}-{request.destination}")),
                (trade, GESLI.repeatPurchaseCount, Literal(4 + index * 6)),
                (trade, GESLI.quoteWinRate, Literal(0.5 + index * 0.08, datatype=XSD.decimal)),
                (trade, GESLI.customerPriceSensitivity, Literal(0.45 + index * 0.15, datatype=XSD.decimal)),
                (trade, GESLI.customerTimeSensitivity, Literal(0.75 - index * 0.1, datatype=XSD.decimal)),
                (trade, GESLI.customerReliabilitySensitivity, Literal(0.7 + index * 0.1, datatype=XSD.decimal)),
                (credit, RDF.type, GESLI.CustomerCreditProfile),
                (credit, GESLI.overdueDays, Literal(index)),
            ])
            generated_customers.append(customer_name)

        added_triples = self.repository.add_triples(triples)
        self.market_service.reset_runtime()
        return {
            "batch_id": token,
            "added_triples": added_triples,
            "suppliers": generated_suppliers,
            "customers": generated_customers,
            "catalog": self.catalog(),
        }

    def reset(self) -> dict:
        self.repository.reload()
        self.market_service.reset_runtime()
        return self.catalog()

    def _route_transport_mode(self, origin: str, destination: str) -> str:
        rows = self.repository.query_file(
            self.root / "queries" / "market-candidates.rq"
        )["rows"]
        return next(
            (
                row.get("transportMode", "海运")
                for row in rows
                if row.get("originLocation") == origin
                and row.get("destinationLocation") == destination
            ),
            "海运",
        )

    @staticmethod
    def _transit_days(transport_mode: str, index: int) -> int:
        return (3 + index) if transport_mode == "空运" else (16 + index * 2)

    @staticmethod
    def _supplier_profile(index: int) -> dict:
        profiles = [
            {
                "supplier_label": "船公司直营",
                "supplier_class": GESLI.Carrier,
                "supplier_type": "船公司",
                "product_label": "优先直航",
                "route_type": "直航",
                "service_form": "优先舱位包柜",
                "offering_channel": "船公司直营",
                "agency_level": "产品品牌本地分支机构",
                "capacity_commitment": "年度协议优先保舱20TEU",
                "spot_space_status": "可立即订舱",
                "spot_space_teu": 18,
                "relationship_tier": "战略合作",
                "relationship_strength": 0.96,
                "low_price_access_probability": 0.90,
                "priority_space_probability": 0.95,
                "relationship_evidence": "年度协议、价格窗口直连与优先放舱额度",
                "solution_assurance": 0.98,
                "service_support_score": 0.94,
                "booking_success_rate": 0.97,
                "space_release_success_rate": 0.96,
                "cost_ratio": 0.91,
            },
            {
                "supplier_label": "一级订舱代理",
                "supplier_class": GESLI.ShippingAgencyCompany,
                "supplier_type": "订舱代理",
                "product_label": "稳定直航",
                "route_type": "直航",
                "service_form": "稳定舱位包柜",
                "offering_channel": "一级订舱代理",
                "agency_level": "一级代理",
                "capacity_commitment": "月度保舱12TEU",
                "spot_space_status": "现舱有限",
                "spot_space_teu": 8,
                "relationship_tier": "核心合作",
                "relationship_strength": 0.87,
                "low_price_access_probability": 0.82,
                "priority_space_probability": 0.86,
                "relationship_evidence": "核心代理授权与固定订舱窗口",
                "solution_assurance": 0.90,
                "service_support_score": 0.90,
                "booking_success_rate": 0.94,
                "space_release_success_rate": 0.92,
                "cost_ratio": 0.86,
            },
            {
                "supplier_label": "稳定合作代理",
                "supplier_class": GESLI.FreightForwarderSupplier,
                "supplier_type": "货代公司",
                "product_label": "常规直航",
                "route_type": "直航",
                "service_form": "标准包柜",
                "offering_channel": "长期合作货代",
                "agency_level": "稳定合作代理",
                "capacity_commitment": "周度滚动确认8TEU",
                "spot_space_status": "现舱有限",
                "spot_space_teu": 4,
                "relationship_tier": "稳定合作",
                "relationship_strength": 0.74,
                "low_price_access_probability": 0.70,
                "priority_space_probability": 0.72,
                "relationship_evidence": "长期出货记录与周度保舱协同",
                "solution_assurance": 0.83,
                "service_support_score": 0.82,
                "booking_success_rate": 0.90,
                "space_release_success_rate": 0.88,
                "cost_ratio": 0.79,
            },
            {
                "supplier_label": "市场订舱代理",
                "supplier_class": GESLI.ShippingAgencyCompany,
                "supplier_type": "订舱代理",
                "product_label": "中转经济",
                "route_type": "中转",
                "service_form": "经济包柜",
                "offering_channel": "市场订舱代理",
                "agency_level": "二级代理",
                "capacity_commitment": "按单票确认舱位",
                "spot_space_status": "需确认",
                "spot_space_teu": 0,
                "relationship_tier": "常规合作",
                "relationship_strength": 0.52,
                "low_price_access_probability": 0.46,
                "priority_space_probability": 0.48,
                "relationship_evidence": "按市场舱位和临时报价执行",
                "solution_assurance": 0.67,
                "service_support_score": 0.68,
                "booking_success_rate": 0.84,
                "space_release_success_rate": 0.82,
                "cost_ratio": 0.72,
            },
            {
                "supplier_label": "区域货代",
                "supplier_class": GESLI.FreightForwarderSupplier,
                "supplier_type": "货代公司",
                "product_label": "加班船直航",
                "route_type": "直航",
                "service_form": "加班船包柜",
                "offering_channel": "区域货代",
                "agency_level": "合作货代",
                "capacity_commitment": "加班船窗口优先确认6TEU",
                "spot_space_status": "现舱有限",
                "spot_space_teu": 3,
                "relationship_tier": "稳定合作",
                "relationship_strength": 0.69,
                "low_price_access_probability": 0.63,
                "priority_space_probability": 0.70,
                "relationship_evidence": "区域市场协同与加班船资源预留",
                "solution_assurance": 0.80,
                "service_support_score": 0.79,
                "booking_success_rate": 0.89,
                "space_release_success_rate": 0.88,
                "cost_ratio": 0.82,
            },
        ]
        return profiles[index % len(profiles)]


def _as_float(value: str | None, default: float = 0.0) -> float:
    try:
        return float(value) if value not in (None, "") else default
    except (TypeError, ValueError):
        return default


def _as_date(value: str | None) -> date | None:
    if not value:
        return None
    try:
        return date.fromisoformat(value[:10])
    except ValueError:
        return None


def _within(current: date, start: str | None, end: str | None) -> bool:
    start_date = _as_date(start)
    end_date = _as_date(end)
    return (not start_date or start_date <= current) and (
        not end_date or current <= end_date
    )


def _date_time_literal(value: date) -> Literal:
    dt = datetime.combine(
        value,
        time(hour=12),
        tzinfo=timezone(timedelta(hours=8)),
    )
    return Literal(dt.isoformat(), datatype=XSD.dateTime)